^5, 3r^ iFS^J-^/pf^hS 
 
 PB-245 345 
 
 
 FISH PROTEIN GiSM^lNTRATE INFORMATION PACKAGE 
 PARTS I, II AND 
 
 NORTH SERVICES, INCORPORATED 
 
 PREPARED FOR 
 
 NATIONAL MARINE FISHERIES SERVICE 
 
 DISTRIBUTED BY: 
 
 *MP 
 
 National Technical Information Senrice 
 U. S. DEPARTMENT OF COMMERCE 
 
BIBLIOGRAPHIC DATA 
 SHEET 
 
 1 . Report No. 
 
 NOAA-75091704 
 
 14. Title and Subtitle 
 
 Fish Protein Concentrate Information Package 
 Parts I, II and III 
 
 PB 245 345 
 
 5. Report Date 
 
 1973 
 
 6. 
 
 7. Author(s) 
 
 8. Performing Organization Rept. 
 No. 
 
 9. Performing Organization Name and Address 
 
 North Services, Inc. 
 1700 N. Lynn Street 
 Arlington, VA 22209 
 
 10. Proiect/Task/Work Unit No. 
 
 11. Contract/Grant No. 
 
 DOC 3-33189 
 
 12. Sponsoring Organization Name and Address 
 
 NCAA, National Marine Fisheries Service 
 Office of Resource Utili.^ation 
 Regents Drive 
 College Park, MD 20740 
 
 13. Type of Report & Period 
 Covered 
 
 1961-1973 
 
 14. 
 
 15. Supplementary Notes ^^^^ ^^^ NOM" 7 5091. 7 04- 1 . (Part IV) Reel of microfilin (16 mm), 
 giving selected documentation oi the report. 
 
 a 
 o 
 
 1 
 
 u 
 
 a 
 
 16. Abstracts Tj-j^g ^5 a collection of all pertinent docuraentacion relative to the involve- 
 ment of National Marine Fisheries Service (NMFS) in the Fish Protein Concentrate (FPC) 
 Program. The purpose of this Information Bank is to provide the user with a ready 
 reference and easy access to the work on FPC undertaken or sponsored by NMFS over a 
 12-year period 1961-1973. The Information Package was selectively compiled to most 
 effectively meet potential user needs. To facilitate access to the information, the 
 package has been divided into four parts. Part I provides a summary statement for each 
 of six categories: General; Product Characteristics; Product Uses; Industrial/Economic 
 Aspects; Laboratory Processes; and Production Processes. Part II is the selected NMFS- 
 FPC Bibliography, which consists of titles, authors, type of document and call number. 
 Part III of this file contains abstracts of selected documents considered to be of 
 particular significance in the FPC Program. They are arranged in the same order as the 
 bibliography. Part IV of the file consists of the reproduction on microfilm of the 
 
 documentation determined to be of prime im- 
 portance, particularly with regard to the de- 
 tailed content thereof. These documents have 
 been microfilmed in their entirety and are in 
 the order given in the bibliography. (Part IV 
 is sold as a separate item.) 
 
 17. Key Words and Document Analysis 
 
 Fish protein concentrate 
 Fish meal 
 Documentation 
 Bibliographies 
 Food industry 
 Food processing 
 Food analysis 
 Production engineering 
 Product development 
 
 17b. Identifiers/Open-Ended Te-ms 
 
 12-year period 
 
 17a. Descriptors 
 
 Production control 
 Economic analysis 
 
 Roproducod by 
 
 NATIONAL TECHNICAL 
 INFORMATION SERVICE 
 
 U S DepArtmenI of Commerce 
 Springfield, VA. 22151 
 
 17c. COSATl Field 'Gro-ip 
 
 
 Prices subject to change 
 
 18. Availability Statement 
 
 19. Security Class (This 
 Report) 
 
 UNn.ASSIFIF.n 
 
 20. Security ( Liss (This 
 Page 
 
 UNC l.ASSlFll-n 
 
 21. No. of P.iges 
 
 22. l^icc 
 
 b» 
 
 FORM NTis-35 (REV io-'>3i ENDORSED BY ANSI AND UNESCO. 
 
 THIS FORM, MAY BE RFPRODUCFD 
 
 USCOMM-DC e26S-P74 
 
FISH PROTEIN CONCENTRATE' 
 INFORMATION PACKAGE 
 
 Work supported by: 
 
 Office of Resource Utilization 
 National Marine Fisheries Service 
 National Oceanic & Atmospheric Administration 
 U. S. Department of Commerce 
 
 1973 
 
 Work performed by: 
 
 Northrop Services, Inc. 
 
 1700 N. Lynn Street 
 
 Arlington, Virginia 22209 
 
 Contract No. DOC 3-35189 
 
 NOTICE 
 This report was prepared as an account of work conducted or sponsored 
 by the United States Government. Neither the United States nor the 
 National Oceanic and Atmospheric Administration, nor any of their 
 employees, nor any of their contractors, subcontractors, or their 
 employees, makes any warranty, express or implied, or assumes any 
 legal liability or responsibility for the accuracy, completeness or 
 usefulness of any information, apparatus, product or process disclosed, 
 or represents that the use would not infringe privately owned rights. 
 
FORWARD 
 
 Purpose 
 
 For over twelve years the National Marine Fisheries Service (NMFS), 
 previously the Bureau of Commercial Fisheries, Fish and Wildlife Service, 
 U. S. Department of the Interior, actively engaged in and sponsored research 
 dedicated to the development of Fish protein Concentrate (FPC) as a product 
 and as a process. During that time a substantial body of information was 
 developed, some of it published and some not. 
 
 The organization within the NMFS having prime responsibility for the 
 FPC program and studies was the Office of Resource Utilization. Program 
 planning and control was exercised by the Fisheries product Research and 
 Inspection (FPR&I) Division and most of the research was conducted by that 
 division's laboratories or contracted to universities and industry. Two 
 laboratories performed the bulk of the inhouse effort. These were the 
 pacific Utilization Research Center (PURC) in Seattle, Washington and the 
 Southeast Utilization Research Center (SURC) in College park, Maryland. 
 Tlie latter was kiiOwn during the height Oi. the FPC program as the National 
 Center for Fish protein Concentrate. It was in these two laboratories that 
 much of the state of the art was developed, whether we consider product 
 forms and uses or the process used in translating the raw resource into a 
 high quality protein concentrate. 
 
 A third center of activity was the Experiment and Demonstration plant 
 (EDP) in Aberdeen, Washington, where production techniques at a semiworks 
 scale were employed and a substantial amount of high grade FPC produced. 
 
 Also of importance was the expertise of the academic community em- 
 ployed in some of the more critical considerations facing the FPC program. 
 Studies of flavor and chemistry of FPC and research into its economic poten- 
 tial illustrate the range of work sponsored by N^FS at the university level. 
 Finally, industry know-how was utilized in the design, construction, and 
 operation of pilot plant facilities. 
 
 In 1972 the U.S. Congress chose not to renew sponsoring legislation. 
 As a result, the work was brought to an orderly close with most of the on- 
 going projects completed by the end of CY 1972. A number of projects car- 
 ried out during this final phase had significant results. These were studies 
 of the Aqueous Phosphate process (presscake), redesign of the process used in 
 
 11 
 
the EDP» product testing, and studies to improve handling and storage of the 
 raw material. One important project was continued until recently: a field 
 testing of product acceptability in American Samoa. 
 
 As the program wound down, several problems became apparent. These were: 
 
 (1) Although NMFS had developed a great deal of information useful to 
 any domestic or international organization interested in studying, 
 producing, or using FPC> much of that information was not in a well 
 organized, readily accessible form. It was also apparent that 
 really useful information should be segregated from that which would 
 be less meaningful to users. 
 
 (2) A number of organizations have requested and continue to request 
 information on FPC, even though the program has been discontinued. 
 These include industry interested in building processing plants 
 
 and developing countries seeking alternatives to other protein forms 
 now available. As NMFS personnel have departed the program and with 
 data not put together in a usable format, the valuable service of 
 providing useful information in a timely and effective manner has 
 become increasingly difficult. 
 
 (3) It would be irresponsible not to ensure the ready availability of 
 the information developed because of: a) the future potential of 
 FPC and b) the fact that the information could substantially assist 
 others in speeding up their research and development, avoiding mis- 
 takes, and pragmatically assessing investment risks. 
 
 Accordingly, in the interest of potential users, NMFS resolved to develop 
 a selective body of information truly representative of the research it con- 
 ducted or sponsored over the twelve year period. It should be clearly pointed 
 out, however, that this information package is restricted to work conducted 
 within the United States under the cognizance of NMFS. The bibliography, etc. 
 does not include the significant volume of information developed throughout 
 the world during this time period and appropriately cited in the Literature. 
 
 LLl 
 
users Guide 
 
 The FPC Information package is a collection of all pertinent documenta- 
 tion relative to the involvement of NMFS in the FPC program. The purpose of 
 this Information Bank is to provide the user, whether scientist, commercial 
 investigator or layman, a ready reference and easy access to the work on FPC 
 undertaken or sponsored by NMFS over a 12-year period, 1961-1973^ . 
 
 The Information package was selectively compiled to most effectively 
 meet potential user needs, primary potential users considered in structur- 
 ing the file were: 
 
 (1) The food processing industry 
 
 (2) Manufacturers (both potential FPC and current fish meal manufacturers) 
 
 (3) Universities 
 
 (4) Foreign governments and agencies 
 
 (5) State governments and agencies 
 
 (6) potential consumer groups, with emphasis on institutions 
 
 (7) NMFS/NOAA 
 
 The approach to compilation of the package consisted of the following 
 parts : 
 
 (1) Define objectives, develop basic document structure, and means of 
 access 
 
 (2) Search file, collect and categorize data 
 
 (3) Review and select pertinent data 
 
 (4) prepare information package 
 
 During data collection, files of the Washington, D.C. office of FPR&I> 
 the College park Laboratory (SURC), including the Beltsville facility, the 
 Seattle Laboratory (PURC) and the Experiment and Demonstration plant were 
 searched and candidate documents collected for review and selection. In 
 addition, all existing bibliographies and abstracts were collected, reviewed 
 and documents retrieved if available. 
 
 _' With the exception of a few selected works on the economics of FPC, all 
 documentation referred to in this file was accomplished under the cognizance 
 of or participated in by NMFS. A very substantial body of published litera- 
 ture on FPC work not sponsored by NMFS and not cited in this package does 
 exist. Suggested reference documents for that material are as follows; (a) 
 Fish protein Concentrate, A Comprehensive Bibliography (Library of Congress), 
 (b) Chemical Abstracts, (c) Biological Abstracts, (d) Marine Fisheries Ab- 
 stracts, (e) Food Science Abstracts, and (f) Nutrition Abstracts and Reviews . 
 
 iv 
 
Based on the data collected, major subject categories were identfified 
 and the documentation placed in the appropriate category. The documentation 
 was again reviewed with regard to the significance, applicability and avail- 
 ability of each. Based on these criteria, a decision was made as to whether 
 the document was to be merely listed in • the bibliography, or if it ^as to 
 be abstracted, microfilmed or both. Following this the material was care- 
 fully analyzed and the summary statements, including conclusions and recom^ 
 mendations, were developed. 
 
 To facilitate access to the information, the package has been divided 
 into four parts, part I provides a summary statement for each of six 
 categories. These categories are: 
 
 (1) General 
 
 (2) product Characteristics 
 
 (3) product Uses 
 
 (4) Industrial/Economic Aspects 
 
 (5) Laboratory processes 
 
 (6) production processes 
 
 The summary statement includes an overview of the data and information avail- 
 able in the category, the work pursued by NMFS, significant successes and 
 failures, and, where appropriate, recommended future investigations and follow- 
 up work, 
 
 part II is the selected NMFS-FPC Bibliography, which consists of titles, 
 authors, type of document and call numbero These are listed in annual chron- 
 ology by the subject categories shown above. Within a given category and 
 year, titles on the same general subject matter are listed together. The 
 types of documents included are: published and unpublished articles and 
 manuscripts; contractor final reports, and, if significant, interim reports; 
 internal NMFS reports and memoranda; miscellaneous titles from speeches, 
 papers presented at various conferences, etc. The current location of these 
 documents is given, as well as whether they are microfilmed and/or abstracted. 
 
 Part III of this file contains abstracts of selected documents considered 
 to be of particular significance in the FPC Program, They are arranged in 
 the same order as the bibliography. 
 
 Part IV of the file consists of the reproduction on microfilm of the 
 documentation determined to be of prime importance, particularly with regard 
 to the detailed content thereof. These documents have been microfilmed in 
 their entirety and are in the order given in the bibliography. 
 
The FPC Information Package will be offered for sale as two separate items. 
 
 The first three parts of the package are available in printed formo Part ;^V is 
 
 available on microfilmo These two items can be ordered from: 
 
 National Technical Information Service (NTIS) 
 U. S. Department of Commerce 
 5285 Port Royal Road 
 Springfield, Virginia 22151 
 
 VI 
 
TABLE OF CONTENTS 
 
 Section Page 
 
 PART I. FPC SUMMARY STATEMENTS 
 
 1 OVERVIEW OF THE FISH PROTEIN CONCENTRATE PROGRAM I. 1-1 
 
 1.1 Background I. 1-1 
 
 1.2 Development of the FPC Program 1.1-5 
 
 1.3 Program Completion 1.1-14 
 
 2 PRODUCT CHARACTERISTICS 1.2-1 
 
 2.1 Conclusions 1.2-2 
 
 2.2 Composition 1.2-2 
 
 2.3 Nutritional Quality 1.2-8 
 
 2.4 Toxicological Studies 1.2-21 
 
 2.5 Recommendations for Further Work 1.2-25 
 
 3 PRODUCT USES 1.3-1 
 
 3.1 Conclusions and Recommendations 1.3-1 
 
 3.2 Uses and Formulations 1.3-2 
 
 3.3 Acceptability 1.3-9 
 
 3.4 Functionality 1.3-11 
 
 4 INDUSTRIAL/ECONOMIC ASPECTS 1.4-1 
 
 4.1 Conclusions and P.ecomniendatTOTis 1.4—1 
 
 4.2 Objectives and Approach 1.4-4 
 
 4.3 Accomplishments 1.4-5 
 
 4.4 Future Outlook 1.4-18 
 
 5 LABORATORY PROCESSES 1.5-1 
 
 5.1 Conclusions 1.5-1 
 
 5.2 Background 1.5-1 
 
 5.3 IPA Process 1.5-3 
 
 5.4 Other FPC Processes 1.5-14 
 
 6 PRODUCTION PROCESSES 1.6-1 
 
 6.1 Conclusions and Recommendations 1.6-1 
 
 6.2 EDP Production 1.6-3 
 
 6.3 Problem Areas /Recommended Work 1.6-8 
 
 vil 
 
PART II. NMFS FPC BIBLIOGRAPHY 
 
 Section 
 1 
 2 
 3 
 4 
 5 
 6 
 
 GENERAL 
 
 PRODUCT CHARACTERISTICS 
 PRODUCT USES 
 
 INDUSTRIAL/ECONOMIC ASPECTS 
 LABORATORY PROCESSES 
 PRODUCTION PROCESSES 
 
 Page 
 II. 1-1 
 II. 2-1 
 II. 3-1 
 II. 4-1 
 II. 5-1 
 II. 6-1 
 
 PART III. SELECTED ABSTRACTS 
 
 1 GENERAL 
 
 2 PRODUCT CHARACTERISTICS 
 
 3 PRODUCT USES 
 
 4 INDUSTRIAL /ECONOMIC ASPECTS 
 
 5 LABORATORY PROCESSES 
 
 6 PRODUCTION PROCESSES 
 
 III. 1-1 
 III. 2-1 
 III. 3-1 
 III. 4-1 
 III. 5-1 
 III. 6-1 
 
 PART IV. SELECTED DOCUMENTATION 
 
 GENERAL 
 
 PRODUCT CHARACTERISTICS 
 PRODUCT USES 
 
 INDUSTRIAL/ECONOMIC ASPECTS 
 LABORATORY PROCESSES 
 PRODUCTION PROCESSES 
 
 IV. 1-1 
 IV. 2-1 
 IV. 3-1 
 IV. 4-1 
 IV. 5-1 
 IV. 6-1 
 
 vin 
 
Figures Page 
 
 2-1 .Weight Gains of Rats Fed Diets Containing 80 
 
 Percent Bread Supplemented with Varying Amounts 
 
 of Either FPC or Lysine 1.2-20 
 
 Tables 
 
 2-1 Essential Amino Acid Contents of Protein in Egg, 
 
 Milk, and FPC 1.2-4 
 
 2-2 Essential Amino Acid Patterns in Protein in Egg, 
 
 Milk, and FPC 1.2-5 
 
 2-3 Mineral Composition of FPC Made by IPA Extraction 
 
 of 12 Species of Fish 1.2-7 
 
 2-4 Nutritive Quality of Fish Protein 1.2-11 
 
 2-5 Nutritive Quality of Fish Protein Concentrate 
 
 Prepared by Isopropyl Extraction of Various Species 
 
 of Fish 1.2-12 
 
 2-6 Nutritive Quality of Fish Protein Concentrates 
 
 Prepared by Different Methods from Various 
 
 Species of Fish 1.2-13 
 
 2-7 Effect of Raw Material Source on the Protein 
 
 Quality of FPC Prepared by Isopropyl Alcohol 
 
 Extraction 1.2-14 
 
 2-8 Protein Quality of FPCs Prepared by Isopropyl 
 
 Alcohol Extraction of Atlantic Menhaden 1.2-15 
 
 2-9 Nutritive Quality of FPCs Prepared by Enzyme 
 
 Hydrolysis of Whole Red Hake and Evaluated 
 
 Either as a Sole Source of Dietary Protein or as 
 
 a Supplement to Wheat Flour 1.2-16 
 
 2-10 Summary of Toxicological Studies of FPC 1.2-24 
 
 3-1 Flavor of Food Products Fortified with FPC 1.3-4 
 
 3-2 Texture of Food Products Fortified with FPC 1.3-5. 
 
 3-3 Appearance of Food Products Fortified with FPC 1.3-6 
 
 4-1 Construction Costs for 200, 400, and 600 Ton/Day 
 
 FPC Plants 1.4-9 
 
 4-2 FPC Process Cost and Design Model 1.4-14 
 
 IX 
 
PhUT i 
 
 ® General 
 
 © Frodoict Ciiara€t@ristl€S 
 
 e Product Uses 
 
 e Bndystrlai Eccisi^mlc Asp@ets 
 
 ® Lals®rat©ry Processes 
 
 © Pr@dMCti©rs Fr©€©ss©s 
 
SECTION 1 
 
 OVERVIEW OF THE FISH PROTEIN CONCENTRATE PROGRAM 
 
 1.1 BACKGROUND 
 
 1.1.1 program Justification 
 
 Fish protein concentrate (FPC) is any stable, wholesome product of high 
 nutritive value, hygienically prepared from fish, in which protein and other 
 nutrient materials are more concentrated than they were in the fresh fish, A 
 more specific definition of fish protein concentrate (FPC) appears in the U.S. 
 Food and Drug Administration (FDA) regulations covering the manufacture and 
 use of FPC— As presently produced, FPC is an almost odorless and tasteless 
 food supplement which does not- create perceptible physical changes to most 
 products to which it is addedo 
 
 Beginning in 1961; thp Rnrpan of Commercial Fisheries (BCF) and its 
 successor, the National Marine Fisheries Service (NMFS), commenced a program 
 to develop the technology of producing fish protein concentrate (FPC). This 
 program accomplished much in FPC research, experimentation and the production 
 of quantities of acceptable FPC in a semiworks plant. 
 
 There were numerous justifications for the program efforts and for its 
 continuation. In the final analysis, however, the overriding consideration 
 prior to its termination was that the program appeared to be on the threshold 
 of demonstrating the technical and economic feasibility of a product that 
 would upgrade a natural resource and could help alleviate worldwide protein 
 malnutrition. Such a demonstration would include a proper engineering 
 evaluation, definitive information on resource availability and cost and a 
 clearer understanding of both the economics of processing and the market 
 potential for FPC. 
 
 - U. S. Federal Register, XXXII, February 2, 1967, p. 1217. 
 
 1. 1-1 
 
During the past two decades the populations of emerging countries* have 
 increased more rapidly than has the supply of food. As a result, much of 
 the world's population is suffering from malnutrition, particularly protein 
 malnutrition. To compound the problem, the population explosion has increased 
 the ratio, of people to land available fqr grazing animals needed to supply 
 valuable animal protein. There is a need to make new sources of protein, 
 especially animal protein, available in large quantities and at low cost. 
 
 The worldwide need for additional protein was well described by the 
 following general conclusions derived from reports of the president's Science 
 Advisory Committee _' and the United Nations Economic and Social Council — ' ; 
 
 (1) For over one-third of the present population in developing 
 countries (750 million people), the ratio of protein to 
 calorie intake in the diet is inadequate because of too 
 little protein. . ' 
 
 (2) The gap between nutritional protein requirements and 
 actual consumption of protein is widening rapidly through- 
 out the world. 
 
 (3) protein deficient diets increase susceptibility to acute 
 and chronic 'infection and disease, reduce the capacity 
 for activity, promote apathy and cause death or mental 
 and physical retardation in young children. 
 
 To alleviate these needs, inexpensive sources of a high-quality protein 
 product, compatible with the basic diets of diverse populations, are needed 
 to supplement and upgrade the lesser quality proteins which predominate in 
 the diets of many developing nations. The protein product must be such that 
 it can be stored, shipped and marketed without refrigeration throughout the 
 world and yet retain a high nutritional value. It must be available in a 
 form that can be readily introduced into a wide' variety of traditional food 
 products to ensure worldwide usage. Fish protein concentrate (FPC) appeared 
 to have the potential to fulfill these requirements. 
 
 - U.S., President, 1963-69 (Johnson), The World Food Problem, A Report of 
 the president's Science Advisory Committee (Washington, D. C, 1967). 
 
 2' United Nations, Economic and Social Council, Increasing the production 
 and use of Edible protein (E/4343, 1967). 
 
 1.1-2 
 
1.1.2 Early Efforts 
 
 Since the 1930 's there has been considerable worldwide FPC research aii<j 
 development activity. The following discussion, while by no means complete, 
 is intended to provide a sense of the international interest and involvement 
 in FPC prior to and concurrent with the program of the United States Govern- 
 ment. 
 
 In 1937 one of the earliest solvent extraction methods for making FPC 
 was developed in the Union of South Africa by Dr. G. M. Dreosti. The process 
 used ethyl alcohol as the solvent. Experimental fish flour was successfully 
 manufactured. In 1956 the Fishing Industry Research Institute in Cape Town 
 built a pilot plant for the production of fish flour. 
 
 The United Nations Children's Fund (UNICEF) commissioned an FPC plant 
 for Chile in 1956. Although the plant encountered technical difficulties 
 and was eventually abandoned, it produced sufficient quantities of FPC for 
 studies of nutritional qualities and large-scale trial use of the product. 
 
 The largest industrial investment in FPC manufacture was made in the 
 late 1950's by Astra of Stockholm, Sweden and culminated in the establish- 
 ment of a model plant in 1962. PvCccntly, they have affiliated v/ith the 
 National Biscuit Company (Nabisco, Inc.) for the marketing of FPC products. 
 
 Within the United States the most prominent industrial involvement was 
 that of Ezra Levin and the VioBin Corporation. In 1961 this company became 
 the first petitioner to seek FDA approval for FPC made from whole fish. 
 The VioBin process involved two stages of extraction using ethylene dichlor- 
 ide and isopropyl alcohol. VioBin FPC was exported to Mexico, other Latin 
 American countries and Africa. 
 
 In the early 1960 's the General Foods Company developed a patented 
 prototype process for FPC manufacture in the United States. Although re- 
 search was terminated, the company maintained an interest in FPC programs. 
 
 In 1961 the Food and Agriculture Organization (FAO) of the United 
 Nations sponsored an International Meeting on Fish Meal to discuss FPC for 
 human consumption. Later that year the FAO International Conference on 
 Fish in Nutrition appointed an ad hoc working group to study tentative FPC 
 specifications. Subsequently, the FAO Expert Panel on Fish Meal and FPC was 
 convened to discuss specifications for FPC and to evaluate action programs 
 
 1.1-3 
 
to introduce FPC into one or two countries. The panel established a list 
 of countries in which such programs might begin immediately. In order of 
 priority, the countries were Chile and Peru, Morocco, Senegal, Ghana and 
 Pakistan. In January 1964 FAO sponsored the Symposium on the Significance 
 of Fundamental Research in which technological and biochemical problems of 
 FPC development were discussed. The FAO Conference, at its 1963 and 1965 
 sessions, continued to reemphasize the potential importance of FPC and the 
 need for its further development. 
 
 Meaningful research has been conducted in other countries including 
 Brazil, Canada,- Sweden, Norway and Iceland. In October 1967, following 
 laboratory investigations of the isopropyl alcohol extraction process, the 
 Fisheries Research Board of Canada sponsored an International FPC Conference 
 in Ottawa to explore the potential of FPC and assess its development from 
 the standpoint of fisheries and the world's nutritional needs. The meeting 
 addressed raw material resources, processing technology, FPC standards, 
 industrial development and markets for FPC. Investigations into the develop- 
 ment of FPC by the Board continued throughout the 1960's. It is interesting 
 to note that current Canadian regulations for FPC are less restrictive than 
 those of the FDA in the United States. 
 1.1.3 Initial FPC Program in the Bureau of Commercial Fisheries 
 
 Support of the FAO-sponsored International Conference on Fish in Nutri- 
 tion in 1961 generated interest in the United States concerning an FPC 
 program. The Bureau, of Commercial Fisheries saw U.S. involvement as an op- 
 portunity for the increased utilization and upgrading of ocean resources, 
 potential improvement of the economic status of our fishing industry and a 
 chance to enhance the image of the United States by alleviating worldwide 
 protein malnutrition. A basic principle of the plan was that it would not 
 place the Government in competition with private industry as a basis for 
 investment in the production and marketing of FPC. 
 
 The initial FPC program in the Bureau of Commercial Fisheries was planned 
 as a three-phase program: 
 
 (1) Conduct a comprehensive literature search and establish 
 an FPC file of source documents, 
 
 (2) Conduct a world wide survey of all kno\>m and existing 
 processes for the manufacture of FPC. 
 
 (3) Establish a laboratory research program to develop 
 methods for FPC production. 
 
 1.1-4 
 
The worldwide survey was conducted in 1961 and 1962 by the BCF laboratory 
 at College park, Maryland. Identification of existing processes for FPC manu- 
 facture would minimize the duplication of research and permit evaluation of 
 the levels of world interest in FPC. Visits were made to several countries 
 in Central America, South America and Europe to obtain firsthand information 
 on the research and development efforts abroad. The survey indicated that 
 there was considerable activity and interest in FPC by government, academic 
 and private industrial institutions in many countries, but that only modest 
 financial support had been made available for FPC programs. Numerous groups 
 were working on the development of an acceptable and economically feasible 
 FPC process. However, there was no apparent coordination of effort and 
 limited sharing of advances and discoveries among the researchers. The sur- 
 vey confirmed the need for protein supplementation of the diets of many 
 developing nations and the probable suitability of FPC for reducing protein 
 malnutrition. One of the most significant conclusions of the survey was 
 that, despite considerable worldwide research and development activity, 
 there was no existing process to manufacture FPC of acceptable quality in 
 the quantities required. The conclusion was clear: BCF was justified in 
 proceeding with the development of a process suitable for the commercial 
 manufacture of FPC. 
 
 The initial program within BCF encompassed broadly based research on 
 all aspects of FPC. Included were studies on raw material requirements, 
 processes and product analyses, process research was conducted in three 
 areas: chemical, biological and physical, 
 
 1.2 DEVELOPMENT OF THE FPC PROGRAM 
 1.2.1 BCF/NMFS Effort 
 
 In July 1963, Secretary of the Interior iJdall directed the rapid 
 development of one commercially feasible method of producing FPC. Consequently, 
 research emphasis in BCF was shifted from a broad base to one specific chemical 
 process, isopropyl alcohol extraction. Experimental work concentrated on 
 developing this process and analyzing the product. ]n addition, plans were 
 made to demonstrate the commercial feasibility of the process by constructing 
 an experiment and demonstration plant. Favorably im.pressed by the results of 
 the research effort, the National Academy of Sciences urged continuing research 
 and strong support by Congress for the FPC research and development program. 
 
 1.1-5 
 
On Nov^ember 2, 1966, public Law 89-701 was signed, authorizing the 
 construction and operation of an Experiment and Demonstration plant (EDP) 
 for fish protein concentrate. To satisfy a funding deficiency, this law was 
 amended in October 1968 by Public Law 90-549. Early in March 1971 the EDP 
 commenced operations which continued into 1972. 
 
 The following is a brief chronology of the highlights of the FPC program. 
 The events and activities are presented by the time periods which identify 
 major phases of the Program. These time periods are: 
 
 July 1961 - July 1963 Covering the period from commence- 
 
 ment of the research program in BCF 
 to the Udall decision. 
 July 1963 - November 1966 Research on the isopropyl alcohol 
 
 extraction process leading to the 
 passage of PL 89-701. 
 November 1966 - October 1968 The period leading to the contract 
 
 for the EDP. 
 October 1968 - March 1971 Design and construction of the EDP. 
 March 1971 - May 1972 EDP operation. 
 
 Specific objectives of the National Marine Fisheries Service (which 
 replaced the BCF in 1970) FPC programs were seen to be as follows: 
 
 (1) Demonstration and evaluation of the continuous extraction 
 process at the semiworks scale, using recycled isopropyl 
 alcohol . 
 
 (2) production of sizable quantities of FPC which meet FDA 
 specifications and the standards acceptable in other 
 countries. 
 
 (3) provision of sample FPC for research purposes and the 
 evaluation of potential uses of the product by both 
 the public and private sectors. 
 
 (4) Development of engineering scale-up data for use by 
 industry in the design and construction of commercial 
 plants . 
 
 (5) Demonstration of the commercial feasibility of FPC 
 production, both domestically and abroad, in terms 
 of resources, economics of processing and market 
 potential of FPC as a food supplement. 
 
 1.1-6 
 
(6) Demonstration of the use of FPC as an additive for various 
 standard manufactured foods to improve their nutritional 
 quality and upgrade the diets of those consuming them. 
 
 The accomplishment of these objectives would provide a basis for 
 meeting the following goals: 
 
 (1> Development of a domestic FPG 'industry to process present!\<y 
 underutilized species, thus making better use of our natural 
 resources. 
 
 (2) Development of a potential outlet for U.S. fishermen for 
 those fishery resources not now utilized, underutilized 
 or which could be upgraded in value. 
 
 (3) production of FPC, making available to people suffering 
 from a protein deficiency (particularly children), a 
 safe, stable, low cost, nutritious concentrate compatible 
 with the foods they presently consume. 
 
 1.2.2 Other Government Participation and Support 
 
 Several government or quasi-governmental agencies played an actiye role 
 in the FPC Program. These interfaces, by the very nature of their organiza- 
 tional charters or assigned functions, provided program support, advisory 
 services, or made individual assessments 'and decisions with significant 
 impact on the overall program. 
 
 The Marine Sciences Council, the National Academy of Sciences, the Food 
 and Drug Administration and the Agency for International Development contri- 
 buted to the total management environment of the FPC Program in advisory 
 and coordinating roles. The Marine Sciences Council was established by Act 
 of Congress in 1966 to serve as a cabinet-level advisory organization within 
 the Executive Office of the President; it ended, by law, on April 30, 1971. 
 The Council had a significant impact on the FPC Program by its high level 
 program planning, goal setting and intra-governmental coordination. MSC 
 leadership in initiating the Food From the Sea program, served to bring the 
 Agency for International Development into the FPC picture. 
 
 The National Academy of Sciences provided advisory services to the FPC 
 Program since 1962, first through an ad hoc committee, and later through a 
 standing committee concerned with FPC research. Results of the NAS efforts 
 were provided to the BCF in annual program reviews, with recommendations for 
 changes in emphasis as needed. The National Academy's status as a highly 
 
 1.1-7 
 
/ 
 
 prestigious, independent research advisory group was particularly useful to 
 the BCF in establishing the credibility of the FPC objectives and program. 
 There is evidence that NAS opinions were influential in bringing about FDA 
 decisions favorable to the FPC program. 
 
 The role of the Food and Drug Administration is that of a Government 
 regulatory agency, operating under a Commissioner and implementing the pro- 
 visions of the Federal Food, Drug and Cosmetic Act of 1938. Since FPC must 
 meet the standards established by the FDA if it is to be used in food prod- 
 ucts, the agency played a critically important role in the FPC Program. 
 
 The Agency for International Development, within the Department of State, 
 is charged with worldwide responsibility for assisting developing countries 
 to help themselves. This objective complemented the objectives of the 
 Bureau of Commercial Fisheries in the FPC Program. The Agency for Inter- 
 national Development supported BCF by urging action favorable to FPC by the 
 Food and Drug Administration and by supporting financial requirements to 
 expedite development of FPC, especially the construction of the Experiment 
 and Demonstration plant. 
 1.2.3 Congressional Activity 
 
 A discussion of the governmental organizations affecting the FPC Program 
 must center on the activities of the Bureau of Commercial Fisheries/National 
 Marine Fisheries Service. These agencies were the focal point of all 
 Government research and development efforts in FPC. However, the Program 
 could never have been implemented without the direction, guidance and funding 
 of the U.S. Congress. 
 
 The first significant Congressional interest in FPC was expressed in 
 1961. Speeches drawing attention to various aspects of fish flour and FPC' 
 were made by a number of Congressmen. Congressional proponents of FPC were 
 able to secure an appropriation of $50,000 for FPC research in FY 1962, under 
 the authority of the Fish and Wildlife Act of 1956. Funding for FPC research 
 continued through this period, with annual appropriations reaching in e'xcess 
 of $400,000. 
 
 In 1962, an important consideration regarding FPC was brought before 
 Congress. The issue was the Food and Drug Administration's reluctance to 
 approve FPC unless the fish were first eviscerated, with heads, fins, intes- 
 tines and intestinal contents discarded, on the grounds that such materials 
 were "adulterated" and "filthy" under clause 3, Section 402(a) of the Food, 
 
 1.1-8 
 
Drug and Cosmetic Act of 1938. Four bills were introduced, each proposing 
 an amendment to the Act to permit FPC to be made from whole fish, provided 
 the fish were safe and pure and handled in a sanitary manner, although por- 
 tions of the raw material might be considered aesthetically objectionable 
 to some people. 
 
 Hearings were held on the bills before the Subcommittee on Health and 
 Safety of the House Committee on Interstate and Foreign Commerce. Witness 
 testimony in favor of the proposed amendment mentioned the worldwide need for 
 the product, its nutritional advantages, and universal availability of fish 
 from which it could be produced, proponents predicted a boost for the fish- 
 ing and fishery-related industries that would result from the development 
 of FPC, thus stimulating the national economy. Testimony opposing the 
 amendment questioned the need for such a product and emphasized the need 
 for maintaining aesthetic and purity standards for foods. The Department 
 of Health, Education and Welfare, parent organization of the FDA, counseled 
 against weakening the authority of the FDA in discharging its responsibilities 
 to protect the consumer. The Department of the Interior, while generally 
 favoring the amendment, finally breached the impasse by suggesting a delay 
 to await the report of the National Academy of Sciences on FPC. On the 
 basis of this recommendation, the bills were tabled and no further action 
 was taken, pending receipt of that report. 
 
 On August 14, 1964, hearings were held before the Subcommittee on 
 Merchant Marine and Fisheries of the Senate Commerce Committee to investi- 
 gate the Federal Government's research program on FPC. These hearings were 
 not in connection with any proposed legislation. The testimony was dominated 
 by personnel of the BCF, who presented a status report on their program. Tes- 
 timony by representatives of the National Academy of Sciences affirmed the 
 support of that body of the BCF efforts. 
 
 Important action on the FPC Program came two years later when bills 
 were introduced proposing the authorization of pilot plants to test the FPC 
 process developed by the Bureau of Commercial Fisheries. The primary bill, 
 S. 2720, was introduced by Senators F. H. Bartlett of Alaska and Warren 
 Magnuson of Washington. Hearings to consider S. 2720 were held on April 26 
 and May 16-17, 1966, before the Senate Commerce Committee. The Department 
 of the Interior submitted a report favoring passage of S. 2720, but recom- 
 mending only one plant instead of the five called for by the bill. The BCF 
 
 I.l- 9 
 
once again described the status of their program and expressed accord with 
 a conclusion made by the NAS that laboratory research had progressed to th&.» 
 point where larger-scale studies were necessary. Testimony was similar to 
 that of the August 1964 hearings, relative to the value of the product and 
 the need for further study. In addition, the Agency for International 
 Development testified in favor of S. 2720, because of their interest in 
 expediting development of commercially feasible FPC. 
 
 During the hearings several witnesses opposed the provision of S. 2720 
 authorizing up to five plants. Since the BCF had only one process ready 
 for further testing, it was argued that one plant would be sufficient to 
 demonstrate" the feasibility of this process. 
 
 In Senate Report 1304 of June 23, 1966, the Commerce Committee con- 
 cluded that S. 2720 was justified by the population explosion and the 
 benefits to be derived from developing a vast, underutilized protein source. 
 The Committee report advocated demonstration plants to reduce private 
 investment risk, to stimulate the development of a market for FPC and to 
 provide adequate quantities of FPC for food and technological research. 
 tviq re'^ort summarized* "'^here is no short cut that would eliminate the 
 pilot plant stage in the development of a fully successful commercial fish 
 protein concentrate." They saw a possible need for the five different plants 
 called for by the bill to test different methods of extraction. 
 
 On August 11, 1966, hearings were held before the House Subcommittee 
 on Fisheries and Wildlife Conservation (of the Committee on Merchant Marine 
 and Fisheries) on some 15 variations of S. 2720. The principal difference 
 among the several proposed bills was the number of demonstration plants to 
 be authorized. Testimony followed the pattern of earlier hearings with 
 Congressional proponents emphasizing the need for continuing and expanding 
 FPC research by means of pilot plant studies. Some representatives of the 
 food industry (General Foods and others) offered testimony in support of 
 passage, while others (National Milk producers Federation) went on record 
 in opposition to the Government's fostering the development of such a product. 
 In House Report 2165 of September 30, 1966, the House Subcommittee recommended 
 passage of S. 2720 with amendments, which would authorize only one plant, 
 either constructed or leased and would require the FDA to certify their approv- 
 al of FPC made from whole fish before any plant was acquired. 
 
 I. 1-10 
 
 / 
 
A committee of conference was convened to resolve the differences 
 between House and Senate proposals. House Report 2290 of October 16, 1966, 
 recommended that S. 2720 be amended to effect a compromise in the number of** 
 plants authorized and the money appropriated for them. The arbitrators 
 compromised on two plants, one to be constructed, the other to be leased. 
 They decided upon an initial appropriation of $1 million for the construc- 
 tion of one plant and five annual appropriations of $1,555,000 each, for 
 leasing a second plant, the operation and management of both plants, and 
 for conducting the entire program authorized by the bill. The compromise 
 was acceptable to both houses, and S. 2720 became Public Law 89-701, 
 signed by president Johnson on November 2, 1966, 
 
 The intent of Congress in PL 89-701 was to use experiment and demon- 
 stration plants to develop practicable and economic means for the produc- 
 tion of FPC by industry. Congress saw the objectives of the BCF Program as 
 the development of a wholesome, nutritious, acceptable, inexpensive and 
 stable product which would stimulate the nation's economy by the boost FPC 
 would give to the fishing and related industries. It was anticipated that 
 the development of an effective and inexpensive deterrent to worldwide 
 malnutrition would enhance che U.S. image abroad. 
 
 PL 89-701 vested the Secretary of the Interior with the authority for 
 administration of the EDP program for a period of five years from the date 
 of enactment. The Secretary was to select sites for the two plants, con- 
 duct necessary studies, research and experiments within his own Department 
 (BCF), and promote research efforts outside the Government through grants 
 and contracts. A significant limitation of the act was that the Secretary 
 could take no action on the construction or lease of plants until FDA 
 certified their approval of FPC made from whole fish. 
 
 The Secretary was to be directly responsible for administration, opera- 
 tion and maintenance of the plants, and their disposition upon expiration 
 of the program. However, he was authorized to use contractors for the 
 operation and management of the plant, and he was given a broad range of 
 discretion in acquiring methods, materials or lands necessary to meet the 
 program objectives. The public was to be allowed access to all operating 
 records and to the plants, and the Secretary was enjoined to cooperate with 
 public and private groups and individuals pursuant to the intent of the act. 
 
 1. 1-11 
 
For the FPC Program, the most significant Congressional action of 1968 
 was the amendment to PL 89-701. prior to hearings on the original act, BCF 
 had contracted with the engineering firm of Ionics, Incorporated, for a 
 cost estimate for construction of an EDP. Ionics provided an estimate of 
 $800,000, and it was based on this estimate that the $1 million had been 
 
 appropriated for plant construction. When BCF published their Request for j 
 
 f 
 proposal for the actual construction, the bids averaged approximately $1.9 
 
 million. There were a number of reasons for the disparity between the bids 
 and the Ionics estimate. FDA's failure to certify approval of FPC made 
 from whole fish legally prevented the Secretary of the Interior from pro- 
 ceeding with the acquisition of plants, thus causing over a year's delay 
 in contracting for construction. During that time, inflation escalated 
 construction costs. Intervening research and stringent FDA restrictions 
 regarding product quality had resulted in a plant design of greater com- 
 plexity than the one analyzed by Ionics, and greater complexity meant 
 higher cost. Also, the costs of required Government inspections increased 
 appreciably. 
 
 A number of bills were introduced to provide additional funds so that 
 the program, now well into the first of its authorized five fiscal years, 
 could proceed. The simplest solution was to increase the construction 
 appropriation to $1.9 million. This proposition was introduced in the House 
 as H.R. 15490 and H.R. 15535 and in the Senate as S. 3030. As it turned 
 out, S. 3030 became the bill which ultimately provided for construction of 
 the EDP. It was amended, however, to provide for the augmentation of 
 construction funds by eliminating the restrictions which PL 89-701 had placed 
 on. the $1,555,000 annual appropriations to the EDP program, so that this 
 money could also be used for construction. The bill amended the provision 
 of the act for two plants, to require that the Secretary of the Interior 
 acquire only one plant, either by construction or lease. With the funds 
 provided in the annual appropriation freed to permit their use for leasing 
 or construction and the other purposes of the act, no additional appropria- 
 tions to the EDP program were required. 
 
 On March 26 and June 26-27, 1968, hearings were held before the House 
 Subcommittee on Fisheries and Wildlife Conservation on H.R. 15490 and 15535. 
 Favorable reports were submitted by the Departments of the Interior, Commerce, 
 
 1.1-12 
 
Agriculture, HEW, the General Services Administration and the Comptroller 
 General. All other witnesses testified favorably with two exceptions. 
 Ezra Levin, the president of the VioBin Corporation, and Illinois Congress- 
 man William Springer objected to the proposed bills and, in fact, -to any 
 further participation by the Goveriament in FPC development on the premise 
 that such development was the prerogative of private industry. Mr. Levin 
 testified that he had already developed a satisfactory process for the 
 manufacture of FPC and that the Government was unfairly competing with him. 
 This view was challenged by Representative Thomas Pelly of Washington, 
 who emphasized that the Government was not attempting to usurp enterprise, 
 but rather to help it by bearing the investment costs of developing a new 
 process. He made the poin-t that the process being developed in BCF was 
 not a duplication of Ezra Levin's process, but would provide an alternative 
 rather than a competitive process. 
 
 In House Report 1782 of July 23, 1966, the Subcommittee acknowledged 
 the merits of opposing testimony. However, they felt that there yas a 
 genuine need for the legislation in order that the FPC program could be 
 continued. As a result, S. 3030 was passed by Congress and signed into law 
 as PL 90-549 on October 4, 1968, finailly clearing the way. to let the con- 
 tract for the design, construction and operation of the EDP. However, some 
 15 months of the 60 month (5 year) time period allowed for in the Act 
 initially had been lost. 
 
 With the initiation of construction on the much delayed EDP in 1970, 
 it became apparent that it would be difficult to complete the planned for 
 EDP operations phase of the Program in the remaining fiscal years author- 
 ized by PL 89-701. Without an extension, it would not be possible to ful- 
 fill completely the objectives of Congress with respect to the Act. In 
 the time available, the EDP would not genetate sufficient data to demon- 
 strate engineering and economic feasibility, and private investors would 
 still be reluctant to proceed with FPC manufacture. 
 
 This fear was communicated to Congress, and in March 1971 a bill 
 (S. 1273) to extend the entire program for an additional fiscal year was 
 introduced in the Senate. The need for extension was related to the 
 several delays, including plant design and construction, which, in turn 
 would result in insufficient data and information being generated from 
 
 1.1-13 
 
plant operations. More time was needed to remedy unexpected complications 
 arising from adapting a novel laboratory process to semiworks -scale production. 
 Unexpected instability of lean fish supplies had also forced a shift to fatfCy 
 fish operation, resulting in insufficient data on the processing of either 
 kind. I 
 
 In October 1971 the Senate passed S. 1273, which proposed appropria- 
 tions for an additional year of operation of the EDP. The bill also amended 
 the wording of the original Act to extend the authority of the Secretary of 
 Commerce to administer the Program through June 30, 1973. However, although 
 hearings were held in the House, no bill was passed. Thus, the FPC Program 
 was terminated in 1972. 
 
 In 1966, another act which was to have great influence on the FPC 
 Program was signed into law. This was the Marine Resources and Engineering 
 Development Act of 1966, Public Law 89-454. It established the National 
 Council on Marine Resources and Engineering Development or, as it was more 
 popularly known, the Marine Sciences Council, within the Executive Office 
 of the president. During the existence of the Council, the president used 
 its annual report as a basis for his own report to Congress on marine 
 sciences and as justification for his budget request on marine programs. 
 
 In 1967 another piece of legislation included a potential source of 
 fiscal support for the FPC Program. The Foreign Assistance Act of 1967 
 amended the Foreign Assistance Act of 1961, giving the president authority 
 to conduct a program to demonstrate the potential, and encourage the use of 
 fish and other protein concentrates to reduce nutritional deficiencies in 
 less developed areas of the world. The program was to include food tech- 
 nology studies, the development of suitable marketing techniques, the 
 fostering of consumer acceptance and feeding programs to demonstrate the 
 value of the concentrates as a food supplement. The president was to en- 
 courage full participation in this effort by private enterprise and to 
 consult with the appropriate Government agencies, including the Marine 
 Sciences Council, Two million five hundred thousand dollars was provided to 
 support the goals of this Act, 
 
 1.3 PROGRAM COMPLETION 
 1.3.1 Alternate plan 
 
 Efforts were made to obtain funds to continue the FPC program through- 
 
 K 1.1-14 
 
out 1972. An alternative program plan was developed under contract to NMFS» 
 The objective of this alternate plan was to complete as expeditiously as 
 possible the major objectives of the original program by resolving the prob- 
 lems not foreseen in the course of initial program planning. Some parts of 
 the alternate program were implemented, but all developmental work was ter- 
 minated in 1972. 
 
 1.3.2 Conclusions 
 
 A review and analysis of the accomplishments and status of the FPC 
 program at its termination enable a number of conclusions to be drawn. 
 These conclusions are presented in the following Sections of this Part, 
 Recommendations for the additional work that is necessary to meet program 
 objectives are also listed below. 
 
 1.3.3 Future Work 
 
 At the time of program termination, additional work remained to com- 
 pletely fulfill the goals of the FPC program and satisfy the intent of 
 Congress, as stated in PL 89-701. This work is as follows: 
 process Development and EDP Operations 
 
 (1) Resolve knov^^l EDP processing problems (in the laboratory) and 
 modify the EDP as required. 
 
 (2) Conduct sustained plant operation to determine yields, improve 
 basic IPA extraction processes, and evaluate major process 
 variations to provide valid engineering and operating cost 
 
 data. 
 
 (3) produce substantial quantities of FPC meeting FDA specifica- 
 tions for evaluation purposes. 
 
 Engineering Evaluation 
 
 (1) Evaluate the validity of design modifications based on past 
 EDP experience and current laboratory experiments. 
 
 (2) Collect and analyze all engineering data, assess for techni- 
 cal integrity and publish selectively. 
 
 (3) prepare an Engineering Design package to assist industry in 
 the development of commercial FPC plants. 
 
 Economics and Marketing 
 
 (1) Determine locations of adequate supplies of suitable low- 
 cost raw material for processing in established FPC plants. 
 
 1.1-15 
 
(2) Establish the economics of processing, based on cost infor- 
 mation produced in the EDP and further expansion and re- 
 finement of the Engineering Economic Model (EEM). 
 
 (3) Further investigate and analyze the potential FPC market 
 within the U.S. and abroad. 
 
 product Acceptability 
 
 (1) Continue effort to improve FPC as a product with regard 
 to organoleptic qualities and use in food products. 
 
 (2) Evaluate results on distribution of FPC to AID, Samoa, 
 Jamaica and food processors. 
 
 Although the above work has not been completed, much valuable informa- 
 tion bearing on the technological development and commercial feasibility of 
 FPC has been derived from the program. The Sections summarize the effect 
 and results evolving from these efforts. 
 
 1.1-16 
 
SECTION 2 
 
 PRODUCT CHARACTERISTICS SUMMARY- 
 
 2.1 CONCLUSIONS 
 
 Fish protein concentrate (FPC) is an effective protein supplement 
 and could become a significant source of dietary protein provided other 
 aspects such as relative cost and organoleptic characteristics are also 
 satisfied . 
 
 From the evidence available, it is clear that the balance of amino 
 acids in fish protein and its quality can be preserved during processing 
 into FPC. With few exceptions, FPCs prepared by solvent extraction have 
 been shown to be high in protein with a quality at least equivalent to 
 that of casein. FPC prepared by extraction with ethylene dichloride has, 
 in some cases, been shown to be slightly lower in quality. When evaluated 
 in protein-depleted animals, FPC has usually been shown to be as effective 
 as casein in stimulating weight gains and regeneration of serum proteins. 
 FPC is also an excellent source of many essential minerals, especially 
 when prepared from whole fish. Numerous studies have amply demonstrated 
 that FPC, at .levels up to 6 to 8 percent, can effectively enhance the 
 quantity and quality of protein in foods that contain protein of vegetable 
 origin. FPC may be most effective nutritionally and economically when 
 combined with other supplements. Toxicological studies have shown that 
 FPC prepared by several processes from different species of fish is safe 
 and wholesome. 
 
 The remainder of this summary highlights the results of studies on the 
 composition, nutritional characteristics, and safety of FPC and identifies 
 areas which may need further study. Emphasis is placed on FPC prepared by 
 solvent extraction and designed for nutritional purposes as an ingredient 
 in foods. 
 
 £' Specific references are not footnoted in this summary. However, all 
 reference material used is noted under item B066 of the microfilmed docu- 
 ments. Nutritional and Safety Characteristics of FPC . 
 
 1.2-1 
 
2.2 COMPOSITION 
 
 2.2.1 proximate Composition 
 
 The proximate composition o-f FPC (crude protein, ash, lipids and vola- 
 tiles) will be dealt with only in a general way, since this is affected 
 greatly by the source of raw material and the process used. Crude protein 
 in FPC ranges from less than 75 percent to as high as 95 percent, ash from 
 2 to over 25 percent, lipids from 0.1 to over 0,5 percent and volatiles 
 from 1 to over 10 percent. 
 
 The protein and ash contents varied inversely and were affected by 
 the species of fish processed. For example, when whole fish were used, 
 FPC made from menhaden was lower in protein and higher in ash than FPC made 
 from herring. This variation could be eliminated largely by mechanically 
 deboning the fish prior to extraction. This results in FPC with protein 
 and ash content of about 90 and 5 percent, respectively. 
 
 The lipid content of FPC depends on the efficiency of removal during 
 solvent extraction. Efficient extraction usually resulted in a residual 
 amount of 0.5 percent or less. Medwadowski et al . showed that lipids in 
 FPC are relatively stable when present at levels of 0.1 percent or less. 
 When FPC contained more than 0.5 percent lipids, changes occurred in the 
 lipids during a six-month storage at 37 and 50 C that were reflected by 
 losses in extractibility and severe losses of highly unsaturated fatty 
 acids. 
 
 As a point of interest, U.S. FDA regulations specify that FPC shall 
 not contain less than 75 percent crude protein and not more than 0.5 
 percent lipid and 10 percent moisture. For these comiponents, Canadian FDD 
 regulations specify only that "fish protein" shall not contain less than 
 74 percent crude protein. 
 
 2.2.2 Amino Acids 
 
 The essential amino acid compositions of FPC and of egg and milk are 
 shown in Table 2-1. Values shown for FPC are averages and ranges for 
 several samples prepared by isopropyl alcohol extraction of whole hake, 
 menhaden, alewife, herring, anchovy and ocean pout. In general, the essen- 
 tial amino acids are somewhat lower in FPC than in egg protein, used as 
 an optimum reference standard. Lysine is an exception; it is higher in 
 FPC than in egg protein or milk protein. Additional data have been 
 
 1.2-2 
 
published in the literature but, in general, the amino acid values do not 
 vary significantly from those reported here. 
 
 The essential amino acid patterns of these three proteins are shown 
 in Table 2-2. Although the total amounts of essential amino acids were 
 lower in FPC, data in this table show that the pattern of amino acids in 
 FPC compared favorably with those of egg and milk. Isoleucine, total 
 aromatic amino acids, and valine were somewhat lower in FPC and lysine 
 higher. 
 
 These data do not completely agree with the results from studies on 
 the limiting amino acids in fish protein. Evidence available from the 
 literature indicated that methionine was usually the first limiting amino 
 acid in FPC and in fish meal. Miller reported that the supplementation 
 of five commercial fish meals with methionine increased the net protein 
 utilization (NPU) 13 to 58 percent. Likewise, Yanez et al, showed that 
 the NPU of FPC prepared by ethanol extraction of hake increased signifi- 
 cantly when supplemented with methionine but^ did not increase when supple- 
 mented with lysine. Stillings et al. studied the limiting amino acids in 
 isopropyl alcohol-extracted FPC prepared from red hake and grouped the 
 amino acids according to their -limitation from greatest to least as 
 follows: (1) methionine, (2) histidine, tryptophan and threonine, (3) va- 
 line, isoleucine and phenylalanine and (4) leucine, lysine and arginine. 
 Makdani et al. reported that histidine and methionine appeared to be equally 
 first limiting in FPC prepared by either isopropyl alcohol extraction or 
 by extraction with ethylene dichloride. 
 
 1.2-3 
 
TABLE 2-1 
 
 ESSENTIAL AMINO ACID CONTENTS OF. PROTEIN IN EGG, MILK, AND FP|] 
 
 (mg/gm of total nitrogen) 
 
 
 Hen's . 
 egg - 
 
 Cow's 
 milk -' 
 
 
 FPc y 
 
 
 
 Amino Acids 
 
 Average 
 
 Range 
 
 Isoleucine 
 
 A15 
 
 407 
 
 265 
 
 (250 
 
 _ 
 
 273) 
 
 Leucine 
 
 553 
 
 630 
 
 450 
 
 (415 
 
 - 
 
 476) 
 
 Lysine 
 
 403 
 
 496 
 
 509 
 
 (493 
 
 - 
 
 533) 
 
 Total aromatic AAs 
 
 627 
 
 634 
 
 442 
 
 (421 
 
 - 
 
 464) 
 
 Phenylalanine 
 
 365 
 
 311 
 
 246 
 
 (231 
 
 - 
 
 254) 
 
 Tyrosine 
 
 262 
 
 323 
 
 196 
 
 (188 
 
 - 
 
 210) 
 
 Total Sulfur AAs 
 
 346 
 
 211 
 
 235 
 
 (221 
 
 _ 
 
 2^6) 
 
 Cystine 
 
 149 
 
 57 
 
 44 
 
 ( 36 
 
 - 
 
 54) 
 
 Methionine 
 
 197 
 
 154 
 
 191 
 
 (182 
 
 - 
 
 203) 
 
 Threonine 
 
 317 
 
 292 
 
 259 
 
 (248 
 
 _ 
 
 271) 
 
 Tryptophan 
 
 100 
 
 90 
 
 73 
 
 ( 61 
 
 - 
 
 82) 
 
 Valine 
 
 454 
 
 440 
 
 314 
 
 (282 
 
 - 
 
 341) 
 
 Total essential 
 
 
 
 
 
 
 
 amino acids 
 
 3215 
 
 3200 
 
 2547 
 
 (2391 
 
 — 
 
 2696) 
 
 - Source: FAO 
 
 2/ 
 
 — FPC made from red hake, Atlantic menhaden, alewife, Atlantic 
 
 herring, Pacific anchovy, and ocean pout. Source: Sidwell et al. 
 
 1.2-4 
 
TABLE 2-2 
 
 ESSENTIAL AMINO ACID PATTERNS IN PROTEIN IN EGG, MILK, AND FPC 
 (mg/gm of total essential amino acids) 
 
 Amino Acid 
 
 Hen's 
 egg 
 
 Cow's 
 
 Milk 
 
 FPC 
 
 Isoleucine 
 
 129 
 
 127 
 
 Leucine 
 
 172 
 
 196 
 
 Lysine 
 
 125 
 
 155 
 
 Total aromatic AAs 
 
 195 
 
 197 
 
 Phenylalanine 
 
 114 
 
 97 
 
 Tyrosine 
 
 81 
 
 100 
 
 Total sulfur AAs 
 
 107 
 
 65 
 
 Cystine 
 
 46 
 
 17 
 
 Methionine 
 
 61 
 
 48 
 
 Threonine 
 
 99 
 
 91 
 
 
 31 
 
 28 
 
 Valine 
 
 141 
 
 137 
 
 104 
 177 
 200 
 
 174 
 97 
 77 
 
 92 
 17 
 75 
 
 102 
 
 28 
 
 123 
 
 1.2-5 
 
2.2.3 Minerals 
 
 In considering the nutritional characteristics of FPC, most emphasis 
 and attention was given quite logically to the protein component. It is 
 often overlooked, however, that FPC is an excellent source of several 
 essential minerals. Table 2-3 shows the mineral composition of FPC made 
 from several species of whole fish by isopropyl alcohol extraction. 
 
 There is considerable variation in the content of most minerals. 
 This can be attributed primarily to the species of fish used since all 
 samples were processed by the same general method. Values for fluoride, 
 especially, vary widely and were found to range up to 600 ppm in FPC made 
 from non-deboned Pacific hake. However, FPC could contribute significantly 
 to the requirements for several essential minerals. Based on the median of 
 the ranges shown in Table 2-3, as little as 10 grams of FPC would furnish 
 the following percentages of the recommended daily allowances (RDAs) for an 
 8- to 10-year old child: calcium, 40 percent; phosphorus, 25 percent; iron, 
 45 percent and magnesium, 10 percent. FPC is also a good source of other 
 essential trace elements such as copper, zinc, chromium, fluoride and 
 selenium. It has been suggested that FPC may not be an adequate source of 
 potassium for optimum protein metabolism. Also, it should be pointed out 
 that the removal of bone during processing will have an effect on the miner- 
 al content of FPC. 
 
 Another important aspect is the biological availability of minerals 
 in FPC. Studies conducted in humans have shown that calcium and phosphorus 
 are highly available. On the other hand, the iron in FPC has been reported 
 to be only 28 percent as available as that in ferrous sulfate, but similar 
 to that in egg yolk. Several studies on the availability of fluoride have 
 indicated a high availability at high intakes. 
 
 1.2-6 
 
TABLE 2-3 
 MINERAL COMPOSITION OF FPC MADE BY IPA EXTRACTION OF 12 SPECIES OF FISH - 
 
 % 
 
 Calcium 2.75 - 5.05 
 
 Phosphorus 1.87 - 3.39 
 
 Potassium 0.59 - 1.10 
 
 Sodium 0.34 - 0.44 
 
 Magnesium 0.16 - 0.28 
 
 ppm 
 
 Iron 147 - 770 
 
 Copper 6 - 21 
 
 Zinc 112 - 163 
 
 Chromium 8 - 19 
 
 Fluorine 58 - 162 
 
 Selenium 1.8 - 3.8 
 
 1/ 
 
 Source: Finch 
 
 1.2-7 
 
2.2.4 Vitamins 
 
 FPC prepared by solvent extraction was not a significant source of 
 vitamins. Extraction of most of the lipids during processing removed 
 fat-soluble vitamins. Water-soluble vitamins were removed in the aqueous 
 fraction and were also shown to be low in FPC. 
 
 2.3 NUTRITIONAL QUALITY 
 2.3.1 General 
 
 Fish flesh has long been known as an excellent source of high quality 
 protein. Although few published data are available, examples of the 
 quality of protein in fish are shown in Table 2-4. As indicated by these 
 data, the quality of protein in a variety of fish was shoivrn to be high and 
 consistently above that of beef protein. In processing fish into FPC to 
 be used as a protein supplement, one obvious objective was to retain to the 
 fullest extent possible the quality of the original fish protein. Dubrow 
 et al. found that it is indeed possible to retain the quality of the protein 
 during the processing of fish into FPC. In that study, the protein quality 
 of freeze-dried, whole red hake caught during different seasons was compared 
 to that of FPC made from the same starting material by isopropyl alcohol 
 extraction. With the exception of one sample, the protein efficiency ratio 
 (PER) value for the FPC was virtually the same as that for the starting 
 freeze-dried raw hake. There were, however, considerable differences among 
 the PER values of starting material caught during different seasons of the 
 year . 
 
 The raw material source, as well as the processing procedures, could 
 affect the quality of protein in FPC. Since the quality of protein in FPC 
 could not be expected to be higher than that in the raw material, it was 
 important that the protein in the starting m.aterial be undamaged. Table 2-5 
 shows the quality of FPCs prepared from several species of fish. In these 
 studies, FPC was prepared by isopropyl alcohol extraction of frozen, whole 
 raw fish at the NMFS College park Laboratory, except for hake, ( Merluccius 
 productus), which was prepared at the EDP. The latter was held in brine and 
 deboned prior to extraction. The data from these studies showed that the 
 quality of protein in FPC prepared from these species of fish was reasonably 
 consistent and equal to or slightly higher than that of casein. Thus, it 
 could be expected that if fish were adequately preserved after catching, 
 
 1.2-8 
 
the species of fish used in the prgduction of FPC would have only a slight 
 effect on the quality of protein in the FPC prepared therefrom. It was 
 noted that the quality of protein in FPC prepared from some whole, fatty 
 fish such as menhaden and sardines tended to be slightly lower than that 
 of FPC prepared from lean fish such as hake. 
 
 The literature contains numerous references to studies on the nutri- 
 tional quality of FPC prepared from different species and by different 
 processes. Table 2-6 summarizes data from some of these studies. With 
 the exception of the sample prepared by ethylene dichloride extraction, 
 the PER values were high and consistently above that for casein. Several 
 investigators have used methods other than PER for determining the protein 
 quality of FPC, such as net protein utilization (NPU) and relative nutri- 
 tive value (RNV) . As with PER evaluations, these methods have also shown 
 the protein quality of FPCs to be generally equal to or higher than that 
 of casein. Thus, from the data available it was evident that FPC with 
 protein of high quality could be produced from several species and by 
 several solvent extraction processes. Compared to other protein supplements, 
 the PER for isopropyl alcohol-extracted FPC was reported to be about 60 
 percent higher than that for soy and cottonseed flour. 
 
 In addition to the effect of different species of fish, the protein 
 quality of FPC was also affected by whether the whole fish or only parts of 
 the fish were used. For example, if fish were deboned prior to extraction, 
 the protein quality of the final FPC was slightly higher than that made 
 from whole fish. presumably, this slight difference in quality could be 
 attributed to the removal of lower quality protein associated with the 
 bone fraction and the retention of the higher quality flesh protein in the 
 final FPC. 
 
 H. E. power compared the nutritive value of FPC prepared from whole 
 fish, fillets, headed and eviscerated fish, trimmings and presscake. As 
 shown in Table 2-7, FPC prepared from fillets was considerably higher in 
 quality than that made from the whole fish. The use of headed and evis- 
 cerated fish or of trimmings from fish had little effect on the protein 
 quality of FPC compared to that made from whole fish. Wlien FPC was made 
 from presscake prepared from cod trimmings, however, the protein quality 
 of the FPC was considerably lower than that made from whole fish, pre- 
 sumably due to heat damage during processing. Other work has sho\>m that 
 
 1.2-9 
 
it is possible to produce a high quality FPC from presscake otj- from fish 
 meal. As shown in Table 2-8, there was little difference in the protein 
 quality of menhaden FPC prepared from whole fish, deboned fish, presscake 
 or fish meal . 
 
 A recent study reported that FPC made from decomposed fish x^ras higher 
 in quality than that made from fresh fish. In this study freshwater fish 
 were allowed to decompose by storing them in *the laboratory for 24 hours, 
 then they were processed into FPC. Although the yield and organoleptic 
 qualities were poorer, the PERs of FPC made from decomposed fish were 
 about 25 percent higher than those of FPC made from fresh fish. 
 
 Mention should also be made of the nutritional quality of FPC produced 
 by enzymatic hydrolysis. This type of FPC was not consistently as high in 
 quality as FPC produced by solvent extraction because of a less favorable 
 balance of amino acids. As shown in Table 2-9, enzymatically produced 
 FPCs could be an effective protein supplement. When evaluated as a sole 
 source of protein, the PERs ranged from 75 to 82 percent of that for casein. 
 When added to wheat flour on a protein basis, the PERs were 91 to 99 percent 
 of the values obtained with comparable mixtures of casein and wheat flour. 
 
 Although the quality of protein in solvent-extracted FPC was generally 
 reported to be at least equal to that in casein, there were some exceptions. 
 Morrison and McLaughlan found that the PER values for several FPCs varied 
 from 60 to 100 percent of the PER for casein. Specifically, FPC prepared 
 by ethylene dichloride extraction was shown to be of lesser quality than 
 that prepared by alcohol extraction. 
 
 1.2-10 
 
TABLE 2-4 
 
 NUTRITIVE QUALITY OF FISH PROTEIN _' 
 
 1/ 
 
 Source of 
 
 Rats 
 
 
 PER 
 
 
 dietary protein 
 
 No Sex 
 
 Actual 
 
 
 % of beef 
 
 Halibut 
 
 
 20 
 
 M 
 
 
 
 20 
 
 F 
 
 Lingcod 
 
 
 15 
 
 M 
 
 Lemon Sole 
 
 
 15 
 
 M 
 
 White spring 
 
 salmon 
 
 15 
 
 M 
 
 Red snapper 
 
 
 15 
 
 M 
 
 Herring 
 
 
 15 
 
 F 
 
 Beef 
 
 
 20 
 
 M 
 
 
 
 20 
 
 F 
 
 3.27 
 
 114 
 
 2.98 
 
 104 
 
 3.60 
 
 126 
 
 3.66 
 
 128 
 
 3.68 
 
 129 
 
 3.86 
 
 135 
 
 3.18 
 
 111 
 
 2.86 
 
 100 
 
 2.63 
 
 92 
 
 _' Source: 
 
 Beveridge 
 
 1.2-11 
 
TABLE 2-5 
 
 NUTRITIVE QUALITY OF FISH PROTEIN CONCENTRATES PREPARED BY LSOPROPYL 
 ALCOHOL EXTRACTION OF VARIOUS SPECIES OF FISH U 
 
 
 Species of fish 
 
 No. 
 Samples 
 
 PER 
 
 Hake, Urophycis chuss 
 Hake, Merluccius productus 
 Menhaden, Brevoortia tyrannus 
 Herring, Clupea harengus harengus 
 Anchovy, Engraulis mordax 
 Ocean poun, Macrozoarces amerlcanus 
 Alewife, Alosa pseudoharengus 
 Moroccan sardine, Sardinia pilchardus 
 
 
 % of casein 
 
 58 
 
 107 
 
 10 
 
 112 
 
 6 
 
 102 
 
 1 
 
 105 
 
 1 
 
 108 
 
 1 
 
 102 
 
 1 
 
 106 
 
 1 
 
 99 
 
 1/ 
 
 Source: Sidwell et al. and NMFS, unpublished. 
 
 1.2-12 
 
TABLE 2-6 
 
 NUTRITIVE QUALITY OF FISH PROTEIN CONCENTRATES PREPARED BY DIFFERENT 
 
 METHODS FROM VARIOUS SPECIES OF FISH 
 
 Species of fish and No. 
 
 processing method samples Per Ref , 
 
 % of casein 
 
 Hake, Merluccius gayi ; dried, hexane 4 106 18, 19 
 
 or ethanol extraction; Chile 
 
 Sardine, Sardinella longiceps ; cooked 1 117 20 
 
 and pressed, ethanol extraction; 
 India 
 
 Bombay Duck, Harpoden nehereus ; 6 108 21 
 
 cooked and pressed, ethanol 
 extraction; India 
 
 Krill, Euphausia superba Dana; isopropyl 1 103 22 
 
 alcohol extraction; Australia 
 
 Cod, Gadus morhua ; isopropyl alcohol 1 106 4 
 
 extraction; Canada 
 
 Herring, Clupea harengus harengus ; 1 110 4 
 
 isopropyl alcohol extraction; 
 Canada 
 
 Hake, Urophycis chuss ; ethylene di- 1 89 23 
 
 chloride extraction; U.S.A. 
 
 1.2-13 
 

 TABLE 2-7 
 
 EFFECT OF RAW MATERIAL SOURCE ON THE PROTEIN QUALITY OF FPC PREPARED BY 
 
 ISOPROPYL ALCOHOL EXTRACTION 1.1 
 
 1/ 
 
 2/ 
 
 Raw material 
 source from Cod 
 
 PER 
 
 % of casein 
 
 Whole 106 
 
 Fillets 119 
 
 Headed & eviscerated 103 
 
 Trimmings 103 
 
 2/ 
 
 Trimmings press cake— ' 88 
 
 Source: Power 
 
 Trimmings cooked with indirect heat and pressed to 
 60 percent moisture. 
 
 1.2-14 
 
TABLE 2-8 
 
 PROTEIN QUALITY OF FPCs PREPARED BY ISOPROPYL ALCOHOL EXTRACTION OF 
 
 ATLANTIC MENHADEnI' 
 
 Raw material 
 Source 
 
 PER 
 
 % of casein 
 Whole fish 101 
 
 Deboned fish 104 
 
 Whole f ish-presscake 98 
 
 Whole fish-fish meal 99 
 
 1/ 
 
 Source: Stillings, Unpublished. 
 
 1.2-15 
 
TABLE 2-9 
 
 NUTRITIVE QUALITY OF FPCs PREPARED KY ENZYME HYDROLYSIS OF WHOLE RED 
 HAKE AND EVALUATED EITHER AS A SOLE SOURCE OF DIETARY PROTEIN OR AS" 
 
 A SUPPLEMENT TO WHEAT FLOUR U 
 
 
 
 
 
 PER 
 
 Enzyme used 
 prepare FPC 
 
 to 
 
 Sole source 
 of protein 
 
 2 / 
 Supplement to wheat flour — ' 
 
 2% protein 4% protein 
 
 
 
 % of casein 
 
 
 % of casein 
 
 Pancreatin 
 
 
 82 
 
 
 95 94 
 
 Alcalase 
 
 
 75 
 
 
 92 91 
 
 Autolysis 
 
 
 82 
 
 
 99 95 
 
 1/ 
 
 2/ 
 
 bource: 
 
 ]<jioD± et a±. 
 
 Diets contained 10 percent of which either FPC or casein 
 supplied 2 or 4 percent and the remaining 8 or 6 percent 
 was supplied by wheat flour. 
 
 1.2-16 
 
2.3.2 Utilization of FPC When Fed to Malnourished Animals and Humans 
 
 With few exceptions, studies with normal animals showed that the quality 
 of protein in FPC was high and comparable to casein. Questions were raised 
 regarding the adequacy of fish protein when fed to malnourished animals and 
 humans. 
 
 Several studies have shown that the overall results of animal feeding 
 studies have indicated with few exceptions that FPC, when properly processed 
 and when little or no damage occurs to the protein during processing, was 
 equal to casein in rehabilitating malnourished animals. Other studies in- 
 dicated that FPC prepared in Quintero, Chile, was high in quality and should 
 have been as effective as milk protein in stimulating weight gains and re- 
 generating serum protein in malnourished animals. 
 
 Studies with humans have been more contradictory. Results of several 
 of these were reviewed by Scrimshaw et al., including studies conducted by 
 the Institute of Nutrition for Central America and Panama (INCAP) . 
 
 The studies at INCAP clearly indicated that the FPC had a very satis- 
 factory protein quality for children and that it induced rapid and satis- 
 factory nutritional recovery in malnourished children. Recoveries of serum 
 protein levels and body weights when FPC was used as the sole source of 
 protein were similar to those obtained when milk was used as the sole source 
 of protein. In general, nitrogen absorption and retention values were 
 higher with FPC when compared to results obtained in similar children treated 
 with milk. It was also observed that diarrhea, which is common in mal- 
 nourished cliildren, disappeared rapidly in the course of treatment with FPC, 
 at rates similar to those obtained with casein supplemented with methionine, 
 and at a faster rate than with milk, possibly this was related to the ab- 
 sence of or very low lactose content in FPC and in casein diets compared to 
 milk diets. 
 
 Results from the INCAP study and others indicated that the adequacy of 
 FPC as a protein source for malnourished children depended on whether any 
 damage had occurred during processing. When properly processed, FPC proved 
 to be an adequate source of protein for malnourished, as well as normal 
 children. 
 
 2.3.3 Supplementation of Other Proteins 
 
 FPC was intended to bo used to supplement and enhance the quality of 
 dietary protein rather than as a sole source of dietary protein. Numerous 
 
 1.2-17 
 
studies have been reported in which FPC was evaluated as a supplement to 
 other proteins. Again, it is not feasible to review each study in detail. 
 However, a few general conclusions can be drawn from the information avail- 
 able. 
 
 The addition of FPC to proteins that are naturally lower in quality, 
 produced an increase in overall protein quality. FPC was particularly ef- 
 fective as a supplement to rice, wheat and corn. For example, the addition 
 of about 3.0 to 3.5 percent FPC to either rice, wheat or corn was found to 
 increase the PER by 35, 100 and 300 percent, respectively. Barnett found 
 that the addition of 1, 3, and 5 percent FPC to diets containing corn, 
 wheat, rice, rye, grain sorghum and millet increased growth and PER, with 
 the greatest responses obtained when FPC was added to sorghum, millet and 
 corn. Several other studies showed the enhancement of protein quality 
 through the addition of FPC to proteins of vegetable origin or to food 
 mixtures . 
 
 Few studies, hov/ever, have been designed to determine the optimum level 
 of FPC supplementation. Bressani concliided that a minimum of 3 percent FPC 
 added to a diet based on lime-treated corn produced the maximum PER, al- 
 though 8 percent was needed to produce maximum weight gains by rats. 
 
 Stillings et al. studied diets containing 10 percent protein and reported 
 that 6 percent FPC added to the wheat flour produced maximum weight gains, 
 PER and NPU. A similar optimum level was found when FPC was added to semo- 
 lina and processed into pasta. 
 
 Figure 2-1 shows that maximum weight gains of rats were obtained when 
 10 percent of the flour in bread was replaced by FPC and the bread was 
 added to diets at an 80 percent level. Although 10 percent of the flour was 
 replaced by FPC, the diets contained only a net 7 percent FPC since the diets 
 contained only 80 percent bread, the remaining 20 percent consisted of 
 other standard ingredients. 
 
 The optimum level of FPC supplementation was shown to depend on several 
 factors, especially the combination and amounts of other proteins in the 
 diets. In general, however, the addition of not more than 6 to 8 percent 
 FPC to foods containing protein mainly of vegetable origin appeared justified. 
 
 Some might argue that this level is in excess of the amount needed to 
 meet amino acid requirements. For example, based on a computer study of the 
 amino acids supplied by diets containing vegetable protein, it was concluded 
 
 1.2-18 
 
that more than 3 percent FPC would rarely be needed to meet 200 percent of 
 the FAO reference amino acid pattern. This assumed, however, that the total 
 diet would be supplemented with FPC, rather than individual food items, and 
 it did not account for differences in amino acid availability or processing 
 losses . 
 
 The combination of FPC with other protein supplements and synthetic 
 amino acids was another necessary consideration. It could have been more 
 effective and economical to combine small amounts of FPC with other supple- 
 ments, such as oilseed proteins and lysine, rather than using only one 
 supplement. For example, in Figure 2-1 it is evident that a greater response 
 was obtained by adding FPC to bread than by adding lysine. Jansen, however, 
 showed that a combination of 6 percent FPC and 0.25 percent lysine produced 
 a greater response in PER when added to bread than when only 6 percent FPC 
 was added. When either FPC or lysine were added separately to wheat flour, 
 Hegstead reported that the cost per unit of utilizable protein was less for 
 FPC than for lysine. 
 
 Brief mention should be made of losses in protein quality during process- 
 ing of foods. Adrian and Frange found that during the baking of biscuits 
 the PER of those enriched with peanut flour declined 30 percent, those with 
 FPC declined AO percent and those with skim milk powder declined 100 per- 
 cent; others found somewhat different results. No loss in protein quality 
 was noted during the baking of crackers supplemented with FPC. With bread 
 a 10 to 15 percent loss in protein quality was found. With small loaves of 
 bread, however, Chilean investigators found a considerably larger loss dur- 
 ing processing, presumably because of the higher proportion of crust. proc- 
 essing of pasta supplemented with FPC was shown to have little effect on 
 protein quality. 
 
 From the data available it appeared that a 10 to 20 percent loss in 
 protein quality could be expected during processing of baked products. For 
 pasta-type products the loss was less than 10 percent. 
 
 1.2-19 
 
'^ FPC 
 
 ' ^LYSINE 
 
 O'.l 0.2 
 
 25% FPC 
 
 i 
 
 .0% LYSINE 
 
 FIGURE 2-1 WEIGHT GAINS OF RATS FED DIETS CONTAINING 80 PERCENT BREAD SUPPLE- 
 
 ML-NTED WITH VARYING AMOUI^TS OF EITHER FPC (FISH PROTEIN CONCENTRATE) 
 OR LYSINE 
 
 1.2-20 
 
2.4 TOXICOLOGICAL STUDIES 
 
 In contrast to the voluminous reports on the nutritional quality of FPC 
 and on its use in foods, relatively few studies on the wholesomeness and 
 toxicological safety of FPC have been published. A sununary of these studies 
 is shown in Table 2-10. 
 
 2.4.1 Safety of Deodorized FPC 
 
 One of the first toxicological studies of FPC was conducted by Allison 
 et al , on the investigation of the safety of deodorized FPC prepared in 
 Quintero, Chile. The FPC was prepared from dried hake (Merluccius gayi ) by 
 solvent extraction with heptane and ethanol. The FPC was incorporated into 
 diets at levels ranging from to 25 percent in 5 percent increments, and 
 the diets were fed to male and female weanling rats for periods ranging up 
 to 12 months. Periodically during the 12 months animals were selected from 
 each group, autopsied and their organs examined histopathologically. Over 
 the entire period there was no evidence of any effect of the FPC on blood 
 cytology, body and organ weights, and histopathology of tissues and organs, 
 nor was there any other evidence of toxicity. 
 
 2.4.2 Safety of FPC Extracts 
 
 Yanezet.al, also reported a study in which the toxicological safety ot 
 FPC prepared by extraction of hake with hexane and ethanol was evaluated. 
 The FPC was incorporated into diets at a level of 30 percent o.nd fed to rats 
 for three generations, with no evidence of toxicity observed. Another study 
 involving a new approach to the evaluation of the toxicological safety of 
 foods was reported by Friedman et al. This approach consisted of preparing 
 extracts of isopropyl alcohol-extracted FPC and studying the toxicity of 
 the extracts. The FPC was extracted with solvents of increasing polarity, 
 beginning with hexane, followed by chloroform, then by ethanol and finally 
 by water. Each extraction was expected to concentrate substances of similar 
 solubility properties. The study was designed so that the extracts could be 
 fed at a level equivalent to 10 times that obtainable by feeding 50 percent 
 of the diet as FPC. In this study, the original FPC sample, the extracts 
 and the residue remaining after extraction were each included in diets at 
 a level of 18 percent and were compared to a control diet containing 18 
 percent casein. Diets were fed to mice and rats for a period of four weeks, 
 followed by exliaustive evaluation of tissues, blood, excreta, etc. All 
 
 1.2-21 
 
parameters failed to demonstrate any significant difference between test and 
 control animals, and, in summary, the study failed to uncover any toxic 
 principles in FPC. 
 2.4.3 Conventional Studies 
 
 The studies conducted by Friedman et al.were continued using the con- 
 ventional approach. FPC prepared by isopropyl 'alcohol extraction by the 
 U.S. National Marine Fisheries Service was included in diets at a level of 
 25 percent and fed to rats and mice over five generations. Control groups 
 were fed a diet containing casein or a commercial rat chow. In a very 
 exhaustive evaluation of the results, no evidence of any toxicity from the 
 FPC diet was observed. Of somewhat academic interest, it was observed that 
 after a continuous period of time on a diet in which menadione was provided 
 to meet Vitamin K requirements, many of the rats fed FPC developed widespread 
 hemorrhages characteristic of Vitamin K deficiency. It turned out that this 
 was indeed a deficiency of Vitamin K, and that menadione could not be uti- 
 lized as a source of Vitamin K when the animals were fed FPC. When Vitamin 
 K was substituted for menadione, all animals fed FPC responded normally and 
 no hemorrhages v^ere observed. 
 
 A 12-week feeding study was conducted by Stillings on FPC prepared by 
 isopropyl alcohol extraction of menhaden (Brevoortia tyrannus). Control 
 diets contained similar amounts of protein from casein. Seven male and seven 
 female weanling rats were assigned to each diet. No significant differences 
 were detected between experimental and control animals in respect to growth, 
 organ weights and histopathology of tissues and organs. 
 
 The reports of Morrison and McLaughlan on the lowered nutritional qual- 
 ity and possible toxicological effects of ethylene dichloride FPC were based 
 on the extraction of fish for long periods of time with the ethylene dichlor- 
 ide. Schendel and Johnson studied the effect of feeding, to rats over several 
 generations, FPC prepared by extraction with ethylene dichloride for short 
 periods of time. The FPC was prepared by the VioBin Corporation and was in- 
 corporated into diets at a level of 25 percent. Diets containing the FPC 
 were superior to those containing casein in respect to reproduction and 
 general performance of rats through three matings of the first generation, 
 seven matings of the second generation, and seven matings of the third gen- 
 eration. Histological examination of 12 organs and tissues from animals 
 of the first, second and third generations showed no differences between 
 
 1.2-22 
 
animals receiving FPC and those receiving casein. However, problems were 
 encountered in this study wi tli animals on both the control and the FPC 
 diets with respect to poor reproductive performance. All animals and espe- 
 cially animals consuming the casein diet showed poorer reproductive perfor- 
 mance from the first to the third generation. Thus, although there was no 
 indication of toxicity in rats consuming the FPC diet, the reproductive 
 results of both the control and test diets remained in doubt because of an 
 apparent stress, of unknown origin, that affected all animals. 
 
 Hallgren reported the results of a reproductive study of rats fed FPC 
 prepared by the Astra Company of Sweden. The Astra FPC was prepared pre- 
 sumably by isopropyl alcohol extraction of cooked and pressed herring. The 
 fish protein was incorporated into diets at a level of 20 percent and was 
 compared to a diet containing 20 percent casein. Rats were raised for three 
 generations on diets containing fish protein, or casein for the control 
 groups, as the sole source of protein. No differences were found between 
 the groups consuming fish protein and those consuming casein with respect 
 to the number of pregnant females, number of litters born, number of rats 
 per litter, weight of young when weaned, or the number of dead rats per 
 litter. Again, from these studies there was no evidence of toxicity. 
 
 Jaffe also reported on the results of a toxicological study using 
 Venezuelan whole sardine meal or defatted and deodorized FPC from the United 
 States. Two generations of pregnant rats were fed a diet containing 80 
 percent corn and 20 percent FPC while the control group was fed a standard 
 rat chow diet. Reproduction and growth of litters were good, and histo- 
 logical examination of organs revealed no toxic effect of the FPC. Pre- 
 sumably, the FPC from the U. S. was prepared by the VioBin Company. 
 
 Dreosti reported that no evidence of toxicity was found in studies on 
 growth, breeding performance and his topathology of rats receiving for 199- 
 203 days a diet containing 28 percent FPC prepared in South Africa. Growth 
 and histopathological studies on the second generation of rats raised on 
 the same diets for 87 days also showed no evidence suggestive of toxicity. 
 
 Several types of FPCs prepared from different species of fish were 
 examined for evidence of toxicological properties. In all cases, no evi- 
 dence of toxicity was revealed and the studies indicated that the FPCs 
 examined were indeed safe. 
 
 1.2-23 
 
TABLE 2-10 
 
 SUMMARY OF TOXICOLOGICAL STUDIES ON FPC 
 
 Investigator 
 
 Species 
 of fish 
 
 Solvent used 
 in process 
 
 Method 
 
 Allison et al. Hake 
 
 Yanez et al. Hake 
 
 Friedman et al. Hake 
 
 Newberne (unpublished) Hake 
 
 Schendel and 
 Johnson 
 
 Hallgren 
 
 jaffe 
 
 Dreosti 
 
 Herring 
 
 Sardine 
 meal 
 
 Heptane, 
 ethanol 
 
 Hexane, 
 ethanol 
 
 IPA 
 
 IPA 
 
 Stillings (unpublished) Menhaden IPA 
 
 Ethylene 
 dichloride 
 
 IPA 
 
 Ethylene 
 dichloride 
 
 Fish meal extraction 
 
 1 generation rats 
 
 3 generations rats 
 
 Highly concentrated 
 fractions fed the 
 rats 
 
 5 generations rats 
 and mice 
 
 1 generation rats 
 A generations rats 
 
 3 generations rats 
 
 2 generations rats 
 
 2 generations rats 
 
 1.2-24 
 
2.5 RECOMMENDATIONS FOR FUETHER WORK 
 
 2.5.1 Nutritional Quality 
 
 Little additional work is needed on the nutritional quality. Consider- 
 able evidence exists to demonstrate the nutritional adequacy of FFC prepared 
 by IPA extraction for several species of fish. Thus, only limited effort 
 is indicated in this area. Additional studies might be indicated, for example* 
 when significant changes are made in the process or when new species of fish 
 are processed. 
 
 2.5.2 Safety 
 
 Although numerous studies have demonstrated the safety of specific FPCs, 
 careful consideration should be given to evaluating the toxicological safety of 
 FPC. 
 
 2.5.3 Feeding Studies 
 
 In relation to human feeding studies, emphasis in the future should be 
 given to acceptability and economics of using FPC. In addition, from an 
 effectiveness and economical viewpoint, combinations of FPC and other supple- 
 ments should be more strongly considered. Acceptability problems are dealt 
 witii uiui-'e luj-j-y li'i tuc suminary on product uses. 
 
 1.2-25 
 
SECTION 3 
 PRODUCT USES- 
 
 3.1 CONCLUSIONS AND RECOMMENDATIONS 
 
 Fish protein concentrate can be used as an effective protein fortifier 
 in a wide range of food products — breads, pasta, snacks, soups and bever- 
 ages. Since FPCs can be manufactured by several processes and from various 
 types of fish, there are FPCs of varying odors, flavors and colors. These 
 FPCs should be made available to food manufacturers for use in marketable 
 products. Wliich FPC the user may choose will depend upon the food product 
 and the desired level of fortification. 
 
 Work done at the NMFS laboratories suggests that the optimum fortifica- 
 tion level is 6 to 8 percent of the dry ingredients of a given food product, 
 Tuis level seeius to ruiiiiiuize problems in handling and of changes in formu- 
 lation and flavor. 
 
 More research on the utilization, acceptability and functional qualities 
 of FPC is recommended. Despite its alleged nonfunctionality, FPC made by 
 isopropanol extraction is as easy or easier to incorporate into established 
 food formulations as other protein supplements. Research has indicated 
 that when a particular food item is fortified with comparable amounts of 
 different protein supplements — FPC, dried skim milk and soy flour — the FPC 
 pose<i the fewest problems. 
 
 Most of the utilization work on FPC fortification has been done with 
 the cereals wheat and corn. Research is necessary on other cereals such as 
 oats and rye, as well as on the problem of blending flavorful FPCs with soy 
 to make an acceptable high protein supplement. 
 
 More extensive utilization studies are required to enlarge the number 
 of food products in which FPCs with varying degrees of flavor can be used. 
 Past studies have indicated that FPC will react differently with various 
 ingredients in a particular formulation, producing fishy flavors that are 
 
 L' Much of the work discussed in this summary is covered in greater detail 
 in item C031 of the bibliography, Utilization of FPC and Acceptability . 
 
 1.3-1 
 
unacceptable in the food product (e.g., when FPC was used in a sponge cake 
 recipe, it reacted with the oil of orange flavoring, resulting in a definite- 
 ly fish-flavored cake). 
 
 Another area requiring additional study is the acceptability of FPC as 
 a protein fortifier. Insufficient information is available on whether FPC- 
 enriched foods will be eaten repeatedly by consumers. Although it has been 
 proven that FPC can be incorporated satisfactorily into various foods, 
 there have not been adequate tests of the acceptability of those foods. 
 
 Relative to the question of acceptability and commercial use of FPC in 
 manufactured foods, the problem of overly restrictive regulation of FPC by 
 the Food and Drug Administration (FDA) has been eliminated. On July 24, 1973 
 the FDA removed the 1-pound packaging requirement, thus making it feasible 
 for food manufacturers to use it on a commercial basis. 
 
 3.2 USES AND FORMULATIONS 
 3.2.1 Utilization in General 
 
 Considerable work has been done at the College park laboratory on the 
 utilization of FPC in various food products. Most of the NMFS work was con- 
 centrated on foods that would be considered acceptable to the American 
 consumer. However, the objective of the work was not to develop protein- 
 rich foods per se, but rather to observe how FPC functioned with other in- 
 gredients in established formulations of various food products. problems 
 in acceptability and functionality were considered in the NMFS work on uti- 
 lization and will be covered in the course of this summary. 
 
 From utilization studies conducted by NMFS, it was determined that the 
 optimum FPC fortification level was 10 percent or less of the dry ingredi- 
 ents of a given food product, particularly for baked products and pasta. 
 At this level of fortification unfavorable changes occurring in flavor, 
 texture or appearance can be remedied by minimal modifications of product 
 formulation or FPC processing. 
 
 The utilization of FPC in a given food product depends primarily on 
 the flavor and odor of the FPC itself rather than on the flavor and odor 
 it may impart to the food product it is to fortify. This premise raises 
 the question of whether FPC must be nearly flavorless and odorless in order 
 to be used in all food products. There is little available data on this 
 problem. NMFS conducted the following study to obtain some information in 
 
 1.3-2 
 
this area: six FPCs with varying flavor and odor characteristics were in- 
 corporated into various food products — grits, bread, plain cookies and choco- 
 late chip cookies — at a 10 percent level of fortification. The food products 
 were then judged according to sensory qualities — aroma, flavor by mouth, 
 aftertaste, color, mouth feel and texture. 
 
 'The six FPCs in this' experiment were processed by isopropyl'^extraction 
 of menhaden as follows: 
 
 (1) Deboned fresh fish 
 
 (2) Whole fish with a four-stage extraction 
 
 (3) Whole fish with a five-stage extraction 
 (A) presscake 
 
 (5) Fresh fish meal 
 
 (6) Cured fish meal 
 
 The results of the flavor evaluation are shown in Table 3-1. FPC pro- 
 cessed from hake was used as a control in the test, as hake FPC is consid- 
 ered to possess desirable characteristics. If the characteristics of a 
 test sample were equal to those of the hake-FPC sample, a score of'^3 was 
 given; if its characteristics were less desirable than the hake-FPC sample, it 
 was scored 1 or 2; if the test sample was superior to the hake-FPC sample, 
 it was scored '^ or 5 accordingly. 
 
 Because different milling methods were used, there was a significant 
 difference in particle size between the hake FPC and the menhaden-FPC 
 samples, as was reflected in the test combining FPC with grits. The results 
 of the texture evaluation are shown in Table 3-2. particle size affected 
 the texture of the final food products as follows: 
 
 (1) When used in broad, menhaden FPCs produced a texture comparable 
 to bread made with hake FPC, except for breads made with deboned 
 menhaden FPC and presscake FPC; the latter samples were doughier. 
 
 (2) Cookies made with the menhaden FPCs were comparable to those 
 made with hake FPCs. 
 
 As shown in Table 3-3, the menhaden FPC produced food products darker 
 in color than those fortified with hake FPC The original dark color of the 
 test samples was lightened somewhat when the FPC was incorporated into bread 
 and cookie formulations; the following menhaden FPCs produced cookies most 
 comparable to those fortified with hake FPC: deboned fresh fish, five-stage 
 extracted whole fish and fish meal. No color differences among the various 
 FPCs were apparent when used in chocolate chip cookie*^. 
 
 1.3-3 
 
TABLE 3-1 
 
 FLAVOR OF FOOD PRODUCTS FORTIFIED WITH FPC 
 
 (Rated on a scale from 1 (strongest, or unfavorable) to 5 (mildest, or more favorable), 
 with 3 equal to the standard (or control) 
 
 Grits 
 
 Bread 
 
 Plain 
 cookies 
 
 Chocolate 
 chip cookies 
 
 Hake FPC (control) 
 Menhaden FPC made from: 
 
 3.0 
 
 3.0 
 
 3.0 
 
 3.0 
 
 Iced fish deboned 
 Iced whole fish— 
 
 2.2 
 
 2.6 
 
 2.8 
 
 3.0 
 
 2.2 
 
 2.5 
 
 2,4 
 
 - 
 
 2 / 
 Iced whole fish—' 
 
 3.0 
 
 2.5 
 
 2.9 
 
 2.6 
 
 Fresscake 
 
 1 / 
 
 1 o 
 X. o 
 
 1.8 
 
 2.7 
 
 Fish meal (fresh) 
 
 1.5 
 
 2.1 
 
 2.4 
 
 - 
 
 Fish meal (cured) 
 
 1.0 
 
 2.0 
 
 2.1 
 
 2.7 
 
 2/ 
 
 Four-sta^e countercurrent extraction 
 Five-stage countercurrent extraction 
 
 1.3-4 
 
TABLE 3-2 
 TEXTURE OF FOOD PRODUCTS FORTIFIED WITH FPC 
 
 Grits 
 
 y 
 
 Bread 
 
 2/ 
 
 Plain ^, Chocolate „ . 
 cookies chip cookies 
 
 Hake FPC (control) 
 Menhaden FPC made from: 
 
 3.0 
 
 Iced fish -deboned 
 
 2.2 
 
 4 
 
 Iced whole 
 
 2.3 
 
 Iced whole 
 
 2.6 
 
 Press cake 
 
 2.4 
 
 Fish meal (fresh) 
 
 2.5 
 
 Fish meal (cured) 
 
 2.1 
 
 3.0 
 
 2.3 
 2.9 
 2.9 
 2.3 
 2.9 
 2.6 
 
 3.0 
 
 3.2 
 2.9 
 3.4 
 3.2 
 3.2 
 3.2 
 
 3.0 
 
 3.0 
 
 2.7 
 3.0 
 
 2.6 
 
 - Equal to standard, 3; more desirable than standard, greater than 3; 1 
 
 desirable, less than 3. 
 2/ 
 
 11 
 
 5/ 
 
 Equal to standard, 3; spongier than the standard, more than 3; doughier 
 
 than the standard, less than 3. 
 
 Equal to standard, 3; crispier than standard, greater than 3; softer, 1 
 than 3, 
 
 ess 
 
 Four-stage countercurrent extraction, 
 — Five-stage countercurrent extraction. 
 
 1.3-5 
 
TABLE 3-3 
 
 APPEARANCE OF FOOD PRODUCTS FORTIFIED WITH FPC 
 (Rated on a scale from 1 (darkest) to 5 (lightest) with 3 equal to the standard) 
 
 Raw material 
 
 Grits 
 
 Bread 
 
 Plain 
 cookies 
 
 Chocolate 
 chip cookies 
 
 Hake FPC (control) 
 Menhaden FPC made from: 
 
 3.0 
 
 1/ 
 
 2/ 
 
 3.0 
 
 Four-stage countercurrent extraction 
 
 Five-stage countercurrent extraction 
 
 3.0 
 
 3.0 
 
 Iced fish deboned 
 
 1.4 
 
 1.6 
 
 3.1 
 
 3.7 
 
 Iced whole — 
 
 1.8 
 
 2.2 
 
 2.0 
 
 - 
 
 1 
 Iced whole.£' 
 
 1.8 
 
 2.1 
 
 2.7 
 
 3.3 
 
 Press cake 
 
 1.6 
 
 2.0 
 
 2.0 
 
 3.1 
 
 Fish meal (fresh) 
 
 1.2 
 
 2.1 
 
 2.8 
 
 - 
 
 Fish meal (cured) 
 
 1.1 
 
 1.3 
 
 1.6 
 
 3.5 
 
 1.3-6 
 
3.2.2 Utilization in Baked products 
 
 3.2.2.1 Breads 
 
 Because some form of bread appears in the dietary regimen of almost 
 every society, bread and bread-like products are among the best vehicles for 
 FPC utilization. NMFS conducted studies on the effect on dough characteris- 
 tics of varying levels of fortification with FPC. It was shown that the 
 addition of FPC increased the required amounts of water and the necessary 
 mixing time of the dough, with fortification levels of up to 15 percent. 
 Also, the dough lost its characteristics of springiness and resiliency 
 (i.e., it became "short"). The dough became shorter with the addition of 
 more FPC, reducing its capacity to retain gas during fermentation. The 
 amount of handling that the dough could withstand was reduced as well, re- 
 sulting in a decrease in loaf volume. 
 
 Fat is an important ingredient in bread formulation. It was found that 
 an 8 percent fortification of fish meal FPC could be used instead of a A 
 percent fortification of dried skim milk in a standard bread formulation. 
 NMFS also studied the problem of making a loaf of FPC-fortif ied bread equal 
 in size and weight to that commonly sold on the American market as a high- 
 protein product. A compar;able, but slightly larger, FPC-fortif ied loaf was 
 produced, although the addition of FPC to the formiulation again caused the 
 dough to become short and therefore delicate to handle. A similar bread 
 formulation was developed at Kansas State University with comparable results. 
 
 3.2.2.2 Chapatis, Tortillas and Biscuits 
 
 Chapatis are an unleavened bread product commonly used in India. A study 
 conducted outside of NMFS investigated the utilization of FPC in chapatis at 
 fortification levels of 5 and 10 percent. Those with 5 percent FPC were re- 
 ported acceptable, with no discernible texture or flavor changes. Chapatis 
 with 10 percent FPC fortification were judged to be noticeably fishy in 
 flavor, although not abnormal in texture. 
 
 The tortilla, made from ground lime-treated corn, is a form of bread 
 widely used in Central America and Mexico. NMFS conducted a study on tor- 
 tillas with FPC fortification levels of 5, 10 and 15 percent. The color of 
 the tortilla became noticeably darker as the level of FPC increased; texture 
 and flavor were not affected to a noticeable degree. 
 
 Studies have shown biscuits to be another product to which FPC can be 
 successfully added, . however, most of the work with this medium has been con- 
 ducted abroad. 
 
 1.3-7 
 
3.2.3 utilization in pasta 
 
 Because it is consumed b}' a large proportion of people in various so- 
 cieties, pasta may be considered another excellent vehicle for FPC Tests 
 were conducted at NMFS to determine the optimum levels of FPC fortification 
 for macaroni made from hake FPC and semolina. FPC was incorporated at levels 
 of 0, 3, 6, 9 and 12 percent; water was added in sufficient quantity to keep 
 the doughs free flowing yet cohesive under pressure, the mixture was extruded 
 and the resulting pasta were dried overnight. The dried pasta samples became 
 darker as the FPC content increased, although cooking washed out much of the 
 darkness. No discernible differences in odor appeared in pasta with 3 percent 
 FPC; there was a definite odor in the 12 percent sample. Similarly, with 
 regard to flavor, the 12 percent sample had a noticeable fishy taste; the 0, 
 3 and 6 percent samples were judged acceptable. 
 
 Another study on FPC-fortif ied pasta showed that noodles fortified with 
 10 and 15 percent FPC had nutritive values greater than those fortified with 
 casein, resulting in PERs of 3.35, 3.83 and 3.21, respectively. 
 
 3.2.4 Utilization in Snacks 
 
 Snacks are a widely consumed type of food product in this country and 
 as such, could be good basic products for FPC fortification. 
 
 3.2.4.1 Crackers 
 
 In cooperation with a commercial food manufacturer, NMFS developed a 
 formulation for FPC-fortif ied saltines. One experiment was conducted to as- 
 sess the effect of processing on the nutritive value of a cracker fortified 
 with 10 percent of FPC in place of flour. The baking process had no effect 
 on nutritive value. A second experiment was conducted to observe the physi- 
 cal and sensory characteristics as well as the nutritive values of crackers 
 fortified with 0, 4, 8, 12 and 16 percent fortification of FPC. The 4 per- 
 cent supplementation almost tripled the nutritive values of the crackers; 
 the 8 percent samples' had almost four times the nutritive value of the 
 percent sample. There were no discernible textural changes in the crackers 
 until the fortification level reached 16 percent. The same was true for 
 flavor. No rancidity or other storage changes were evident in the 0, 4 and 
 8 percent samples for storage periods of 161 days at temperatures up to 
 1260f. 
 
 3.2.4.2 Doughnuts 
 
 Studies were conducted on protein-fortified doughnuts incorporating 
 
 1.3-8 
 
various protein supplements, including FPCs from whole fish and filleted 
 fish. The FFC-fortif ied products were reported to be* very acceptable in 
 comparison to those fortified wi-th the other supplements. 
 
 At the NMFS laboratory FPC was added to yeast-raised doughnut formula- 
 tions with the following results: increasing increments of I^PC interfered 
 with the delicate texture, making the final product more breadlike in tex- 
 ture; fat absorption was consistently low; flavor and appearance were quite 
 acceptable. Extrusion was attempted on doughnuts with a 10 percent FPC 
 fortification with good results. 
 3.2.4.3 Corn Snacks 
 
 A mixture of cornmeal and FPC was cooked under pressure and extruded 
 to produce a puffed, curl-type snack food. The physical and sensory 
 characteristics of the fortified product were comparable to those of the 
 unfortified product with the exception of a change in color evident in the 
 FPC curls. 
 3.2.5 Utilization in Beverages and Soups 
 
 A A percent fortification of FPC was successfully used in a beverage 
 formulation comparable in composition to cow's milk. This beverage formu- 
 lation could be enriched with the necessary vitamins and minerals to produce 
 a completely nutritional food. Since it is quite bland, one potential use 
 of this product would be infant feeding. 
 
 Another successful utilization of FPC is in soups and stews. The FPC 
 can be complemented by spices or flavorings or combined with legumes or 
 other vegetables at fortification levels up to 10 percent. 
 
 3.3 ACCEPTABILITY 
 
 3.3.1 Consumer Acceptability 
 
 There is only a limited amount of information extant on the acceptability 
 of FPC as a food supplement. Clearly this is an area requiring further and 
 more extensive study to provide valid conclusions on how well FPC-fortif ied 
 food products will be accepted into the diet regimens of large- groups of the 
 world's population. 
 
 The work at NMFS concentrated more on the physical characteristics of 
 FPC-fortif ied products--odor , flavor, appearance, etc. — rather than on broad- 
 based consumer acceptance trials. Most of the work in the latter area has 
 been done abroad; however, even these studies are not conclusive as to the ad- 
 vantages or disadvantages of using FPC in a wide range of foods. 
 
 1.3-9 
 
As was indicated earlier in this summary, FPCs have been utilized suc- 
 cessfully in a number of foodstuffs with respect to developing a final prod- 
 uct in which the supplement of FPC was either not discernible or not unfavor- 
 ably so. Now these products should be tested for consumer acceptability and 
 economic feasibility of manufacture on a commercial scale. After all, even 
 the tastiest, most nutritious food will be useless if people do not choose 
 to eat it, and people cannot eat something which manufacturers find too cost- 
 ly or too complicated to make. 
 
 Most of the principal food products mentioned in the discussion of uti- 
 lization have been given limited tests of consumer acceptability. FPC-for- 
 tified bread has been investigated in Chile (the results of these studies 
 remain to be published at this writing). An extensive study on the consumer 
 acceptability of FPC-fortif ied bread was conducted for some three years in 
 South Africa. There was no indication from the public that the bread was un- 
 acceptable, although it can be argued that the study stopped too soon to be 
 really valid. 
 
 Limited consumer acceptability studies have been conducted for tortillas, 
 biscuits, pasta and soups. These studies were done in several countries 
 among various strata of society. While by no means conclusive, the studies 
 did bring forth one interesting, perhaps quite significant, point; to direct 
 the development of FPC-fortif ied foods toward the lower class segments of a 
 population may not be successful. Instead, researchers might ensure better 
 acceptability of FPC by incorporating it into foods favored by the higher 
 classes. Making a "prestige product" with FPC may enhance its acceptability 
 to all levels of a society. Nevertheless, the products must be cheap enough 
 to be afforded by those poorer people who desire to emulate their financial 
 betters. Jt should be remembered that one of the major benefits of FPC is 
 its use as a potential protein source for those unable to afford meat. 
 
 NMFS became involved in plans for a consumer acceptability test in 
 American Samoa. The NMFS laboratory developed formulations for a cookie 
 weighing 30 grams of which 2 grams were FPC The plan called for these 
 cookies to be tested in a school lunch program; the FPC was to be provided 
 by NMFS. This program was never implemented, however, due to financial and 
 personnel problems. A similar project has been slated for Jamaica. 
 3.3.2 Acceptability and the FDA 
 
 A major factor in FPC acceptability or even testing such acceptability 
 
 1.3-10 
 
in the United States has been the overly restrictive attitude of the Food 
 and Drug Administration (FDA) toward the product. Virtually since the begin- 
 ning of NMFS participation in FPC development, actions of the FDA have se- 
 verely limited the potential utilization and testing of the product. 
 
 To briefly summarize, from the beginning the FDA has strongly questioned 
 the safety and acceptability of FPC with regard to some of its components 
 (possible contaminants), manufacturing processes and characteristics. Thus, 
 FDA regulations have curtailed the widespread use of FPC, particularly on a 
 commercial scale. The outstanding example of FDA's conservative attitude was 
 its regulation limiting the packaging of FPC to units of 1 pound or less. 
 
 The Gonrimissioner of Food and Drugs took under consideration the recom- 
 mendations of the National Academy of Science to make FPC available to con- 
 sumers (FED. REG., V. 32 (22); 2-2-67) and found that: 
 
 "Individual consumers are entitled to the opportunity of choice 
 in the matter of whether they desire to include this item in 
 their diets; therefore the product, to be readily distinguish- 
 able, should be more properly identified as 'whole fish protein 
 concentrate' and for dom.estic distribution, should be packaged 
 for household use in consumer sized units not exceeding 1 pound 
 net weight. Bulk sales of packaging in units larger than 1 
 pound can not be authorized unless the Comimissioner is supplied 
 with data demonstrating that the proposed use of the article 
 will not promote deception and will promote honesty and fair 
 dealing in the interest of the dealer." 
 Six years later this restriction was removed (FED. REG., V. 36 (141); 
 7-24-73); and FPC can now be used in commercially processed products, 
 but the final product can not exceed 8 ppm of fluoride on a dry weight basis. 
 Obviously, this restriction precluded the commercial use of the product un- 
 til now and the enthusiasm of private enterprise has waned. It is unkno^m 
 whether the past widespread interest in the use of FPC can or will be 
 revitalized. 
 
 3.4 FUNCTIONALITY 
 
 The lack of functional qualities of FPC compared to other protein supple- 
 ments has been considered a major problem area. practically no data are 
 
 available, however, giving a direct comparison of the final food products 
 
 I. 3-11 
 
containing equal amounts of high protein supplements such as FPC, soy and 
 dried skim milk. 
 
 An experiment was conducted at NMFS adding varying amounts (0, 3, 6, »» 
 9, 12 and 15 percent) of FPC, soy flour or dried skim milk to a bread 
 formulation. The doughs were prepared under the same conditions in all 
 respects except for the amount of water. The doughs that contained higher 
 levels of FPC were less sticky and easier to handle than the doughs with 
 comparable amounts of soy flour or dried skim milk. 
 
 Three percent supplementation of soy or dried skim milk had a marked 
 effect on the loaf volume, a decrease of 9 percent and 8 percent, respec- 
 tively, against 2 percent for FPC. Nine percent FPC can be added before the 
 reduction in loaf volume is comparable (7 percent) to the volume reduction of 
 the loaf containing 3 percent soy or dried skim milk. At the 9 percent level 
 of supplementation the loaf volume with FPC was significantly larger than 
 those with the same percent of soy or dried skim milk. Some of the volume of 
 the loaf of bread can be restored by including a surfactant or a glycolipid 
 in the formulation as shown by workers at Kansas State University. 
 
 In another study, 3 and 6 percent protein from either FPC, soy or dried 
 
 this study was to compare the nutritive values and the sensory and physical 
 characteristics of pasta supplemented with various types of protein. The nutri- 
 tive value of the pasta was determined on the cooked as well as the uncooked 
 pasta. The PER value improved slightly after cooking for each experimental 
 batch except for those supplemented with dried skim milk and lysine. For those 
 the PERs decreased significantly, upon cooking, particularly the pasta fortified 
 with lysine. The PER values were more than doubled when 6 percent FPC protein 
 was added to the semolina. The dimension of improvement was nearly the same 
 for the pasta with dried skim milk. The soy pasta did not show as great an 
 improvement as the pasta with the other two protein supplements. There was a 
 slight improvement over unfortified pasta for the lysine-fortif led pasta, but 
 it was quite small in comparison to the other protein supplements, probably 
 due to the fact that the lysine leaches out upon cooking. 
 
 When 3,5 or 7.0 percent FPC was added to the pasta formulation, the 
 amount of material that leached into the cooking water remained practically 
 the same as it was for unfortified pasta containing only semolina. There 
 was a marked increase in leaching for pasta supplemented with 8.4 or 16.8 
 
 1.3-12 
 
percent dried skim milk. Some of this increase might have been due to the 
 amount used, as well as the high solubility of the supplement. The leaching 
 from the soy-fortified pasta was slightly greater than in unfortified pasta* 
 The solids leached from the pasta enriched v;ith lysine were no greater than 
 for unfortified samples . 
 
 The amount of protein that leached into the cooking water was least for 
 the pasta fortified with FPC ; the largest amount was for the pasta fortified 
 with dried skim milk or lysine. The amount of protein that leached out 
 increased with cooking time; the largest amount being lost at the 10-minute 
 cooking time. After 10 minutes leaching seemed to level off. 
 
 One investigator has reported that the lack of binding quality in FPC 
 has limited its use in pasta products. He found that 20 to 30 percent of 
 the FPC in the fortified pasta leached into the cooking water, although by 
 changing the cooking procedure he was able to reduce the losses to 5 percent, 
 A study conducted at NMFS showed that the binding quality of the semolina 
 was not visibly modified by the addition of 10 percent FPC or less. 
 
 The addition of the protein supplements has a definite bearing on the 
 texture of the cooked pasta. The addition of dried skim milk caused the 
 cooking time for pasta to shorten and the soy supplementation caused cooking 
 time to lengthen, based on an approximate standard time of 10 minutes. The 
 cooked characteristics of pasta with FPC were comparable to those of the un- 
 fortified samples. FPC-fortif led pasta was grayish and soy-fortified pasta 
 was a bright yellow. Textural differences between the milk-fortified pasta 
 and unfortified samples were not discernible, but hardness of the soy-for- 
 tified pasta was evident. The flavor of the FPC-fortif led pasta was compar- 
 able to that of unfortified pasta. pasta with 6 percent protein from dried 
 skim milk was unacceptable, as was pasta with 3 percent protein from soy. 
 
 1.3-13 
 
SECTION 4 
 
 industrial/economic aspects 
 4.1 conclusions and recommendations 
 
 The key questions to be asked regarding any new product are: (1) Is it 
 econorp.ically feasible to produce? and (2) Is there a market for the product? 
 The research and development of FPC followed the normal path of most complex 
 research projects, i.e., extensive laboratory research followed by a semi- 
 works-^scale plant operation to test laboratory techniques on a much larger 
 scale. The government's FPC Experiment and Demonstration plant (EDP) at 
 Aberdeen, Washington, was expected to provide considerable data on the eco- 
 nomics of production and provide the necessary quantities of FPC needed for 
 testing the market potential of the product. The closing of the plant after 
 bcirely one yecir of opetdLlua uiediiL LhaL the economics of px.ouuuLion wei-t; 
 never clearly established. However, through the years, considerable economic 
 and marketing data was accumulated on FPC, primarily through several contract 
 studies. This summary discusses the available economic and marketing data 
 and points out the areas where deficiencies continue to exist. 
 
 FPC has as yet achieved little commercial success. The potentially 
 high cost of food-grade FPC has not been matched by correspondingly attrac- 
 tive properties which will persuade the consumer to buy, the consumer in 
 this case being the food manufacturer. Moreover, there is difficulty in 
 introducing any new foods or food supplements into diets in developing 
 countries . 
 
 Many of the basic problems of FPC production have been solved. The 
 problems of marketing the product at a price acceptable to consumers and in 
 sufficient quantity to support a commercial operation have never received 
 serious study. 
 
 Relatively little was done in the field of market research and consid- 
 erable efforts will be required to develop markets for the product and to 
 establish consumer acceptance. 
 
 1.4-1 
 
4.1.1 Conclusions With Respect to FPC in the Developing Country » 
 
 (1) The first requirement for beneficial protein supplementation in a 
 developing country is that the population is receiving sufficient 
 calories to meet its energy needs. If not, part of the added protein 
 will be converted to energy rather than used in tissue generation; 
 this energy could be supplied more cheaply by other means. 
 
 L 
 
 (2) The weaning and recently weaned lower income child who is receiving 
 adequate calories is the major target group for protein supplementation. 
 
 (3) The estimated present (1970) opportunity cost of IPA-FPC to Chile is 
 25 cents per pound. 
 
 (A) In most cases, FPC was 25 to ^0 percent more expensive than the least 
 expensive alternative in Chile. 
 
 (5) In those countries which must import protein sources other than fish, 
 the opportunity cost of FPC must be less than 20 cents per pound be- 
 fore FPC can be expected to offer any substantial economic advantage 
 over imported alternatives. 
 
 (6) FPC will not be an economically efficient source of protein on the world 
 market at opportunity costs much above 15 cents per pound. 
 
 (7) The refinement of the IPA extraction process is '.unlikely to reduce 
 costs below 20 cents per pound even where the cost of the raw fish 
 is only 1 cent per pound. 
 
 The world's protein requirements are rapidly increasing; this shift in 
 demand may make FPC competitive in the future, even with no further decrease 
 in cost or increase in functionality. There is a substantial probability that 
 the cost of FPC can be reduced, and/or its functionality increased given a 
 proper research program. Even a moderate reduction in the price of FPC could 
 make a substantial change in its attractiveness, especially in a corn-eating 
 country. Furthermore, there are processing' alternatives which promise sub- 
 stantially improved functional qualities. 
 
 Some question arises as to whether further refinement of solvent extrac- 
 tion would result in reduced cost or improved functionality. A lower quality 
 FPC and other processes or a combination of these processes with solvent ex- 
 traction should be investigated. 
 A. 1.2 Conclusions Regarding FPC in the United States 
 
 (1) No reason exists to believe that groundfish, such as hake, can be landed 
 re-gularly at less than 3 cents per pound on the East Coast. Herring 
 
 1.4-2 
 
and other oily fish might be landed on the East Coast at 1 to 1.5 cents 
 per pound, but the supply would be extremely irregular. The costs of 
 obtaining hake or anchovy on the West Coast may go as low as 1 cent per 
 pound, assuming an efficient fleet. 
 
 (2) production costs of FPC, using hake, on the East Coast will be 36 cents 
 per pound. The cost based on any available fish on the West Coast may 
 be as low as 20 cents per pound. 
 
 (3) Due to the randomness of the catch in both species and sizes, at least 
 on the East Coast, any profitable FPC operation will have to be inte- 
 grated into a fish processing complex. 
 
 (4) There has been no complete analysis of the variability of supply on 
 the cost of producing FPC 
 
 From NMFS work, it is impossible to make any firm conclusion about the 
 present domestic marketability of FPC. Estimates of the cost of FPC are ex- 
 tremely sensitive to the costs of the raw material, which exhibit consider- 
 able variation. Any prediction about how soon a domestic FPC industry will 
 become economically viable would be at best an educated guess. 
 
 Conclusions about the future role of FPC in the U.S. food industry are 
 
 i-c*.!. i-i-wiu v-j_v:;ci.i.. \jiL\^ i. X. ^ kj. j. y i- y i ^ f i: u t\ iicxu ^xu.v^cva cL (a*::: i-ci^^uU ucxii v-/ii Cue uoc Kj i. 
 
 FPC for human consumption in the United States. 
 
 One must also consider the economic environment and function of FPC and 
 its substitutes, both domestically and in developing countries. FPC, by its 
 very nature, competes in an economic environment in which market processess 
 either have often been judged unsatisfactory or have often been interfered 
 with on a massive scale. The problem of Federal investment in FPC cannot 
 be divorced from social and political judgments which in turn cannot be 
 valued by market processes. Given the present state of the art, these 
 judgments are difficult or impossible to quantitatively value by any means. 
 
 The NMFS view is that a considerable reduction in costs will be re- 
 quired if FPC is to have a large role in world nutrition. In many cases, 
 the differential in net costs of supplementation between FPC and the least 
 expensive alternative was on the order of 50 percent or more. The differ- 
 ential between FPC and the cheapest natural protein source was greater than 
 25 percent. Even in fish-rich, currency-poor countries with very few alter- 
 native protein sources, FPC is unlikely to be the least expensive alternative 
 
 1.4-3 
 
at costs in excess of 20 cents per pound. FPC appears to have no properties 
 other than nutritional which could mitigate this cost differential. 
 
 The fact that FPC is either too expensive or presently lacks functional 
 qualities is a poor argument for not undertaking a further research and de- 
 velopment effort. The respective potential positions of FPC and its competi- 
 tive processed protein forms are rapidly changing in favor of FPC. Recent 
 increasing prices and shortages of supply of soy and casein have substantially 
 lowered the risks of FPC and increased the probability of reasonable profit 
 margins and short-term return on capital. Since the economics studies spon- 
 sored by NMFS were conducted before these changes, the evolving situation 
 calls for a continuing look at the economics of FPC. Perhaps less conserva- 
 tive thinking in the forecasting of the overall future protein situation is 
 also necessary. 
 
 4.2 OBJECTIVES AND APPROACH 
 
 4.2.1 Objectives 
 
 The economics and marketing plans were directed toward determining, with 
 a good level of confidence, the economic feasibility of FPC and, to a limited 
 degree, demonstrating FPC market potential. projects were designed to meet 
 the following basic needs: 
 
 (1) Determination of the cost and availability of raw fish for 
 FPC processing. 
 
 (2) Collection of valid FPC cost information and a true picture 
 of processing costs under probable, commercial operating 
 conditions and process variables. 
 
 (3) Determination, in as much depth as practical, of the market 
 potential for FPC and its by-products, including an assess- 
 ment of the market places and price/volume relationship. 
 
 (4) Consolidation into one concise information package of the 
 results of all economics and marketing efforts and their 
 relationship to the commercial feasibility of FPC. 
 
 4.2.2 Approach 
 
 The total economics and marketing effort was broken do\sm into four parts 
 as follows: 
 
 (1) Resource availability and cost 
 
 (2) Economics of FPC processing 
 
 1.4-4 
 
(3) FPC market potential 
 
 (A) Summary report of commercial feasibility 
 For the resource availability and FPC market potential projects, the general 
 approach was to examine existing information, utilize it to the extent possi- 
 ble through extrapolation, restructuring and synthesis of data when valid, 
 and to expand upon that data base through surveys and research as required. 
 
 The practicality of market testing was to be assessed, and, if the find- 
 ings were positive, a test market plan was to be developed. 
 
 The Engineering Economic Model was to be the focal point for the collec- 
 tion and synthesis of processing economics data and was to be continually 
 modified as necessary to meet the objective. 
 
 4.3 ACCOMPLISHMENTS 
 
 From the objectives stated in subsection A. 2, significant achievements 
 were noted in the first three (resource availability and cost; economics of 
 FPC processing; and FPC market potential). Termination of active effort by 
 NMFS to develop FPC precluded valid conclusions regarding FPC market potential. 
 4.3.1 Resource Availability and Cost 
 
 There are major regional differences in the availability of raw materials 
 for U.S. fisheries development. The Northwest Atlantic and the Gulf of Mexico, 
 with their huge continental shelves, offer the greatest opportunities for in- 
 creasing landings of skates, dogfish, sharks, saury, capelin, argentines, 
 hakes, grenadiers, lantern fishes, etc. The pacific Ocean, with its narrow 
 continental shelf, does not approach the Atlantic in productive potential. 
 Nevertheless, dogfish, capelin, saury, hake, pollock, turbot, sand launce, 
 rockfish, anchovy, grenadiers and lantern fishes offer opportunities for 
 expanding fisheries. 
 
 Although an abundance of raw material is available, selection of resources 
 will be necessary in order to keep costs of production to a minimum. Inshore 
 species now landed for reduction to fish meal would appear to be the cheapest 
 source of raw material for FPC production. On offshore banks, retention of 
 trash fish caught by trawlers could add considerable raw material for FPC 
 production. 
 
 It is advisable to select a schooling type of fish such as the hakes or 
 herrings as raw material for FPC. Such schools of fish generally have almost 
 no other species associated with them, thus assuring a homogenous raw material. 
 
 1. 4-5 
 
To insure optimum quality of the f'inished product, the raw material must be 
 as fresh as possible and indeed be of food grade quality. It would be prefer- 
 able to utilize only fish caught in deep waters, as far removed as possible 
 from the shore where contamination is most likely to occur. The source fish 
 must be of human food quality at all stages. The schooling fish that can be 
 caught in volumes likely to keep cost per ton of production ^^fithin reason are 
 small and fragile, with very active enzymes, a lot of polyunsaturated body 
 oils and with natural bacteria which multiply vigorously in the vessel holds 
 crowded with small, broken fish. 
 
 These problems are not insuperable. Chilling at sea, pumping, use of 
 anti-oxidents , moving part of the manufacturing process onto the vessel at 
 least to the point of stabilizing the raw material, ■ are possible means of 
 improving quality of the raw material. There are two problems — the cost and 
 the fishermen's philosophy. It may well be that the latter cannot be accom- 
 modated in the United States system of separate ownership and management for 
 fishing vessel and processing plant. A fully integrated sea/land operation 
 may be called for. 
 
 If a fleet of vessels were required, this would be the most costly part 
 of the whole operation. A new trawler capable of harvesting 6 million pounds 
 of fish per year cost about $1 million in Canada in 1968. Under these cir- 
 cumstances it is difficult to see the landed cost of fish at less than 3 
 cents per pound on the East Coast of the United States. 
 
 Several raw material costs have been estimated from 1 to 2 cents per 
 pound. Unless the desired fish can be obtained in adequate volume as a 
 by-product from other fishing operations, or large harvesting economics can 
 be attained, it is difficult to see how these low harvesting costs could be 
 achieved. This is the single most important cost item in the final cost of 
 FPC because of the relatively low yield achieved in processing. 
 4.3.2 Economics of FPC Processing 
 
 There have been at least six non-proprietary studies of the cost of 
 producing FPC in developing countries. 
 
 (1) Bureau of Coiiimercial Fisheries 1966 study of a proposed pilot plant. 
 
 (2) A Canadian study of Resources Engineering in 1967. 
 
 (3) Southwest Engineering Comipany, Inc. study in 1969. ■ 
 
 (4) A study by a Montreal consulting firm (Surigor, Nenniger and 
 Chenewort) in 1968 of the potential of FPC plants in eastern Quebec. 
 
 1.4-6 
 
(5) A study by Dr. J. W. Devanney eC al., MIT, 1970, 
 
 (6) A study by professor James Crutchfield, University of Washington, 
 in 1972. 
 
 4.3.2.1 Source Material 
 
 Since it takes about 6.6 pounds of whole fish to produce 1 pound of FPC, 
 raw material costs are the most important variable in the production of FPC. 
 Following are MIT's estimates of cost with varying raw fish prices. 
 
 Cents per pound 
 Cost of fish 12 3 4 
 Cost of FPC 23 29 36 43 
 
 Three cents per pound would be the absolute minimum price for New England 
 trawl-caught hake, thus making the price of FPC about 36 cents per pound. In 
 order to ensure a regular supply, a company would have to be willing to pay 
 the bulk of the costs of supporting a fishing fleet rather than the opportunity 
 cost of a fisherman who happens across a school of hake. 
 
 Hake on the West Coast will be available at an anticipated $22 per ton, 
 and a considerably more regular catch can be expected than on the East Coast. 
 Thus, the cost of West Coast FPC from hake could go as low as 20 cents per 
 pound, assuming a 300-day' operation at capacity. At 240 days it would cost 
 'about 24 cents. 
 
 New England fish meal producers pay about 1 cent per pound for oily fish 
 such as herring, mackerel and capelin. However, this penny-a-pound fish 
 usually arrives at dockside in rather poor condition. In order to make FPC, 
 considerably more care in handling, shorter trips and more careful sanita- 
 tion would be required. This will add substantially to the cost of the fish. 
 A cent and a half per pound would bo about the lowest price. 
 
 The University of Washington study estimated that California anchovy 
 can be landed for $16 per ton. A penny per pound for West Coast pellagics 
 seems a reasonable estimate; this would lead to an FPC cost of about 20 
 cents per pound. Thus, it appears that a West Coast-based FPC industry 
 will have substantial advantages over an East Coast operation. The Gulf of 
 Mexico was not studied. 
 
 A crucial point missing from all analyses is that they do not take into 
 account the fact that, in general, the raw material will not be available on 
 a steady basis throughout the fishing year. There will be storage costs when 
 several days catch of raw material is received in a few hours. There will be 
 
 1.4-7 
 
intermittent periods when the plant is unable to operate at full capacity. 
 A conservative estimate is that costs due to variability in the catch will 
 add at least 20 percent to East Coast figures. 
 
 Although the literature would indicate that commercial quantities of 
 FPC would be relatively cheap to produce, little has been published that 
 would- help a prospective manufactured make a decision to build a "plant. 
 There have been few definitive estimates of plant, construction costs, dir- 
 ect and indirect operating costs for production rates geared to different 
 fishing seasons, and other data needed for calculation of return on invest- 
 ment. 
 A. 3. 2. 2 production Plant Cost 
 
 SWECO, Inc., through a joint venture company. Ocean Harvesters, Inc., 
 designed and built the EDP in Aberdeen, Washington. SWECO had also designed, 
 constructed and operated a pilot-scale FPC plant at Texas A&M University 
 since 1967. Using the: single-solvent process developed by the National Marine 
 Fisheries Service, the plant is capable of producing 120 to 150 pounds of 
 FPC per day. 
 
 SWECO has estimated plant construction costs for commercial-sized FPC 
 plants with one, two and three parallel production lines, each capable of 
 processing 200 tons per day of raw fish. Table 4-1 lists 1970 cost estimates 
 for each major item, for the three plant sizes. Starting with the basic cost 
 of $2,500,000 for a 200-ton per day line, parallel processing lists at an 
 additional incremental cost of $1,000,000, Thus, a 400-ton per day plant 
 would cost $3,500,000 and a 600-ton per day plant, $4,500,000. SWECO, Inc., 
 estimated manufacturing costs for 200, 400 and 600 tons of raw fish per day 
 based on prices of $20, $30 and $40 per ton for raw fish, 
 
 A factor that SWECO considered second in importance to the basic cost 
 of raw fish was the length of the fishing season. Four months (150 days) 
 of plant operation were used because the average fishing season for most 
 species of fish ranges between four and six months. Utilities are another 
 major cost in the manufacture of FPC, with the principal cost being for the 
 steam required in the solvent-recovery process. 
 
 Because of the economics of large-scale operation, the total manufac- 
 turing costs, exclusive of raw material cost, for a 600-ton per day plant 
 are only twice the manufacturing cost for a 200-ton per day plant. 
 
 1.4-8 
 
 / 
 
TABLE 4-1 
 
 Construction Costs for 200,400, and 600 ton/day FPC plants 
 
 Plant Capacity 
 
 200-TPD 400-TPO 60Q-TPD 
 
 land: 
 
 Surrtyj. Fees, Property Cost 
 
 $ 80.000 
 
 S 105,000 
 
 $ 130.000 
 
 SITE development: 
 
 CViriog, Oriding, Roads Railroads, Fence, Picrj, Lijidicapiog 
 
 85.000 
 
 89.000 
 
 92.000 
 
 PROCESS buildings: 
 
 St^irvayj, Koisjs, ElevatorJ 
 
 104.000 
 
 115.000 
 
 125.000 
 
 AUXILIARY buildings: 
 
 Adminisuauon. Ntaiotenince, Comrol Laboratory, Locker Room 
 
 61.000 
 
 61,000 
 
 61,000 
 
 BUILDING services: 
 
 Plombln;. Heating, Air Conditioning, Sprinkler S)stems, Building Ughung, Telephone, Painting 
 
 45.000 
 
 50,000 
 
 55,000 
 
 PROCESS equipment: 
 
 Ajper Row Sheet 
 
 1,200,000 
 
 2.000,000 
 
 2.800,000 
 
 NON-PROCESS EQUIP.VtENT: 
 
 Office t'urmture. Laboratory Equipment. Maintenance Equipment, Safety and pirc i*rotectioo Equipment 
 
 80.000 
 
 85.000 
 
 50,000 
 
 PROCESS APPURTENANCES: 
 
 taping, InsiUauoD, Instrumentation, Electrical Substations 
 
 utilities: 
 
 Boiler, Refrigtration Plant, Air Compressor, Cooling Towers, Waste Treatment Facilitiej 
 
 100.000 130.000 160.000 
 
 110,000 175,000 230,000 
 
 YARD DISTRIBUTION FACILITIES: 
 
 Process Pipe Lines, Electric Lines. Produa Loading Statiotu, Raw Material 
 
 Receiving, Fuel and Solvent Storage 
 
 Plant Stnrt-Up Expense (60 Days) - 
 EsliniatctJ Cost of Fishing PIcct — 
 
 100.000 110.000 120,000 
 
 MISCELLANEOUS: 
 
 Solvent Inventory, Spare Pans, Freight, Tales and Insurance 
 
 70.000 
 
 80,000 
 
 90.000 
 
 ENGINEERING COSTS: 
 
 Admirusirative, Process, Project and General Engineering 
 
 75,000 
 
 80.000 
 
 85.000 
 
 CONSTRUCTION- EXPENSE: 
 
 Construction Supcoision, Labor, Tools and Equipment 
 
 390.000 
 
 420,000 
 
 450,000 
 
 TOTAL CAPITAL COST- 
 
 $2,500,000 
 
 $3,500,000 
 
 $4,500,000 
 
 
 i 400,000 $ 500.000 S 600,000 
 $2,500,000 $3,000,000 $3,500,000 
 
 1.4-9 
 
TABLE 4-1 (Cont'd) 
 
 200 Ton/Day Raw Fish (150 Day Operation) 
 
 DIRECT PRODUCTIO:i COSTS: 
 Labor, Utilities, Operating Supplies, 
 Kalntenance, Plant Office Expense, Di- 
 rect Payroll Expense and Packaging Cost 
 
 INDIRECT PRODUCTIO.'I COSTS: 
 Supervision, Control Laboratory, I:. di- 
 rect Payroll Expense, Interest on Worli- 
 Ing Capital, General Plant Overhead 
 
 FIXED PRODUCTION COSTS: 
 Depreciation (10 years). Taxes and In- 
 surance 
 
 MANUFACTURING (PROCESSING) COST:^ 
 
 RAH MATERIAL COST: 
 
 TOTAL MANUFACTURING COST:* 
 
 Coit.of Raw Fish 
 
 $20/Ton 
 
 $30 /Ton 
 
 J40/Ton 
 
 J/lb. 
 Total $ FPC' 
 
 Total $ 
 
 J/lb. 
 FPC' 
 
 Total $ 
 
 147,720 .0164 
 
 325,020 .0361 
 
 1,017.360 .1130 
 
 600.000 .0667 
 
 156.120 .0173 
 
 325,020 .0361 
 
 $1,617,360 $.1797 
 
 $/l>. 
 FPC' 
 
 $ 544.620 $.0605 $ 544.620 $.0605 $ 544.620 $.0605 
 
 164.520 .0183 
 
 325.020 .0361 
 
 1,025.760 .1139 1.034,160 .1149 
 
 900",000 .1000 1,200.000 .1333 
 
 $1,925,760 $.2130 $2,234,160 $.2482 
 
 400 Ton/Day Raw Fish (150 Day Operation) 
 
 DIRECT PRODUCTION COSTS:' 
 Labor, Utilities, Operating Supplies, 
 Maintenance, Plant Office Expense, Di- 
 rect Payroll Expense and Packaging Cost 
 
 INDIRECT PRODUCTION COSTS: 
 Supervision, control Laporatory, inoi- 
 rect Payroll Expense, Interest on Work- 
 ing Capital, General Plant Overhead 
 
 FIXED PRODUCTION COSTS: 
 Depreciation (10 years), Taxes and In- 
 surance 
 
 MANUFACTURING (PROCESSING) COST:^ 
 
 HMt MATERIAL COST: 
 
 TOTAL MANUFACTURING COST:* 
 
 $ 950,040 $.0528 $ 950,040 $.0528 $ 950,040 $.0528 
 
 180,960 .0100 
 
 455,040 .0253 
 
 1,586,040 .0881 
 
 1.200,000 .0667 
 
 197,760 .0109 
 
 455.040 
 
 .0253 
 
 1,602,840 .0890 
 1, 800,000 .1000 
 
 214.560 
 
 455.040 
 
 .0119 
 
 .0253 
 
 1.619.640 .0900 
 2,400,000 .1333 
 
 $2,786,040 $.1548 $3,402,840 $.1890 
 
 $4,019,640 $.2233 
 
 600 Ton/Day Raw Fish (150 Day Operation) 
 
 DIRECT PRODUCTION COSTS:' 
 Labor, Utilities, Operating Supplies, 
 Maintenance, Plant Office Expense, Di- 
 rect Payroll Expense and Packaging Cost 
 
 INDIRECT PRODUCTION COSTS: 
 Supervision, Control Laboratory, Indi- 
 rect Payroll Expense, Interest on Work- 
 ing Capital, General Plant Overhead 
 
 FIXED PRODUCTION COSTS: 
 Depreciation (10 years). Taxes and In- 
 surance 
 
 MANUFACTURING (PROCESSING) COST:' 
 
 RAH MATERIAL COST: 
 
 TOTAL MANUFACTURING COST:* 
 
 $1,334,520 $.0494 $1,334,520 $.0494 $1,334,520 $0494 
 
 213,660 .0079 
 
 585,000 .0217 
 
 2.133,180 .0790 
 
 1. POO .000 $.0C C7 
 
 $3,933,180 $.1457 
 
 238,860 .0083 
 
 585,000 
 
 .0217 
 
 2;i58,380 .0799 
 
 2, 703,000 .1000 
 
 264,060 .0098 
 
 505,000 
 
 $4,8b8,3aJ ».1799 
 
 .0217 
 
 2,183,550 .0809 
 3.600,000 .1337 
 
 >7, 967, ICO >.2955 
 
 Unit costs ire based on a 151 yield of FPC from raw fish 
 Excluding raw material cost 
 
 Exclusive of row rratcrlal, sales and G 4 A expense 
 Including raw ratcrlal, exclusive of sales and C & A expense 
 
 1.4-10 
 
'^.S.Z.S By-products 
 
 The FPC process produces two valuable by-products: (1) fish oil and (2-.) 
 solubles. Sales values achieved for these by-products would be credited to 
 the FPC operation, thereby reducing the cost of producing the FPC. The fish 
 oil obtained in the FPC process will be higher in quality than fish oil pro- 
 duced in conventional fish meal reduction and should have a selling value of 
 6 cents per pound or greater according to SWECO. 
 
 Soluble and insoluble proteins can be recovered in a form consisting of 
 50 percent by weight, of solids, and acidified to a pH of 4.5, This product 
 could be sold as an animal foodstuff or supplement. According to SWECO, the 
 income realized from the sale of solubles should offset the recovery process- 
 ing expenses. It can be assumed these solubles would have a minimum sales 
 value of 0.5 cents per pound. 
 4.3.2.4 profit Calculations and production Costs 
 
 Assuming the average fishing season to be five months, according to 
 SWECO calculations, even the least attractive alternative — a 200-ton per day 
 plant paying $40 per ton for raw fish and selling FPC for 25 cents per pound- 
 
 t^ »^v^ouo. £■ ex y j.ii(~j y^w poi- i^wAi i-v^i- j-j_oii, tAn\a <j ^ ^ j- j.in^ 
 
 for 30 cents per pound, the plant would produce a gross profit, after manu- 
 facturing costs and property taxes, as follows: (1) $800,000 for 200-ton 
 per day plant (or 32% of the $2,500,000 investment); (2) $2,000,000 for the 
 400-ton per day plant (or 57% of the $3,500,000 investment); or (3) $3,200,000 
 for the 600-ton per day plant (or 71% of the $4,500,000 investment). 
 
 The data were calculated on the basis of a 15 percent yield of FPC from 
 raw fish. No credit was included for by-products such as fish oil and con- 
 densed solubles, although the cost of their recovery was included in the 
 plant operating costs. Also, no attempt was made to estimate selling and 
 distribution expenses and general and administrative costs. 
 
 If credits are included for the sale of fish oil and condensed solu- 
 bles, the gross profit could be increased from 10 to 40 percent, depending 
 upon the species of fish being processed. For example, a 400-t;on per day 
 plant on low-oil fish (3%) has a gross profit Vvfithout oil and solubles of 
 $2,000,000 ($30 per ton for raw fish, 30 cents per pound for FPC, 150-clay 
 operation). If oil and solubles are included, the gross profit is increased 
 to $2,250,000, a 12.5 percent increase (oil at 64q per pound and solubles 
 at ^c;: per pound). 
 
 1.4-11 
 
No costs were given for removal of bone fraction (resulting in lower 
 yield of FPC) where stringent fluoride specifications must be met. 
 
 Aside from the detailed computer programs used by SWECO, Inc. to esti- 
 mate manufacturing costs of FPC, the only other detailed cost study involved 
 an Engineering Economic ^odel developed by the University of Maryland under 
 contract to NMFS. The objective of this study was to develop a process cost 
 and design model which would predict construction and operating costs for 
 a wide range of FPC process conditions. 
 
 The model contains five subprograms for different type processes: fish 
 meal; biological; isopropyl alcohol extraction; presscake-biological ; and 
 presscake-isopropyl alcohol extraction. Of the processes in the model only 
 fish meal processing is a commercial operation at present. The model IPA 
 process was based on an analysis of the modified NMFS pilot plant at Aber- 
 deen, Washington. The cost data developed from this program is summarized 
 in Table A-2 for four of the subprograms, excluding fish meal, 
 
 production costs are rather insensitive to assumptions about interest 
 rates and the cost and life of capital assets over plausible ranges; using 
 a 6 percent rate of interest instead of 12 percent reduces costs bv less 
 than 1 cent per pound. This is due pri.marily to the assumption of a rather 
 short life for durable capital and the fact that the process is not very 
 capital-intensive. As one might expect, the cost of fish and the number of 
 days the plant can operate are the crucial variables. 
 
 To provide a partial check on the validity of these results, similar 
 analysis was performed on the cost estimates provided in a study done by 
 General Oceanology (GO) on FPC production in Chile. The effect of interest 
 rafes was seen to be small. production costs are quite sensitive to depre- 
 ciation periods in ranges near the 5-year figure chosen by GO; in the range 
 of 8- to 12-years. However, the effect is minimal. The cost of raw material 
 and the length of the annual operating season are of prime importance. 
 
 One interesting feature of the GO study is the estimate of by-product 
 revenues (from fish oil and fish solubles). Since that study assumed the use 
 of hake as raw material, the choice of 3 percent oil yield (weight as % of 
 weight in whole fish) was appropriate. However, the oil content of anchoveta 
 and other fatty fish is considerably greater. Oil content will vary from 
 season to season and average content for Peruvian anchoveta has been 
 
 1.4-12 
 III 
 
estimated at 6 percent. For other species, such as the Northern <California) 
 anchovy, this figure may range closer to 9 percent depending on the season 
 chosen for fishing operations. 
 
 The recovery of marketable fish oil from commercial operation of the 
 IPA process remains to be demonstrated. 
 
 1.4-13 
 
CM 
 I 
 
 W 
 
 1^ 
 
 M 
 Q 
 
 i 
 
 CO 
 
 M 
 Q 
 
 H 
 C/3 
 
 O 
 O 
 
 CO 
 CO 
 
 M 
 U 
 
 § 
 CJ 
 
 FE4 
 
 
 
 u 
 
 
 
 
 
 
 
 
 
 u 
 
 
 
 
 
 
 
 
 u 
 
 J 
 
 
 
 
 
 
 
 
 U) -u| 
 
 
 
 
 
 
 
 
 o 
 
 
 
 
 
 
 
 CO 
 
 ^ 
 
 CO 
 
 r-4 
 
 
 ;j 1 
 
 U. XI 
 
 •-I 
 
 
 o» 
 
 
 CO 
 
 •-* 
 
 
 iJ 
 
 
 
 
 
 
 
 
 > u 
 
 
 
 
 
 
 
 
 n 
 
 3 
 
 
 
 
 
 
 
 
 -o -b 
 
 
 
 
 
 
 
 
 o 
 
 
 
 
 
 
 
 
 )-. U 1- 
 
 t.v 
 
 
 tv 
 
 t* 
 
 t> 
 
 ^ 
 
 
 CJ ui o 
 
 i O 1 
 
 u > 
 
 ■J- 
 
 
 
 >-< 
 
 O 
 
 • 
 
 o 
 
 
 
 c^ 
 
 
 t-i 
 
 •H 
 
 .-1 
 
 •-* 
 
 
 o o< o 
 
 
 
 
 
 
 
 
 iJ U. c 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 o 
 
 
 o 
 
 
 o 
 
 o 
 
 o 
 
 o 
 
 
 o 
 
 
 o 
 
 
 o 
 
 o 
 
 o 
 
 o 
 
 
 tn i. 
 
 J 
 
 o 
 
 
 o 
 o 
 
 o 
 o 
 
 o 
 
 CO 
 
 o 
 
 
 U u 
 
 T 
 
 •—4 
 
 
 ,~t 
 
 en 
 
 CM 
 
 CJ 
 
 
 ^ c 
 
 > 
 
 
 
 o 
 
 00 
 
 
 O 
 
 * 
 
 
 o 
 
 
 •v> 
 
 
 v> 
 
 v> 
 
 </> 
 
 v>- 
 
 
 u 
 
 
 
 
 
 
 
 
 
 u 
 
 
 
 
 
 
 
 
 
 U 3 
 
 
 
 
 
 
 
 
 
 W T3 
 
 
 
 
 
 
 
 
 
 O O 
 
 
 -«* 
 
 
 •e- 
 
 f* 
 
 t^ 
 
 fj 
 
 
 D 
 
 
 Cm 
 
 • 
 
 
 • 
 
 • 
 
 vO 
 
 o 
 
 
 O 1 
 
 Oi > 
 
 
 
 
 CSI 
 
 .-1 
 
 CO 
 
 .-A 
 
 
 NO 
 f— 1 
 
 
 >- *J 
 
 
 
 
 
 
 
 
 
 rj o 
 
 
 
 
 
 
 
 
 
 O 3 
 
 
 
 
 
 
 
 
 
 ■o 
 
 
 
 
 
 
 
 
 
 H o 
 
 
 
 
 
 
 
 
 
 o u u 
 
 
 
 
 
 
 
 
 
 C. U) o. 
 
 
 
 
 
 
 
 
 
 O 1 
 
 
 ■t> 
 
 
 -«* 
 
 t* 
 
 i». 
 
 ■fj. 
 
 
 C XI 
 
 
 • 
 
 
 CO 
 
 CM 
 
 00 
 
 00 
 
 J 
 
 D a 
 
 ■J Hi o 
 
 
 u-i 
 CM 
 
 
 m 
 .-1 
 
 i-H 
 CM 
 
 
 ON 
 .-1 
 
 r 
 
 D 
 
 
 
 
 
 
 
 
 t 
 
 ■^ 
 
 
 
 j 
 
 
 
 
 
 c 
 
 ^ 
 
 
 
 
 
 
 
 
 
 1 
 
 
 O 
 
 
 o 
 
 CJ 
 
 o 
 
 o 
 
 
 
 
 O 
 
 
 O 
 
 o 
 
 o 
 
 o 
 
 
 
 
 o 
 cT 
 
 
 o 
 
 CD* 
 
 o 
 o 
 
 o 
 o 
 
 o 
 o 
 
 
 -< o 
 
 a. o 
 
 
 o 
 
 CM 
 
 
 CM 
 
 CO 
 CM 
 
 CM 
 
 00 
 
 CM 
 
 C^l 
 OO 
 
 CM 
 
 . 
 
 o 
 
 
 v»- 
 
 
 vy 
 
 ■V>- 
 
 v> 
 
 vy 
 
 
 o 
 
 
 
 
 
 
 
 
 
 irt -o 
 
 
 ■y. 
 
 
 ■«. 
 
 ■w- 
 
 t^ 
 
 •^ 
 
 
 o o 
 
 
 o 
 
 • 
 
 
 CO 
 
 • 
 
 f-« 
 
 •-H 
 
 't 
 
 
 c 
 
 
 o 
 
 
 o 
 
 O 
 
 •O 
 
 f^ 
 
 
 O 1 
 
 Q4 > 
 
 
 fo 
 
 
 CM 
 
 CM 
 
 CM 
 
 CM 
 
 
 U. X) 
 
 
 
 
 
 
 
 
 > 
 
 iJ 
 
 
 
 
 
 
 
 
 <t 
 
 o 
 
 
 
 
 
 
 
 
 X 
 
 3 
 •D 
 
 
 
 
 
 
 
 
 u 
 
 O 
 
 
 
 
 
 
 
 
 rj 
 
 l~) U 
 
 
 t* 
 
 
 t*- 
 
 •«* 
 
 -U* 
 
 
 P 
 
 0) Cl 
 o » 
 
 
 •-I 
 
 
 CO 
 
 00 
 
 CM 
 
 
 
 • 
 
 
 • 
 
 • 
 
 • 
 
 
 u- 
 
 o >- 
 
 
 vO 
 
 
 CM 
 
 .-1 
 
 r-4 
 
 f—t 
 
 r 
 o 
 
 X) 
 
 o 
 
 
 ro 
 
 
 CM 
 
 n 
 
 CO 
 
 CO 
 
 i~i 
 
 a, o 
 
 
 
 
 
 
 
 
 c 
 
 
 
 
 
 
 
 
 
 kO 
 
 
 
 
 
 
 
 
 
 
 
 
 O 
 
 
 O 
 
 o 
 
 O 
 
 o 
 
 
 
 
 o 
 
 
 O 
 
 o 
 
 o 
 
 o 
 
 
 --1 « 
 
 
 o 
 
 
 O 
 
 o 
 
 o 
 
 o 
 
 
 (TJ iJ 
 
 
 o 
 
 
 O 
 
 o 
 
 o 
 
 o 
 
 
 U w 
 
 
 «r» 
 
 
 .-« 
 
 00 
 
 r» 
 
 f^ 
 
 
 ■-< o 
 
 
 O 
 
 
 «» 
 
 lO 
 
 in 
 
 m 
 
 
 CJ 
 
 
 — < 
 
 
 
 .-4 
 
 
 
 
 
 
 
 
 *i 
 
 
 
 
 
 
 
 
 
 ca 
 
 
 
 
 
 
 
 
 
 U4 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 '""^ 
 
 
 1 
 
 .-1 
 
 ; 
 
 
 1 
 
 
 jC 
 
 •--« 
 
 
 « 
 
 rS 
 
 Q) 
 
 
 V C 
 
 
 W) 
 
 <9 
 
 
 ^ 
 
 O 
 
 J£ 
 
 
 
 ^^ 
 
 U 
 
 
 •^ 
 
 •^ 
 
 C n 
 
 u 
 
 cQ 41 
 
 **-# 
 
 UJ 
 
 •r4 
 
 
 o 
 
 «) 
 
 t u 
 
 <» 
 
 U'>-i 
 
 o 
 
 -^ 
 
 M) 
 
 
 
 o 
 
 v 
 
 (^ 
 
 
 
 aJ 
 
 o 
 
 4J 
 
 in 
 
 •-^ 
 
 .-< lA 
 
 •~» 
 
 It < 
 
 CI 
 
 C 
 
 •— * 
 
 >« 
 
 u> 
 
 o 
 
 — «) 
 
 < 
 
 ex 
 
 <» 
 
 o 
 
 U4 
 
 « 
 
 ■-4 
 
 < « 
 
 CL. 
 
 41 »-i 
 
 '. ^ 
 
 i— « 
 
 ■VH 
 
 
 ki 
 
 U3 
 
 CV< W 
 
 r-t 
 
 u 
 
 o. 
 
 M 
 
 
 (U 
 
 
 M (U 
 
 
 (U 
 
 1.4-14 
 
4.3.3 FPC Marker, potential 
 
 There are txvo general characteristics to be examined in this area: (1) 
 the functional properties needed in the final food product to be made from 
 the protein supplement and (2) the nutritional value of the material to the 
 user of the final food product. 
 
 There are many protein-rich materials which possess either unique func- 
 tional or nutritional benefits, and the food industry is constantly examining 
 these in various applications. The functional pr6perties are very important 
 because the structure, texture and eating characteristics are dependent in a 
 large part upon the functional properties of the proteins. It is not neces- 
 sary however that a protein material possess beneficial functional properties 
 to allow its use; a neutral effect may be perfectly satisfactory. 
 
 If any protein material does not seem to be of value from the standpoint 
 of functional properties, it is possible that it may possess outstanding 
 characteristics in the area of nutrition. 
 
 Guidelines for successful commercial development of FPC are: 
 
 (1) Only the production of a practically odorless, flavorless 
 and nonreverting FPC should be considered. 
 
 (2) FPC will be used almost exclusively as an ingredient in 
 other food products and will not be consumed by itself. 
 
 (3) If FPC is to be accepted by the food industry, its good 
 nutritional properties (i.e., a protein efficiency ratio 
 at least equal to casein) must be taken for granted. In 
 addition, the food industry will consider using FPC as an 
 ingredient only if it can economically replace other in- 
 gredients without a sacrifice in the quality of the final 
 food product in which it is used. 
 
 4.3.3.1 potential Markets 
 
 Several of the suggested markets are based on the nutritional charac- 
 teristics of FPC They include: (1) pet foods, (2) institutional diets, 
 (3) geriatric and reducing diets, and (4) high protein snacks. 
 
 Since it has been reported that a significant portion of canned pet food 
 is consumed by humans, this product generally has to meet human standards 
 with regard to sanitation and toxicity. Hence, there is an argument for a 
 market for hvmian-grade FPC in pet foods. The high protein snack market would 
 
 1.4-15 
 
/ 
 
 requires a Madison Avenue approach, in which the producer would attempt to 
 convince the consumer that he should pay more for an FPC-supplemented snack. 
 
 It has also been suggested that FPC might find a substantial market in 
 the processed meat field which uses and distinguishes between functional 
 protein and filler protein. Functional protein supplies texture, binding 
 power and taste, and presently is comprised of meats costing the industry 
 AO cents per pound of protein and more. Filler protein is expected to supply 
 little more than bulk. Currently, by-products from the packing industry are 
 available at about 3 cents per pound or about 10 cents per pound of protein, 
 presently, FPC cannot compete with functional protein on the basis of its 
 properties, or with filler protein on the basis of its costs. If the binding 
 power of fish protein could be improved, it would open up a tremendous 
 market for the product. 
 
 While it was recognized that the greatest potential markets were in 
 developing countries, much of the impetus of FPC development work came from 
 developing countries. Food processors want protein with inexpensive ingredi- 
 ents for the manufacture of a large variety of food products ranging from 
 soups to cookies to sausages. The following were believed to be potential 
 markets for FPC in developed countries: 
 
 (1) Sale of a high-protein value product directly to the consumer 
 market for "fad foods," 
 
 (2) Sale of a tasteless and odorless concentrate to food processors 
 for incorporation in staple food products, 
 
 (3) Sale to food processors as a protein supplement for incorporation 
 in dietary or specialty foods, 
 
 (4) Sale to hospitals, prisons, schools, etc,, for incorporation as 
 a protein supplement in institutional feeding, 
 
 (5) Sale to the government for use in domestic food-aid programs, 
 
 (6) Sale to the government to provide a store of strategic food 
 supplies for emergency conditions, 
 
 A. 3. 3. 2 protein Utilization 
 
 Hammonds and Call of Cornell University described the current protein 
 food additive market in the United States, protein ingredients are added to 
 processed food products for their pliysical properties and for the improvement 
 in the physical properti.es of the end product. These functional characteris- 
 tics may be more important than the nutritional role for highly processed foods. 
 
 ti' I.A-16 
 
In a follow-up report, Hammonds and Call projected the future market 
 for protein ingredients. Each of 16 major product categories was examined 
 to determine the reasons for protein ingredient use and an assessment was 
 made for the ease of penetration of new proteins for each category. 
 
 protein needs were combined with ingredient characteristics, and three 
 categories were established: (1) those with highly specialized uses, not 
 likely to be subjected to replacement; (2) those with minimum functionality, 
 highly vulnerable to replacement; and (3) the middle ground, those ingredi- 
 ents vulnerable to replacement by a protein source with the correct functional 
 properties . 
 
 Hammonds and Call made an important point about ingredient image and 
 the choice of markets to penetrate. It is easy for the producers of a new 
 protein source to fall into the trap of serving the speciality food market 
 segments. Once any early reputation is established as a "poor people's 
 food" or as "animal feed," this reputation may hinder later marketing efforts. 
 Care must be taken to establish and maintain an image suitable to the long- 
 term goals for the protein ingredient. 
 
 Development of functional characteristics and solution of flavor problems 
 will determine the rate of growth for individual protein ingredients. How- 
 ever, one trend seems clear; the continued decline in use of nonfat dry milk 
 and dry whole milk. While the total level of protein in ingredient form 
 grows at 9.3 percent annually, protein sources other than milk must grow at 
 13.9 percent in order to make up for declining milk use. 
 
 Cornell University undertook a study for NMFS on consumer attitudes 
 towards alternative protein ingredients, probably the most important fea- 
 ture of the study was the ignorance of the consumer regarding protein ingredi- 
 ents presently eaten and those likely to be purchased in the future. 
 
 Prices and brand preferences were the most important determinents in 
 purchasing. Consumer knowledge appears to be limited to the inference given 
 to the name of the ingredient. In the direct comparison of ingredients, the 
 product in which they are incorporated influences the consumer. Without any 
 knowledge of FPC, consumers found soy protein concer.trate and textured vege- 
 table protein quite acceptable, nonfat dry milk was less so, and whole fish 
 protein concentrate came in last. On informing the consumer of the proper- 
 ties of whole FPC, less than 20 percent objected to it. The consumer tests 
 did not include real food products; hypothetical situations were constructed. 
 
 1.4-17 
 
If FPC can be produced with beneficial functional properties for use in 
 food systems, then it will receive consideration for use in products in the 
 U.S. market. If the main beneficial characteristic is its nutritive value, 
 then it will be of interest primarily in the lesser developed areas of the 
 world. 
 
 At present, FPC has few, if any, functional properties, though these may 
 be improved in the future. The extent of the potential use of FPC cannot be 
 defined with any degree of accuracy until the price range and functional 
 properties of the product are more fully known. The easiest markets to pene- 
 trate would be those of beverages and breakfast foods. If whole fish protein 
 concentrate could preserve freshness, prolong shelf life and retain moisture, 
 then it could compete with soy flour in baked products. The ability to ab- 
 sorb fat, carry flavors and provide some texture improvement could mean that 
 this ingredient could compete with the soy concentrate/soy isolate market in 
 baby foods, baked goods and processed meats; that is, assuming that this 
 could be done at a competitive price. 
 
 The food industry will consider using FPC as an ingredient only if it 
 can economically replace other ingredients without a sacrifice in quality. 
 Its price would have to be competitive first of all with other animal pro- 
 teins. If a maximum wholesale price of 25 to 30 cents a pound, at a minimum 
 protein content of 80 percent can be assumed, FPC would be cheaper than other 
 animal proteins. However, FPC will also have to compete with oilseed pro- 
 teins and possibly with cereal proteins fortified with synthetic amino acids 
 and would face serious competition from soybean products. Food processors 
 have to convince themselves of the desirability of including FPC as an in- 
 gredient in their existing products. Consumer acceptance may have psycho- 
 logical implications but is of lesser importance since FPC should not in 
 any manner change customary properties of the product. 
 
 A. A FUTURE OUTLOOK 
 
 The marketing of products labeled "with fish protein" may impair the 
 position of FPC even if its price is competitive with other proteins, as 
 pointed out in the study on consumer attitudes to FPC. Also, the nonfunc- 
 tional properties of FPC make a cost comparison with other protein sources 
 impossible at present. 
 
 The only market arijas for IPA-FPC are in uses where nutritional quality 
 
 1.4-18 
 
is important, e. g., baby foods, breakfast food, and canned and processed 
 meats. These uses are presently supplied by nonfat dry milk (costing 40 
 to 50 cents per pound of protein) and soy concentrates and isolates (costing 
 27 to A2 cents per pound of protein). 
 
 The importance of the raw material source in deciding where to locate 
 FPC production facilities is quite clear. Among the most crucial factors 
 are the cost and regularity of supply. Also, the oil content of the species 
 utilized is potentially very important. Other factors of importance in 
 choosing an efficient location are r a labor pool possessing necessary skills, 
 and back-up facilities needed to maintain the fishing fleet and the FPC plant 
 including recovery os by-products. No cost estimates to date have included the 
 equipment needed to control emission of pollutants into the air and water. 
 A plant located in the United States must take these costs into account. 
 
 The value of by-products and the technology for their recovery is of 
 vital importance to the commercial feasibility of FPC. Furthermore, this 
 factor will increase in importance as increased utilization of lean fish for 
 direct human consumption forces reduction operations (fish meal and FPC) to 
 rely more and more upon clupeoids. 
 
 One of the most severe impediments to the determination of the role of 
 FPC in world protein markets is the lack of even rough estimates of processing' 
 costs based on actual production experience. 
 
 The very name "protein" suggests nutrition. This has been the traditional 
 rolfe of protein and the focus of most work has been on the development of 
 new protein sources. The problem of developing functional properties is 
 often not appreciated during the early phases of research. However, it would 
 be a mistake to abandon nutritional consideration entirely. Relatively bland, 
 unsophisticated protein ingredients are used in some domestic products to 
 raise the general protein level. 
 
 There can be no doubt that there is a tremendous market for protein in 
 the United States and that a very small penetration of this market percent- 
 agewise would support a rather large FPC industry. 
 
 I.A-19 
 
SECTION 5 
 
 LABORATORY PROCESSES 
 
 This section summarizes the laboratory and processing development ef- 
 forts made by NMFS facilities throughout the course of the FPC program. 
 
 5.1 CONCLUSIONS 
 
 FPC can be produced by various proven methods as tested in the labora- 
 tory at both bench-scale and model-scale unit levels. 
 
 The major effort of NMFS since 1963 has been concentrated on develop- 
 ment of a solvent extraction process using isopropyl alcohol to remove 
 water, lipids and flavors from whole raw fish. Various parameters of this 
 process have been investigated on both lean and fatty species. While it 
 appears to be a viable means of producing FPC, from both ',an engineering and 
 economical standpoint, several process modifications require further explora- 
 tion. In other vwrds, although enough work has been done to confirm the 
 potential of producing FPC by the IPA process, a complete process has not 
 been engineered and several modifications should be investigated. 
 
 It may be generalized that the greater the degree of separation of 
 protein from other constituents of fish (lipids, water and flavors), the 
 more complex and costly the process will be. Thus, to bring FPC to commer- 
 cial feasibility, a process must be developed that combines maximum simplic- 
 ity and economy with an acceptable degree 'of protein concentration. 
 
 5.2 BACKGROUND 
 
 5.2.1 Early NMFS Participation 
 
 Upon initiation of the FPC effort in 1961, laboratory work was devoted 
 to a number of objectives. One of these was a worldwide investigation and 
 literature survey to determ.ine the types of FPC bein^ produced, the avail- 
 able or potential processes to produce them and the state of the art in 
 general. One basic conclusion to be drawn from this initial survey and 
 
 1.5-1 
 
literature search was the substantial level of interest throughout various 
 nations in the development of FPC as a viable protein source. 
 
 The survey reviewed a broad spectrum of FPCs, including fish pastes, 
 sauces and protein isolates, as well as the more conventional FPCs. Types 
 of FPCs ranged in characteristics from bland, odorless and tasteless prod- 
 ucts to distinctive fish, beef or cheese flavored products. A variety of 
 FPC processes had been proposed, which could be generally categorized into 
 three types of processes: (1) chemical (solvent extraction) processes, 
 (2) biological (enzymatic and bacterial hydrolysis) processes and (3) physi- 
 cal (thermal and mechanical) processes. Based on these findings, NMFS 
 designed a laboratory research program to investigate the three types of 
 processes . 
 
 The decision was made in 1963, to concentrate NMFS effort on rapidly 
 developing what appeared to be the most promising FPC process at that time — 
 the solvent extraction of fish by isopropyl alcohol (IPA) . The major por- 
 tion of NMFS laboratory work, thereafter, was devoted to thoroughly inves- 
 tigating and developing the JPA process from bench scale to the point of 
 demonstrated commercial use. 
 
 Most of the work was done by NMFS at their various laboratory facili- 
 ties (particularly College Park and Beltsville, Maryland), with some speci- 
 fic projects being done by contractors. 
 5.2.2 FPC processes 
 
 In simplest terms, the principle of most FPC processes is the removal 
 of water and lipids from whole fish by whatever means (chemical, biological 
 or physical). Because lipids will rapidly oxidize, resulting in distinc- 
 tive flavor and odor characteristics, maximum removal of lipids is necessary 
 to minimize unfavorable characteristics in the final product. More complex 
 separation and protein isolation processes have also been developed in the 
 laboratory but the principle remains basically the same: separate the de- 
 sired constituent (protein) from the undesired constituents (mainly water, 
 lipids and flavors). From a general economic and engineering viewpoint, 
 the greater the degree of separation achieved, the more complex and costly 
 the processing will be. The simplest process is dehydration of whole fish 
 followed by solvent extraction of the lipids. The most complex is the 
 mechanical isolation of the most desirable raw muscle tissues (i.e., evis- 
 ceration, skinning, deboning, etc.), followed by chemical isolation of the 
 
 1.5-2 
 
protein from the lipids, water, etc. It was the aim of NMFS to develop a 
 process for FPC manufacture which v/ould produce the most nutritious protein 
 supplement by the most economical and least complex means, while not over-»» 
 looking such important factors as product acceptability and safety. 
 
 5.3 IPA PROCESS 
 
 5.3.1 principle of the IPA process 
 
 As a result of both the world survey and preliminary NMFS work, it was 
 determined that solvent extraction processes promised the most favorable 
 potential -for rapid development to a commercial scale. An important factor 
 in the selection of the most promising type of process was whether the 
 processing method and the resulting FPC would meet the standards and regula- 
 tions of the U.S. Food and Drug Administration; alcohol (solvent) extraction 
 seemed to pose the fewest problems in this respect. 
 
 The choice of solvent was dictated by a number of factors: extraction 
 efficiency, safety, toxicity, availability, cost and stability of the sol- 
 vent during its redistillation and recovery. Laboratory tests indicated 
 that two solvents--ethanol and isopropanol — gave the best results in rela- 
 tion to the above factors. IPA demonstrated better lipid extraction charac- 
 teristics, was less costly and relatively free from the regulatory controls 
 of the U.S. government. Thus, it was chosen over ethanol. 
 
 The principle of the IPA process to produce FPC from whole raw fish is 
 relatively simple. A successive series of countercurrent extractions of 
 the comminuted fish are made with azeotropic IPA which dehydrates, defats 
 and deodorizes the remaining fish solids. Residual IPA is removed by evap- 
 oration and the remaining solids are milled and packaged. 
 
 5.3.2 Source Species 
 
 NMFS considered and tested many different species of fish and marine 
 animals for FPC manufacture. probably any edible aquatic animal can be 
 used, by itself or in combination with others. Toxicity would be and is a 
 principal constraint in selection, although toxic species of fish are rela- 
 tively rare. However, because of the possibility of toxic fish existing in 
 certain areas of the world, NMFS concluded that the processing of such 
 species requires more investigation before they can bo considered in the 
 manufacture of FPC \ 
 
 Successful FPCs have been processed from shellfish, crustaceans, fin 
 
 i.5-3 
 
fish, telestoids and others. NMFS has studied (either cursorily or in detail) 
 FPCs from hake, whiting, herring, menhaden, pollock, sardine, anchovy, ale- 
 wives, smelt, ocean pout and flounder. Most have produced acceptable FPC 
 with high nutritional value. 
 
 It was recognized from the beginning that the fatty species of fish such 
 as herring, menhaden, anchovy and sardines offered the greatest potential 
 from a resource availability standpoint. Hake, however, was an abundant, low 
 cost, underutilized resource on both the East and West Coasts of the U.S. in 
 the early sixties and was selected for the initial investigations. Another 
 advantage of hake was their lower fat and flavor content, thus process de- 
 velopment was considered easier. Success with the development of an accept- 
 able product from hake would encourage the more difficult and complex devel- 
 opment of processing the species of higher fat content. 
 5.3.3 processing parameters 
 
 NMFS investigated a number of processing parameters for IPA extraction 
 at both the laboratory and pilot plant level. These investigations were 
 aimed at extrapolation (i.e., engineering scale-up) to a large-scale demon- 
 stration plant. Most of the parameters were related directly to the engi- 
 neering aspects of the IPA process. Much of the data for establishing per- 
 formance requirements for large-scale equipment were provided by laboratory 
 work. 
 5.3.3.1 Storage and Handling 
 
 Since it is impossible to process a catch of fish instantaneously, even 
 if the FPC processing plant were to be located on the vessel, satisfactory 
 short-range storage (one day minimum) is necessary to hold the fish from time 
 of harvest to time of processing. NMFS conducted several studies to inves- 
 tigate various methods for storing the source fish, both on a long- and short- 
 range basis. 
 5.3.3.1.1 Frozen 
 
 Frozen storage was the most frequently used method of storage during 
 the FPC program, particularly for the laboratory and pilot plant work, but 
 was not considered commercially feasible for FPC production. Very few studies 
 were made of frozen storage because it was already a highly developed commer- 
 cial method of fish storage. No particular problems were encountered with 
 the use of frozen lean fish. It was found that freezer burn (partial dehy- 
 dration in storage), normally undesirable, did not appear to adversely affect 
 
 1.5-4 
 
 I 
 
the quality of FPC . Later experiments with fatty fish indicated that several 
 months frozen storage was generally satisfactory. Freshness of the fish 
 prior to freezing Vs/as probably the most important factor in the quality of* 
 the FPC produced. Losses of soluble protein (solubles) during extraction 
 of the fish tended to be less for frozen fish than for fresh or thawed fish. 
 This was probably due to a lower degree of enzymatic hydrolysis taking place 
 from time of comminution to processing. 
 
 5.3.3.1.2 Freeze Dried 
 
 Freeze drying was not considered to be commercially feasible for a low- 
 cost FPC and was never investigated as a means of dehydration prior to ex- 
 traction of lipids. Freeze drying was used to prepare standard samples for 
 nutritional evaluation and in studying methods and processing parameters 
 for making FPC. The nutritional quality of protein in ipA-extracted fish 
 (FPC) was found to be virtually the same as the protein in the freeze-dried 
 standard . 
 
 5.3.3.1.3 IPA 
 
 Limited studies were conducted on the possibility of storing comminuted 
 fish in IPA. Good preservation was obtained in one study with menhaden of 
 low-fat content where the t ish-and-aicohoi slurry was heated to 170^ F prior 
 to storage. Other studies using either fresh fish or cooked and pressed 
 fish, stored in room temperature alcohol, indicated potential problems with 
 FPC flavor and yield. 
 
 No definitive studies were made on the stability of fish which was par- 
 tially extracted with IPA. However, there were numerous occasions when the 
 partially extracted fish solids were held several weeks at room temperature 
 with apparently no adverse effect. There appears, however, to be no advantage 
 of utilizing this method for storage. One may as well complete the process- 
 ing and store the FPC. 
 5. 3. 3.1. A Refrigerated Sea Water or Brine 
 
 Storage of fresh fish in refrigerated sea water (RSW) is known to be 
 effective for varying periods, depending on the species and conditions. 
 Studies showed that pacific hake autolyzed rather quickly in RSW and pro- 
 longing the storage period resulted in a decreased yield of FPC RSW was 
 also an unsatisfactory storage medium for anchovy (a fatty fish) because of 
 the occurrence of off flavors in the FPC Frozen storage of anchovy worked 
 best. However, if the landed fisli could be prechilled promptly in refrigerated 
 
 1.5-5 
 
sea water, storage was satisfactory on a short-term basis. Storage of hake 
 in refrigerated brine at the EDP was not entirely satisfactory (see Section 
 6.2). 
 
 5.3.3.1.5 Dehydration 
 
 Dehydration of raw fish prior to extraction was considered and investi- 
 gated to a limited extent by NMFS .and extensively by others. Qooking, press- 
 ing and drying (fish meal); vacuum drying; drum drying; flash drying; hetero- 
 azeotropic dehydration and defatting; and vacuum dehydration in an oil medium 
 are methods of dehydration that have been used or proposed. Generally, the 
 use of such methods has resulted in a lower quality FPC Many of these 
 methods, however, offer potential economies and possible advantages for in- 
 termediate product preprocessing. 
 
 5.3.3.1.6 Intermediate product preprocessing 
 
 Some investigative work was done on the possibility of processing fish 
 into an intermediate, product and then storing the intermediate product for 
 subsequent processing into FPC. Like frozen storage, preprocessing offers 
 potential long-range storage capability. There are several advan<tages to 
 the concept: fish could be processed into an intermediate product and stock- 
 piled to permit year round proauccion of I'ft^', excess scores could be diverted 
 to other products as the FPC market fluctuated; and more uniform quality FPC 
 could be produced by blending of intermediate products. Disadvantages are 
 indicated as well: double processing would mean increased production costs 
 and greater sanitation problems, quality standards would require multiple 
 considerations, and storage stability of various intermediate products might 
 not be uniform. 
 
 Some intermediate product work has been done commercially, including 
 azeotropic dehydration and defatting folloived by alcohol extraction (the 
 VioBin Corporation and Alpine Marine Corporation); and a heat transfer method, 
 or HTM> to produce a fish meal-type product (Zapata-Haynie Corporation). 
 Conventional fish meal has been extracted to make FPC, However, the FPCs 
 thus produced were not as acceptable organoleptically as those produced by 
 direct IPA extraction. Some preprocessing methods show favorable potential, 
 such as presscake from a food-grade fish meal operation or centrifuge cake 
 produced by substituting a centrifuge for a conventional press in such an 
 operation. The NMFS-developed aqueous extraction process has also shown 
 promise in producing a storageable fish cake for subsequent processing into 
 FPC or fish' protein isolates with desirable functional properties. 
 
 1.5-6 
 
5.3.3.2 Comminution and Evisceration 
 
 Comminution, or grinding, of fish is an important initial part of the 
 IPA process, however, it has not been studied separately and problems have** 
 been encountered in this area. 
 
 Size of the comminuted particles does not appear to be critical below 
 3mm (or ^ inch). Particles twice that size may be satisfactory for large- 
 scale plants. Frozen fish comrninuted much differently than did fresh, the 
 frozen material showing a greater tendency to fracture. Some difficulties 
 have been encountered with the fish clogging the comminution machinery and 
 in stringy solids remaining in the resulting slurry. Studies utilized 
 various machines, including a Reitz — mill, a Bibun-type deboner and a 
 Beehive deboner (these combined deboning and comminution). 
 
 Evisceration was not widely investigated, the principal approach being 
 the utilization of whole fish for economic and technical reasons. However, 
 very limited work was done on the development of an enzymatic evisceration 
 process. No conclusive results are available. 
 
 5.3.3.3 Deboning 
 
 Deboning was studied in the Model Scale Unit (pilot plant, or MSU) , the 
 laboratory and by contractors. Since the bones in fish account tor most of 
 the minerals, including fluorides, their removal reduces the mineral content 
 (thereby increasing the protein content) of the resulting FPC Two principal 
 means of deboning were investigated. Wet deboning was accomplished by utiliz- 
 ing any of several types of commercial meat-bone separators to remove bone 
 from the raw fish before extraction. Dry deboning was carried out by the re- 
 moval of bone particles from the dry extracted fish solids. 
 
 The major purpose of bone removal was to decrease the fluoride content 
 from those species of fish high in fluorides in order to meet the FDA require- 
 ment for fluoride residue in FPC (100 ppm) . Deboning could also be used to 
 increase the protein content if the FPC was made from excessively bony fish 
 or simply to produce an especially high-protein FPC Normally it was con- 
 sidered undesirable to lower the bone content of FPC, since the bones contain 
 nutritionally valuable minerals and proteins, and their removal consituted 
 not only added processing cost but a waste of material. 
 
 The advantages of wet versus dry deboning are: 
 
 (1) proven methods and equipment are available. 
 
 — Reference to trade names does not imply endorsement by the National Marine 
 
 fisheries Service, NOAA. 
 
 1.5-7 
 
(2) Bones and other material (sometimes scales and skin) to be 
 removed do not have to be solvent extracted, thus leaving 
 extraction capacity in the plant to produce either higher 
 quality protein or more through-put. 
 
 The disadvantage is that a highly perishable by-product is created and 
 constitutes a disposal/utilization problem. 
 
 Dry deboning would have the advantages of; 
 
 (1) permitting closer control of desired composition of the 
 final product by use of blending or controlled classifi- 
 cation. 
 
 (2) A dry, stable by-product would be of higher value and be 
 more easily sold. 
 
 The disadvantage is that exposure of dried product to the deboning 
 process greatly increases the opportunity for bacterial and other product 
 contamination. 
 
 Considerable work was done on dry deboning by classifying and gravity 
 separation techniques, but no satisfactory operation was established. 
 5.3.3.4 Solvent Ratios and Extraction Efficiency 
 
 Various extraction parameters were studied, including solvent ratios, 
 temperatures, times of extraction, particle size, number of stages, cross- 
 current and countercurrent extraction, mixing and separating methods and 
 equipment design. 
 
 Because fish is a highly variable, complex, physical (colloidal) and 
 chemical system which changes constantly, it is virtually impossible to com- 
 pletely and accurately define and express the many complex and changing re- 
 lationships . 
 
 A multistage system of mixing and separating was selected for the basic 
 process because of the extreme changes in physical properties that occur 
 during extraction and the apparent lack of availability of any single com- 
 mercial extraction system to adequately handle these changes. Filtra- 
 tion ( vacuum, , gravity, screening, pressure, and centrifugal) and decanting 
 (gravity and centrifugal) were studied. In spite of many disadvantages, 
 centrifugal decantation has continually proven to be the most effective and 
 consistently satisfactory method of separation applicable to the highly 
 variable conditions enccjuntcred in alcohol extraction of raw fish. 
 
 Many experiments were conducted to study the important variables and to 
 
 1.5-8 
 
attempt to express the relationships between them in an accurate and meaning- 
 ful manner necessary for good process design. The high degree of variability 
 of the raw material and the behavior of the materials in processing, make'^t 
 imperative that the process and plant design be flexible enough to be operable 
 within wide limits. 
 
 Key results of the extraction research carried out to permit development 
 of the multistage mix and separation process selected by the NMFS may be sum- 
 marized as follows; 
 
 (1) Countercurrent extraction is satisfactory and necessary for 
 economy of solvent usage. 
 
 (2) Temperature of extraction should be held near the boiling 
 point of the solvent, 80°C (65° to 80°C is satisfactory). 
 
 (3) With moderately intense mixing, high temperatures and particle 
 sizes of \ inch maximum, the extraction rate is rapid and con- 
 tact time need not exceed 15 minutes per stage, 
 
 (A) Efficiency of extraction is mostly influenced by the degree 
 of separation of liquids and solids at each stage, and the 
 highest efficiency is achieved by a centrifugal decanter. Com- 
 plete solubility of lipids in the alcohol phase is not required 
 in the first stage and would require a prohibitive solvent ratio. 
 (5) Solvent to fish (or deboned flesh) weight ratios and number of 
 stages required are inversely related. Neither are as greatly 
 influenced by lipid content of the fish as would normally be 
 anticipated from solubility phase diagrams. practical solvent 
 to fish ratios are limited to between 1.5 and 2.5 pounds of 
 azeotropic IPA per pound of fish and the number of stages limited 
 to between 3 and 5, 
 5.3.3.5 Desolventization and Drying 
 
 Desolventizing is the reduction to the desired level of residual solvent 
 in the extracted solids. With the variabilities of the process, prob- 
 lems with analytical procedures and the effect of this processing step on 
 the product characteristics, desolventizing has required much attention and 
 yet remains to bo definitively resolved. 
 
 Solvent extraction of water from protein tissues with alcohol removes 
 free water and much of the "bound" water, replacing it with alcohol, some 
 of which itself becomes "bound." Simple drying will not remove this "bound 
 
 1.5-9 
 
alcohol" much below the 1 percent level in the final product. The release 
 and replacement of the alcohol by water is necessary to meet the difficult 
 specification limit of 250 ppm residual IPA set by the FDA. 
 
 Drying temperatures were initially limited (below 65°C) through vacuum 
 drying but gradually raised to as high as 120°C, as feeding studies proved 
 little damage was done to the nutritional value of the FPC 
 
 Steam stripping of the solvent-wet solids for desolventization, fol- 
 lowed by drying to remove condensed moisture, was used and shown effective. 
 As more attention was given to color of the product (the lighter, the better), 
 it was shown that this direct steaming of wet solids resulted in darker FPC. 
 Lighter color resulted if the solids were at least partly dried before steam 
 stripping. 
 
 Desolventizing of solvent-wet solids can best be accomplished by partial 
 drying followed by steam stripping and final redrying, if necessary, to as- 
 sure a final moisture content below 10 percent and residual alcohol below 
 250 ppm. The drying and desorption rates are still inadequately defined to 
 permit precise design of such equipment. 
 
 Since desolventizing contributes greatly to the removal of residual odors 
 in the FPC, complete deodorization of the recycled solvent is not required in 
 the recovery system. 
 5.3.3.6 Milling 
 
 The desired final particle size of FPC is defined by several factors, 
 among which are its composition (bone and hard particle content) and its 
 intended use. 
 
 FPC is not easily milled and considerable difficulties have been ex- 
 perienced in milling. A number of vendor's tests were made to select a satis- 
 factory mill. Hammer mills have not proven adequate, and although air at- 
 trition or fluid energy mills have been used with considerable success, their 
 operation on FPC would be very expensive. Impact mills with classifiers 
 appear to be a reasonable compromise, but control of input moisture can be 
 difficult and capacities are low. 
 
 Since fine milling is expensive any FPC producer will probably desire 
 the flexibility of producing particle sizes to meet the specific customer 
 demand. Specifications for fineness of FPC from the EDP wore set so that 95 
 percent would pass a standard 200-mosh screen. This resulted in a widely 
 acceptable product. 
 
 1.5-10 
 
5.3.3.7 Solvent Rncovery 
 
 Miscella (solvent effluent from the extraction system) from the IPA 
 extraction process is comprised of isopropyl alcohol, water, lipids, mixed 
 soluble proteins, nitrogenous and mineral constituents and fine-suspended 
 solids, including high-lipid flocculcnt solids. The composition is 
 variable, depending upon the initial composition and condition of the fish, 
 solvent ratio used, and degree of separation achieved with the first-stage 
 separator. Depending on the lipid content of the fish, the miscelia can 
 exist as either a two or three-phase system. Centrifugal separation will 
 result in a heavy solids phase, a heavy oil phase (containing lipids and 
 solvent), and a light solvent phase (containing solvent, water, solubles 
 and dissolved oil). All components exist as complex mixtures of compounds. 
 
 Solvent (the water azeotrope of isopropyl alcohol) must be recovered 
 for reuse with minimum loss, but must be of a purity suitable for such reuse 
 in the extraction system. Also, the extracted lipids and proteins must be 
 recovered as by-products (if commercially feasible) and any waste products 
 must be treated and discarded in an acceptable manner (pollution control). 
 
 The rectification of isopropanol is a well known chem.ical engineering 
 operation and presents no significant problem. Deodorization of the sol- 
 vent is unique from the standpoint of the particular contaminants involved, 
 but numerous standard techniques exist for purification. They include 
 distillation and stripping, adsorption and chemical reaction. Definition of 
 the degree of purity required has been one major problem of solvent deodori- 
 zation because of difficulties in characterizing and quantifying odors in 
 either the solvent or the FPC 
 
 Several recovery systems were designed under contract but no complete 
 recovery system has been adequately evaluated. One glassware system demon- 
 strating a method of alcohol deodorization was constructed, evaluated and 
 products tested. A follow-up design of a more complete^ 90 gallon per hour, 
 pilot plant system was never tested. The EDP design did not utilize this 
 system principle. 
 
 The addition of acid (i.e., phosphoric) to the distillation column, if 
 used in conjunction with proper desolventization design with bleed-off, v;ill 
 adequately resolve the odor problem. 
 
 The additional steps of oil, fines and solubles recovery are primarily 
 economic and are intimately related to waste treatment. Their separation 
 
 1.5-11 
 
and recovery is rather complex and costly. N>'lFS has contributed data on 
 fish oil solubility in alcohol, investigated separation of oil from secondr, 
 stage miscella and has shown that high quality triglycerides can be removed 
 by cooling and centr ifuging. This solubility data was extended to several 
 other solvents and more complex mixed lipids. 
 
 5.3.4 Model Scale Unit (MSU) 
 
 The purpose of the model scale unit (MSU) was to augment bench-level 
 laboratory efforts on a larger scale. The MSU was located in Beltsville, 
 near the College park Laboratory. This facility V'/as designed as an inter- 
 mediate step between the laboratory and the EDP . The MSU was an assembly 
 of pumps, tanks, mixing vessels, a Bird centrifuge, a patterson-Kelley 
 vacuum tumble dryer and a simple packed distillation column constructed of 
 stainless steel. The assembly was mounted on a 10- by 40-foot platform and 
 was capable of various types of batch operations under vacuum or pressure. 
 Most of the system parameters developed in the laboratory were further 
 tested in the MSU. Test samples were eventually produced in the MSU from 
 East and West Coast hake, ocean pout, alewife, Alaskan pollock, Atlantic 
 and Qui f menhaden and even beef (BFC)" Samples were also prepared from 
 various dehydrated products of menhaden: fish meal, presscake, HTM meal, 
 HTM centrifuge cake and flash-dried menhaden. Most engineering data was 
 obtained for hake and menhaden. 
 
 5.3.5 problem Areas 
 
 As indicated by the above discussions, research and development on FPC 
 by IPA extraction can not be considered complete. Some parts of the system 
 have received considerable attention, but more work is necessary in the less 
 extensively studied areas. Moreover, during the course of the program new 
 problems became apparent as unresolved parts of the system were studied. 
 Among the most significant problem.s remaining were those in the areas of 
 lipid recovery and waste treatment. 
 5.3.5.1 Lipid Recovery 
 
 Detailed work on lipid recovery was not conducted during the early 
 phases of the FPC program because it had little economic consequence in 
 processing non-oily species of fish such as hake. However, as increasing 
 consideration was given to processing oily species such as menhaden and 
 anchovy, the recovery of lipids became an important factor in determining 
 the cost associated with the processing of FPC. 
 
 1.5-12 
 
when comminuted fish is contacted with alcohol a large porfion of the 
 oil can be removed by simple physical means. That is, when the fish/lPA 
 slurry is separated into a solid and liquid fraction by either centrifuga- 
 tion or pressing, a separate oil-rich phase may appear in the liquid efflu- 
 ent. Subsequent extractions, howey,er, remove oil from the solids by solu- 
 bilizing it into the alcohol. To recover oil in the IPA process, it then 
 becomes necessary to separate it from the alcohol in the miscella. In 
 addition (because of the mutual solubility of alcohol in oil), residual al- 
 cohol must also be removed from the recovered oil and returned to the proc- 
 ess. Laboratory work showed that a high quality oil can be prepared from 
 fish processed by the IPA process. This oil can be distinctly superior to 
 oil prepared in typical fish meal operations because it has been exposed to 
 less heating and oxidation. Insufficient operating time and problems with 
 existing plant equipment in the EDP, however, prevented further scaled-up 
 testing and economic evaluation on oil recovery processing and characteri- 
 zation of the oil for food or industrial use. 
 5.3.5.2 Waste Effluents 
 
 Only preliminary studies were made on the treatment of waste effluents 
 from an FPC plant. These studies were done in conjunction with actual 
 plant operations at the EDP at Aberdeen, Washington. Laboratory studies 
 were provided in support of EDP efforts in this area. 
 
 The major sources of waste effluents from the plant were identified 
 from the following areas of plant operation: 
 
 (1) Still bottoms from the alcohol recovery distillation unit 
 
 (2) plant cleanup water, particularly that water used to keep the 
 area around the deboner free of fish wastes, etc. 
 
 (3) Waste brine from the fish holding tanks 
 (A) Pvun-off water from the fish conv'eyors 
 
 These effluents were normally discharged untreated into the local 
 municipal waste treatment system, but occasionally caused a serious overload. 
 Laboratory tests showed that the still bottoms were the major source of 
 objectionable materials in the plant waste effluents. This effluent stream 
 sometimes contained about 5.0 percent total solids, of which 40 percent 
 wore classified as lipids. The 5-day biochemical oxygen demand (BOD) of 
 this stream frequently exceeded 80,000 ppm. When such high loads were dis- 
 charged into the municipal sewer system, they responded very poorly to 
 
 1.5-13 
 
primary treatment and clarification, causing serious foaming because of the 
 high phospholipid content of the effluent stream. 
 
 The aqueous and brine effluents depended on the fish processing load 
 and related cleanup water used. They were composed mainly of nitrogenous 
 components with a BOD range from 2,000 to 13,000 ppm. 
 
 Laboratory tests showed that the treatment of still bottoms by classi- 
 cal alum-lime treatments could reduce the BOD load by over 70 percent. The 
 foaming characteristics of the effluents were destroyed by this treatment. 
 
 Other treatments consisted of dissolving small quantities of sodium 
 hexametaphosphate (less than 0.1%) in the aqueous plant effluents and then 
 combining these treated effluents with the acidified still bottoms. This 
 treatment produced a flock (easily removed by centrifugation) that reduced 
 the nitrogenous compounds in the waste stream by 90 percent and the BOD 
 level by 80 percent. 
 
 This waste treatment work was never developed into a demonstrable work- 
 ing unit and further work is indicated. 
 5.3.6 product Modification 
 
 A considerable amount of work has been done under contract and by NMFS 
 on modification of IPA-extracted FPC and has been reported in other sections. 
 While pos tproduction modifications are highly desirable and justified for 
 many applications, more research is needed on relatively simple modifications 
 to tlie basic process to produce, for example, a more wettable, less dusty 
 product. Bleaching can also be accomplished at the same time. Although 
 water dispersion followed by spray drying has been investigated, this neces- 
 sitates a high evaporative load approaching that of the basic process and it 
 is costly. 
 
 5.4 OTHER FPC PROCESSES 
 
 By 1968 the preparation of FPC by various solvent extraction systems 
 (IPA and other solvents) had been evaluated to the point where their virtues 
 and weak points were apparent. The major advantages of solvent processes 
 were that: (1) organoleptically acceptable FPC could be produced in higher 
 yield and therefore at lovvcr projected costs than FPC prepared by biological 
 processes and (2) the nutritional (}uality of the FPC was higher (approximately 
 equivalent to the raw material) than that produced by biological methods. 
 
 1.5-14 
 
The negative aspects of solvent extraction systems were the unsolved 
 problem of effective, efficient oil recovery and the lack of functional 
 properties of the FPC solvent extraction of whole, raw fish. Economics 
 dictated that an efficient process for use with fatty fish should produce 
 both a high quality FPC and a utilizable oil. Functionality was not con- 
 sidered critical for FPC to be used solely as a nutritional supplement. 
 However, the lack of functional properties could be a deterrent to the 
 full utilization and potential usefulness of FPC in various foods. This 
 is particularly true in the U. S. where food manufacturers blend ingredients 
 to complement one another and to achieve the characteristics desired by 
 consumers. 
 
 Although from 1963 on most laboratory research was directed toward 
 the successful development of the IPA process to produce FPC for use as a 
 protein supplement, alternate processes for preparing functional FPCs con- 
 tinued to be investigated. At the laboratory level, biological (enzymatic) 
 and aqueous-solvent processes were primarily studied. Some work was also 
 done on an HTM process, 
 5.4,1 Biological Processes 
 
 Biological processes can be divided into two categories: 
 
 (1) those using enzymes to solubilize (hydrolyze) the proteinaceous 
 tissues of fish in water and thereby releasing the lipid and bone fractions, and 
 
 (2) those using microorganisms to convert the fish protein or lipids 
 for use as a nutrient substrate. 
 
 An attractive feature of the latter method is that selected microorganisms can 
 use the lipids as a carbon source to produce additional amounts of protein. 
 
 During the early phases of work on biological processes, over 400 micro- 
 organisms were screened for their suitability as fermentation agents for the 
 biological production of FPC, Organisms selected for detailed study were 
 those showing high lipolytic activity for converting lipids into nonprotein 
 nitrogen and protein. The results of this study indicated that two strains of 
 Geotrichum candidum and one strain of Candida lipolyticus offered the best 
 possibility for FPC preparation, FPC was prepared by biological processes 
 using the above organisms. Evaluation of these processes indicated conversion 
 of the lipids to protein wa<; not quantitative. With fatty fish, oil conversion 
 often did not exceed 60 percent. 
 
 1.5-15 
 
Freshly prepared FPC made by the biological process generally had good 
 flavor and odor characteristics, but during storage flavor and odor reversion 
 (development of objectionable odors and flavors) were often noted. In addition, 
 research at the Seattle laboratory showed that sterilizing the raw fish prior 
 to inoculation with the lipolytic organisms was difficult, with the resulting 
 possibility of growth of other organisms during the fermentation step. 
 
 A limited amount of work was also done with the organism Rhizopus 
 oligosporus . This organism liberates a powerful antioxidant during its growing 
 cycle. Efforts to produce storage-stable FPCs using this organism, however, 
 were never satisfactorily demonstrated, and work on fermentation processes 
 utilizing microorganisms was terminated in 1971. 
 
 Considerably more success was achieved in preparing FPC by utilizing 
 enzymes to hydrolyze the proteins so that they could be separated from the 
 lipid and bone fractions. Using commercially available proteolytic enzyraes, 
 an extensive study was made on various processing conditions for preparation 
 of FPC. The nutritional characteristics and yields of the FPCs prepared by 
 enzymic processes were generally quite inferior to those prepared by IPA 
 extraction. This was due_ to the loss of tryptophan and histidine that were 
 bound in the insoluble sludges formed by enzymic methods. However, an alkaline 
 protease of Bacillus subtilis did release sufficient tryptophan and histidine 
 into the resulting hydrolysate (protein fraction) so that the protein efficiency 
 ratios (PERs) of the FPC were equivalent to casein. 
 
 FPCs prepared by enzymic methods, while being soluble, were not bland 
 because they are composed of amino acids and low molecular weight peptides. 
 Although this characteristic would not hamper their use as a nutritional 
 Supplement in certain types of foods such as soups and gravies, their use in 
 many other types of foods would be restricted. 
 5,4.2 Aqueous-Solvent Processes 
 
 In 1968, the Seattle laboratory began investigations on a promising 
 process that used water washes on the insolubilized protein to remove the 
 major portion of the lipids followed by solvent to remove residual lipids 
 to produce an organoleptically stable FPC. This process utilized sodium 
 hexametaphosphate to insolubilize the fish proteins and thus prevent the 
 loss of protein during the water extraction step. Residual lipids were 
 removed by extracting the insoluble proteins (presscake) with IPA. FPC 
 
 1.5-16 
 
prepared by this method was organoleptically stable, nutritionally equiva- 
 lent to FPC prepared by the IPA process and displayed several functional 
 properties. In addition, oil recovery presented no problems because it was 
 recovered from the aqueous washes employing currently available technology. 
 A public service patent was granted on the process in August 1971. More 
 recent work in this area showed that the presscake prepared by the fore- 
 going process could be stabilized against microbiological and oxidative 
 deterioration for 30 days at 20*^C. Stabilization of the presscake was ac- 
 complished by the addition of sodium metabisulf ite and a mixture of anti- 
 oxidants (propyl gallate and butylated hydroxytoluene) during processing. 
 The stored presscake could then be processed into FPC with acceptable or- 
 ganoleptic properties and no loss of nutritional properties. 
 5.4.3 Other processes 
 
 The heat transfer method (HTM) of vacuum drying a comminuted slurry of 
 fish in fish oil followed by various degrees of oil removal by centrifuga- 
 tion and pressing has been patented and commercially developed. Research 
 into its applicability as a preprocessing step was begun at NMFS but was 
 discontinued because of other priorities. Additional biological or solvent 
 extraction processing would be necessary to produce an acceptable FPC. 
 
 Extraction research with different solvents, mixed solvents, use of 
 cooked fish, pH modification and prewashing did not indicate any signifi- 
 cant improvements that would justify modifications to the basic IPA process. 
 Later work with low temperature extraction of lean fish with isopropyl al- 
 cohol shov;od im.provements in functionality of the FPC but neu methods of 
 low temperature desolventizing had not been satisfactorily developed. Other 
 priorities replaced this work. 
 
 1.5-17 
 
SECTION 6 
 
 PRODUCTION PROCESSES 
 
 6.1 CONCLUSIONS AND RECOMMENDATIONS 
 
 The general conclusions that can be drauTi regarding the technical and 
 economic feasibility of producing FPC are based in part on work done at the 
 Experiment and Demonstration plant (EDP), built under contract, by NMFS at 
 Aberdeen, Washington. The purpose of the EDP was to test extraction of fish 
 by isopropyl alcohol under various engineering conditions on a semiworks 
 scale. There wore several objectives envisioned for EDP operations; a num- 
 ber of these were fulfilled in the course of these operations. 
 6.1.1 Accomplishments of the EDP 
 
 ' One of the primary objectives of the plant was to demonstrate the en- 
 
 IPA extraction of fish (originally lean fish — pacific hake--although fatty 
 species, menhaden and anchovy, were also used). The counterrurrent extrac- 
 tion process had been developed and tested at bench level and at the model- 
 scale unit level. This process was brought up to semix-;orks at the EDP and 
 was proven feasible. 
 
 Specific accomplishments of EDP operations were as follows: 
 
 (1) Continuous, multistage extraction was effectively demonstrated 
 at the semiworks scale. Although there were repeated inter- 
 ruptions in operations caused by technical problems, these 
 indicated and emphasized the necessity for a high degree of 
 reliability and capability of controlling the subprocesses 
 and components of the total process. 
 
 (2) The EDP was able to produce sizable quantities of FPC of a high 
 quality, meeting specifications set forth by the Food and 
 
 Drug Administration. Although not as much FPC was produced as 
 had been originally anticipated, the ED? was- able to provide 
 
 1.6-1 
 
enough of the product for use m large-scale feeding and product 
 evaluation studies both domestically and abroac^. 
 
 (3) Despite the sometimes erratic operation of the plant, useful 
 scale-up data was obtained for expanding operations to a 
 commercial scal^. 
 
 (4) Many production problems were identified; most of these were 
 resolved or significant insights acquired toward their reso- 
 lution. 
 
 (5) Although little concrete economic data was obtained which would 
 be directly applicable to commercial operation, a better per- 
 spective was achieved in this area. 
 
 Neither economic nor engineering design data are sufficiently complete 
 for the design and construction of a large FPC plant. If this is desired, 
 then these data must be obtained at the pilot plant level. There are no 
 insurmountable technical problems. 
 
 The two most critical problems remaining in the process desigif are in 
 the desolventizing and waste treatment systems. If a high fat fish is to 
 be processed and high quality oil to be produced, then more pilot plant work , 
 is required in this area also. Many ffeel that neither the present nor possi- 
 ble future value of oil warrants the additional capital investment that might 
 be required in a commercial plant. Less critical problems exist in the areas 
 of fish storage, centrifugal decanter sizing, mill selection and sizing, and 
 solvent recovery system design (including solvent deodorization) . These 
 could be reasonably overcome by overdesign. 
 6.1.2 Reconimendations 
 
 The recommended approach to the design and construction of a large com- 
 mercial plant would be to design and construct a prototype pilot plant to 
 process 300 to 600 pounds of fish per hour and to operate this at the desired 
 location on the specific species of fish to be used. All major design prob- 
 lems could then be reasonably resolved to permit the design of a large com- 
 mercial scale plant. 
 
 Heretofore most considerations have been given to eventually building a 
 large FPC plant capable of processing hundreds of tons of fish into a low- 
 cost product daily. This approach would permit the supplying of significant 
 quantities of high quality animal protein to supplement human nutritional 
 
 1.6-2 
 
deficiencies. Alternatively, it is recommended that consideration be given 
 to the construction of a moderate-sized commercial plant (possibly batch), 
 capable of producing several tons of FPC per day and operating in conjunc- 
 tion with a large, food-fish processing plant, to produce FPC for use in 
 higher value speciality markets such as health foods, medically controlled 
 diets and pharmaceuticals. The associated food-fish plant would be capable 
 of supplying any amount of fish required for this operation as well as main- 
 tenance, supervision, laboratory control, by-product and waste processing 
 and disposal, product storage, shipping and marketing. It is believed that 
 this approach could result in a profitable operation, be self-supporting 
 and supply a basis and information for future expansion. 
 
 6.2 ED? PRODUCTION 
 
 6.2.1 Background of the EDP 
 
 6.2.1.1 Establishment 
 
 The construction and operation of a semiworks experiment and demonstra- 
 tion plant was an important step in a logical sequence of development toward 
 bringing FPC to production on a commercial scale. The EDP was the testing 
 ground for further evaluation of the various subprocesses of FPC production 
 established by the laboratory and MSU work. 
 
 In 1965 the National Academy of Sciences passed a resolution recommend- 
 ing that the Federal Government expand the research work on FPC by establish- 
 ing a suitable pilot plant facility, capable of producing 10 tons of FPC per 
 day. This resolution reflected a high level of interest by various compo- 
 nents of the fishing industry and other food interests. Congress favored 
 governmental sponsorship of such a plant to reduce private investment risk, 
 to develop a market for FPC and to provide adequate quantities of FPC for 
 human feeding studies. Accordingly, the EDP was authorized by Public Law 
 89-701, which was signed into law on November 2, 1966. The stated objec- 
 tive of PL 89-701 was to develop practical and economical means of FPC 
 production for use by the U.S. fishing industry. 
 
 Following the tentative selection of a contractor (Ocean Harvesters, 
 Inc.) for the design and construction of a semiworks plant, it became evi- 
 dent that the financial allocations of PL 89-701 were inadequate. An amend- 
 ment, PL 90-549, secured ir^ October 1968, reduced the number of plants from 
 tv>/o to one and diverted available funds for construction and operation of that 
 plant. The contract was then let to Ocean Harvesters, Inc. 
 
 T.6-3 
 
prior to the design and construction of a serniworks scale plant a pilot 
 plant was originally planned by NMFS to further refine the laboratory proc- 
 esses, plans were completed in 1967 for building such a plant, referred to 
 as the modified model scale unit (MMSU) , to accommodate up to 300 pounds of 
 raw material per hour. Contracts were let to design and build the MMSU and . 
 to develop an engineering economic model to assist in evaluating the capital 
 investment and processing costs. However, work on the MMSU was halted prior 
 to completion, based on a decision that such a pilot plant was an unnecessary 
 expenditure of time and money in light of the impending EDP. All inhouse 
 engineering effort was subsequently devoted to evaluation of the contractor's 
 design and operation of the EDP. 
 
 plant operations commenced at the EDP in March, 1971 and continued for 
 some 14 months with various interruptions. During this period much of the 
 engineering and analytical work in the laboratory was focused toward solving 
 production problems which arose in the EDP, developing modified processes 
 and adapting the established lean fish process to accommodate fatty fish. 
 6.2.1.2 Design 
 
 The EDP was originally designed for the IPA extraction of FPC from lean 
 fish. Subsequent modifications were necessary to permit limited experimental 
 fatty fish processing. 
 
 The proposed preliminary process design of the contractor deviated con- 
 siderably from that of NMFS (at the time, the Bureau of Commercial Fisheries), 
 The NMFS process design was based on a continuous-flow, multistage extraction 
 process using centrifugal decanters as the liquid-solids separator at each 
 stage. This was the method used in the MSU to prepare samples and conduct 
 process studies. The contractor proposed and demonstrated the use of vi- 
 brating screens and presses as a means of liquid-solids separation. The 
 contractor's design was accepted as representing a lower potential capital 
 investment and possible time savings in plant construction. 
 
 The design and operation phases of the Ocean Harvesters, Inc., contract 
 were : 
 
 (1) Batch pilot plant process development 
 
 (2) process design 
 
 (3) plant design 
 ('0 Construction 
 
 I.6-A 
 
(5) Start-up 
 
 (6) Lean fish (hake) processing operations 
 
 (7) Modifications 
 
 (8) Fatty fish (menhaden and anchovy) processing operations 
 
 (9) Shut-down 
 
 6.2.1.3 Objectives of the EDP 
 
 Major objectives to be fulfilled by successful operation of the EDP 
 were envisioned as follows: 
 
 (1) Demonstration and evaluation of a continuous multistage process 
 for FPC (using IPA) at the semiworks scale. 
 
 (2) production of sizable quantities of FPC (for use in human feeding 
 studies) meeting FDA specifications and applicable standards of 
 other countries. 
 
 (3) Stimulation of industry participation by development of engineer- 
 ing scale-up data, including economic estimates, for use by in- 
 dustry in the design and construction of commercial-scale FPC 
 production plants. 
 
 (A) Identification and resolution of processing problems not evident 
 at laboratory -or pilot plant level; function as a testing ground 
 for process modifications. 
 (5) provision of personnel training and experience for possible fu- 
 ture demonstration plants, domestic and foreign. 
 6.2.2 EDP processing 
 6.2.2.1 pre-design Research 
 
 The basis for FPC production in the EDP was provided by the several years 
 of product research and evaluation at the bench and pilot plant scale, both 
 by NMFS and by contractors. At the time of contract award to Ocean Harvesters 
 for the design, construction and operation of the EDP, the solvent extraction 
 process had not been adequately defined. Various subprocesses required fur- 
 ther work for scale-up to a semiworks plant. Ocean Harvesters conducted 
 further pre-plant design development work in several subprocesses, including 
 comminution, deboning, screen/press separation of liquids and solids, de- 
 solventizing, milling and solvent recovery. 
 
 In addition to producing data necessary to scale-up the process design, 
 another objective of the above work was the evaluation of various types of 
 equipment for potential plant use. 
 
 1.6-5 
 
6.2.2.2 FPC Subprocesses 
 
 The following descriptions detail the stops involved in processing fish 
 into FPC in the EDP. 
 
 6.2.2.2.1 Fish procurement 
 
 Most of the fish processed in the plant were obtained by contract 
 from three trawlers equipped with midwater trawling gear and refrigerated 
 sea water equipment for holding the fish prior to delivery. Additional fish, 
 
 either iced or frozen, were shipped in by truck for special purposes of equip- 
 ment testing or to supply fish not locally available. 
 
 6.2.2.2.2 Fish Unloading, Storage and Refrigeration 
 
 Fresh fish unloading was accomplished by a vacuum system commonly used 
 by fish reduction plants. The fish were rinsed with fresh water, inspected 
 by a U.S. Government fish inspector, measured volumetrically and conveyed 
 to storage. 
 
 Ocean Harvesters designed a brine storage system for fresh fish. This 
 storage system required a subsystem for making brine, refrigeration and fish 
 storage tanks . 
 
 6.2.2.2.3 Fish preparation 
 
 Deboning of some species of fish was necessary in order to reduce the 
 fluoride content of the FPC Although bone may be removed before or after 
 extraction, the greatest success had been obtained by use of meat/bone sepa- 
 rators on fresh fish. This deboning approach was selected and resulted in 
 sufficient comminution. 
 
 6.2.2.2.4 Extraction 
 
 Extraction was accomplished in the EDP by contacting the comminuted 
 fish with hot azeotropic isopropyl alcohol for up to a maximum of thirty 
 minutes in a jacketed vessel with a turbine agitator. Separation of the 
 solids from the liquid was accomplished by pumping the hot mixture of fish 
 and alcohol over a fine mesh, vibrating screen. The wet solids were further 
 reduced in liquid content by pressing before being discharged to the next 
 stage of extraction or to desolventizing. Only three stages of counter- 
 current extraction were required when using a 2:1 soivent-to-f ish ratio 
 for fish containing less than 6 percent lipids. Many operating problems 
 were encountered with the use of screens and presses (see Section 6.3). A 
 centrifugal decanter was later installed for evaluation. 
 
 1.6-6 
 
6.2.2.2.5 Do s ol von ti zinc/ Drying 
 
 Desolventization was achieved in the EUP in a four-stage, continuous, 
 
 , »» 
 
 jacketed dryer. Steam was introduced in the tirst and second stages counter- 
 current to the flow of solids. A small amount of air was introduced counter- 
 current to the solids in the third and fourth stages. Considerable difficul- 
 ties were experienced in consistantly removing residual traces of solvent 
 to below 250 ppm. 
 
 6.2.2.2.6 Milling 
 
 Following the desolventization phase, the dried solid concentrate was 
 milled by an impact mill with an internal classifier and a product collector. 
 Modifications to this mill were necessary to effect the desired particle 
 size. Moisture content of the solid concentrate had to be kept below 7 per- 
 cent to avoid jamming or overloading of the mill. 
 
 6.2.2.2.7 Solvent and Oil Recovery 
 
 The solvent effluent (miscella) contains alcohol, water, lipids, soluble 
 protein, salts and fine suspended fish solids. These fine suspended solids 
 were removed by centrifugation to prevent excessive fouling and foaming in 
 the distillation column. The fines thus separated (sludge) contained a high 
 percentage of lipids and solvent. Solvent was recovered from the sludge by 
 stripping with steam and the fines discarded as waste. 
 
 Alcohol was recovered by distillation and concentrated to approximately 
 90 percent by volume (azeotrope is 91% by volume). The remaining water, solu- 
 bles and lipids, from the bottom of the still, were discarded. 
 
 When high fat content fish such as anchovy or menhaden were processed, 
 the miscella consisted of three phases: oil (containing solvent), alcohol 
 (containing proteins, lipids, minerals and water), and dispersed solids (con- 
 sisting of protein and lipids). A three-phase centrifuge was used to clarify 
 the miscella, remove an oily phase and remove an oily sludge. Because of 
 the high alcohol content of the miscella, the oil phase was the heavier of 
 the two liquid phases. Oil was salvaged for sale but the oily sludge was 
 wasted. Additional development work is necessary in this area in order to 
 design an efficient recovery system. 
 
 6.2.2.2.8 Waste Water Treatment 
 
 Waste water effluents from the plant consisted of waste brine, wash 
 water from fish and equipment, aqueous sludge from th? centrifuge after 
 steam stripping to remove alcohol and aqueous effluent from the bottom of 
 
 1.6-7 
 
the still. These effluents were screened (tangential and vibratjing screens), 
 skimmed of oil and light solids in a skimming tank and finally subjected to 
 a dissolved air flotation system. This treatment removed considerable soli4s 
 and oil from the effluent but had "no significant effect on dissolyed solids and 
 little impact on BOD because of the high amount of dissolved organic solids.. 
 
 6.3 PROBLEM AREAS /RECOMMENDED WORK 
 
 During the operational stages of the EDP (March 1971 through May 1972) 
 various operating and process design deficiencies were encountered. Some 
 plant modifications were made during that period both to improve operation 
 and to modify the plant to permit processing of fatty fish. It became ob- 
 vious that extensive modification of the plant and process were necessary 
 in order to achieve extended, trouble-free operation and production of an 
 acceptable product. 
 
 During the summer of 1972, NMFS contracted with a private engineering 
 firm, p/E Development Company (PEDCO), to assist in a review and evaluation 
 of the plant and to develop specifications and cost estimates to modify the 
 plant for further operation and processing of high fat fish. 
 
 The following dlGCuGGicn highlights seme cf the major problems identi- 
 fied by PEDCO and offers several proposed solutions. 
 
 6.3.1 Storage 
 
 The storage system of holding hake in chilled brine was only partially 
 satisfactory. The fish handling system proved to be too destructive of the 
 soft-fleshed fish, resulting in leaching of protein into the brine. Protein 
 losses were significant after 2- or 3-days-. storage. The high fat and protein 
 content of the used brine also presented a serious disposal problem. 
 
 Freezing, although not economical for commercial application, was rec- 
 ommended during further operation of the EDP on menhaden from the Gulf. 
 
 Alternative methods could be dry refrigerated storage or preprocessing, 
 such as partial dehydration by cooking and pressing followed by temporary 
 preservation with alcohol. More investigation was required. 
 
 6.3.2 Fish preparation 
 
 Deboning of fresh or thawed, chopped or whole fish was not completely 
 resolved in the EDP. None of the deboners tested in the EDP demonstrated 
 adequate reliability, efficiency of separation or sanitation. New deboners 
 have been reported since.' EDP shutdown; these need to be subjected to pro- 
 duction testing. 
 
 1.6-8 
 
Dry deboning methods offer an alternative, but, again, more investi- 
 gation is needed before a valid evaluation can be made of such processes. 
 6.3.3 Extraction 
 
 Extraction in the EDP was a complex, multistage system with many in- 
 terdependent unit operations. FPC production would be greatly expedited 
 by combining these operations into a single countercurrent extractor. Such 
 a system would resolve many of the problems identified in the extraction 
 subprocess. 
 
 Objectives in FPC extraction which were not achieved during EDP opera- 
 tions were: 
 
 (1) The optimum range of time for all stages of extraction was not 
 determined for all species likely to be processed. Extraction 
 contact times would probably be best investigated at pilot plant 
 level, since the time factor will vary with the size of the plant. 
 The experimental plants should be designed so that contact times 
 
 can be varied up or down at least 50 percent from the best estimates. 
 
 (2) Optimum removal of free oil in the first stages was not achieved. 
 Extraction efficiency will be improved if free oil can be removed 
 more completely during the first stage. One method of removal 
 
 was given limited laboratory testing: a dip weir ring was installed 
 just ahead of the liquid discharge from the decanter centrifuge 
 bowl to direct the removal of liquid from the bottom of the cylin- 
 drical pool in the decanter bowl. Since free oil is the heavier 
 of the two liquid phases of the miscella, it should be removed 
 preferentially. No improvement was evident when the weir ring vias 
 used in a small laboratory machine, and no comparative runs with the 
 weir ring were made in the EDP. The method is promising, but re- 
 quires more complete testing. Other means of free oil removal 
 should be sought as well . 
 
 (3) Optimum interstage separation of finely divided solids from the 
 miscella was not achieved. The use of screens and presses in the 
 EDP proved unsatisfactory because of a high degree of fines loss 
 through the screens and frequency of screen fouling and breakage. 
 Both the NMFS and the PEOCO follow-up studies recommended replace- 
 ment of screen and ptress separators with decanter centrifuges. 
 The centrifuges proved superior in separating fine solids from 
 
 1.6-9 
 
miscella both at laboratory and pilot plant level. However, even 
 with the decanter added to the first stage in the EDP, from 0.2 to 
 1.0 percent of solids by volume remained in the first miscella. 
 This level would have to be reduced. 
 (A) Interstage slurry transfer and regulation needed improvement. The 
 interstage slurry transfer pumps v^/ith the recycle controls, as 
 recommended in the PEDCO study, should do the job of level control- 
 ling satisfactorily, but the system is both costly and complex. 
 Efforts should be made to develop an alternative approach. 
 6.3.4 Miscella Clarification and Oil Recovery 
 
 These is an intrinsic limitation in the EDP method of oi.l removal from 
 miscella: in centrifugal separation of an oil phase from the solvent phase^ 
 the zone of operation is narrow with high oil loss on either side. If the 
 IPA content is increased, the solubility of oil in the miscella will also 
 increase, and the centrifuge can remove this dissolved oil. If the IPA 
 content is decreased, the specific gravity of the miscella will become too 
 close to that of the fish oil phase and the centrifuge will again fail to ef- 
 fect separation. Even the best operations in the EDP resulted in oil residuals 
 in the miscella of 0.25 to 0.5 percent by volume. This oil uLt.'atus> problerfis 
 in the still, as well as waste disposal difficulties. The oil itself is 
 degraded by the heat treatment in the IPA still . 
 
 The problem of specific gravity was brought under control, in the en- 
 gineering review study, by using excess IPA and a small amount of water. 
 Adding a large excess of water to convert the miscella to the heavier liquid 
 phase would solve many problems, but, this would be too costly in terms of 
 additional load on the still. 
 
 Another approach could possibly be better and needs to be investigated. 
 The oil could be removed by centrifugation of a liquid stream taken from the 
 second or third plate belov^; the column feed in the still. The exposure to 
 column temperatures may be short enough to avoid degradation of the oil and 
 oil would definitely be kept out of the reboiler isfhere most damage is done. 
 There would also be some stripping of IPA out of the oil in such a system. 
 
 The removal of IPA from the oil produced by methods used in the labora- 
 tory has as yet to be accomplished in a pilot plant. A countorcurrent wash 
 with water has been successful at bench scale in the laboratory, but it has 
 not been tried in pilot scale equipment or in a continuous manner. 
 
 1.6-10 
 
6.3.5 DGSolventizing and Drying 
 
 The drying operation at the EDP was an unstable source of too much off- 
 standard product. These sources of instability would be corrected by niodi^* 
 fications recommended in the PEDCO review, but the revised system has yet 
 to be tested. At this point, the remaining problems are of an engineering 
 nature and very little will be gained from further laboratory testing of the 
 current techniques. 
 
 6.3.6 Dry Milling 
 
 Dry milling requires further design. In the EDP operation, the FPC was 
 not coming out of the process at a constant moisture content. The energy of 
 the impact mill overdries the solids duri.ng mil.ling, whereas feeding the mill 
 wetter material to compensate this only jams the mill. Moisture adjustment 
 of the milled solids or changes in milling equipment are required. 
 
 6.3.7 Solvent Recovery 
 
 There are many divergent viewpoints on solvent recovery which have not 
 been resolved. The major controversies involve oil recovery, odor removal 
 and column control. The control problems are not too serious, since there 
 is plenty of expertise available from the petroleum derivative industry to 
 Ldckle that dLed., Ouul LtiiiiuvcLl iLuiii IPA iiicty uuu evtii bt; utiCtiSsaLy. D'-^ijI" 
 ventizing as performed in the EDP deodorized the product even though the 
 recovered IPA was heavily laden with odor compounds. The changes in the 
 dryers and strippers as recommended by tlie engineering review study may trap 
 and recycle these odor compounds back into the process. This can be answered 
 only through actual pilot operation on a continuous basis. 
 
 6.3.8 Waste Water Treatment 
 
 A satisfactory solution to the treatment of plant effluents was not 
 accomplished during the operation of the EDP. The effluents from a plant 
 modified according to the engineering review study would be greatly reduced 
 in volume and biological oxygen demand. Whenever a plant is built in this 
 country, the effluents must be under control from the start. One of the most 
 practical alternatives to resolving the remaining problems in this area would 
 be to build a pilot plant small enough that its wastes can be tolerated in a 
 local sewer system, then develop the most economical treatment system to make 
 the effluents acceptable in public waterways. 
 
 1.6-11 
 
m^S FPC 
 
 
Z3Z 
 
 
 o- 
 
 
 «raC 
 
 
 m 
 
 hJ 
 
 
 ^ 
 
 — 1 
 
 g 
 
 C3 
 
 w 
 
 r/i 
 
 o 
 
 CO 
 
 <-3 
 
 
 
 
 
 
 
 
 
 
 
 «^ 
 
 
 
 
 
 
 
 
 
 
 
 rc: 
 
 
 
 
 
 
 
 
 
 
 
 1 — ' 
 
 CI 
 
 
 
 
 
 
 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 
 ** 
 
 
 
 
 
 
 
 
 
 
 
 t5 
 
 
 
 
 
 
 
 
 
 
 
 LaJ 
 
 
 
 
 
 
 
 
 
 
 
 CD 
 
 
 
 
 
 X 
 
 
 
 
 
 
 CJ 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 o 
 
 •r-l 
 
 
 
 
 
 
 
 
 
 ^ 
 
 ^ 
 
 4-lf 
 
 
 ^ 
 
 
 
 ^ 
 
 
 M 
 
 
 M 
 
 M 
 
 H-l 
 
 
 u 
 
 
 
 p 
 
 
 u 
 
 as 
 
 td 
 
 m 
 
 o 
 
 
 to 
 
 
 
 td 
 
 
 10 
 
 CO 
 
 PUI 
 
 P4 
 
 
 
 fU 
 
 
 
 P4 
 
 
 PM 
 
 f— 
 
 
 
 c 
 
 
 
 
 
 
 
 
 «« 
 
 <u 
 
 <u 
 
 o 
 
 
 0) 
 
 
 
 0) 
 
 
 <U 
 
 £J> 
 
 CO 
 
 60 
 
 •H 
 
 
 60 
 
 
 
 60 
 
 
 60 
 
 CD 
 
 (U 
 
 (U 
 
 to 
 
 
 <J 
 
 
 
 <u 
 
 
 <U 
 
 
 1-4 
 
 t-l 
 
 •r-l 
 
 
 I-l 
 
 
 
 •H 
 
 
 1-1 
 
 
 r-l 
 
 r-4 
 
 > 
 
 
 r-l 
 
 
 
 i-l 
 
 
 1-1 
 
 
 O 
 
 O 
 
 •rH 
 
 
 o 
 
 
 
 O 
 
 
 o 
 
 
 u 
 
 CJ 
 
 Q 
 
 
 o 
 
 
 
 O 
 
 
 o 
 
 
 
 
 
 
 
 
 
 "rt 
 
 
 
 
 « 
 
 
 
 
 
 
 
 
 u 
 
 
 
 
 o 
 
 
 
 
 
 
 
 
 •rl 
 
 
 
 
 •r^ 
 
 
 
 
 
 
 
 
 V4 
 
 
 
 
 U 
 
 
 
 
 
 
 
 
 2 
 
 
 
 3: 
 
 1 
 
 o 
 
 
 
 
 
 
 
 1 
 
 
 ^ 
 
 f— 
 
 
 •r-l 
 
 
 
 
 
 
 
 >M 
 
 
 to 
 
 -t: 
 
 ■H 
 
 U 
 
 
 
 
 
 
 
 o 
 
 
 o 
 
 tj 
 
 O 
 
 •r^ 
 
 
 
 
 
 
 
 
 
 •r4 
 
 —J 
 
 
 U 
 
 
 
 
 
 
 
 to 
 
 
 60 
 
 CO 
 
 CO 
 
 t-> 
 
 
 
 
 
 
 
 0) 
 
 C 
 
 O 
 
 =3 
 
 0) 
 
 3 
 
 
 
 
 
 
 
 4-1 
 
 o 
 
 /-I 
 
 O. 
 
 
 K 
 
 
 
 
 'c 
 
 
 
 CO 
 
 1 
 
 o 
 
 •r4 
 
 
 4-) 
 
 -a 
 
 
 
 
 •H 
 
 
 
 CO 
 
 o 
 
 O 
 
 
 CO 
 
 
 
 
 
 tu 
 
 4J 
 
 
 
 •xi 
 
 hJ 
 
 O 
 
 CO 
 
 
 X) 
 
 
 
 
 
 o 
 
 
 
 (U 
 
 ♦ • 
 
 
 
 0) 
 
 j:: 
 
 
 
 
 u 
 
 
 
 u 
 
 •a 
 
 •» 
 
 
 4J 
 
 to 
 
 
 
 
 fu 
 
 
 
 •H 
 
 u 
 
 1-4 
 
 
 •H 
 
 •H 
 
 
 
 
 
 
 
 c 
 
 ►^ 
 
 CO 
 
 
 C 
 
 P4 
 
 
 
 
 r-4 
 
 
 
 t3 
 
 
 o 
 
 
 3 
 
 c 
 
 
 
 
 e 
 
 
 
 UJ 
 
 to 
 
 t-l 
 
 4J 
 
 
 0) 
 
 o 
 
 
 
 
 •H ■ 
 
 
 
 x: 
 
 ^ 
 
 •H 
 
 
 J_ 
 
 
 
 
 
 C 
 
 
 
 4-) 
 
 O 
 
 1-4 
 
 
 r-l 
 
 u 
 
 tU CM 
 
 
 
 
 < 
 
 
 
 
 o 
 
 o 
 
 
 VD 
 
 
 ' O vO 
 
 
 
 
 
 
 
 c 
 
 M 
 
 A4 
 
 
 C\ 
 
 c 
 
 a o^ 
 
 c 
 
 
 
 fs 
 
 
 
 •H 
 
 ^^ 
 
 ' 
 
 
 r-l 
 
 •r-l 
 
 0) 1-1 
 
 o 
 
 
 
 4J 
 
 
 
 
 
 •* 
 
 
 
 >-l 
 
 •r^ 
 
 
 
 •H 
 
 
 
 to 
 
 • 
 
 o 
 
 
 (0 
 
 0) 
 
 u 
 
 
 
 ■-1 
 
 
 
 iJ 
 
 to 
 
 •r4 
 
 
 <-) 
 
 "4-1 
 
 a 
 
 
 
 to 
 
 
 
 c 
 
 3 
 
 e ^ 
 
 CO 
 
 C 
 
 c 
 
 <o 
 
 
 
 3 
 
 
 
 (U 
 
 <u 
 
 O CO 
 
 
 <u 
 
 o 
 
 > 
 
 
 
 C 
 
 
 lO 
 
 E 
 
 •z 
 
 C b^ 
 
 ccT 
 
 g 
 
 CJ 
 
 c 
 
 
 
 
 
 1 
 
 cu 
 
 
 O IS 
 
 CD 
 
 & 
 
 
 o 
 
 
 
 j: 
 
 
 1-1 
 
 o 
 
 • 
 
 o g 
 
 
 o 
 
 f-< 
 
 u 
 
 
 
 60 
 
 
 
 1-1 
 
 rC 
 
 w ^ 
 
 r-l 
 
 to 
 
 
 
 
 1-1 
 
 
 • • 
 
 <u 
 
 to 
 
 
 «cX 
 
 O 
 
 c 
 
 H ■ 
 
 
 to 
 
 K 
 
 
 <!■ 
 
 > 
 
 t-l 
 
 • u 
 
 
 > 
 
 o 
 
 g 
 
 
 to 
 
 
 
 CSJ 
 
 tu 
 
 '^ 
 
 <u 
 
 
 <u 
 
 •rH 
 
 
 <u 
 
 <: 
 
 
 
 Q 
 
 
 O M X4 
 
 
 o 
 
 4J 
 
 
 
 V4 
 
 
 
 . *» 
 
 
 *» 
 
 
 
 to 
 
 t-< 
 
 
 60 
 
 1 
 
 
 3 
 
 f-l 
 
 <t 
 
 
 I-l 
 
 c 
 
 ■to 
 
 — 
 
 c 
 
 
 
 0) 
 
 to 
 
 fO 
 
 r-l to V 
 
 
 to 
 
 >-l 
 
 3 
 
 "<U 
 
 o 
 
 0) 
 
 
 -H 
 
 o 
 
 <J- 
 
 fe a, ^ 
 
 
 u 
 
 <u 
 
 (U 
 
 c 
 
 u 
 
 o 
 
 4J 
 
 
 > 
 
 T^ 
 
 1 
 
 -H 
 
 
 •r-l 
 
 J.J 
 
 u 
 
 c 
 
 to 
 
 
 to 
 
 
 4) 
 
 60 
 
 CM 
 
 -C • 7-^ 
 
 
 « 
 
 c 
 
 to 
 
 < 
 
 u 
 
 • 
 
 u 
 
 
 «; 
 
 O 
 
 n 
 
 to eii 
 
 
 o 
 
 M 
 
 ^1 
 
 
 iJ 
 
 01 
 
 4-1 
 
 
 
 1-1 
 
 <r 
 
 •rl 
 
 
 r-l 
 
 ^-s 
 
 u 
 
 /-N j: 
 
 c 
 
 
 c 
 
 y^N 
 
 to 
 
 O 
 
 /'^ 
 
 f^ • ^ 
 
 
 o 
 
 to O 
 
 C 
 
 C/5 U 
 
 tu 
 
 • 
 
 o 
 
 CO 
 
 ti) 
 
 c 
 
 CO 
 
 -"2 
 
 
 c 
 
 
 0) 
 
 1- 
 
 o 
 
 :=> 
 
 ') 
 
 b 
 
 •r4 
 
 x: 
 
 b *,: 
 
 
 a 
 
 o 
 
 c 
 
 c 
 o 
 
 n 
 
 c 
 o 
 
 1 
 
 
 o 
 
 1 § 
 
 o .« ^ 
 
 UJ 
 
 pi 
 
 OJ 
 
 v-/ -a 
 
 o 
 
 V— ' 4J 
 
 u 
 
 0) 
 
 o 
 
 *^^ 
 
 X. 
 
 H 
 
 ^^ -H 
 
 <ii ^ 1< 
 
 8h 
 
 OJ 
 
 u 
 
 to 
 
 
 to 
 
 
 
 to 
 
 
 4-> 
 
 o "n "^ 
 
 
 (J i-> 
 
 
 u 
 
 c 
 
 D 
 
 c 
 
 u 
 
 •r^ 
 
 1 
 
 ^4 •H 
 
 c 
 
 p= 
 
 1 
 
 a) c 
 
 c 
 
 Oi C 
 
 •rJ 
 
 o 
 
 •rH 
 
 OJ 
 
 fl4 
 
 
 0) I-l 
 
 CO "'(U 
 
 
 
 
 w <u 
 
 •r-l 
 
 to 0) 
 
 <1» 
 
 p^ 
 
 'OJ 
 
 to 
 
 
 M 
 
 to 4-1 
 
 4J 1-1 U 
 
 
 
 Vj 
 
 •r^ to 
 
 0) 
 
 •r-l > 
 
 4J 
 
 
 iJ 
 
 •H 
 
 r-l 
 
 3 
 
 •r4 3 
 
 V4 ^ U 
 
 
 
 D 
 
 1-4 0) 
 
 4J 
 
 U -r^ 
 
 
 
 r^ 
 
 o 
 
 U 
 
 to 
 
 o 
 
 ^ Si 
 
 o o tu 
 
 
 
 O 
 
 to Vj 
 
 o 
 
 to 60 
 
 u 
 
 to 
 
 u 
 
 to 
 
 •r^ 
 
 rH 
 
 to 
 
 Du C N 
 
 
 
 —1 
 
 (^ a. 
 
 )^ 
 
 PU 
 
 cu 
 
 00 
 
 PLi 
 
 Pu 
 
 o 
 
 b4 
 
 &4 C 
 
 S « fl 
 
 
 
 Ci^ 
 
 
 fi^ 
 
 x: 
 
 
 ,-s C 
 
 
 
 »-l 
 
 
 •r-l 
 
 M O 
 
 
 
 
 • u 
 
 
 • o 
 
 .c 
 
 W -H 
 
 J= 
 
 • 
 
 (U 
 
 x: 
 
 • 
 
 • 
 
 
 
 ^ 
 
 «: o 
 
 X- 
 
 ttJ (U 
 
 to 
 
 is to 
 
 to 
 
 Di 
 
 g 
 
 •to 
 
 a! x: 
 
 <u S jc 
 
 
 
 </> 
 
 ■ a. 
 
 U) 
 
 (1) 
 
 •r-l 
 
 •r^ 
 
 
 g 
 
 •r-i 
 
 to 
 
 S . -IS 
 
 
 
 ■r^ 
 
 • <a 
 
 •H 
 
 • D- 
 
 u. 
 
 U^ 
 
 • 
 
 o 
 
 u. 
 
 • •r-l 
 
 
 
 ■j^ 
 
 U P-i 
 
 fl4 
 
 W CO 
 
 
 ^-^ ^ 
 
 - 
 
 U 
 
 o 
 
 
 W Ci4 
 
 = (J fll 
 
 - - . 
 
 cd 
 
 
 
 
 
 
 
 
 
 
 
 ac 
 
 
 
 
 
 
 
 
 
 
 
 
 r-< 
 
 N 
 
 fo 
 
 
 <t 
 
 
 
 «o 
 
 
 vO 
 
 O 
 
 o 
 
 o 
 
 
 o 
 
 
 
 o 
 
 
 o 
 
 o 
 
 o 
 
 o 
 
 
 o 
 
 
 
 <5 
 
 
 q 
 
 < 
 
 < 
 
 < 
 
 
 < 
 
 
 
 ■<! 
 
 
 <: 
 
 II. 1-1 
 
a. 
 ■a: 
 
 ca 
 
 S 
 
 _i 
 
 CO 
 CO 
 
 tx. 
 
 t — 
 
 
 
 
 
 
 
 
 <-•> 
 
 
 
 
 
 
 
 
 rrrr 
 
 
 
 
 
 
 
 
 r-2 
 
 
 
 
 
 
 
 
 1 — 
 
 
 
 
 
 
 
 
 t-o 
 
 
 
 
 
 
 
 
 f--) 
 
 
 
 
 
 
 
 
 »- -^ 
 
 
 
 
 
 
 
 
 *Fr 
 
 
 
 
 
 
 
 
 ilZ 
 
 
 
 
 
 ■5c 
 
 •J: 
 
 
 CZ3 
 
 
 
 
 
 X 
 
 ><; 
 
 
 t-3 
 
 
 
 
 
 
 
 
 ril 
 
 
 
 
 
 
 
 
 """* 
 
 O 
 
 
 
 
 
 OJ 
 
 
 
 o 
 
 
 
 
 
 o 
 
 
 
 •r^ 
 
 
 
 
 
 •^^ 
 
 
 
 y-l ^ 
 
 -^ 
 
 J4 
 
 
 A! 
 
 14-1 
 
 
 
 U-J yr 
 
 Vj 
 
 u 
 
 
 U 
 
 14-1 
 
 
 ITS 
 
 O (fl 
 
 (TJ 
 
 (tf 
 
 
 ti 
 
 o 
 
 
 C3 
 
 P-. 
 
 Q-, 
 
 PU 
 
 
 tu 
 
 
 
 1 — 
 
 C 
 
 
 
 
 
 c 
 
 
 ■»^ 
 
 O CJ 
 
 <u 
 
 <u 
 
 
 (U 
 
 o 
 
 
 CJ 
 
 •H t>0 
 
 CO 
 
 00 
 
 
 CO 
 
 •H 
 
 
 C3 
 
 M <U 
 
 (U 
 
 <u 
 
 
 <0 
 
 to 
 
 
 _J 
 
 •rH i-l 
 
 I-l 
 
 r•^ 
 
 
 I-l 
 
 tA 
 
 
 
 > .-4 
 
 1-1 
 
 •-I 
 
 
 1-4 
 
 > 
 
 
 
 •H O 
 
 o 
 
 o 
 
 
 o 
 
 •H 
 
 
 
 Q O 
 
 u 
 
 o 
 
 
 o 
 
 Q 
 
 
 
 
 
 
 
 
 (0 
 
 
 
 
 
 
 
 
 0) 
 
 
 
 
 
 
 
 
 •H . 
 
 
 
 
 
 
 
 
 . Jj 
 
 
 
 
 
 
 
 
 0) 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 (0 
 
 
 
 
 
 
 
 
 •H 
 
 
 
 
 60 
 
 
 
 
 Pl4 
 
 
 
 
 C 
 
 ^-s 
 
 
 
 
 
 a: 
 
 
 •1-1 
 
 w 
 
 
 
 TJ 
 
 
 CJ 
 
 
 4-1 
 
 fl4 
 
 
 
 tJ 
 
 
 1— 
 
 
 0) 
 
 "^^ 
 
 
 
 CO 
 0} 
 
 
 _i 
 
 
 a 
 
 V4 
 
 
 
 Tt 
 
 
 en 
 
 
 3 
 
 <u 
 
 
 
 u 
 
 
 
 
 w 
 
 •r-l 
 
 o 
 ^1 
 o 
 
 •a 
 
 
 u 
 
 
 
 
 
 j:: 
 
 o 
 
 O 
 
 CO 
 
 CO 
 
 ji 
 
 
 
 [/) 
 
 to 
 
 CO 
 
 
 ca 
 
 (U 
 
 o 
 
 
 
 W 
 
 •H 
 
 •H 
 
 • 
 
 (U 
 
 M 
 
 u 
 
 
 
 0) 
 
 f^ 
 
 > 
 
 Q 
 
 V^ 
 
 CO 
 
 i 
 
 
 
 ^ 
 
 
 ■^ 
 
 
 3) 
 
 a 
 
 
 
 60 
 
 r-4 
 
 <; 
 
 *4.J A 
 
 d 
 
 o 
 
 
 
 
 a 
 
 nJ 
 
 
 C 1-1 
 
 o 
 
 o 
 
 ti 
 
 
 
 o 
 
 •r^ 
 
 c 
 
 <U 0) 
 
 o 
 
 
 o 
 
 
 
 r ^ 
 
 t) 
 
 •-I 
 
 c: rrt 
 
 
 " 
 
 
 
 
 <t ~ 
 
 (4 
 
 OJ 
 
 53 T-i 
 
 ^ 
 
 CO 
 
 ^ OJ 
 
 
 
 VO 
 
 0) 
 
 •u 
 
 4J ^1 
 
 C/5 
 
 
 ^ So) 
 
 
 
 <r> CO 
 
 
 o 
 
 n) <i3 
 
 
 • 
 
 as 4J w 
 
 
 
 •-I 
 
 ^ 
 
 ;-! 
 
 4J PU 
 
 P 
 
 E3 
 
 ■-I 4J (U 
 
 
 
 E3 
 
 O 
 
 Ph 
 
 w 
 
 
 
 •H ^4 
 
 
 
 
 o 
 
 
 • 
 
 •» 
 
 •* 
 
 (5 CO 
 
 
 
 ^ 
 
 
 t4 
 
 bO Od 
 
 <U 
 
 <u 
 
 B C 
 
 
 Zo 
 
 <u 
 
 iw 
 
 w 
 
 r^ 
 
 u 
 
 p 
 
 
 
 
 
 o 
 
 O 
 
 o 
 
 •1-1 • 
 
 Vi 
 
 u 
 
 CJ CJ 
 
 
 
 0) 
 
 3 
 rt 
 
 H 
 
 >4-l Ci3 
 •W " 
 
 <u 
 
 1 
 
 03 • 
 <U V3 
 
 
 =3 
 
 ^ 
 
 OJ 
 
 "^^ 
 
 u • 
 
 o 
 
 3 
 
 > • 
 
 
 «« 
 
 o 
 
 ^ 
 
 Q 
 
 fq n 
 
 o 
 
 o 
 
 •S & 
 
 
 
 o 
 
 D 
 
 <d 
 
 ^ •^ 
 
 
 
 4J 
 
 
 
 
 m 
 
 fa 
 
 <J 
 
 CJ 
 
 g 
 
 (\J M 
 
 
 
 c 
 
 
 "•"•^^ 
 
 •V 
 
 o 
 
 o 
 
 4J d) 
 
 
 
 o 
 
 <u 
 
 o 
 
 1 f-l 
 
 
 
 c 
 
 
 
 
 x: 
 
 p^ 
 
 ,Q 
 
 0) 
 
 = <u 
 
 = Q) U 
 
 ►• 
 
 
 "(U (U 
 
 4-J 
 
 3 
 
 d) O 
 
 (1) 
 
 <U (U 
 
 <u to cu 
 
 
 CS 4J 
 
 «-l 
 
 aj 
 
 t^^ 
 
 4J 
 4J 
 
 4.1 4J 
 CO 4-1 
 
 
 •^ 
 
 
 l-l 4J 
 
 o 
 
 j:: 
 
 Vj 
 
 -H 
 
 U •r^ 
 
 M (X 
 
 
 
 
 iJ -,-) 
 
 
 4-! 
 
 4-> • 
 
 e 
 
 4-1 E 
 
 ti OJ 
 
 
 
 "^ a 
 
 p 
 
 
 a s 
 
 2 
 
 C B 
 
 c ^ ^ 
 
 a 
 
 
 <U E 
 
 c3 
 
 ^•N M- 
 
 0) 
 
 o 
 
 <u o 
 
 cu "^ c 
 
 4J 
 
 Si 
 
 
 O o 
 
 p 
 
 tyi o 
 
 o 
 
 o 
 
 O CJ 
 
 CJ C4-I 
 
 
 c U 
 
 bt, 
 
 tij U-l 
 
 c o 
 
 
 c 
 
 S C3 
 
 1 
 
 
 o 
 
 o 
 
 •^ 
 
 o 
 
 (U 
 
 O cu 
 
 <u 
 
 LaJ 
 
 u 0) 
 
 Vj 
 
 g -a 
 
 CJ « 
 
 JJ 
 
 CJ 4-) 
 
 CJ <U (U 
 
 4) 
 
 p! 
 
 
 P-, 
 
 
 0) 
 C r-l 
 
 CO 
 
 c 
 
 CO 
 
 C 3 4J 
 
 4J 
 
 (It 
 
 p: 
 
 •H c: 
 
 £ 
 
 U c« 
 
 •w u 
 
 (U 
 
 •r-l cu 
 
 •^ -H 
 
 ■4J 
 
 
 CJ (U 
 
 o 
 
 (u a. 
 
 0) 0) 
 
 w 
 
 4) en 
 
 <U «- E 
 
 (0 
 
 
 4-> (y) 
 
 u 
 
 -o 0) 
 
 4-1 E 
 
 
 u 
 
 *J ■■ p 
 
 
 
 o 
 
 « 
 
 >. 1-^ 
 
 O E 
 
 A 
 
 o 
 
 p - § 
 
 •0 
 
 
 )-l CO 
 
 (1) 
 
 C CU 
 
 U ca 
 
 CO 
 
 M CO 
 
 }* to CJ 
 
 Q) 
 
 
 a, to 
 
 V5 
 
 w 
 
 CM X 
 
 00 
 
 Ph CO 
 
 p4 ei 
 
 V4 
 
 
 ^ c 
 
 OJ 
 
 4-1 
 
 
 c 
 
 ^N c 
 
 _ /-^ C Q> 
 
 «9 
 
 
 j:: M -H 
 
 OS 
 
 • u 
 
 JZ ' 
 
 •H 
 
 X en -H 
 
 Si yi .M u 
 
 Da 
 
 
 Fis 
 
 N>IF 
 ear 
 
 
 o o 
 
 CO <; 
 
 M 
 
 CO ^ Vj 
 
 <5 & J-" CO 
 
 2 
 
 04 
 
 * 
 
 
 u 
 
 a, 
 
 •r-l 
 
 CO 
 
 •H ^ CO 
 
 fa fe; cu 
 
 :'-' ^- CO c 
 
 
 CL, 
 
 • <u 
 
 lil • 
 
 0) 
 
 f^ 2: (U m 
 
 
 1 <-' K 
 
 PIH 
 
 a Pi 
 
 : O 
 
 X 
 
 : ^-' X 
 
 = ^ X en 
 
 CD 
 
 
 
 
 
 
 
 
 :az 
 
 
 
 
 
 
 
 
 .^ 
 
 (^ 
 
 00 
 
 o\ 
 
 
 o 
 
 r-l 
 
 
 C3 
 
 o 
 
 o 
 
 o 
 
 
 iH 
 
 1-1 
 
 
 o 
 
 < 
 
 o 
 
 <: 
 
 5 
 
 
 5 
 
 5 
 
 
 II. 1-2 
 
» — 
 
 1 ■ 
 
 
 
 " ' 
 
 
 
 
 
 
 
 
 
 
 
 C_3 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 «n: 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ex.: 
 1 — 
 
 
 
 
 X 
 
 
 X 
 
 
 
 
 
 X 
 
 
 
 
 vo 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Ja.1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 -■c 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 --J 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ij. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CJ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CxX 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CJ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Zc£ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 (U 
 
 
 
 
 
 Q) 
 
 
 
 
 (U 
 
 
 
 
 
 
 (J 
 
 
 
 
 
 o 
 
 
 
 
 o 
 
 
 
 
 
 
 •H 
 
 
 
 
 
 ••-i 
 
 
 
 
 •H 
 
 
 
 
 
 
 M-l 
 
 
 X 
 
 
 X 
 
 <« 
 
 
 ^ 
 
 
 <M 
 
 
 
 
 
 
 U-l 
 
 
 u 
 
 
 u 
 
 M-l 
 
 
 V-i 
 
 
 IW 
 
 
 
 
 vz 
 
 
 O 
 
 
 cd 
 
 
 flj 
 
 o 
 
 
 ctJ 
 
 
 O 
 
 
 
 
 en 
 
 
 
 
 P4 
 
 
 PUl 
 
 
 
 Ph 
 
 
 
 
 
 
 »— 
 
 
 a 
 
 
 
 
 
 c 
 
 
 
 
 c 
 
 
 
 
 •a: 
 
 
 o 
 
 
 <u 
 
 
 o> 
 
 o 
 
 
 V 
 
 
 o 
 
 
 
 
 C3 
 
 
 •rH 
 
 
 CO 
 
 
 tiO 
 
 •H 
 
 
 bO 
 
 
 •rl 
 
 
 
 
 C3 
 
 
 W 
 
 
 <u 
 
 
 0) 
 
 CO 
 
 
 <u 
 
 
 CO 
 
 
 
 
 .—1 
 
 
 Q 
 
 
 t-l 
 I-t 
 o 
 
 
 f-l 
 
 r-l 
 O 
 
 o 
 
 •H 
 > 
 
 Q 
 
 
 l-« 
 
 l-l 
 o 
 o 
 
 
 > 
 •i-t 
 
 Q 
 
 
 
 
 
 
 
 
 
 <4-< 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 o 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 u 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 c 
 
 
 
 
 
 
 
 
 
 
 :e 
 
 
 
 
 
 0) 
 
 
 
 
 
 
 
 
 
 
 cr> 
 
 
 
 
 
 s 
 
 
 
 
 
 
 
 
 
 
 t— 
 
 
 
 w 
 
 
 Vi 
 
 
 
 
 
 t3 
 
 
 
 
 
 •oc 
 
 
 
 to 
 
 
 <d 
 
 
 
 
 
 B 
 
 
 
 
 
 C-> 
 
 
 
 OJ 
 
 
 o- 
 
 
 
 
 
 ttf 
 
 
 
 
 
 Zj 
 
 
 
 >-l 
 
 
 (U 
 
 
 
 
 
 i-i 
 
 
 
 
 
 ca 
 
 
 
 00 
 
 
 Q 
 
 
 
 
 
 (U 
 
 
 
 
 
 i=3 
 
 
 
 c 
 
 
 
 
 *-s 
 
 
 
 o 
 
 
 
 
 
 f^ 
 
 
 
 o 
 o 
 
 W5 
 
 
 • 
 
 
 c 
 
 CO 
 l-l 
 Pu, 
 
 
 
 H 
 
 «4-l 
 O 
 
 CO 
 
 u 
 
 
 
 
 
 
 
 
 :3 
 
 
 o 
 
 (U 
 
 n 
 
 >^ 
 
 
 
 4) 
 
 
 
 
 
 
 u 
 
 
 
 
 s-l 
 
 4-i 
 
 u 
 
 oc 
 
 
 
 4) 
 
 
 
 
 
 
 c 
 
 
 «k 
 
 
 > 
 
 cfl 
 
 
 o 
 
 
 
 f: 
 
 
 
 
 
 
 o 
 
 
 0) 
 
 
 (-1 
 
 >-l 
 
 
 u 
 
 
 
 T* 
 
 
 
 
 
 
 u 
 
 
 4-1 
 
 
 <u 
 
 4J 
 
 d) 
 
 p-l 
 
 
 
 bO 
 
 
 
 
 
 
 v^ 
 
 
 2 
 
 
 M 
 
 C 
 
 0) 
 
 <li 
 
 
 
 c 
 
 
 
 
 
 
 vO 
 
 
 3 
 
 
 0) 
 
 u 
 
 4-1 
 
 > 
 
 r^ 
 
 
 
 
 
 
 
 
 \D 
 
 
 w 
 
 
 v-l 
 
 C 
 
 •H 
 
 •H 
 
 vO 
 
 
 T) 
 
 
 
 
 
 
 <T> 
 
 
 
 
 iH 
 
 o 
 
 P 
 
 ■ 4-) 
 
 a^ 
 
 
 <u 
 
 
 
 
 
 
 r-4 
 
 
 
 
 ■-I 
 
 1-1 
 
 c 
 
 o 
 
 cfl 
 
 4J 
 
 c 
 
 I-l 
 
 
 V4 
 
 
 
 
 
 'U> 
 
 
 
 <u 
 
 
 S 
 
 0) 
 
 CO 
 
 H 
 
 
 
 cd 
 
 
 
 
 
 
 
 
 w 
 
 
 
 4-) 
 
 (U 
 
 
 
 
 J3 
 
 
 
 
 
 
 
 
 3 
 
 
 T) 
 
 o 
 
 •1-1 
 
 <: 
 
 
 
 O 
 
 
 
 
 
 
 
 o 
 
 
 fl 
 
 1-1 
 
 M 
 
 
 
 
 
 
 
 
 
 
 
 K 
 
 
 CO 
 
 p-l 
 
 OJ 
 
 
 
 
 <« 
 
 
 
 
 
 rrj 
 
 
 
 
 
 
 
 j:: 
 
 s 
 
 
 
 o 
 
 e 
 
 
 
 
 «« 
 
 
 
 <u 
 
 4J 
 
 J-l 
 
 "(U 
 
 
 en 
 
 CO 
 •r-l 
 
 y 
 
 cU 
 Oi 
 
 o 
 
 
 
 60 
 C 
 •W 
 
 4-1 
 
 l-l 
 
 w 
 o 
 )-l 
 
 
 
 
 
 
 — 
 
 •I-! 
 
 4J 
 
 M 
 
 c 
 
 ■1-1 
 
 PL, 
 
 
 
 <u 
 
 pn 
 
 
 
 
 
 
 "(I) 
 
 p 
 
 nj 
 
 CO 
 
 o 
 
 4J 
 
 
 
 
 <u 
 
 
 
 >. 
 
 
 
 
 4J 
 
 p 
 
 u 
 
 
 
 c 
 
 <v 
 
 
 <u 
 
 e 
 
 OJ 
 
 
 I-I 
 
 
 
 
 (0 
 
 o 
 
 u 
 
 0) 
 
 (0 
 
 4-1 
 
 
 4-1 
 
 
 4-1 
 
 
 c 
 
 
 
 
 ^1 
 
 u 
 
 c 
 
 u 
 
 i-H 
 
 rJ 
 
 
 (0 
 
 (U 
 
 CO 
 
 
 o 
 
 
 
 
 4-1 
 
 
 (U 
 
 
 u 
 
 4J 
 
 1-1 
 
 
 >-l 
 
 jr 
 
 u 
 
 
 
 
 
 
 C 
 
 0) 
 
 o 
 
 A 
 
 4) 
 
 <: 
 
 4J 
 
 
 4J 
 
 ^>> 4-i 
 
 4J 
 
 
 to 
 
 
 
 
 0) 
 
 o 
 
 c 
 
 <r 
 
 
 Vj 
 
 
 C 
 
 
 c 
 
 CO 
 
 c 
 
 
 4J 
 
 
 
 
 o 
 
 c 
 
 o 
 
 CO 
 
 
 d) 
 
 1-1 
 
 <u 
 
 
 OJ 
 
 
 (U 
 
 
 c 
 
 
 
 
 c 
 
 <1) 
 
 u 
 
 u-l 
 
 
 <4-l 
 
 (0 
 
 u 
 
 
 o 
 
 o 
 
 
 <u 
 
 
 
 
 o 
 
 ^1 
 
 
 
 
 c 
 
 •H 
 
 c 
 
 
 c 
 
 c 
 
 
 S 
 
 
 LaJ 
 
 
 o 
 
 OJ 
 
 c 
 
 4J 
 
 OU 
 
 o 
 
 O 
 
 o 
 
 
 o 
 
 N-^ -O 
 
 o 
 
 
 lU 
 
 
 ^ 
 
 
 
 l4-( 
 
 •■-1 
 
 (U 
 
 D. 
 
 u 
 
 n 
 
 o 
 
 
 o 
 
 i) 
 
 CJ 
 
 
 ij 
 
 ' 
 
 
 c 
 
 c 
 
 01 
 
 r-H 
 
 
 
 ■1-1 
 
 
 
 
 tl 4-1 
 
 
 
 CO 
 
 
 F= 
 
 
 ■H 
 
 o 
 
 4-1 
 
 U-l 
 
 r-^ 
 
 
 > 
 
 c 
 
 
 c 
 
 4) C 
 
 n 
 
 - 
 
 u 
 
 
 
 
 a 
 
 o 
 
 o 
 
 n 
 
 CM 
 
 w 
 
 o 
 
 •l-l 
 
 4-1 
 
 1-1 
 
 CO (U 
 
 •l-l 
 
 4J 
 
 to 
 
 
 
 
 4-1 
 
 
 u 
 
 V 
 
 
 Of 
 
 1-1 
 
 0) 
 
 u 
 
 o 
 
 •H CO 
 
 4J 
 
 U 
 
 
 
 
 
 o 
 
 «\ 
 
 Pm 
 
 hJ 
 
 • • 
 
 c 
 
 cu 
 
 4-1 
 
 o 
 
 4J 
 
 t-l <U 
 
 4-1 
 
 o 
 
 T3 
 
 
 
 
 u 
 
 w 
 
 
 
 u 
 
 •H 
 
 1 
 
 o 
 
 a- 
 
 o 
 
 n) u 
 
 O 
 
 a. 
 
 tu 
 
 
 
 
 PL. 
 
 00 
 
 0) 
 
 ?^ 
 
 o 
 
 -o 
 
 l-l 
 
 u 
 
 (U 
 
 u 
 
 P-l O- 
 
 u 
 
 <u 
 
 u 
 
 
 
 
 
 <~^ c 
 
 c 
 
 /-^ J-1 
 
 •H 
 
 CD 
 
 CTJ 
 
 cu 
 
 /-s eJ 
 
 Ph 
 
 
 Ph 
 
 /^ CcJ 
 
 rt 
 
 
 
 
 x; 
 
 CO ••"! 
 
 •H 
 
 to OJ 
 
 jj 
 
 o 
 
 IJ 
 
 
 w 
 
 
 • i-i 
 
 
 CO 
 
 a. 
 
 
 
 
 (/, 
 
 2: oj 
 
 l-l 
 
 tl — 
 
 <u 
 
 U 
 
 (U 
 
 X. 
 
 b w 
 g y 
 
 ^-^ ^ 
 
 ^ 
 
 Oi (U 
 
 .C 
 
 11 
 
 (U 
 
 
 
 
 •H 
 
 j3 
 
 S w 
 
 4_) 
 
 o 
 
 -o 
 
 (0 
 
 (0 
 
 D. 
 
 to 
 
 V4 
 
 
 
 
 Cj< 
 
 "^ 
 
 2: -^ 
 
 c 
 
 U 
 
 QJ 
 
 •H 
 
 •f^ 
 
 • to 
 
 •H 
 
 PL4 
 
 
 
 
 
 *-' x 
 
 
 v_^ u. 
 
 M 
 
 PU 
 
 t4 
 
 t. 
 
 [■- 
 
 CJ PU 
 
 fe 
 
 * 
 
 
 CJ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 =1 
 
 arc 
 
 
 <M 
 
 
 a 
 
 
 ^* 
 
 m 
 
 
 vO 
 
 
 l»» 
 
 
 
 
 
 — « 
 
 
 
 t-l 
 
 tH 
 
 
 r-l 
 
 
 I-t 
 
 
 
 
 t_> 
 
 
 O 
 
 
 5 
 
 
 O 
 < 
 
 o 
 <: 
 
 
 % 
 
 
 5 
 
 
 
 
 II. 1-3 
 
cv. 
 
 
 •a: 
 
 
 Dc: 
 
 CO 
 
 ^ 
 
 c-D 
 
 cd 
 
 — 1 
 
 ^ 
 
 OJl 
 
 W 
 
 ca 
 
 o 
 
 r -» 
 
 
 O. 
 
 
 ►— 
 
 
 
 ' 
 
 '■"" 
 
 
 
 ■■ ' ■' 
 
 
 
 
 
 
 
 ■ ■ ' 
 
 
 
 
 C-> 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 -< 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ex: 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 t— 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 X 
 
 
 
 to 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 oo 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 «=c 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 2£ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 _j 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CC 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CD 
 
 
 
 
 
 
 
 X 
 
 
 
 X 
 
 
 X 
 
 
 
 X 
 
 
 
 C_> 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 X 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 J<! 
 
 
 ^ 
 
 
 
 ^ 
 
 
 ^ 
 
 
 
 ^ 
 
 
 
 
 
 M 
 
 
 
 U 
 
 
 u 
 
 
 
 u 
 
 
 u 
 
 
 
 u 
 
 
 
 2C 
 
 
 nj 
 
 
 
 03 
 
 
 03 
 
 
 
 0) 
 
 
 «« 
 
 
 
 03 
 
 
 
 CD 
 
 
 (X, 
 
 
 
 (X, 
 
 
 P^ 
 
 
 
 P4 
 
 
 PL, 
 
 
 
 P4 
 
 
 
 »— 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 «: 
 
 
 0) 
 
 
 
 4) 
 
 
 (U 
 
 
 
 <u 
 
 
 <u 
 
 
 
 03 
 
 
 
 t_3 
 
 
 60 
 
 
 
 OO 
 
 
 CO 
 
 
 
 60 
 
 
 00 
 
 
 
 CO 
 
 
 
 CD 
 
 
 Q) 
 
 
 
 <D 
 
 
 0) 
 
 
 
 (U 
 
 
 v 
 
 
 
 03 
 
 
 
 
 
 i-l 
 O 
 
 o 
 
 
 
 r-l 
 
 i-l 
 
 o 
 o 
 
 
 r-l 
 
 I-l 
 o 
 
 O 
 
 
 
 r-l 
 r-l 
 
 o 
 o 
 
 
 r-l 
 
 o 
 
 o 
 
 
 
 r-H 
 r-( 
 
 U 
 
 
 
 
 
 
 
 
 
 
 
 
 
 03 
 
 
 -o 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 U) 
 
 
 )-l 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 tH C 
 
 
 03 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 (-1 03 
 
 
 3 
 
 
 
 3 
 
 
 
 
 
 
 
 
 
 
 
 
 
 03 e 
 
 
 o 
 
 
 
 (d 
 
 
 
 
 
 
 
 
 
 
 
 
 
 P4 a 
 
 
 H 
 
 
 
 C3 
 
 
 
 
 
 
 
 
 
 
 
 
 
 03 
 
 
 
 
 
 5-1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 • x: 
 
 
 CO 
 
 
 
 3 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Di O 
 
 
 4-1 
 U 
 
 
 
 PQ 
 
 
 
 
 
 * 
 
 
 
 
 
 
 
 
 • • 
 
 
 o 
 
 
 
 
 
 
 as 
 
 
 "cu 
 
 
 
 
 
 
 
 
 W S 
 
 
 14-1 
 
 
 
 • 
 
 
 
 CD 
 
 
 u 
 
 
 
 
 
 
 
 
 
 
 CM 
 
 
 
 m 
 
 
 
 
 
 nJ 
 
 
 
 
 
 
 
 
 1 • 
 
 
 W 
 
 
 
 1 
 
 
 
 H^ 
 
 
 M 
 
 
 
 
 
 
 
 
 3 
 
 
 
 
 
 
 
 
 «* 
 
 
 u 
 
 
 
 
 
 
 
 
 
 
 -o 
 
 
 
 03 
 
 
 
 CD 
 
 
 C 
 
 
 
 
 w 
 
 
 
 
 CJ 1 
 
 
 1-H 
 
 
 
 g 
 
 
 
 — 1 
 
 
 0) 
 
 
 
 
 c 
 
 
 
 
 Pj 
 
 
 tl 
 
 
 
 
 
 
 
 CO 
 
 
 o 
 
 
 
 
 (U 
 
 
 
 
 t^ u 
 
 
 o 
 
 
 
 X 
 
 
 
 =D 
 
 
 c 
 
 
 
 
 g 
 
 
 
 
 a, 
 
 
 s 
 
 
 
 
 
 
 Oi. 
 
 196/ (Cont'd) 
 
 o 
 o 
 
 c 
 
 o 
 u 
 
 x: 
 CO 
 
 •H 
 
 G 
 O 
 
 e 
 
 
 CJ 
 
 1 
 
 
 5 
 
 0) 
 
 M 
 
 C 
 
 0) 
 
 tn 
 
 tfl 
 
 03 
 
 OO 
 o 
 
 o 
 
 <u 
 w 
 
 1—1 
 
 
 
 U-l [J-i 
 
 o 
 
 3-. 
 
 a o 
 
 U u-l 
 03 
 
 >H ^ 
 
 CI 
 
 a. ^-1 
 
 03 03 
 03 CJ 
 Q 
 
 (11 tn 
 H fS 
 
 
 1 
 
 "4-1 
 
 03 
 03 
 
 00 
 C 
 •H 
 
 00 
 
 
 
 4J 
 CJ 
 
 3 
 •0 
 
 U 
 Pi 
 
 03 
 
 x: 
 
 60 
 r^ 
 
 •^^ 
 60 
 C 
 
 •H 
 
 V4 
 
 
 
 . . 
 
 
 03 
 
 
 ^ 
 
 
 C 
 
 
 
 
 
 
 to 
 
 
 
 
 
 
 oo 
 
 
 u 
 
 
 r— < 
 
 
 OS 
 
 1 
 
 
 
 1 
 
 
 03 
 
 
 
 
 
 
 
 
 OO 
 
 
 
 
 <u 
 
 
 
 
 
 
 V4 
 
 
 
 • • 
 
 
 
 ex: 
 
 CD 
 
 
 O 
 
 
 u 
 
 u 
 
 
 o 
 o 
 
 >> 
 
 S-1 
 
 
 
 S^ 
 
 
 
 
 
 
 C^ 
 
 
 
 
 Oi 
 
 
 1 
 
 C 
 
 
 
 o 
 
 
 
 O 
 
 
 
 
 
 
 
 
 
 zz> 
 
 
 
 
 •<r 
 
 <u 
 
 
 c 
 
 u 
 
 
 
 u 
 
 
 • • 
 
 
 
 4-1 : 
 
 
 
 «« 
 
 
 (0 
 
 (1) 
 
 
 ^^ 
 
 e 
 a 
 
 
 o 
 
 CO 
 
 
 
 CO 
 
 
 (U 
 
 u 
 
 
 
 CO B 
 
 03 
 
 
 
 
 
 •H 
 
 
 .. 
 
 o 
 
 
 03 
 
 0) 
 
 
 <M 
 
 03 
 
 CM 
 
 0; 
 
 ,-— N 
 
 CT\ 
 
 03 Vj 
 
 SO 
 
 
 
 
 H 
 
 
 ^^ 
 
 1— ( 
 
 
 o 
 
 ■u 
 
 
 r^ 
 
 4-1 
 
 r~^ 
 
 .^ 
 
 CO 
 
 CO 
 
 4-) 00 
 
 1 — 
 
 
 
 
 .-i; 
 
 
 u 
 
 (U 
 
 
 c 
 
 03 
 
 
 1 
 
 03 
 
 1 
 
 H 
 
 u-l 
 
 1 
 
 03 
 
 1 
 
 
 
 
 x: 
 
 
 D 
 
 > 
 
 
 <u 
 
 u 
 
 
 o 
 
 U 
 
 O 
 
 
 ^ 
 
 vO 
 
 1-1 3-1 
 
 
 
 
 
 
 m 
 
 
 < 
 
 OJ 
 
 
 1-1 
 
 u 
 
 
 r^ 
 
 4-) 
 
 r>. 
 
 X) 
 
 ^ 
 
 en 
 
 4J p^ 
 
 r^ 
 
 
 
 
 ■r^ 
 
 
 
 Q 
 
 
 (U 
 
 r^ 
 
 
 
 C 
 
 
 c 
 
 ■^^ 
 
 
 C 
 
 
 
 
 
 U^ 
 
 
 00 
 
 CO 
 
 
 14-1 
 
 c 
 
 o 
 
 
 ^ 
 
 03 
 
 o 
 
 ^ 
 
 03 
 
 , 
 
 ^ 
 
 03 CO 
 03 
 
 ^ 
 
 
 
 
 1—1 
 
 
 <r 
 
 4_» 
 
 
 o 
 
 c 
 
 
 r^ 
 
 c 
 
 r^ 
 
 03 
 
 u 
 
 00 
 
 C -H 
 
 a> 
 
 
 
 
 03 
 
 /^N 
 
 
 •H 
 
 
 t) 
 
 O 
 
 
 N-^ 
 
 o 
 
 v-^ 
 
 U 
 
 >-l 
 
 Nw^ 
 
 U 
 
 "v-^ 
 
 
 
 
 •r^ 
 
 CO 
 
 r. 
 
 
 •-^ 
 
 
 CJ 
 
 ,^-N 
 
 . * 
 
 u 
 
 ^-( 
 
 0) 
 
 
 r— 1 
 
 CJ 01 
 
 1^-N < 
 
 
 
 
 o 
 
 b 
 
 C 
 
 U.I 
 
 W 
 
 u 
 
 
 OO 
 
 CM 
 
 
 OJ 
 
 ^ 
 
 »• 
 
 CN 
 
 jn 
 
 CO rsi 
 
 
 
 
 1-1 
 
 03 
 
 O 
 
 U. 
 
 m 
 
 C 
 
 
 
 ^ 
 
 
 
 — ' 
 
 
 ^- «** 
 
 
 
 
 
 dJ 
 
 J:^ 
 
 e 
 
 
 U 
 
 
 •H 
 
 1 
 
 r% 
 
 •H 
 
 *^ 
 
 1 
 
 Xi 
 
 *» 
 
 •M •r4 
 
 
 UJ 
 
 
 
 ^^ 
 
 > 
 
 •o 
 
 0) 
 
 >i 
 
 03 
 
 >> 
 
 
 
 
 > 
 
 03 Ft. 
 
 
 p: 
 
 
 p 
 
 
 a 
 
 u 
 
 ^-^ 
 
 0) 
 
 iJ 
 
 ^«^ 
 
 Oi 
 
 4J 
 
 oc 
 
 >^ 
 
 c 
 
 t.( 
 
 4J 
 
 v-^ 00 
 
 
 
 d 
 
 U 
 
 .r 
 
 o 
 
 
 u 
 
 o 
 
 
 o 
 
 o 
 
 o 
 
 u 
 
 i^ 
 
 
 
 r-< 
 
 
 
 
 f= 
 
 
 u 
 
 <u 
 
 10 
 
 i-> 
 
 )-i 
 
 c 
 
 (J 
 
 (-1 
 
 r— 1 
 
 14 
 
 r-l 
 
 c 
 
 
 r-l 
 
 U 03 
 
 V-l r-l 
 
 
 
 
 
 CO 
 
 •rJ 
 
 ■-0 
 
 0) 
 
 CJ 
 
 P4 
 
 01 
 
 O 
 
 P4 
 
 O 
 
 4_) 
 
 • 
 
 
 
 Pi -H 
 
 03 
 
 
 
 
 •4-1 
 
 •r-l 
 
 Ci. 
 
 •1-1 
 
 X) 
 
 t/i 
 
 
 TD 
 
 C 
 
 
 c 
 
 CO 
 
 ^ 
 
 C 
 
 CJ 
 
 -0 c 
 
 
 
 
 o 
 
 u 
 
 
 re 
 
 >. 
 
 01 
 
 x: 
 
 >^ 
 
 .c 
 
 x: 
 
 j: 
 
 
 
 j=; 
 
 x: Vj 
 
 >, x: 
 
 
 
 
 
 ri 
 
 1—1 
 
 
 c 
 
 u 
 
 to 
 
 c 
 
 u 
 
 CO 
 
 o 
 
 CJ 
 
 • 
 
 
 
 CO 0) 
 
 c 
 
 
 
 
 D 
 
 Pi 
 
 oj 
 
 <: 
 
 C/2 
 
 o. 
 
 •r-l 
 
 CO 
 
 03 
 
 •H 
 
 03 
 
 a, 
 
 ;j 
 
 0) 
 
 ■H F 
 
 CO 03 
 
 
 
 
 03 
 
 
 C 
 
 
 
 
 U. 
 
 
 H 
 
 Ci4 
 
 H 
 
 t- 
 
 
 H 
 
 tl4 E 
 
 H 
 
 
 
 
 OJ 
 
 • 
 
 o 
 
 1 
 
 • 
 
 M 
 
 
 > 
 
 
 
 
 
 1 
 
 
 
 
 • 
 
 
 
 
 >-i 
 
 a; 
 
 •w 
 
 
 o 
 
 <u 
 
 03 
 
 CJ 
 
 -a 
 
 03 
 
 •o 
 
 01 
 
 
 T3 
 
 ^ u 
 
 a -o 
 
 
 
 
 
 
 _03 
 
 u 
 
 p- 
 
 
 D. 
 03 
 
 jS 
 
 
 o 
 o 
 
 e5 
 
 o 
 o 
 
 P 
 
 
 
 
 
 
 p 14.4 
 
 
 • 
 
 
 
 
 
 w 
 
 J^' 
 
 u, 
 
 Q 
 
 Oi 
 
 
 Q 
 
 Ii4 
 
 - 
 
 C«4 
 
 - 
 
 Ui- 
 
 U. 
 
 z 
 
 Q Pu, 
 
 
 CD 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 a= 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 00 
 
 
 
 o\ 
 
 
 O 
 
 
 
 I-l 
 
 
 p>i 
 
 
 
 ro 
 
 
 
 
 >-H 
 
 
 
 r-l 
 
 
 <M 
 
 
 
 CM 
 
 
 CvJ 
 
 
 
 CN 
 
 
 
 CD 
 
 
 
 
 
 O 
 < 
 
 
 ^ 
 
 
 
 O 
 < 
 
 
 
 < 
 
 
 
 
 
 1: 
 
 
 
 II. 1-4 
 
>- 
 =1= 
 ex. 
 
 tC 
 
 cc 
 
 CO 
 
 CO 
 
 I 
 
 CO 
 
 C-9 
 
 w 
 
 1 — 
 
 
 
 
 
 
 
 
 
 
 
 
 CJ 
 
 
 
 
 
 
 
 
 
 
 
 
 "J. 
 
 
 
 
 
 
 
 
 
 
 
 
 ceL 
 
 
 
 
 
 
 
 
 
 
 
 
 h- 
 
 
 
 
 
 
 
 
 
 
 
 
 OO 
 
 
 
 
 
 
 
 
 
 
 
 
 CI 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Lj 
 
 
 
 
 
 
 
 
 
 
 
 
 i»_ 
 
 
 
 
 
 
 
 
 
 
 
 
 C3 
 
 
 
 
 >< 
 
 
 
 
 X 
 
 
 
 
 <-:> 
 
 
 
 
 
 
 
 
 
 
 
 
 t.i 
 
 
 
 
 
 
 
 
 
 
 
 
 
 OJ 
 
 (U 
 
 
 1) 
 
 
 
 
 
 
 
 
 
 o 
 
 o 
 
 
 u 
 
 
 
 
 
 
 
 
 
 •rH 
 
 'H 
 
 
 •r^ 
 
 
 
 
 
 
 
 
 
 **-( 
 
 U-l 
 
 
 IH 
 
 
 j« 
 
 
 X 
 
 
 ^ 
 
 
 
 U-l 
 
 IM 
 
 
 U-( 
 
 
 u 
 
 
 u 
 
 
 M 
 
 
 o 
 
 O 
 
 
 O 
 
 
 J!> 
 
 
 ti 
 
 
 TO 
 
 
 
 
 
 
 
 
 04 
 
 
 fu 
 
 
 P-i 
 
 
 H^ 
 
 c 
 
 c 
 
 
 B 
 
 
 
 
 
 
 
 
 «: 
 
 o 
 
 o 
 
 
 O 
 
 
 <u 
 
 
 o 
 
 
 0) O 
 
 
 <-^ 
 
 T-l 
 
 •rt 
 
 
 iH 
 
 
 t>0 
 
 
 00 
 
 
 60 .-1 
 
 
 
 05 
 
 W 
 
 
 CO 
 
 
 (U 
 
 
 CI 
 
 
 0) u 
 
 
 •H 
 
 •W 
 
 
 •H 
 
 
 r-4 
 
 
 r-l 
 
 
 r-l 4-1 
 
 
 
 > 
 
 > 
 
 
 > 
 
 
 l-l 
 
 
 r-l 
 
 
 i-l TO 
 
 
 
 •H 
 
 •r^ 
 
 
 f-l 
 
 
 O 
 
 
 O 
 
 
 O (U 
 
 
 
 o 
 
 Q 
 
 
 Q 
 
 
 u 
 
 
 o 
 
 
 O CO 
 
 
 
 
 
 
 
 u 
 
 
 
 
 
 
 
 
 
 
 
 
 lU 
 
 
 
 
 
 
 
 
 
 
 
 
 f-l 
 
 
 
 
 
 
 
 
 
 
 
 
 u 
 
 
 
 
 
 
 
 
 
 
 
 
 <u 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 Ul 
 
 
 
 
 
 
 
 
 
 
 
 
 •r-l 
 
 
 
 
 
 
 
 
 
 
 
 
 U* 
 
 
 
 
 
 
 
 :« 
 
 
 
 
 
 •a 
 
 
 CM 
 
 
 
 
 
 C3 
 »— 
 
 
 
 
 
 S 
 
 
 1 
 
 in 
 
 
 
 
 
 •CC 
 
 
 
 
 
 
 
 vO 
 
 
 
 
 
 CJ 
 
 
 
 
 
 s 
 
 
 CM 
 
 
 
 
 
 >— J 
 
 
 
 
 
 •rl 
 
 
 
 
 
 
 
 CO 
 
 => 
 a. 
 
 
 4-> 
 
 
 
 u 
 
 a 
 to 
 
 o 
 
 a 
 
 
 • 
 
 o 
 o 
 
 r-l 
 
 o 
 
 •r-l 
 X) 
 
 
 
 •H 
 
 <u 
 
 o 
 u 
 
 
 
 00 
 
 
 (d 
 
 
 
 <U oi 
 
 
 TO 
 
 - 
 
 
 4J 
 
 
 
 vc 
 
 a 
 
 
 u 
 
 o 
 
 
 (U vo 
 
 
 H 
 
 "e 
 
 
 •r^ -a 
 
 
 
 en 
 
 u 
 
 
 iJ 
 
 r^ 
 
 
 w o> 
 
 
 4-1 
 
 TO 
 
 
 r-l . en 
 
 
 
 r-l 
 
 TO 
 
 
 c 
 
 1 
 
 
 4J r-l 
 
 
 CO 
 
 U 
 
 o 
 
 TO I 
 
 
 
 (-1 
 
 
 <U 
 
 ro 
 
 
 •H 
 
 
 O 
 
 60 
 
 CO 
 
 3 -^ 
 
 
 
 iJ 
 
 
 o 
 
 r>» 
 
 
 6 
 
 
 Si 
 
 o 
 
 1 
 
 O" en 
 
 
 «— k 
 
 C 
 
 
 C 
 
 
 
 g 
 
 
 
 u 
 
 in 
 
 1 
 
 
 </> 
 
 (P 
 
 
 o 
 
 .• 
 
 
 O 
 
 
 C 
 
 p-l 
 
 CM 
 
 x: 
 
 
 rx; 
 
 u 
 
 
 CJ 
 
 •'>• 
 
 
 o 
 
 
 TO 
 
 
 
 60 ''^ 
 
 
 CD 
 
 c 
 
 
 
 >, 
 
 
 
 
 > 
 
 (J 
 
 • • 
 
 •H CT» 
 
 
 
 o 
 
 
 c 
 
 r-t 
 
 
 « 
 
 
 
 4J 
 
 ^-\ 
 
 P^ J^ 
 
 
 P^ 
 
 o 
 
 
 •H 
 
 3 
 
 
 <u 
 
 
 n 
 
 TO 
 
 r— ) 
 
 t» 
 
 
 =D 
 
 
 
 <U 
 
 •n 
 
 
 > 
 
 
 to 
 
 n 
 
 s_^ 
 
 CM v--- 
 
 
 «X 
 
 c 
 
 
 4J 
 
 v-/ 
 
 
 T-I 
 
 
 <D 
 
 4-) 
 
 ,— 1 
 
 O -1 
 
 
 
 -rH 
 
 
 O 
 
 o 
 
 
 «J 
 
 
 O 
 
 c 
 
 cn 
 
 /-N ^ 
 
 
 
 0) 
 
 
 Vj 
 
 1-4 
 
 
 TO 
 
 
 u 
 
 0) 
 
 
 <u CO 
 
 
 
 4J 
 
 
 Oi 
 
 
 
 U 
 
 
 3 
 
 o 
 
 M 
 
 Ufa" 
 
 
 
 o 
 
 
 
 A 
 
 
 c 
 
 
 O 
 
 c 
 
 3 
 
 V4 § 5 
 
 
 
 u 
 
 
 JS 
 
 to 
 
 
 (U 
 
 
 to 
 
 o 
 
 01 
 
 3 fe; 4) 
 
 
 
 tu 
 
 
 Ul 
 
 4J 
 
 Q) 
 
 CO 
 
 
 <u 
 
 o 
 
 •H 
 
 O ^-^ -H 
 
 Ss 
 
 
 
 •'^ 
 
 •H 
 
 (J 
 
 •u 
 
 OJ 
 
 
 on, 
 
 
 > 
 
 CO > 
 
 r-l 
 
 
 A 
 
 (^ 
 
 U< 
 
 o 
 
 TO 
 
 u 
 
 
 
 c 
 
 G 
 
 V4 fU 
 
 C 
 
 
 <A 
 
 o 
 
 
 a 
 
 t-i 
 
 c 
 
 
 tu 
 
 •H 
 
 «: 
 
 r-l 4) Cli 
 
 o 
 
 
 *H 
 
 •H 
 
 c 
 
 o 
 
 4-1 
 
 <u 
 
 
 c 
 
 tu 
 
 
 TO C 
 
 
 
 Ua 
 
 Vj 
 
 o 
 
 ei 
 
 c 
 
 (13 
 
 
 <^ -H 
 
 J-l 
 
 to 
 
 •H >» to 
 
 to 
 
 
 
 <y 
 
 
 •-S 
 
 <u 
 
 
 
 CO Vj 
 
 o 
 
 tu 
 
 4-1 O 1) 
 
 4J 
 
 
 
 4-1 
 
 U) 
 
 CO tn 
 
 o 
 
 >4-l 
 
 
 & J3 
 
 u 
 
 •r^ 
 
 C >-) •- 1 
 
 c 
 
 
 in 
 
 c 
 
 (0 
 
 2 'u 
 
 c 
 
 o 
 
 
 ^ ^ 
 
 Oi 
 
 Vj 
 
 (U l-l 
 
 0) 
 
 
 1 
 
 M 
 
 OJ 
 
 o 
 
 
 
 g 
 
 
 0) 
 
 4J • OJ 
 
 6 
 
 Lcl 
 
 o 
 
 
 l-l 
 
 o 
 
 0) 
 
 
 v_x M-i 
 
 J= 
 
 <■"• x: 
 
 o H x: 
 
 0) 
 
 ..J 
 
 r^ 
 
 <4-l 
 
 60 
 
 v-/ a) 
 
 
 0] 
 
 
 o 
 
 to 
 
 CO to 
 
 &4 to 
 
 4J 
 
 :**- 
 
 O 
 
 o 
 
 x: 
 
 c 
 
 D 
 
 
 u 
 
 •H 
 
 |E 
 
 •rH 
 
 « 
 
 
 
 u 
 
 Vj (/) 
 
 1-1 
 
 o 
 
 
 <U TO 
 
 ft, 
 
 <j: - tiL, 
 
 4J 
 
 
 u 
 
 O u 
 
 Ph 
 
 (U -H 
 
 <u 
 
 •X. 
 
 
 CO •r^ 
 
 
 •rl 
 
 tn 
 
 
 0) 
 
 C 0) 
 
 
 •a u. 
 
 4J 
 
 
 
 •H -O 
 
 • 
 
 -^ r-l 
 
 r-l r-l 
 
 
 
 D 
 
 « a. 
 
 J= 
 
 >» 
 
 o 
 
 M 
 
 
 Wi 01 
 
 en 
 
 td 
 
 • • r-l TO 
 
 •a 
 
 
 TO 
 
 S a. 
 
 o 
 
 c c 
 
 >-l 
 
 Oi 
 
 
 TO a 
 
 
 x: -H 
 
 to 4) -H 
 
 tu 
 
 
 Oi 
 
 u 
 
 CO n 
 
 PU 
 
 60 
 
 
 Pk O 
 
 • 
 
 u u 
 
 r-l C O 
 
 u 
 
 
 
 u 
 
 TO 
 
 •r-l 
 
 
 /-N C 
 
 
 1— 1 
 
 5 
 
 C P 
 
 o; tH U 
 
 cd 
 
 
 OJ 
 
 TO OJ 
 
 O 
 
 • T3 
 
 x: 
 
 CO -H 
 
 — 
 
 ; " 
 
 
 •r^ O 
 
 to Q. 0) 
 
 CU 
 
 
 3 
 
 CL 3 
 
 en 
 
 O TO 
 
 CT 
 
 2 <y 
 
 U 
 
 « > 
 
 <u 
 
 ^ 6 
 
 to CO g 
 
 (U 
 
 
 (/) 
 
 O W) 
 
 <u 
 
 C 
 
 ■H 
 
 cu 
 
 o 
 
 s 
 
 £ 
 
 3 6 
 
 n 
 
 
 VI 
 
 Q CO 
 
 ci 
 
 • TO 
 
 t< 
 
 Cl4 
 
 • c 
 
 • o 
 
 ?: • o 
 
 Pui 
 
 
 M 
 
 •-^ M 
 
 : 
 
 Q O 
 
 : 
 
 v_^ re 
 
 : 
 
 tj CJ 
 
 : 
 
 Oi CJ 
 
 Z <-! <^ 
 
 • J^ 
 
 rrj 
 
 
 
 
 
 
 
 
 
 
 
 
 ac 
 
 
 
 
 
 
 
 
 
 
 
 
 —1 
 
 <)• 
 
 u-» 
 
 
 NO 
 
 
 r* 
 
 
 a 
 
 
 tjN 
 
 
 r^ 
 
 rsi 
 
 <N 
 
 
 «M 
 
 
 <M 
 
 
 c^ 
 
 
 CM 
 
 
 CJ 
 
 O 
 
 o 
 
 
 O 
 
 
 O 
 
 
 g 
 
 
 o 
 
 
 
 < 
 
 < 
 
 
 < 
 
 
 < 
 
 
 < 
 
 
 < 
 
 
 II. 1-5 
 

 X 
 
 
 ■X. 
 
 C3 
 
 -a: 
 
 C_3 
 
 u 
 
 to 
 
 O 
 
 <u 
 
 M 
 0) 
 i-l 
 r-l 
 O 
 O 
 
 J< 
 
 
 u 
 
 n. 
 
 cd 
 
 u 
 
 PU 
 
 Pi 
 
 00 
 
 Ml 
 
 lU 
 
 Q) 
 
 i-i 
 
 i-H 
 
 i-H 
 
 t-4 
 
 O 
 
 o 
 
 o 
 o 
 
 P4 
 
 (U 
 bO 
 U 
 
 o 
 
 re 
 cr: 
 
 C3 
 
 CO 
 CU 
 
 «£ 
 
 «_•> 
 
 CO 
 CL. 
 
 C 
 
 o 
 u 
 
 ON 
 
 OS 
 
 cc: 
 
 C3 
 
 (« 
 
 
 D 
 
 U 
 
 w 
 
 Oi 
 
 o 
 
 (^ 
 
 >4-l 
 
 o 
 
 
 
 4-1 
 
 
 c 
 
 w 
 
 M 
 
 01 
 
 
 ■r-1 
 
 tJ" 
 
 u 
 
 c 
 
 OJ 
 
 ■H 
 
 a 
 
 (/) 
 
 w 
 
 to 
 
 
 CJ 
 
 (U 
 
 o 
 
 F 
 
 o 
 
 o 
 
 Vj 
 
 C/3 
 
 fu 
 
 CO 
 
 2 
 
 U < 
 
 O 2 
 
 •u 
 
 rj o 
 
 U 4-1 
 O 
 
 rO 4-1 
 
 t-1 O 
 
 o. 
 
 (1) <u 
 
 t-l Di 
 
 4-1 
 
 4-1 CO 
 
 a) fi^ 
 
 CO Ja 
 
 O 
 
 
 u 
 
 a 
 a 
 o 
 u 
 a 
 w 
 
 u 
 
 o 
 a. 
 
 0) 
 
 iM CO 
 
 (U 
 
 .a 
 
 § 
 u 
 
 Q 
 
 > 
 
 CU 
 o 
 
 I-H 
 
 o 
 c 
 
 o 
 
 H 
 
 •u 
 
 •a 
 
 o 
 
 o 
 
 o 
 
 o 
 
 Cb 
 
 U4 
 
 c 
 
 c 
 
 03 
 
 -.-1 
 
 R 
 
 
 -I 
 
 CO 
 
 ffi 
 
 3 
 
 
 4) 
 
 V4 
 
 •H 
 
 o 
 
 > 
 
 CO Ci 
 
 (0 
 
 a u 
 
 tL4 
 
 o 
 
 03 CJ 
 
 CO 
 
 
 - 
 
 Oj 
 
 t-< 
 
 >. W 
 
 r^ 
 
 n) 
 
 fO 
 
 <y> 
 
 T3 
 
 • -3 
 
 i-i 
 
 p 
 
 3: " 
 
 
 EH 
 
 
 
 w 
 
 >< 
 
 
 3 
 
 o .. 
 
 
 ^ 
 
 Cn /-\ 
 
 
 flj 
 
 00 
 
 
 4J 
 
 • 1 
 
 
 • CO 
 
 W 1^ 
 
 
 CO 
 
 • n 
 
 
 4J 
 
 S en 
 
 
 t 
 
 CO <u 
 
 
 <u 
 
 C -r-l 
 
 
 4-1 
 
 •w > 
 
 
 CO 
 
 1-1 0) 
 
 
 u 
 
 .-t « 
 
 
 4J 
 
 •r^ 
 
 
 c 
 
 4J to 
 
 
 (U 
 
 CO <D 
 
 
 o 
 
 •H 
 
 
 c 
 
 • Vj 
 
 
 o 
 
 PQ <U 
 
 
 o 
 
 x; 
 
 •> en 
 
 
 c 
 
 • v4 
 
 
 •H 
 
 M [i- 
 
 
 (U 
 
 •-) 
 
 
 u 
 
 i-< 
 
 
 o 
 
 « 03 
 
 
 u 
 
 .-; -r^ 
 
 
 PL. 
 
 o M 
 
 
 X 
 
 C (U 
 
 
 OJ 
 
 >i 6 
 
 
 •H 
 
 C 
 
 
 fe 
 
 • o 
 
 
 
 O CJ 
 
 o 
 
 o 
 
 < 
 
 lo 
 
 5 
 
 CM 
 
 en 
 
 5 
 
 o 
 
 < 
 
 «n 
 o 
 
 II. 1-6 
 
cu 
 en 
 
 C3 
 
 en 
 
 CO 
 
 1 — 
 
 
 "^' " ' 
 
 
 
 
 
 
 
 
 
 
 
 CJ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 «x 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 X 
 
 
 
 
 X 
 
 
 
 X 
 
 
 
 X 
 
 
 t/0 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 
 
 
 -t 
 
 
 
 
 
 
 
 
 
 
 
 
 
 s 
 
 
 
 
 
 
 
 
 
 
 
 
 
 — i 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CD 
 
 
 
 
 
 
 
 X 
 
 
 * 
 X 
 
 
 X 
 
 
 CJ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 s 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 •r-l 
 
 
 01 
 
 a 
 
 •iH 
 
 
 •r-( 
 
 11 
 
 
 •r^ 
 
 OJ 
 
 
 
 •r^ 
 
 
 
 
 
 ^ 
 
 
 <4^ 
 
 
 IM 
 
 
 H-l 
 
 ll-l 
 
 t*-l 
 
 
 .^ 
 
 
 3; 
 
 u 
 
 
 *H 
 
 
 U-l 
 
 
 l*-l 
 
 n-l 
 
 <W 
 
 
 W 
 
 
 CD 
 
 a, 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CO 
 
 CM 
 
 
 H^ 
 
 
 c 
 
 
 c 
 
 
 c 
 
 C 
 
 c 
 
 
 
 «x 
 
 0) 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 0) 
 
 
 CD 
 
 CO 
 
 
 •r-l 
 
 
 •H 
 
 
 •H 
 
 •r^ 
 
 ■H 
 
 
 M ■ 
 
 
 CD 
 
 d) 
 
 
 CO 
 
 
 M 
 
 
 W 
 
 M 
 
 CO 
 
 
 <u 
 
 
 — 1 
 
 <-> 
 
 
 t-l 
 
 
 •i-l 
 
 
 T-l 
 
 •rl 
 
 •H 
 
 
 r-l 
 
 
 
 i-i 
 
 
 > 
 
 
 > 
 
 
 > 
 
 > 
 
 > 
 
 
 I-I 
 
 
 
 
 
 
 ■H 
 
 
 ■t-l 
 
 
 •H 
 
 •rH 
 
 •r^ 
 
 
 
 
 
 
 
 
 
 a 
 
 
 Q 
 
 
 
 
 Q 
 
 Q 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 uT 
 
 
 
 
 
 
 
 
 
 
 
 
 
 (U 
 
 
 
 
 
 
 
 
 
 
 
 
 
 •r^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 U 
 
 
 
 
 
 
 
 
 
 
 
 
 
 <u 
 
 
 
 
 
 
 
 
 
 
 
 
 
 X 
 
 
 
 
 
 
 
 
 
 
 
 
 
 to 
 
 •H 
 
 
 00 
 
 a 
 
 
 3= 
 
 
 
 
 
 
 
 
 
 •0 
 
 C 
 CO 
 
 
 
 
 CD 
 
 
 
 
 
 
 
 
 
 
 S 
 
 
 K^ 
 
 
 
 
 
 
 
 
 
 (U 
 
 
 
 
 •cC 
 
 
 
 
 
 
 
 
 
 a 
 
 
 CO 
 
 u 
 
 CO 
 
 
 CD 
 
 
 
 
 
 
 
 
 
 •^^ 
 
 
 
 
 
 
 
 
 
 
 
 
 v^ 
 
 
 
 1 
 
 
 
 
 
 
 
 
 
 C3 
 
 
 •I-I 
 
 
 OQ 
 
 ZD 
 
 
 
 
 
 
 
 
 
 s 
 
 
 60 
 • 
 
 
 O. 
 
 
 
 
 
 
 
 
 
 tJ 
 
 
 r-l 
 
 
 
 
 
 
 
 
 
 
 
 c 
 
 
 </. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 to 
 
 
 <U 
 
 c 
 
 
 
 
 
 
 
 
 
 
 
 j:; 
 
 
 Ut 
 
 JZ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 U 
 
 
 1— 1 
 
 
 
 0) 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 H 
 
 
 
 
 0) 
 
 
 
 
 
 
 
 S 
 
 
 en 
 
 f 
 
 
 u 
 < 
 
 
 
 
 5) 
 
 
 
 3 
 
 o; 
 
 >r-( 
 
 
 
 
 
 13 
 
 
 0) 
 (U 
 
 
 C 
 
 'r-l 
 <V 
 i-i 
 
 • 
 
 C 
 
 U-l 
 
 
 
 x: 
 
 4-1 
 
 
 C 
 
 
 
 
 > 
 
 4) 
 
 
 
 
 
 
 
 
 u 
 
 CM 
 
 CO 
 
 
 CC 
 CD 
 
 1 — 
 
 =3 
 
 
 
 
 u 
 nj 
 
 
 00 
 
 
 
 c 
 
 JZ 
 
 
 
 « 
 
 
 u 
 
 
 
 
 
 
 u 
 
 to 
 0) 
 
 > 
 
 
 JZ 
 U) 
 
 •r-l 
 
 14-1 
 
 
 v-^ 4-) 
 
 •H 
 N 4J 
 4-1 to 
 
 <u c 
 
 2: M 
 
 
 <£ 
 
 
 H 
 
 
 
 
 H 
 
 
 
 
 
 Cm 
 
 
 
 trj 
 
 
 
 ■ 4J 
 
 to 
 u 
 
 c 
 
 ^ CJ 
 
 
 tr. 
 
 •r-l 
 
 • I-I 
 
 < to 
 
 
 
 
 
 0) 
 
 
 
 
 
 
 6 
 
 
 
 
 Vj 
 
 CO C 
 
 
 
 1 
 
 
 
 
 
 y-*N 
 
 
 
 .^N 
 
 ^-N 
 
 <U U) 
 
 
 0) 
 
 to < 
 
 >. 
 
 
 
 
 
 
 
 
 (U 
 
 
 
 g 
 
 • 
 
 • 
 
 
 
 OJ 
 
 c 
 
 1 — ' 
 
 
 li 
 
 
 JZ 
 
 
 u 
 
 u 
 
 
 
 
 C/5 
 
 
 
 
 
 
 
 
 ■^ -a 
 
 c 
 
 
 4-1 
 
 ''^ 
 
 01 
 
 
 
 
 ni 
 
 c 
 
 
 
 c 
 
 c: 
 
 
 u 
 
 .-1 c 
 
 
 
 
 CO 
 
 
 3 
 
 •u 
 
 M 
 
 
 1) 
 > 
 
 •r-l 
 
 M 
 
 M 
 
 0) 
 
 
 
 I-I CM 
 4-) 
 
 
 
 C 
 
 Zi 
 
 
 
 <u 
 
 OJ 
 
 Ui 
 
 
 1-1 
 
 CO 
 
 W 
 
 U_l 
 
 
 c: 
 
 CO 4-1 
 
 C 
 
 
 
 
 
 to 
 
 
 (0 
 
 u 
 
 c 
 
 
 
 
 3 
 
 
 u 
 
 r: 
 
 
 
 g ° 
 
 
 ty 
 
 to 
 
 • 
 
 
 B 
 
 
 c 
 
 * 
 
 
 
 
 
 ■r-l 
 
 u 
 
 OJ 
 
 ■r-l 0-1 
 
 t\) 
 
 •r-l U 
 
 ••_) «J 
 
 
 c 
 
 Di T3 
 
 0) 
 
 
 
 
 u 
 
 13 
 
 /•*\ 
 
 C/i 
 
 CJ 
 
 > 
 
 u 
 
 y. 
 
 > VJ 
 
 1— 1 
 
 > vj 
 
 0) 
 
 
 
 
 OJ 
 
 u 
 
 1 
 
 
 
 '~1 
 
 01 
 
 CO 
 
 6C 
 
 c 
 
 •r-i 
 
 
 u 
 
 
 
 Id 
 
 Vj 
 
 P-. 
 
 v. 
 
 C D 
 
 Ul 
 
 ■r^ 
 
 « 4J 
 
 to 
 
 
 c 
 
 
 c 
 
 c 
 
 11 
 
 0. 
 
 
 d) o. 
 
 
 Q) D- 
 
 •rH 
 
 U] 
 
 u 
 
 m c 
 
 4J 
 
 ^^ 
 
 r— < 
 
 G 
 
 a 
 
 •r^ 
 
 m 
 
 1) 
 
 § 
 
 CO <u 
 Ci 
 
 
 CO <U 
 
 
 0) 
 
 u 
 
 4) 
 
 - to 
 
 -.0 
 
 
 il 
 
 .Q 
 
 I/) 
 
 nj 
 
 ^^ 
 
 CD 
 
 AJ 
 
 a, 
 
 
 )-i 
 
 & 
 
 Vj 
 
 IX 
 
 
 
 ^ 
 
 nJ 
 
 x: <u 
 
 T3 
 
 
 j-J 
 
 
 C 
 
 CD 
 
 ^ 
 
 
 
 
 
 
 u 
 
 &f1 
 
 J-J 
 
 ti 
 
 J-> 
 
 u in 
 
 c 
 
 a 
 
 ^ u 
 
 01 
 
 
 
 
 1-1 
 
 ac 
 
 JZ 
 
 •H 
 
 1-1 
 
 i-i 
 
 
 
 
 
 iM 
 
 
 
 Vj 
 
 CM to 
 
 
 
 a> 
 
 ty ci, 
 
 u 
 
 
 
 
 01 
 D. 
 tU 
 O, 
 
 
 
 CI- 
 
 'tI 
 
 ca 
 
 CO 
 
 Ci 
 
 x: 
 00 
 
 u 
 u 
 
 
 tr) 
 >-i 
 u 
 C 
 
 
 
 CM 
 
 CJ 
 
 CM 
 
 ^ ID 
 
 C 
 J: 
 
 CM 
 
 x: <D 
 
 C 
 •<<; 
 
 v-' CJ 
 
 c 
 
 JZ -^ 
 
 •r-l nj 
 
 CJ 
 CO 
 
 1-1 
 
 CM 
 
 to r. 
 
 >-> CO 
 
 to 
 0. 
 
 01 
 
 ca 
 
 
 
 
 
 
 
 
 
 
 
 
 
 21 
 
 m 
 
 
 vO 
 
 
 r» 
 
 
 00 
 
 ON 
 
 
 
 
 t—t 
 
 
 I 
 
 fO 
 
 
 rn 
 
 
 ro 
 
 
 <*i 
 
 <n 
 
 <r 
 
 
 <t 
 
 
 — i 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 «t 
 
 < 
 
 
 < 
 
 
 < 
 
 
 < 
 
 < 
 
 < 
 
 
 < 
 
 
 CD 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 -. 
 
 II. 1-7 
 
en 
 I — 
 oo 
 en 
 
 CD 
 
 
 CJ 
 
 U-l 
 
 O 
 
 d 
 o 
 
 CD 
 
 
 CO 
 
 CI. 
 
 C3 
 
 ^ 
 
 > M 
 
 H O 
 
 a; o. 
 
 CO 0) 
 Q. 
 
 o 
 u 
 x: 
 
 CD 
 
 
 o 
 
 < 
 
 II. 1-8 
 
re 
 o- 
 •a: 
 cxz 
 
 eg 
 ca 
 
 
 00 
 
 M 
 H 
 CO 
 l-l 
 
 oi 
 
 u 
 
 H 
 U 
 
 < 
 
 a; 
 
 H 
 O 
 
 Q 
 O 
 
 ►— 
 
 
 
 
 
 
 
 
 t_> 
 
 
 
 
 
 
 
 
 B 
 
 X 
 
 
 X 
 
 
 
 
 X 
 
 CO 
 
 
 
 
 
 
 
 
 Cj3 
 
 -a: 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 ^j 
 
 
 
 
 
 
 
 
 Li^ 
 
 
 
 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 X 
 
 QC 
 
 
 
 
 
 
 
 
 CJ 
 
 
 
 
 
 
 
 
 -rj: 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 ^ 
 
 
 Ji! 
 
 
 ^ 
 
 
 n 
 
 
 V4 
 
 
 U 
 
 
 M 
 
 
 03 
 
 
 n) 
 
 
 01 
 
 
 f 
 
 2= 
 
 On 
 
 
 (ii4 
 
 
 Cli 
 
 
 CLi 
 
 CD 
 
 
 
 
 
 
 
 
 1— > 
 
 <U 
 
 
 u 
 
 
 <u 
 
 
 O 
 
 "C 
 
 00 
 
 
 00 
 
 
 00 
 
 
 00 
 
 ej> 
 
 <U 
 
 
 at 
 
 
 0) 
 
 
 V 
 
 o 
 
 .— < 
 
 
 i-i 
 
 
 I-I 
 
 
 1-1 
 
 —J 
 
 #— 1 
 
 
 i-i 
 
 
 •-4 
 
 
 I-I 
 
 
 o 
 
 
 o 
 
 
 o 
 
 
 O 
 
 
 u 
 
 
 CJ 
 
 
 o 
 
 
 CJ 
 
 
 
 
 u 
 
 
 
 
 1 
 
 
 
 
 tu 
 
 
 
 
 u 
 
 
 
 
 Xi 
 
 
 
 
 tl) : 
 
 
 
 
 CJ 
 
 
 1 
 
 
 
 4-1 (0 
 
 
 
 
 c; 
 
 
 o 
 
 ,*-^ 
 
 
 
 0) TD 
 
 
 o 
 
 
 
 •r-l 
 
 CO 
 
 u 
 
 «3 
 
 
 ^ 
 
 to 
 3 
 
 C 
 
 
 
 Q 0) 
 T3 [14 
 
 
 > 
 
 /'^ 
 
 
 (0 
 
 
 
 u-l 
 
 ■H 
 
 •— \ 
 
 
 •r-l 
 
 
 t) 
 
 to 
 
 
 t-4 
 
 , 
 
 
 o 
 
 ^4 
 
 CO 
 
 
 a -H 
 
 
 •r^ 
 
 Cn 
 
 
 3 
 
 Cl, 
 
 
 
 to 
 
 Cn 
 
 
 *^ 2 
 
 CD 
 
 CO 
 
 s-^ 
 
 
 O 
 
 f—A 
 
 H 
 
 
 > 
 
 4J 
 
 •r4 
 
 CO 
 
 ^ 
 
 ^-^ 
 
 
 O T^ 
 
 K— 
 
 O 
 
 
 
 
 
 
 .—1 
 
 0) 
 
 
 
 •H < 
 
 •a: 
 
 
 Vl 
 
 
 x: 
 
 
 
 CO 
 
 4J 
 
 M 
 
 
 e 
 
 CD 
 
 T3 
 
 01 
 
 
 CO 
 
 •^ 
 
 
 3 
 
 to 
 
 (U 
 
 
 <i -o 
 
 — 1 
 
 C 
 
 x: 
 
 
 •r-l 
 
 u 
 
 
 o 
 
 rj 
 
 x: 
 
 
 OJ 
 
 ca 
 
 n] 
 
 o 
 
 
 U-, 
 
 3 
 
 
 
 XI 
 
 o 
 
 
 C X 
 
 •~^ 
 
 
 c 
 
 
 
 ^ 
 
 
 T3 
 
 tu 
 
 3 
 
 
 •r-l -r-l 
 
 a- 
 
 u 
 
 « 
 
 
 to 
 
 to 
 
 
 ^ 
 
 C/3 
 
 CO 
 
 
 s 
 
 
 C 
 
 M 
 
 
 3 
 
 •r-l 
 
 
 v^^ 
 
 M 
 
 
 to 
 
 
 <U 
 
 m 
 
 
 O 
 
 a, 
 
 
 
 
 M 
 
 
 4-1 T3 
 
 
 u 
 
 
 
 •H 
 
 
 
 > 
 
 x: 
 
 
 
 .-1 C 
 
 
 c 
 
 , 
 
 
 U 
 
 , 
 
 
 u 
 
 o 
 
 • 
 
 
 t^ CO 
 
 
 o 
 
 U^ 
 
 
 CO 
 
 H 
 
 
 •r^ 
 
 M 
 
 Pm 
 
 
 to 
 
 
 u 
 
 
 
 > 
 
 
 
 4J 
 
 c 
 
 (X, 
 
 
 
 00 
 
 -a to VD 
 
 
 r; 
 
 H 
 
 
 14-1 
 
 IjJ 
 
 
 CO 
 
 
 oJ 
 
 
 C (U -CM 
 
 
 •H 
 
 
 
 O 
 
 
 
 3 
 
 C 
 
 
 
 to 4-1 '-v 1 
 
 
 OJ 
 
 
 
 
 «k 
 
 
 (D- 
 
 CO 
 
 A 
 
 
 CO c/j vc 
 
 
 4-1 
 
 u 
 
 
 c 
 
 ij 
 
 
 
 01 
 
 4J 
 
 
 » (-1 P-. VO 
 
 
 O 3 
 
 
 •r-l 
 
 3 
 
 
 ni 
 
 M 
 
 ^ 
 
 
 "> 4-1 v; /^ J 
 
 
 H O 
 
 
 -J 
 
 ■D 
 
 
 ^ 
 
 ■^ 
 
 
 
 
 C/ >-< x^ 
 
 
 a, 
 
 j:: 
 
 
 4-) 
 
 x: 
 
 
 4J 
 
 
 x: 
 
 
 4J 0) ^^^ •■ 
 
 
 
 u 
 
 
 o 
 
 to 
 
 M 
 
 
 
 6 
 
 M 
 
 
 CO O ^-^ 
 
 
 c 
 
 0) 
 
 
 ^ 
 
 4-1 
 
 OJ 
 
 
 C 
 
 o 
 
 0) 
 
 
 1-1 C • n 
 
 
 o 
 
 j-i 
 
 
 c 
 
 4-) 
 
 
 O 
 
 M 
 
 4-1 
 
 
 XI o >-i -^ 
 
 
 
 to 
 
 
 
 tl) 
 
 to 
 
 
 
 tw 
 
 to 
 
 
 >. o ^ 
 
 
 f— ^ 
 
 w 
 
 t§ 
 
 
 U-l 
 
 E 
 
 3 
 
 • on 
 
 (O 
 
 
 3 
 
 LO 
 
 -C CO 
 
 
 vO 
 
 (U 
 
 
 O 
 
 0) 
 
 O 
 
 vD 
 
 i) 
 
 OJ 
 
 O 
 
 vD 
 
 O d ' .-1 
 
 Co~ 
 
 a\ 
 
 r-H 
 
 
 
 
 T—\ 
 
 
 tr. 
 
 f-4 
 
 X) 
 
 
 ON 
 
 XI •r^ r-l 
 
 
 J3 
 
 , 
 
 
 > 
 
 a 
 
 . 
 
 I-I 
 
 XI 
 
 JO 
 
 • 
 
 f-H 
 
 H QJ r43 •> 
 
 Oi 
 
 to 
 
 w 
 
 
 4J 
 
 a 
 
 cu 
 
 
 CO 
 
 s 
 
 CxJ 
 
 
 to 4-1 o >, 
 
 CD 
 
 ■ r-4 
 
 
 
 ■A 
 
 3 
 
 
 
 •r4 
 
 
 
 
 O O C fj 
 
 re 
 1 — 
 
 W 
 
 , 
 
 
 4-' 
 
 to 
 
 , 
 
 
 V4 
 
 0) 
 
 • 
 
 
 1-1 fc4 JJ 
 
 nj 
 
 kj 
 
 
 c 
 
 
 ►-5 
 
 
 to 
 
 4-1 
 
 ►J 
 
 
 •> A4 (0 
 
 > 
 
 
 
 CO 
 
 c 
 
 
 
 > 
 
 CO 
 
 
 
 to • -r^ 
 
 
 «« 
 
 
 3 
 
 •H 
 
 •• 
 
 
 
 u 
 
 <^ 
 
 
 " x: s: E^ 
 
 
 M 
 
 u 
 
 
 O 
 
 0) 
 
 >-i 
 
 
 cc 
 
 u 
 
 M 
 
 
 to to til 
 
 
 C 
 
 (U 
 
 
 
 4-1 
 
 0) 
 
 
 c 
 
 c 
 
 <u 
 
 
 U^ •H • JC 
 
 
 •H 
 
 -o 
 
 
 -o 
 
 O 
 
 -a 
 
 
 •r^ 
 
 OJ 
 
 T3 
 
 
 [^ O CJ 
 
 
 « 
 
 >, 
 
 
 c 
 
 ^ 
 
 >, 
 
 
 to 
 
 o 
 
 >. 
 
 
 u-l 
 
 
 U) 
 
 G 
 
 
 CO 
 
 d 
 
 
 to 
 
 c 
 
 c 
 
 
 o - "-:: 
 
 
 01 
 
 CO 
 
 
 
 
 to 
 
 
 <u 
 
 o 
 
 00 
 
 
 to tu o 
 
 
 O 
 
 
 u 
 
 >i 
 
 u 
 
 
 4-t 
 
 o 
 
 o 
 
 
 U 
 
 01 .H (0 O 
 
 
 
 
 , 
 
 D. 
 
 4J 
 
 OJ 
 
 , 
 
 Q. 
 
 o 
 
 
 . 
 
 Q. 
 
 o to o u. 
 
 
 (5^ 
 
 O 
 
 •r-l 
 
 ■r^ 
 
 .n 
 
 u 
 
 •H 
 
 i: 
 
 c 
 
 u 
 
 •r-l 
 
 c o u 
 
 
 
 t-l 
 
 ,— ( 
 
 4J 
 
 
 t-i 
 
 •r4 
 
 
 u 
 
 OJ ?^ XI *c 
 
 
 
 , 
 
 u 
 
 CO 
 
 o 
 
 , 
 
 O 
 
 
 01 
 
 . 
 
 o 
 
 1-1 EC 
 
 
 ,_! 
 
 Q 
 
 to 
 
 3 
 
 
 Q 
 
 to 
 
 c 
 
 4-1 
 
 Q 
 
 to 
 
 11 rt < tt 
 
 
 Cu 
 
 
 3 
 
 CjT 
 
 J=. 
 
 
 3 
 
 •r-l 
 
 o 
 
 
 3 
 
 u-l to 
 
 
 
 
 
 c 
 
 u 
 
 C 
 
 CO 
 
 to 
 
 4J 
 
 £ 
 
 C 
 
 C 
 
 to 
 
 >-l T^ -r-l 
 
 CI [i-i W o3 
 
 
 > 
 
 t^ 
 
 '^, 
 
 •r-l 
 
 's 
 
 to 
 
 2 
 
 ^4 
 
 
 to 
 
 ^ 
 
 ti ,, ^ 
 
 CD 
 
 E 
 
 
 
 
 fi 
 
 
 0) 
 
 x: 
 
 B 
 
 
 C C • 3 
 
 P= 
 
 to 
 
 N 
 
 t3 
 
 to 
 
 to 
 
 N 
 
 T3 
 
 u 
 
 « 
 
 M 
 
 T3 
 
 1-1 -r^ Z: 4-) 
 
 
 
 u 
 
 4-1 
 
 01 
 
 OJ 
 
 c 
 
 4J 
 
 OJ 
 
 
 •r^ 
 
 4-1 
 
 <U 
 
 r-> 
 
 
 >4-l 
 
 
 
 V-i 
 
 ^ 
 
 o 
 
 o 
 
 V-i 
 
 JS 
 
 14-1 
 
 U. 
 
 w 
 
 x: 
 
 ty CO •> p 
 
 
 
 
 o 
 
 3 
 
 10 
 
 c 
 
 to 
 
 3 
 
 to 
 
 o 
 
 
 3 
 
 to 
 
 -1 c x: o 
 
 
 
 
 i^ 
 
 •r-l 
 
 tl) 
 
 --I 
 
 l<i 
 
 •r4 
 
 
 t4-l 
 
 S«J 
 
 •r-l 
 
 x: o *-" .H 
 
 
 «J 
 
 tu 
 
 
 ^H 
 
 i-l 
 
 u 
 
 
 *— ' 
 
 4J 
 
 
 
 
 1-4 
 
 •■-I -H -rH P 
 
 
 u 
 
 
 , 
 
 J3 
 
 0) 
 
 "i 
 
 . tyl 
 
 XI 
 
 o 
 
 
 . 
 
 -Q 
 
 to u e CO 
 
 
 <u 
 
 .c 
 
 .x 
 
 3 
 
 >4-( 
 
 a 
 
 s: u^ 
 
 3 
 
 0) 
 
 C 
 
 ^ 
 
 3 
 
 to to oo < 
 
 
 y-j 
 
 tn 
 
 
 D. 
 
 IM 
 
 K 
 
 . 9. 
 
 D. 
 
 14-1 
 
 •r4 
 
 
 D- 
 
 ,? ° 
 
 
 U-, 
 
 •^ 
 
 , 
 
 C 
 
 •r-l 
 
 o 
 
 C 
 
 U-l 
 
 0) 
 
 . 
 
 c 
 
 a, .rJ . . 
 
 
 w 
 
 u. 
 
 u 
 
 13 
 
 Q 
 
 u 
 
 O '-^ 
 
 :=) 
 
 ■-U 
 
 4J 
 
 CJ 
 
 3 
 
 c> 
 
 
 
 
 
 
 
 
 -^ 
 
 fA 
 
 
 <s 
 
 
 n 
 
 
 -* 
 
 1 
 
 o 
 
 
 o 
 
 
 o 
 
 
 o 
 
 *-J 
 
 o 
 
 
 o 
 
 
 o 
 
 
 o 
 
 CD 
 
 ca 
 
 
 CO 
 
 
 CQ 
 
 
 CO 
 
 II. 2-1 
 

 CO 
 
 
 u 
 
 
 M 
 
 *^M 
 
 H 
 
 3C 
 
 00 
 
 O- 
 
 c: 
 
 •«3: 
 
 w 
 
 cc: 
 
 H 
 
 CJ 
 
 tJ 
 
 cu 
 
 < 
 
 
 rr 
 
 —i 
 
 < 
 
 — — 
 
 _u 
 
 en 
 
 o 
 
 CJJ 
 
 H 
 
 O™ 
 
 U 
 
 Uu. 
 
 U 
 
 
 CJ 
 
 
 o 
 
 
 Pi 
 
 
 a. 
 
 ►— 
 
 
 
 
 
 
 
 
 ■"*■*■ 
 
 
 
 
 
 
 <_} 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 1 — 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 X 
 
 
 X 
 
 
 X 
 
 
 
 X 
 
 
 
 
 C/O 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 "i 
 
 
 
 
 
 
 
 
 
 
 
 
 ' 
 
 
 g 
 
 
 
 
 
 
 
 
 
 
 
 
 ' % 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 lH 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CD 
 
 
 
 
 
 
 
 X 
 
 
 
 
 
 
 •■ 
 
 ce 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 C_) 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 :£ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 ^ 
 
 
 ^ 
 
 
 j«: 
 
 
 
 ^ 
 
 
 .y 
 
 
 
 )-> 
 
 
 u 
 
 
 V4 
 
 
 >-i 
 
 
 
 VI 
 
 
 I-I 
 
 
 
 13 
 
 
 <0 
 
 
 CO 
 
 
 CO 
 
 
 
 (0 
 
 
 CO 
 
 
 a- 
 
 a. 
 
 
 04 
 
 
 CUi 
 
 
 a. 
 
 
 
 a. 
 
 
 cv, 
 
 
 o 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 >~ 
 
 <u 
 
 
 <u 
 
 
 v 
 
 
 01 
 
 
 
 <u 
 
 
 CU 
 
 
 «c 
 
 to 
 
 
 60 
 
 
 00 
 
 
 !30 
 
 
 
 00 
 
 
 CO 
 
 
 C-3 
 
 <y 
 
 
 (U 
 
 
 <u 
 
 
 01 
 
 
 
 <u 
 
 
 
 
 
 C3 
 
 r-l 
 
 
 i-t 
 
 
 l-l 
 
 
 t~i 
 
 
 
 I-I 
 
 
 1— 1 
 
 
 _J 
 
 o 
 o 
 
 
 t-i 
 
 o 
 o 
 
 
 
 
 
 
 
 
 
 f-i 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 4J 
 
 c 
 
 01 
 
 
 
 
 
 
 c 
 
 
 CO 
 
 
 
 
 
 ^~\ 
 
 
 
 > 
 
 
 
 32 
 
 
 
 o 
 
 
 •u 
 
 
 
 
 
 r^ 
 
 
 
 f— 1 
 
 
 
 c^ 
 
 
 
 t-l 
 
 
 CO 
 
 ,—1 
 
 
 
 
 
 v^ 
 
 
 
 
 
 
 
 
 
 
 4-1 
 
 
 •r-l 
 
 CO 
 
 
 
 
 
 o\ 
 
 
 
 CO 
 
 
 
 1 — 
 
 
 
 CO 
 
 
 to 
 
 
 
 
 
 
 
 1-1 
 
 
 
 
 
 
 
 
 4J 
 
 
 o 
 
 •r^ 
 
 
 
 
 
 • 
 
 
 
 
 
 
 
 C3 
 
 
 
 c 
 
 
 rH 
 
 E 
 
 
 
 
 
 > 
 
 
 
 4-1 
 
 
 
 —J 
 
 
 CO 
 
 OJ 
 
 
 O 
 
 OJ 
 
 
 
 
 
 
 
 
 
 
 
 OQ 
 
 
 u 
 
 s 
 
 •-^ 
 
 c 
 
 X 
 
 
 
 
 
 
 
 
 u 
 
 
 
 :=> 
 
 
 w 
 
 M 
 
 >, 
 
 X 
 
 u 
 
 
 30 
 
 
 
 *» 
 
 
 
 
 
 
 
 o. 
 
 
 ■r-l 
 
 00 
 
 01 
 
 4-1 
 •r-l 
 
 o 
 
 0) 
 
 
 
 ■r4 
 
 
 c 
 
 ■r^ 
 
 
 
 CO 
 
 iJ 
 
 
 
 ■r^ 
 
 
 
 
 
 o 
 
 
 '/I 
 
 H 
 
 CQ 
 
 
 4J 
 
 
 
 u 
 
 
 
 
 
 
 OJ 
 
 
 ,-< 
 
 > 
 
 S-l 
 
 
 
 
 a; 
 
 
 
 0) 
 
 
 
 
 
 
 
 u 
 
 
 
 
 XI 
 
 QJ 
 
 TD 
 
 c:; 
 
 
 ij 
 
 
 
 CO 
 
 
 
 X 
 
 
 
 
 P3 
 
 
 c 
 
 
 > 
 
 O 
 
 
 
 
 71 
 
 " 
 
 
 3 
 
 
 
 CO 
 
 
 
 
 M 
 
 
 ^ 
 
 CO 
 
 ■-J 
 
 o 
 
 
 
 
 
 
 
 x 
 
 
 
 •r^ 
 
 
 
 
 <-i 
 
 
 o 
 
 f— ( 
 
 c 
 
 Cl, 
 
 /) 
 
 
 CO 
 
 4-J 
 
 
 a 
 
 
 
 Ci. 
 
 
 
 
 c 
 
 
 (U 
 
 CO 
 
 o 
 
 
 01 
 
 
 u 
 
 a 
 
 
 CO 
 
 
 
 
 
 
 
 a 
 
 
 H 
 
 •r-< 
 
 
 C4-I 
 
 u 
 
 
 CO 
 
 v^ 
 
 
 03 
 
 
 
 j: 
 
 
 
 
 o 
 
 
 
 u 
 
 CO 
 
 o 
 
 3 
 
 
 •r^ 
 
 -U 
 
 
 CO 
 
 • 
 
 
 01 
 
 
 
 
 c 
 
 
 ■c 
 
 0) 
 
 •r^ 
 
 
 -o 
 
 
 M 
 
 c 
 
 
 ra 
 
 CU 
 
 
 
 re: 
 
 
 \ 
 
 
 
 
 
 JJ 
 
 
 
 -J 
 
 
 
 v^ 
 
 
 ^-. 
 
 ^ 
 
 
 
 
 
 
 
 u 
 
 
 o 
 
 CO 
 
 E 
 
 4II 
 
 a 
 
 
 ^H 
 
 
 
 
 
 CO 
 
 
 
 CO 
 
 
 
 
 
 
 tl 
 
 T. 
 
 ri 
 
 D 
 
 
 
 
 
 
 C 
 
 
 *■ 
 
 1-1 
 
 
 
 c 
 
 
 
 
 c 
 
 
 
 
 f— 1 
 
 4J 
 
 i-i 
 
 
 C 
 
 
 
 
 CO 
 
 ■4-1 
 
 
 
 •r^ 
 
 
 
 
 •.-I 
 
 
 y-l 
 
 > 
 
 o 
 
 ■r-l 
 
 0-. 
 
 
 X 
 
 a 
 
 
 X 
 
 c 
 
 
 
 4-1 
 
 
 
 
 0) 
 
 
 o 
 
 u 
 
 CLJ 
 
 XJ 
 
 
 
 CJ 
 
 
 
 CO 
 
 OJ 
 
 
 
 CO 
 
 
 
 
 4-1 
 
 /-^ 
 
 
 01 
 
 
 CO 
 
 c 
 
 
 OJ 
 
 c 
 
 
 1-3 
 
 
 
 
 
 01 
 
 
 
 
 vO 
 
 o 
 
 H 
 
 01 
 
 x; 
 
 
 c 
 
 
 
 
 H r^ 
 
 •r4 
 
 
 
 c 
 
 
 
 X 
 
 
 
 ZTT 
 
 0^ 
 
 £ 
 
 1— 1 
 
 
 CO 
 
 ■r-l 
 
 x; 
 
 H-l 
 
 •r-l 
 4-) 
 
 ,^ 
 
 X> ON 
 
 OJ 
 4J 
 
 
 CJ 
 
 
 
 CJ 
 
 
 
 Ci-I 
 
 
 
 
 .— ( 
 
 
 ^— ^ 
 
 iJ 
 
 U-, 
 
 CJ 
 
 
 
 
 
 CO 
 
 -H 
 
 
 
 
 •r-l 
 
 
 
 
 
 
 
 
 13 
 
 x: 
 
 
 ■r4 
 
 
 
 X 
 
 CO 
 
 ^ 
 
 
 
 i: 
 
 
 4-1 
 
 c 
 
 
 
 
 
 
 aj 
 
 -o 
 
 W 
 
 C4-I 
 
 . 
 
 iJ 
 
 V-i 
 
 C14 
 
 
 CO 
 
 •H 
 
 
 
 4-1 
 
 
 
 •r-( 
 
 r-< 
 
 Ul 
 
 o 
 
 <; 
 
 
 4-t 
 
 :s 
 
 
 
 
 iC 
 
 01 
 
 
 
 u 
 
 
 
 b-. 
 
 i) 
 
 G 
 
 
 
 M-l 
 
 X 
 
 ^— ' 
 
 Ci-I 
 
 X 
 
 
 
 4-1 
 
 
 
 0) 
 
 
 
 «C 
 
 U-l 
 
 
 M 
 
 C 
 
 c/) 
 
 
 
 u 
 
 1-1 
 
 
 
 CO 
 
 •r4 
 
 
 1 
 
 
 
 
 
 C44 
 14-1 
 
 
 
 
 O 
 
 tc 
 
 M 
 
 0) 
 
 r-l 
 
 «) 
 
 .— » 
 
 0) 
 
 OJ 
 
 U, 
 
 
 u 
 
 
 
 to 
 
 
 
 
 
 o 
 
 o 
 
 4-1 
 
 01 
 
 c 
 
 
 
 •y) 
 
 4-1 
 
 
 
 < 
 
 
 
 
 
 
 
 
 tn 
 
 v^ 
 
 M-l 
 
 C 
 
 o 
 
 ■r^ 
 
 X 
 
 ■r-l 
 
 D 
 
 U~i 
 
 
 
 X 
 
 
 
 
 
 
 
 0) 
 
 03 
 
 
 o 
 
 Pi 
 
 4J 
 
 
 
 w 
 
 ^ 
 
 
 
 
 • 
 
 CO 
 
 
 
 • • 
 
 
 
 
 •H 
 
 
 00 
 
 u 
 
 
 01 
 
 
 
 CO 
 
 ■M 
 
 
 
 00 
 
 •r^ 
 
 
 
 y 
 
 
 
 
 4-1 
 
 • 
 
 c 
 
 
 , 
 
 01 
 
 , — 1 
 
 X 
 
 J-i-l 
 
 u 
 
 C 
 
 
 • 
 
 U4 
 
 
 
 
 
 
 ^1 
 
 J 
 
 •r-l 
 
 c 
 
 < 
 
 6 
 
 < 
 
 CO 
 
 
 CO 
 
 
 
 ^^ 
 
 3 
 
 
 
 
 t:-, 
 
 ^ 
 
 
 
 o 
 
 
 •u 
 
 •r-l 
 
 
 
 
 ■r-l 
 
 • 
 
 c: 
 
 ■,-i 
 
 '.0 
 
 
 C4-I 
 
 
 
 
 IX 
 
 
 
 D 
 
 , 
 
 <y 
 
 QJ 
 
 , 
 
 r— 1 
 
 , — 1 
 
 Pm 
 
 0; 
 
 i-i 
 
 4-1 
 
 1 
 
 C4-1 
 
 
 
 
 
 U-l 
 
 ■r^ 
 
 
 
 O 
 
 12 
 
 dj 
 
 4-) 
 
 o 
 
 to 
 
 > 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 ^-v U 
 
 
 
 t 
 
 
 S 
 
 
 
 
 D 
 
 a 
 
 c^ 
 
 
 4-1 
 
 3 
 
 
 c 
 
 .'^. 
 
 
 
 00 
 
 
 
 A 
 
 
 i: 
 
 .^ 
 
 C 
 
 
 
 c 
 
 J 
 
 CO 
 
 1— t 
 
 ^r^* 
 
 4-1 
 
 
 
 H 
 
 
 >, 
 
 U- CO 
 
 
 
 
 g 
 
 r-l 
 
 1-1 
 
 c 
 
 ^ 
 
 
 
 
 CO 
 
 
 t-1 
 
 •^ 
 
 1— 1 
 
 
 4-1 
 
 X 3 
 
 
 
 •a 
 
 3 
 
 « 
 
 
 Z) 
 
 nj 
 
 c 
 
 CO 
 
 . f« 
 
 -0 
 
 > 
 
 CT 
 
 
 
 1_) 
 
 "^ 
 
 4J 
 
 ■r4 
 
 ^ c 
 
 
 ■ » ■ 
 
 c 
 
 CO 
 
 D 
 
 1J 
 
 13 
 
 
 
 
 u 
 
 OJ 
 
 01 
 
 U 
 
 !10 
 
 c 
 
 03 
 
 v_x 
 
 >-l 
 
 ■ 1 
 
 •.^ 
 
 
 p! 
 
 o 
 
 J3 
 
 C 
 
 :S 
 
 
 01 
 
 CO 
 
 ■r^ 
 
 —- 1 
 
 4-1 
 
 
 c 
 
 0) 
 
 n: 
 
 
 
 
 2 
 
 _ Zl 
 
 
 
 c 
 
 C 
 
 O 
 
 X 
 
 f— 1 
 
 4-1 
 
 s-l 
 
 C 
 
 ■— 1 
 
 •r-4 
 
 OS 
 
 —J 
 
 G 
 
 a 
 
 
 3 
 
 
 p: 
 
 c 
 
 dJ 
 
 CO 
 
 
 X 
 
 4-1 
 
 
 CO 
 
 OJ 
 
 01 
 
 CO 
 
 1—1 
 
 
 •—* 
 
 1-1 
 
 01 
 
 6' 
 
 XI 
 
 
 
 o 
 
 c 
 
 
 u 
 
 _t: 
 
 
 C4-I 
 
 •r^ 
 
 K 
 
 CO 
 
 c 
 
 f— 1 
 
 OT 
 
 ■^ 
 
 CO 
 
 QC 
 
 
 c 
 
 M 01 
 
 
 
 •^ 
 
 c 
 
 Vj 
 
 c 
 
 .u 
 
 u 
 
 
 
 u 
 
 s 
 
 01 
 
 
 
 ■^ 
 
 0; 
 
 X 
 
 ■J 
 
 
 ly 
 
 
 
 X X 
 
 
 
 u 
 
 o 
 
 o 
 
 ■<-i 
 
 3 
 
 
 
 
 
 
 CO 
 
 >-l 
 
 •r-i 
 
 iJ 
 
 ■^1 
 
 3 
 
 X 
 
 u 
 
 > 
 
 •r-l 
 
 3 -.n 
 
 
 
 •H 
 
 H 
 
 U-( 
 
 CO 
 
 CO 
 
 C4-I 
 
 VI 
 
 4-1 
 
 3: 
 
 0. 
 
 4-1 
 
 to 
 
 u 
 
 r-» — ' 
 
 < 
 
 
 
 ■r-l 
 
 U 
 
 C ■r-l 
 
 
 
 tn 
 
 
 
 CO 
 
 
 
 4-1 
 
 u 
 
 
 
 •r-l 
 
 
 .—1 
 
 —1 
 
 
 CO 
 
 4-1 
 
 CJ 
 
 -—4 
 
 
 
 O 
 
 , 
 
 )-i 
 
 1/ 
 
 
 u 
 
 
 
 CO 
 
 , 
 
 I-I 
 
 u 
 
 . 
 
 > 
 
 r-l 00 
 
 ■ 
 
 u 
 
 ■rl 
 
 (0 
 
 . X 
 
 
 
 o 
 
 a: 
 
 <u 
 
 1-1 
 
 Di 
 
 01 
 
 01 
 
 l-l 
 
 < 
 
 
 
 4-1 
 
 Di 
 
 ■r-l 
 
 1 
 
 
 
 4-1 
 
 I-I 
 
 1-1 
 
 -1 3 
 
 
 
 fl 
 
 
 D. 
 
 (J 
 
 
 tx 
 
 U-l 
 
 CO 
 
 
 D. 
 
 3 
 
 
 4-1 
 
 Oj 
 
 
 G 
 
 4-1 
 
 4-1 
 
 ex. 
 
 
 
 o 
 
 , 
 
 CO 
 
 c 
 
 
 CO 
 
 U-l 
 
 X 
 
 , 
 
 CO 
 
 y. 
 
 . 
 
 
 
 X 
 
 • 
 
 
 
 3 
 
 X 
 
 . c 
 
 
 
 u 
 
 00 
 
 a, 
 
 t— 1 
 
 a! 
 
 ft. 
 
 UJ 
 
 <J 
 
 
 
 CU 
 
 ~ 
 
 CQ 
 
 < 
 
 -< E-i 
 
 ''^ 
 
 
 
 x: 
 
 u; 
 
 Q :j 
 
 
 CD 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 3C 
 
 u-1 
 
 
 vO 
 
 
 rv 
 
 
 00 
 
 
 
 OX 
 
 
 
 
 
 =1 
 
 o 
 
 
 o 
 
 
 
 
 
 
 
 
 
 
 
 
 —1 
 
 
 o 
 
 
 o 
 
 
 
 
 
 
 
 
 
 S 
 
 
 
 
 
 CO 
 
 
 03 
 
 
 to 
 
 
 CO 
 
 
 
 ca 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ,- 
 
 II. 2-2 
 
a. 
 CO 
 
 0) 
 
 o 
 o 
 
 u 
 
 CO 
 
 a> 
 
 60 
 
 0) 
 
 o 
 o 
 
 IV, 
 0) 
 
 00 
 
 o 
 o 
 
 u 
 
 00 
 
 o 
 o 
 
 u 
 
 
 
 •rl 
 
 
 
 <« 
 
 ^ 
 
 ^ 
 
 >M 
 
 M 
 
 Vl 
 
 o 
 
 <0 . 
 
 CO 
 
 
 o. 
 
 CU 
 
 c 
 
 
 
 o 
 
 <u 
 
 o 
 
 1-1 
 
 M 
 
 30 
 
 w 
 
 di 
 
 o 
 
 •r4 
 
 ^^ 
 
 --t 
 
 > 
 
 f-H 
 
 rH 
 
 1-1 
 
 o 
 
 O 
 
 Q 
 
 o 
 
 O 
 
 V 
 
 o 
 o 
 
 
 en 
 
 
 J 
 
 
 M 
 
 
 H 
 
 >- 
 
 W 
 
 ac 
 
 M 
 
 -a: 
 
 
 cc 
 
 C5 
 
 CD 
 
 < 
 
 CD 
 
 k; 
 
 — 1 
 
 S 
 
 CQ 
 
 u 
 
 
 
 
 U) 
 
 
 a 
 
 
 o 
 
 g 
 
 c 
 o 
 
 o 
 
 CQ 
 
 O 
 Q. 
 
 00 
 
 o 
 
 6i 
 
 H to 
 
 CO 
 
 o 
 
 o 
 
 <0 
 
 u 
 
 w 
 
 a> 
 
 4J 
 
 > 
 
 B 
 
 •^^ 
 
 01 
 
 c: 
 
 o 
 
 =3 
 
 c 
 
 ^ — ' 
 
 o 
 
 
 u 
 
 4J 
 
 a o 
 
 (D* f—i 
 u O 
 
 o 
 
 a u 
 
 CL Q 
 
 
 a 
 -a 
 
 0) 
 
 O 
 OQ 
 
 l/I 
 
 OJ 
 
 5 
 
 CJ 
 
 o 
 
 c 
 
 TI 
 
 o 
 
 CO 
 
 •r-l 
 
 3 
 
 o 
 
 ■a 
 
 CO 
 
 0) 
 
 
 zc 
 
 T) 
 
 
 O 
 
 . 
 
 O 
 
 tL, 
 
 Ul, 
 
 CQ 
 
 >-) 
 
 o 
 
 G 
 
 O '-N 
 O H 
 
 i-l 
 
 O CO 
 
 c o 
 
 x; to r^ 
 
 t/3 0-, CO 
 
 •H I 
 
 Ci-i . LO 
 
 10 (1) 
 
 4-1 C 
 
 C vl 
 
 3 Vj 
 
 ■-I XJ 
 
 *-> c 
 
 t/J 3 
 c 
 o 
 
 o w 
 
 tj 
 
 o 
 > 
 
 en 
 
 o 
 
 CQ 
 
 ■V 
 
 a 
 
 CO 
 
 y 
 
 o 
 
 c 
 
 •r-l 
 
 lU 
 
 o 
 
 a 
 
 Si 
 <o 
 •1-1 
 
 U-l 
 
 o 
 
 CO 
 
 >! tfl 
 
 0-1 4-1 
 
 to 
 01 ce; 
 
 g CN 
 ^-^ .-I 
 
 00 
 
 o 
 
 -o 
 
 0) 
 
 CJ CM 
 
 . 
 
 to 
 
 < 
 
 a( 
 
 
 c 
 
 . 
 
 •r-J 
 
 o 
 
 "n 
 
 
 a 
 
 «% 
 
 aj 
 
 (/)• 
 
 <> 
 
 00 
 
 o 
 
 c 
 
 ^ 
 
 ^ 
 
 c 
 
 •^ 
 
 o 
 
 U 
 
 •r-l 
 
 V) 
 
 1-1 
 
 
 to 
 
 , 
 
 u 
 
 Pi 
 
 o 
 
 
 T3 
 
 . 
 
 iJ 
 
 CQ 
 
 U. 
 
 o 
 n 
 
 ON 
 I 
 
 00 
 
 O H 
 
 >— I 
 
 (« s: 
 
 03 — ' 
 
 c c 
 
 D to 
 
 e E 
 
 O TJ 
 
 to <U 
 
 il -r-l 
 
 o Uj 
 
 o 
 
 OQ 
 
 
 (0 
 
 
 
 •r-l 
 
 
 
 C! 
 
 
 
 M 
 
 
 
 O 
 
 
 
 U-( 
 
 
 " 
 
 -r-l 
 
 
 V 
 
 
 
 u 
 
 to 
 
 
 cv! 
 
 o 
 
 to 
 
 4_) 
 
 •4-i 
 
 ■ H 
 
 c 
 
 o 
 
 U^ 
 
 01 
 
 
 
 u 
 
 >, 
 
 01 
 
 
 XJ 
 
 r— 4 
 
 o 
 
 •H 
 
 o 
 
 u 
 
 to 
 
 ^ 
 
 c 
 
 ;j 
 
 
 
 > 
 
 E 
 
 
 ■H 
 
 o 
 
 c 
 
 OJ 
 
 -o 
 to 
 
 OJ 
 
 ■a 
 
 •H 
 
 ex 
 
 ■r-l 
 
 —I 
 
 to 
 
 OJ 
 
 to 
 
 J-i o 
 O r^ 
 O 
 
 CO tA 
 
 C 
 
 0) 
 
 01 
 
 > 
 
 u 
 
 OJ 
 
 a 
 
 o 
 -o 
 
 to 
 
 -a 
 
 OJ 
 
 o 
 
 DQ 
 
 ^ 
 
 o 
 
 CQ 
 
 u 
 
 w 
 
 a 
 
 o 
 
 > 
 
 CL 
 
 o 
 
 01 
 
 Di 
 
 Cti 
 
 U 
 
 V5 
 
 o 
 
 to 
 
 > 
 
 a 
 
 {-• 
 
 ^ 
 
 o 
 
 OQ 
 
 u 
 
 n 
 
 II. 2-3 
 
>- 
 a- 
 
 •cC 
 DC 
 
 S 
 
 I 
 
 rn 
 
 O- 
 
 C/2 
 
 Qi 
 
 < 
 
 H 
 O 
 
 a 
 o 
 ai 
 a, 
 
 ►— 
 
 
 
 ■ ■ 
 
 
 
 
 ■"■■"•" 
 
 
 k— «■ 
 
 
 ■— * 
 
 
 
 — "* 
 
 
 
 
 
 CJ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 -tC 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ct.- 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 >— 
 t/1 
 
 
 
 
 X 
 
 
 
 
 X 
 
 
 
 
 X 
 
 
 
 
 
 1 
 
 CQ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 «t 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 S 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 m 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 cz> 
 
 
 
 
 
 
 
 
 X 
 
 
 
 
 X 
 
 
 
 X 
 
 
 
 
 CJ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 s 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 JiS 
 
 ^ 
 
 
 ^ 
 
 
 
 
 -M 
 
 
 
 
 Ji 
 
 
 
 Jil 
 
 
 
 
 
 M 
 
 VI 
 
 
 M 
 
 
 
 
 n 
 
 
 
 
 V4 
 
 
 
 Vi 
 
 
 
 
 C3 
 
 Q 
 
 CO 
 
 
 CO 
 
 
 
 
 to 
 
 
 
 
 i" 
 
 
 
 ^ 
 
 
 
 
 Oi 
 
 (U 
 
 
 CU 
 
 
 
 
 PC 
 
 
 
 
 (Li 
 
 
 
 ■ a, 
 
 
 
 
 C3 
 
 OJ 
 
 0) 
 
 
 d) 
 
 
 
 
 tu 
 
 
 
 
 <U 
 
 
 
 a> 
 
 
 
 
 60 
 
 00 
 
 
 00 
 
 
 
 
 CO 
 
 
 
 
 CO 
 
 
 
 00 
 
 
 
 
 <U 
 
 a> 
 
 
 <u 
 
 
 
 
 CJ 
 
 
 
 
 a> 
 
 
 
 cu 
 
 
 
 
 t-i 
 
 >-i 
 
 
 .-( 
 
 
 
 
 r-( 
 
 
 
 
 i-i 
 
 
 
 ■— ( 
 
 
 
 
 i-t 
 
 f-H 
 
 
 .— 1 
 
 
 
 
 1— 1 
 
 
 
 
 ■-I 
 
 
 
 .-1 
 
 
 
 
 
 o 
 
 o 
 
 
 o 
 
 
 
 
 O 
 
 
 
 
 o 
 
 
 
 
 
 
 
 
 
 u 
 
 CJ 
 
 
 CD 
 
 
 
 
 o 
 
 
 
 
 o 
 
 
 
 CJ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 x; 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 4J 
 
 (0 
 
 
 
 
 
 
 
 
 1 
 
 
 
 
 
 
 
 
 
 
 
 4J 
 
 •r4 
 
 
 
 
 
 
 
 
 c 
 
 
 
 
 
 
 
 
 >» 
 
 
 
 CO 
 
 U-i 
 
 
 
 
 
 
 
 
 o 
 
 
 
 
 
 
 
 
 ua 
 
 
 
 ^4 
 
 
 
 
 
 
 
 
 
 2 
 
 
 4J 
 
 to 
 
 
 
 
 
 
 •o 
 
 
 
 4-1 
 
 d 
 
 d 
 
 •H 
 
 
 
 
 
 
 
 
 u 
 
 
 
 
 OJ 
 
 
 
 
 <u 
 
 
 
 01 
 
 
 
 
 
 
 
 
 
 O 
 
 
 -o 
 
 
 4J 
 
 
 
 
 u 
 
 
 
 
 
 V4 
 
 
 
 
 
 
 
 
 (1) 
 
 
 c 
 
 
 to 
 
 
 
 
 3 
 
 
 
 d 
 
 
 
 
 
 
 ar 
 
 
 
 
 u 
 
 
 to 
 
 
 u 
 
 
 
 
 T> 
 
 
 
 
 
 > 
 
 
 
 
 <=> 
 
 
 
 
 •H 
 
 
 
 
 4J 
 
 
 
 
 O 
 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 Q 
 
 
 
 
 c 
 
 (U 
 
 
 
 
 iJ 
 
 
 
 c 
 
 r-l 
 U4 
 
 
 
 
 C-i 
 
 
 
 
 CO 
 
 
 OJ 
 
 
 u 
 
 
 
 
 
 
 
 •r4 
 
 
 
 
 
 — 1 
 
 Ci3 
 
 
 
 
 
 
 •r< 
 
 
 c 
 
 
 
 
 cu 
 
 
 
 01 
 
 ■-H 
 
 
 
 
 
 
 
 > 
 
 
 o 
 
 
 o 
 
 
 
 
 .u 
 
 
 
 4-1 
 
 
 
 
 
 
 Z3 
 
 
 
 
 JD 
 
 
 o 
 
 
 a 
 
 
 
 
 to = 
 
 
 
 C 
 
 
 
 
 
 O. 
 
 
 
 
 
 
 M 
 
 
 
 
 
 
 u ^ 
 
 
 
 £ 
 
 > 
 
 
 
 
 
 
 « 
 
 
 
 0) 
 
 
 
 
 c 
 
 
 
 
 4.) tn 
 
 
 
 1-1 
 
 
 
 
 
 
 c 
 
 
 
 iJ 
 
 
 r-J 
 
 
 •H 
 
 
 
 
 C tn 
 
 
 
 
 4J 
 
 
 
 
 
 
 •r^ 
 
 
 
 to 
 
 
 CO 
 
 
 <U 
 
 
 
 
 QJ 3 
 
 
 
 rC 
 
 to 
 
 
 
 
 
 
 •o 
 
 
 
 >-l 
 
 
 o 
 
 
 4-) 
 
 
 
 
 o x: 
 
 
 
 '/) 
 
 
 
 
 
 
 
 (U 
 
 
 
 *_) 
 
 
 •w 
 
 
 O 
 
 
 
 
 C o 
 
 
 
 •r-l 
 
 p 
 
 
 
 
 
 
 (U 
 
 
 
 c 
 
 
 e 
 
 
 ^ 
 
 
 
 
 o 
 
 ^-N 
 
 
 U. 
 
 
 
 
 
 
 
 
 fl4 
 
 
 
 CJ 
 
 
 <u 
 
 
 
 
 
 u to 
 
 C/3 
 
 
 
 X 
 
 
 
 
 
 
 
 
 
 a 
 
 
 x: 
 
 o 
 
 
 
 
 
 •r-l 
 
 t, 
 
 
 ^-1 
 
 CJ 
 
 
 
 
 
 
 e 
 
 
 
 c 
 
 
 u 
 
 r^ 
 
 X, 
 
 
 
 
 C CJ 
 
 T. 
 
 
 
 
 
 
 
 
 
 
 u 
 
 
 
 o 
 
 
 
 a\ 
 
 to 
 
 
 
 
 •H >> 
 
 Z 
 
 
 H-l 
 
 (U 
 
 
 
 
 
 
 <u 
 
 
 
 CJ 
 
 
 c 
 
 .—1 
 
 •t-l 
 
 
 
 
 OJ .c 
 
 v^ 
 
 
 
 x: 
 
 
 
 
 
 
 H 
 
 
 
 
 
 CO 
 
 
 t-, 
 
 
 
 r^ 
 
 4-1 a 
 
 
 
 -0 4-J 
 
 
 
 
 
 
 
 
 
 c 
 
 
 u 
 
 « 
 
 
 
 
 (') 
 
 O u 
 
 M 
 
 
 01 
 
 
 
 
 
 
 
 M 
 
 
 
 •H 
 
 
 •r-l 
 
 01 
 
 T3 
 
 
 
 1 
 
 >■ ^ 
 
 o 
 
 
 u 
 
 XI 
 
 
 
 
 
 
 C 
 
 
 
 o 
 
 ^^ 
 
 U 
 
 o 
 
 ■-^ 
 
 
 
 <f 
 
 PL, 3 
 
 XI 
 
 
 •r-l 
 
 c 
 
 
 
 
 
 
 O 
 
 
 
 4J 
 
 0-1 
 
 <y 
 
 c 
 
 r— 1 
 
 
 
 t^ 
 
 '^.>' 
 
 >, 
 
 
 D 
 
 CO 
 
 
 
 
 
 
 kJ 
 
 
 
 o 
 
 Ci-, 
 2^ 
 
 ^ 
 
 OJ 
 
 
 CO 
 
 
 
 .. 
 
 to 01 
 
 d 
 
 CO 
 
 
 CT 
 
 0; 
 
 01 
 
 
 
 
 
 
 c 
 
 
 
 2; 
 
 
 tu 
 
 
 
 
 /-X 
 
 ■r-l J^ 
 
 
 
 Oi 
 
 Q 
 
 
 
 
 
 c^ 
 
 
 ■H 
 
 
 
 
 N->' 
 
 t4-l 
 
 U-l 
 
 c 
 
 
 
 r-l 
 
 fc-. to 
 
 . 
 
 
 
 
 
 
 
 . — . 
 
 o 
 
 
 
 
 
 x; 
 
 
 o 
 
 c 
 
 -r-l 
 
 
 
 S.^ 
 
 X 
 
 o 
 
 
 > 
 
 U 
 
 
 
 
 oo 
 
 CT. 
 
 
 
 o 
 
 
 
 U3 
 
 .-1 
 
 
 o 
 
 
 
 
 r^ 
 
 !2 
 
 
 
 4J 
 
 i-i 
 
 
 
 
 oc: 
 
 rH 
 
 U 
 
 
 CM 
 
 
 
 ■r-t 
 
 TD 
 
 XI 
 
 00 
 
 CJ 
 
 r-l 
 
 
 
 r-J 
 
 •r-4 X) 
 
 • 
 
 
 •rj 
 
 
 
 
 
 
 C3 
 
 PL, 
 
 
 fo 
 
 
 
 b- 
 
 O 
 
 o 
 
 c 
 
 
 O 
 
 
 
 
 0) 
 
 Q 
 
 
 ^4 
 
 01 
 
 
 
 
 ac 
 
 flH 
 
 
 
 
 
 
 X 
 
 c 
 
 •r-l 
 
 >, 
 
 r; 
 
 
 ^r^ 
 
 *• 
 
 'ji CC 
 
 
 
 3 
 
 N 
 
 
 
 
 (— 
 
 
 
 c 
 
 
 
 c 
 
 u 
 
 5^ 
 
 •U 
 
 OO 
 
 O 
 
 
 t/5 
 
 > 
 
 •o 
 
 M 
 
 
 0-, 
 
 < 
 
 
 
 
 =D 
 
 tM 
 
 
 •H 
 
 
 
 •r-l 
 
 o 
 
 
 0) 
 
 o 
 
 CJ 
 
 
 & 
 
 1-1 
 
 •r4 14-1 
 
 01 
 
 
 
 
 
 
 
 «C 
 
 O 
 
 
 
 
 
 
 ^J 
 
 , 
 
 <u 
 
 f— 1 
 
 
 
 u 
 
 o o 
 
 i-^ 
 
 
 4-1 
 
 < 
 
 
 
 
 
 
 
 m 
 
 
 
 CU 
 
 
 s 
 
 ^ 
 
 o 
 
 < 
 
 r 
 
 zs 
 
 ■J) 
 
 < 
 
 u 
 
 
 d 
 
 PU 
 
 
 
 
 
 (0 
 
 
 0) 
 
 
 
 XI 
 
 OJ 
 
 
 
 c 
 
 
 >> 
 
 ^-^ 
 
 •r4 
 
 c 
 
 <u 
 
 
 0) 
 
 M 
 
 
 
 
 
 Cfl 
 
 
 u 
 
 
 
 •r-l 
 
 •a 
 
 , 
 
 , — 1 
 
 -C 
 
 , — 1 
 
 
 
 e 
 
 o o 
 
 c 
 
 
 > 
 
 
 
 
 
 
 lU 
 
 
 c 
 
 
 
 M 
 
 o 
 
 o 
 
 to 
 
 CJ 
 
 >, 
 
 "q. 
 
 c 
 
 (U 
 
 C •-! 
 
 ^ 
 
 (X3 
 
 1—4 
 
 XI 
 
 
 
 
 
 B 
 
 
 CO 
 
 
 
 
 
 p 
 
 
 c 
 
 (U 
 
 c 
 
 to 
 
 3 
 
 Si 
 
 •r-l 4-1 
 
 CO 
 
 r-^ 
 
 
 
 (U 
 
 
 
 
 
 <u 
 
 
 XJ 
 
 
 
 D 
 
 4J 
 
 r^ 
 
 o 
 
 H 
 
 
 
 1-1 
 
 O 
 
 CJ 
 
 E CJ 
 
 3: 
 
 1 
 
 C/3 
 
 u 
 
 
 
 
 
 e 
 
 
 (0 
 
 
 
 r-l 
 
 O 
 
 ^ 
 
 ■r-l 
 
 
 u 
 
 ^0 
 
 1-1 
 
 
 -5 to 
 
 
 
 
 
 <u 
 
 
 
 
 
 o 
 
 
 J3 
 
 
 
 U, 
 
 Oi 
 
 o 
 
 U 
 
 tn 
 
 a 
 
 
 
 03 
 
 •-a 
 
 )-i 
 
 
 r~- 
 
 '4-4 
 
 > 
 
 
 
 
 
 (fl 
 
 
 D 
 
 
 
 
 1 — 1 
 
 o 
 
 ra 
 
 V 
 
 o 
 
 4-) 
 
 
 
 CO 4-1 
 
 < 
 
 
 
 
 
 
 
 u 
 
 
 
 (U 
 
 '-'^ 
 
 to 
 
 
 
 U-l 
 
 LU 
 
 N) 
 
 z 
 
 ■r4 
 
 V) 
 
 2 
 
 . 
 
 ■a 
 
 d X 
 
 
 
 
 
 
 
 I-I 
 
 
 
 .-( 
 
 H 
 
 
 
 
 O 
 
 
 
 
 1-4 
 
 )-i 
 
 
 r-3 
 
 o 
 
 •r-l UJ 
 
 . 
 
 •— s 
 
 d 
 
 01 
 
 ^ 
 
 
 
 
 
 o 
 
 i-i 
 
 M 
 
 
 
 
 n 
 
 . 
 
 u 
 
 "J 
 
 
 r\ 
 
 
 r\ 
 
 JJ 
 
 
 
 — ' 
 
 Q 
 
 .•^ 
 
 :-! 
 
 ^ ' 
 
 
 
 ^ 
 
 <e_ 
 
 •r^ 
 
 ^— .^ 
 
 
 c 
 
 -a 
 
 u 
 
 to 
 
 x: 
 
 14-4 
 
 U 
 
 • 
 
 ul 
 
 •r4 r— I 
 
 
 ^— ■' 
 
 •H 
 
 
 tt 
 
 0; 
 
 
 
 s-^ 
 
 X 
 
 H 
 
 
 o 
 
 •r-) 
 
 
 
 U5 
 
 O 
 
 JZ 
 
 2 
 
 
 6 O 
 
 *• 
 
 r^ 
 
 4J 
 
 cj: 
 
 1-1 
 
 pli 
 
 
 
 
 U 
 
 o 
 
 1— 1 
 
 4-1 
 
 •H 
 
 u 
 
 , 
 
 -a 
 
 ■r4 
 
 
 tj 
 
 
 r— 1 
 
 •r-l X 
 
 tfl 
 
 CJN 
 
 CO 
 
 d 
 
 XJ 
 
 
 
 UlJ 
 
 <u 
 
 0) >-i 
 
 H 
 
 y^ 
 
 (-1 
 
 iJ 
 
 o 
 
 UJ 
 
 QJ 
 
 Ci-, 
 
 C 
 
 
 »» 
 
 to 
 
 J 
 
 00 
 
 
 3 
 
 •-4 
 
 d 
 
 >^ 
 
 
 
 ^ 
 
 C O 
 
 
 ^-^ 
 
 o 
 
 ra 
 
 3 
 
 
 u 
 
 
 o 
 
 X) 
 
 . 
 
 Vj 
 
 CJ 
 
 d 
 
 r 
 
 . — I 
 
 to 
 
 
 
 ^■^ U 
 
 
 tJ 
 
 u ex 
 
 U-l 
 
 
 Q. 
 
 c 
 
 r— < 
 
 •t 
 
 C 
 
 o 
 
 •r-l 
 
 
 1-1 
 
 D 
 
 <« r-l 
 
 •r-l 
 
 d 
 
 to 
 
 :::i 
 
 tj 
 
 H 3 
 
 
 
 <U OJ 
 
 o 
 
 ^ 
 
 C' 
 
 •-^ 
 
 u. 
 
 i 
 
 tu 
 
 •r-l 
 
 4-) 
 
 D 
 
 •-> 
 
 4-1 
 
 o < 
 
 f—* 
 
 
 
 > 
 
 
 C 
 
 1-4 E 
 
 
 
 c 
 
 XI si 
 
 
 Pi 
 
 p 
 
 
 en 
 
 4-1 
 
 to 
 
 V 
 
 
 ^^ 
 
 
 r— 1 
 
 •r4 
 
 ClJ 
 
 d 
 
 r) 
 
 >^ c: 
 
 
 
 o 
 
 :i 
 
 .—J 
 
 * E 
 
 
 p 
 
 c 
 
 <u 
 
 a 
 
 c 
 
 c 
 
 •r-l 
 
 r. 
 
 D 
 
 01 -H 
 
 •r-l 
 
 4-J 
 
 
 
 
 6 
 
 ^-^ 
 
 
 
 
 0) 4-1 
 
 r: 
 
 -5 
 
 *J 
 
 a 
 
 o 
 
 
 u 
 
 to 
 
 •r4 
 
 kJ 
 
 x: 
 
 U 
 
 CJ >. 
 
 iJ 
 
 'r^ 
 
 r— 1 
 
 ■-4 
 
 
 C/5 
 
 
 
 tn 
 
 J^ u 
 
 •^ 
 
 i» 
 
 o 
 
 u 
 
 •^ 
 
 y. 
 
 Q. 
 
 , — ( 
 
 c 
 
 1 
 
 4J 
 
 'f-i 
 
 d o. 
 
 CO 
 
 u 
 
 CO 
 
 u 
 
 d 
 
 .^ 
 
 
 
 OJ 
 
 n] 
 
 u 
 
 •H 
 
 C3 
 
 o 
 
 u 
 
 
 
 4-< 
 
 fj 
 
 V) 
 
 •r4 
 
 u 
 
 01 O 
 
 
 4J 
 
 
 
 
 
 •r-l 
 
 u ^ 
 
 
 
 •H 
 
 • ^ 
 
 c; 
 
 >-l 
 
 w 
 
 a 
 
 r: 
 
 , 
 
 W 
 
 < 
 
 o 
 
 to 
 
 G 
 
 OC 
 
 D M 
 
 
 3 
 
 •r-l 
 
 D 
 
 01 
 
 .-4 to 
 
 
 
 •o 
 
 r: ij 
 
 OJ 
 
 u- 
 
 u 
 
 
 , — 1 
 
 CQ 
 
 OJ 
 
 
 4J 
 
 ;j 
 
 OO 
 
 < 
 
 cr a 
 
 a: 
 
 >'. 
 
 K 
 
 X3 
 
 4J 
 
 3 D 
 
 
 
 D 
 
 c 
 
 4-1 
 
 
 c 
 
 a 
 
 o 
 
 
 a. 
 
 0) 
 
 1) 
 
 
 
 
 iJ o 
 
 
 
 01 
 
 
 
 
 
 d 
 
 
 
 j-i 
 
 • o 
 
 o 
 
 , 
 
 o 
 
 !5 
 
 ■■fl 
 
 , 
 
 to 
 
 x: 
 
 Q 
 
 >i 
 
 , 
 
 , 
 
 C/) iSl 
 
 . 
 
 . 
 
 X 
 
 M 
 
 1-4 
 
 . d 
 
 
 
 t/3 
 
 0-1 u 
 
 a, 
 
 ►J 
 
 u 
 
 M 
 
 Oti 
 
 a. 
 
 4-1 
 
 -■ 
 
 -O 
 
 (X, 
 
 <-) 
 
 Z I— 1 
 
 03 
 
 ^ 
 
 CJ 
 
 a- 
 
 D- 
 
 LO < 
 
 
 C3 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 z: 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 =i 
 
 o\ 
 
 o 
 
 
 rH 
 
 
 
 
 CN 
 
 
 
 
 CO 
 
 
 
 sr 
 
 
 
 
 l-l 
 
 <M 
 
 
 rsi 
 
 
 
 
 CNI 
 
 
 
 
 cs 
 
 
 
 CM 
 
 
 
 
 "t: 
 
 o 
 
 O 
 
 
 O 
 
 
 
 
 O 
 
 
 
 
 O 
 
 
 
 
 
 
 
 
 " 
 
 BQ 
 
 □3 
 
 
 03 
 
 
 
 
 CQ 
 
 
 
 
 03 
 
 
 
 CQ 
 
 
 
 
 ,, II. 2-4 
 

 to 
 
 
 u 
 
 
 n 
 
 >- 
 
 C-l 
 
 rn 
 
 w 
 
 Ou 
 
 M 
 
 ■sC 
 
 tj 
 
 cc 
 
 H 
 
 ej3 
 
 r T 
 
 c:> 
 
 < 
 
 1 
 
 CO 
 
 
 
 
 CO 
 
 O 
 
 C_J 
 
 M 
 
 lX. 
 
 C) 
 
 U- 
 
 n 
 
 
 n 
 
 
 o 
 
 g 
 
 ,_ 
 
 
 
 
 
 ■ ■ ' 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 C-i 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 •a; 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Cfl 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 X 
 
 
 
 y— 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 cr> 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ra 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 _-=£ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 a^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 _j 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Ca- 
 
 
 
 
 
 
 
 
 
 
 
 
 J 
 
 
 
 
 
 
 
 CD 
 
 
 
 
 
 
 
 X 
 
 
 
 
 
 
 X 
 
 
 
 
 
 
 C-l 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ;h 
 
 
 
 
 
 
 
 " 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 j«: 
 
 
 ^ 
 
 
 .a! 
 
 
 ^ 
 
 
 
 ji: 
 
 
 
 
 
 
 (0 
 
 
 
 IS' 
 
 
 ^ 
 
 
 
 
 CO 
 
 
 
 vi 
 
 
 
 ISC 
 C3 
 
 (X, 
 
 
 
 oa 
 
 
 
 a: 
 
 
 CL, 
 
 
 Oi 
 
 
 a, 
 
 
 
 cu 
 
 
 
 ►— 
 
 i) 
 
 O 
 
 u 
 
 
 
 01 
 
 
 
 <u 
 
 
 tt) 
 
 
 <u 
 
 
 <u 
 
 
 
 <u 
 
 
 
 ■^ 
 
 
 
 00 
 
 
 
 00 
 
 
 00 
 
 
 00 
 
 
 30 
 
 
 
 10 
 
 
 
 C_i 
 
 
 
 v 
 
 
 
 0) 
 
 
 01 
 
 
 n 
 
 
 01 
 
 
 
 01 
 
 
 
 o 
 
 
 
 l-l 
 
 
 
 i—i 
 
 
 1— 1 
 
 
 l-l 
 
 
 l-l 
 
 
 
 .—1 
 
 
 
 —1 
 
 
 
 I-l 
 
 
 
 r-l 
 
 
 r~i 
 
 
 l-l 
 
 
 l-l 
 
 
 
 l-l 
 
 
 
 
 
 
 o 
 
 
 
 O 
 
 
 O 
 
 
 o 
 
 
 o 
 
 
 
 o 
 
 
 
 
 
 
 o 
 
 
 
 O 
 
 
 U 
 
 
 o 
 
 
 o 
 
 
 
 o 
 
 
 
 
 o 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 u 
 
 
 
 
 
 l-H 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 c 
 
 
 
 
 
 •r4 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 o 
 
 
 
 
 
 U 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 •r4 
 
 
 
 
 
 w 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 u 
 
 
 
 
 
 r-H 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 m 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 - D 
 
 
 
 
 
 > 
 
 
 
 
 
 
 
 
 
 
 
 
 M 
 
 
 
 r-l 
 
 ro 
 
 
 
 
 
 c 
 
 
 
 
 
 
 
 
 
 
 
 
 U) 
 
 
 
 > 
 
 
 
 
 z: 
 
 o 
 
 
 
 c 
 
 
 
 
 
 
 
 
 
 OJ 
 
 
 
 w 
 
 
 
 
 C3 
 
 
 
 
 w 
 
 
 
 
 
 
 
 10 
 
 
 
 o 
 
 
 
 
 
 
 
 -o: 
 
 CO 
 
 (0 
 
 c 
 o 
 
 
 
 l-l 
 
 
 
 U CO 
 
 
 
 
 •H 
 
 
 
 o 
 
 
 
 •— 1 
 ro 
 o 
 
 
 
 
 
 •H 
 
 
 
 3 
 
 
 
 •H lx^ 
 
 
 
 
 
 
 
 
 
 
 •r4 
 
 
 
 
 —J 
 
 ■u 
 
 
 
 l-l 
 
 
 
 
 
 
 
 > 
 
 
 
 01 
 
 
 
 fi 
 
 
 
 
 O- 
 
 •--1 
 
 
 
 o 
 
 
 
 
 
 
 XI 
 
 
 
 c 
 
 
 
 o 
 
 
 
 
 ■V 
 
 
 
 CO 
 
 
 
 3 v^ 
 
 
 
 
 « 
 
 
 
 •r^ 
 
 
 
 x: 
 
 
 
 
 c 
 
 
 
 
 
 
 C 
 
 
 <u 
 
 
 
 CQ 
 
 
 
 1— 1 
 
 
 
 u 
 
 
 
 
 
 o 
 
 
 
 .c 
 
 
 
 
 
 4-1 
 
 
 
 
 
 
 ro 
 
 
 
 
 
 
 
 
 o 
 
 
 
 CO 
 
 1-1 
 
 
 
 0) i^ 
 
 - n 
 
 
 « 
 U 
 
 
 
 <U 
 
 
 
 r-l 
 
 
 
 -a 
 
 c 
 
 
 
 
 
 c 
 
 
 
 fa 
 
 
 
 l-l 
 
 
 iJ 
 
 
 
 u 
 
 
 
 < 
 
 
 
 ro 
 
 
 
 
 
 o 
 
 
 
 
 
 
 D 
 
 
 c 
 
 
 
 Q) 
 
 
 
 
 ^— s 
 
 
 « 
 
 
 
 
 
 •H 
 
 
 
 ■ c 
 
 
 
 •r^ u 
 
 
 a 
 
 
 
 l-l 
 
 
 
 ^ 
 
 CO 
 
 
 
 
 
 
 u 
 
 
 
 <u 
 
 
 
 U 01 
 
 
 o 
 
 
 
 <U 
 
 
 
 Cij 
 
 
 4_» 
 
 
 
 
 
 o 
 
 
 
 T3 
 
 
 
 OJ .—1 
 
 
 c 
 
 
 
 c 
 
 
 
 
 ^ 
 
 
 ro 
 
 
 
 
 
 N 
 
 
 
 n 
 
 
 
 U 1— 1 
 
 
 o 
 
 
 
 i) 
 
 
 
 • 
 
 JC 
 
 
 >H 
 
 
 
 
 
 --X •'H 
 
 
 
 A 
 
 
 
 n) -r^ 
 
 
 a 
 
 
 
 a 
 
 
 
 i-H 
 
 ^-^ 
 
 
 4-1 
 
 
 
 
 
 -o t-" 
 
 
 
 c 
 
 
 
 S S 
 
 
 
 
 
 
 
 
 
 
 
 C 
 
 
 
 
 
 c 
 
 
 
 <u 
 
 
 
 
 
 c 
 
 
 
 c 
 
 
 
 1 
 
 tn 
 
 
 01 
 
 c 
 
 
 
 ) U ZJ 
 
 
 
 x: 
 
 
 
 '< • 
 
 
 •t-i 
 
 
 
 •r-« 
 
 
 
 
 ^' 
 
 
 ; i 
 
 
 
 - 
 
 
 u ^ 
 
 
 
 
 
 
 qj X 
 
 
 OJ 
 
 
 
 (U 
 
 
 
 a 
 
 4-1 - 
 
 
 C 
 
 •1^ 
 
 
 • 
 
 
 O — ' 
 
 
 
 (n 
 
 
 
 Oi / 
 
 r— ( 
 
 4-1 
 
 
 
 ■U 
 
 
 
 4J 
 
 ro 
 
 
 o 
 
 4J 
 
 
 
 
 u o 
 
 
 
 1— 1 
 
 
 
 
 ro 
 
 o 
 
 
 
 o 
 
 
 
 . ra 
 
 CO 
 
 
 u 
 
 O 
 
 
 
 
 — ' en 
 
 
 
 
 /-S 
 
 
 in C 
 u 5 
 
 1 
 
 ^ 
 
 
 
 <ii 
 
 
 
 l-l 
 
 4-1 
 
 , 
 
 
 c 
 
 3 
 
 
 
 
 a\ 
 
 Q 
 
 
 o 
 
 
 CO 
 
 
 M O 
 
 CN 
 
 
 
 
 
 
 
 C 
 
 Oh 
 
 
 ■r-l 
 
 
 
 
 
 _ 
 
 ^ 
 
 
 
 r^ 
 
 T3 
 
 1 
 
 
 J-i 
 
 
 Xi 
 
 
 
 •a 
 
 
 
 01 
 
 
 
 01 
 
 ^ 
 
 
 
 I/? 
 
 a\ 
 
 C 
 
 
 CTi 
 
 d 
 
 
 to oa 
 
 ,, 
 
 0) 
 
 
 
 § 
 
 
 
 o 
 
 • 
 
 
 4J 
 
 
 
 or: 
 
 f—t 
 
 -.-1 
 
 
 r-H 
 
 n) 
 
 
 ro 
 
 ^,~^ 
 
 •r-l 
 
 
 
 
 
 c 
 
 Di 
 
 
 o 
 
 
 
 
 > 
 
 
 
 y 
 
 •v^ 
 
 
 <u J 
 
 
 l^ 
 
 
 
 1— 1 
 
 
 
 o 
 o 
 
 •* 
 
 
 £ 
 
 je 
 
 
 
 1 — 
 
 u 
 
 
 
 •r^ 
 
 <D 
 
 
 C 
 
 ^^ 
 
 H-l 
 
 
 
 to 
 
 
 
 
 *'-\ 
 
 
 
 b-i 
 
 
 
 =5 
 
 •r-( 
 
 
 
 4J 
 
 U) 
 
 
 c9 • 
 
 CM 
 
 O 
 
 
 
 E 
 
 ^N 
 
 
 c 
 
 H 
 
 
 x; 
 
 
 
 
 •«C 
 
 u 
 
 
 
 c 
 
 
 
 
 00 z; 
 
 ro 
 
 
 
 
 >H 
 
 >, 
 
 
 •rl 
 
 ^-l 
 
 
 en 
 
 V-i 
 
 
 
 
 <2 
 
 
 
 (0 
 
 M 
 
 
 
 
 > 
 
 
 
 o 
 
 4-1 
 
 
 01 
 
 2 
 
 ON 
 
 ■r-l 
 
 O 
 
 
 
 
 
 
 1— 1 
 
 ^ 
 
 
 OJ 
 
 •^ 
 
 j-i 
 
 
 
 ■z. 
 
 •r-l 
 
 
 4-J 
 
 ■! 
 
 o> 
 
 u, 
 
 U-l 
 
 
 
 
 
 
 
 4J 
 
 1 
 
 
 "S 4J 
 
 3 
 
 •r-l 
 
 
 
 
 7) 
 
 
 O 
 
 
 1 
 
 
 
 
 
 
 u 
 
 
 
 < 
 
 
 OJ 
 
 ,-1 
 
 
 
 l-l 
 
 U 
 
 
 'C 
 
 c 
 
 ^D 
 
 C 
 
 -o 
 
 
 
 
 a 
 
 
 
 
 
 
 (n 
 
 •r^ 
 
 (0 
 
 
 
 o 
 
 o 
 
 
 ^-< 
 
 ON 
 
 ■r-l 
 
 o 
 
 
 
 
 OJ 
 
 ^-s 
 
 
 tl-l 
 
 , 
 
 
 >. -l-l 
 
 > 
 
 D 
 
 
 
 4-1 
 
 > 
 
 
 
 01 
 
 
 
 l-l 
 
 
 
 
 > 
 
 H 
 
 
 o 
 
 sr 
 
 LA 
 
 -H Vj 
 
 (^ 
 
 O 
 
 
 
 
 ■r^ 
 
 
 sz 
 
 x: 
 
 
 V-i 
 
 ■r-l 
 
 
 
 
 .-1 
 
 M 
 
 
 
 
 vO 
 
 ^ 03 
 
 05 
 
 
 
 
 ^ 
 
 C 
 
 
 in 
 
 <i; 
 
 
 o 
 
 3 
 
 
 
 
 o 
 
 s 
 
 
 c 
 
 K 
 
 
 r3 P-1 
 
 
 c 
 
 ^^ 
 
 
 3 
 
 
 ■r4 
 
 
 UO 
 
 > 
 
 a 
 
 4-1 
 
 
 
 C/2 
 
 s-x 
 
 
 
 
 »-l 
 
 , , 
 
 3 
 
 en 
 
 'r~f 
 
 H 
 
 
 Du 
 
 
 
 U. 
 
 
 
 ro 
 
 01 
 
 1-1 
 
 
 
 
 y 
 
 
 
 •r-( 
 
 a 
 
 ^^% 
 
 u 
 
 >u 
 
 0) 
 
 l-l 
 
 
 
 .—1 
 
 
 
 "^ 
 
 <t 
 
 .-H 
 
 Oi 
 
 o 
 
 
 
 u-l 
 
 • 
 
 
 iJ 
 
 t-* 
 
 n 
 
 V-l Oi 
 
 •r^ 
 
 i~i 
 
 >^ 
 
 
 U_( 
 
 r-l 
 
 
 U-l 
 
 
 rvj 
 
 Ci^ 
 
 
 c 
 
 
 
 o 
 
 u. 
 
 f— < 
 
 
 •r\ 
 
 1— ( 
 
 ro 
 
 •r-( 
 
 Vj 
 
 o 
 
 
 
 O 
 
 01 
 
 
 O 
 
 " 
 
 
 
 > 
 
 :i 
 
 
 
 
 
 r^ 
 
 
 w" 
 
 •^ 
 
 
 r> • 
 
 ^ 
 
 £ 
 
 
 
 
 C 
 
 
 
 ^ 
 
 - 
 
 •■t-t 
 
 — 
 
 ^ 
 
 
 
 U) 
 
 U-l 
 
 
 iJ 
 
 o 
 
 2^ 
 
 <N| 
 
 UJ 
 
 x: 
 
 3 
 
 4-1 
 
 11 
 
 u 
 
 ij 
 
 c 
 
 3 
 
 > 
 
 o 
 
 ■r-l 
 
 
 
 UJ 
 
 0) 
 
 
 
 u 
 
 >-l 
 
 Q. 
 
 
 <t 
 
 >^ 
 
 •/) 
 
 
 m 
 
 l-l 
 
 3 
 
 o 
 
 l-l 
 
 o 
 
 ro 
 
 U 
 
 
 Vi 
 
 >^ 
 
 
 
 o 
 
 
 V 
 
 o 
 
 G 
 
 
 
 u 
 
 — 1 
 
 U_) 
 
 x; 
 
 o 
 
 .— 1 
 
 o 
 
 o 
 
 ■r-l 
 
 -Q 
 
 o 
 
 ^^ 
 
 3 
 
 '-N 1-1 
 
 
 c 
 
 (/) 
 
 
 a. 
 
 o 
 
 C 
 
 
 —I 3 
 
 P-. 
 
 O 
 
 in 
 
 CL 
 
 ro 
 
 ^-^ 
 
 ex 
 
 iJ 
 
 C 
 
 ^H 
 
 u 
 
 (X 
 
 H ro 
 
 
 0) 
 
 u 
 
 I-. 
 
 01 
 
 o 
 
 
 
 -■ O 
 
 
 
 E 
 
 01 
 
 ^ 
 
 
 :j 
 
 ro 
 
 :j 
 
 c 
 
 4-1 
 
 
 t~M E 
 
 
 
 D 
 
 C 
 
 '/) 
 
 OS 
 
 
 « 
 
 tn 
 
 ID U 
 
 . — I 
 
 C 
 
 •l-l 
 
 oc 
 
 
 ■o 
 
 csi 
 
 M 
 
 c 
 
 c 
 
 X 
 
 4-1 
 
 -*. E 
 
 
 
 ,^ 
 
 <u 
 
 n3 
 
 
 .— ( 
 
 w 
 
 U-J 
 
 D -Q 
 
 a 
 
 O 
 
 )-i 
 
 
 a 
 
 c 
 
 
 ■r-l 
 
 d 
 
 XI 
 
 ■r4 
 
 c 
 
 ^^ 3 
 
 
 
 U-l 
 
 D 
 
 Vj 
 
 u 
 
 nj 
 
 o 
 
 U-* 
 
 ^5 
 
 •rH 
 
 ■ rj 
 
 o 
 
 u 
 
 > 
 
 o 
 
 u 
 
 •—1 
 
 ro 
 
 o 
 
 E 
 
 G 
 
 CO 
 
 
 
 C 
 
 u 
 
 a> 
 
 O 
 
 ti 
 
 )-l 
 
 a 
 
 o 
 
 i-t 
 
 in 
 
 o 
 
 •i-i 
 
 a, 
 
 O 
 
 ■r-l 
 
 H 
 
 c^ 
 
 i> 
 
 > 
 
 _•<: 
 
 
 
 1-1 
 
 • ^( 
 
 c 
 
 t;j 
 
 -r-l 
 
 o 
 
 i-j 
 
 •s\ 
 
 1-1 
 
 ^ 
 
 
 Q 
 
 •u 
 
 
 ro 
 
 XI 
 
 
 f-l 
 
 x: 
 
 ^H 
 
 U r-l 
 
 
 
 
 u 
 
 rj 
 
 u 
 
 c 
 
 
 c/; 
 
 
 iJ 
 
 o 
 
 , 
 
 >-l 
 
 •^ 
 
 
 l-l 
 
 3 
 
 
 
 (LJ 
 
 
 
 •rJ (TJ 
 
 
 
 a 
 
 in 
 
 D^ 
 
 u 
 
 o 
 
 c}5 
 
 •a 
 
 k ^" 
 
 
 
 , — 1 
 
 C/5 
 
 4J 
 
 l-l 
 
 d 
 
 u 
 
 rH 
 
 ai 
 
 T3 
 
 
 CO 
 
 :2 3 
 
 
 
 c 
 
 C 
 
 
 c 
 
 j:: 
 
 
 ■J 
 
 CD 
 
 
 a 
 
 i-l 
 
 
 C 
 
 o 
 
 
 o 
 
 ly 
 
 
 c: 
 
 
 
 
 
 o 
 
 
 o 
 
 o 
 
 , 
 
 'J 
 
 Cu 
 
 
 
 > 
 
 , 
 
 o 
 
 3 
 
 . 
 
 O 
 
 j/1 
 
 . 
 
 
 
 -C 
 
 U-( 
 
 . c 
 
 
 
 in 
 
 u 
 
 UJ 
 
 CJ 
 
 
 •-) 
 
 u^ 
 
 I Q 
 
 o 
 
 W 
 
 ?: 
 
 u 
 
 X. 
 
 •^ 
 
 u 
 
 " 
 
 CO 
 
 u- 
 
 H 
 
 o 
 
 CJ < 
 
 
 CD 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ac 
 
 u- 
 
 > 
 
 
 vO 
 
 
 
 r^ 
 
 
 00 
 
 
 ON 
 
 
 o 
 
 
 
 I— 1 
 
 
 
 =1 
 
 -X 
 CJ 
 
 C\ 
 
 1 
 
 
 Ol 
 
 
 
 CM 
 
 
 CM 
 
 
 (S 
 
 
 
 
 CO 
 
 
 
 c 
 
 > 
 
 1 
 
 
 s 
 
 
 
 O 
 
 
 o 
 
 CO 
 
 
 s 
 
 
 O 
 
 CQ 
 
 
 
 O 
 
 crt 
 
 
 
 II. 2-5 
 

 n 
 
 
 o 
 
 
 t-t 
 
 >- 
 
 H 
 
 rr: 
 
 c/: 
 
 o_ 
 
 1— 1 
 
 •*£ 
 
 tj 
 
 ar. 
 
 ^ 
 
 CJ3 
 
 r 1 
 
 CD 
 
 <- 
 
 1 
 
 « 
 
 CQ 
 
 <, 
 
 
 U, 
 
 CQ 
 
 o 
 
 a 
 o 
 
 1 — 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CJ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 -t 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 cs: 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 X 
 
 
 
 
 
 X 
 
 
 
 
 
 
 
 
 X 
 
 
 
 
 
 ca 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 -c 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^^^_ 
 
 • 
 
 
 rt 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 —J 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 i^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CD 
 
 X 
 
 
 
 
 
 X 
 
 
 
 
 X 
 
 
 
 
 X 
 
 
 
 
 
 CJ> 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 :h 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ij: 
 
 
 
 ^ 
 
 
 ^ 
 
 
 A! 
 
 
 ^ 
 
 
 
 
 J£ 
 
 
 
 u 
 
 to 
 fa 
 
 
 
 u 
 
 
 
 M 
 
 
 Vl 
 
 
 V4 
 
 
 u 
 
 
 
 
 U 
 
 
 
 
 C3 
 
 !D 
 
 
 
 <o 
 
 
 r: 
 
 
 (\J 
 
 
 CO 
 
 
 
 
 CO 
 
 
 
 
 a. 
 
 
 
 a. 
 
 
 PL. 
 
 
 CLi 
 
 
 (U 
 
 
 
 
 tU 
 
 
 
 
 
 a; 
 
 
 
 di 
 
 
 OJ 
 
 
 (U 
 
 
 <u 
 
 
 
 
 CD ■ 
 
 
 
 41 
 SO 
 
 <y 
 1— 1 
 
 
 00 
 
 
 
 00 
 
 
 00 
 
 
 00 
 
 
 60 
 
 
 
 
 00 
 
 
 
 
 <u 
 
 
 
 (U 
 
 
 (U 
 
 
 <u 
 
 
 01 
 
 
 
 
 <U 
 
 
 
 
 ^^ 
 
 
 
 t-i 
 
 
 ■-I 
 
 
 1— 1 
 
 
 ■-I 
 
 
 
 
 l-l 
 
 
 
 
 
 
 f—i 
 
 
 
 r-l 
 
 
 ,-1 
 
 
 f— 1 
 
 
 l-l 
 
 
 
 
 f-i 
 
 
 
 
 
 o 
 
 
 
 o 
 
 
 
 O 
 
 
 o 
 
 
 o 
 
 
 o 
 
 
 
 
 o 
 
 
 
 
 
 o 
 
 
 
 O 
 
 
 o 
 
 
 u 
 
 
 o 
 
 
 
 
 o 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 <u 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Xi 
 
 
 
 
 
 
 
 
 
 x: 
 
 
 
 
 
 -a 
 
 
 
 
 
 
 «J 
 
 
 
 
 
 
 
 
 
 w 
 
 
 
 
 
 0) 
 
 
 J3 
 
 
 
 
 
 
 
 
 
 
 c 
 
 
 
 •r^ 
 
 
 
 
 
 ^4 
 
 
 CO 
 
 
 
 
 c 
 
 
 
 
 
 
 •r-l 
 
 
 
 t. 
 
 
 
 
 
 CO 
 
 
 •l-l 
 
 
 
 
 •H 
 
 
 
 
 
 
 (U 
 
 
 
 
 
 
 
 
 D 
 
 
 b 
 
 
 
 
 
 
 
 
 
 
 iJ 
 
 
 
 G 
 
 
 
 
 
 <U 
 
 
 
 
 
 
 (U 
 
 
 
 
 
 7^ 
 
 o 
 
 
 
 ° 
 
 
 
 
 
 <ii 
 
 
 ■l-l 
 
 
 
 
 4J 
 CO 
 
 
 
 
 
 C3 
 
 
 
 C 
 
 
 
 
 
 
 
 
 
 
 
 u 
 
 
 
 
 
 ►— 
 
 
 
 
 o 
 
 
 
 
 
 0) 
 
 
 <u 
 
 
 
 
 4J 
 
 
 
 
 
 tC 
 
 A 
 
 
 
 •r^ 
 
 
 
 
 
 w 
 
 
 •o 
 
 
 
 
 a 
 
 
 
 
 
 C_J 
 
 w 
 
 
 
 4-1 
 
 
 
 
 
 CO 
 
 
 ■r4 
 
 
 
 
 <u 
 
 
 
 
 
 
 ■A 
 
 
 
 o 
 
 
 
 
 
 Vi 
 
 
 Vl 
 
 
 
 
 o 
 
 
 
 
 
 CQ 
 
 U-, 
 
 
 
 < 
 
 
 
 
 
 4J 
 
 
 o 
 
 
 
 
 c 
 
 
 
 
 
 ZD 
 
 
 
 
 
 
 
 
 
 C 
 
 
 3 
 
 
 
 
 o 
 
 
 
 
 
 O. 
 
 o 
 
 
 
 o 
 
 •A 
 
 4-) 
 
 
 
 
 
 o 
 
 c 
 o 
 
 
 60 
 
 
 
 vO 
 
 1 
 
 u 
 
 d 
 
 •r^ 
 
 
 
 
 
 
 
 
 1— 1 
 
 
 
 
 
 o 
 
 
 c 
 
 
 
 ■ ■ 
 
 0> 
 
 
 
 01 
 
 4-1 
 
 CO 
 
 
 
 
 •r-l 
 
 
 
 o 
 
 
 
 
 
 
 
 •r-l 
 
 
 
 y^^ 
 
 4-1 
 
 
 
 
 
 
 ^H 
 
 
 
 cj 
 
 
 
 
 
 c 
 
 
 C 
 
 
 
 r— 1 
 
 o 
 
 
 
 
 
 
 •rj 
 
 
 
 ■r7 
 
 
 
 
 
 •r^ 
 
 
 •^^ 
 
 
 
 ^^y 
 
 ^ 
 
 
 
 u 
 
 
 
 
 XI 
 
 
 
 -J 
 
 
 
 
 
 o 
 
 
 E 
 
 
 
 r-l 
 
 
 
 
 
 
 nj 
 
 
 
 
 
 
 
 
 4J 
 
 
 u 
 
 
 
 to 
 
 
 
 
 d 
 
 01 
 CJ 
 
 d 
 
 
 
 
 
 4-1 
 
 
 
 -0 
 
 
 
 
 
 o 
 
 
 0) 
 
 
 
 
 X 
 
 
 
 
 
 
 '-v CO 
 
 
 
 c 
 
 
 
 
 
 ^ 
 
 
 ■U 
 
 
 
 to 
 
 tn 
 
 •r-l 
 
 ,^ 
 
 
 
 < 
 
 o 
 
 
 0) 
 
 
 
 
 
 
 fj 
 
 
 
 
 
 Q 
 
 
 
 .u 
 
 fa 
 
 CO 
 
 
 
 c -1^ 
 
 
 
 .;:::; 
 
 
 
 CO 
 
 
 
 In 
 
 
 VI 
 
 
 
 •r^ 
 
 U-l 
 
 "Z 
 
 
 u 
 
 
 ^■•4 
 
 
 O ^ 
 
 
 
 4-1 
 
 
 
 u 
 
 
 
 •^^ 
 
 
 o 
 
 
 
 e 
 
 o 
 
 2 
 
 
 d 
 
 
 t4-l 
 
 
 a •r^ 
 
 
 
 > 
 
 5^^ 
 
 
 •o 
 
 
 
 Cl4 
 
 
 >+-l 
 
 
 
 OJ 
 
 
 v_^ 
 
 
 
 o 
 
 
 ^-^ u 
 
 
 
 ,—1 
 
 >^ 
 
 
 C 
 
 
 
 
 
 
 
 
 x: 
 
 01 
 
 
 
 
 
 
 4-1 
 
 
 
 o 
 
 4-1 
 
 
 <u 
 
 
 
 c 
 
 
 •a 
 
 
 
 u 
 
 XI 
 
 tn 
 
 
 
 :^o 
 
 
 o 
 
 D 
 
 
 
 0) 
 
 •r-l 
 
 
 o 
 
 
 
 •w 
 
 
 o 
 
 
 
 
 •r-l 
 
 00 
 
 
 o 
 
 
 c 
 
 . 
 
 r- 
 
 Z 
 
 
 
 4-1 
 
 '/) 
 
 
 c 
 
 
 
 
 
 x: 
 
 
 
 ,—4 
 
 U 
 
 d 
 
 
 
 CO 
 
 ON 
 
 
 
 •-^ 
 
 o 
 
 U 
 
 
 o 
 
 ^-^ 
 
 
 "O 
 
 
 u 
 
 
 
 to 
 
 O 
 
 •r-l 
 
 
 
 ij 
 
 cr: 
 
 1 — 1 
 
 -o 
 
 
 CO 
 
 £ 
 
 0) 
 
 
 u 
 
 n) 
 
 
 •r^ 
 
 r-N 
 
 <D 
 
 
 
 o 
 
 3 
 
 •—1 
 
 
 
 
 0) 
 
 C3 
 
 C 
 
 
 tu 
 
 > 
 
 
 
 •r4 
 
 
 c 
 
 CO 
 
 s 
 
 
 
 •--1 
 
 r-H 
 
 r-l 
 
 
 -1— ( 
 
 
 CJ 
 
 3: 
 
 tU 
 
 
 io 
 
 
 •r4 
 
 
 d 
 
 d 
 
 
 •r-l 
 
 tn 
 
 
 
 •-N 
 
 4J 
 
 fa 
 
 •r^ 
 
 
 
 
 h— 
 
 
 
 Vj 
 
 c 
 
 
 •A 
 
 u 
 
 
 -J 
 
 g 
 
 cu 
 
 
 c/^ 
 
 > 
 
 
 4-1 
 
 
 
 •''-• 
 
 =3 
 
 r— 1 
 
 
 ^-^ CO 
 
 o 
 
 :=) 
 
 
 OJ 
 
 o 
 
 
 
 c 
 
 
 Ci. 
 
 
 0) 
 
 CO 
 
 
 U-l 
 
 CO 
 
 
 
 "X 
 
 nj 
 
 
 vO 
 
 4-1 
 
 
 
 4-1 
 
 u-l 
 
 
 1— 1 
 
 N— •' 
 
 ■r-l 
 
 
 g 
 
 t\3 
 
 X 
 
 
 
 
 
 
 o 
 
 
 (0 1 
 
 
 n) 
 
 
 O 
 
 •r^ 
 
 
 tfl 
 
 
 Vj 
 
 
 d 
 
 4-1 
 
 , 
 
 CT> 
 
 ^ 
 
 
 ■rV 
 
 
 OO r^ 
 
 
 •H 
 
 
 i: 
 
 (—1 
 
 
 
 
 d 
 
 . O 
 
 
 ^w.* 
 
 < 
 
 
 02 
 
 t3> 
 
 • l— J 
 
 ^ 
 
 
 G 
 
 
 C r-^ 
 
 cn 
 
 XI 
 
 
 to 
 
 
 XI 
 
 o 
 
 3 
 
 
 
 
 U-l 
 
 
 CN4 
 
 
 
 ^ 
 
 
 OJ 
 
 
 •r-l v£) 
 
 £ 
 
 e 
 
 
 
 o 
 
 
 ■r^ 
 
 tn 
 
 ,—1 
 
 
 0) 
 
 1 — 1 
 
 O 
 
 . 
 
 1 
 
 O 
 
 •r-1 
 
 
 
 ^ 
 
 
 r-l 
 
 w 
 
 :3 
 
 
 rC 
 
 
 
 (0 
 
 1-1 
 
 Ci, 
 
 
 to 
 
 to 
 
 
 CQ 
 
 rn 
 
 iJ 
 
 
 o 
 
 
 ,—1 .. 
 
 •rl 
 
 .—1 
 
 
 W 
 
 u-l 
 
 
 01 
 
 d 
 
 01 
 
 
 
 o 
 
 •A 
 
 >. 
 
 
 <Ti 
 
 
 j::: 
 
 
 
 
 •H --X 
 
 c 
 
 o 
 
 
 ■r-l 
 
 o 
 
 
 Pi 
 
 o 
 
 -o 
 
 u 
 
 
 M 
 
 o 
 
 4-1 
 
 •» 
 
 CS 
 
 o 
 
 
 4_J 
 
 
 <u 
 
 
 4J U-l 
 
 rt 
 
 o 
 
 
 [1. 
 
 
 
 
 ■r-l 
 
 d 4-1 
 
 < 
 
 
 _D 
 
 •f~A 
 
 •r-l 
 
 to 
 
 
 00 
 
 
 xi 
 
 
 CO '—^ 
 
 oi: 
 
 s— ' 
 
 
 
 >^ 
 
 
 i) 
 
 4J 
 
 < c 
 
 o 
 
 
 1 
 
 u-l 
 
 .—1 
 
 to 
 
 ■ • 
 
 J^ 
 
 
 •U 
 
 
 LO 
 
 ^j 
 
 
 
 U-l 
 
 4J 
 
 
 x: 
 
 t1 
 
 •r-l 
 
 < 
 
 r 
 
 U-l 
 
 •r-l 
 
 CJ 
 
 ^^ 
 
 , cc 
 
 
 
 
 
 • n 
 
 o 
 
 3 
 
 
 O 
 
 •r-l 
 
 
 4_) 
 
 d 
 
 . ^1 
 
 
 OJ 
 
 
 o 
 
 X 
 
 3 
 
 1^ 
 
 r 
 
 
 j_i 
 
 
 c 
 
 r 
 
 Qi 
 
 o 
 
 -C 
 
 
 
 U) 
 
 
 
 o 
 
 rJ O 
 
 o 
 
 u 
 
 , 
 
 
 to 
 
 M 
 
 ^-^ 
 
 
 
 o 
 
 
 o 
 
 '-N 
 
 
 1-1 
 
 a 
 
 
 C 
 
 u 
 
 
 o 
 
 ■r-l 
 
 t/l 
 
 X 
 
 to 
 
 1t^ 
 
 u-l 
 
 r— 1 
 
 
 O 
 
 U-J 
 
 t— ^ 
 
 
 
 
 K 
 
 *" * ' 
 
 
 
 ^^ 
 
 ^ 
 
 
 U"J 
 
 4-r 
 
 3 
 
 4-, 
 
 l-l 
 
 
 
 
 •r-l 
 
 • 
 
 O 
 
 O 
 
 ' 
 
 
 
 JJ 
 
 CO- U 
 
 •r-l 
 
 . 
 
 4-1 
 
 •r-l 
 
 > 
 
 4-1 
 
 to 
 
 o 
 
 rr: d 
 
 
 u 
 
 • 
 
 
 to 
 
 -C 
 
 1— 1 
 
 
 Ixi 
 
 03 
 
 u. 
 
 o 
 
 -C 
 
 < 
 
 1-1 
 
 4-) 
 
 'A 
 
 U 
 
 o 
 
 t\; 
 
 to 
 
 u-l 
 
 r^ 
 
 ^ 
 
 d 
 
 > 
 
 
 
 ^ 
 
 d 
 
 —J 
 
 (= 
 
 i! 
 
 v-^ 
 
 
 
 
 o 
 
 •A 
 
 c 
 
 o 
 
 Vj 
 
 1-1 
 
 •s ^ 
 
 
 
 01 
 
 
 O 
 
 < 
 
 • 
 
 •• 
 
 o 
 
 •r-l 
 
 c 
 
 -^ 
 
 a; 
 
 O -I 
 
 4^ 
 o 
 
 U5 
 
 D. 
 0) 
 
 O 
 
 := 
 
 ex 
 
 a) 
 
 o 
 
 [i^ 
 
 —1 
 
 d 
 
 o 
 
 d 
 
 i 
 
 •r-l 
 
 4-1 
 
 ,—1 
 
 CO 
 
 d 
 o 
 
 o 
 
 '-0 
 
 
 l« 
 
 ij 
 
 u o 
 
 
 r— 1 
 
 a; 
 
 C 
 
 
 oi 
 
 o 
 
 tn 
 
 a u 
 
 
 
 c 
 
 to 
 
 to 
 
 •• 
 
 ■r-l 
 
 r3 
 
 ^ 
 
 
 
 
 nj 
 
 Xi CO 
 
 ofl 
 
 OJ 
 
 
 e 
 
 4-1 
 
 
 
 D 
 
 > -C 
 
 ■r-l 
 
 a 
 
 o 
 
 •A 
 
 o 
 
 d 
 
 4-1 
 
 
 
 N] 
 
 u 
 
 
 
 M 
 
 p 
 
 c 
 
 o 
 
 4-1 
 
 o 
 
 u 
 
 4J 
 
 o 
 
 O 
 
 to (/) 
 
 4-1 
 
 
 —4 
 
 u 
 
 •r-l 
 
 ■r-l 
 
 •A 
 
 •1-4 
 
 ex 
 
 
 u 
 
 4-) 
 
 Q T3 
 
 ."r-l 
 
 (X. 
 
 o 
 
 o 
 
 
 
 y 
 
 4-1 
 
 CJ 
 
 ai ■H 
 
 to 
 
 d 
 
 ^ 
 
 o 
 
 SO 
 
 .-■.i 
 
 U 
 
 V-i 
 
 
 
 
 o 
 
 C 
 
 o 
 
 C 
 
 
 nj 
 
 
 O 
 
 to 
 
 
 3 
 
 f— 1 
 
 4-1 
 
 ■r-l 
 
 
 1/1 
 
 
 
 a 
 
 4-1 
 
 
 v^ 
 
 
 0) 
 
 <u 
 
 • O 
 
 -J 
 
 • 
 
 ^4 
 
 X) 
 
 1 — 1 
 
 w 
 
 > 
 
 cr 
 
 . XI 
 
 D. 
 
 o 
 
 . 
 
 to 
 
 < — 1 
 
 •r-l 
 
 3 
 
 i-) 
 
 6 
 
 -U 
 
 
 M-l 
 
 o 
 
 J tu, 
 
 o 
 
 < 
 
 4-) 
 
 •r-l 
 
 o 
 
 4-1 
 
 -o 
 
 < 
 
 x; 3 
 
 to 
 
 u 
 
 CQ 
 
 < 
 
 : 
 
 Nl 
 
 2 
 
 n. 
 
 
 y-i 
 
 c 
 
 
 1-1 
 
 
 c 
 
 G 
 
 
 C 
 
 D 
 
 
 D. 
 
 ■o 
 
 o 
 
 
 
 •r-l 4-1 
 
 
 
 OJ 
 
 
 o 
 
 
 u 
 
 o 
 
 • • 
 
 o 
 
 • 
 
 o 
 
 •r-l 
 
 ■ 
 
 o 
 
 AJ 
 
 > 
 
 • d 
 
 _< 
 
 cfc 
 
 • 
 
 • 
 
 oa ro 
 
 • 
 
 ■ 
 
 f,l 
 
 a, 
 
 
 - 
 
 u 
 
 a -) 
 
 CO 
 
 o 
 
 o 
 
 ►J 
 
 a: 
 
 CJ 
 
 CO 
 
 X) 
 
 w :j 
 
 - 
 
 ct: 
 
 ■-) 
 
 - 02 
 
 l-l 
 
 •-5 
 
 
 
 CD 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ae 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 =d 
 
 CM 
 
 
 
 r^ 
 
 
 <f 
 
 
 ITl 
 
 
 vO 
 
 
 
 
 r^ 
 
 
 
 00 
 
 
 O 
 
 
 
 rn 
 
 
 n 
 
 
 r^ 
 
 
 n 
 
 
 
 
 to 
 
 
 
 tn 
 
 
 -c 
 
 O 
 
 
 
 O 
 
 
 o 
 
 
 O 
 
 
 o 
 
 
 
 
 O 
 
 
 
 o 
 
 
 c_> 
 
 CQ 
 
 
 
 33 
 
 
 CO 
 
 
 CQ 
 
 
 CQ 
 
 
 
 
 CQ 
 
 
 
 OQ 
 
 
 II. 2-6 
 
to 
 
 re 
 cu 
 
 H 
 O 
 
 CO 
 
 
 a 
 o 
 a: 
 a. 
 
 >•>»" 
 
 
 
 
 
 
 
 '^■"" 
 
 ""^"^ 
 
 
 *"*"* 
 
 
 ■M-w^in 
 
 **■"*" 
 
 
 
 
 
 *"■■ 
 
 """ ' " ■ 
 
 
 
 
 <_> 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 «I 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 oc 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ►— 
 
 
 
 X 
 
 
 
 X 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 1^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 •-C 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 nr 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ••- 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 — * 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 . . 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 C3 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 C_3 
 
 
 
 X 
 
 
 
 
 
 
 
 
 
 
 X 
 
 
 
 
 
 
 X 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 J<! 
 
 
 
 
 
 
 .:i: 
 
 
 
 
 
 ^ 
 
 
 Ji 
 
 
 
 j<: 
 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 J3 
 
 
 
 
 
 u 
 
 
 M 
 
 
 
 M 
 
 
 
 
 
 Vl 
 
 
 04 
 
 
 
 
 
 
 0. 
 
 
 
 
 2: 
 
 « 
 
 
 CO 
 
 
 
 to 
 
 
 
 
 
 (Q 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 (X 
 
 
 a. 
 
 
 
 a< 
 
 
 
 
 
 CU 
 
 
 ID 
 
 
 
 
 
 
 V 
 
 
 
 
 ^— 
 
 
 
 
 
 
 
 
 
 
 
 
 
 00 
 
 
 
 
 
 
 50 
 
 
 
 
 «r 
 
 0) 
 
 
 
 
 
 
 «j 
 
 
 
 
 
 u 
 
 
 o 
 
 
 
 
 
 
 41 
 
 
 
 
 CO 
 
 JO 
 
 
 00 
 
 
 
 CO 
 
 
 
 
 
 00 
 
 
 f-l 
 
 
 
 
 
 
 f-l 
 
 
 
 
 CD 
 
 <u 
 
 
 <u 
 
 
 
 4) 
 
 
 
 
 
 v 
 
 
 t-l 
 
 
 
 
 
 
 l-l 
 
 
 
 
 
 t— ( 
 
 
 .-1 
 
 
 
 t-H 
 
 
 
 
 
 l-l 
 
 
 
 
 
 
 
 
 
 
 CJ 
 
 
 
 
 
 .-( 
 
 
 rH 
 
 
 
 r-l 
 
 
 
 
 
 f-l 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ■■^rfr 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 , 
 
 
 
 CO 
 (50 
 C 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 •r^ 
 
 
 
 ^ 
 
 
 
 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 r-l 
 
 
 
 r— 1 
 
 
 
 ?s 
 
 
 
 
 
 
 
 
 
 
 
 u^ 
 
 
 
 
 
 
 TD 
 
 
 1—1 
 
 
 
 •r-l 
 
 
 
 
 
 c 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 4-) 
 
 
 •r^ 
 4J 
 
 
 
 s 
 
 
 
 CO 
 
 ■rH 
 
 1) 
 
 
 
 
 
 
 
 
 
 
 W 
 
 
 
 
 
 
 nj 
 
 
 CO 
 
 
 
 e 
 
 
 
 XI 
 
 4-1 
 
 
 •« 
 
 
 
 
 
 
 
 
 *-J 
 
 
 
 
 
 
 »-H 
 
 
 
 
 
 -r-l 
 
 
 
 cu 
 
 
 
 
 • 
 
 
 
 
 U-l 
 
 
 
 
 1— ( 
 
 
 
 
 
 
 
 
 
 , 
 
 
 
 ^ 
 
 
 
 4-1 
 
 ^ 
 
 
 u 
 
 
 
 
 
 
 
 
 
 D 
 
 
 
 
 
 
 CO 
 
 
 oi 
 
 
 
 00 
 
 
 
 CO 
 
 
 '^ 
 
 
 
 
 
 
 
 ■^ 
 
 CO 
 
 
 
 
 
 
 h- 1 
 
 
 
 
 
 
 
 
 t-H 
 
 
 
 
 
 
 
 >, 
 
 
 
 a 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 ci; 
 
 
 
 
 
 CU 
 
 
 «^ 
 
 
 
 
 4J 
 
 
 
 CO 
 
 oi 
 
 
 0) 
 
 
 
 
 -0 
 
 
 CQ 
 
 
 
 
 
 CO 
 
 rC 
 
 
 Vh 
 
 
 
 
 •H 
 
 
 
 ^ 
 
 
 
 u 
 
 
 
 
 C 
 
 
 
 
 
 •H 
 
 
 
 M 
 
 4-1 
 
 
 CU 
 
 
 
 
 CO 
 
 
 
 •It 
 
 ' 
 
 
 c 
 
 
 
 
 CO 
 
 
 M 
 
 
 
 V^ 
 
 
 
 
 
 
 ^ 
 
 
 
 
 U 
 
 
 
 c_> 
 
 > 
 
 
 0) 
 
 
 
 
 
 
 « 
 
 
 
 Q 
 
 
 1-4 
 
 XI 
 
 UH 
 
 
 r-l 
 
 
 
 
 CU 
 
 
 
 u 
 
 
 u 
 
 
 
 
 CU 
 
 
 u 
 
 
 
 
 
 1— 1 
 
 G 
 
 
 
 
 CO 
 
 
 
 
 > 
 
 
 
 —J 
 
 D 
 
 
 q; 
 
 ~ 
 
 
 
 4-t 
 
 
 •-) 
 
 
 
 # 
 
 
 M 
 
 CO 
 
 
 
 :x 
 
 
 
 
 •rH 
 
 
 
 CO 
 
 
 
 
 y-i 
 
 c 
 
 
 
 CO 
 
 
 
 
 
 1) 
 
 
 
 
 G 
 
 
 
 
 
 
 G 
 
 
 
 rD 
 
 W 
 
 
 G 
 
 •r-l 
 
 
 
 l-l 
 
 
 ^ 
 
 
 
 iJ 
 
 
 CO 
 
 4J 
 
 
 
 
 • 
 
 
 
 r 
 
 := 
 
 
 
 cu 
 
 2: 
 
 
 
 U 
 
 0) 
 
 
 
 
 
 4J 
 
 c 
 
 OJ 
 
 CO 
 
 to 
 
 
 
 
 to 
 
 4-1 
 
 
 CO 
 
 OJ 
 
 u 
 
 4J 
 
 CO 
 
 u 
 
 •rH 
 4-1 
 
 CO 
 
 
 til 
 
 
 
 4.1 
 
 CO 
 
 
 
 
 
 G 
 
 
 1— ( 
 CD 
 
 i: 
 
 
 
 
 
 c 
 
 a, 
 
 .—1 
 CO 
 
 
 
 G 
 <U 
 
 
 CO 
 
 c 
 
 u 
 
 G 
 
 l-l 
 
 
 tq 
 
 
 
 u 
 
 4J 
 
 
 
 
 
 — ^ 
 
 
 CJ 
 
 
 
 
 
 
 
 1^ 
 
 
 
 
 
 
 
 
 1) 
 
 c 
 
 
 •» 
 
 
 
 c 
 
 CJ 
 
 
 
 
 c 
 
 
 •r-l 
 
 >> 
 
 
 
 
 
 >, 
 
 
 
 
 C 
 
 
 u 
 
 
 
 u 
 
 
 -^ 
 
 
 
 cu 
 
 r-4 
 
 
 
 
 •r-l 
 
 
 10 
 
 (-1 
 
 
 
 
 X) 
 
 , 
 
 
 
 
 
 
 
 c 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 C-) 
 
 
 
 E 
 
 
 
 
 c^ 
 
 
 <f 
 
 c 
 
 CO 
 
 t. 
 
 
 
 CJ 
 
 
 c 
 
 
 
 K 
 
 
 
 
 
 
 d 
 
 
 00 
 
 
 
 M 
 
 
 ^H 
 
 4-1 
 
 
 r-l 
 
 •r4 
 
 133 
 
 
 
 
 
 
 
 
 u 
 
 Cli 
 
 
 d 
 
 
 
 
 
 • 
 
 r^ 
 
 
 
 0) 
 
 
 
 
 cy 
 
 
 -<^ 
 
 OJ 
 
 
 , 
 
 
 
 c 
 
 
 •r-l 
 
 
 
 
 CO 
 
 
 
 
 
 C/1 
 
 1 
 
 
 
 ti 
 
 
 S 
 
 ••A 
 
 
 
 4J 
 
 14-1 
 
 w 
 
 
 
 •r^ 
 
 
 4-1 
 
 c 
 
 •^1 
 
 11 
 
 
 
 
 
 r- 
 
 
 CN 
 
 
 
 Q 
 
 
 
 
 
 
 r-l 
 
 k 
 
 '^ 
 
 ^ 
 
 
 
 Zj 
 
 
 1^ 
 
 
 
 >^ 
 
 
 H 
 
 
 
 •rH 
 
 X 
 
 r^ 
 
 
 
 
 
 CJ 
 
 "4-1 
 
 
 <f 
 
 x: 
 
 )-l 
 
 
 
 
 
 
 u 
 
 XJ 
 
 N) 
 
 
 
 
 
 CU 
 
 
 
 
 
 V^ 
 
 
 H 
 
 
 
 
 
 
 4-1 
 
 0) 
 
 
 
 k 
 
 
 D 
 
 
 
 c 
 
 
 • 
 
 ^•"S 
 
 
 ■u 
 
 •» 
 
 • • 
 
 
 
 
 
 
 
 
 ^->^ 
 
 ,, 
 
 x: 
 
 3 
 
 ■0 
 
 
 
 
 z 
 
 £ 
 
 w 
 
 
 >-> 
 
 ^ 
 
 
 
 
 d 
 
 ^■^ 
 
 
 
 
 . — r 
 
 ^ 
 
 
 CO 
 
 
 
 CO 
 
 >— ^ 
 
 CO 
 
 
 
 >, 
 
 
 
 
 
 
 
 
 
 Cu 
 
 
 £ 
 
 cu 
 
 CN 
 
 
 (/> 
 
 r-^ 
 
 
 
 <y 
 
 
 
 tu 
 
 CXI 
 
 •r-l 
 
 1-1 
 
 G 
 
 
 
 x; 
 
 
 11 
 
 
 
 
 
 •* 
 
 S 
 
 
 cu 
 
 f—i 
 
 
 
 a^ 
 
 OJ 
 
 
 •r-l 
 
 u 
 
 g 
 
 v.-* 
 
 r., 
 
 
 
 CO 
 
 
 
 CO 
 
 ,— ^ 
 
 V-j 
 
 x: 
 
 •rH 
 
 
 V4 
 
 z 
 
 
 
 > 
 
 *•— ' 
 
 
 ol: 
 
 
 ^ 
 
 
 u 
 
 z> 
 
 00 
 
 
 
 
 
 
 •r-l 
 
 CO 
 
 OJ 
 
 CO 
 
 4-1 
 
 
 - 
 
 v-^ 
 
 
 -C 
 
 
 00 
 
 
 cz> 
 
 '"' D 
 
 
 0) 
 
 
 
 v_^ 
 
 <f 
 
 M 
 
 1-1 
 
 , 
 
 
 
 bn 
 
 Lu 
 
 x; 
 
 •rH 
 
 CO 
 
 
 X 
 
 
 
 CO 
 
 u 
 
 ■<r 
 
 
 re 
 
 -0 
 
 
 jn 
 
 C/0 
 
 
 
 c 
 
 
 
 3 
 
 
 , ( 
 
 
 g 
 
 CL 
 
 t^ 
 
 CU 
 
 
 >, 
 
 10 
 
 
 •rH 
 
 CU 
 
 
 
 » — 
 
 CI 
 
 
 CO 
 
 
 , 
 
 
 •r-l 
 
 >4-( 
 
 
 
 Ov 
 
 C+-1 
 
 CO 
 
 
 l-l 
 
 
 c 
 
 CO 
 
 
 Ct, 
 
 Q 
 
 •> 
 
 
 rD 
 
 
 
 ^~N 
 
 — ^ 
 
 J. 
 
 1-1 
 
 >, 
 
 01 
 
 
 ^ 
 
 
 in 
 
 
 
 
 •r-l 
 
 •» 
 
 a 
 
 
 CO 
 
 c 
 
 
 
 
 > 
 
 
 «s 
 
 
 
 CO 
 
 tu, 
 
 4J 
 
 ■-) 
 
 4J 
 
 CO 
 
 CO 
 
 .--N 
 
 
 1 
 
 
 
 e 
 
 c 
 
 § 
 
 
 
 •rH 
 
 
 
 <4-l 
 
 d 
 
 4J 
 
 
 
 t 
 
 U-. 
 
 
 yc 
 
 
 
 
 CO 
 
 4-1 
 
 >, 
 
 
 r^ 
 
 OJ 
 
 r~\ 
 
 
 
 •rH 
 
 r 
 
 • 
 
 ^H 
 
 
 
 
 
 CO 
 
 11 
 
 
 
 4J 
 
 ^^ 
 
 u 
 
 
 r. 
 
 ■r-l 
 
 
 
 0) 
 
 4-1 
 
 
 00 
 
 D 
 
 r-l 
 
 X 
 
 OJ 
 
 CL, 
 
 20 
 
 "^ 
 
 ,-H 
 
 CM 
 
 
 > 
 
 •rH 
 
 
 
 
 1—1 
 
 ■-^ 
 
 •r-l 
 
 nj 
 
 , — 1 
 
 
 
 
 •r-l 
 
 •r-l 
 
 
 in 
 
 
 0) 
 
 
 CO 
 
 
 •rH 
 
 
 ■rH 
 
 ,—1 
 
 CO 
 
 
 u 
 
 
 
 ' — 1 
 
 
 ^^ 
 
 4-1 
 
 
 JO 
 
 
 
 U-l 
 
 Q 
 
 CO 
 
 
 
 "3 
 
 3 
 
 c 
 
 CO 
 
 -0 
 
 Cu, 
 
 * 
 
 4J 
 
 1 
 
 -a 
 
 • 
 
 
 
 
 
 w 
 
 J= 
 
 
 C 
 
 . - 
 
 
 
 CO 
 
 
 
 
 u 
 
 
 ,, 
 
 > 
 
 ■a 
 
 M 
 
 
 
 G 
 
 
 r-N 
 
 CO 
 
 f-l 
 
 •■-1 
 
 -1 
 
 CO 
 
 
 
 CJ 
 
 1-1 
 
 >> 
 
 CD 
 
 1) 
 
 c 
 
 
 
 T3 
 
 CLI 
 
 
 ^•^ 
 
 
 •r4 
 
 aj 
 
 
 CO 
 
 >^ 
 
 • 
 
 
 a\ 
 
 a. 
 
 
 
 
 
 •H 
 
 •r-l 
 
 11 
 
 r— 1 
 
 4-) 
 
 i<i 
 
 (/■ 
 
 G 
 
 •-( 
 
 > 
 
 
 ro 
 
 -^ 
 
 CO 
 
 4-1 
 
 Ci-I 
 
 
 .Q 
 
 > 
 
 > 
 
 t—* 
 
 -^ 
 
 •» 
 
 CO 
 
 
 
 iJ 
 
 3 
 
 f— 1 
 
 U 
 
 CTJ 
 
 
 4- 
 
 c 
 
 p 
 
 •r^ 
 
 
 
 CO 
 
 
 CO 
 
 
 
 -JG 
 
 CO 
 
 •r4 
 
 Oi 
 
 ,—1 
 
 rJ 
 
 >^ 
 
 4-1 
 
 
 
 > 
 
 
 ^ 
 
 < 
 
 1-1 
 
 , 
 
 (0 
 
 'rA 
 
 cr 
 
 c 
 
 
 CO 
 
 4J 
 
 
 CJ 
 
 
 4J 
 
 CQ 
 
 G 
 
 
 
 
 u 
 
 CO 
 
 
 
 1— < 
 
 w 
 
 
 
 
 4-1 
 
 s: 
 
 •r^ 
 
 i-l 
 
 •r4 
 
 rl 
 
 
 CO 
 
 G 
 
 Q 
 
 2 
 
 G 
 
 3 
 
 
 ID 
 
 ■ 
 
 • • 
 
 d 
 
 ^H 
 
 ■r4 
 
 
 
 c: 
 
 1* 
 
 r3 
 
 x: 
 
 c 
 
 
 E 
 
 CO 
 
 J 
 
 
 
 
 
 
 
 
 
 
 
 
 X> 
 
 
 CO 
 
 ^— V 
 
 
 
 CO 
 
 E 
 
 
 
 c 
 
 -rH 
 
 -.^ 
 
 i-j 
 
 u 
 
 , 
 
 (LI 
 
 N 
 
 
 .—1 
 
 
 
 E 
 
 ^ 
 
 ij 
 
 '.-^ 
 
 1-1 
 
 01 
 
 ,— J 
 
 
 CT^ 
 
 
 X 
 
 V 
 
 
 
 < 
 
 ~J 
 
 
 ^ 
 
 u 
 
 CJ 
 
 
 --4 
 
 c 
 
 
 
 , 
 
 
 
 >' 
 
 ro 
 
 4-1 
 
 
 
 X 
 
 --1 
 
 ■a 
 
 ^■-^ 
 
 1) 
 
 
 x: 
 
 
 
 
 3 
 
 • 
 
 • — 1 
 
 c 
 
 
 
 
 , — 1 
 
 •H 
 
 cu 
 
 r*s 
 
 11 
 
 
 
 
 CO 
 
 
 ■-0 
 
 
 
 G 
 
 r-4 
 
 CO 
 
 • 
 
 
 
 
 
 XI 
 
 4„) 
 
 ^ 
 
 
 
 
 ^ 
 
 
 •r-I 
 
 
 c 
 
 CO 
 
 
 
 a 
 
 ^ 
 
 -T3 
 
 M 
 
 >-i 
 
 •rH 
 
 M 
 
 to 
 
 
 
 CO 
 
 < 
 
 
 
 
 
 
 CO 
 
 
 -a 
 
 CJ 
 
 CO 
 
 r— < 
 
 4J 
 
 c 
 
 l-l 
 
 u- 
 
 
 G 
 
 1) 
 
 CO 
 
 OJ 
 
 •r-l 
 
 
 
 V4 
 
 Vj 
 
 
 r-l 
 
 ^H 
 
 
 f— 1 
 
 
 LkJ 
 
 u 
 
 
 • 
 
 <u 
 
 
 ^ 
 
 .^ 
 
 ■Zi 
 
 •-I 
 
 
 
 ^ 
 
 1> 
 
 D 
 
 c 
 
 10 
 
 4-1 
 
 -H 
 
 >— 1 
 
 3 
 
 
 
 * 
 
 
 
 
 * 
 
 •rH 
 
 
 ' 
 
 nj 
 
 ■D 
 
 33 
 
 ij 
 
 c 
 
 C 
 
 
 
 
 CD 
 
 CJ 
 
 ;7, 
 
 ■r4 
 
 CO 
 
 
 c: 
 
 c 
 
 •rH 
 
 
 
 
 u 
 
 ' — 1 
 
 •» 
 
 4-1 
 
 -r-* 
 
 C 
 
 
 1 — 
 
 E 
 
 > 
 
 
 d 
 
 •r-l 
 
 •r-l 
 
 
 11 
 
 4-) 
 
 
 
 
 
 
 CO 
 
 •r^ 
 
 11 
 
 4J 
 
 G 
 
 G 
 
 ^w^ 
 
 1— 1 
 
 c 
 
 oO 
 
 _-<: 
 
 
 
 ►— 
 
 
 
 •r^ 
 
 ^ 
 
 OJ 
 
 'i( 
 
 r-l 
 
 c 
 
 > 
 
 
 
 
 
 CO 
 
 u 
 
 > 
 
 
 CO 
 
 r) 
 
 'M 
 
 ■ — 1 
 
 
 • >-t 
 
 
 
 
 CO 
 
 ^> c 
 
 
 
 iJ 
 
 u 
 
 .id 
 
 (O 
 
 4_) 
 
 r-l 
 
 c^ 
 
 •r4 
 
 c5^ 
 
 "O 
 
 . t 
 
 
 > 
 
 J 
 
 , 1 
 
 
 
 
 V 
 
 CO 
 
 -0 
 
 'J 
 
 ■—< 
 
 U-1 
 
 
 
 <0 <0 
 
 
 
 D 
 
 nj 
 
 
 
 XI 
 
 
 
 •r-l 
 
 u 
 
 4-1 
 
 G 
 
 ,-^ 
 
 , J 
 
 • —1 
 
 
 •r-( 
 
 1-1 
 
 cu 
 
 U 
 
 >" 
 
 cl 
 
 Z 
 
 4-J 
 
 
 
 T3 
 
 ■rH 
 
 
 
 < 
 
 M 
 
 c 
 
 l-l 
 
 .1-1 
 
 4-t 
 
 •r-l 
 
 CO 
 
 
 
 
 OJ 
 
 CO 
 
 4-1 
 
 XI 
 
 4J 
 
 cx 
 
 > 
 
 
 
 
 Q 
 
 a 
 
 •.-< 
 
 
 n 
 
 C -H 
 
 
 
 1 
 
 C 
 
 rs; 
 
 ex 
 
 0. 
 
 CO 
 
 )-l 
 
 l-l 
 
 c 
 
 (X, 
 
 3 
 
 E 
 
 CO 
 
 
 CO 
 
 
 '^ 
 
 n 
 
 11 
 
 cu 
 
 >: 
 
 J-< 
 
 u 
 
 3 
 
 Vh >•< 
 
 
 
 ■ -J 
 
 XI 
 
 
 
 
 
 
 
 CO 
 
 CO 
 
 
 TJ 
 
 'T-* 
 
 k-i 
 
 CO 
 
 
 x; 
 
 4-1 
 
 ■rH 
 
 
 
 J-l 
 
 u 
 
 -0 
 
 :- 
 
 
 
 G 
 
 ^ 
 
 • 
 
 l-l 
 
 c 
 
 , 
 
 E 
 
 C 
 
 y 
 
 ^ 
 
 'M 
 
 c. 
 
 CO 
 
 
 , 
 
 
 
 CO 
 
 
 U 
 
 • 
 
 • 
 
 D 
 
 11 
 
 a 
 
 •rH -eS 
 
 
 
 Xt 
 
 *— * 
 
 
 
 SJ 
 
 40 
 
 OS 
 
 < 
 
 c 
 
 JD 
 
 CJ 
 
 C/1 
 
 < 
 
 c. 
 
 , 
 
 a: 
 
 cu 
 
 -—I 
 
 c 
 
 
 
 
 
 < 
 
 X 
 
 U-l 
 
 •" 
 
 
 
 ISi 
 
 
 
 , 
 
 Q. 
 
 Cfl 
 
 •r4 
 
 
 
 
 
 
 
 
 
 
 E 
 
 
 
 > 
 
 
 
 
 CO 
 
 
 
 CO 
 
 , 
 
 , 
 
 . 
 
 U-4 
 
 , 
 
 f4 
 
 (0 
 
 
 
 < 
 
 c_> 
 
 U) 
 
 Q. 
 
 
 CQ 
 
 •-) 
 
 CO 
 
 3' 
 
 >' 
 
 1-) 
 
 u 
 
 CO 
 
 cr: 
 
 lyi 
 
 - 
 
 XI 
 
 
 
 -^ 
 
 a 
 
 '^ 
 
 ■ 
 
 ao 
 
 <J -5 
 
 
 CD 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 — 1 
 
 CO 
 
 
 
 en 
 
 
 
 
 
 
 CO 
 
 
 
 —4 
 •J- 
 
 03 
 
 
 
 
 
 CM 
 
 
 
 
 CO 
 
 
 
 
 
 
 CQ 
 
 
 
 
 , ,. , 
 
 
 
 
 
 
 
 ^^_^ 
 
 IM^ 
 
 
 __ 
 
 
 »^M« 
 
 
 
 
 
 
 
 
 
 
 
 II. 2-7 
 

 w 
 
 
 ■J) 
 
 
 HH 
 
 >- 
 
 H 
 
 w^ 
 
 M 
 
 Ci_ 
 
 05 
 
 •CC 
 
 UJ 
 
 rr: 
 
 H 
 
 CTJ 
 
 < 
 
 CD 
 
 :< 
 
 — 1 
 
 <. 
 
 m 
 
 -:^ 
 
 
 t ) 
 
 cu 
 
 
 
 C-J 
 
 CO 
 
 - 1 
 
 Q_ 
 
 H) 
 
 u_ 
 
 n 
 
 
 o 
 
 
 «-. 
 
 
 a, 
 
 
 
 
 
 ■"■"""" 
 
 
 
 
 
 
 ■*~*~^ 
 
 — 
 
 «_■' 
 
 
 
 
 
 
 
 
 
 
 
 
 
 -i 
 
 
 
 
 
 
 
 
 
 
 
 
 
 cc 
 
 
 
 . 
 
 
 
 
 
 
 
 X 
 
 
 
 oo 
 
 
 
 X 
 
 
 X 
 
 
 X 
 
 
 
 
 
 
 ca 
 
 
 
 
 
 
 
 
 
 
 
 
 
 «c 
 
 
 
 
 
 
 
 
 
 
 
 
 
 i^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CD 
 
 
 
 X 
 
 
 X 
 
 
 X 
 
 X 
 
 
 
 
 
 C_) 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ,^ . 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 ^ 
 
 
 
 ^ 
 
 
 
 -a: 
 
 
 ^ 
 
 
 
 2= 
 CD 
 
 a. 
 
 a, 
 
 
 
 u 
 a, 
 
 
 
 u 
 to 
 
 
 CO 
 
 cu 
 
 
 
 t— 
 
 a 
 
 <u 
 
 (U 
 
 
 OJ 
 
 
 
 o 
 
 
 u 
 
 
 
 -I 
 
 to 
 
 00 
 
 f-i 
 
 
 c:0 
 
 
 60 
 
 
 00 
 
 
 
 CD 
 
 o 
 
 0) 
 
 4J 
 
 
 0) 
 
 
 u 
 
 <u 
 
 
 CJ 
 
 
 
 C^ 
 
 .— ( 
 
 r~~i 
 
 U 
 
 
 >— I 
 
 
 4J 
 
 r-t 
 
 
 f— 1 
 
 
 
 —J 
 
 t—i 
 
 r-t 
 
 nj 
 
 
 r-l 
 
 
 C3 
 
 r-< 
 
 
 t-H 
 
 
 
 
 o 
 
 O 
 
 <u 
 
 
 o 
 
 
 o 
 
 
 o 
 
 
 
 
 o 
 
 U 
 
 M 
 
 
 CJ 
 
 
 C/2 
 
 CJ 
 
 
 o 
 
 
 
 
 
 
 
 
 
 
 J3 
 
 
 
 
 
 
 
 
 
 
 
 
 
 XJ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 •r4 
 
 
 
 
 
 
 
 
 
 
 
 
 
 3 
 
 0) 
 
 
 
 
 
 
 
 
 (0 
 
 
 
 
 
 X 
 
 
 V4 
 
 
 
 
 
 
 a> 
 
 
 <u 
 
 
 X) 
 
 4J 
 
 
 cu 
 
 
 
 
 
 
 •r^ 
 
 
 > 
 
 
 (U 
 
 
 
 •o 
 
 
 
 rtt 
 
 se 
 
 
 
 3 
 
 
 4J 
 
 
 0) 
 
 s 
 
 
 c 
 
 3 
 
 
 
 CD 
 
 tl. 
 
 ^-N 
 
 
 
 
 CO 
 .—1 
 
 
 f-4 
 
 a 
 
 x: 
 
 
 X) 
 
 <u 
 u 
 
 o 
 
 u 
 
 CNj 
 
 
 
 0) 
 
 
 
 -d 
 
 
 
 
 1 
 
 o 
 
 CJ 
 
 
 
 CD 
 
 ;_j 
 
 s 
 
 
 CJ 
 
 
 
 
 CJ 
 
 o 
 
 
 1 
 
 
 — 
 
 
 v.^ 
 
 
 •U 
 
 ^ 
 
 en 
 
 
 
 ^1 
 
 ^"•^ 
 
 o 
 
 CNI 
 
 
 _i 
 
 
 
 CO 
 
 H 
 
 D 
 
 
 CO 
 
 D. 
 
 H 
 
 <~n 
 
 
 CO 
 
 so 
 
 c 
 
 
 cq 
 
 
 o 
 
 •r-l 
 
 
 
 D. 
 < 
 
 
 UJ 
 
 
 O- 
 
 j3 
 
 CO 
 
 P3 
 3 
 
 
 
 
 U 
 
 CO 
 
 
 0) 
 
 4-1 
 
 3 
 
 •^-^ 
 
 CO 
 
 
 
 
 u 
 
 CJ 
 
 OJ 
 u 
 
 s 
 
 C/l CO 
 
 > 
 
 
 "o 
 
 7^ 
 
 V 
 ■Ji 
 
 )-i 
 
 u 
 
 c 
 
 o 
 
 C-1 
 
 
 
 
 
 4_) 
 
 u 
 
 11 X) 
 
 CO 
 
 ^^\ 
 
 
 < 
 
 •r-l 
 
 V 
 
 ^ 
 
 
 
 ^ 
 
 ^ 
 
 i—i 
 
 fl 
 M 
 
 
 en C 
 
 <u 
 
 fa 
 
 CO 
 
 
 CO CM 
 0-, lTi 
 
 CJ 
 
 CO 
 
 
 
 
 
 J-J 
 
 o 
 
 o c ^ 
 
 4-) 
 
 i 
 
 •r-l 
 
 • • 
 
 o 
 
 
 
 ? o 
 
 , 
 
 c 
 
 CJ 
 
 o c ^ 
 
 CO 
 
 Ix, 
 
 (1) 
 
 CNI 
 
 o 
 
 O 
 
 
 
 (r 
 
 Q 
 
 ni 
 
 c 
 
 O C3 1 
 
 u 
 
 >^-' 
 
 
 i_} 
 
 . 1 
 
 
 
 o 
 
 
 (t: 
 
 CJ 
 
 C 
 
 D 
 
 ' 
 
 CO 
 
 CO 
 
 
 U-l 
 
 n 
 
 G^ 
 
 O 
 
 1) 
 
 c 
 
 ' 
 
 c 
 
 c 
 
 
 o 
 
 CO 
 
 
 ■H 
 
 H 
 
 CO 
 
 c 
 
 )-l 
 
 o 
 ex 
 
 a) 
 
 •K 
 
 Di 
 
 O 
 
 o 
 
 e 
 
 N 
 
 c 
 u 
 
 o 
 
 a 
 
 0) 
 
 o 
 
 o 
 
 CO 
 
 E 
 
 l-l CJ 
 
 x: 
 • a 
 
 Q 
 
 X) 
 
 ^ o 
 c o 
 
 CLI 
 
 x; ^•a 
 
 c 
 o 
 o 
 c 
 o 
 
 CJ 
 
 c 
 
 ■H 
 CJ 
 4J 
 
 o 
 
 3 
 
 o 
 
 S-i 
 
 D 
 Q 
 
 O 
 -. r^ 
 
 CO I 
 
 to •^ 
 
 c o 
 
 A 
 o 
 o 
 )-l 
 
 CO 
 CO 
 
 o 
 
 c 
 o 
 o 
 
 c 
 o 
 
 CO 
 
 U CJ 
 CL H 
 
 o -o 
 
 o 
 
 4J 
 
 x; 
 a 
 
 o 
 
 ca 
 
 o 
 
 ca 
 
 : u 
 
 N) ^ CJ < 
 
 c 
 
 ex . 
 < o 
 
 o 
 
 CQ 
 
 . 
 
 ^ 
 
 CQ 
 
 CJ 
 
 
 o 
 
 *. 
 
 c 
 
 x: 
 
 V 
 
 .^ 
 
 ■^ 
 
 o 
 
 CJ 
 
 tji 
 
 C/1 
 
 CO 
 
 
 Cr; 
 
 -a 
 
 
 o 
 
 , 
 
 o 
 
 O 
 
 '■^ 
 
 —1 
 
 -)' 
 
 C30 
 
 o 
 
 CQ 
 
 < 
 
 c 
 
 c 
 
 •r-l 
 
 < 
 
 o 
 
 ^ -C 
 
 o ex 
 
 CO 
 
 >. J-" 
 
 iJ o 
 
 •r-l E 
 
 ,-( CO 
 
 •r-l X 
 
 X) fiJ 
 
 CO 
 
 -^ E 
 
 •r-l 3 
 
 a: -r^ 
 
 > X) 
 
 < c 
 
 o 
 
 ca 
 
 
 
 Cli CJ 
 
 (-1 x: 
 
 CJ Cl. 
 
 CJ 
 
 c •• 
 
 •r-l vD 
 
 ex n 
 
 l-l 
 
 4) 
 
 
 o 
 
 n 
 
 c 
 
 >, 
 
 il 
 
 o 
 
 ■rA 
 
 n 
 
 o 
 
 .— 1 
 
 CO 
 
 '^) 
 
 
 CL, 
 
 T3 
 
 
 O 
 
 . 
 
 O 
 
 < 
 
 u- 
 
 u 
 
 •-) 
 
 CO --I 
 c/5 O 
 
 o -o 
 
 o 
 
 CJ 
 
 x: 
 
 H 
 
 o 
 
 O 
 
 CO 
 
 c 
 
 CO 
 
 E 
 ■o 
 
 CJ 
 
 o 
 
 c 
 
 CO 
 
 to :: 
 
 u c/ 
 
 c 
 
 1 c 
 O ■- 
 
 to 
 
 q 
 
 -a 
 c 
 o 
 o 
 
 O ■'^ CI 
 
 r-l E E 
 
 .-I "O c 
 
 O --1 c 
 
 Vj o — 
 
 1-1 CJ 
 
 a 
 o 
 
 u c 
 
 ll.2-i 
 

 CO 
 
 
 ■ ) 
 
 >- 
 
 1— 1 
 H 
 
 zxz 
 
 V) 
 
 cu 
 
 M 
 
 •cC 
 
 Oi 
 
 CO 
 CD 
 
 UJ 
 
 H 
 
 1 
 
 c^ 
 
 to 
 
 <, 
 
 CO 
 
 o 
 
 C3 
 
 
 Li- 
 
 • ; 
 
 u. 
 
 C) 
 
 
 TJ 
 
 
 Q 
 
 
 O 
 
 g 
 
 
 ^ 
 
 JS 
 
 
 w 
 
 u 
 
 
 to 
 
 CL. 
 
 CO 
 
 (X. 
 
 » — 
 
 <u 
 
 4) 
 
 «« 
 
 CiO 
 
 00 
 
 <_> 
 
 o 
 
 V 
 
 CD 
 
 r-l 
 
 .-< 
 
 — ^ 
 
 1— 1 
 
 .-< 
 
 
 o 
 
 o 
 
 
 o 
 
 o 
 
 o 
 o 
 
 bO 
 
 o 
 o 
 
 u 
 
 CO 
 
 a> 
 a; 
 
 00 
 
 o 
 
 u 
 
 C8 
 
 v 
 
 00 
 
 o 
 o 
 
 u 
 
 CO 
 
 a. 
 
 5J 
 
 iO 
 o 
 
 o 
 o 
 
 ^ 
 
 CL 
 •r-l 
 
 o o 
 
 
 OJ 
 
 c 
 
 CO 
 
 
 CO 
 
 ^-J 
 
 TD 
 
 -1- 
 
 u 
 
 G 
 
 
 > 
 
 n) 
 
 -a 
 
 OJ 
 
 
 o 
 
 J 
 
 X 
 
 x: 
 
 
 o 
 
 « 
 
 >~ 
 
 lu, 
 
 ■r-l 
 
 u 
 
 
 r-A 
 
 
 
 . 
 
 .a 
 
 o 
 
 UJ 
 
 3 
 
 01 
 
 o 
 
 c 
 
 0) 
 M 
 
 <u 
 
 a 
 o 
 o 
 
 CO 
 
 o 
 
 ■r-l 
 00 
 
 o 
 
 .—I 
 o 
 
 -G 
 JG 
 CJ 
 
 <U 
 H 
 
 0) 
 
 •r-l 
 V-l 
 
 ^"■^ o 
 
 CO j::: 
 C--1 u) 
 
 ?: -H 
 
 u 
 
 CJ < 
 
 a 
 u 
 C 
 4) 
 u) 
 (U 
 )-l 
 
 a 
 >^ 
 
 D. 
 
 <0 
 
 o 
 
 03 
 
 O 
 
 <U 
 U 
 
 o 
 
 
 0) 
 
 M 
 C 
 
 r-l 
 ft, 
 
 >. 
 
 J3 
 CO 
 
 CO 
 > 
 
 3 : 
 
 V4 
 
 x; 
 
 4-1 
 
 u 
 
 3 
 
 o 
 
 C- 
 
 M 
 
 a; 
 
 o 
 o 
 
 >^ u~l 
 
 r-l ro 
 
 r-l I 
 
 CO ^o 
 
 U) CNJ 
 
 
 r 
 
 to 
 
 C 
 
 CD 
 
 to 
 
 O 
 
 j-i 
 
 c 
 
 •r^ 
 
 <0 
 
 •>-l 
 
 J-I 
 
 u 
 
 r-l 
 
 •.-1 
 
 *-> 
 
 r-l 
 
 )4 
 
 c 
 
 •r4 
 
 J-) 
 
 o 
 
 JJ 
 
 D 
 
 c; 
 
 C/1 
 
 X. 
 
 c 
 
 
 
 o 
 
 • 
 
 • 
 
 u 
 
 03 
 
 •-) 
 
 m 
 O 
 
 oa 
 
 c 
 c 
 
 o <o 
 
 ■rJ U 
 
 . o 
 
 W CJ 
 
 CO 
 
 o 
 > 
 
 •r^ 
 
 4-1 
 
 D 
 Z 
 
 o 
 
 ex 
 e 
 o 
 u 
 
 o 
 o 
 
 OJ 
 
 d 
 -a 
 
 o u 
 
 u-l 
 
 UJ U-i 
 
 : O 
 
 =0 n 
 c: cN 
 •H <r 
 
 . 
 
 J 
 
 a: 
 
 "J 
 
 
 U 
 
 
 c 
 
 03 
 
 ■J 
 
 
 ■rl 
 
 r. 
 
 C) 
 
 .c 
 
 CO 
 
 j; 
 
 
 o 
 
 -o 
 
 U) 
 
 o 
 
 CO 
 
 o 
 
 fti 
 
 u. 
 
 •-5 
 
 •-i 
 
 o 
 
 c 
 c 
 
 o 
 
 o 
 
 c 
 o 
 o 
 
 y 
 
 u-l CO 
 O j=3 
 
 'J. 
 
 o 
 
 V) 
 
 o 
 
 O 
 
 ct 
 
 i-) 
 
 
 ,0 
 
 t—t 
 
 ■r-l 
 
 c 
 
 VJ 
 
 
 o 
 
 I/) 
 
 J-) 
 
 ij 
 
 CJ 
 
 y 
 
 r3 
 
 1^ 
 
 >-■ 
 
 U-i 
 
 CO 
 
 >*.( 
 
 sz 
 
 UJ 
 
 o 
 
 o 
 
 \r\ 
 O 
 
 CO 
 
 II. 2-9 
 
S 
 ' I 
 
 CQ 
 
 H 
 
 H 
 < 
 
 
 
 o 
 
 Q 
 o 
 
 0_ 
 
 ►— 
 
 
 
 
 
 
 
 
 
 
 t_> 
 
 
 
 
 
 
 
 
 
 
 
 "t 
 
 
 
 
 
 
 
 
 
 
 
 cl; 
 
 
 
 
 
 
 
 
 
 
 
 00 
 
 X 
 
 XJ 
 
 « 
 
 
 
 X 
 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 
 2: 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 " 
 
 
 
 
 
 
 —1 
 
 
 
 
 
 
 
 
 
 
 
 CD 
 
 
 
 
 X 
 
 
 
 X 
 
 
 
 
 cr: 
 
 
 
 
 
 
 
 
 
 
 
 Ci 
 
 
 
 
 
 
 
 
 
 
 
 " 
 
 
 
 
 
 
 
 
 
 
 
 *"~ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 ^ 
 
 ^ 
 
 ^ 
 
 ^ 
 
 ji: 
 
 
 ji 
 
 
 
 ^ 
 
 >4 
 
 u 
 
 )-i 
 
 v< 
 
 l-l 
 
 u 
 
 
 1^ 
 
 
 CD 
 
 
 
 0- 
 
 2 
 
 CO 
 CVi 
 
 <13 
 
 a, 
 
 
 03 
 
 a. 
 
 
 £ 
 
 
 ►-^ 
 
 
 0) 
 
 v 
 
 (U 
 
 (U 
 
 01 
 
 
 
 
 :> 
 
 
 •«s 
 
 m 
 
 bO 
 
 00 
 
 M 
 
 00 
 
 yj 
 
 so 
 
 
 =0 
 
 
 C_3 
 
 oO 
 
 Q) 
 
 <u 
 
 W 
 
 <u 
 
 01 
 
 
 
 
 ^ 
 
 
 CD 
 
 
 
 r-( 
 
 rH 
 
 1— ( 
 
 r-4 
 
 —( 
 
 
 
 f-H 
 
 
 ■ 
 
 r-l 
 
 r-l 
 
 !-• 
 
 .—( 
 
 i-l 
 
 1—1 
 
 _t 
 
 
 ^H 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 u 
 
 u 
 
 CJ 
 
 
 u 
 
 
 
 
 
 
 
 
 
 
 
 «4-l 
 
 
 
 ,— J 
 
 
 
 
 
 
 
 
 
 
 
 
 nl 
 
 
 
 to 
 
 
 
 
 
 
 
 
 c 
 
 
 
 l-l 
 
 
 
 
 
 10 
 
 
 
 
 
 
 
 M 
 
 
 
 
 
 c 
 
 
 
 •H 
 
 
 
 
 
 
 
 
 
 
 
 
 4J 
 
 
 
 ■0 
 
 
 
 
 
 •tA 
 
 
 
 
 
 
 
 C 
 
 
 
 
 
 4J 
 
 
 
 c 
 
 
 
 <« 
 
 
 .^^ 
 
 
 
 •rJ 
 
 
 zn 
 
 D 
 
 
 
 
 
 CO 
 
 
 
 -a 
 
 
 CD 
 
 tLl 
 
 
 
 Vj 
 
 
 
 
 
 c 
 
 
 
 
 
 
 
 
 
 V4 
 
 
 r ^"-^ 
 
 
 
 
 
 
 
 CD 
 
 XJ 
 
 
 
 
 
 
 c 
 
 
 
 « 
 
 
 t 
 
 
 
 
 •r^ 
 
 
 ■H (0 
 
 
 
 c 
 
 
 CO 
 
 C 
 
 
 
 x: 
 
 
 OJ 30 
 
 
 
 •H 
 
 
 
 
 
 
 
 w 
 
 
 i-l C 
 
 
 
 >■ 
 
 
 D- 
 
 1-1 
 
 
 
 0) 
 U 
 
 
 CO 
 3 
 
 
 -H 
 
 •r4 
 
 
 
 > 
 
 -J) 
 
 
 N 
 
 i-> 
 
 
 >4-l 
 
 
 x: JJ 
 
 
 
 Sj 
 
 ■u 
 
 
 'M 
 
 c 
 
 
 
 
 
 to CO 
 
 
 CJ 
 
 
 
 ■J2 
 
 
 ,_( 
 
 3J 
 
 ^-N 
 
 
 
 ■r* 
 
 
 a. 
 
 -o 
 
 • -J 
 
 
 .^ 
 
 
 
 H 
 
 U 
 
 
 tL, . 
 
 
 tu 
 
 c 
 
 ti 
 
 
 C~ -Q 
 
 c 
 
 h- 1 
 
 c 
 
 
 05 
 
 
 
 D 
 
 
 
 
 T3 n3 
 
 
 
 S 
 
 0) 
 
 
 c 
 
 
 c 
 
 
 1 — ' 
 
 
 
 
 
 Nw^ 
 
 
 
 CQ 
 
 ^1 
 
 
 
 CD 
 J-J 
 
 
 
 c 
 
 o 
 u 
 
 'I ._, 
 
 OJ >^ 
 
 to 
 
 2 
 
 CO 
 
 c 
 
 Q 
 
 C? 
 
 
 G- 
 
 >J 
 
 Co., 
 
 CO 
 
 
 u 
 
 U~l 
 
 0. 
 
 
 
 in 
 
 
 
 tn 
 
 U_( 
 
 
 
 
 
 •r-* 
 
 
 i-i 
 
 LJ 
 
 M 
 
 U 
 
 
 
 IJ 
 
 C 
 
 U-l 
 
 
 
 UJ 
 
 £ 
 
 
 X) 
 
 a 
 
 D. 
 
 c 
 
 r3 
 
 <X3 
 
 O 
 00 
 
 O 
 
 CO 
 
 c 
 o 
 00 
 
 0) 
 
 (-( 
 o 
 
 ^c 
 
 GJ 
 
 •U 
 
 x: 
 
 C 
 
 
 
 OJ 
 
 
 
 
 • 
 
 t- 
 
 CJ 
 
 
 
 
 C) 
 
 A 
 
 
 00 
 
 c 
 
 c 
 
 •r^ 
 
 a 
 
 <D 
 
 :2 
 
 u 
 
 o 
 a. 
 
 c 
 o 
 
 o 
 o 
 
 c 
 
 u 
 
 
 H 
 
 c 
 
 
 0) 
 
 CJ 
 
 u 
 
 ^ 
 
 -a 
 
 u 
 
 r-* 
 
 
 •^ 
 
 C 
 
 x: 
 
 •rJ 
 
 u 
 
 01 
 
 CU 
 
 u 
 
 « 
 
 to 
 
 < 
 
 i-l 
 
 
 4-1 
 
 f— t 
 
 C 
 
 
 
 QJ 
 
 
 
 
 
 -C 
 
 C 
 
 CJ 
 
 
 
 C/2 
 
 u 
 
 1 
 
 
 01 
 
 n 
 
 u 
 
 •H 
 
 IX 
 
 OJ 
 
 
 l_J 
 
 c 
 
 
 
 
 
 >-i 
 
 
 CU 
 
 
 
 
 D 
 
 j: 
 
 r-l 
 
 ■J) 
 
 CO 
 
 ■r-l 
 
 ^ 
 
 [J-, 
 
 
 
 c 
 
 U-i 
 
 • w 
 
 
 
 "J 
 
 
 I-t 
 
 V 
 
 
 
 (/■i 
 
 u 
 
 r> 
 
 a. 
 
 CO 1-1 
 
 x: o 
 
 7) D. 
 
 o ^ 
 
 CO (J 
 
 c 
 • o 
 
 CO 
 
 o 
 
 >, 
 
 S-l 
 
 H 
 !-i M 
 
 0) S 
 
 S ^ 
 
 '-I c 
 
 >> "3 
 
 ^ E 
 
 01 OJ 
 
 S ■^^ 
 »-i 
 
 O Uh 
 
 CO 
 
 a: 
 
 OJ <f 
 
 u 
 
 o 
 
 1) 
 a; 
 
 
 V) O cy-i 
 
 O 
 
 a. 
 
 X 
 
 c o 
 
 U CO 
 '- u 
 
 C i-l 
 c 
 o 
 
 J3 
 
 il 
 
 o 
 
 CQ 
 
 CM 
 O 
 
 o 
 
 OQ 
 
 LJ 
 
 O 
 
 c 
 o 
 u 
 
 vD 
 O 
 CQ 
 
 M 
 
 
 J-l 
 
 
 c: 
 
 
 
 
 li 
 
 u 
 
 r: 
 
 c: 
 
 ■ —1 
 
 
 
 r— ' 
 
 c; 
 
 
 
 -0 
 
 c 
 
 
 •r-( 
 
 CO 
 
 iJ 
 
 
 i-l 
 
 > 
 
 
 
 4-1 
 
 i 
 
 -a 
 
 
 •^ 
 
 r; 
 
 p 
 
 :/! 
 
 3 
 
 .r-< 
 
 ;x; 
 
 Ci. 
 
 
 
 "J 
 
 U_l 
 
 > 
 
 
 
 • —1 
 
 
 j-i 
 
 >, 
 
 c^ 
 
 1-J 
 
 ■ — ■ 
 
 •r-t 
 
 -'' 
 
 .— ' 
 
 ^1 
 
 •r^ 
 
 
 XI 
 
 " 
 
 to 
 
 \j 
 
 iJ 
 
 u 
 
 •-0 
 
 u 
 
 
 —1 
 
 ^ 
 
 ~ 
 
 :j 
 
 l-l 
 
 to 
 
 '_, 
 
 l-l 
 
 a 
 
 c 
 
 ti: 
 
 u 
 
 ■J 
 
 !J1 
 
 H 
 
 
 vO 
 
 o 
 en 
 
 II. 2-10 
 
cx: 
 en 
 
 O 
 
 M 
 
 M 
 UJ 
 
 H 
 O 
 < 
 
 f-3 
 
 o 
 
 Q 
 O 
 
 ► - 
 
 c .- 
 
 
 
 
 
 
 
 
 
 
 
 CjI 
 
 
 
 
 
 
 
 
 
 
 
 ►-— 
 
 
 
 
 
 
 
 
 
 
 
 cn 
 
 
 
 
 
 X 
 
 
 
 X 
 
 X 
 
 
 c.-i 
 
 
 
 
 
 
 
 
 
 
 
 ■^ 
 
 
 
 
 
 
 
 
 
 
 
 T- 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 L*. 
 
 
 
 
 
 
 
 
 
 
 
 CD 
 
 
 
 
 
 
 X 
 
 
 
 
 
 CC 
 
 
 
 
 
 
 
 
 
 X 
 
 
 C_> 
 
 
 
 
 
 
 
 
 
 
 
 *'J 
 
 
 
 
 
 
 
 
 
 
 ' 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 V4 
 
 
 ^ 
 
 
 .M 
 
 ^ 
 
 
 .^<: 
 
 u 
 
 to 
 a, 
 
 1 
 
 
 n) 
 
 
 Wi 
 
 
 M 
 
 u 
 
 
 M 
 
 
 ^^ 
 
 a, 
 
 
 to 
 
 
 to 
 
 tu 
 
 
 to 
 
 » 
 
 CD 
 
 
 
 Oi 
 
 
 pu, 
 
 a.- 
 
 
 a. 
 
 • 
 
 
 o 
 to 
 
 
 <1) 
 
 
 OJ 
 
 tu 
 
 
 <a 
 
 I-I 00 
 
 U r-4 
 
 
 C_5 
 
 01 
 
 
 OO 
 
 
 OO 
 
 00 
 
 
 60 
 
 
 CD 
 
 »— 4 
 
 
 <u 
 
 
 4J 
 
 OJ 
 
 
 0) 
 
 • 
 
 
 ^H 
 
 
 —( 
 
 
 rH 
 
 r-l 
 
 
 r-l 
 
 to .-< 
 
 <U O 
 
 CO O 
 
 
 
 O 
 
 u 
 
 
 i-i 
 o 
 o 
 
 
 1—1 
 
 O 
 
 u 
 
 ■-I 
 
 o 
 
 
 «— 1 
 
 o 
 
 CJ 
 
 
 
 
 
 "(U 
 
 
 w 
 
 
 
 
 
 i 
 
 
 
 
 y 
 
 
 4-1 
 
 
 
 
 
 t 
 
 
 
 
 •r< 
 
 
 to 
 
 
 
 
 
 ^ \ 
 
 
 to 
 
 
 a 
 
 
 ci 
 
 
 
 x: 
 
 
 
 I 
 
 
 "O 
 
 
 
 <M 
 
 
 U-l 
 
 
 
 CO 
 
 
 J3 
 
 
 
 •r^ 
 
 
 
 o 
 
 
 o 
 
 
 
 •1-1 
 
 
 to 
 
 
 
 C 
 
 
 
 tn ^N 
 
 
 tn 
 
 
 
 b 
 
 
 •r< 
 fa 
 
 
 
 nj 
 
 
 
 C CO 
 
 
 C 
 
 
 
 M-l 
 
 
 
 
 3: 
 
 O 
 
 
 
 O t4 
 
 •r-l v; 
 
 
 o 
 
 •H ,-x 
 
 
 
 O 
 
 
 0) 
 
 \ 
 
 CD 
 
 -o 
 
 
 
 4J 2 
 
 
 iJ CO 
 
 
 
 to 
 
 
 
 c 
 
 
 
 to ^^ 
 
 
 2 ^ 
 
 
 
 <U 
 
 
 1 
 
 »— 
 
 ra 
 
 
 
 )-i 
 
 
 
 
 •f-l 
 
 
 
 
 tX. 
 
 
 
 
 01 to 
 
 
 0) -z, 
 
 
 
 J-l 
 
 
 ■o 
 
 CO i 
 
 CD 
 
 e 
 
 
 
 C OO 
 
 
 c ^-^ 
 
 
 
 u 
 
 
 c 
 
 ^ 1 
 
 CD 
 ZD 
 
 -o 
 
 
 
 O •r< 
 
 r-l 
 
 
 0) 
 
 U tn 
 
 00 
 
 
 
 a 
 o 
 
 
 to 
 
 D 
 X 
 
 
 o_ 
 
 c 
 
 
 
 > -r-l 
 
 
 
 
 
 i: 
 
 
 C i 
 
 to 
 
 
 P3 
 
 
 
 ■H U 
 
 
 •r^ ,-1 
 
 
 
 
 
 •r-l 
 
 
 o 
 
 
 
 CI^ to 
 
 
 ft. .-1 
 
 • r-( 
 
 
 
 r-l 
 OH 
 
 
 .—1 
 
 to 
 
 g< 
 
 
 
 
 
 V-i 
 
 
 V4 U . 
 
 OJ CO 
 
 
 
 C 
 
 o 
 
 
 ■X. 
 
 00 ( 
 
 
 j:: 
 
 
 
 > 
 
 
 > 
 
 
 
 •I-I 
 
 
 ■a 
 
 • ^ . 
 
 
 o 
 
 
 
 o 
 
 
 O . r-l 
 
 
 
 iJ 
 
 
 01 
 
 rJ fS 
 
 1 
 
 
 
 ^^ -rH 
 
 •-N 
 
 
 
 
 C3 <t 
 
 
 
 y 
 
 
 4-1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ■" iJ ttj 
 
 
 
 
 O ■ — ' 
 U > 1 
 
 o 
 
 
 
 '5 
 
 
 c 
 
 r-l VD 
 
 
 
 
 1 
 
 
 tT3 ^ 
 
 
 
 —1 
 
 /'J 
 
 
 
 r.u 
 
 
 e ■ 
 
 —1 t~-J 
 
 to 
 
 
 
 o 
 
 
 
 4-J 
 
 
 4-1 M ^^ 
 
 o 
 
 
 
 <U 
 
 
 c 
 
 3C •• 
 
 
 
 a ^ 
 
 
 
 c c 
 
 CNI 
 
 C S 
 
 to 
 
 
 
 j: 
 
 
 o 
 
 ^•^1 
 
 
 
 XI 
 
 N 
 
 
 0) to 
 
 f— 1 
 
 0) v^ .. 
 
 4J 
 
 
 
 u 
 
 
 
 • CNI 
 
 ■ 
 
 CM 
 
 
 G 
 
 
 C "O 
 
 1 
 
 C C tN 
 
 X 
 
 
 
 c 
 
 
 r-- 5 
 
 f-H i 
 
 t/O 
 
 r~~ 
 
 >-i 
 
 •r-l 
 
 
 o 0) 
 
 rH 
 
 O to 
 
 w 
 
 
 
 
 
 
 o^ -^ 
 
 - CN i 
 
 CXI 
 
 ON 
 
 01 
 
 e 
 
 
 O •'-' 
 
 00 
 
 u e -» 
 
 
 
 
 
 
 I-I 
 
 U ! 
 
 CD 
 
 "I 
 
 '^ 
 
 to 
 
 .—1 
 
 
 
 ,—1 
 
 d 01 o£ 
 
 5I^ 
 
 _, 
 
 
 0( 
 
 c 
 
 
 Q 
 C 
 
 OJ 
 
 r-l > 
 
 ; 
 
 5 
 
 ^ 
 
 o 
 o 
 
 
 •r-l 
 0) 
 
 ^'X 
 
 •r^ •r^ O 
 
 OJ ^4 — 1 
 
 b4 
 
 
 
 •I-I 
 ^4 
 
 
 •r4 
 
 01 
 
 .—1 ^4 
 ■r4 4-1 
 
 1 
 
 
 
 
 4-> J 
 
 o 
 
 
 U tu O 
 
 o o 
 
 c 
 
 
 
 •■-» 
 ■u 
 
 <t 
 
 4D 
 
 to 
 
 T. to 
 
 ■r-l 
 
 . e 
 
 oi o 
 
 \ 
 
 
 4-1 
 CO 
 
 d 
 
 
 (C )-i 
 
 r^ 
 
 i^ J §> 
 
 a; 
 
 
 
 C 
 
 o 
 I-I 
 
 >N 
 
 
 
 ^ 
 
 
 OJ 
 
 n-l 
 
 vj 
 
 •a 
 
 
 
 > 
 
 1 
 
 
 X 
 
 » 
 
 
 Uh 
 
 .— ( 
 
 
 x; tn 
 
 to 
 
 x; - to 
 
 •r-l 
 
 
 
 r-l 
 
 as 
 
 X 
 
 - CJ 
 
 ■ 
 
 
 o 
 
 .— ( 
 
 
 c/) to 
 
 c 
 
 tfl (-1 X 
 
 X 
 
 
 
 o 
 
 ^•s 
 
 (J\ 
 
 4-1 
 
 00 
 
 
 
 
 e) 
 
 
 •r-l r-l 
 
 o 
 
 ■r-l 01 CUi 
 
 o 
 
 
 
 to 
 
 ^ 
 
 
 ■r-l 
 
 u -o 
 
 . 
 
 
 U) 
 
 ^ 
 
 
 [L, O 
 
 •.-1 
 
 U-, to 
 
 1-4 
 
 
 
 0) 
 
 • • 
 
 3 
 
 oj o 
 
 
 
 c 
 
 C/l 
 
 
 
 iJ 
 
 to -o 
 
 11 
 
 
 
 Q 
 
 U- 
 
 /--N 
 
 
 X o 
 
 
 
 o 
 
 nj 
 
 
 14-1 
 
 to 
 
 U-l ^ OJ 
 
 Ou, 
 
 
 
 
 ^-' 
 
 r-l 
 
 >■ 
 
 C U-. 
 
 • 
 
 
 • r-l 
 
 Cu 
 
 
 o o 
 
 c 
 
 o o -r-l 
 
 
 
 
 X! 
 
 
 ^^-' 
 
 u 
 
 Z •r^ 
 
 
 
 ■ JJ 
 
 
 
 
 >-i 
 
 
 C 
 
 
 
 c 
 
 o 
 
 ^-s —1 
 
 3 
 
 0) 1) -o 
 
 { 
 
 
 w 
 
 , 
 
 
 >-. 
 
 OJ 
 
 c • o 
 
 CJ 
 
 
 
 to 
 
 4.J 
 
 t/^ r^ 
 
 O 
 
 4J 4-1 C 
 
 
 
 u 
 
 J 
 
 
 o cu 
 
 4-1 
 
 o c 
 
 50 
 
 
 
 
 to 
 
 U-, 
 
 >-i 
 
 CO C/1 <0 
 
 * 
 
 
 c 
 
 
 
 tu 
 
 u 
 
 •M -^ 
 
 O 
 
 
 
 Cm 
 
 M 
 
 A d 
 
 ■v 
 
 M 
 
 L 
 
 
 4) 
 
 , 
 
 -a 
 
 3 
 
 1— 1 
 
 4-1 
 
 V4 
 
 r— N 
 
 
 c 
 
 4-1 
 
 2; •H 
 
 T. 
 
 4J . ^-1 
 
 f 
 
 
 E 
 
 ty^ 
 
 o 
 
 a" 0) 
 
 
 to o -a 
 
 -o 
 
 CO 
 
 
 •rH 
 
 c 
 
 v^ U 
 
 
 c S <o 
 
 LxJ 
 
 u 
 
 
 XI 
 
 OJ c 
 
 tn 
 
 D c c 
 
 >n 
 
 a, 
 
 
 > 
 
 0) 
 
 U 
 
 y-i 
 
 01 In 
 
 
 1 
 
 <u 
 
 ^ 
 
 U/ 
 
 -o u 
 
 4J 
 
 --t l4 T 
 
 X 
 
 S 
 
 
 1-1 
 
 o 
 
 3 -* 
 
 ■0 
 
 O r D 
 
 " 
 
 b^ 
 
 c 
 
 •r-l 
 
 < 0) 
 
 >-. 
 
 to 01 
 
 
 4-/ 
 
 Q 
 
 c 
 
 O '-' 
 
 
 C -r^ 4-1 
 
 1 
 
 f— 
 
 
 CO 
 
 o 
 
 ^^ 
 
 X 
 
 O 
 
 > X >, 
 
 U-) 
 
 ^-^ 
 
 P 
 
 
 o 
 
 V4 D 
 
 C 
 
 O -1 r-- 
 
 
 
 U 
 
 u 
 
 J 
 
 XI 
 
 o S 
 
 c. 
 
 UJ 3 y: 
 
 
 
 
 
 
 Vi 
 
 u 
 
 X ca 
 
 
 
 U -' 3 
 
 : 
 
 
 
 •r^ 
 
 V-i 
 
 D 
 
 •r-l 1) 
 
 o 
 
 Oi o 
 
 
 4-1 
 
 Ql 
 
 O 
 
 
 3 
 
 '^ 
 
 o o 
 
 k 
 
 
 _.j 
 
 i_J 
 
 o 
 
 CL 
 
 iO >2 
 
 cc 
 
 >, i'- —< 
 
 OJ 
 
 ■y) 
 
 1> 
 
 
 c 
 
 Q > 
 
 u 
 
 c c -^ 
 
 1 
 
 
 > 
 
 >. 
 
 
 
 o 
 
 
 u O 
 
 7) 
 
 C 
 
 oc 
 
 4-1 
 
 •r^ 
 
 I-I 
 
 O 
 
 •r^ •r^ V-l 
 
 
 
 , 
 
 o 
 
 
 
 OJ • o 
 
 rj 
 
 )-i 
 
 
 o 
 
 0) 
 
 • 0) 
 
 3 
 
 (i> d. U 
 
 
 c 
 
 o 
 
 X 
 
 X 
 
 IC 
 
 u 
 
 t4-i jr — 4 
 
 
 UJ 
 
 CO 
 
 
 
 4-1 
 
 kJ X 
 
 T3 
 
 4.4 CO < 
 
 
 c 
 
 I a 
 
 
 
 ■r-l 
 
 4_) 
 
 to y. 
 
 'J 
 
 
 u. 
 
 u- 
 
 o 
 
 in 
 
 0> 
 
 O 
 
 t 
 
 
 
 ■r-< 
 
 
 o 
 
 CO 
 
 -J 
 
 in ■ o 
 
 X 
 
 . 
 
 p" 
 
 u~t 
 
 »-l 
 
 . •r-l 
 
 a: 
 
 /^ -: r-: 
 
 1 
 
 
 
 -1 
 
 -)' 
 
 H 
 
 r a. 
 
 X 
 
 : a- H 
 
 H 
 
 a: 
 
 ;^. 
 
 u. 
 
 D- 
 
 Q U. 
 
 ' 
 
 li- •-) "->! : 
 
 CD 
 
 
 
 
 
 
 
 
 
 
 
 -jr. 
 
 
 
 
 
 
 
 
 
 
 
 
 vO 
 
 
 r»» 
 
 
 CO 
 
 o\ 
 
 
 O 
 
 r-4 
 
 
 ■ J 
 
 vO 
 
 
 vO 
 
 
 vO 
 
 vO 
 
 
 r>« 
 
 r^ 
 
 
 <_3 
 
 o 
 
 
 O 
 
 
 o 
 
 o 
 
 CQ 
 
 
 o 
 
 ta 
 
 o 
 
 CO 
 
 
 II.2-11 
 
tx: 
 
 »— 
 
 en 
 
 X 
 
 C3 
 
 <U 
 00 
 
 60 
 4> 
 
 u 
 nt 
 
 m 
 00 
 
 (U 
 
 a: 
 
 0) 
 00 
 
 o 
 
 O 
 
 o 
 u 
 
 o 
 o 
 
 a. 
 
 cc: 
 
 — I 
 £9 
 
 CO 
 
 w 
 
 u 
 
 H 
 U 
 
 Q 
 O 
 
 0> 
 
 CO 
 
 ex: 
 
 
 (X 
 
 E 
 
 D 
 05 
 C 
 
 o 
 o 
 
 c 
 
 > 
 
 0) 
 
 CO >-i 
 fa ^ 
 5^ 
 
 c 
 
 0) 
 r-t i-l ---I 
 
 [JM • O 
 
 
 *J 
 
 
 
 
 
 
 c 
 
 
 
 
 
 0) 
 
 
 
 
 
 6 
 
 
 
 
 
 <u 
 
 
 
 
 
 r-( 
 
 
 
 •\ 
 
 
 CL 
 
 
 
 ^-s 
 
 
 o. 
 
 
 
 OvI 
 
 
 D 
 
 
 
 V--' 
 
 
 CO 
 
 
 
 *. 
 
 
 c 
 
 
 
 CSJ 
 
 
 •l-< 
 
 
 
 CM 
 
 
 <u 
 
 
 
 
 
 iJ 
 
 
 ^^* 
 
 ^ 
 
 
 o 
 
 
 CO 
 
 trS 
 
 
 
 (-1 
 
 
 u. 
 
 4-1 
 
 
 
 p-i 
 
 
 2 
 
 S 
 
 
 
 0) 
 
 
 N-^ 
 
 r-l 
 
 
 
 l-l 
 
 
 
 nJ 
 
 
 
 XI 
 
 
 (-1 
 
 3 
 
 
 
 n) 
 
 
 CJ 
 
 c 
 
 
 
 iJ 
 
 
 -o 
 
 ^ 
 
 
 
 <u 
 
 
 >^ 
 
 
 
 to 
 
 
 c 
 
 
 
 
 CJ 
 
 
 C/2 
 
 rC 
 
 
 
 > 
 
 
 , 
 
 4-J 
 
 
 
 w 
 
 
 o 
 
 o- 
 
 
 
 
 
 
 
 
 
 
 
 
 
 w 
 
 
 
 
 
 >^ 
 
 •r-l 
 
 
 
 f) 
 
 
 cr> 
 
 ^ 
 
 
 n) 
 
 •H 
 
 
 r-l 
 
 tl3 
 
 > 
 
 
 c 
 o 
 
 Q 
 
 a 
 
 
 
 o 
 
 
 1-1 
 
 X 
 
 
 
 >, 
 
 
 Xi 
 
 
 
 
 4J 
 
 
 o 
 
 !-4 
 
 
 
 •r^ 
 
 
 c 
 
 O 
 
 
 
 i—l 
 
 
 ^ 
 
 U-l 
 
 
 
 cfl 
 
 
 
 
 
 
 d 
 
 
 • 
 
 > 
 
 
 
 C 
 
 
 s 
 
 OJ 
 
 
 
 c 
 
 
 QJ 
 
 ■r-l 
 
 
 
 •H 
 
 
 cu 
 
 O 
 
 
 
 <U 
 
 
 Vj 
 
 O 
 
 
 
 iJ 
 
 
 o 
 
 CO 
 
 
 
 o 
 
 
 0) 
 
 
 
 
 >-. 
 
 - 
 
 CJ 
 
 c 
 
 
 
 a. 
 
 ^^^^ 
 
 
 tc 
 
 
 
 
 U 
 
 . .\ 
 
 o 
 
 
 
 CJ 
 
 3 
 
 1-1 
 
 •r-l 
 
 
 
 ,c; 
 
 o 
 
 r~< 
 
 Lj 
 
 
 
 J-) 
 
 1—1 
 
 0) 
 
 1) 
 
 
 
 y-i 
 
 til 
 
 3 
 
 '0 
 
 1 
 
 
 
 o 
 
 X. 
 
 •rH 
 
 
 
 
 
 </) 
 
 CO 
 
 UJ 
 
 
 
 u 
 
 •H 
 
 
 o 
 
 
 
 c 
 
 [1^ 
 
 • 
 
 
 ox 
 
 
 g 
 
 ' 
 
 Q 
 
 c 
 o 
 
 
 
 OJ 
 
 C3 
 
 n) 
 
 •r^ 
 
 
 
 > 
 
 iXi 
 
 •r-l 
 
 i-J 
 
 
 
 o 
 
 u. 
 
 c 
 
 « 
 
 ■ • 
 
 
 V4 
 
 
 •H 
 
 V-i 
 
 1-4 
 
 
 o. 
 
 42 
 
 CO 
 
 (U 
 
 
 
 E 
 
 iJ 
 
 Vi 
 
 X! 
 
 • 
 
 
 M 
 
 •^^ 
 
 ■r-l 
 
 0) 
 
 u 
 
 
 2 
 
 3 
 
 > 
 
 u. 
 
 a. 
 
 1—1 
 
 c 
 
 •r-l 
 
 o 
 
 ,o 
 
 c; 
 
 ni 
 
 
 4-J 
 
 C 
 
 P- 
 
 •r^ 
 
 QJ 
 
 c 
 
 O 
 
 .u 
 
 O 
 
 O 
 
 < 
 
 l-i 
 
 
 Oi 
 
 0) 
 
 
 x; 
 
 X 
 
 ■u 
 
 c/: 
 
 
 •1- 
 
 X) 
 
 iu 
 
 c 
 
 
 ca 
 
 X 
 
 
 4-1 
 
 c 
 
 •r-l 
 
 •rH 
 
 3 
 
 OJ 
 
 
 4-1 
 
 -c 
 
 o 
 
 0/ 
 
 u 
 
 4-J 
 
 a, 
 
 c 
 
 o 
 
 
 u 
 
 
 PU 
 
 
 -n 
 
 *^^ 
 
 c 
 
 X) 
 
 cfl 
 
 C 
 
 
 n) 
 
 # 
 
 rH 
 
 > 
 
 4-1 
 
 &> 
 
 •r-l 
 r— 1 
 
 ^ 
 
 •M 
 
 
 Xi 
 
 14-1 
 
 cd 
 
 O 
 
 4-1 
 
 
 D. 
 
 >, 
 
 01 
 
 4-1 
 
 o 
 
 •r-l 
 
 ~- 
 
 
 
 0) 
 
 <J) 
 
 4J 
 
 o 
 
 « 
 
 •H 
 
 J-1 
 
 4-1 
 
 4J 
 
 W 
 
 C 
 
 ■r-1 
 
 0) 
 
 S-l 
 
 c> 
 
 a> 
 
 c 
 
 4-1 
 
 o 
 
 U 
 
 CJ 
 
 « 
 
 
 M 
 
 c 
 
 W 
 
 ■H 
 
 rC 
 
 OJ 
 
 CJ 
 
 4-1 
 
 
 o 
 
 4-1 
 
 u 
 
 o 
 
 p^ 
 
 D 
 
 
 T3 
 
 •C 
 
 o 
 
 OT 
 
 u 
 
 •H 
 
 c^ 
 
 u. 
 
 11 
 
 ^ 
 
 
 
 X 
 
 4-1 
 
 4-1 
 
 •r-l 
 
 
 ■:? 
 
 U-l 
 
 
 O 
 
 -a 
 
 
 n 
 
 4-1 
 
 -) 
 
 C 
 
 ij 
 
 "J 
 
 PQ 
 
 f: 
 
 
 "J 
 
 c 
 
 > 
 
 ^^ 
 
 O 
 
 c 
 
 u 
 
 CJ 
 
 a 
 
 
 H 
 
 u-^ 
 
 M 
 
 o 
 
 0) 
 
 c 
 x: 
 a, 
 
 ij 
 
 03 
 S-i 
 O 
 CJ 
 
 E 
 
 N T3 
 
 4-) 0) 
 
 )-■ JZ 
 
 D 05 
 
 O 
 
 J3 
 
 D 
 
 C5. 
 
 c 
 :3 
 
 C_3 
 
 O 
 
 o 
 
 CJ 
 
 CM 
 O 
 
 o 
 
 CJ 
 
 CO 
 
 o 
 o 
 
 CJ 
 
 o 
 
 o 
 o 
 
 II. 3-1 
 
-^m 
 
 rn 
 
 a. 
 
 '•1 
 
 -qC 
 
 m 
 
 cx: 
 
 13 
 
 CO 
 
 
 CD 
 
 H 
 
 
 r 1 
 
 _J 
 
 3 
 
 CQ 
 
 Q 
 
 CO 
 
 o 
 
 C3 
 
 ►— 
 
 
 
 
 
 
 
 
 
 
 
 
 CJ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 x 
 
 
 
 t/O 
 
 
 
 
 
 
 
 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 
 
 -r 
 
 
 
 
 
 
 
 
 
 
 
 
 :e 
 
 
 
 
 
 
 
 
 
 
 
 
 _i 
 
 
 
 
 
 
 
 
 
 
 
 
 Lj- 
 
 
 
 
 
 
 
 
 
 
 
 
 C3 
 
 ci: 
 
 X 
 
 
 
 X 
 
 
 
 
 
 
 
 
 c_> 
 
 
 
 
 
 
 
 
 
 
 
 
 =E 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 M 
 
 
 Ji 
 
 
 
 ^ 
 
 
 ^ 
 
 ^ 
 
 
 
 M 
 
 u 
 
 
 U 
 
 
 
 Vj 
 
 
 M 
 
 u 
 
 
 a: 
 
 <D 
 
 n) 
 
 
 q) 
 
 
 
 nj 
 
 
 (U 
 
 ca 
 
 
 o 
 
 P-. 
 
 Pui 
 
 
 (Li 
 
 
 
 (Xi 
 
 
 O. 
 
 cu 
 
 
 ^— 
 
 
 
 
 
 
 
 
 
 
 
 
 «c 
 
 0) 
 
 v 
 
 
 « 
 
 
 
 0) 
 
 
 UJ 
 
 <u 
 
 
 <_> 
 
 00 
 
 00 
 
 
 00 
 
 
 
 CO 
 
 
 60 
 
 60 
 
 
 C3 
 
 <u 
 
 (U 
 
 
 Ci 
 
 
 
 <u 
 
 
 OJ 
 
 <u 
 
 
 •H.J 
 
 l-( 
 
 r-4 
 
 
 r-l 
 
 
 
 r-t 
 
 
 l-l 
 
 r-l 
 
 
 
 t-l 
 
 r-l 
 
 
 l-l 
 
 
 
 r-l 
 
 
 1—1 
 
 l-l 
 
 
 
 o 
 
 O 
 
 
 o 
 
 
 
 O 
 
 
 o 
 
 o 
 
 
 
 u 
 
 U 
 
 
 o 
 
 
 
 o 
 
 
 u 
 
 o 
 
 
 
 
 
 
 
 
 
 
 
 0) 
 
 
 
 
 
 
 
 
 
 
 
 
 
 4-1 
 
 l-l 
 
 
 
 
 
 
 
 
 
 
 to 
 
 
 
 ta 
 
 t3 
 
 
 
 
 
 
 
 
 
 
 3 
 
 
 
 )-i 
 
 OJ 
 
 
 
 
 
 
 
 
 
 
 O 
 
 
 
 4_) 
 
 Vj 
 
 
 
 
 
 
 
 
 
 
 •l-l 
 
 
 r-l 
 
 c 
 
 CJ 
 
 
 
 
 
 
 
 
 
 
 u 
 
 
 ttJ 
 
 (D 
 
 o 
 
 
 
 
 
 
 
 
 
 
 « 
 
 
 OJ 
 
 O 
 
 
 
 
 
 
 
 
 
 
 
 > 
 
 
 u 
 
 ^ 
 
 t4-l 
 
 
 
 
 
 
 
 
 
 ,^-N 
 
 
 
 (U 
 
 o 
 
 o 
 
 
 
 2E 
 
 
 
 
 
 
 iH 
 
 cc 
 
 
 CJ 
 
 o 
 
 
 
 
 C3 
 
 
 
 
 
 
 •• — 
 
 c 
 
 
 
 
 c 
 
 
 
 
 
 
 
 
 
 ON 
 
 •H 
 
 
 t4-4 
 
 c 
 
 o 
 
 
 
 H^ 
 
 
 
 
 
 
 .H 
 
 c 
 
 
 o 
 
 •H 
 
 •l-l 
 
 
 
 «« 
 
 
 
 
 
 
 
 •H 
 
 
 
 tu 
 
 u 
 
 
 
 t_> 
 
 
 
 
 
 
 • 
 
 nJ 
 
 
 c 
 
 4-1 
 
 c3 
 
 
 
 * 
 
 
 
 
 
 
 > 
 
 iJ 
 
 
 o 
 
 o 
 
 ■rA 
 
 
 
 CQ 
 
 
 
 
 
 
 
 p 
 
 
 •r-l 
 
 u 
 
 O 
 
 
 
 =3 
 
 
 
 
 
 
 
 o 
 
 
 4-1 
 
 (X, 
 
 o 
 
 
 
 a. 
 
 
 
 
 
 
 M 
 •U 
 
 CJ 
 
 s 
 
 u 
 
 
 tfl 
 •rH 
 O 
 O 
 
 to 
 to 
 
 x: 
 to 
 
 •r-l 
 ^4 
 
 to 
 to 
 
 < 
 
 
 
 
 
 
 
 
 
 D 
 
 fC 
 
 0) 
 
 v^ 
 
 < 
 
 X 
 
 CO u 
 
 
 
 
 
 
 
 
 
 X 
 
 
 c 
 
 
 
 4-) 
 
 ClH —1 
 
 
 
 
 
 
 >, 
 
 
 
 o 
 
 u-l 
 
 •r-l 
 
 z! 
 
 c 
 
 ■rJ 
 
 S (J 
 
 /-v 
 
 
 
 
 
 f-l 
 
 
 
 nl 
 
 o 
 
 to 
 
 ^ 
 
 n 
 
 ;> 
 
 2 a) 
 
 r-l 
 
 
 
 
 
 o. 
 
 
 
 CO 
 
 
 > 
 
 ^— ' 
 
 o 
 
 
 ^ 3 
 
 nj 
 
 
 
 
 
 D- 
 
 
 
 to 
 
 to 
 
 kJ 
 
 
 •r-l 
 
 TD 
 
 < 
 
 4J 
 
 
 
 
 C| 
 
 3 
 
 
 
 « 
 
 o 
 
 
 OJ 
 
 Vj 
 
 CJ 
 
 CJ 
 
 •r-l 
 
 
 
 X 
 
 nl 
 
 rn 
 
 
 
 
 .—1 
 
 T-l 
 
 —J 
 
 "1' 
 
 ' » 
 
 — 1 ^ 
 
 C- 
 
 
 
 •^\ < 
 
 _jl 
 
 
 r-l 
 
 
 
 ' !^ — 
 
 •^ 
 
 i 
 
 — i-» Xi 
 
 v'j 
 
 
 
 Vi^ (_J 
 
 w 
 
 
 ■a 
 
 ^O 
 
 
 
 (0 
 
 tc 
 
 V 
 
 QJ 
 
 CJ -u 
 
 o 
 
 
 
 c> t, 
 
 (0 
 
 
 o 
 
 0^ 
 
 
 CO 
 
 •l-l 
 
 
 ^ 
 
 
 B 
 
 p 
 
 x 
 
 
 
 
 4-) 
 
 
 o 
 
 l-l 
 
 
 ^ 
 
 Vj 
 
 
 H 
 
 LW 
 
 CJ 
 
 g t4-l 
 
 
 
 
 ""^ o 
 
 C 
 
 
 (^ 
 
 
 
 (C 
 
 CJ 
 
 G 
 
 Cu 
 
 O 
 
 1—1 
 
 to O 
 
 r-l 
 
 
 
 u 
 
 (U 
 
 
 
 
 
 hJ 
 
 4J 
 
 a. 
 
 ^ 
 
 
 a 
 
 X 
 
 ta 
 
 
 
 
 g 
 
 
 •X3 
 
 
 
 
 o 
 
 P^ 
 
 
 00 
 
 a 
 
 00 
 
 )-i 
 
 
 Zo 
 
 c 
 
 (U 
 
 
 r-l 
 
 
 
 ^ 
 
 tfl 
 
 
 . 
 
 c 
 
 3 
 
 . c 
 
 o 
 
 
 "__' 
 
 o 
 
 r-l 
 
 
 >-l 
 
 
 
 o 
 
 )-l 
 
 
 < 
 
 •r-l 
 
 CO 
 
 < •H 
 
 c 
 
 
 Cr: 
 
 •l-l 
 
 D 
 
 
 o 
 
 
 
 •r-l 
 
 tfl 
 
 0) 
 
 
 XJ 
 
 
 4-1 
 
 CJ 
 
 
 as 
 
 u 
 
 a 
 
 
 3 
 
 
 
 ■1-1 
 
 x: 
 
 u 
 
 • 
 
 <U 
 
 1-1 
 
 . OJ 
 
 o 
 
 
 •H 
 
 p 
 
 
 
 
 
 rt 
 
 u 
 
 tt 
 
 o 
 
 CJ 
 
 3 
 
 O 0) 
 
 
 
 Z3 
 
 J-l 
 
 C/D 
 
 
 C 
 
 
 
 JS 
 
 
 >-4 
 
 
 ;^': 
 
 O 
 
 s 
 
 cS 
 
 
 «C 
 
 0) 
 
 
 
 o 
 
 
 
 
 >^ 
 
 4-1 
 
 *■ 
 
 
 r— 1 
 
 
 •r4 
 
 
 
 CL, 
 
 > 
 
 
 
 
 
 > 
 
 >-l 
 
 C 
 
 to 
 
 1—1 
 
 Cij 
 
 1—1 r-l 
 
 x: 
 
 
 
 
 
 
 i-H 
 
 
 
 
 o 
 
 0) 
 
 CO 
 
 ri 
 
 
 l-l ttj 
 
 a. 
 
 
 
 ^ 
 
 nj 
 
 
 0) 
 
 
 
 n 
 
 to 
 
 o 
 
 C 
 
 3 
 
 4J 
 
 0) 3 
 
 r-l 
 
 
 
 (0 
 
 ■U 
 
 
 c 
 
 
 
 < 
 
 c 
 
 c: 
 
 •H 
 
 C 
 
 tt3 
 
 3 C 
 
 <U 
 
 
 
 •H 
 
 0) 
 
 
 (d 
 
 
 
 
 0) 
 
 o 
 
 i-H 
 
 c: 
 
 CJ 
 
 ■o c 
 
 ■o 
 
 
 
 fe 
 
 ■r-l 
 Q 
 
 
 Oi 
 
 
 
 CO 
 
 ^ 
 
 CJ 
 
 r-l 
 •r-l 
 
 < 
 
 x: 
 IS 
 
 •H < 
 CO 
 
 tT3 
 
 
 
 01 
 
 
 
 u 
 
 
 
 • 
 
 T3 
 
 c 
 
 4-) 
 
 x> 
 
 
 -o 
 
 •r-l 
 
 
 
 ^ 
 
 U 
 
 
 < 
 
 
 
 tD 
 
 c 
 
 •r-l 
 
 CO 
 
 c 
 
 t4-| 
 
 • c 
 
 x: 
 
 
 
 •1-1 
 
 o 
 
 
 CO 
 
 
 
 
 « 
 
 CJ 
 
 
 CM 
 
 o 
 
 Q ITM 
 
 . CXi 
 
 
 
 ►J 
 
 iw 
 
 ^— s 
 
 H 
 
 
 
 U-l 
 
 
 4-1 
 
 • 
 
 tA 
 
 
 iTl 
 
 ^•■^ 
 
 
 
 1 
 
 
 CO 
 
 
 
 
 O 
 
 r-l 
 
 O 
 
 cc 
 
 
 >1 
 
 * 
 
 u 
 
 
 
 
 V 
 
 CJ 
 
 b 
 
 ci 
 z--. 
 
 u 
 
 
 
 
 03 
 
 ^J 
 
 
 4J 
 
 4-» 
 
 > 4-1 
 
 <J 
 
 • 
 
 
 
 J«5 
 
 (X, 
 
 nj 
 
 
 /— V 
 
 4-J 
 
 u 
 
 CX, 
 
 • 
 
 tT) 
 
 •r4 
 
 (Tl 
 
 ■v 
 
 c; 
 
 
 
 w 
 
 H-* 
 
 
 
 CO 
 
 u 
 
 ■H 
 
 
 PQ 
 
 
 r-l 
 
 ^ 
 
 •1— 
 
 
 
 iAj 
 
 X 
 
 
 ^^ 
 
 X) 
 
 
 
 o 
 
 to 
 
 x: 
 
 
 -o 
 
 tU 
 
 to -o 
 
 o 
 
 • 
 
 u 
 
 _J 
 
 
 <4-) 
 
 
 0) 
 
 - 
 
 ex 
 
 >- 
 
 to 
 
 ** 
 
 u 
 
 3 
 
 00 CJ 
 
 Vj 
 
 z 
 
 Vj 
 
 H- 
 
 •o 
 
 O 
 
 >-l 
 
 JJ 
 
 "w 
 
 CJ 
 
 ^ 
 
 ■r-l 
 
 r— 1 
 
 4-J 
 
 O' 
 
 C 4-1 
 
 (Xi 
 
 
 O 
 
 F= 
 
 c 
 
 
 (U 
 
 c 
 
 •o 
 
 
 Di 
 
 D-i 
 
 [J^ 
 
 r-l 
 
 C 
 
 
 •H C 
 
 
 •• 
 
 O. 
 
 
 C0 
 
 c 
 
 to 
 
 0) 
 
 o 
 
 
 
 
 
 CJ 
 
 OJ 
 
 r-4 
 
 l-l CJ 
 
 x: 
 
 >> 
 
 OJ 
 
 
 
 o 
 
 •l-l 
 
 Oi 
 
 o 
 
 I-l 
 
 (0 
 
 C 
 
 M_l 
 
 3 
 
 to 
 
 n 
 
 r-l to 
 
 D 
 
 00 
 
 Di 
 
 
 Q) 
 
 •H 
 
 u 
 
 (U 
 
 tu 
 
 .-1 
 
 o 
 
 ■r-l 
 
 o 
 
 -o 
 
 CJ 
 
 c 
 
 <v 
 
 •r^ CJ 
 
 D 
 
 Vj 
 
 
 
 ^ C 
 
 i-l 
 
 rt 
 
 Vj 
 
 
 <u 
 
 •H 
 
 
 
 •r-l 
 
 ^J to 
 
 o 
 
 c 
 
 4-) 1-1 to 
 
 il 
 
 O 
 
 4-1 
 
 
 (0 O 
 
 03 
 
 a, 
 
 a. 
 
 c 
 
 3 
 
 U 
 
 to 
 
 to 
 
 CO 
 
 CL 4J 
 
 'f■^ 
 
 •rl 
 
 CO CX u 
 
 CJ 
 
 >^ 
 
 u 
 
 
 W '-^ •'-' 
 
 N 
 
 
 
 •I-l 
 
 TD 
 
 •w 
 
 V 
 
 4-1 
 
 
 to 
 
 4J 
 
 1/1 
 
 to 
 
 H 
 
 o 
 
 (0 
 
 
 CO i-l 
 
 ■t-l 
 
 • 
 
 ^1 
 
 
 •f-i 
 
 > 
 
 tf 
 
 c 
 
 • 
 
 U 'r-l 
 
 •r^ 
 
 > 
 
 • U ■r4 
 
 
 
 u 
 
 
 •O t^ -t-l 
 
 1— ( 
 
 cd 
 
 01 
 
 u 
 
 C/3 
 
 ■H 
 
 c. 
 
 3 
 
 Q 
 
 OJ tl 
 
 (-1 
 
 .-J 
 
 Ci o E 
 
 X 
 
 r-l 
 
 ij 
 
 
 
 •H 
 
 
 o. 
 
 (X, 
 
 
 i-J 
 
 r3 
 
 O 
 
 
 Cu CJ 
 
 4-J 
 
 
 n. o 
 
 to 
 
 3 
 
 c 
 
 
 iJ 
 
 • 
 
 <TJ 
 
 [Jl 
 
 • 
 
 O 
 
 J= 
 
 .§ 
 
 • 
 
 -■^ j:: 
 
 ^3 
 
 Vj 
 
 • (tj x: 
 
 •r4 
 
 ttj 
 
 o 
 
 
 : ^-^ Pj 
 
 3 
 
 u 
 
 (X. 
 
 : 
 
 > 
 
 <; 
 
 O 
 
 > 
 
 (X, CJ 
 
 • '• 
 
 o 
 
 CQ Cu CJ 
 
 U- 
 
 CXi 
 
 o 
 
 c? 
 
 
 
 
 
 
 
 
 
 
 
 
 z 
 
 
 
 
 
 
 
 
 
 
 
 
 
 in 
 
 >o 
 
 
 r«. 
 
 
 
 00 
 
 
 o 
 
 o 
 
 
 O 
 
 O 
 
 
 O 
 
 
 
 o 
 
 
 o 
 
 .—1 
 
 
 CJ 
 
 O 
 
 o 
 
 
 o 
 
 
 
 o 
 
 
 o 
 
 o 
 
 
 
 o 
 
 o 
 
 
 CJ 
 
 
 
 CJ 
 
 
 u 
 
 CJ 
 
 
 ir.3-2 
 
Q. 
 
 -a: 
 
 CD 
 
 _j 
 
 C9 
 
 IX. 
 
 W 
 CO 
 
 H 
 
 Q 
 
 o 
 
 »— 
 
 
 
 
 
 
 
 
 
 
 
 
 t3 
 
 
 
 
 
 
 
 
 
 
 
 
 -< 
 
 
 
 
 
 
 
 
 
 
 
 
 o: 
 
 
 
 
 
 
 
 
 
 
 
 
 1 — 
 
 
 
 
 
 
 
 
 
 
 
 
 t>o 
 
 
 
 
 
 
 
 
 
 
 
 
 OQ 
 
 
 
 
 
 
 
 
 
 
 
 
 «.£ 
 
 
 
 
 
 
 
 
 
 
 
 
 .— 
 
 
 
 
 
 
 
 
 
 
 
 
 3 
 
 
 
 
 
 
 
 
 
 
 
 
 lA. 
 
 
 
 
 
 
 
 
 
 
 
 
 C3 
 
 
 
 
 
 
 
 
 
 X 
 
 
 
 tJ 
 
 
 
 
 
 
 
 
 
 
 
 
 K 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 tu 
 
 
 
 
 
 
 
 
 
 
 
 
 a 
 
 
 
 
 
 
 
 
 
 
 
 
 1-1 
 
 
 
 
 /<: 
 
 
 ^ 
 
 
 4<5 
 
 
 J^ 
 
 
 <w 
 
 ^ 
 
 
 
 V4 
 
 
 V4 
 
 
 H 
 
 
 M 
 
 
 «w 
 
 M 
 
 
 2: 
 
 n> 
 
 
 ca 
 
 
 CD 
 
 
 qS 
 
 
 
 
 CIt 
 
 
 CD 
 
 PL4 
 
 
 P4 
 
 
 CI4 
 
 
 PU 
 
 
 
 P4 
 
 
 1 — 
 
 
 
 
 
 
 
 
 
 ft 
 
 
 
 «« 
 
 0) 
 
 
 <u 
 
 
 OJ 
 
 
 V 
 
 
 
 
 (U 
 
 
 CJ> 
 
 00 
 
 
 to 
 
 
 C>Q 
 
 
 00 
 
 
 •H 
 
 t>0 
 
 
 
 
 <u 
 
 
 0) 
 
 
 (U 
 
 
 4) 
 
 
 CO 
 
 (U 
 
 
 —J 
 
 »H 
 
 
 r-t 
 
 
 1-4 
 
 
 rH 
 
 
 •H 
 
 r-l 
 
 
 
 r-l 
 
 
 i-H 
 
 
 rH 
 
 
 I-l 
 
 
 > 
 
 i-< 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 CJ 
 
 
 CJ 
 
 
 Q 
 
 CJ 
 
 
 
 
 
 
 
 
 
 « 
 
 
 
 
 
 
 
 
 
 
 
 
 
 M 
 
 
 
 
 
 
 
 
 
 
 
 
 
 -j: 
 
 
 
 
 ti 
 
 
 
 
 
 
 
 
 
 .C 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 •rl 
 
 
 
 
 
 
 
 
 
 •rl 
 
 
 
 
 U 
 
 
 
 
 
 
 
 
 
 g 
 
 
 
 
 
 
 3 
 
 
 
 
 
 
 
 
 
 
 
 
 
 •0 
 
 
 •ss. 
 
 
 
 
 
 
 
 0) 
 
 
 
 
 p 
 
 
 C3 
 
 
 
 
 
 
 
 C 
 
 
 
 
 U 
 
 
 ►— 
 
 
 
 
 
 
 
 1-1 
 
 
 
 
 r^ 
 
 
 «: 
 
 
 
 
 
 
 
 U 
 
 
 
 
 4) 
 
 
 CJ 
 
 
 <u 
 
 
 rH 
 
 
 
 
 
 
 
 
 .c 
 
 
 — ^ 
 
 
 
 
 
 
 CM 
 
 
 
 U 
 
 
 
 
 4J 
 
 
 ca 
 
 
 c 
 
 
 
 (U 
 
 CU 
 
 
 
 
 
 
 £3 
 
 
 <u 
 
 
 I-l 
 
 
 
 
 
 
 
 
 e 
 
 
 a. 
 
 
 
 
 
 
 0) 
 
 V4 
 
 
 ^ 
 
 
 
 
 
 00 
 
 
 
 
 C 
 
 
 ■H 
 
 
 <u 
 
 CO 
 
 CU 
 
 
 
 <u 
 
 c 
 
 
 
 
 
 
 
 f-i 
 
 
 m 
 
 4J 
 
 4J 
 
 
 
 4-> 
 
 1-1 
 
 
 
 
 
 
 
 ■• 
 
 
 
 
 0) 
 0( 
 
 3 
 
 
 
 )-i 
 
 
 
 
 
 &" 
 
 
 c 
 
 
 
 
 <u 
 
 •H 
 
 
 
 4J 
 
 <u 
 
 
 
 
 
 
 
 
 
 > 
 
 4J 
 
 
 
 C 
 
 ^ s 
 
 
 
 
 u 
 
 
 •H 
 
 
 r-l 
 
 
 CO 
 
 
 
 <u 
 
 CO 
 
 
 
 
 M 
 
 
 W 
 
 * 
 
 ca 
 
 •a 
 
 (3 
 
 
 
 
 
 § 3 
 
 
 
 
 3 
 
 
 •H 
 
 a) 
 
 3 
 
 C 
 
 M 
 
 
 
 c 
 
 
 
 
 •o 
 
 
 T3 
 
 CJ 
 
 c 
 
 CO 
 
 
 
 
 
 
 $? 
 
 
 
 
 C 
 
 
 W 
 
 3 
 
 ^ 
 
 
 ^ 
 
 
 
 CJ 
 
 <-> ^ 
 
 
 
 00 
 
 
 H 
 
 
 
 •u 
 
 (U 
 
 
 
 CJ^ 
 
 
 CJ) 
 
 
 
 VC 
 
 
 1 
 
 
 .C 
 
 c 
 
 
 -u 
 
 M 
 
 vO 
 
 c 
 
 r-l 
 
 
 
 CPi 
 
 
 1 
 
 
 u 
 
 Q) 
 
 J3 
 
 CO 
 
 ca 
 
 CT> 
 
 1-1 
 
 X> « 
 
 
 
 i-l 
 
 
 nJ 
 
 
 <i 
 
 z 
 
 4J 
 
 >-l 
 
 0) 
 
 r-l 
 
 0) 
 
 I^ 
 
 
 
 
 (U 
 
 
 
 
 
 
 4-1 
 
 (0 
 
 
 
 4-1 
 
 C 0) 
 
 
 
 
 CO 
 
 
 ^ 
 
 00 
 
 ^-s n 
 
 c 
 
 dl 
 
 
 
 
 
 ><< O' 
 
 
 
 
 
 
 <u 
 
 c 
 
 CO 
 
 <u 
 
 H 
 
 oi 
 
 
 
 V4 
 
 X 
 
 
 10" 
 
 
 (U 
 
 
 en 
 
 iH 
 
 Ft . *i 
 
 
 
 r-l 
 
 
 
 
 P-. 
 
 . w 
 
 
 
 
 .c 
 
 
 ca 
 
 >i 
 
 § """ 
 
 c 
 
 I-l 
 
 4J 
 
 
 
 
 s 
 
 
 
 
 u 
 
 
 to 
 
 A 
 
 s.^ 
 
 
 
 to 
 
 
 to 
 
 
 
 CO 
 
 .^ 
 
 
 
 £ 
 
 
 1-1 
 
 4-1 
 
 c 
 
 
 •H 
 
 St 
 
 
 
 1-1 
 
 ta 
 
 
 ^ 
 
 
 Q 
 
 
 
 
 iH 
 
 !-i 3 
 
 c 
 
 p 
 
 -o 
 
 
 
 fl4 
 
 1 
 
 
 •< 
 
 
 (4 
 
 
 
 s 
 
 0) 
 
 •H 
 
 
 
 •r^ 
 
 
 
 
 « 
 
 
 
 
 (4-1 
 
 " 
 
 •0 
 
 
 •a 
 
 a 
 
 c 
 
 s 
 
 
 
 14-1 
 
 W Q 
 
 
 
 
 
 •0 
 
 c 
 
 (1) 
 
 >^ 
 
 4-1 
 
 
 
 C> 
 
 
 
 
 
 00 M 
 
 
 
 
 "O 
 
 
 
 ca 
 
 C 
 
 c 
 
 
 
 
 
 
 
 » 
 
 
 C Z 
 
 
 
 
 
 
 
 
 
 1-1 
 
 CO -r-l 
 
 u 
 
 w 
 
 ^ 
 
 
 5t! 
 
 to 
 
 1-1 5 
 
 <u 
 
 
 c 
 
 
 
 
 tij 
 
 x; 
 
 Q 
 
 w 
 
 Ph 
 
 
 to 
 
 
 d 
 
 0) 
 
 r-l 
 
 4-> 
 
 
 
 
 
 u^ 
 
 
 4J 
 
 
 • u 
 
 
 t3 
 
 •r-l 
 
 
 b 
 
 to 
 
 r-4 4-1 
 
 to 
 
 
 1-1 
 
 
 
 XI 
 
 I-l 
 
 M 
 
 3 
 
 x: 
 
 C 
 
 u 
 
 
 
 ^ 
 
 lH C 
 
 u 
 
 
 4-1 
 
 
 4J 
 
 c 
 
 CO 
 
 
 
 E 
 
 to 
 
 ca 
 
 u 
 
 
 
 
 
 4J -rl 
 
 4J 
 
 
 •H 
 
 
 U 
 
 CO 
 
 /^ <u 
 
 
 • 01 
 
 •H 
 
 
 (U 
 
 
 4J 
 
 TJ 
 
 CO 
 
 c 
 
 
 M 
 
 ^^ 
 
 r-l 
 
 
 CO "X. 
 
 *• 
 
 Q £ 
 
 fn 
 
 > 
 
 
 
 
 
 c 
 
 •«-) 
 
 0) 
 
 
 4-1 
 
 CO 
 
 
 CO 
 
 IXj 
 
 ca 
 
 U 
 
 
 I-l 
 
 
 
 a 
 
 to 
 
 • 
 
 
 
 
 D 
 
 b 
 g 
 
 4-1 
 
 4-1 
 
 § "^ 
 
 OJ 
 
 13 
 
 <4-l 
 
 t-i 
 
 , 
 
 
 
 
 
 cc: 4J 
 
 c 
 
 
 ^: 
 
 ca 
 
 c 
 
 
 CO 
 
 C 0) 
 
 
 
 to 
 
 eo 
 
 
 •H 
 
 
 CO 
 
 
 
 
 
 
 4) 
 
 N-^ 
 
 
 ca x; 
 
 
 c 
 
 
 
 4J 
 
 b£ 
 
 • 
 
 CJ 
 
 
 •a 
 
 v^ 
 
 •0 
 
 E 
 
 c 
 
 0) 
 
 4-i 
 
 to 
 
 
 
 « 
 
 
 ■rH ^-N 
 
 c 
 
 PQ T) 
 
 
 IJkJ 
 
 c 
 
 
 <u 
 
 cU 
 
 tn 
 
 X. 
 
 CO 
 
 <u 
 
 •H 
 
 
 
 4J 
 
 iJ CO 
 
 1-1 
 
 <U 
 
 c 
 
 pi 
 
 ca 
 
 0) 
 
 4J 
 
 r-l 
 
 00 -H 
 
 4-1 
 
 00 •4-1 
 
 u 
 
 4-» 
 
 
 u 
 
 (U [14 
 
 4-) 
 
 « u 
 
 1-1 
 
 
 rH 
 
 c 
 
 a. 
 
 C 4J 
 
 
 c 
 
 D 
 
 •H 
 
 . », 
 
 
 
 ^ e 
 
 to 
 
 r-4 C 
 
 <u 
 
 t= 
 
 tn 
 
 u 
 
 0) 
 
 0- 
 
 •H -H 
 
 e 
 
 ■r-l 
 
 4-1 
 
 Vj 
 
 CO 
 
 a 
 
 
 01 
 
 r-l 0) 
 
 u 
 
 
 TD 
 
 0) 
 
 <n 
 
 D 
 
 r-l U 
 
 
 
 r-H W 
 
 >; 
 
 4-» 
 
 S 
 
 0) 
 
 X ^-^ 
 
 H 
 
 <u to 
 
 
 
 
 
 
 
 <u 
 
 CO 
 
 1-1 4J 
 
 u 
 
 r-l U 
 
 •H 
 
 3 
 
 
 
 eti 
 
 to 
 
 
 3 (U 
 
 u 
 
 
 
 
 c 
 
 tj 
 
 
 •H D 
 
 <4-l 
 
 •H C 
 
 td 
 
 S 
 
 )-i 
 
 
 1-1 r-l 
 
 
 •a u 
 
 (U 
 
 
 fi. 
 
 ty 
 
 a. 
 
 c 
 
 l-> 2^ 
 
 
 4_) 1-1 
 
 
 
 4J 
 
 4J 
 
 fn <u c 
 
 to 
 
 1-1 D. 
 
 
 
 
 s 
 
 
 •r-l 
 
 CO 
 
 >^ 
 
 CO T3 
 
 a 
 
 1—1 
 
 to 
 
 
 
 XI 
 
 •rl 
 
 CO 
 
 s: 
 
 
 c 
 
 
 X 
 
 a; 
 
 p 
 
 4-1 
 
 aj 
 
 
 
 to 
 
 g 
 
 ta 
 
 >s a) 1-1 
 
 to 
 
 s: 
 
 to 
 
 
 •H 
 
 • 
 
 u 
 
 4-1 
 
 • u 
 
 c 
 
 • OJ 
 
 r-l 
 
 £ 
 
 V4 
 
 4.1 4-1 
 
 ?- 
 
 • tj 
 
 1-1 
 
 
 
 < 
 
 Q) 
 
 
 
 Oi d) 
 
 
 
 cd u 
 
 0) 
 
 •H 
 
 <: 
 
 4-1 
 
 4J Oi 1-1 
 
 r-< 
 
 Q <U 
 
 ta 
 
 
 
 
 
 (1) 
 
 u 
 
 ■a 
 
 
 
 
 
 > 
 
 4-) 
 
 
 c 
 
 CO 4-1 
 
 to 
 
 tu 
 
 
 
 Cu 
 
 • 
 
 D- 
 
 &4 
 
 • 
 
 m 
 
 • u 
 
 
 
 a 
 
 • 
 
 
 
 ft, ♦ (U 
 
 (= 
 
 • 0. 
 
 (4-1 
 
 
 fl^ 
 
 
 
 CO 
 
 
 « 2; 
 
 - 
 
 « PU 
 
 Q 
 
 
 
 Q 
 
 
 
 = P4 fri 
 
 < 
 
 > CO 
 
 
 
 C3 
 
 
 
 
 
 
 
 
 
 
 
 
 se 
 
 
 
 
 
 
 
 
 
 
 
 
 C3 
 
 •-4 
 
 
 CVJ 
 
 
 fO 
 
 
 »t 
 
 
 ir> 
 
 NO 
 
 
 rH 
 
 
 I-l 
 
 
 i-t 
 
 
 1-4 
 
 
 iH 
 
 1-4 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CJ 
 
 
 CJ 
 
 
 U 
 
 
 U 
 
 u 
 
 
 II. 3-3 
 
CU 
 
 -a: 
 ex: 
 
 CD 
 
 cs 
 
 CO 
 QQ 
 CO 
 
 a. 
 
 CO 
 
 w 
 w 
 
 H 
 U 
 
 ra 
 
 Q 
 
 o 
 
 erf 
 
 »— 
 
 
 
 
 
 
 
 
 
 (-3 
 
 
 
 
 
 
 
 
 
 -a: 
 
 
 
 
 
 
 
 
 
 
 f< 
 
 
 
 
 
 X 
 
 
 
 t/O 
 
 
 
 
 
 
 
 
 
 en 
 
 
 
 
 
 
 
 
 
 ::s 
 
 
 
 
 
 
 
 
 
 _J 
 
 
 
 
 
 
 
 
 
 k<- 
 
 
 
 
 
 
 
 
 
 CD 
 
 cc: 
 
 X 
 
 
 
 
 
 X 
 
 
 
 <_> 
 
 
 
 
 
 
 
 
 
 S= 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 ^ 
 
 ^ 
 
 
 
 
 
 u 
 
 
 
 u 
 
 V4 
 
 
 
 
 3= 
 
 rt 
 
 
 
 ci 
 
 (tJ 
 
 
 
 
 c=> 
 
 f^ 
 
 
 
 (U 
 
 PM 
 
 
 
 
 >— 
 
 
 
 
 
 
 
 
 
 «c 
 
 aj 
 
 
 
 Q) 
 
 0) 
 
 <u 
 
 
 
 c_> 
 
 bO 
 
 
 
 60 
 
 60 
 
 r-l 
 
 
 
 C3 
 
 <u 
 
 
 
 ■ <U 
 
 (U 
 
 4J 
 
 
 
 _i 
 
 i-i . 
 i-i 
 
 o 
 u 
 
 
 
 t-t 
 t-l 
 
 o 
 o 
 
 r-l 
 r-l 
 O 
 O 
 
 4-1 
 
 td 
 
 CO 
 
 
 
 
 (0 
 
 
 
 
 
 
 
 
 
 u 
 
 
 
 nsl 
 
 
 
 
 
 
 
 o 
 
 
 
 •o 
 
 c 
 
 
 
 0) 
 
 
 
 
 •rl 
 
 
 
 c 
 
 •1-1 
 
 1-4 
 
 <4-l 
 
 
 t-l 
 
 
 
 
 CU 
 
 
 
 rt 
 
 r-4 
 
 ca 
 
 O 
 
 
 ,Q 
 
 
 
 
 CD 
 
 
 
 
 o 
 
 (U 
 
 
 <-i 
 
 1-1 
 
 
 
 
 H 
 
 
 
 w 
 
 6 
 
 u 
 
 > 
 
 nJ 
 
 CO 
 
 
 
 
 
 
 
 o 
 
 (U 
 
 <D 
 
 ■U 
 
 (U 
 
 CO 
 
 
 
 
 "d 
 
 
 
 1-1 
 
 w 
 
 O 
 
 1-1 
 
 v< 
 
 o 
 
 
 
 
 c 
 
 
 
 4J 
 
 
 
 f-4 
 
 .(U 
 
 P^ 
 
 
 
 CD 
 
 a 
 
 
 
 OT 
 
 •a 
 
 4-1 
 
 (fl 
 
 u 
 
 
 
 
 < 
 
 
 
 1-1 
 
 c 
 
 o 
 
 3 
 
 
 "H 
 
 
 
 
 
 
 
 u 
 
 nj 
 
 
 O 
 
 M-l 
 
 C 
 
 
 
 P^ 
 
 «d 
 
 
 
 OJ 
 
 
 c 
 
 
 O 
 
 cd 
 
 
 
 «C 
 
 >» o 
 
 
 
 +J 
 
 OJ 
 
 o 
 
 r-l 
 
 1 
 
 
 
 
 <-S 
 
 o 
 
 
 
 u 
 
 u 
 
 •H 
 
 nJ 
 
 c 
 
 C! 
 
 
 
 1 
 
 w » 
 
 
 
 rt 
 
 (d 
 
 4J 
 
 C 
 
 o 
 
 O 
 
 
 
 C33 
 
 >^ 
 
 
 
 >-l 
 
 vj 
 
 (d 
 
 O 
 
 1-1 
 
 1-1 
 
 
 
 13 
 
 ^ 4) 
 
 
 
 rt 
 
 vj 
 
 1-1 
 
 ■H 
 
 •4J 
 
 .4-1 
 
 
 
 a. 
 
 C --I 
 
 
 
 ^ 
 
 C 
 
 o 
 
 4-1 
 
 nJ 
 
 C^ 
 
 
 
 
 Vj 1-1 
 
 
 
 CJ 
 
 a; 
 
 o 
 
 •rH 
 
 1-1 
 
 c 
 
 
 
 
 O S 
 
 
 
 
 o 
 
 w 
 
 n 
 
 o 
 
 w 
 
 
 
 
 u 
 
 
 
 1-1 
 
 a 
 
 to 
 
 4J 
 
 o 
 
 O 
 
 
 
 
 
 
 
 ca 
 
 o 
 
 <: 
 
 3 
 
 cd 
 
 n 
 
 (^ 
 
 
 
 
 <^ •> o 
 
 
 
 o 
 
 o 
 
 /-N 
 
 K 
 
 C 
 
 CO '< 
 
 
 
 
 
 -a <u 
 
 
 
 1-1 
 
 
 w c 
 
 
 ■H 
 
 U, 
 
 
 
 
 CJ 
 
 
 
 w 
 
 e 
 
 & <d 
 
 CU 
 
 I-I 
 
 P-i 
 
 •r4 
 
 
 
 
 ■u 1-1 OS 
 
 
 
 > 
 
 •H 
 
 53 o 
 
 x: 
 
 o 
 
 § .c 
 
 .OJ 
 
 00 
 
 
 
 d oS 
 
 
 
 x: 
 
 O 
 
 2 T" 
 
 4-1 
 
 E 
 
 g td 
 
 C! 
 
 00 
 
 
 
 o 
 
 
 
 l-M 
 
 
 '^^' M 
 
 
 ■^ 
 
 "— ' w 
 
 1-1 
 
 CM 
 
 
 
 u e '-i 
 
 
 
 
 o 
 
 a) 
 
 
 CO 
 
 •H 
 
 
 1 
 
 
 
 v_X O f-^ 
 
 
 
 <u 
 
 p 
 
 ^ -3 
 
 0) 
 
 
 <U i-l 
 
 • M 
 
 ■^t 
 
 
 
 Vj (U 
 
 
 
 x: 
 
 CU 
 
 4-1 
 
 -o 
 
 t-l C) 
 
 CO 
 
 00 
 
 
 
 CTi 
 
 4-1 3 
 
 
 
 H 
 
 
 >-i 
 
 n 
 
 n 
 
 J-i 6 O 
 
 CJ <d r~- 
 
 c 
 
 tN 
 
 
 
 vD 
 
 •o 
 
 
 
 
 4-1 
 
 0) 0) 
 
 u 
 
 CO 
 
 1-1 
 
 
 
 
 a> 
 
 0) •!-< 
 
 
 
 
 03 
 
 g x: 
 
 4-j 
 
 
 g 0^ 
 
 tu 
 
 •• 
 
 
 
 fH 
 
 Tl C/5 
 
 
 
 • • 
 
 0) 
 
 E 4-1 
 
 c 
 
 <u 
 
 g U-l .-1 
 
 *-l 
 
 ^-\ 
 
 
 
 nJ 
 
 
 
 to 
 
 
 rt 
 
 0) 
 
 ■U 
 
 3 o 
 
 o 
 
 CM 
 
 
 
 S = 
 
 
 
 "O 
 
 ^ 
 
 K <4-l 
 
 o 
 
 td 
 
 K 
 
 u 
 
 v-x 
 
 
 
 Q) > 
 
 
 
 o 
 
 
 O 
 
 c 
 
 u 
 
 60 
 
 Pt 
 
 00 
 
 
 Co" 
 
 rt ij 
 
 
 
 o 
 
 e 
 
 
 o 
 
 4-) 
 
 1 ^ 
 
 
 I-I 
 
 
 
 ■u cd « 
 
 
 
 p^ 
 
 o 
 
 <: M 
 
 u 
 
 c 
 
 < 1-1 
 
 si 
 
 
 
 ci: 
 
 ta (-1 .--\ 
 
 
 
 
 u 
 
 c 
 
 
 0) 
 
 4J 
 
 0} 
 
 A 
 
 
 3= 
 
 nj 4J T3 
 
 
 
 c 
 
 M-l 
 
 • 1-1 
 
 c 
 
 o 
 
 • 0) 
 
 •y-t 
 
 t 
 
 
 
 P^ C O 
 
 
 
 1-1 
 
 
 O iJ 
 
 •r-l 
 
 
 
 O CJ 
 
 b 
 
 
 
 rs 
 
 (U rJ 
 
 
 
 
 OJ 
 
 0) 
 
 (U 
 
 o 
 
 ^ s 
 
 
 4-1 
 
 
 
 «c 
 
 M-4 U r-l 
 
 
 
 <u 
 
 ■V 
 
 » <u 
 
 4-) 
 
 a 
 
 
 0) 
 
 to 
 
 
 
 
 ° § c? 
 
 
 <!■ 
 
 4-1 
 
 cd 
 
 to S 
 
 o 
 
 
 1-4 r-4 
 
 t-l 
 
 1-1 
 
 
 
 
 
 00 
 
 cd 
 
 S 
 
 bO 
 
 I-I 
 
 c 
 
 I-I td 
 
 -9 
 
 e 
 
 
 
 
 (1) u ^ 
 
 
 1 
 
 Vj 
 
 
 C i-t 
 
 Ph 
 
 •H 
 
 0) 3 
 
 3 
 
 /^ a) 
 
 
 
 
 
 00 
 
 ■u 
 
 n) 
 
 •r-l nj 
 
 
 0) 
 
 3 g 
 
 I-I 
 
 CO x: 
 
 
 
 
 r2 C 
 
 
 r^ 
 
 c 
 
 4-) 
 
 1-4 ■ 3 
 
 x: 
 
 4-1 
 
 -o c 
 
 o 
 
 fa CJ 
 
 
 
 
 ni 1-1 M-i 
 
 
 
 dJ 
 
 <J) 
 
 I-I C 
 
 w 
 
 o 
 
 •r4 <j: 
 
 CO 
 
 )^ 
 
 
 
 
 > dJ o 
 
 
 ,, 
 
 o 
 
 nj 
 
 •r4 C 
 
 •r4 
 
 Vj 
 
 CO 
 
 
 S 13 
 
 
 
 
 u 
 
 
 ^N 
 
 C 
 
 PLI 
 
 4J <: 
 
 U-> 
 
 PM 
 
 j: 
 
 (4-1 
 
 v-/ O 
 
 
 
 
 <U O >^ 
 
 
 1— ( 
 
 o 
 
 
 w 
 
 
 
 • 4-1 
 
 o 
 
 O 
 
 
 
 
 > H 4-1 
 
 
 Nw' 
 
 u 
 
 U-l 
 
 <t 
 
 IM 
 
 4-1 
 
 Q <r 
 
 
 >» f^ 
 
 
 
 
 •H flj 1-1 
 
 
 ^ 
 
 
 o 
 
 • LO 
 
 o 
 
 (d 
 
 m 
 
 01 
 
 u 
 
 
 
 
 4J U 
 
 
 vT 
 
 c 
 
 
 oi 
 
 
 <u 
 
 • 
 
 <u 
 
 3 -O 
 
 
 
 
 'ri JH U 
 
 
 
 •r^ 
 
 c 
 
 u 
 
 0) 
 
 r^ 
 
 > iJ 
 
 X 
 
 O C 
 
 
 
 
 u <n d) 
 
 
 *« 
 
 01 
 
 o 
 
 • <d 
 
 3 
 
 ^ 
 
 (« 
 
 <u 
 
 ^ cd 
 
 
 
 
 JJ -rl .> 
 
 
 t 
 
 u 
 
 •H 
 
 m 
 
 I-I 
 
 
 •* 
 
 r-l 
 
 
 
 
 ■ » ■ 
 
 V fM •<-> 
 
 /-^ 
 
 o 
 
 4J 
 
 T3 
 
 rt 
 
 E 
 
 to -a 
 
 CU 
 
 • r-4 
 
 
 
 f=t 
 
 Z C 
 
 M 
 
 u 
 
 u 
 
 <d 
 
 « (U 
 
 > 
 
 o 
 
 60 o 
 
 e 
 
 pq td 
 
 
 
 ,C S 
 
 fa 
 
 Ul 
 
 P-, 
 
 3 
 
 I-l u 
 
 
 )-l 
 
 C iJ 
 
 o 
 
 J-l 
 
 
 
 P= 
 
 ■« iJ v.^ 
 
 •H 
 
 
 I-I 
 
 r-4 C 
 
 I-I 
 
 U-l 
 
 •H C 
 
 o 
 
 " 3 
 
 
 
 
 C -H 
 
 E 
 
 SI 
 
 rt 
 
 0) <u 
 
 m 
 
 
 r-l O 
 
 
 1-1 4-1 
 
 
 
 
 <tf 3 <U 
 
 ^-^ 
 
 0) 
 
 m 
 
 > 
 
 3 W 
 
 4-1 
 
 <u 
 
 I-I OJ 
 
 0} 
 
 r-l 1-1 
 
 
 
 
 CJ 
 
 
 j:: 
 
 •r-( 
 
 w 
 
 T) <U 
 
 c 
 
 -o 
 
 •r^ <U 
 
 4-1 
 
 1-1 3 
 
 
 
 
 >» -a 3 
 
 <U 
 
 o 
 
 ptH 
 
 
 ■H U W 
 
 1) 
 
 J3 
 
 u u w 
 
 ctf 
 
 J5 il 
 
 
 
 
 U <U i^ 
 
 I-I 
 
 
 
 >- 
 
 CO Cu 4-1 
 
 g 
 
 CO cu 4J 
 
 x: 
 
 C "^ 
 
 
 
 
 1-1 jr 
 
 M 
 
 1—1 
 
 U-4 
 
 u 
 
 05 
 
 0) 
 
 
 « 
 
 a. 
 
 •M U 
 
 
 
 
 t-l o • 
 
 (1) 
 
 rt 
 
 O 
 
 o 
 
 • )-l -H 
 
 r-l 
 
 (d 
 
 . Vj -I-I 
 
 CO 
 
 - a. 6C 
 
 
 
 
 rt -H 3: 
 
 
 a 
 
 
 (/I 
 
 Q <u e 
 
 a 
 
 4J 
 
 cti ty 6 
 
 o 
 
 0) CO <; 
 
 
 
 
 D iJ 
 
 E 
 
 1-1 
 
 OJ 
 
 c 
 
 a. CJ 
 
 D 
 
 05 
 
 Q. CJ 
 
 x: 
 
 o 
 
 
 
 
 cy c • 
 
 CO 
 
 <1J 
 
 03 
 
 0) 
 
 • (U x; 
 
 3 
 
 td 
 
 • (TJ x: 
 
 Oi 
 
 0) • • 
 
 
 
 
 : w 3 
 
 S 
 
 o 
 
 3 
 
 M 
 
 > PU u 
 
 CO| fX, 
 
 m ct, o 
 
 - 
 
 P •-) •-) 
 
 
 
 <=> 
 
 
 
 
 
 
 
 
 
 ae 
 
 
 
 
 
 
 
 
 
 
 f>i 
 
 
 
 00 
 
 o> 
 
 o 
 
 
 
 1-4 
 
 
 
 I-I 
 
 r-i 
 
 CM 
 
 
 
 CJ 
 
 O 
 
 
 
 8 
 
 o 
 o 
 
 O 
 
 o 
 
 
 
 II. 3-4 
 
re 
 ex.' 
 
 «x 
 a: 
 
 CO 
 
 W 
 
 H 
 U 
 to 
 
 § 
 
 
 AS 
 
 
 14 
 
 ar 
 
 (d 
 
 o 
 
 ►4 
 
 •a: 
 
 4) 
 
 C.3 
 
 (SO 
 
 (=} 
 
 Q> 
 
 _j 
 
 i-l 
 
 
 r-l 
 
 
 O 
 
 
 CJ 
 
 c 
 o 
 
 X 
 
 X 
 
 X 
 
 u 
 
 nt 
 
 P4 
 0) 
 
 w 
 
 r-l 
 1-4 
 
 O 
 O 
 
 (U 
 
 
 o 
 
 
 ^ 
 
 ^ 
 
 H-l 
 
 U 
 
 o 
 
 n> 
 
 
 A4 
 
 (3 
 
 
 O 
 
 9) 
 
 1-i 
 
 00 
 
 CO 
 
 <U 
 
 1-1 
 
 i-l 
 
 > 
 
 r-4 
 
 •1-1 
 
 o 
 
 Q 
 
 O 
 
 15 
 
 CO 
 P4 
 
 0) 
 bO 
 0) 
 
 o 
 
 CO 
 
 •H 
 0) 
 
 .e: 
 
 4J 
 
 •H 
 
 Xi 
 U 
 
 0) 
 
 c 
 
 0) 
 
 e 
 <u 
 I-l 
 a. 
 a. 
 
 -o 
 
 0) 
 >-i 
 (4 
 
 •O 
 
 cd 
 
 )^ 
 
 3 
 o 
 t-l - 
 
 
 Q 
 
 CM 
 
 I-J 
 
 . O 
 2 "^ 
 
 -a «^ 
 CO •• 
 
 0) 
 
 01 
 PQ 
 
 a 
 u 
 
 Vi 
 
 3 
 
 o 
 
 
 XI 
 
 a 
 o 
 
 P4 
 
 o 
 
 6 
 P-. 
 
 to 
 
 M 
 
 o 
 
 CO fO 
 
 >4-l O 
 
 o 
 
 0) 
 
 >^ *-" 
 
 00 
 
 > -o o 
 
 O 0) -r-l 
 
 X O U 
 
 y Dd -r-l 
 
 <: • 4) 
 
 1 (i< (i< 
 
 c 
 
 <u o 
 
 > U 
 
 •H 
 
 •u C 
 
 •r-l -H 
 
 U 0) 
 
 4-1 U 
 
 D O 
 
 en 
 
 ^ 
 
 W) 
 
 tj 
 
 c 
 
 to 
 
 tH 
 
 ■r-l 
 
 1-1 
 
 e 
 
 1-1 
 
 (U 
 
 •H 
 
 x: 
 
 JJ 
 
 o 
 
 M 
 
 
 
 1-1 
 
 • 
 
 « 
 
 Qi 
 
 0) 
 
 
 I-l 
 
 • 
 
 4) 
 
 pa 
 
 O 
 
 w 
 
 4-1 
 
 O 
 
 1-1 
 
 0) 
 
 o 
 c 
 <u 
 
 •r-t 
 
 > 
 
 T^ 
 
 S 
 
 t-< 
 
 0) 
 
 1-1 
 
 1-1 
 
 1-1 
 
 > 
 
 u 
 
 cu 
 
 Vi 
 
 od 
 
 • 
 
 0) 
 
 oi 
 
 <u 
 
 
 1-1 
 
 • 
 
 V4 
 
 pq 
 
 0) 
 
 
 ^ 
 
 M 
 
 to 
 
 iH 
 
 •iH 
 
 1-1 
 
 p^ 
 
 «) 
 
 
 :? 
 
 r-< 
 
 •Tj 
 
 «fl 
 
 •H 
 
 iH 
 
 CO 
 
 o 
 
 
 I-l 
 
 • 
 
 (U 
 
 Q 
 
 g 
 
 • 
 
 o 
 
 > 
 
 o 
 
 C4 
 
 O 
 
 o 
 
 CM 
 CM 
 O 
 
 u 
 
 o 
 
 •* 
 
 CM 
 
 CM 
 
 O 
 
 O 
 
 O 
 
 o 
 
 CM 
 
 O 
 
 II. 3-5 
 
C_5 
 -rC 
 
 X 
 
 X 
 
 CD 
 
 C3 
 
 u 
 
 (X. 
 (U 
 0) 
 
 o 
 
 01 
 
 u 
 
 •H 
 <M 
 
 li-l 
 
 o 
 
 c 
 o 
 
 M 
 
 > 
 
 •r-l 
 Q 
 
 Ph 
 
 4) 
 00 
 
 lU 
 
 O 
 
 o 
 
 > 
 
 Q 
 
 u 
 
 0) 
 00 
 
 o 
 
 u 
 CO 
 
 o 
 o 
 
 (U 
 00 
 
 <u 
 
 o 
 o 
 
 CO 
 
 >- 
 3: 
 ci. 
 -a: 
 
 C3 
 
 pq 
 
 CO 
 
 W 
 H 
 
 u 
 
 a 
 o 
 
 
 
 0) 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 u 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 c 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 <u 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 u 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 (U 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 in 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 c 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 o 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 o 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 >> 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 00 
 
 
 
 
 
 
 
 
 
 
 
 
 z 
 
 
 
 o 
 
 
 
 
 
 
 
 
 
 
 
 
 •-S 
 
 
 
 r-l 
 
 
 
 
 
 
 
 
 
 
 
 
 o 
 
 
 
 o 
 
 
 
 
 
 
 /-> 
 
 
 
 
 
 
 pj 
 
 
 
 p: 
 
 
 
 
 
 - 
 
 c 
 
 
 
 
 
 
 ^4 
 
 
 
 j3 
 
 
 
 
 
 "to 
 
 o 
 
 
 
 
 
 
 ^^-^ 
 
 
 
 o 
 
 
 
 
 
 T) 
 
 ■r-l 
 
 
 
 
 
 CJ 
 
 
 
 
 u 
 
 
 
 
 
 O 
 
 4J 
 
 
 
 
 
 pul 
 
 tu 
 
 o 
 
 
 H 
 
 
 
 
 
 o 
 
 •r-l 
 
 
 
 
 
 fi. 
 
 4-1 
 CIJ 
 
 00 
 
 CNJ 
 
 
 w 
 
 - 
 
 
 
 
 
 U 
 
 
 
 
 
 c 
 
 W 
 
 1 
 
 
 01 
 
 rfl 
 
 
 
 
 •n 
 
 o 
 
 
 
 
 
 o 
 
 4-1 
 
 vD 
 
 ■M 
 
 : 
 
 ^ 
 
 
 
 
 w 
 
 Cri 
 
 
 
 
 
 
 c- 
 
 "^^ 
 
 r- 
 
 CO 
 
 ^j 
 
 PL, 
 
 
 
 
 u 
 
 ^ 
 
 
 
 
 
 <u 
 
 (U 
 
 CN 
 
 
 4-1 
 
 cn x: 
 
 o 
 
 
 
 
 2 
 
 
 
 
 
 ^, 
 
 H 
 
 
 
 •H 
 
 1 - 
 S ^ 
 
 u 
 
 
 
 
 CJ 
 
 r-l 
 
 
 
 
 
 <u 
 
 o 
 
 •• 
 
 
 o 
 
 
 
 
 
 rj 
 
 •l^ 
 
 
 
 
 
 Vj 
 
 o 
 
 ^~^ 
 
 
 
 C 
 
 
 
 
 CM 
 
 o 
 
 
 
 
 
 0) 
 
 
 CO 
 
 
 1-1 
 
 ■^>' 
 
 o 
 
 
 
 
 3 
 
 
 
 
 
 
 CM 
 
 c 
 
 .— » -~^ 
 
 
 (U 
 
 o 
 
 •H 
 
 
 
 
 C 
 
 -d 
 
 
 
 t^ 
 
 
 c 
 
 •r-l 
 
 CO r-l 
 
 
 o 
 
 >-l -r-l 
 
 0) 4J 
 
 T-l 
 
 
 
 
 ;S 
 
 to 
 
 
 
 4-1 
 
 
 O 
 O 
 
 (U 
 
 4-1 
 
 ttj vO 
 
 
 t—i 
 
 r-l C 
 
 4-) 
 
 
 
 
 
 
 
 
 I— 1 
 
 
 
 o 
 
 ^ * 
 
 
 a 
 
 r-l n) 
 
 tu 
 
 
 
 
 c 
 
 1-4 
 
 
 
 •r^ 
 
 
 r-l 
 
 l-l 
 
 ^-^ c 
 
 
 0) 
 
 ■r-l 1— 1 
 
 PM 
 
 
 ^^ 
 
 
 •r-l 
 
 ta 
 
 «-N 
 
 
 .Q 
 
 
 tlj 
 
 P-I 
 
 o 
 
 
 Qi 
 
 ^ iJ 
 
 
 
 CO 
 
 
 
 (U 
 
 w 
 
 
 trJ 
 
 
 C 
 
 
 to •H 
 
 
 
 <; 
 
 l-l 
 
 
 fc. 
 
 
 (J 
 
 B 
 
 ^ 
 
 
 4-1 
 
 
 o 
 
 sz 
 
 CO 4-1 
 
 
 ^ 
 
 
 0) 
 
 
 g 
 
 
 P^ 
 
 Xi 
 
 s 
 
 
 o. 
 
 
 •r-l 
 
 to 
 
 C ctJ 
 
 
 t-1 
 
 O t-l 
 
 > 
 
 
 
 Pu, 
 
 CO 
 
 z 
 
 
 tu 
 
 
 4-1 
 
 •r-l 
 
 •M •M 
 
 
 •r^ 
 
 (X) 
 
 0) 
 
 
 ' V— • 
 
 
 
 •rl 
 
 v^ 
 
 
 o 
 
 
 (TJ 
 
 \^ 
 
 r-l O 
 
 
 •^ 
 
 X 3 
 
 hJ 
 
 
 to 
 
 
 
 CM 
 
 Pn 
 
 
 
 o 
 
 
 C 
 
 
 r-l O 
 
 
 
 -C C 
 
 
 
 tu 
 
 c 
 
 
 o 
 
 
 c 
 
 
 < 
 
 
 Vj 
 
 X. 
 
 •r-l to 
 
 
 13 
 
 ^ C 
 
 <: 
 
 
 G 
 
 
 
 
 
 r-l 
 
 o 
 
 
 
 
 OJ 
 
 •M 
 
 4-1 to 
 
 
 c 
 
 OJ < 
 
 PL| 
 
 
 •l-l 
 
 to 
 
 4-1 
 
 r-l 
 
 a 
 
 to 
 
 4-1 
 
 T3 
 
 
 4-1 
 
 •r-l 
 
 CO < 
 
 
 nJ 
 
 to 
 
 M 
 
 
 u 
 
 l-l 
 
 O- 
 
 <^ 
 
 c 
 
 ^ 
 
 D. 
 
 C 
 
 
 c 
 
 3 
 
 
 
 
 cd X. 
 
 
 
 tu 
 
 o 
 
 •r4 
 
 > 
 
 o 
 
 to 
 
 o 
 
 •H 
 
 CJ 
 
 /— ^ 
 
 M 
 
 
 . O 
 
 
 c 
 
 cC *-i 
 
 t-l 
 
 
 k: 
 
 o 
 
 (J 
 
 o 
 
 •H 
 
 Z) 
 
 CJ3 
 
 u 
 
 
 CO 
 
 
 T3 
 
 aU -rA 
 
 
 •H 
 
 r-^ 
 
 n1 
 
 
 
 
 O 
 
 u 
 
 4-1 
 
 a 
 
 
 u 
 
 CJ 
 
 u^ 
 
 4J 
 
 OJ 
 
 4-1 
 
 
 OJ 
 
 • r-l 
 
 3 
 
 
 .-1 
 
 • 
 
 to 
 
 o- 
 
 (Tl 
 
 •^ 
 
 • 
 
 t.n 
 
 P? 
 
 s 
 
 rrt 
 
 •H 
 
 • !> 
 
 
 ■u 
 
 I-) 
 
 T3 
 
 •-\ 
 
 rt 
 
 -) 
 
 p 
 
 D. 
 
 2 
 
 CJ 
 
 n 
 
 3 
 
 P^ 
 
 
 U-l 
 
 aa 4-1 
 
 
 O 
 
 iJ 
 
 • H 
 
 cn 
 
 c 
 
 
 C 
 
 < 
 
 S_^ 
 
 OJ 
 
 
 C 
 
 
 v-^ 
 
 T3 
 
 •rl 
 
 01 
 
 
 U 
 
 ~ nj 
 
 to 
 
 1 
 
 o 
 
 • 
 
 J3 
 
 
 
 02 
 
 • 
 
 J3 
 0^ 
 
 c:-l 
 
 
 OJ 
 
 4-1 
 
 - •r-l 
 
 
 CU 
 
 T3 
 
 0) 
 
 •r^ 
 
 CQ 
 
 ^ 
 
 u 
 
 C 
 
 
 □a 
 
 O 
 
 r-l 
 
 4-1 
 
 . l-l 
 
 r-l Q 
 
 
 
 .-I n 
 
 cd 
 
 4-» 
 
 
 
 o 
 
 ID 
 
 l-l 
 
 
 
 
 r-l 
 
 c: 
 
 o 
 
 r-l 
 
 
 x: 
 
 <1> o 
 
 
 v_x 
 
 3 
 
 " 
 
 •a 
 
 CM 
 
 CO 
 
 (U 
 
 •• 
 
 •o 
 
 C 
 
 01 
 
 OJ 
 
 u^ 
 
 01 C 
 
 
 to 
 
 U-l -r-l 
 
 T3 
 
 
 4J 
 
 Vj 
 
 ty 
 
 
 4J 
 
 p 
 
 l-l 
 
 0) 
 
 O 
 
 3 
 
 to 
 
 
 3 nj 
 
 
 ■r-* 
 
 I/l 4-1 
 
 OJ 
 
 ,H 
 
 ■r-l 
 
 j-i 
 
 -C 
 
 c 
 
 ^1 
 
 3 
 
 l-l 
 
 J= 
 
 •r-l 
 
 X) 
 
 OJ 
 
 to 
 
 -O CJ 
 
 
 ti- 
 
 1-1 m 
 
 to- 
 
 (U C 
 
 4-1 
 
 01 
 
 CO 
 
 o 
 
 til c 
 
 to 
 
 tu 
 
 to 
 
 4J 
 
 ■H 
 
 IJ 
 
 u 
 
 •r-l •M 
 
 
 
 (y »-) 
 
 <ti 
 
 T3 O 
 
 to 
 
 U. 
 
 ■r4 
 
 ■r-l 
 
 ^ o 
 
 C 
 
 U 
 
 •rl 
 
 rC 
 
 to 
 
 a. 
 
 y 
 
 CO iJ 
 
 
 U-l 
 
 D C 
 
 iV 
 
 4) -r-l 
 
 C 
 
 
 r-l 
 
 U 
 
 r-l -r^ 
 
 o 
 
 
 r-l 
 
 N 
 
 
 
 ^ 
 
 OJ 
 
 
 o 
 
 n) 0) 
 
 ^ 
 
 O iJ 
 
 M 
 
 • 
 
 ^ 
 
 •rl 
 
 3 4J 
 
 CJ 
 
 • 
 
 J3 
 
 •H 
 
 • 
 
 l-l 
 
 CJ 
 
 a k 
 
 
 
 fQ (0 
 
 o 
 
 o: -I 
 
 
 o 
 
 3 
 
 U 
 
 O rl 
 
 
 CJ 
 
 3 
 
 r— 4 
 
 Q 
 
 ^ 
 
 (fl 
 
 
 01 
 
 D 
 
 c 
 
 4-1 
 
 u 
 
 
 O. 
 
 0) 
 
 4J 
 
 CJ 
 
 
 a 
 
 •H 
 
 
 o. 
 
 ij 
 
 
 
 to 
 
 • U 
 
 M 
 
 • 0) 
 
 a. 
 
 • 
 
 C 
 
 (X, 
 
 • 0» 
 
 pj 
 
 • 
 
 c 
 
 4-1 
 
 • 
 
 <TJ 
 
 o 
 
 • • 
 
 
 ::j 
 
 (ii fu 
 
 
 Oi CU 
 
 U-. 
 
 oi 
 
 ;=> 
 
 : 
 
 >-> a^ 
 
 u< 
 
 Di 
 
 13 
 
 :=i 
 
 > 
 
 Oi 
 
 = 
 
 > "I 
 
 =1 
 
 VD 
 
 r^ 
 
 00 
 
 a\ 
 
 o 
 
 I-l 
 
 fM 
 
 (VJ 
 
 CM 
 
 «M 
 
 CM 
 
 CO 
 
 n 
 
 CO 
 
 o 
 
 o 
 
 o 
 
 o 
 
 o 
 
 o 
 
 o 
 
 CJ 
 
 o 
 
 o 
 
 CJ 
 
 CJ 
 
 o 
 
 o 
 
 II. 3-6 
 
1— 
 or: 
 
 X 
 
 C_3 
 
 
 
 o 
 
 O 
 
 .!«! 
 
 X 
 
 U 
 
 u 
 
 tS 
 
 cfl 
 
 P* 
 
 cu 
 
 V 
 
 d) 
 
 bO 
 
 t>0 
 
 0) 
 
 0) 
 
 p-l 
 
 r-t 
 
 1-1 
 
 r-^ 
 
 o 
 
 o 
 
 o 
 
 u 
 
 na 
 
 CI. 
 C9 
 
 CQ 
 Cl. 
 
 CO 
 
 w 
 
 H 
 u 
 :=) 
 
 Q 
 
 •a: 
 
 
 
 
 «W 
 
 
 
 
 
 o 
 
 
 
 
 
 c 
 
 
 
 
 
 o 
 
 
 
 
 
 1-1 
 
 
 
 
 
 u 
 
 
 
 
 
 ca 
 
 
 
 
 
 •H 
 
 
 
 
 
 U 
 
 
 
 
 
 O 
 
 
 
 
 
 CO 
 
 
 
 
 
 (0 
 
 
 
 
 
 <: 
 
 
 •o 
 
 ^-^ 
 
 
 g 
 
 
 4J 
 
 u 
 
 
 o 
 
 
 c 
 
 PM 
 
 
 •H 
 
 
 o 
 
 U-, 
 
 
 H 
 
 
 
 
 
 
 
 N-^ 
 
 <a 
 
 
 *** 
 
 -1 
 
 
 CVJ 
 
 u 
 
 
 
 
 r^ 
 
 « 
 
 
 <W 
 
 
 <^ 
 
 Vj 
 
 
 o 
 
 
 r-l 
 
 J-J 
 
 c 
 
 o 
 c 
 o 
 u 
 
 c 
 
 1-4 
 0) 
 
 o 
 
 y-\ 
 
 ^1 
 <u 
 u 
 a. 
 
 •H 
 
 -§. 
 1-1 
 
 ,c 
 
 
 
 4J 
 
 C/1 
 
 u 
 
 w 
 
 
 •H 
 
 h 
 
 (fl 
 
 4-1 
 
 
 3 
 
 fe: 
 
 
 Ul 
 
 
 
 ■V! 
 
 •H 
 
 
 TS 
 
 \^^ 
 
 (U 
 
 bO 
 
 
 Q) 
 
 
 J-l 
 
 o 
 
 
 •r-l 
 
 .-1 
 
 c 
 
 l-< 
 
 
 U-l 
 
 i-( 
 
 OJ 
 
 o 
 
 
 •H 
 
 OJ 
 
 CO 
 
 c 
 
 
 4J 
 
 :? 
 
 <u 
 
 jr 
 
 
 kJ 
 
 T) 
 
 u 
 
 o 
 
 
 O 
 
 •f-l 
 
 a. 
 
 0) 
 
 
 f^ 
 
 V5 
 
 jr 
 
 H 
 
 
 > 
 
 • 
 
 o 
 
 >> 
 
 
 TJ 
 
 « 
 
 <u 
 
 T) 
 
 
 C 
 
 
 (U 
 
 C 
 
 
 Q 
 
 • 
 
 tx 
 
 nj 
 
 > C/l U 
 
 C 
 cd 
 ex 
 J? 
 o 
 
 >-• 
 o 
 
 3 
 
 o 
 o 
 
 C 
 4J 
 
 n) 
 
 <4-l 
 
 o 
 
 60 
 
 m 
 
 e 
 
 s: 
 
 •H 
 
 60 
 
 >^ 
 
 3 
 
 U 
 
 O 
 
 Cfl 
 
 o 
 
 > 
 
 
 
 u-l 
 
 U-l 
 
 o 
 
 O 
 
 
 
 ?^ 
 
 
 60 
 
 iJ 
 
 o 
 
 o 
 
 r-J 
 
 4) 
 
 o 
 
 u-i 
 
 (1) 
 
 <J-| 
 W 
 
 s 
 
 X) 
 </3 
 
 U3 
 
 3 
 
 4) 
 
 • O 
 > H 
 
 4: 
 
 CO 
 
 O 
 
 •* 
 
 »r> 
 
 CO 
 
 en 
 
 o 
 
 O 
 
 (J 
 
 U 
 
 J 
 
 II. 3-7 
 

 co 
 
 
 H 
 
 
 O 
 
 
 w 
 
 
 PL. 
 
 
 w 
 
 >- 
 
 < 
 
 ac 
 
 CI 
 
 U- 
 
 M 
 
 «x 
 
 ?: 
 
 oc: 
 
 O 
 
 C3 
 
 z 
 
 C3 
 
 o 
 
 
 r 1 
 
 ,1, 1 
 
 tij 
 
 ca 
 
 
 
 . 1 
 
 "^ 
 
 <d 
 
 ej 
 
 r— 1 
 
 O- 
 
 Di 
 
 a) 
 CO 
 
 o 
 
 cu| 
 
 (X) 
 
 CU 
 
 
 r— ( 
 
 ^^ 
 
 a 
 
 o 
 
 C 
 
 
 n) 
 
 u 
 
 C/1 
 
 C 
 
 
 CJ 
 
 u 
 
 H 
 
 CO 
 
 a 
 
 (U 
 
 o 
 
 H 
 
 0) 
 
 tfl 
 
 > 
 
 
 (1) 
 
 14-1 
 
 a 
 
 o 
 
 1 
 
 M 
 
 
 0> 
 
 0) 
 
 •H 
 
 4J 
 
 u 
 
 cd 
 
 )-l 
 
 M 
 
 <i) 
 
 XJ 
 
 a 
 
 C 
 
 o 
 
 3) 
 
 u 
 
 
 
 
 
 C 
 
 
 O 
 
 o 
 
 u 
 
 r" 
 
 
 i-) 
 
 c 
 
 
 •r-l 
 
 -o 
 
 o 
 
 c 
 
 4-J 
 
 rt 
 
 o 
 
 
 l-l 
 
 U) 
 
 C-. 
 
 in 
 
 
 O 
 
 r* 
 
 u 
 
 1/) 
 
 O 
 
 •f-^ 
 
 u 
 
 U. 
 
 C; 
 
 u 
 (« 
 a* 
 
 « 
 
 00 
 
 o 
 o 
 
 Q 
 
 o 
 
 u 
 
 0) 
 
 c 
 o 
 
 o 
 
 o 
 o 
 
 Q 
 
 o 
 
 o 
 
 Q 
 
 u 
 
 td 
 
 60 
 
 o 
 o 
 
 o 
 o 
 
 Q 
 
 cit 
 
 oo 
 
 (U 
 
 o 
 o 
 
 
 
 
 C 
 
 
 
 
 
 
 
 
 
 
 •r-l 
 
 
 
 
 
 
 
 
 
 
 01 
 
 
 
 
 
 
 etf 
 
 
 
 
 iJ 
 
 
 
 
 
 
 •o 
 
 
 
 
 o 
 
 
 
 
 
 
 <4 
 
 
 
 
 u 
 
 
 
 
 
 
 c 
 
 
 
 
 (U 
 
 
 
 
 
 
 
 
 
 
 <u 
 
 
 
 
 
 
 
 
 
 u 
 
 E 
 o 
 
 
 
 
 
 
 
 
 
 c 
 
 w 
 
 
 
 
 
 
 ?; 
 
 
 
 0) 
 
 <u 
 
 
 
 
 
 
 CO 
 
 
 
 g 
 
 I-l 
 
 
 
 
 
 ^ 
 
 4-1 
 
 
 
 ij 
 
 o 
 
 
 
 
 
 • 
 
 4-1 
 
 
 
 M 
 
 JZ 
 
 
 
 
 
 -a 
 
 o 
 
 
 
 <u 
 
 D£ 
 
 
 
 
 
 u 
 
 
 
 
 > 
 
 
 — 
 
 
 
 
 ►4 
 
 *» 
 
 
 
 c 
 
 m 
 
 "c 
 
 
 
 
 
 0) 
 
 
 
 M 
 
 
 o 
 
 
 
 
 A 
 
 4J 
 
 
 
 
 in 
 
 •r-l 
 
 
 
 c 
 
 to 
 
 to 
 
 
 
 c 
 
 o 
 
 4J 
 
 
 
 
 
 XI 
 
 1-1 
 
 
 
 nJ 
 
 
 CX 
 
 
 
 •H 
 
 to 
 
 4J 
 
 
 
 
 c 
 
 E 
 
 
 
 4-1 
 
 c 
 
 c 
 
 
 
 w 
 
 o 
 
 3 
 
 
 
 to 
 
 to 
 
 <u 
 
 
 
 « 
 
 •l-l 
 
 XJ 
 
 
 
 
 u 
 
 (1) 
 
 o 
 
 
 
 
 5^ 
 
 O 
 
 b 
 
 
 
 n 
 
 u-l 
 
 o 
 
 
 
 x: 
 
 3 
 
 o 
 
 
 
 o 
 
 o 
 
 o 
 
 
 
 D. 
 
 XI 
 
 
 
 
 
 
 
 
 
 « 
 
 o 
 
 c 
 
 
 
 u 
 
 50 
 
 c 
 
 
 
 >j 
 
 u 
 
 to 
 
 
 
 Ph 
 
 c 
 
 •r-l 
 
 
 
 00 
 
 Pm 
 
 
 
 
 P^H 
 
 -,-1 
 
 <u 
 
 
 
 o 
 
 
 3 
 
 
 
 
 t-1 
 
 4-1 
 
 
 
 c 
 
 1— 1 
 
 3: 
 
 
 
 ■— 1 
 
 o 
 
 o 
 
 
 
 ca 
 
 to 
 
 
 
 
 to 
 
 OJ 
 
 1-1 
 
 
 
 0) 
 
 •H 
 
 Vi 
 
 
 
 •H 
 
 c 
 
 CM 
 
 
 
 o 
 
 U 
 
 o 
 
 
 
 O 
 
 •H 
 
 
 
 
 o 
 
 
 «M 
 
 
 
 
 ao 
 c 
 
 CO 
 
 
 
 ti 
 
 
 J= 
 
 
 
 E 
 
 w 
 
 •r^ 
 
 
 
 o 
 
 O 
 
 CO 
 
 •l-l 
 
 
 
 O 
 
 to 
 
 b 
 
 
 
 o 
 
 O 
 
 Pm 
 
 
 
 O 
 
 tu 
 
 c 
 
 
 
 o 
 
 
 
 
 
 
 CJ 
 
 o 
 
 
 
 c 
 
 (U 
 
 OJ 
 
 
 
 to 
 
 V4 
 
 
 
 
 (U 
 
 ^ 
 
 r-l 
 
 
 
 
 3 
 
 OJ 
 
 
 
 V4 
 
 OJ 
 
 o 
 
 
 
 14-1 
 
 o 
 
 o 
 
 
 
 0) 
 
 
 
 
 ,^^^ 
 
 o 
 
 (0 
 
 c 
 
 
 
 IM 
 
 oo 
 
 5 
 
 
 est 
 
 
 01 
 
 CJ 
 
 
 
 c 
 
 c 
 
 
 
 (Nl 
 
 4J 
 
 oi 
 
 S-i 
 
 
 ^-^ 
 
 o 
 
 •H 
 
 E 
 
 
 ->-y 
 
 c 
 
 •-• 
 
 CJ 
 
 
 t/1 
 
 u 
 
 c 
 
 o 
 
 
 CN 
 
 CJ 
 
 
 y-1 
 
 
 fa 
 
 
 Vj 
 
 l-l 
 
 
 f^ 
 
 B 
 
 r-J 
 
 c 
 
 >^ 
 
 iJ 
 
 0) 
 
 U-l 
 
 
 
 to 
 
 to 
 
 ^o 
 
 ^ 
 
 
 w 
 
 CJ 
 
 
 
 *• 
 
 w'i 
 
 
 ^-^ 
 
 J-l 
 
 v_x 
 
 
 c 
 
 Z) 
 
 /■'^ 
 
 u 
 
 CJ 
 
 4-1 
 
 
 05 
 
 
 T3 
 
 o 
 
 ■o 
 
 <; 
 
 <u 
 
 CO 
 
 CJ 
 
 o 
 
 3 
 
 0) 
 
 c; 
 
 u 
 
 JO 
 
 Q 
 
 tJ 
 
 CO 
 
 
 4J 
 
 •a 
 
 l-l 
 
 i-i 
 
 
 ^ 
 
 U. 
 
 Ul 
 
 < 
 
 " 
 
 
 c 
 
 )-l 
 
 c 
 
 i-) 
 
 
 v«^ 
 
 'r-l 
 
 
 u 
 
 c 
 
 M 
 
 0) 
 
 <i) 
 
 c 
 
 0) 
 
 
 w 
 
 o 
 
 CJ 
 
 o 
 
 
 i 
 
 (0 
 
 CJ 
 
 u 
 
 -o 
 
 CJ 
 
 •H 
 
 XI 
 
 •r-l 
 
 c 
 
 ■y 
 
 B 
 
 to 
 
 u 
 
 Di 
 
 G 
 
 r-l 
 
 4-1 
 
 (0 
 
 t^ 
 
 ^ 
 
 o 
 
 p 
 
 to 
 
 
 o 
 
 o 
 
 to 
 
 
 
 a 
 
 u 
 
 u 
 
 X) 
 
 r-l 
 
 c 
 
 x 
 
 ul 
 
 U) 
 
 
 
 to 
 
 c 
 
 •o 
 
 to 
 
 o 
 
 
 c 
 
 to 
 
 • 
 
 Ui 
 
 LJ 
 
 0) 
 
 o 
 
 i-l 
 
 u 
 
 • 
 
 <u 
 
 
 < 
 
 y 
 
 00 
 
 (J 
 
 o 
 
 V 
 
 w 
 
 Q 
 
 (.0 
 
 u 
 
 
 Cu 
 
 
 c 
 
 
 ■o 
 
 
 
 aj 
 
 0- 
 
 • 
 
 nj 
 
 -< 
 
 o 
 
 • 
 
 o 
 
 C 
 
 • 
 
 V-l 
 
 (J-, 
 
 o 
 
 Cm 
 
 - 
 
 C_) 
 
 •-1 
 
 u. 
 
 < 
 
 u 
 
 Oi 
 
 o 
 
 o 
 
 ON 
 
 a. 
 
 00 
 
 CU 
 
 o 
 
 XI 
 to 
 
 4-1 
 
 w 
 > 
 
 X) 
 
 c 
 
 CO 
 
 CJ 
 
 H 
 
 o 
 
 l-l 
 
 c 
 
 r-l 
 
 o 
 
 to 
 
 o 
 
 o 
 
 
 •r4 
 
 c 
 
 E 
 
 ■r-l 
 
 o 
 
 01 
 
 c 
 
 4-1 
 
 o 
 
 o 
 
 o 
 
 u 
 
 w 
 
 Oi 
 
 
 
 -a 
 
 r- 
 
 c 
 
 to 
 
 to 
 
 .r4 
 
 
 t- 
 
 > 
 
 
 r-l 
 
 U-J 
 
 r-l 
 
 O 
 
 c^ 
 
 
 — 
 
 CO 
 
 o 
 
 CJ 
 
 •rl 
 
 1-1 
 
 i-J 
 
 3 
 
 ■H 
 
 4-1 
 
 M 
 
 X 
 
 4J 
 
 •H 
 
 -) 
 
 ^ 
 
 ^i 
 
 C. 
 
 r-l 
 
 o 
 
 rJ 
 
 r-l 
 
 1- 
 
 OJ 
 
 — I 
 
 > 
 
 i-l 
 
 u 
 
 D 
 
 Q 
 
 o 
 
 o 
 
 o 
 to 
 u 
 u 
 C 
 
 o 
 u 
 
 o 
 o 
 
 Q 
 
 II. 4-1 
 
OS 
 
 est 
 
 oa 
 
 CO 
 CO 
 
 H 
 
 o 
 
 w 
 
 to 
 
 H 
 
 o 
 o 
 
 w 
 
 H 
 
 H 
 
 CO 
 
 t3 
 
 1 — 
 
 
 
 
 
 
 
 
 
 
 
 
 
 C_l 
 
 
 
 
 
 
 
 
 
 
 
 
 
 «^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 X 
 
 
 X 
 
 
 
 
 
 X 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 
 
 
 m 
 
 
 
 
 
 
 
 
 
 
 
 
 
 «ii 
 
 
 
 
 
 
 
 
 
 
 
 
 
 T~ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 _j 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ij. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 C3 
 
 
 
 
 
 
 
 
 
 
 
 
 
 cc 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 «•= 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 <u 
 
 
 
 
 
 
 
 
 
 
 
 
 
 u 
 
 
 
 
 
 
 
 
 
 
 
 
 
 •iH 
 
 
 
 
 
 
 
 ^ 
 
 
 M 
 
 
 AJ 
 
 
 >4-l 
 
 X 
 
 
 X 
 
 
 
 
 u 
 
 
 U 
 
 
 U 
 
 
 •4-1 
 
 u 
 
 
 u 
 
 
 2r 
 
 
 ttf 
 
 
 ti 
 
 
 (tf 
 
 
 o 
 
 CO 
 
 
 cd 
 
 
 o 
 
 
 (X, 
 
 
 t^ 
 
 
 PU 
 
 
 
 (U 
 
 
 P^ 
 
 
 1— 
 
 
 
 
 
 
 
 
 c 
 
 
 
 
 
 •cC 
 
 
 <u 
 
 
 (U 
 
 
 <u 
 
 
 o 
 
 (U 
 
 
 OJ 
 
 
 CJ 
 
 
 60 
 
 
 to 
 
 
 60 
 
 
 •H 
 
 CO 
 
 
 60 
 
 
 CD 
 
 
 01 
 
 
 a) 
 
 
 <U 
 
 
 CO 
 
 (U 
 
 
 0) 
 
 
 
 
 I-l 
 I-l 
 o 
 
 
 1-1 
 I-l 
 o 
 
 
 T-l 
 
 f-l 
 
 o 
 
 
 1-1 
 > 
 
 l-^ 
 1-1 
 o 
 
 
 t-l 
 r-l 
 
 O 
 
 
 
 
 u 
 
 
 u 
 
 
 o 
 
 
 Q 
 
 u 
 
 
 O 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 0) 
 
 <u 
 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 4J 
 
 4-1 
 
 
 
 
 
 d) 
 
 
 
 
 
 
 
 
 
 
 nJ 
 
 ri 
 
 
 
 
 
 1-1 
 
 
 
 
 
 
 
 
 
 
 4J 
 
 u 
 
 
 
 
 
 I-l 
 
 
 
 
 
 
 
 
 
 
 w 
 
 4.J 
 
 
 
 
 
 4-1 
 
 c 
 
 
 
 
 
 
 
 
 
 
 -a 
 
 OJ 
 
 
 
 
 
 3 
 
 
 
 
 <u 
 
 
 
 
 
 PS 
 
 
 (U 
 
 o 
 
 
 
 
 
 O 
 
 
 
 
 Xi 
 
 
 
 
 0-) 
 
 cu 
 
 
 4-1 
 
 c 
 
 
 
 
 
 o 
 
 
 
 
 4J 
 
 
 
 
 00 
 
 p; 
 
 
 c 
 
 o 
 o 
 
 
 
 
 
 6£ 
 
 
 
 
 c 
 
 
 
 
 1 
 
 r -^ 
 
 
 ^ 
 
 
 
 
 
 
 C 
 
 
 
 
 o 
 
 
 
 
 r>. 
 
 
 
 
 c 
 
 
 
 
 
 •r-l 
 
 
 
 
 o 
 
 
 
 
 
 ITi 
 
 
 (U 
 
 •r^ 
 
 
 
 
 
 a 
 
 
 
 
 t> 
 
 
 
 
 •• 
 
 oa 
 
 
 j:: 
 
 <u 
 
 
 
 
 
 o 
 
 
 
 
 
 
 
 
 <u 
 
 
 z:^ 
 
 
 4-) 
 
 4-1 
 
 
 
 
 
 M 
 
 
 
 
 ca 
 
 
 
 
 4J 
 
 
 ex. 
 
 o 
 o 
 
 TO 
 C 
 « 
 
 cn 
 
 <D 
 •H 
 
 d) 
 rn 
 
 o 
 
 u 
 fu 
 
 X. 
 w 
 
 •H 
 "4-1 
 
 o 
 
 
 
 
 
 01 
 
 > 
 
 0) 
 
 o 
 c 
 
 1-1 
 
 0) 
 
 4-1 
 
 n 
 
 4J 
 
 
 
 Oi 
 
 Cl4 
 
 -01 
 
 cd 
 u 
 
 Ui 
 
 c 
 o 
 
 1-1 
 
 4J 
 
 o 
 to 
 1-1 
 
 
 
 ■u 
 u 
 
 3 
 
 s 
 
 •H 
 4J 
 (0 
 C 
 H 
 
 m 
 
 (U 
 
 1-1 
 u 
 
 
 
 
 fij 
 
 c 
 
 
 
 
 
 c 
 
 
 
 to 
 
 
 '4-1 
 
 
 
 
 s^ 
 
 
 
 CO 
 
 
 rl 
 
 
 
 
 
 CJ> 0) 
 
 
 
 0! 
 
 C 
 
 X 
 
 
 o 
 
 ^ 
 
 'Si 
 
 
 
 ».D 
 
 t-( 
 
 e 
 
 
 
 
 
 vD O 
 
 
 
 •H 
 
 o 
 
 w 
 
 
 r^ 
 
 to 
 
 •H 
 
 
 
 ON 
 
 « 
 
 a 
 
 
 
 ^ 
 
 
 
 as C 
 
 
 
 Vj 
 
 
 
 
 <JN 
 
 
 ft, 
 
 
 
 i-l 
 
 •H 
 
 o 
 
 
 
 
 
 I-l o 
 
 
 
 4-1 
 
 vD 
 
 >4-( 
 
 
 I-l 
 
 t4-l 
 
 
 
 
 
 O 
 
 I-l 
 
 
 c 
 
 AJ 
 
 
 
 o 
 
 
 
 c 
 
 1—1 
 
 o 
 
 
 
 o 
 
 c 
 
 
 
 
 J-i 
 
 o 
 
 
 3 
 
 to 
 
 
 
 
 
 
 3 
 
 =S= 
 
 
 
 
 
 CO 
 
 
 ^ 
 
 
 0) 
 
 > 
 
 
 •rH 
 
 3 
 
 
 
 c 
 
 
 
 O 
 
 
 u 
 
 
 
 X. 
 
 0) 
 
 
 t?-. 
 
 
 
 e) 
 
 
 W 
 
 •c 
 
 
 
 •H 
 
 
 
 O 
 
 4-1 
 
 <i) 
 
 
 
 *J 
 
 ,Q 
 
 
 cf 
 i~~> 
 
 
 p 
 
 Q 
 
 
 O 
 
 c 
 
 
 
 0) 
 
 
 
 
 c 
 
 U3 
 
 
 
 s 
 
 ^ 
 
 
 
 o 
 
 
 
 a, 
 
 M 
 
 
 
 4-1 
 
 
 
 w 
 
 <u 
 
 g 
 
 
 
 o 
 
 1-1 
 
 
 
 
 u 
 
 r-l 
 
 
 ^ 
 
 5 
 
 
 
 o 
 u 
 
 
 
 c 
 1-1 
 
 s 
 
 0) 
 
 3 
 
 
 
 u 
 o 
 
 U 
 
 CO 
 
 
 =3 
 
 
 iw 
 
 •H 
 
 
 m 
 
 0) 
 
 
 
 Cn 
 
 
 
 D- 
 
 4J 
 
 
 
 
 
 u 
 
 
 «: 
 
 
 O 
 
 O 
 
 
 
 Z 
 
 
 
 X. 
 
 
 
 O 
 I— 1 
 
 CO 
 iJ 
 
 ■3 
 C 
 
 > 
 
 
 0) 
 
 •o 
 
 
 
 
 3 
 
 dJ 
 
 
 to 
 
 CS 
 
 
 
 CO 
 
 
 
 0) 
 
 CO 
 
 to 
 
 I-l 
 
 
 c 
 
 
 
 
 CtJ 
 
 g 
 
 
 1-4 
 
 
 
 
 1-1 
 
 
 
 > 
 
 
 
 r-l 
 
 
 
 CO 
 
 
 
 
 <U 
 
 d 
 
 
 o 
 
 H 
 
 
 
 P4 
 
 
 
 (U 
 
 y^ 
 
 O 
 
 to 
 
 
 
 .'-N 
 
 
 
 
 U 
 
 o 
 
 
 C 
 
 O 
 
 
 
 
 ^N 
 
 
 Q 
 
 *Z 
 
 ■H 
 
 u 
 
 
 •• 
 
 CO "4-1 
 
 
 
 
 3 
 
 u 
 
 
 >j 
 
 U-l 
 
 
 
 14-1 
 
 • 
 
 
 
 pn 
 
 4-1 
 
 •H 
 
 
 0) 
 
 b '-' 
 
 
 
 
 PQ 
 
 
 
 <u 
 
 
 
 
 o 
 
 CJ 
 
 
 C 
 
 ^— / 
 
 to 
 
 e 
 
 
 4-1 
 
 & ^ 
 
 
 
 
 
 <y 
 
 
 > 
 
 I-l 
 
 ^-N 
 
 
 
 C 
 
 
 •H 
 
 
 Di 
 
 o 
 
 
 to 
 
 2 o 
 
 
 
 
 O 
 
 ^ 
 
 
 o 
 
 0) 
 
 
 
 f^ 
 
 M 
 
 Q 
 
 
 a. 
 
 
 c 
 
 4-1 
 
 I-l 
 
 v-^ 
 
 
 
 
 x: 
 
 4-1 
 
 
 o 
 
 X) 
 
 o 
 
 
 4-1 
 
 
 M 
 
 CJ 
 
 3 
 
 4-1 
 
 o 
 
 o- 
 
 4-1 
 
 1-1 
 
 
 
 
 4-1 
 
 
 
 
 o 
 
 C 
 
 
 1-1 
 
 •V 
 
 <; 
 
 Pul 
 
 O 
 
 c 
 
 CJ 
 
 •iH 
 
 c 
 
 • CO 
 
 
 
 
 
 c 
 
 ^-\ 
 
 •u 
 
 S 
 
 M 
 
 
 I-l 
 
 >^ 
 
 
 Pm 
 
 Vj 
 
 0) 
 
 u 
 
 ^ 
 
 (U 
 
 U 3 
 
 
 
 
 U-l 
 
 •H 
 
 W 
 
 03 
 
 
 
 
 1-1 
 
 60 
 
 o 
 
 
 a 
 
 > 
 
 
 o 
 
 o 
 
 >-3 C 
 
 
 
 
 o 
 
 
 b 
 ^ 
 
 
 u 
 
 A 
 
 
 ,43 
 
 O 
 
 u 
 
 1-1 
 
 
 i~! 
 
 o 
 
 /-N to 
 
 c 
 
 C 
 
 
 
 
 
 U) 
 
 ■o 
 
 •r-l 
 
 O 
 
 
 iH 
 
 1-1 
 
 
 o 
 
 e^ 
 
 o 
 
 Cu 
 
 CO 3 
 
 o 
 
 - < 
 
 
 LU 
 
 
 4-1 
 
 g 
 
 (U 
 
 g 
 
 o 
 
 
 CO 
 
 o 
 
 4-1 
 
 >4-l 
 
 CO 
 
 [^ 
 
 Pm C 
 
 o 
 
 I-l 
 
 
 pi 
 
 
 U) 
 
 v—' 
 
 u 
 
 o 
 
 0) 
 
 a. 
 
 n] 
 
 c 
 
 U 
 
 
 o 
 
 
 
 1 ^ 
 
 
 U3 fO 
 
 
 
 0) 
 
 •H 
 
 
 C 
 
 c 
 
 3 
 
 a. 
 
 (U 
 
 n] 
 
 o 
 
 I— 1 
 
 to 
 
 0) 
 
 IJ-I 
 
 c 
 
 O CM 
 
 
 »— 
 
 
 u 
 
 P^ 
 
 ^J 
 
 cu 
 
 
 
 CO 
 
 
 f^ 
 
 0) 
 
 a 
 
 03 
 
 •H 
 
 x; 
 
 o 
 
 s— ' 
 
 ■r-l 
 
 « to 
 
 
 
 
 4) 
 
 
 <u 
 
 CO 
 
 o 
 
 v_x 
 
 o 
 
 
 u 
 
 <u 
 
 •H 
 
 > 
 
 4-1 
 
 
 -o 
 
 OJ 
 
 
 
 
 4J 
 
 1-4 
 
 T3 
 
 <u 
 
 w 
 
 
 Csl 
 
 1-1 
 
 o 
 
 Di 
 
 4-1 
 
 •o 
 
 
 c 
 
 to 0) 
 
 4-1 
 
 tc 
 
 
 
 
 C 
 
 rj 
 
 >, 
 
 )J 
 
 
 4-1 
 
 
 CC 
 
 
 
 c 
 
 < 
 
 U-l 
 
 o 
 
 r-l J: 
 
 o 
 
 • c 
 
 
 
 
 M 
 
 •M 
 
 C 
 
 D. 
 
 c 
 
 X. 
 
 M 
 
 1-1 
 
 I-l 
 
 4J 
 
 CJ 
 
 
 O 
 
 ■H 
 
 r-l CO 
 
 u 
 
 S 1-1 
 
 
 
 
 
 )-i 
 
 00 
 
 
 < 
 
 CO 
 
 u 
 
 O 
 
 <TJ 
 
 O 
 
 4-1 
 
 c 
 
 
 4-1 
 
 •I-l -H 
 
 a. 
 
 •xp 
 
 
 
 
 C 
 
 4-1 
 
 
 J= 
 
 
 •H 
 
 <u 
 
 iJ 
 
 I-l 
 
 n) 
 
 o 
 
 •r4 
 
 4J 
 
 CJ 
 
 ^4 1-1 
 
 
 <u o 
 
 
 
 
 o 
 
 m 
 
 • 
 
 o 
 
 
 )-l 
 
 I-l 
 
 OJ 
 
 (1) 
 
 ^1 
 
 Pu 
 
 0) 
 
 u 
 
 3 
 
 3 XI 
 
 J3 
 
 CO 01 
 
 
 
 
 (= 
 
 
 o 
 
 0) 
 
 • • 
 
 2 
 
 ^ 
 
 p 
 
 c 
 
 4-1 
 
 
 4-1 
 
 o 
 
 -o 
 
 O 3 
 
 to 
 
 vj tJ 
 
 
 
 
 g 
 
 -0 
 
 
 01 
 
 u 
 
 
 o 
 
 g 
 
 0) 
 
 c 
 
 ty 
 
 O 
 
 u_i o 
 
 a. 
 
 1-1 
 
 o o 
 
 
 
 o 
 
 C 
 
 » 
 
 Q. 
 
 Pu 
 
 o 
 
 o 
 
 CJ 
 
 o 
 
 -C 
 
 IJ 
 
 u-l »-. 
 
 • K 
 
 U^ 
 
 OJ Vj 
 
 
 
 u 
 
 M| Q 
 
 CO 
 
 Cn CO 
 
 eQ 
 
 CJ 
 
 
 u 
 
 H 
 
 cu 
 
 U &4 
 
 1-1 ^ 
 
 - 
 
 O (^ 
 
 CD 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ac: 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 vO 
 
 
 r- 
 
 
 00 
 
 
 <T\ 
 
 o 
 
 
 I-l 
 
 
 xC 
 
 
 o 
 
 
 o 
 
 
 o 
 
 
 O 
 
 I-l 
 
 
 1—1 
 
 
 CJ 
 
 
 o 
 Q 
 
 
 o 
 
 Q 
 
 
 o 
 
 Q 
 
 
 O 
 
 Q 
 
 o 
 
 Q 
 
 
 o 
 
 Q 
 
 
 II. 4-2 
 
CO 
 
 H 
 CJ 
 
 ■-2 
 
 o.. 
 
 X 
 
 Q 
 
 en 
 
 ■?^ 
 
 C3 
 
 o 
 
 C3 
 
 c -) 
 
 _-J 
 
 w 
 
 CO 
 
 *^ 
 
 CO 
 
 ^ 
 
 
 
 <_:< 
 
 cr; 
 
 Ou 
 
 H 
 
 Ll_ 
 
 r/1 
 
 
 
 
 
 
 a 
 
 
 ^ 
 
 »— 
 
 
 
 
 
 
 
 
 ■-MMM 
 
 
 
 
 ' ' 
 
 
 
 
 
 CJ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 «« 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 X 
 
 
 
 
 X 
 
 
 
 
 
 X 
 
 
 
 X 
 
 
 
 oo 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 5 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Ul. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CD 
 
 ck: 
 
 
 
 
 
 
 X 
 
 
 
 
 
 X 
 
 
 
 
 
 
 CJ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ac 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 a) 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 o 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 •H 
 
 
 ^ 
 
 
 Ji4 
 
 
 
 J< 
 
 
 
 
 
 u 
 
 
 
 
 
 U-l 
 
 
 Vi 
 
 
 U 
 
 
 
 ^ 
 
 
 
 
 
 
 
 C3 
 
 
 o 
 
 
 03 
 
 PL) 
 
 
 
 
 
 
 
 
 
 
 (0 
 
 
 
 ►— 
 
 
 c 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 «t 
 
 
 o 
 
 
 4) 
 
 
 0) 
 
 
 
 lU 
 
 
 u 
 
 
 
 60 
 
 
 
 CJ 
 
 
 •H 
 
 
 &0 
 
 
 60 
 
 
 
 60 
 
 
 •H 
 
 
 
 
 
 C3 
 
 
 (0 
 
 
 0) 
 
 
 Q> 
 
 
 
 <U 
 
 
 iJ 
 
 
 
 CJ 
 
 
 
 ^ 
 
 
 > 
 
 •r-l 
 
 a 
 
 
 r-l 
 1-1 
 O 
 O 
 
 
 r-l 
 r-l 
 O 
 
 CJ 
 
 
 
 r-l 
 r-l 
 O 
 CJ 
 
 
 u 
 at 
 
 (U 
 CO 
 
 
 
 r-l 
 
 r-l 
 
 o 
 <J 
 
 
 
 
 
 
 
 o 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 r- 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ON 
 
 
 
 (U 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 r-l 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 c 
 
 
 
 
 
 
 
 1 
 
 
 
 
 
 
 
 
 
 
 
 
 n] 
 
 
 
 1 
 
 u 
 
 
 
 
 X 
 
 
 
 
 
 
 
 
 ' 
 
 
 
 
 
 u 
 
 
 
 
 o 
 
 
 
 
 w 
 
 
 
 
 
 
 
 
 
 
 
 
 
 to 
 
 
 
 M 
 
 14-1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 1-1 
 
 
 
 
 
 
 
 
 4-1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 J<5 
 
 
 
 i-> 
 
 4-1 
 
 
 
 
 e 
 
 
 
 
 
 
 
 
 
 a= 
 
 
 
 
 CO 
 
 
 
 U 
 
 CJ 
 
 
 
 
 <u 
 
 
 
 
 
 
 
 
 
 C3 
 
 
 
 
 04 
 
 
 
 tC 
 
 ^ 
 
 
 
 
 > 
 
 
 
 
 
 
 
 
 
 CJ 
 
 
 
 
 •rl 
 
 
 
 p-l 
 
 u 
 
 
 
 
 r-l 
 O 
 CO 
 
 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 —J 
 
 Cl3 
 
 
 
 
 u 
 
 3 
 
 
 
 u 
 
 CJ 
 
 u 
 
 4J 
 
 
 
 
 U-4 
 
 o 
 
 
 
 
 
 
 r-l 
 
 u 
 
 
 
 
 
 
 
 4J 
 
 
 
 p 
 
 D 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 o 
 
 
 
 X) 
 
 tL, 
 
 
 
 
 CO 
 
 
 • 
 
 
 
 
 o 
 
 
 
 
 
 
 
 n) 
 
 
 
 c 
 
 
 
 
 
 u 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 iU 
 
 
 
 M 
 
 CJ 
 
 
 
 
 C 
 
 
 
 r-l 
 
 
 
 
 
 
 
 
 
 c 
 
 
 
 "O 
 
 H 
 
 
 
 
 3 
 
 
 
 CO 
 
 ■J 
 
 
 
 >% 
 
 
 
 
 iJ 
 
 
 
 ;g 
 
 
 
 o 
 o 
 
 1 
 
 
 
 
 r-l 
 
 
 
 s 
 
 
 
 CO' ■■ 
 en 
 
 •H 
 
 
 
 
 c 
 
 
 
 
 
 
 l^ 
 
 
 
 
 
 U-l 
 
 
 
 x: 
 
 
 
 a 
 
 CO 
 
 
 
 
 o 
 
 
 
 0) 
 
 
 
 
 M 
 
 
 
 
 U 
 
 
 
 to 
 
 
 
 c 
 
 Vl 
 
 
 
 
 o 
 
 
 
 4-1 
 
 
 
 • 
 
 M 
 
 
 
 
 
 
 
 •r4 
 
 
 
 o 
 
 > 
 
 •r4 
 
 
 
 
 v.^ 
 
 
 
 0) 
 
 
 
 CO 
 
 iJ 
 
 
 
 
 CO 
 
 
 
 P^ 
 
 
 
 i- 
 
 
 
 
 o 
 
 
 
 4J 
 
 
 
 « 
 
 ^1 
 
 
 
 
 0) 
 
 
 r-4 
 
 X! 
 
 
 
 
 
 
 
 
 r^ 
 
 
 
 c 
 
 
 
 3 
 
 CO 
 
 
 
 
 u 
 
 
 r~. 
 
 C 
 
 
 
 < 
 
 
 
 
 a^ 
 
 
 
 w 
 
 
 
 
 P-i 
 
 
 
 
 o 
 
 
 o> 
 
 a 
 
 
 
 ft. 
 
 ^^-^ 
 
 
 
 
 1— ( 
 
 0) 
 
 
 .-H 
 
 o 
 
 
 
 0) 
 
 
 
 
 
 n 
 
 
 r-l 
 
 
 
 
 h- 1 
 
 u 
 
 
 
 
 
 i_) 
 
 
 •■-( 
 
 c 
 
 
 
 x; 
 
 
 
 
 
 a. 
 
 
 
 CJ 
 
 
 
 
 
 
 
 
 rt 
 
 
 o 
 
 o 
 
 
 
 U 
 
 • 
 
 
 ^^ 
 
 
 
 
 
 ij 
 
 
 
 1 
 
 
 
 _ 
 
 
 (J 
 
 
 c 
 
 CJ 
 
 
 
 
 0) 
 
 
 • 
 
 
 g 
 
 4-1 
 
 
 
 to 
 
 
 
 
 4-) 
 
 
 
 oo 
 
 
 u 
 
 
 D 
 
 
 
 
 c 
 
 iJ 
 
 
 3 
 
 
 o 
 
 c 
 
 
 
 1-1 
 
 
 1— 1 
 
 0) 
 
 c 
 
 
 
 c^ 
 
 
 C 
 
 
 O 
 
 c 
 
 
 
 •r^ 
 
 C 
 
 
 
 
 Vj 
 
 to 
 
 
 
 4-1 
 
 
 r^ 
 
 .^ 
 
 OJ 
 
 
 
 CD 
 
 
 OJ 
 
 
 u 
 
 •r^ 
 
 
 
 
 CJ 
 
 
 r—i 
 
 
 M-l 
 
 i-H 
 
 
 
 c 
 
 
 * — » 
 
 CC 
 
 CJ 
 
 
 
 rn 
 
 
 o 
 
 
 
 (U 
 
 
 
 en 
 
 •r-l 
 
 
 ,—1 
 
 
 
 p-l 
 
 
 
 CJ 
 
 
 o 
 
 CJ 
 
 
 
 
 H- 
 
 
 c 
 
 
 dJ 
 
 4J 
 
 
 
 4-J 
 
 X) 
 
 
 OJ 
 
 
 > 
 
 
 
 
 o 
 
 
 ,—1 
 
 
 
 
 
 =J 
 
 
 o 
 
 
 o 
 
 o 
 
 
 
 c 
 
 OJ 
 
 
 C 
 
 
 1-1 
 
 u 
 
 
 
 c 
 
 
 — -^ 
 
 to 
 
 (J) 
 
 
 
 «C 
 
 
 u 
 
 c 
 
 
 c 
 
 
 
 
 •r-l 
 
 XI 
 
 
 
 1-1 
 
 o 
 o 
 
 
 OJ 
 
 > 
 
 o 
 
 4-1 
 
 CO 
 l-i 
 
 
 
 o 
 u 
 
 
 00 
 XI 
 
 CO 
 
 0) 
 
 1-1 
 
 to 
 
 
 
 
 
 •H 
 
 
 o 
 
 j: 
 
 
 
 QJ 
 
 M 
 
 
 V-r* 
 
 
 o 
 
 4-) 
 
 
 
 c 
 
 
 CJ 
 
 Cu 
 
 CO 
 
 c 
 
 
 
 
 
 QJ 
 
 
 w 
 
 w 
 
 
 
 U 
 
 
 
 
 
 CJ 
 
 C 
 
 
 
 •r-l 
 
 
 D 
 
 
 
 
 
 
 u 
 
 
 
 •rH 
 
 
 
 0£ 
 
 c 
 
 
 ,—1 
 
 
 Di 
 
 a 
 
 
 
 OJ 
 
 
 CO 
 
 ■o 
 
 o 
 
 
 
 
 
 o 
 
 
 <u 
 
 f^ 
 
 
 
 C 
 
 •rJ 
 
 
 ■ — 1 
 
 
 
 o 
 
 /-S 
 
 
 4-1 
 
 
 to 
 
 c 
 
 G 
 
 
 
 
 
 IJ 
 
 
 C 
 
 
 
 
 M 
 
 CJ 
 
 
 CO 
 
 
 4-1 
 
 c 
 
 § 
 
 
 o 
 
 
 •r4 
 
 CO 
 
 1— 1 
 
 
 
 
 
 Ol, 
 
 
 ■H 
 
 u 
 
 
 
 
 i-J 
 
 
 u 
 
 
 o 
 
 o 
 
 
 u 
 
 
 » 
 
 
 <: 
 
 
 
 
 
 
 
 n 
 
 o 
 
 
 ij 
 
 c 
 
 o 
 
 
 
 
 D 
 
 u 
 
 < 
 
 
 cu 
 
 
 vD 
 
 <^ 
 
 
 
 
 
 
 — 
 
 
 to 
 
 14-I 
 
 
 CU 
 
 •rH 
 
 iJ 
 
 
 • 
 
 
 X) 
 
 
 
 
 
 
 O 
 
 cu 
 
 • 
 
 
 
 
 
 </) 
 
 .«~« 
 
 s; 
 
 
 
 •^^ 
 
 CJ 
 
 Pi 
 
 
 rJ 
 
 
 O 
 
 c 
 
 CO 
 
 
 x: 
 
 /— V 
 
 vO 
 
 >-* 
 
 >^ 
 
 
 
 
 
 •n^ 
 
 H 
 
 
 r-l 
 
 
 Vj 
 
 iJ 
 
 
 to 
 
 
 
 iJ 
 
 •r^ 
 
 to 
 
 
 CO 
 
 CO 
 
 •• 
 
 
 
 
 
 
 
 u. 
 
 M 
 
 o 
 
 <0 
 
 ,— ^ 
 
 u 
 
 o 
 
 Jj 
 
 4-) 
 
 > 
 
 
 Ch 
 
 1) 
 
 X 
 
 
 •r-l 
 
 h 
 
 00 
 
 t4-l 
 
 " 
 
 
 
 
 
 
 ^_l 
 
 «j 
 
 ■r-i 
 
 lO 
 
 Oj 
 
 ;_i 
 
 o 
 
 U 
 
 C^ 
 
 
 1 
 
 i-J 
 
 Ol 
 
 
 MLi 
 
 £; 
 
 tr. 
 
 o 
 
 /-^ 
 
 
 
 
 
 U-l 
 
 v_^ 
 
 
 iJ 
 
 
 
 
 P-. 
 
 U_i 
 
 CJ 
 
 
 
 > 
 
 O 
 
 H 
 
 
 
 2C 
 
 u~> 
 
 
 CO 
 
 s 
 
 
 LU 
 
 
 o 
 
 
 4J 
 
 c 
 
 
 
 
 •r-l 
 
 * 
 
 4-) 
 
 pa 
 
 Vj 
 
 v-^ 
 
 
 y-i 
 
 N— ' 
 
 •» 
 
 c 
 
 fc 
 
 
 
 
 
 0) 
 
 U 
 
 1) 
 
 1 
 
 U~I 
 
 t-l 
 
 X! 
 
 U) 
 
 IJ 
 
 
 pL. 
 
 
 V-l 
 
 O 
 
 
 en 
 
 o 
 
 f>- 
 
 3 
 
 
 
 U) 
 
 1—1 
 
 o 
 
 4-1 
 
 v^^ 
 
 O 
 
 V 
 
 a 
 
 X) 
 
 o 
 
 o 
 
 
 r-l 
 
 01 
 
 
 ■M 
 
 
 to 
 
 .^ 
 
 XI 
 
 
 
 (J 
 
 >^ 
 
 a. 
 
 o 
 
 
 
 
 ^ 
 
 1-1 
 
 c 
 
 CL 
 
 u 
 
 JZ 
 
 r— 1 
 
 Cu 
 
 c 
 
 f—i 
 
 u 
 
 •r-l 
 
 s^ 
 
 C 
 
 
 
 
 • H 
 
 :ii 
 
 V 
 
 0-. 
 
 c 
 
 ■o 
 
 C 
 
 u 
 
 w 
 
 o 
 
 1) 
 
 
 to 
 
 •r-l 
 
 to 
 
 o 
 
 r-4 
 
 C 
 
 u 
 
 
 d 
 
 
 
 
 cl 
 
 
 OC 
 
 
 3 
 
 CJ 
 
 c 
 
 i3 
 
 c 
 
 p 
 
 Di 
 
 10 
 
 •r-l 
 
 CJ 
 
 o, 
 
 •r^ 
 
 CJ 
 
 CJ 
 
 CO 
 
 CO 
 
 ^ 
 
 
 
 
 O 
 
 • 
 
 
 IJ-I 
 
 O 
 
 x: 
 
 •r-l 
 
 M 
 
 c? 
 
 
 CJ 
 
 ;i- 
 
 7'. 
 
 
 u 
 
 c 
 
 4-1 
 
 D- 
 
 .—1 
 
 o 
 
 
 
 
 C 
 
 "^ 
 
 i~t 
 
 o 
 
 t-< 
 
 CO 
 
 i-l 
 
 
 
 CI 
 
 Ul 
 
 -C 
 
 
 - 
 
 -C 
 
 to 
 
 •rj 
 
 to 
 
 G 
 
 •— ' 
 
 £: 
 
 
 
 
 o 
 
 
 o 
 
 
 ca 
 
 
 n] 
 
 4-J 
 
 C 
 
 X 
 
 U 
 
 o 
 
 XI 
 
 O 
 
 u 
 
 1-1 
 
 r- 
 
 cu 
 
 o 
 
 ■H 
 
 _cu 
 
 
 
 
 u 
 
 • 
 
 fl 
 
 > 
 
 
 r-l 
 
 N 
 
 c 
 
 •r-l 
 
 
 CO 
 
 n 
 
 ^ 
 
 
 r-* 
 
 CO 
 
 co 
 
 
 CJ 
 
 Vj 
 
 ^ 
 
 
 
 
 U) 
 
 <; 
 
 k-l 
 
 V 
 
 • 
 
 .Q 
 
 ■ ^ 
 
 :j 
 
 il 
 
 • 
 
 u 
 
 o 
 
 4_l 
 
 • 
 
 to 
 
 G. 
 
 
 * 
 
 
 3 
 
 
 
 
 
 
 
 iJ 
 
 > 
 
 J 
 
 D 
 
 ,_v 
 
 u 
 
 4-1 
 
 ^^ 
 
 u 
 
 u 
 
 U 
 
 v-^ 
 
 CJ 
 
 CJ 
 
 c 
 
 CO 
 
 4J 
 
 D 
 
 CO 
 
 
 
 
 1) 
 
 • 
 
 c 
 
 u 
 
 
 D. 
 
 ■r-» 
 
 1-1 
 
 o 
 
 
 c 
 
 c 
 
 CO 
 
 
 to 
 
 w 
 
 •^ 
 
 
 to 
 
 
 Ci-i 
 
 
 
 
 JZ] u 
 
 
 
 D 
 
 • 
 
 C 
 
 u 
 
 ^ 
 
 1-. 
 
 • 
 
 o 
 
 c 
 
 1-1 
 
 • 
 
 CJ 
 
 Cu 
 
 o 
 
 • 
 
 o 
 
 • 
 
 s 
 
 
 
 
 Hi O 
 
 o 
 
 col x 
 
 U 
 
 :_) 
 
 6 
 
 (X, 
 
 H 
 
 o 
 
 < 
 
 4-1 
 
 H 
 
 Di 
 
 - 
 
 -) 
 
 Z2 
 
 CJ 
 
 ^ 
 
 ^. 
 
 
 CD 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ac: 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 rl 
 
 
 (SI 
 
 
 n 
 
 
 ^ 
 
 
 
 m 
 
 
 \0 
 
 
 
 r«. 
 
 
 
 -< 
 
 
 .— t 
 
 
 r-l 
 
 
 r-l 
 
 
 
 r-l 
 
 
 r-l 
 
 
 
 f-i 
 
 
 
 CJ 
 
 
 o 
 
 
 o 
 
 Q 
 
 
 o 
 
 
 
 O 
 
 Q 
 
 
 o 
 
 Q 
 
 
 
 o 
 
 Q 
 
 
 
 II. 4-3 
 
>- 
 
 c- 
 -a: 
 OS 
 CD 
 CD 
 
 CC3 
 
 
 CO 
 
 H 
 
 o 
 
 w 
 
 o 
 
 M 
 
 g 
 
 o 
 
 
 h- 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ■*** 
 
 CJ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 c2 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 »— 
 
 
 
 
 
 
 
 
 
 
 • X 
 
 
 X 
 
 
 
 
 
 oo 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 mxt; 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^^^^^ 
 
 
 
 •«* 
 
 V- 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 |1J 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Li.. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 C3 
 
 ru- 
 
 
 
 
 
 
 
 
 X 
 
 
 X 
 
 
 
 
 
 
 
 es 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 __^ 
 
 
 
 ,, 
 
 
 
 0) 
 
 
 
 
 
 
 
 <u 
 
 
 
 
 
 
 
 
 
 
 o 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 •H 
 
 
 
 
 
 
 
 •H 
 
 
 
 
 
 
 
 
 ^ 
 
 
 y-i 
 
 
 X 
 
 
 
 ^ 
 
 
 U-l 
 
 
 ;!«5 
 
 
 
 .!<! 
 
 
 
 M 
 
 
 U-( 
 
 
 u 
 
 
 
 M 
 
 
 «M 
 
 
 M 
 
 
 
 Vi 
 
 
 nJ 
 
 
 O 
 
 
 CO 
 
 
 
 Id 
 
 
 
 
 
 ^ 
 
 
 
 R) 
 
 
 
 CW 
 
 
 
 
 P^ 
 
 
 
 P4 
 
 
 
 
 P-l 
 
 
 
 P4 
 
 
 P^ 
 
 
 
 C 
 
 
 
 
 
 
 
 c 
 
 
 
 
 
 
 
 «: 
 
 H> 
 
 
 o 
 
 
 CU 
 
 
 
 0) 
 
 
 
 
 
 0) 
 
 
 
 lU 
 
 
 CO 
 
 M 
 
 
 T-l 
 
 
 M 
 
 
 
 00 
 
 
 ■1-1 
 
 
 t>0 
 
 
 
 U 
 
 
 
 QJ 
 
 
 to 
 
 
 CU 
 
 
 
 <u 
 
 
 to 
 
 
 CU 
 
 
 
 0) 
 
 
 i-l 
 
 
 •H 
 
 
 r-l 
 
 
 
 I-l 
 
 
 •M 
 
 
 t-l 
 
 
 
 iH 
 
 
 
 t-l 
 
 
 > 
 
 
 I-l 
 
 
 
 i-4 
 
 
 > 
 
 
 I-l 
 
 
 
 M 
 
 
 
 O 
 
 
 1-( 
 
 
 
 
 
 
 
 
 
 •rl 
 
 
 
 
 
 
 
 
 
 
 o 
 
 
 Q 
 
 
 CJ 
 
 
 
 CJ 
 
 
 Q 
 
 
 
 
 
 
 CJ 
 
 
 
 
 
 en 
 
 0) 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 u 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 3 
 
 
 
 lU 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 4-) 
 
 
 
 u 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 c 
 
 
 
 Rj 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 OJ 
 
 
 
 u 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 J> 
 
 
 
 4-1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 F-l 
 
 
 
 CU 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 a 
 
 
 
 
 
 
 
 
 
 
 
 m 
 
 • 
 
 
 
 
 
 2 
 
 
 
 •r4 
 
 
 
 C 
 
 
 
 
 
 
 
 CU 
 
 
 
 
 
 
 
 pll 
 
 
 
 o 
 
 
 
 
 
 
 
 
 
 
 
 CO 
 
 • 
 
 
 
 
 
 «t 
 
 
 
 u 
 
 
 
 
 
 
 
 
 
 
 
 CO 
 
 iJ 
 
 
 
 
 
 ej 
 
 
 
 m 
 
 
 
 
 
 
 
 
 
 
 lU 
 
 
 
 
 
 
 —1 
 Ota 
 ^3 
 
 
 
 
 
 
 J3 
 CO 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 o 
 
 
 
 •H 
 
 
 
 
 
 
 
 u 
 
 "2 
 
 
 
 
 
 Q_ 
 
 
 
 o 
 
 
 
 fX4 
 
 
 
 
 
 
 
 p^ 
 
 a 
 
 
 
 
 
 
 OT 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CO 
 
 
 
 
 
 
 <u 
 
 
 
 a 
 
 
 
 <4-l 
 
 
 
 to 
 
 
 
 
 
 C 
 
 r-l 
 
 
 
 
 
 
 ^ 
 
 
 
 PM 
 
 
 
 
 
 
 
 4J 
 
 
 
 
 
 
 
 >, 
 
 
 
 
 
 
 TO 
 
 
 
 (^ 
 
 
 
 
 
 
 C 
 
 
 
 
 
 •r^ 
 
 M 
 
 
 
 
 
 
 CJ 
 
 
 
 u 
 
 
 
 CO 
 CU 
 
 
 
 <u 
 
 1-4 
 
 
 
 
 
 4J 
 
 to 
 
 CO 
 
 S 
 
 
 
 
 
 
 g 
 
 
 
 o 
 
 
 
 •H 
 
 
 
 'O 
 
 
 
 
 
 u 
 
 
 
 
 
 
 
 
 
 U-l 
 
 
 
 T3 
 
 
 
 (U 
 
 
 
 
 
 4J 
 
 >*H 
 
 
 
 
 
 
 X 
 
 
 
 CO 
 
 
 
 3 
 
 4-1 
 
 "oj 
 
 
 
 
 
 
 
 CU 
 
 
 
 
 
 
 
 
 e 
 
 
 
 u 
 
 
 
 CO 
 
 0), 
 
 
 e 
 
 
 
 
 
 
 
 • 
 
 
 
 
 
 
 o 
 
 
 
 u 
 
 
 
 
 WW 
 
 
 
 
 
 y<^ 
 
 
 rt 
 
 •n 
 
 
 
 
 
 
 »-l 
 
 
 
 (U 
 
 
 
 
 CO 
 
 
 
 
 
 c 
 
 
 b 
 
 v> 
 
 
 
 
 
 
 M-l 
 
 
 
 e 
 
 
 
 • 
 
 u 
 
 
 c 
 
 
 
 
 
 
 CJ 
 
 
 
 
 
 
 
 eg 
 
 
 
 
 § 
 
 
 
 x: 
 
 u 
 
 
 1-1 
 
 
 
 u 
 
 
 
 
 
 
 CO 
 
 
 
 
 I^ 
 
 O 
 
 
 
 1^ 
 
 
 
 
 
 
 
 
 <u 
 
 
 
 ^ 
 
 
 c 
 
 1-1 
 
 
 r- 
 
 
 
 
 ON 
 
 P-( 
 
 
 
 •H 
 
 
 
 ij 
 
 J3 
 
 
 4-) 
 
 
 
 
 •H 
 
 r-l 
 
 
 e3> 
 
 
 
 
 1—1 
 
 Du 
 
 
 
 3 
 
 
 
 CO 
 
 CO 
 
 
 
 
 
 
 CU 
 
 •H 
 
 
 CU 
 
 (U 
 
 
 r-l 
 
 
 
 
 
 
 
 cr 
 
 
 
 <U 
 
 
 
 u 
 
 
 
 4J 
 
 X! 
 
 
 u 
 
 J3 
 
 
 
 
 
 
 M-l 
 
 
 
 Q) 
 
 
 
 CO 
 
 i-t 
 
 
 p-l 
 
 
 
 CO 
 
 to 
 
 
 
 
 CJ 
 
 •'-» 
 
 
 
 
 , 
 
 O 
 
 
 
 Oi 
 
 
 
 CU 
 
 CO 
 
 
 
 
 
 iJ 
 
 CO 
 
 
 u 
 
 Vj 
 
 CO 
 
 
 
 
 y> 
 
 
 
 
 
 
 
 oi 
 
 c 
 
 
 CU 
 
 
 
 4-1 
 
 :2 
 
 
 P4 
 
 to 
 
 ^ 
 
 
 
 
 Cct 
 
 0) 
 
 
 
 CO 
 
 
 
 
 
 
 
 > 
 
 
 
 C 
 
 
 
 
 S 
 
 g 
 
 
 
 
 cb 
 
 1-1 
 
 
 
 C 
 
 
 
 c 
 
 t4 
 
 
 1-4 
 
 
 
 tU 
 
 m 
 
 
 .c 
 
 
 § 
 
 r~ 
 
 
 
 ^ 
 
 3 
 
 
 
 o 
 
 
 
 •r^ 
 
 4J 
 
 
 4-1 
 
 
 
 
 
 
 
 
 CO 
 
 • 
 
 v-/ 
 
 vD 
 
 
 
 4-> 
 
 
 
 •H 
 
 
 
 CU 
 
 •H 
 
 
 CO 
 
 
 
 c 
 
 
 
 •r^ 
 
 S 
 
 
 fO 
 
 
 
 => 
 
 O 
 
 
 
 4-1 
 
 
 
 4J 
 
 h 
 
 
 C 
 
 
 
 
 
 • 
 
 
 fe 
 
 
 lU 
 
 1 
 
 > 
 
 
 
 «t: 
 
 CO 
 
 14-1 
 
 3 
 C 
 
 u 
 
 o 
 
 
 
 CO 
 1-1 
 w 
 
 CO 
 
 e 
 o 
 B 
 <u 
 
 Q 
 CO 
 
 
 
 
 I-l 
 
 P-l 
 
 i 
 1-1 
 
 C4-1 
 
 <u 
 
 60 
 
 3 
 •T3 
 
 F-l 
 
 n 
 
 
 12 
 
 
 
 m 
 
 r-l 
 
 1 
 
 CN 
 
 t-l 
 
 1-1 
 
 <u 
 
 4-1 
 1-1 
 
 <: 
 
 ■a 
 1-4 
 
 CO 
 3 
 
 
 
 t3 
 
 
 CJ 
 
 c 
 
 •r-l 
 
 0) 
 
 4-1 
 
 U 
 P-4 
 
 ^^ 
 
 c: 
 
 
 
 CO 
 tJJ 
 
 
 
 1-1 
 
 
 4-1 
 
 I-l 
 4) 
 
 X) 
 
 
 >^ 
 
 1-1 
 
 4-1 
 
 c 
 <u 
 
 CJ 
 
 1-H 
 
 CO 
 
 
 
 u 
 CO 
 
 4J 
 
 4J 
 CO 
 
 rH 
 •H 
 .0 
 
 
 
 
 iJ 
 
 ^ 
 
 
 CO 
 
 
 
 ^1 
 
 4-1 
 
 
 
 H 
 
 
 
 CO 
 
 Di 
 
 
 
 
 • 
 
 4J 
 
 u 
 
 (0 
 
 
 
 
 c 
 
 T) 
 
 
 0) 
 
 /-v 
 
 
 CU 
 
 
 u 
 
 
 1-4 
 
 
 •1-1 
 
 
 
 •H 
 
 3 
 
 CO 
 
 
 
 *J 
 
 
 
 
 <u 
 
 C 
 
 
 o 
 
 O 
 
 
 6 
 
 C 
 
 ^ 
 
 CO 
 
 r-l 
 
 
 Ua 
 
 T3 
 
 
 e 
 
 
 c 
 
 CU 
 
 Q 
 
 
 
 
 g 
 
 ctf 
 
 
 o 
 
 CJ 
 
 
 w 
 
 
 
 0) 
 
 0) 
 
 
 
 d 
 
 
 
 
 • 
 
 M 
 
 0) 
 
 CU 
 
 y-^ 
 
 
 
 a 
 
 I-l 
 
 
 u 
 
 Q 
 
 
 
 tH 
 
 
 •d 
 
 c 
 
 
 U-l 
 
 CO 
 
 
 c 
 
 •-) 
 
 W 
 
 Pi 
 
 
 
 CO 
 
 
 LU 
 
 •H 
 
 CO 
 
 u 
 
 pL, 
 
 W 
 
 u 
 
 >» 
 
 4-1 
 
 
 3 
 
 1-1 
 
 4J 
 
 
 
 
 4-1 
 
 
 
 
 
 
 u 
 
 pL| 
 
 
 pi 
 
 F= 
 
 3 
 
 M 
 
 ij 
 
 
 a. 
 
 1-1 
 
 CO 
 
 3 
 
 > 
 
 4-1 
 
 
 
 u 
 
 
 X) 
 
 V4 
 
 
 
 •» 
 
 • 
 
 t-l 
 
 < 
 
 ^ 
 
 
 CT 
 
 
 o 
 
 -o 
 
 v^-' 
 
 o 
 
 s 
 
 1-1 
 
 to 
 
 •H 
 
 a 
 
 
 
 Cfl 
 
 I-l 
 
 
 
 w 
 
 to 
 
 CJ 
 
 CO 
 
 
 ^ 
 
 
 w 
 
 O 
 
 o. 
 
 c 
 
 
 04 
 
 
 
 
 T3 
 
 4-1 
 
 ^— / 
 
 Q. 
 
 
 
 a) 
 
 o. 
 
 
 CO 
 
 
 
 
 •a 
 
 N^' 
 
 
 
 
 "O 
 
 0) 
 
 CO 
 
 c 
 
 <D 
 
 CO 
 
 CO 
 
 
 
 4J 
 
 
 0) 
 
 •H 
 
 •r^ 
 
 tU 
 
 6C 
 
 c 
 
 • 
 
 •H 
 
 r-l 
 
 4J 
 
 
 
 .— I 
 
 o 
 
 Dd 
 
 
 CO 
 
 Oi 
 
 4-) 
 
 
 H 
 
 < 
 
 1—1 
 
 oi 
 
 E 
 
 ci-l 
 
 ed 
 
 C 
 
 0) 
 
 w 
 
 c 
 
 <u 
 
 I-l 1-1 
 
 
 
 w 
 
 jr 
 
 
 1-1 
 
 e 
 
 
 c 
 
 60 
 
 
 
 1—1 
 
 
 
 
 x: 
 
 
 •H 
 
 e 
 
 
 x; 
 
 •r^ 
 
 r-4 
 
 
 
 •H 
 
 a; 
 
 4-1 
 
 CO 
 
 a; 
 
 4-t 
 
 to 
 
 c 
 
 CU 
 
 V4 
 
 to 
 
 4-1 
 
 c 
 
 
 
 4-1 
 
 1-1 
 
 r— 1 
 
 1* 
 
 
 
 f^ 
 
 lU G. 
 
 
 
 u 
 
 
 o 
 
 u 
 
 ■c 
 
 
 
 r-l 
 
 •r-l 
 
 
 
 <u 
 
 u 
 
 
 
 
 
 4J 
 
 
 
 tu 
 
 < 
 
 CO 
 
 tU 
 
 
 3 s 
 
 
 
 1-1 
 
 Ci-I 
 
 CD 
 
 •r^ 
 
 c 
 
 cO 
 
 pL. 
 
 U /-N 
 
 1-1 
 
 E 
 
 
 to 
 
 
 
 3 
 
 CO 
 
 OJ 
 
 
 t-l 
 
 H 
 
 c 
 
 x> pS 
 
 
 
 <D 
 
 o 
 
 1-1 
 
 C 
 
 *H 
 
 1-1 
 
 
 w CO 
 
 0) 
 
 3 
 
 • 
 
 1-1 
 
 w 
 
 1-1 
 
 u 
 
 c 
 
 • 
 
 1-1 
 
 
 to 
 
 ■r-l 
 
 
 
 
 
 4-) 
 
 C 
 
 u 
 
 kJ 
 
 i-l 
 C 
 
 3 
 
 0) 
 
 CO 2: 
 
 
 CO 
 
 C 
 
 >J 
 
 J-l 
 
 c 
 
 0) 
 
 U 
 
 4-1 
 
 c 
 
 •H 
 
 or 
 
 y: 
 
 •H 
 U 
 
 ^ 
 
 
 
 e 
 
 CO CO 
 
 
 
 o 
 
 o 
 
 O 
 
 <u 
 
 • 
 
 
 
 z; 
 
 
 
 
 
 • 
 
 
 
 pC 
 
 • 
 
 
 
 C 
 
 • 
 
 3 
 
 
 
 to 
 
 • *^ 
 
 
 
 CJ 
 
 — ' 
 
 CJ 
 
 H 
 
 o 
 
 u 
 
 
 S ^ 
 
 
 
 CJ 
 
 Q 
 
 
 
 H 
 
 >-) 
 
 
 
 w 
 
 i^ 
 
 Q 
 
 2: 
 
 • CO 
 
 > z 
 
 
 CD 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 a= 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ri 
 
 00 
 
 
 a\ 
 
 
 
 
 
 
 .-1 
 
 
 C<1 
 
 
 *n 
 
 
 
 5 
 
 
 «c 
 
 i-i 
 
 
 1-4 
 
 
 CS 
 
 
 
 CN 
 
 
 eg 
 
 
 Ol 
 
 
 
 CM 
 
 
 CJ 
 
 o 
 
 Q 
 
 
 o 
 
 Q 
 
 
 
 
 Q 
 
 
 
 
 
 Q 
 
 
 
 
 Q 
 
 
 
 
 Q 
 
 
 
 § 
 
 
 II. 4-4 
 
>- 
 
 Ou 
 
 ca 
 
 „ i 
 pa 
 
 ca 
 
 CD 
 
 CO 
 
 w 
 
 CO 
 C/3 
 
 w 
 o 
 
 o 
 a; 
 (u 
 
 >i 
 
 o 
 
 o 
 
 pa 
 
 ►-- 
 
 
 
 
 
 
 
 
 .-iM^i 
 
 
 
 
 ^^^^ 
 
 
 
 
 
 
 CJ> 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 X 
 
 
 
 
 
 
 
 
 »— 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 c/» 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 «x 
 
 
 
 
 
 
 
 
 
 
 
 
 .A*^ 
 
 
 
 
 
 
 «,— 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Lj 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ll- 
 
 
 
 
 
 . ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 C3 
 
 
 
 
 
 K 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CJ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 ^ 
 
 
 .AS 
 
 
 j^ 
 
 
 
 ^ 
 
 
 
 
 .!^ 
 
 ^ 
 
 
 
 
 u 
 
 
 M 
 
 
 »4 
 
 
 )-i 
 
 
 
 u 
 
 
 
 
 VI 
 
 h 
 
 
 
 as: 
 cz> 
 
 t8 
 
 
 to 
 
 
 09 
 
 
 CO 
 
 
 
 CD 
 
 
 
 
 CO 
 
 cS 
 
 
 
 Pj 
 
 
 1^4 
 
 
 (^ 
 
 
 P4 
 
 
 
 PL4 
 
 
 
 
 P4 
 
 Pui 
 
 
 
 1— 
 
 0) 
 
 
 <u 
 
 
 <0 
 
 
 0) 
 
 
 
 gj 
 
 
 
 
 « 
 
 «■ 
 
 
 
 «0C 
 
 bO 
 
 
 CiO 
 
 
 00 
 
 
 (so 
 
 
 
 eo 
 
 
 
 
 bO 
 
 C>0 
 
 
 
 C3 
 
 <u 
 
 
 <u 
 
 
 (U 
 
 
 0) 
 
 
 
 0) 
 
 
 
 
 <0 
 
 0) 
 
 
 
 o 
 
 r-l 
 
 
 1-4 
 
 
 i-H 
 
 
 r-l 
 
 
 
 r-l 
 
 
 
 
 r-l 
 
 r-l 
 
 
 
 i-( ■ 
 
 
 r-l 
 
 
 t-l 
 
 
 I-l 
 
 
 
 r-l 
 
 
 
 
 r-l 
 
 r-l 
 
 
 
 
 o 
 
 
 o 
 
 
 O 
 
 
 O 
 
 
 
 o 
 
 
 
 
 o 
 
 o 
 
 
 
 
 o 
 
 
 o 
 
 
 u 
 
 
 u 
 
 
 
 o 
 
 
 
 
 o 
 
 o 
 
 
 
 
 
 
 
 
 
 
 
 
 
 1 
 u 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 * 
 
 
 
 u 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 >» 
 
 
 
 *J 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 r-l 
 
 
 
 X 
 
 
 
 
 
 
 C 
 
 
 
 
 
 
 
 
 
 r-l 
 
 
 
 u 
 
 
 
 
 
 
 o 
 
 
 
 
 
 
 
 
 
 <1> 
 
 
 
 
 
 
 
 
 i 
 
 tH 
 
 
 
 w 
 
 
 
 
 
 
 
 i>C| 
 
 c 
 
 
 
 «J 
 
 
 
 
 
 o 
 
 A-l 
 
 
 
 to 
 
 
 
 
 
 
 
 c 
 
 c 
 
 
 
 c 
 
 
 
 
 
 (M 
 
 to 
 
 
 
 <u 
 
 
 
 
 
 
 
 •H 
 
 o 
 
 
 
 0) 
 
 
 
 
 
 fl4 
 
 N 
 
 
 2C 
 
 o 
 
 
 
 
 
 
 
 N 
 
 CJ 
 
 
 
 > 
 
 
 
 
 
 
 T-l 
 
 
 o 
 
 o 
 
 
 
 
 
 
 
 •T-< 
 
 
 
 
 r-l 
 
 
 
 
 
 00 
 
 ^S 1-4 
 
 
 ^— 
 
 u 
 
 
 
 
 
 
 
 r-l 
 
 • 
 
 
 
 o 
 
 
 
 
 
 c 
 
 CO -r-l 
 
 
 «x 
 
 (U 
 
 
 
 
 
 
 
 T-l 
 
 •n 
 
 
 
 CO 
 
 
 
 
 
 •H 
 
 fcj 4-) 
 
 
 C3 
 
 
 
 
 
 
 
 
 ■U 
 
 
 
 
 
 
 
 
 
 tn 
 
 § S 
 
 
 
 I-l 
 
 
 
 
 
 
 
 ■s 
 
 • 
 
 
 
 fiC 
 
 
 
 
 
 tn 
 
 2^ 
 
 
 —J 
 
 ca 
 
 to 
 
 
 
 
 
 
 
 
 >-) 
 
 
 
 c 
 
 
 
 
 
 <U 
 
 ^ x; 
 
 
 u 
 
 •H 
 
 
 
 
 
 
 
 4-1 
 
 A 
 
 
 
 T-l 
 
 N 
 
 
 
 
 
 CJ 
 
 o 
 
 tn 
 
 M T-l 
 
 
 M 
 
 
 
 
 
 
 
 C 
 
 >> 
 
 
 
 •rl 
 
 
 
 
 
 1-1 
 
 <U fn 
 
 
 
 O 
 
 
 
 
 
 
 
 to 
 
 r-l 
 
 
 
 . iH 
 
 
 
 
 
 P4 
 
 X) 
 
 
 
 i-l 
 
 
 
 
 
 
 
 
 <u 
 
 
 
 •H 
 
 
 
 
 
 
 >, <« 
 
 
 
 O 
 
 
 
 
 
 
 
 Of 
 
 > 
 
 
 
 4J 
 
 
 
 
 
 u 
 
 a o 
 
 
 
 •>H 
 
 
 
 
 
 »-x 
 
 
 c 
 
 
 
 
 D 
 
 
 
 
 
 o 
 
 CO 
 
 
 
 m 
 
 
 
 
 
 >, 
 
 
 •r4 
 
 • 
 
 
 
 
 
 
 
 
 ■4-1 
 
 s^ 
 
 
 
 
 
 
 
 
 c 
 
 
 tn 
 
 o 
 
 
 
 u 
 
 
 
 
 
 
 . 0£; 1 
 
 
 ^ 
 
 
 
 
 
 c 
 
 to 
 
 
 tn 
 
 
 
 
 T-l 
 
 
 
 
 
 to 
 
 o or 
 
 
 
 o 
 
 
 
 
 
 o 
 
 a 
 
 
 4) 
 
 • 
 
 
 
 c 
 
 
 
 
 
 0) 
 
 r-l 
 
 
 
 <*^ 
 
 
 
 
 
 4J 
 
 a 
 
 o 
 
 
 O 
 
 • > 
 
 
 
 t> 
 
 
 
 
 
 T-l 
 4J 
 
 Q . C 
 
 
 
 a 
 
 
 
 
 
 o 
 
 q 
 
 
 h 
 
 M 
 
 
 
 « 
 
 
 
 
 
 T-l 
 
 x: 
 
 
 
 M 
 
 
 
 
 
 to 
 
 / 
 
 
 P4 
 
 ^ 
 
 
 
 c 
 
 
 
 
 
 1— i 
 
 "» CJ 
 
 
 
 •J- 
 
 s 
 
 
 
 
 
 n 
 
 00 
 
 
 
 c 
 
 
 
 .r4 
 
 
 
 
 
 T-l 
 
 n tu 
 
 
 
 O 
 
 
 
 
 
 
 iJ 
 
 d 
 
 
 0) 
 
 -A 
 
 
 
 tn 
 
 
 ^N 
 
 
 
 y 
 
 O H 
 
 
 
 C^ 
 
 y-i 
 
 
 
 
 
 X 
 
 •^^ 
 
 
 4J 
 
 a 
 
 
 
 to 
 
 
 (U 
 
 
 
 CO 
 
 tn 
 
 
 
 i-( 
 
 C3 
 
 
 
 
 
 w 
 
 u 
 
 
 CO 
 
 
 
 
 (U 
 
 
 4J 
 
 
 
 Pn 
 
 t4 <U 
 
 
 
 
 
 
 
 
 
 <u 
 
 
 M 
 
 • 
 
 
 
 O 
 
 
 3 
 
 
 
 
 V4 jz; 
 
 
 
 c 
 
 
 
 
 
 X 
 
 <u 
 
 
 4-1 
 
 M 
 
 
 
 o 
 
 
 4-1 
 
 
 
 T) 
 
 to H 
 
 
 
 o 
 
 
 
 
 
 m 
 
 d 
 
 
 c 
 
 
 
 
 ^ 
 
 
 T-l 
 
 
 
 c 
 
 P4 
 
 
 ^ > 
 
 •H 
 
 
 
 
 
 •H 
 
 •r-l 
 
 
 <u 
 
 • 
 
 
 
 p-l 
 
 
 '4J 
 
 
 
 CO 
 
 
 
 CO 
 
 U 
 
 
 
 
 
 U- 
 
 00 
 
 
 o 
 
 S 
 
 
 
 
 
 to 
 
 
 o 
 
 
 
 * M 1 
 
 oc 
 
 to 
 
 
 
 
 
 
 a 
 
 
 c 
 
 
 
 
 o 
 
 
 a 
 
 
 P4 
 
 
 u 
 
 Pi c 
 
 
 C3 
 
 r-t 
 
 
 
 
 
 Vj 
 
 w 
 
 
 o 
 
 * 
 
 
 
 PL| 
 
 
 ,M 
 
 
 fe 
 
 
 C 
 
 o 
 
 
 3C 
 
 1-1 
 
 
 
 
 
 o 
 
 
 
 o 
 
 •a 
 
 
 
 (14 
 
 
 
 
 
 
 tu 
 
 . -o 
 
 
 >— 
 
 CO 
 
 
 
 
 
 tl-l 
 
 ■a 
 
 
 
 I-l 
 
 
 
 
 
 r-t 
 
 
 tl 
 
 
 fi 
 
 w c 
 
 
 => 
 
 4J 
 
 
 
 
 
 
 ^ 
 
 
 c 
 
 tn 
 
 <u 
 
 ^N 
 
 
 c 
 
 
 .<0 
 
 
 o 
 
 
 o. 
 
 o 
 
 
 "S 
 
 tn 
 
 
 
 
 
 tn 
 
 
 •r^ 
 
 <u 
 
 •rl 
 
 <u 
 
 
 to 
 
 
 T-l 
 
 
 14-1 
 
 
 T-l 
 
 " ►.! 
 
 
 
 t: 
 
 
 
 
 
 u 
 
 
 
 <u 
 
 ^1 
 
 U-l 
 
 JJ 
 
 
 
 
 u 
 
 
 
 
 3 
 
 r-l 
 
 
 
 M 
 
 •^ 
 
 
 /-» 
 
 
 C 
 
 x: 
 
 
 4-1 
 
 3 
 
 x: 
 
 rf 
 
 
 >4-l 
 
 
 O 
 
 
 x: 
 
 ^N 
 
 cr 
 
 J3 - i 
 
 
 
 • 
 
 
 s 
 
 
 tu 
 
 o 
 
 
 o 
 
 13 
 
 o 
 
 4-1 
 
 
 o 
 
 
 g 
 
 
 tn 
 
 2 
 
 u 
 
 O • 
 
 
 
 T3 
 
 y 
 
 
 
 
 > 
 
 u 
 
 
 u 
 
 (U 
 
 4-1 
 
 T-l 
 
 
 
 
 <u 
 
 
 T-l 
 
 
 
 c -o 
 
 
 
 c 
 
 c 
 
 
 ^ 
 
 
 I-l 
 
 CO 
 
 
 PL| 
 
 o 
 
 •r4 
 
 *J 
 
 
 c 
 
 
 S 
 
 
 fl4 
 
 ••a 
 
 r-l 
 
 ^ H 
 
 
 
 to 
 
 M 
 
 
 
 
 o 
 
 JJ 
 
 
 
 o 
 
 hJ 
 
 tn 
 
 
 o 
 
 
 
 
 
 
 to 
 
 1-3 
 
 
 
 
 
 
 < 
 
 
 CO 
 
 en 
 
 
 J= 
 
 u 
 
 
 C 
 
 
 T-l 
 
 
 0) 
 
 
 X) 
 
 <: 
 
 4J 
 
 • 
 
 
 
 c 
 
 •« 
 
 
 
 
 
 01 
 
 
 tn 
 
 p-l 
 
 • 
 
 M 
 
 
 4J 
 
 
 1— 1 
 
 
 (U 
 
 
 c 
 
 a '-- 1 
 
 
 o 
 
 to 
 
 
 m 
 
 CO 
 
 
 U-l 
 
 Pi 
 
 
 -A 
 
 
 X 
 
 
 
 CJ 
 
 
 r-4 
 
 
 4-> 
 
 to 
 
 CU 
 
 to 
 
 1 
 
 
 •r-l 
 
 <u 
 
 
 <u 
 
 to 
 
 
 O 
 
 
 
 1X4 
 
 c 
 
 
 I-l 
 
 
 3 
 
 
 <U 
 
 
 to 
 
 to 
 
 e 
 
 • ^ 
 
 1 
 
 
 1-1 
 
 •rl 
 
 
 1-1 
 
 X 
 
 
 
 
 
 
 
 o 
 
 • 
 
 CO 
 
 
 ^4 
 
 
 4J 
 
 
 <u 
 
 X 
 
 •r4 
 
 O a 
 
 1 
 
 
 o 
 
 M 
 
 
 TS 
 
 v 
 
 
 > 
 
 U5 
 
 
 CO 
 
 •r4 
 
 >-) 
 
 •H 
 
 
 4J 
 
 
 U 
 
 
 u 
 
 ,<" 
 
 Vl 
 
 • 
 
 1 
 
 
 D 
 
 4-1 
 
 U 
 
 3 
 
 H 
 
 4J 
 
 T3 
 
 <n 
 
 4J 
 
 
 4-1 
 
 
 14 
 
 4-1 
 
 tn 
 
 tn 
 
 10 
 
 4-1 
 
 H 
 
 H 4J 
 
 <U 
 
 - cc 
 
 1 
 
 
 M 
 
 to 
 
 >-l 
 
 •u 
 
 ^^^ 
 
 M 
 
 D 
 
 u 
 
 M 
 
 14-1 
 
 o 
 
 •s 
 
 o 
 
 u 
 
 C 
 
 (U 
 
 « 
 
 u 
 
 
 ^^ u 
 
 Ou 
 
 01 ^ 1 
 
 
 iJ 
 
 3 
 
 o- 
 
 CO 
 
 
 O 
 
 4-1 
 
 v_^ 
 
 o 
 
 o 
 
 to 
 
 1-1 
 
 e 
 
 o 
 
 o 
 
 p 
 
 ■^w^ 
 
 o 
 
 4) 
 
 o 
 
 X 
 
 r-l 
 
 
 ULl 
 
 t/I 
 
 T3 
 
 cx 
 
 
 0) 
 
 CI. 
 
 CO 
 
 
 a 
 
 / 
 
 u 
 
 <u 
 
 o 
 
 a, 
 
 o 
 
 3 
 
 
 a. 
 
 1 
 
 a) o- 
 
 w 
 
 >j or 
 
 
 pi 
 
 C 
 
 C 
 
 a; 
 
 C 
 
 ^ 
 
 (U 
 
 
 0) 
 
 0) 
 
 ' 4.J 
 
 4-1 
 
 tn 
 
 S 
 
 <u 
 
 
 x> 
 
 >% 
 
 OJ 
 
 ^ a) 
 
 
 <U 3 
 
 eg 
 
 O 
 
 M 
 
 « 
 
 o 
 
 c 
 
 CtJ 
 
 (U 
 
 o 
 
 ej 
 
 C 
 
 X 
 
 x: 
 
 
 ets 
 
 TD 
 
 0) 
 
 1-1 
 
 Pi 
 
 N 
 
 C 'CcJ 
 
 >4-l 
 
 g 4 
 
 CM 
 
 o 
 
 
 
 •I-l 
 
 •r-l 
 
 
 u 
 
 •r4 
 
 
 <U 
 
 to 
 
 o 
 
 <u 
 
 
 c 
 
 o 
 
 .-< 
 
 
 c 
 
 T-l 
 
 o 
 
 g 2 
 
 CM 
 
 
 
 C 
 
 u 
 
 X-i 
 
 <u 
 
 u 
 
 D 
 
 V4 
 
 4J 
 
 E 
 
 1 
 
 <0 
 
 ■— 1 
 
 4-1 
 
 to 
 
 o 
 
 to 
 
 4-1 
 
 w 
 
 d) 4-) 
 
 
 <o 
 
 1 
 
 
 
 to 
 
 u 
 
 3 
 
 Si 
 
 o 
 
 4J 
 
 Oi 
 
 O 
 
 a 
 
 4-1 
 
 CO 
 
 1-4 
 
 o 
 
 
 V4 
 
 c 
 
 O 
 
 
 S CJ 
 
 c 
 
 K Ot r-. 1 
 
 
 c 
 
 (0 
 
 to 
 
 C 
 
 
 CO 
 
 to 
 
 
 to 
 
 o 
 
 c 
 
 
 0) 
 
 (0 
 
 c 
 
 p-l 
 
 c 
 
 CO 
 
 U-l 
 
 to 
 
 00 
 
 c 
 
 *-^ 1 
 
 
 bC 
 
 •r-l 
 
 u 
 
 •■-1 
 
 • 
 
 M 
 
 ^4 
 
 • 
 
 I-l 
 
 1— 1 
 
 o 
 
 • 
 
 4J 
 
 u 
 
 oc 
 
 
 o 
 
 M 
 
 O 
 
 • u 
 
 •r-l 
 
 • T-l CSl 
 
 
 •r-l 
 
 iJ 
 
 u 
 
 g 
 
 3 
 
 u 
 
 <u 
 
 < 
 
 4-1 
 
 01 
 
 > 
 
 [14 
 
 4-1 
 
 4-1 
 
 •r4 
 
 r^ 
 
 CJ 
 
 4-1 
 
 
 :3 j-i 
 
 to 
 
 < x: 
 
 
 tn 
 
 V4 
 
 c 
 
 g 
 
 
 C 
 
 4-1 
 
 
 c 
 
 > 
 
 1— 1 
 
 
 to 
 
 c 
 
 tn 
 
 O 
 
 
 c 
 
 <U 
 
 c 
 
 <u 
 
 tn • 
 
 
 0) 
 
 < 
 
 o • 
 
 o 
 
 • 
 
 o 
 
 •H 
 
 • 
 
 o 
 
 01 
 
 o 
 
 • 
 
 ca 
 
 o 
 
 <V 
 
 •r4 
 
 • 
 
 o 
 
 tn 
 
 • o 
 
 p 
 
 • T-l a. 
 
 
 Q 
 
 ^.^ 
 
 o 
 
 u 
 
 3 
 
 u 
 
 .J 
 
 •-) 
 
 a 
 
 Q 
 
 CO 
 
 o 
 
 
 CJ 
 
 Q 
 
 4J 
 
 •-} 
 
 o 
 
 O 
 
 S CJ 
 
 
 O Cti D. 
 
 o 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ac 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ri 
 
 t~t 
 
 
 <VJ 
 
 
 CO 
 
 
 •* 
 
 
 
 lO 
 
 
 
 
 vO 
 
 r» 
 
 
 
 O 
 
 
 O 
 
 
 O 
 
 
 O 
 
 
 
 o 
 
 
 
 
 O 
 
 o 
 
 
 
 -< 
 
 o 
 
 
 o 
 
 
 O 
 
 
 o 
 
 
 
 o 
 
 
 
 
 o 
 
 o 
 
 
 
 (J 
 
 u 
 
 
 U] 
 
 
 u 
 
 
 Cd 
 
 
 
 M 
 
 
 
 
 H 
 
 M 
 
 
 
 II. 5-1 
 
C3 
 
 cat 
 
 s S 
 
 "^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CJ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 «c 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ee 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ►- 
 
 X 
 
 
 
 
 
 
 
 
 
 X 
 
 
 
 
 </) 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ea 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 OS 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Z 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 u. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 oe 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 <-> 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 £ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 J<! 
 
 Jd 
 
 j<; 
 
 
 
 ^ 
 
 
 ^ 
 
 
 j£ 
 
 
 
 
 
 U 
 
 M 
 
 M 
 
 
 
 >-i 
 
 
 u 
 
 
 |4 
 
 
 
 
 se: 
 
 flj 
 
 (0 
 
 CO 
 
 
 
 a 
 
 
 CO 
 
 
 0) 
 
 
 
 
 o 
 
 Pk 
 
 P4 
 
 »4 
 
 
 
 e^ 
 
 
 P4 
 
 
 P4 
 
 
 
 
 »— 
 
 <a 
 
 0) 
 
 u 
 
 
 
 4) 
 
 
 « 
 
 
 01 
 
 
 
 
 c^ 
 
 60 
 
 bO 
 
 60 
 
 
 
 M 
 
 
 60 
 
 
 60 
 
 
 
 
 ^ 
 
 u 
 
 0) 
 
 9> 
 
 
 
 0) 
 
 
 « 
 
 
 a> 
 
 
 
 
 
 1-1 
 o 
 
 1-1 
 o 
 
 o 
 
 
 
 l-l 
 i-f 
 o 
 
 
 1-4 
 r-4 
 
 o 
 
 CJ 
 
 
 t-i 
 l-l 
 o 
 o 
 
 
 
 
 
 
 X! 
 
 
 
 
 
 .^» 
 
 
 
 , 
 
 
 
 
 
 
 
 
 
 m 
 
 
 
 
 
 4J 
 
 
 
 m 
 
 
 
 
 
 
 
 
 
 1-1 
 
 
 
 
 
 c 
 
 
 
 V 
 
 
 
 
 
 
 
 
 c 
 
 
 (£4 
 
 I-« 
 
 
 
 
 0) 
 
 
 
 1-1 
 
 
 
 CQ 
 
 
 
 
 
 
 1-1 
 
 
 
 CO 
 
 
 
 
 3 
 
 
 
 "O 
 
 
 
 CO 
 
 
 
 
 
 
 0) 
 
 
 U-l 
 
 u 
 
 
 
 
 4J 
 
 
 
 3 
 
 
 
 IV 
 
 
 
 
 
 
 4J 
 
 
 o 
 
 1-1 
 
 
 
 
 1-1 
 
 
 
 4-1 
 
 
 
 o 
 
 
 
 
 
 
 o 
 
 
 
 e 
 
 
 
 
 U 
 
 
 
 to 
 
 
 
 o 
 
 
 
 
 
 ^— 
 
 Vl 
 
 
 c 
 
 <u 
 
 09 
 
 
 
 CO 
 
 
 
 
 
 
 u 
 
 
 
 
 
 CD 
 
 Pm 
 
 
 o 
 
 X 
 
 •o 
 
 
 
 c 
 
 
 
 60 
 
 
 
 P4 
 
 
 
 
 
 
 
 
 1-1 
 
 u 
 
 o 
 
 
 
 o 
 
 
 
 c 
 
 
 
 
 
 
 
 
 ^ 
 
 x: 
 
 
 J-l 
 
 
 o 
 
 
 
 o 
 
 
 
 1^ 
 
 
 
 4) 
 
 
 
 
 
 «t 
 
 M 
 
 
 a 
 
 IM 
 
 (14 
 
 
 
 
 
 
 CO 
 
 
 
 4J 
 
 
 
 
 
 U 
 
 T-l 
 
 
 u 
 
 o 
 
 
 
 
 h 
 
 
 
 CO 
 
 
 
 CO 
 
 
 
 
 
 
 Pl4 
 
 
 CD 
 
 
 Xi 
 
 
 
 o 
 
 
 
 <u 
 
 
 
 ;4 
 
 
 
 
 
 ^ 
 
 
 
 a 
 
 0) 
 
 o 
 
 
 
 "O 
 
 
 
 o 
 
 
 
 u 
 
 
 
 
 
 OQ 
 
 >4-l 
 
 
 0} 
 
 4-) 
 
 1-1 
 
 
 
 o 
 
 
 
 o 
 
 
 
 c 
 
 
 
 
 
 ^ 
 
 o 
 
 
 M 
 
 3 
 
 P!i 
 
 
 
 
 
 
 u 
 
 
 
 0) 
 
 
 
 
 
 OL. 
 
 3 
 
 o 
 
 to 
 
 3 
 
 
 &4 
 4-1 
 
 o 
 
 4J 
 
 4-1 
 lH 
 4J 
 0) 
 
 c 
 
 H 
 
 *^N ffl 
 
 c 
 
 0) 
 4J 
 O 
 
 Ph 
 
 
 
 u-l 
 o 
 
 I-t 
 
 CO 
 
 > 
 o 
 
 
 
 4J 
 CO 
 
 u 
 
 4J 
 
 c 
 
 lU 
 
 
 
 o 
 
 c 
 o 
 
 CJ 
 
 c 
 1-1 
 
 4J 
 
 
 
 
 
 
 W 
 
 c 
 
 
 m 
 
 CA u 
 
 0) 
 
 
 
 CC 
 
 
 
 o 
 
 
 
 o 
 
 
 
 
 
 
 ■o 
 
 
 
 0) 
 
 
 u 
 
 a 
 
 
 
 
 
 
 c 
 o 
 
 
 ^ 
 
 u 
 
 
 
 
 
 
 u 
 
 
 3 
 
 Z 0) 
 
 •o 
 
 
 
 4J 
 
 
 
 CJ 
 
 
 ^-\ 
 
 
 
 
 
 
 
 c 
 o 
 
 a) 
 
 j2 
 
 
 0) 
 
 ^1 
 
 o 
 
 
 
 c 
 
 0) 
 
 
 
 c 
 
 
 en 
 
 c 
 
 /-> 
 
 
 
 
 
 o 
 
 4_} 
 
 
 u 
 
 (U 
 
 P4 
 
 
 
 > 
 
 
 
 1-4 
 
 
 ^ 
 
 1-1 
 
 0) 
 
 
 
 
 
 
 
 
 o 
 
 1-1 •« 
 
 
 
 
 l-l 
 
 
 
 0) 
 
 
 z 
 
 u 
 
 l-l 
 
 
 
 
 _ 
 
 
 ^ 
 
 
 m 
 
 ti 00 
 
 o 
 
 
 
 o 
 
 
 
 4J 
 
 
 >*^ 
 
 CO 
 
 i-H 
 
 
 
 
 t<0 
 
 -* 
 
 o 
 
 
 Ph 
 
 at c ^ 
 
 V 
 
 
 
 w 
 
 
 
 o 
 
 
 
 s 
 
 1^ 
 
 
 
 
 
 vO 
 
 lU 
 
 
 
 e -H vo 
 
 
 /-^ 
 
 
 
 
 
 ^4 
 
 
 u 
 
 
 > 
 
 
 
 
 oc 
 
 a\ 
 
 
 
 c 
 
 E 4-1 ON 
 
 <n 
 
 >N 
 
 
 0} 
 
 
 
 P4 
 
 
 V 
 
 c 
 
 CO 
 
 
 
 
 o 
 
 i-( 
 
 en 
 
 
 o 
 
 CO V 1-1 
 
 <U 
 
 4J 
 
 
 4J 
 
 
 
 
 
 l-l 
 
 1-4 
 
 / t4 
 
 
 
 
 3= 
 
 m 
 
 
 1-t 
 
 53 0) 
 
 4J 
 
 lH 
 
 
 CO 
 
 
 
 .c 
 
 
 l-l 
 
 
 3 
 
 
 
 
 ►— 
 
 a> 
 
 
 4J 
 
 s 
 
 CD 
 
 CO 
 
 
 u 
 
 
 
 0) 
 
 
 1-1 
 
 < 
 
 o 
 
 
 
 
 £3 
 
 u 
 
 
 u 
 
 
 CO 
 
 V4 
 
 
 4J 
 
 
 
 1-1 
 
 
 S 
 
 P^ 
 
 ■-:] 
 
 
 
 
 «C 
 
 o 
 
 
 a 
 
 <! '-< 
 
 Oi 
 
 01 
 
 
 c 
 
 
 
 [14 
 
 
 
 M 
 
 
 
 
 
 
 u 
 
 
 14 
 
 CO 
 
 
 > 
 
 
 (U 
 
 ^-s 
 
 
 
 c 
 
 • 
 
 < 
 
 u-l 
 
 
 
 
 
 (X< 
 
 
 4J 
 
 • 3 
 
 14-1 
 
 1-1 
 
 
 o 
 
 <u 
 
 
 u 
 
 o 
 
 ac 
 
 
 O 
 
 
 
 
 
 
 X 
 
 o c 
 
 O 
 
 c 
 
 
 c 
 
 4J 
 
 
 
 
 iH 
 
 
 •o 
 
 
 
 
 
 
 <o 
 
 
 Cd 
 
 " C 
 * < 
 
 4J 
 
 D 
 
 
 o 
 
 3 
 
 4J 
 
 
 u-l 
 
 4-1 
 
 o 
 
 3* 
 
 % 
 
 4-1 
 
 
 
 
 
 VM 
 
 
 1 
 
 > 
 
 c 
 
 f-> 
 
 
 
 1-4 
 
 
 CO 
 
 CO 
 
 o 
 
 1-t 
 
 1-4 
 
 
 
 
 
 o 
 
 
 
 O J= 
 
 V 
 
 1—1 
 
 
 c 
 
 4J 
 
 
 4) 
 
 u 
 
 )-l 
 
 (0 
 
 CO 
 
 
 
 
 
 
 
 4J 
 
 U 4J 
 
 B 
 
 0) 
 
 
 1-( 
 
 en 
 
 
 u 
 
 u 
 
 J3 4J 
 
 t-l 
 
 U 
 
 
 
 
 
 c 
 
 
 c 
 
 J3 r» 
 
 u 
 
 c 
 
 
 0) 
 
 c 
 
 
 3 
 
 X 
 
 3 D- 
 
 u 
 
 4) 
 
 
 
 
 
 o 
 
 
 0) 
 
 3 li^ 
 
 CO 
 
 )-4 
 
 
 4-1 
 
 M 
 
 
 •o 
 
 u 
 
 Q T^ 
 
 (U 
 
 > 
 
 
 
 
 
 1-1 
 
 
 > 
 
 <u 
 
 a 
 
 0) 
 
 o 
 
 
 o 
 
 
 
 (U 
 
 
 ^ 
 
 Qc: 
 
 1-1 
 
 
 
 
 
 4J 
 
 
 r-l 
 
 4J 
 
 o 
 
 u 
 
 o 
 
 
 u 
 
 J= 
 
 
 o 
 
 4-1 
 
 • O 
 
 
 c 
 
 
 
 
 
 <a 
 
 
 o 
 
 CO 
 
 • 4J . 
 
 H 
 
 >*• 
 
 
 &4 
 
 o 
 
 
 o 
 
 c 
 
 J CO 
 
 £ 
 
 o 
 
 
 
 
 
 3 
 
 
 M 
 
 V4 
 
 a 
 
 
 
 
 
 l-l 
 
 
 l-l 
 
 4) 
 
 3 
 
 o 
 
 ^ 
 
 
 
 
 
 fl 
 
 4J 
 
 
 4J 
 
 T3 
 
 1-4 
 
 c 
 
 4J 
 
 OJ 
 
 CO 
 
 4J 
 
 P>4 
 
 > 
 
 • d 
 
 u 
 
 
 4J 
 
 
 
 
 (S 
 
 /-N l-l 
 
 u-l 
 
 c 
 
 " 4) 
 
 CO 
 
 lU 
 
 V4 
 
 c 
 
 (U 
 
 V4 
 
 
 1-1 
 
 Q CO 
 
 u-l 
 
 - 60 
 
 >4 
 
 
 
 UJ 
 
 > 
 
 • O 
 
 o 
 
 <u 
 
 l-l 4.1 
 
 o 
 
 <u 
 
 o 
 
 t4 
 
 CO 
 
 o 
 
 T3 
 
 o 
 
 ^ X 
 
 
 c 
 
 o 
 
 
 
 —J 
 
 b3 
 
 a d. 
 
 
 a 
 
 (U c 
 
 -H 
 
 > 
 
 a 
 
 u 
 
 4) 
 
 o- 
 
 c 
 
 C/3 
 
 
 l-l 
 
 
 1-1 
 
 ex. 
 
 
 
 J^ 
 
 
 (U 
 
 c a) 
 
 c 
 
 c 
 
 0) (U 
 
 Cl£ 
 
 
 <u 
 
 nj 
 
 oi 
 
 (U 
 
 to 
 
 
 C T3 
 
 CO 
 
 ^•* 
 
 -o 
 
 4) 
 
 
 
 H* 
 
 "O 
 
 4J 
 
 M BS 
 
 O 
 
 o 
 
 •H tn 
 
 O 
 
 u 
 
 OJ 
 
 s 
 
 
 a; 
 
 
 :^ 
 
 3 0) 
 
 > 
 
 f— 1 
 
 ■—1 
 
 C3d 
 
 
 
 
 c 
 
 CQ 
 
 
 1-1 
 
 CJ 
 
 U lU CO 
 
 l-l 
 
 CU 
 
 
 
 • 
 
 
 en 
 
 M 
 
 o s: 
 
 o 
 
 ^ 
 
 <0 
 
 
 
 
 
 CO 
 
 1.1 
 
 •> u 
 
 j-i 
 
 
 <0 V4 )-J 
 
 o 
 
 T3 
 
 4-1 
 
 U.4 
 
 H 
 
 4-1 
 
 3 
 
 o 
 
 u to 
 
 E 
 
 
 Q. 
 
 4-1 
 
 
 
 
 
 XJ 
 
 M U 
 
 03 
 
 c 
 
 04 a. lu 
 
 1-1 
 
 
 u 
 
 o 
 
 
 o 
 
 u 
 
 4J 
 
 CQ 1-t 
 
 <u 
 
 4-1 
 
 C/3 
 
 O 
 
 
 
 
 c 
 
 C 
 
 cj to 
 
 O 
 
 1-1 
 
 <u 
 
 M 
 
 c 
 
 CO 
 
 
 • 
 
 CO 
 
 (0 
 
 CO 
 
 l-l 
 
 Bi 
 
 U 
 
 
 <0 
 
 
 
 
 M 
 
 (U 
 
 M 1-1 
 
 i-J 
 
 0) 
 
 ' u c 
 
 O 
 
 CO 
 
 M 
 
 > 
 
 H 
 
 h 
 
 u 
 
 l-l 
 
 • XI 
 
 
 o 
 
 • 
 
 u 
 
 
 
 
 1-1 
 
 o 
 
 2 -U 
 
 .-1 
 
 J-l 
 
 Oi 0) ^ 
 
 u 
 
 > 
 
 4J 
 
 •a 
 
 
 4J 
 
 CO 
 
 o 
 
 J 3 
 
 U 
 
 a 
 
 o 
 
 4.1 
 
 
 
 
 n 
 
 c 
 
 o c 
 
 & 
 
 o 
 
 D. ao 
 
 o 
 
 
 c 
 
 3 
 
 • 
 
 c 
 
 D. 
 
 -O 
 
 O. 
 
 o 
 
 0) 
 
 
 c 
 
 
 
 
 OJ 
 
 o 
 
 M O 
 
 a 
 
 u 
 
 • CO c 
 
 1^ 
 
 • 
 
 o 
 
 4J 
 
 H 
 
 o 
 
 (X 
 
 CO 
 
 • C 
 
 •o 
 
 Q^ 
 
 • 
 
 o 
 
 
 
 
 O 
 
 CJ 
 
 w o 
 
 < 
 
 p^ 
 
 CiJ Pli U 
 
 X 
 
 •^ 
 
 CJ 
 
 Cfl 
 
 >— ' 
 
 CJ 
 
 < 
 
 ij 
 
 Z 3 
 
 o 
 
 
 CO 
 
 o 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 a: 
 —J 
 
 00 
 o 
 o 
 
 ON 
 
 o 
 o 
 u 
 
 o 
 
 i 
 
 
 
 l-l 
 l-l 
 o 
 u 
 
 
 1-4 
 
 o 
 
 
 
 
 
 
 II. 5-2 
 
X 
 
 u 
 
 V 
 
 60 
 O 
 
 O 
 
 o 
 
 eg 
 
 0) 
 CO 
 
 o 
 
 M 
 
 C9 
 P« 
 
 (1) 
 t>0 
 lU 
 
 O 
 
 09 
 
 0) 
 bO 
 
 o 
 o 
 
 .i£ 
 
 J^ 
 
 h 
 
 Wl 
 
 CO 
 
 CS 
 
 tu 
 
 PU 
 
 9) 
 
 V 
 
 CsO 
 
 00 
 
 (U 
 
 a> 
 
 i-l 
 
 f-i 
 
 f-l 
 
 f-i 
 
 o 
 
 o 
 
 o 
 
 u 
 
 
 --0 
 
 
 til 
 
 >- 
 
 ■ji 
 
 rn 
 
 V3 
 
 Q_ 
 
 '-J 
 
 •a: 
 
 H 
 
 cc: 
 
 
 CD 
 
 c 
 
 <=> 
 
 
 
 >H 
 
 — 1 
 
 
 CQ 
 
 o 
 
 oa 
 
 ^ 
 
 C-3 
 
 g 
 
 c 
 o 
 
 o 
 
 u 
 o 
 
 c 
 
 > 
 
 I— I 
 o 
 
 o 
 
 u 
 
 (u 
 
 IJ 
 
 o 
 J= 
 
 o 
 
 UJ 
 
 ■ ^ CI > 
 
 U ^ O 
 
 rj (/) C 
 
 Oi •'-I o 
 
 • JO 
 
 Di D In 
 
 C O 
 
 • c -J 
 
 o 
 
 :3 
 o 
 
 > 
 
 C 
 
 00 
 
 c 
 
 r- ij 
 
 C/5 o 
 
 c 
 . o 
 
 CO u 
 
 i-l 
 
 
 a. 
 
 o 
 
 c 
 
 u 
 
 u 
 
 o 
 
 o 
 
 (^ 
 
 •r-l 
 CO 
 
 o 
 
 JZ 
 
 c 
 
 03 
 
 (fi 
 
 0) 
 
 r-( 
 
 •r^ 
 
 G 
 
 n) 
 
 P^ 
 
 t/) 
 
 u 
 
 
 D 
 
 •M 
 
 c 
 
 C/3 
 
 H 
 
 •H 
 
 
 dJ 
 
 
 f— ( 
 
 x: 
 
 c 
 
 O 
 
 u 
 
 o 
 
 c 
 
 
 •H 
 
 (3 
 
 ij 
 
 4-1 
 
 a 
 
 >-( 
 
 u 
 
 o 
 
 (0 
 
 3 
 
 u 
 
 w 
 
 XI 
 
 a 
 
 c 
 
 
 ^^ 
 
 M 
 
 <y 
 
 
 c 
 
 B 
 
 -r-l 
 
 O 
 
 u 
 
 M 
 
 o 
 
 U-l 
 
 D 
 
 
 1— 1 
 
 > 
 
 U. 
 
 u 
 
 
 GJ 
 
 u 
 
 > 
 
 
 
 O 
 
 U-l 
 
 C) 
 
 
 0) 
 
 (/) 
 
 cd 
 
 T) 
 
 
 O 
 
 c 
 
 x: 
 
 •r-( 
 
 jj 
 
 01 
 
 0) 
 
 u 
 
 r. 
 
 o 
 
 
 u 
 
 f— 1 
 
 P-. 
 
 ni 
 
 
 cr 
 
 r* 
 
 D 
 
 71 
 
 u-. 
 
 •r-l 
 
 -H 
 
 u- 
 
 U 
 
 
 i-l 
 
 T3 
 
 c 
 
 C 
 
 OJ 
 
 a 
 
 u 
 
 
 
 o 
 
 -J 
 
 u 
 
 o 
 
 a 
 
 
 u 
 
 > 
 
 *_t 
 
 TJ 
 
 c 
 
 3 
 
 CI 
 
 u 
 
 CI 
 
 CO 
 
 ^ 
 
 
 D 
 
 < 
 
 u 
 
 o 
 
 o 
 
 o 
 
 T3 
 
 a 
 
 CO 
 
 u 
 to 
 o 
 
 3 
 
 x: 
 u 
 
 dj 
 -o 
 .-( 
 
 o 
 
 x: 
 
 3 
 
 ca 
 
 o 
 
 OS iJ 
 
 00 
 
 O 
 
 u 
 
 ai 
 C 
 
 •H 
 
 u 
 
 00 
 o 
 
 o 
 
 CM 
 
 O 
 r-i 
 I 
 
 o> 
 cr. 
 
 
 O ti3 X 
 
 X -r^ U 
 
 ^ X :j 
 
 c 
 • o 
 
 •::i fa 
 
 T3 ^-^ 
 C 
 
 TO l-l 
 
 •H >, 
 
 i-l c 
 
 nj CO 
 
 C 
 
 o 
 
 o 
 
 UJ 
 
 o 
 
 ml 
 
 
 01 
 
 E 
 
 c 
 
 0) 
 
 •H 
 
 i-l 
 
 t: 
 
 tr. 
 
 QJ 
 
 > 
 
 01 
 
 C/1 
 
 ^ 
 
 
 o 
 
 > 
 
 )-l 
 
 Vi 
 
 P-. 
 
 0) 
 
 
 > 
 
 « 
 
 o 
 
 M 
 
 o 
 
 M 
 
 0) 
 
 
 cd 
 
 cn 
 
 
 t« 
 
 < 
 
 OJ 
 
 CL, 
 
 (J 
 
 M 
 
 w. 
 
 
 c 
 
 0) 
 
 o 
 
 t-l 
 
 o 
 
 r3 
 
 
 o 
 
 c 
 
 CO 
 
 o 
 
 
 ■r-( 
 
 X 
 
 i-l 
 
 u 
 
 •r^ 
 
 c 
 
 Vj 
 
 01 
 
 i-l 
 
 CC 
 
 D 
 
 
 '^. 
 
 r3 
 
 CJ 
 
 ._j 
 
 1-1 
 
 .— ' 
 
 o 
 
 n 
 
 ■— 
 
 i_) 
 
 a 
 
 i/i 
 
 C/l 
 
 c 
 
 ■r-l 
 
 H-l 
 
 e; 
 
 
 o 
 
 -a 
 
 
 
 
 
 
 n: 
 
 c 
 
 
 u 
 
 c 
 
 o 
 
 ts. 
 
 i-l 
 
 "H 
 
 tr. 
 
 V) 
 
 0) 
 
 o 
 
 3 
 
 Q 
 
 o 
 
 D. 
 O 
 
 c 
 o 
 
 o 
 
 a 
 o 
 u 
 u 
 o 
 
 N 
 
 m 
 
 Cfl 
 
 3 
 
 CJ 
 
 0) 
 
 CO 
 
 rr 
 
 Ea 
 
 
 g 
 
 -n 
 
 
 OJ 
 
 V4 
 
 Di 
 
 OJ 
 
 
 .:i: 
 
 nj 
 
 o 
 
 
 V 
 
 c 
 
 >-A 
 
 •r^ 
 
 CQ 
 
 u 
 
 
 o 
 
 O 
 
 u_< 
 
 
 OT 
 
 , 
 
 C 
 
 2C 
 
 nj 
 
 
 Sj 
 
 ^ 
 
 H 
 
 v/ 
 
 o 
 
 ^ 
 
 o 
 
 c 
 
 (T) 
 
 X 
 
 c 
 
 -r-t 
 
 01 
 
 a 
 
 iJ 
 
 ■r-l 
 
 o 
 
 ^ 
 
 r—i 
 
 1-. 
 
 r3 
 
 X 
 
 D. 
 
 — 
 
 3 
 
 O 
 
 
 Q. 
 
 to 
 
 • 
 
 c 
 
 I-l 
 
 Q 
 
 13 
 
 o 
 
 CM 
 
 O 
 
 10 
 
 II. 5-3 
 
X 
 
 ^ 
 
 ,i<5 
 
 ^ 
 
 ^ 
 
 J4 
 
 ^ 
 
 u 
 
 VI 
 
 14 
 
 u 
 
 14 
 
 n 
 
 to 
 
 a 
 
 «0 
 
 to 
 
 « 
 
 to 
 
 (U 
 
 P^ 
 
 Pn 
 
 P4 
 
 P4 
 
 PL| 
 
 Q) 
 
 <o 
 
 0) 
 
 . at 
 
 0) 
 
 .0) 
 
 bO 
 
 U) 
 
 00 
 
 60 
 
 60 
 
 60 
 
 0) 
 
 <u 
 
 0) 
 
 <U 
 
 0) 
 
 0) 
 
 r-t 
 
 i-i 
 
 r-l 
 
 t-l 
 
 •-I 
 
 1-1 
 
 i-l 
 
 r-l 
 
 1-1 
 
 1-1 
 
 i-« 
 
 1-1 
 
 o 
 
 O 
 
 o 
 
 o 
 
 O 
 
 O 
 
 o 
 
 u 
 
 u 
 
 o 
 
 o 
 
 O 
 
 ^ 
 
 ^ 
 
 M 
 
 u 
 
 CO 
 
 to 
 
 ru 
 
 P4 
 
 « 
 
 0) 
 
 60 
 
 60 
 
 0) 
 
 <U 
 
 t-l 
 
 1-1 
 
 r-l 
 
 1-1 
 
 o 
 
 o 
 
 o 
 
 o 
 
 C3 
 
 CZ3 
 CO 
 
 
 a; 
 
 
 C3 
 
 
 
 
 ^— 
 
 
 «« 
 
 
 <_J 
 
 CO 
 
 » 
 
 w 
 
 cn 
 
 to 
 
 r-> 
 
 CO 
 
 w 
 
 o_ 
 
 Pi 
 
 o 
 
 fn 
 
 Di 
 
 CO 
 
 c 
 
 3 U 
 
 o o 
 ^< E 
 
 iJ CO 
 
 . & 
 
 (T3 
 CO 
 
 ex 
 a 
 c« 
 
 OJ 
 i-l 
 
 CO 
 ■r-l 
 
 3 
 
 O 
 O. 
 
 dl 
 to Di 
 
 r-l to 
 
 0) u 
 
 © u 
 
 O 
 
 6 
 
 D 
 -t3 
 
 C 
 
 to 
 p 
 o 
 e 
 
 CO 
 
 e 
 
 D 
 C 
 
 to 
 o 
 
 e 
 
 .J w 
 
 z S 
 
 to 
 C 
 o 
 1-1 
 JJ 
 eg 
 u 
 1-1 
 tM 
 
 tI 
 
 u 
 a 
 
 CO 
 
 Q 
 
 •C 
 
 C 
 
 to 
 
 <^ 
 
 c 
 
 H 
 
 e 
 
 D 
 -O 
 C 
 
 to 
 >-( 
 o 
 
 E 
 a 
 
 iJ CO 
 
 a a; 
 
 o 
 a. 
 <u 
 oi 
 
 o 
 
 CO 
 
 u 
 
 c 
 o 
 o 
 
 to 
 
 D 
 C 
 
 c 
 <: 
 
 en 
 
 w 
 
 V-l - 
 
 o 
 
 to oi 
 o 
 
 D- to 
 
 -r-l V-l 
 C 
 
 • o 
 CJ o 
 
 a. 
 u 
 o 
 
 CJ 
 
 to 
 
 <-< 
 
 to • 
 
 •r^ M 
 
 e o 
 0) a. 
 x: <u 
 
 —I o 
 to to 
 
 <=> 
 
 
 
 
 
 
 
 
 
 2c: 
 
 
 
 
 
 
 
 
 
 
 i-i 
 
 CN 
 
 fn 
 
 <t 
 
 lO 
 
 vO 
 
 r-- 
 
 00 
 
 —J 
 
 CM 
 
 CM 
 
 CM 
 
 Cvl 
 
 CM 
 
 CM 
 
 CM 
 
 CM 
 
 —J 
 
 O 
 
 O 
 
 o 
 
 o 
 
 O 
 
 O 
 
 q 
 
 O 
 
 CJ 
 
 u 
 
 W 
 
 u 
 
 bl 
 
 U 
 
 U3 
 
 Pd 
 
 u 
 
 II. 5-4 
 

 to 
 
 
 CJ 
 
 >- 
 
 C/5 
 
 Qu 
 
 CO 
 
 It 
 
 o 
 
 OS 
 
 Di 
 
 en 
 
 P-> 
 
 o 
 
 
 CO 
 
 >> 
 
 o 
 
 CO 
 
 H 
 
 r > 
 
 ;S 
 
 Ol. 
 
 o 
 
 I I 
 
 0) 
 CO 
 
 o 
 o 
 
 In 
 
 U 
 
 0) 
 
 a 
 
 o 
 o 
 
 c 
 
 CO 
 (U 
 
 c 
 
 CO 
 0) 
 
 OJ 
 
 •r4 
 
 •d 
 
 a 
 
 o 
 
 1^ 
 
 o 
 
 o 
 n 
 o 
 
 U 
 
 (0 
 
 00 
 
 o 
 o 
 
 m 
 
 •r4 
 
 w 
 >» 
 
 r-l 
 O 
 
 Si 
 
 c 
 
 J3 
 
 to 
 
 CO 
 > 
 
 O 
 
 > 
 •l-l 
 
 u 
 
 4) 
 00 
 
 o 
 u 
 
 ^ 
 
 ^ 
 
 )-l 
 
 u 
 
 to 
 
 to 
 
 P4 
 
 fvl 
 
 (U 
 
 (U 
 
 tsO 
 
 00 
 
 0) 
 
 0) 
 
 l-l 
 
 t-l 
 
 l-< 
 
 >-l 
 
 o 
 
 o 
 
 u 
 
 o 
 
 > o 
 
 •r^ u 
 
 1-1 (Xi 
 (0 
 
 —I ^ 
 
 & 
 
 -'-. 
 
 > 
 
 ii 
 
 o 
 
 •>-y 
 
 1 — 1 
 
 
 o 
 
 QJ 
 
 c 
 
 i-M 
 
 ^c 
 
 ra 
 
 o 
 
 — ^ 
 
 01 
 
 
 H 
 
 CO 
 
 ■n 
 
 
 o 
 
 • 
 
 o 
 
 X- 
 
 ti_ 
 
 0) 
 
 u 
 o 
 
 CO 
 
 .•fl 
 
 O 
 P4 
 
 VI 
 
 o 
 tl-l 
 
 o 
 
 jc 
 o 
 u 
 
 ex 
 o 
 u 
 a 
 o 
 
 CO 
 M 
 
 "^ C 
 
 X) O 
 
 QJ -H 
 
 QC i-> 
 
 •^ O 
 
 3 
 
 x: -a 
 
 U5 o 
 
 U-. PL, 
 
 (TJ O 
 
 00 V-i 
 
 C iJ 
 
 CO <N 
 
 c 
 o 
 
 Q 
 
 CO 
 
 c 
 to 
 o. 
 S 
 o 
 o 
 
 o 
 u 
 
 CO 
 
 o 
 
 CJ 
 
 en 
 
 o 
 
 u 
 
 u 
 
 <u 
 
 > 
 
 c 
 
 3 
 
 .a 
 
 E 
 3 
 
 o 
 o 
 
 
 
 c 
 
 U 
 
 
 o 
 
 
 o 
 
 11 
 
 
 :n 
 
 1-) 
 
 -r-l 
 
 
 u 
 
 ;/l 
 
 Vj 
 
 i-J 
 
 « 
 
 Vj 
 
 LU 
 
 o 
 
 ro 
 
 U 
 
 O 
 
 v_^ 
 
 n. 
 
 4-1 
 
 a 
 
 CL 
 
 
 0) 
 
 C 
 
 -o 
 
 a 
 
 Ul 
 
 Qi 
 
 a 
 
 r-l 
 
 ci 
 
 o 
 
 
 h 
 
 
 
 
 •r-l 
 
 u 
 
 Vj 
 
 c 
 
 i-i 
 
 U 
 
 u 
 
 il 
 
 .M 
 
 o 
 
 1^. 
 
 a 
 
 U-, 
 
 Vi 
 
 o 
 
 ,r-i 
 
 t-i 
 
 
 3 
 
 Vj 
 
 t. J 
 
 i_) 
 
 r" 
 
 "■^ 
 
 i-l 
 
 
 c 
 
 (/) 
 
 
 c 
 
 ■ 
 
 o 
 
 ■rJ 
 
 • 
 
 o 
 
 a 
 
 u 
 
 Lu 
 
 J 
 
 t_) 
 
 o 
 to 
 
 II. -5 -5 
 
>- 
 ■a: 
 
 Cxi. 
 C3 
 
 CO 
 CO 
 
 cu 
 
 CO 
 Cd 
 C/3 
 CO 
 
 w 
 o 
 o 
 
 >^ 
 
 o 
 
 H 
 
 O 
 
 ^ ;3 
 
 »— 
 
 
 
 
 
 
 
 
 
 
 
 
 C_3 
 
 
 
 
 
 
 
 
 
 
 
 
 •KC 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 X 
 
 
 X 
 
 
 
 
 
 t/> 
 
 
 
 
 
 
 
 
 
 
 
 
 on 
 
 
 
 
 
 
 
 
 
 
 
 
 -^c 
 
 
 
 
 
 
 
 
 
 
 
 
 S£ 
 
 
 
 
 
 
 
 
 
 
 
 
 1 
 
 
 
 
 
 
 
 
 
 
 
 
 IJL. 
 
 C3 
 
 
 
 
 
 
 
 >{ 
 
 
 
 
 
 o: 
 
 
 
 
 
 
 
 
 
 
 
 
 £_■> 
 
 
 
 
 
 
 
 
 
 
 
 
 I^ 
 
 
 
 
 
 
 
 
 
 
 
 
 «i_ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Jrf 
 
 
 a 
 
 
 a 
 
 
 JiS 
 
 a 
 
 
 a, 
 
 
 
 u 
 
 
 u 
 
 
 u 
 
 
 n 
 
 u 
 
 
 u 
 
 
 35 
 C3 
 
 CO 
 
 
 ca 
 
 
 CO 
 
 
 CO 
 
 CO 
 
 
 CO 
 
 
 P4 
 
 
 PLl 
 
 
 P4 
 
 
 p< 
 
 fa 
 
 
 fa 
 
 
 y— 
 
 0) 
 
 
 0) 
 
 
 (U 
 
 
 0) 
 
 (U 
 
 
 0) 
 
 
 <«£ 
 
 00 
 
 
 bO 
 
 
 bO 
 
 
 M 
 
 60 
 
 
 60 
 
 
 C^ 
 
 <u 
 
 
 0) 
 
 
 0) 
 
 
 Q) 
 
 (U 
 
 
 (U 
 
 
 C3 
 
 r-l 
 
 
 t-l 
 
 
 l-l 
 
 
 r-l 
 
 r-l 
 
 
 r-l 
 
 
 .^ 
 
 f-t. 
 
 
 t-l 
 
 
 r-l 
 
 
 r-l 
 
 r-l 
 
 
 r-l 
 
 
 
 O 
 
 
 o 
 
 
 O 
 
 
 O 
 
 O 
 
 
 o 
 
 
 
 o 
 
 
 o 
 
 
 U 
 
 
 CJ 
 
 O 
 
 
 o 
 
 
 
 
 
 
 
 (U 
 
 
 
 
 
 
 
 
 
 
 
 
 
 > 
 
 
 
 
 
 
 4J 
 
 
 
 
 
 
 
 
 ■rl 
 
 
 
 
 
 
 c 
 
 
 
 
 
 
 
 
 4J 
 
 
 
 
 
 
 0) 
 
 
 
 
 
 
 ^ 
 
 
 Ci 
 
 
 
 
 
 
 > 
 
 
 
 
 
 
 4J 
 
 
 (U 
 
 
 
 
 
 
 r^ 
 
 
 
 
 
 
 •H 
 
 
 l-l 
 
 
 
 
 
 
 o 
 
 
 
 2E 
 
 
 
 > 
 
 
 <u 
 
 
 
 
 
 
 CO 
 
 
 
 C3 
 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 1— 
 
 
 
 
 
 !;► 
 
 
 
 4J 
 
 
 
 
 CD 
 
 
 ■ 
 
 CO 
 
 
 
 •H 
 
 
 ,o 
 
 
 
 CO 
 
 
 
 
 u-l 
 
 
 
 
 
 
 fa 
 
 
 
 
 
 n 
 
 
 
 
 o 
 
 
 
 —J 
 
 
 
 
 
 0) 
 
 
 
 4J 
 
 
 
 
 
 
 
 CO 
 
 
 
 U^ 
 
 
 4-) 
 
 
 
 C 
 
 
 
 
 to 
 
 
 
 
 
 o 
 
 
 CO 
 
 ^^ 
 
 
 (U 
 
 
 
 
 u 
 
 
 
 
 
 
 
 l-l 
 
 0) 
 
 
 o 
 
 
 
 
 c 
 
 
 
 
 
 
 c 
 
 
 i-i 
 
 ■U 
 
 
 c 
 
 
 
 
 0) 
 
 
 
 
 
 
 o 
 
 
 C 
 
 3 
 
 
 o 
 
 
 
 
 3 
 
 
 
 
 
 
 1-1 
 
 
 ■<u 
 
 4J 
 
 
 CJ 
 
 
 
 
 r-l 
 
 
 
 
 
 
 u 
 
 
 o 
 
 •r4 
 
 
 
 /-N 
 
 
 
 14-1 
 
 
 
 
 
 
 CO 
 
 
 c 
 
 4-1 
 
 
 c 
 
 >> 
 
 
 
 14-1 
 
 
 
 
 
 
 *J 
 
 
 o 
 
 CO 
 
 
 •r4 
 
 c 
 
 
 
 Ui 
 
 
 
 
 
 
 c 
 
 
 o 
 
 c 
 
 
 0) 
 
 CO 
 
 
 
 
 
 
 
 
 
 (U 
 
 
 
 M 
 
 
 4J 
 
 a 
 
 
 
 m 
 
 
 
 
 
 
 a 
 
 
 c 
 
 — 1 
 
 r- 
 
 
 o 
 
 6 
 o 
 
 ^_^ ' 
 
 
 CO 
 
 
 
 
 
 
 0) 
 
 
 0) 
 
 o 
 
 
 fa 
 
 CJ 
 
 o 
 
 
 
 C) 
 
 
 
 
 
 
 fa 
 
 
 ■u 
 o 
 
 u 
 
 CO 
 
 
 £ 
 
 60 
 
 CO 
 
 
 
 O 
 t-l 
 
 
 
 
 /-> 
 
 
 > 
 
 /-N 
 
 ;- 
 
 g; 
 
 
 CO 
 
 C 
 
 
 
 
 fa 
 
 4J 
 
 
 
 
 >. 
 
 
 J3 
 
 >. 
 
 P-I 
 
 CO' 
 
 
 1-1 
 
 •r^ 
 
 
 
 
 
 c 
 
 
 
 
 u 
 
 
 
 iJ 
 
 
 01 
 
 
 fa 
 
 V4 
 
 g 
 
 
 
 e 
 
 CO 
 
 
 
 
 •H 
 
 
 <u 
 
 -r-l 
 
 .c 
 
 a! 
 
 
 
 cu 
 
 <u 
 
 
 
 o 
 
 I-l 
 
 
 
 
 U) 
 
 
 iJ 
 
 (0 
 
 CO 
 
 
 
 y-i 
 
 (U 
 
 4J 
 
 
 
 u 
 
 fa 
 
 
 
 
 O 
 
 u 
 
 
 (0 
 
 u 
 
 •H 
 
 •o 
 
 
 o 
 
 c 
 
 CO 
 
 
 
 M-l 
 
 
 
 
 
 r^ 
 
 0) 
 
 
 n 
 
 <u 
 
 fa 
 
 M 
 
 
 
 •r4 
 
 > 
 
 
 
 
 0) 
 
 
 
 yo 
 
 c^ 
 
 > 
 
 
 XJ 
 
 > 
 
 
 O 
 
 
 OJ 
 
 00 
 
 CO 
 
 
 
 :^ 
 
 4-1 
 
 
 
 C3 
 
 t-( 
 
 •H 
 
 
 (1) 
 
 •H 
 
 c 
 
 >4-l 
 
 o 
 
 C4-I 
 
 
 3 
 
 
 CJC 
 
 
 
 Vi 
 
 CO 
 
 t-4 
 
 
 
 3; 
 
 3 
 
 
 o 
 
 p 
 
 
 CO 
 
 
 4J 
 
 H 
 
 c 
 
 
 
 > 
 
 4J 
 
 
 
 ►— 
 
 
 
 c 
 
 
 I-l 
 
 4-1 
 
 
 o 
 
 l-H -O 
 
 1-1 
 
 
 
 o 
 
 c 
 
 
 
 =3 
 
 <g 
 
 
 o 
 
 « 
 
 0) 
 
 CO 
 
 
 CO 
 
 S C 
 
 N 
 
 
 
 o 
 
 0) 
 
 
 
 -x 
 
 
 
 o 
 
 •r-l 
 
 J3 
 
 > 
 
 0) 
 
 ■^^ 
 
 
 •4-1 
 
 3 
 
 ^-^ (0 
 
 1-1 
 
 u 
 
 
 
 4) 
 OS 
 
 o 
 
 c 
 
 ^'^ 
 
 S 
 
 
 
 
 
 c 
 
 en 
 
 S 
 
 ►-5 
 
 c 
 
 
 r; 
 
 p- -C 
 
 o 
 
 ^-\ 
 
 
 
 o 
 
 
 
 3 
 
 
 1-1 
 
 e 
 
 P 
 
 
 o 
 
 
 CO 
 
 0) o 
 
 T3 
 
 • 
 
 
 CO 
 
 o 
 
 .•a 
 
 
 
 rH 
 
 
 0) 
 
 C/) 
 
 1— 1 
 
 0) 
 
 CO 
 
 
 s 
 
 CO t-l 
 
 o 
 
 o 
 
 
 4-1 
 
 
 
 
 O 
 
 
 4-1 
 
 •H 
 
 o 
 
 13 
 
 H 
 
 
 
 1-1 n) 
 
 0) 
 
 c 
 
 
 o 
 
 c: 
 
 < 
 
 
 
 U 
 
 
 o 
 
 c 
 
 CJ 
 
 •r4 
 
 <u 
 
 
 (U 
 
 Sj (U 
 
 Q 
 
 M 
 
 
 3 
 
 1-1 
 
 
 
 >-* 
 
 
 u 
 
 CD 
 
 — ' 
 
 u 
 
 & 
 
 
 rC 
 
 CO <« 
 
 
 
 
 •a 
 
 0) 
 
 
 
 
 
 
 Pj 
 
 M 
 
 
 o 
 
 e 
 
 
 *J 
 
 fa ii 
 
 •a 
 
 n 
 
 
 o 
 
 4-1 
 
 CO 
 
 
 
 01 
 
 
 
 u 
 
 3 
 
 3 
 
 CO 
 
 
 
 Pi 
 
 c 
 
 CO 
 
 
 u 
 
 O 
 
 CO 
 
 
 
 l-l 
 
 
 jr 
 
 o 
 
 S^ 
 
 I-l 
 
 K 
 
 
 c 
 
 • 
 
 CO 
 
 cU 
 
 
 fa 
 
 t-l 
 
 X 
 
 
 
 C 
 
 0) 
 
 
 U3 
 
 o 
 
 u 
 
 fa 
 
 
 
 1-) 
 
 cc; 
 
 
 1-1 
 
 
 
 fa 
 
 H 
 
 
 
 O 
 
 o 
 
 
 •r-l 
 
 V-i 
 
 
 
 CO 
 
 • 
 
 
 
 CO 
 
 t^ 
 
 U 
 
 
 t^ 
 
 
 
 
 ^ 
 
 »J 
 
 J-l 
 
 fa 
 
 o 
 
 ' *J 
 
 01 
 
 0) 
 
 :3 
 
 u 
 
 CO 
 
 • CO 4-1 
 
 u 
 
 4-) 
 
 u 
 
 PQ 
 
 rC 
 
 ••-^ 
 
 
 
 4J 
 
 
 V4 
 
 
 ■rH 
 
 <; u 
 
 £. 
 
 c 
 
 
 U 
 
 u 
 
 W W M 
 
 0) 
 
 CO 
 
 u 
 
 
 CO 
 
 
 u 
 
 
 « 
 
 • 
 
 o 
 
 U-l 
 
 ,£^ 
 
 o 
 
 u 
 
 o 
 
 •» 
 
 o 
 
 C 
 
 ^ O 
 
 > 
 
 3 
 
 o 
 
 O 
 
 •rt 
 
 r-l 
 
 OJ 
 
 UJ 
 
 i-1 
 
 < 
 
 c 
 
 o 
 
 
 - a 
 
 
 PQ 
 
 c 
 
 a 
 
 U 
 
 Ou 
 
 o 
 
 X) 
 
 D. 
 
 4J 
 
 fa 
 
 r-l 
 
 o. 
 
 
 c 
 
 
 <u 
 
 
 o 
 
 CO <u 
 
 6C 
 
 
 •r-l 
 
 0) 
 
 > 
 
 C 0) 4) 
 
 o 
 
 c 
 
 OJ 
 
 
 
 •r-l 
 
 CO 
 
 a; 
 
 • 
 
 OS 
 
 c 
 
 •r-l 
 
 ^ CtJ 
 
 C 
 
 U-l 
 
 CO 
 
 OS 
 
 I— 1 
 
 S o oJ 
 
 OJ 
 
 M 
 
 OS 
 
 CO 
 
 c 
 
 "J 
 
 fa 
 
 £ 
 
 o 
 
 
 o 
 
 iJ 
 
 (U 
 
 ■r-l 
 
 O 
 
 6 
 
 
 o 
 
 O 1-1 
 
 OS 
 
 
 
 0) 
 
 o 
 
 2 
 
 
 
 J_l 
 
 
 J-l 
 
 •H 
 
 ?^ 
 
 O iJ 
 
 1—1 
 
 
 <y 
 
 4-1 
 
 CO 
 
 (-1 )-l 4-) 
 
 
 C 
 
 4J 
 
 .c 
 
 •H 
 
 
 J= 
 
 
 0) 
 
 ^ 
 
 o 
 
 iJ 
 
 f— 1 
 
 ai o 
 
 I— 1 
 
 1—1 
 
 )-i 
 
 o 
 
 
 CG fa O 
 
 4-) 
 
 CO 
 
 o 
 
 o 
 
 4J 
 
 O 
 
 u 
 
 
 fa 
 
 3 
 
 CO 
 
 u 
 
 o 
 
 CO 
 
 O 
 
 (D 
 
 •r-l 
 
 CO 
 
 U-l 
 
 CO 
 
 c 
 
 CO 
 
 CO 
 
 CO 
 
 o 
 
 
 u 
 
 
 
 x: 
 
 u 
 
 D 
 
 CJ 
 
 • u 
 
 U 
 
 > 
 
 (U 
 
 u 
 
 o 
 
 • • u 
 
 <u 
 
 1-1 
 
 u 
 
 o 
 
 CO 
 
 • 
 
 CO 
 
 
 £ 
 
 u 
 
 u 
 
 ■o 
 
 i-l 
 
 < J-l 
 
 4-J 
 
 o 
 
 ca 
 
 4J 
 
 
 -3 <: 4-t 
 
 > 
 
 4-) 
 
 4J 
 
 M 
 
 IJ 
 
 S 
 
 0) 
 
 
 CO 
 
 
 c 
 
 o 
 
 o 
 
 c 
 
 c 
 
 E 
 
 
 c 
 
 0) 
 
 c 
 
 r-l 
 
 U 
 
 c 
 
 a 
 
 4_) 
 
 
 CO 
 
 
 •1-1 
 
 • 
 
 o 
 
 ^1 
 
 u 
 
 • o 
 
 o 
 
 0) 
 
 • 
 
 o 
 
 (0 
 
 • • o 
 
 o 
 
 <_ 
 
 o 
 
 P- 
 
 x: 
 
 
 <u 
 
 
 fa 
 
 < 
 
 o 
 
 Pm 
 
 P-i 
 
 o o 
 
 CJ 
 
 Pi 
 
 >A 
 
 u 
 
 o 
 
 Z i-j tj 
 
 CO 
 
 
 u 
 
 < 
 
 w 
 
 H 
 
 OS 
 
 ci 
 
 
 
 
 
 
 
 
 
 
 
 
 ac 
 
 
 
 
 
 
 
 
 
 
 
 
 
 m 
 
 
 vO 
 
 
 r«. 
 
 
 00 
 
 <JN 
 
 
 o 
 
 
 rd 
 
 o 
 
 
 CO 
 
 
 *n 
 
 
 cn 
 
 en 
 
 
 <f 
 
 
 acC 
 
 o 
 
 
 o 
 
 
 o 
 
 
 o 
 
 o 
 
 
 o 
 
 
 (-> 
 
 w 
 
 
 u 
 
 
 (d 
 
 
 w 
 
 ux 
 
 
 U3 
 
 
 II. 5-6 
 
a. 
 oi. 
 
 ca 
 
 QQ 
 
 to 
 
 CO 
 CO 
 
 o 
 o 
 
 04 
 
 1 — 
 
 
 
 
 ■"" 
 
 
 
 
 *"" 
 
 
 
 
 ""^ 
 
 
 
 
 
 CJ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 -=c 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 or: 
 
 ;«: 
 
 
 
 
 
 
 X 
 
 
 X 
 
 
 
 
 X 
 
 
 
 
 oo 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CXD 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 5 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CD 
 
 
 
 X 
 
 
 
 
 
 
 
 
 X 
 
 
 
 
 
 
 cc 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 <_> 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 SE 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 j«: 
 
 
 
 
 .« 
 
 
 J<5 
 
 
 ,« 
 
 
 ^ 
 
 
 
 
 
 w 
 
 
 M 
 
 
 
 
 Vi 
 
 
 >.4 
 
 
 ki 
 
 
 u 
 
 
 
 
 
 CO 
 
 
 n) 
 
 
 
 
 n) 
 
 
 CO 
 
 
 CO 
 
 
 C9 
 
 
 
 
 f^ 
 
 
 fu 
 
 
 
 
 PU 
 
 
 PM 
 
 
 tu 
 
 
 (U 
 
 
 
 
 ►— 
 
 0) 
 
 
 <U 
 
 
 
 
 0) 
 
 
 V 
 
 
 0) cu 
 
 
 lU 
 
 
 
 
 «s 
 
 60 
 
 
 60 
 
 
 
 
 CO 
 
 
 60 
 
 
 .-1 CO 
 
 
 00 
 
 
 
 
 c^ 
 
 (U 
 
 
 <U 
 
 
 
 
 ti 
 
 
 V 
 
 
 4-1 (U 
 
 
 tu 
 
 
 
 
 C3 
 
 t— 1 
 
 
 f— 4 
 
 
 
 
 r-l 
 
 
 r-l 
 
 
 4-1 r-) 
 
 
 r-l 
 
 
 
 
 -^ 
 
 .-I 
 
 o 
 
 
 i-t 
 
 o 
 
 u 
 
 
 
 
 O 
 
 
 r-l 
 
 
 
 to 1-1 
 
 CO CJ 
 
 
 r-l 
 
 CJ 
 
 
 
 
 
 
 
 
 
 
 
 1 '■"^^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 4J C/3 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 d i^ 
 
 
 
 
 
 
 VI 
 
 
 
 
 
 
 
 
 
 
 
 S ?: 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 > 2: 
 
 
 
 
 
 
 It-I 
 
 
 
 
 
 
 
 
 
 
 
 t-i ^ 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 0) 
 
 
 
 
 
 
 
 
 
 00 
 
 
 (U 
 
 
 
 
 
 
 
 u 
 
 
 
 
 CO 1-1 
 
 
 
 
 a 
 
 
 c 
 
 
 ^> 
 
 
 
 
 
 CO 
 
 
 
 
 r-l 
 
 
 
 
 T^ 
 
 
 •H 
 
 
 CO 
 
 
 
 
 
 u 
 
 
 
 
 U-l <u 
 
 
 
 
 OS 
 
 
 VI 
 
 
 Ex^ 
 
 
 
 
 
 u 
 
 
 
 
 3 
 
 
 
 
 CO 
 
 
 (Q 
 
 
 )S 
 
 
 ae 
 
 
 
 c 
 
 
 
 
 •5 
 
 
 
 
 <u 
 
 
 
 
 ^Z 
 
 
 C3 
 
 
 
 0) 
 
 
 
 
 CO -r^ 
 
 
 
 
 
 
 
 •0 
 
 
 \^ 
 
 
 r~" 
 
 
 
 o 
 
 
 
 
 d) CO 
 
 
 
 
 
 
 
 <U 
 
 
 
 
 
 
 
 c 
 o 
 
 
 
 
 4-1 • 
 
 
 
 
 ^ 
 
 
 CO 
 
 
 u 
 tu 
 
 
 
 
 
 CJ 
 
 
 
 
 u > 
 
 
 
 
 
 
 VI 
 
 
 60 
 
 
 —J 
 
 
 
 
 
 
 
 01 
 
 
 
 
 <u 
 
 
 
 
 
 •r-l 
 
 
 C3Q 
 O. 
 
 
 
 c 
 
 •r-l 
 
 0) 
 u 
 
 o 
 
 Vj 
 
 Oh 
 
 CO 
 
 CO 
 
 D 
 
 
 
 ive Propi 
 er, Jr., 
 
 
 
 
 u 
 
 cd 
 Ji 
 c\. 
 to 
 
 
 
 00 
 
 •H 
 Vl 
 
 14-1 
 (U 
 
 C 
 
 
 C 
 
 Vi 
 
 
 
 • 
 
 
 
 
 
 r* 
 
 o 
 
 
 
 4-1 r 
 
 ■-I 
 
 
 
 
 
 
 M-t 
 
 
 * 
 
 
 
 
 
 00 
 
 
 
 
 •r^ CO 
 
 r-l 
 
 
 
 
 to 
 
 
 
 
 VI 
 
 
 
 CO 
 
 
 •,-1 
 
 CO 
 
 
 
 U t/) 
 
 •r-l 
 
 
 
 
 D -^ 
 
 
 TI 
 
 
 d) 
 
 
 
 0) 
 
 
 t-l 
 
 ■H 
 
 
 
 u 
 
 ^ 
 
 
 
 
 CO 
 
 
 (U 
 
 
 3 
 
 
 
 •u 
 
 
 
 O 
 
 
 
 3 ^ 
 
 
 
 ^^ 
 
 
 a> Cu 
 
 
 > 
 
 
 <0 
 
 
 
 nj 
 
 
 U-J 
 
 > 
 
 
 
 2 <-> 
 
 • 
 
 
 H 
 
 
 ^^ 
 
 
 r-l 
 
 
 PQ 
 
 
 
 >^ 
 
 
 O 
 
 ^ 
 
 
 
 
 X 
 
 
 M 
 
 
 
 
 
 
 
 
 
 u 
 
 
 
 n 
 
 
 
 T-l CO 
 
 
 
 nj 
 
 v; 
 
 
 <1 ^--' 
 
 
 ^0 
 
 
 • 
 
 
 
 C 
 
 
 CO 
 
 o 
 
 b 
 
 
 
 C •'^ 
 CO 
 
 u 
 
 
 CO 
 
 v-> 
 
 
 00 
 
 
 •r^ 
 
 
 rr 
 
 
 
 o 
 
 
 •H 
 
 13 
 
 
 
 > 
 
 0) 
 
 
 >-i 
 
 u 
 
 
 . OJ 
 
 
 a 
 
 
 ^ 
 
 
 
 c 
 
 
 u 
 
 
 
 
 r-l x: 
 
 (0 
 
 
 4-1 
 
 
 
 
 n ;^ 
 
 
 
 
 Vi 
 
 
 
 o 
 
 
 w" 
 
 •\ 
 
 
 
 CO a 
 
 •r-l 
 
 
 C 
 
 x; 
 
 
 -1 s 
 
 
 cu 
 
 
 CU 
 
 
 
 r-( 
 
 CJ 
 
 
 •H 
 
 <u 
 
 y^ 
 
 
 
 
 u 
 
 
 CJ 
 
 
 
 
 1 
 
 
 •0 
 
 
 4-1 
 
 
 . 
 
 r-- 
 
 
 
 Vj 
 
 ^ 
 
 CO 
 
 
 •H u 
 
 CO 
 
 
 
 
 u 
 
 
 >~> ^' 
 
 
 •n4 
 
 • • 
 
 c 
 
 
 CO 
 
 O 
 
 c 
 
 
 (U 
 
 CO 
 
 1 
 
 
 e ^ 
 
 ol 
 
 
 c 
 
 < 
 
 
 V4 Q 
 
 
 X 
 
 x; 
 
 3 
 
 
 Oi 
 
 .— 1 
 
 •H 
 
 
 JJ 
 
 •X. 
 
 
 <u 
 
 
 
 
 
 
 
 
 
 
 
 CO 
 
 X 
 
 
 CD 
 
 GJ 
 
 
 u 
 
 
 
 x: X • 
 
 
 CJ 
 
 • 
 
 
 4-1 >- 
 
 
 •r-l 
 
 •r-l 
 
 
 
 ^ 
 
 iJ 
 
 
 CD 
 
 x> 
 
 ■ — -* 
 
 <f 
 
 CJ cu a; 
 
 
 
 
 CJ 
 
 
 CO c 
 
 
 Q 
 
 (il 
 
 • 
 
 Cvl 
 
 O 
 
 
 M 
 
 o 
 
 
 vj- 
 
 ^ 
 
 LTl 
 
 c 
 
 
 
 to 
 
 
 
 
 1-) 
 
 o- 
 
 =3 
 
 U 
 
 
 CO 
 
 OJ 
 
 en 
 
 r-l 
 
 (U CO • 
 
 t— 1 
 
 ■H 
 
 • 
 
 
 cu 6 
 
 
 C 
 
 tu 
 
 
 -* 
 
 «i: 
 
 cu 
 
 
 S^ 
 
 
 M 
 
 1 
 
 ^ K w 
 
 1 
 
 d) 
 
 ^ 
 
 
 4-1 x: 
 
 r~» 
 
 
 
 r-H 
 
 • 
 
 1 
 
 
 
 
 (J 
 
 o 
 
 C 
 
 r-l 
 
 1-1 
 
 in 
 
 u 
 
 
 
 to QJ 
 
 cri 
 
 XI 
 
 
 
 (Xi 
 
 ro 
 
 
 x: 
 
 
 
 r-l 
 
 •r^ 
 
 <f 
 
 -0 
 
 <t 
 
 
 
 •» 
 
 
 W hJ 
 
 1 
 
 Vi 
 
 t5 
 
 
 n 
 
 
 W 
 
 
 r-< 
 
 o 
 
 i-l 
 
 r-l 
 
 c <u c 
 
 1— 1 
 
 u 
 
 bO 
 
 
 4-1 
 
 ro 
 
 CO 
 
 
 •» 
 
 <t 
 
 
 ■H 
 
 
 CO 
 
 
 r-l 
 
 
 .0 Di 3 
 
 
 ol 
 
 c 
 
 
 c 
 
 LO 
 
 CJ 
 
 
 c 
 
 
 
 U-, 
 
 
 c 
 
 ^ 
 
 •H 
 
 • • 
 
 
 
 ■ • 
 
 
 CO 
 
 
 tu hJ 
 
 
 
 U-l 
 
 
 
 • • 
 
 
 
 
 o 
 
 
 4-1 
 
 ^-s 
 
 0) 1 u 
 
 ,— \ 
 
 '— 
 
 3 
 
 
 CJ 
 
 • • 
 
 Ci-I 
 
 
 
 to 
 
 y-N 
 
 
 0) 
 
 
 •-H 
 
 -o 
 
 CO 
 
 r-l 
 
 00 CQ 
 
 r-l 
 
 CO 
 
 
 
 c « 
 
 ^N 
 
 
 
 
 r-l 
 
 CN 
 
 
 ^ 
 
 4-1 
 
 u 
 
 a 
 
 
 ^--^ 
 
 CO j:: 
 
 ^^ 
 
 •r-l 
 
 • 
 
 
 )-l 
 
 r-^ 
 
 
 C 
 
 tu 
 
 ^ 
 
 
 JS 
 
 a. 
 
 •H 
 
 4-J 
 
 • 
 
 C7^ 
 
 U CO 
 
 a> 
 
 u. 
 
 Q 
 
 
 u <u 
 
 v^* 
 
 0) 
 
 
 
 z: 
 
 ^7^ 
 
 
 ^ 
 
 •H 
 
 )-l 
 
 CO 
 
 Di 
 
 vD 
 
 -H H-I 
 
 
 
 
 
 
 >% 
 
 u-l 
 
 CO 
 
 •r-l 
 
 
 vO 
 
 
 
 •^^ M 
 
 JJ 
 
 ■u 
 
 
 
 4-J [1. 
 
 
 u-l 
 
 r* 
 
 
 C Q 
 
 CNI 
 
 3 
 
 XJ 
 
 • 
 
 
 
 o 
 
 CO o 
 
 D 
 
 
 • 
 
 C 
 
 CO 
 
 c 
 
 
 
 P 
 
 
 •r-l 
 
 
 
 to 
 
 IS 
 
 C 
 
 
 
 u 
 
 '-^ (0 
 
 :s: 
 
 
 CQ 
 
 •r-l 
 
 o lis 
 
 •r-l 
 
 
 3 
 
 4J 
 
 a) 
 
 >^ 
 
 
 
 > 
 
 
 •r^ 
 
 
 
 
 -'- D 
 
 
 e 
 
 
 iJ 
 
 a —1 
 
 4-1 
 
 c 
 
 CO 
 
 V4 
 
 XJ •-) 
 
 0£ 
 
 x: 
 
 Vj 
 
 • 
 
 4-1 
 
 
 
 '.'; 
 
 X, j: 
 
 — ^ 
 
 o 
 
 • n 
 
 -; 
 
 y c - 
 
 u) 
 
 G 
 
 X) 
 
 
 
 
 
 u 
 
 4-1 
 
 ^ 
 
 OS 
 
 CU 
 
 
 UJ 
 
 <y 
 
 ^-^ c5 
 
 £ 
 
 u 
 
 3 
 
 r-l 
 
 5 
 
 f—i 
 
 •H 
 
 c 
 
 c. 
 
 u 
 
 .—1 
 
 
 to 
 
 
 r-4 
 
 
 pi 
 
 p: 
 
 1 
 
 2 
 
 <0 
 
 U-l 
 
 o 
 
 1— 1 
 
 1— 1 
 
 4J 
 
 CLI 
 
 <u 
 
 cu -r-l 
 
 
 
 c 
 
 0) 
 
 * 
 
 1-1 
 
 
 OJ 
 
 
 
 u 
 
 3 
 
 C4-1 (J 
 
 3 
 
 CO 
 
 C 
 
 oi 
 
 r-l 
 
 C 
 
 
 
 Vi 
 
 4-1 
 
 3 
 
 
 t^ 
 
 •-I -o 
 
 f-( 
 
 -o 
 
 jO 
 
 ca 
 
 TJ -O 
 
 ca 
 
 N 
 
 c 
 
 
 x: 1-1 
 
 X 
 
 
 Oj 
 
 4-1 
 
 ca 
 
 
 
 c 
 
 D O 
 
 CO 
 
 <u 
 
 3 
 
 
 <U 3 
 
 
 •r-l 
 
 (0 
 
 4J 
 
 tn <i> 
 
 CJ 
 
 CO 
 
 
 
 
 
 
 
 w 
 
 
 CJ 
 
 (0 
 
 O 
 
 > 
 
 iJ i-l Q 
 
 >~ 
 
 1—1 
 
 H 
 
 
 
 •rH C 
 
 0) 
 
 U 
 
 y 
 
 c 
 
 > 
 
 
 
 
 cn 
 
 ■r-l 
 
 tn 
 
 
 u 
 
 
 
 u 
 
 ■r-l 
 
 
 CO 
 
 U, -H 
 
 H 
 
 ■r-l 
 
 x: 
 
 Vi 
 
 u 
 
 
 
 w 
 
 • -H 
 
 G 
 
 a 
 
 • 
 
 d) 
 
 (U CO • 
 
 
 
 XI 
 
 • 
 
 u 
 
 0. 
 
 
 ■0 
 
 4J 
 
 to 
 
 <u 
 
 
 
 c: 
 
 ca .-1 
 
 (U 
 
 ■ u 
 
 hJ 
 
 X 
 
 ^ V4 hJ 
 
 x: 
 
 3 
 
 Oi 
 
 4-1 
 
 3J CO 
 
 -0 
 
 3 
 
 
 03 
 
 X 
 
 
 
 •H 
 
 J3 
 
 
 o 
 
 
 CO 
 
 U-l A-» 
 
 CO 
 
 1—1 
 
 
 c 
 
 r^ 
 
 
 
 u 
 
 ^ 
 
 
 to 
 
 
 
 (/5 
 
 • D 
 
 CJ 
 
 u 
 
 • 
 
 •H 
 
 LJ X 
 
 -r-l 
 
 
 
 . 
 
 
 
 ^ ^ 
 
 
 
 CO 
 
 
 
 • 
 
 •H 
 
 
 
 Zj 
 
 :s &. 
 
 
 Oi 
 
 o 
 
 U. 
 
 : U Q 
 
 U-l 
 
 CO 
 
 CO 
 
 u 
 
 Un 
 
 
 U-l 
 
 X 
 
 U. 
 
 
 c> 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ac 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 =1 
 
 f-l 
 
 
 cs 
 
 
 
 
 CO 
 
 
 ^ 
 
 
 m 
 
 
 vO 
 
 
 
 
 SJ- 
 
 
 vf 
 
 
 
 
 SC 
 
 
 St 
 
 
 «* 
 
 
 <r 
 
 
 
 
 -c 
 
 o 
 
 
 o 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 o 
 
 UJ 
 
 
 u 
 
 
 
 
 UJ 
 
 
 UJ 
 
 
 w 
 
 
 UJ 
 
 
 
 
 II. 5-7 
 
cxz 
 
 CJD 
 
 CD 
 
 CO 
 CO 
 
 CO 
 
 UJ 
 
 o 
 o 
 
 a: 
 
 o 
 
 -5. 
 
 o 
 
 < 
 ►J 
 
 1 — 
 
 
 
 ■*"*■ 
 
 
 
 
 ' 
 
 
 ■^■" 
 
 
 
 "■■"■ 
 
 ■■"" 
 
 
 
 
 
 
 CJ> 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 -< 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 1 — 
 
 X 
 
 
 
 X 
 
 
 X 
 
 
 
 
 X 
 
 
 
 
 
 
 
 
 
 CT} 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 -< 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 3 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ A 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CD 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CC 
 
 X 
 
 
 
 X 
 
 
 
 
 
 
 
 
 
 
 X 
 
 
 
 
 
 CJ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ;y 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 J£ 
 
 
 
 j«: 
 
 
 
 
 Vj 
 
 
 V4 
 
 
 
 
 
 
 
 
 
 
 U 
 
 
 
 u 
 
 
 CD 
 
 
 TO 
 
 a, 
 
 
 TO 
 
 
 
 
 
 
 
 
 
 
 CO 
 
 
 
 TO 
 fa 
 
 
 K— 
 
 o 
 
 <u 
 
 
 0) 
 
 
 (U 
 
 
 
 
 01 
 
 
 
 
 a> 
 
 
 
 
 
 
 «< 
 
 .-1 
 
 to 
 
 
 oc 
 
 
 ,-t 
 
 
 
 
 r— I 
 
 
 
 
 iO 
 
 
 
 30 
 
 
 C_5 
 CD 
 
 iJ 
 
 
 
 
 <u 
 
 
 u 
 
 
 
 
 U 
 
 
 
 
 1; 
 
 
 
 01 
 
 
 tJ 
 
 •— * 
 
 
 .-( 
 
 
 4.i 
 
 
 
 
 4J 
 
 
 
 
 1— 1 
 
 
 
 .— 1 
 
 
 "— ^ 
 
 (9 
 
 ^ 
 
 
 r-( 
 
 
 TO 
 
 
 
 
 TO 
 
 
 
 
 .—1 
 
 
 
 — 1 
 
 
 
 
 
 u 
 
 
 
 
 
 
 OJ 
 CO 
 
 
 
 
 c5^ 
 
 
 
 
 
 u 
 
 
 
 
 CJ 
 
 
 
 
 
 
 
 
 . CO 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 U3 
 
 
 
 
 
 a> 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 D 
 
 G 
 
 
 
 
 u 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 •H 
 
 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 U 
 
 
 
 
 
 r-l 
 
 
 
 
 
 
 
 
 c 
 
 
 
 
 
 
 O 
 
 u 
 
 
 
 
 
 
 
 
 
 
 J3 
 
 
 
 •r4 
 
 
 
 
 
 
 D 
 
 
 
 
 
 
 CO 
 
 
 
 
 
 0) 
 
 
 
 OJ 
 
 
 
 
 
 
 1— I 
 
 11 
 
 
 
 
 M 
 
 
 
 
 
 ■r^ 
 
 
 
 iJ 
 
 
 
 
 
 
 W 
 
 >+-! 
 
 
 
 
 
 
 
 
 
 fa 
 
 
 
 
 
 
 
 
 
 
 2: 
 
 
 
 
 
 c 
 
 •H 
 
 
 
 
 c 
 
 •r-l 
 
 
 
 
 
 £ 
 
 
 
 
 
 ac 
 
 
 
 
 
 
 <U 
 
 
 
 
 01 
 
 •r-l 
 
 
 
 
 
 
 
 
 o 
 
 •' 
 
 Vi 
 
 
 
 
 U 
 
 
 
 
 4J 
 
 B 
 
 
 
 JC 
 
 
 
 
 
 fc 
 
 <u 
 
 ■M 
 
 
 
 
 
 
 
 
 
 
 
 U5 
 
 
 
 CO 
 
 
 
 
 
 
 ^ 
 qj 
 
 
 
 
 
 ^ 
 
 
 
 
 ^ 
 
 TO 
 
 i-H 
 
 
 
 •r-l 
 fa 
 
 
 
 
 
 CD 
 
 X 
 
 H 
 
 
 
 
 
 
 
 
 
 Q. 
 
 
 
 
 
 
 
 
 n!j 
 
 
 
 
 
 
 <« 
 
 
 
 
 U-I 
 
 
 
 
 
 <M 
 
 
 
 
 
 oa 
 
 c 
 
 ■0 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 13 
 
 •r-l 
 
 c 
 
 
 
 
 
 
 
 
 
 u 
 
 
 
 
 
 
 
 
 o. 
 
 (0 
 
 TO 
 
 
 CO 
 
 
 
 ■ft 
 
 
 
 
 c 
 
 
 TO 
 CO 
 
 
 
 c 
 
 
 ■r-l 
 
 
 
 (U 
 
 
 
 Vj 
 
 •.-1 
 
 
 
 •H 
 
 
 u 
 
 
 
 
 4J 
 
 T3 
 
 
 
 U 
 
 
 
 ^ 
 
 
 
 3 
 
 V) 
 
 
 
 M 
 
 
 TO 
 
 
 
 
 TO 
 
 § 
 
 
 
 u 
 
 
 
 CC 
 
 
 
 4-1 
 
 TO 
 
 ; 
 
 
 > 
 
 
 S-i 
 
 
 
 
 U 
 
 
 
 3 
 
 
 
 u 
 
 
 
 ta 
 
 •— ( 
 
 a; 
 
 
 1— ( 
 
 
 TO 
 
 
 
 
 TO 
 
 
 
 
 "O 
 
 
 
 
 
 
 u 
 
 »— * 
 
 4-1 
 
 
 
 
 
 O- 
 
 z 
 
 
 
 Q. 
 
 ^j 
 
 
 
 
 
 
 
 OJ 
 
 
 
 2i 
 
 n) 
 
 TO 
 
 
 t-l 
 
 
 
 
 c 
 
 
 
 OJ 
 
 TO 
 
 
 
 £ 
 
 ^-^ 
 
 
 ££ 
 
 ,,— ^ 
 
 
 E 
 
 
 
 •a 
 
 
 i 
 
 •r-l 
 0) 
 
 
 
 ^ 
 
 .—1 
 .—1 
 
 
 
 CO 
 
 fa 
 
 
 3 
 
 U-l 
 
 
 
 H 
 
 in 
 
 3 
 
 C 
 
 
 
 a: 
 
 
 OJ 
 
 4J 
 
 
 
 
 1) 
 
 •r-l 
 U 
 
 
 
 01 
 
 x; 
 
 
 
 • ^ 
 
 >-l 
 
 § 
 
 
 ^->. 
 
 U 
 
 
 
 
 u 
 
 
 x: 
 
 U 
 
 
 
 j:: 
 
 Ji 
 
 
 
 4-1 
 
 ^-» 
 
 
 4J 
 
 N— ' 
 
 
 — -1 -m 
 
 f-" 
 
 r* 
 
 
 ••-1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 — 
 
 
 
 C 
 
 
 
 
 
 
 u 
 
 
 
 
 
 
 
 li-i 
 
 
 
 ' 
 
 c 
 
 -* 
 
 5 
 
 Vj 
 
 
 i-J cu 
 
 >-i 
 
 u 
 
 
 g 
 
 
 u 
 
 J3 
 
 
 
 >-■ 
 
 
 
 
 
 4-) 
 
 
 
 
 
 
 
 at 
 
 
 C 
 
 TO 
 
 
 
 
 
 
 (0 
 
 
 
 
 
 >, 
 
 
 
 
 w 
 
 LO 
 
 
 
 
 W 
 
 SZ 
 
 c 
 
 
 >- 
 
 N 
 
 
 m 
 
 •r-l 
 
 ^ 
 
 
 M-i 
 
 2 
 
 __^ 
 
 
 TO 
 
 4-1 
 
 u 
 
 OJ 
 
 1 
 
 ^ 
 
 
 ■ 
 
 v_^ K 
 
 TO 
 
 a 
 
 
 C 
 
 
 x: 
 
 
 CO 
 
 
 x; 
 
 U-I 
 
 CO 
 
 
 TO 
 
 -a 
 
 
 
 X 
 
 ^ 
 
 yT 
 
 D. 
 
 01 
 
 u 
 
 y^N 
 
 w 
 
 r^ 
 
 U) 
 
 Vj 
 
 t, 
 
 
 w 
 
 
 
 fa 
 
 
 Q 
 
 G 
 
 LO 
 
 
 
 f— * 
 
 
 c. 
 
 
 
 C/1 
 
 CNI 
 
 
 u-i 
 
 •r-l 
 
 TO 
 
 g 
 
 
 •r^ 
 
 
 S 
 
 
 
 0) 
 
 
 U-l 
 
 . 
 
 ex: 
 
 r~» 
 
 v> 
 
 3 
 
 i 
 
 u. 
 
 r^ 
 
 > 
 
 vD 
 
 u, 
 
 .— ( 
 
 
 Du 
 
 Ui 
 
 is 
 
 
 E 
 
 X 
 
 
 
 
 PC 
 
 o 
 
 ON 
 
 3 
 
 .—1 
 
 2^ 
 
 o> 
 
 .n 
 
 1 
 
 
 f— 1 
 
 
 
 
 <ii 
 
 
 
 3 
 
 
 • • 
 
 
 
 ac 
 
 .— ( 
 
 
 
 
 
 
 y^ 
 
 .—1 
 
 
 ti-, 
 
 U-I 
 
 •r-l 
 
 
 
 14-1 
 
 -r-l 
 
 
 
 ■r-l 
 
 , 
 
 /-x 
 
 C 
 
 <y 
 
 1-^ 
 
 •r-l 
 
 x; 
 
 ^ 
 
 vO to 
 
 Oi 
 
 
 
 u 
 
 u 
 
 
 
 
 4J 
 
 ^^ 
 
 
 u 
 
 i-i 
 
 cr\ 
 
 
 
 r~4 
 
 ^ 
 
 •> 10 
 
 
 <f 0) 
 
 CO 
 
 
 XI 
 
 0) 
 
 
 
 U 
 
 OJ 
 
 
 -Q 
 
 
 ■-^ 
 
 •r-l 
 
 X 
 
 TO 
 
 ao -H 
 
 00 
 
 CM iJ 
 
 w 
 
 c 
 
 ■r4 
 
 .—1 
 
 
 c 
 
 CJ 
 
 r— 4 
 
 
 ■r-l 
 
 , 
 
 as 
 
 4-) 
 
 
 
 
 > 
 
 C tL, 
 
 n 
 
 1 TO 
 
 
 
 
 ^ 
 
 r-l 
 
 en 
 
 
 
 c 
 
 ,—1 
 
 00 
 
 .— ( 
 
 J 
 
 <t 
 
 a 
 
 
 
 
 •H 
 
 •r-l 
 
 •-I U 
 
 CO 
 
 •r-l 
 
 
 
 ■rj 
 
 
 
 •r^ 
 
 
 
 •r-l 
 
 
 
 ■r-l 
 
 
 
 TO 
 
 :5 
 
 
 u 
 
 U M-( 
 
 Di 
 
 <f 4-> 
 
 fT. 
 
 4-1 
 
 >^ 
 
 S 
 
 ^D 
 
 4-1 
 
 £ 
 
 s 
 
 vD 
 
 ■ 3 
 
 «, 
 
 
 V-i 
 
 
 
 ca 
 
 ^4 
 
 
 CM C 
 
 *2 
 
 TO 
 
 s: 
 
 
 1 
 
 TO 
 
 
 
 0- 
 
 3 
 
 ■ 
 
 4-1 
 
 . 
 
 
 
 01 
 
 • 
 
 <U 
 
 2 
 
 N 
 
 
 • 
 
 CT\ 
 
 N 
 
 
 • 
 
 vt 
 
 UJ 
 
 
 
 
 
 X 
 
 
 
 
 U) 
 
 ■X. c 
 
 Q 
 
 
 
 
 •r^ 
 
 U-l 
 
 oi 
 
 a\ 
 
 •r-l 
 
 •0 
 
 02 
 
 
 
 
 u 
 
 
 
 UJ 
 
 
 
 <u 
 
 
 
 
 '-^ c 
 
 1 
 
 r-l 
 
 
 
 
 u-i 
 
 , 1 
 
 c 
 
 
 sO 
 
 >. 
 
 X 
 
 CO 
 
 
 <« 
 
 
 u 
 
 r^ 
 
 •• 
 
 —1 
 
 
 ■r-l 
 
 
 M 
 
 
 •r-l 
 
 TO 
 
 r. 
 
 
 >-l 
 
 3 
 
 
 u 
 
 c 
 
 
 nj 
 
 — ( 4-1 
 
 ■r^l 
 
 ^ u 
 
 4-) 
 
 4J 
 
 tn 
 
 >> 
 
 .. 
 
 4-1 
 
 
 >^ 
 
 . , 
 
 TO 
 
 a 
 
 'J> 
 
 C 
 
 3 
 
 
 a: 
 
 ^ TO 
 
 f— 1 
 
 a\ 
 
 u 
 
 C3 
 
 c 
 
 u 
 
 r-» 
 
 := 
 
 d 
 
 V4 
 
 r^ 
 
 e 
 
 
 i-> 
 
 
 
 
 
 
 
 ■^ U 
 
 r— 1 
 
 vO C 
 
 
 
 
 
 
 3 
 
 C-) 
 
 
 
 
 3 
 
 ro 
 
 • 
 
 CO 
 
 > 
 
 V4 
 
 
 m 
 
 TO 
 
 0) 
 
 •H 
 
 ex 
 
 <U 
 
 ■r4 
 
 
 
 
 OJ 
 
 -(-4 
 
 
 
 
 i) 
 
 -J 
 
 •r-l 
 
 I — 1 
 
 03 
 
 
 •H 
 
 TO D. 
 
 c 
 
 0) 
 
 ^-N 0) 
 
 x: 
 
 4J 
 
 ii 
 
 
 -C 
 
 4-i 
 
 i(i 
 
 A 
 
 M 
 
 
 G 
 
 r\ 
 
 
 
 w'j 
 
 ^ ^ 
 
 •r-l 
 
 c 
 
 4-1 
 
 in cc 
 
 4-) 
 
 TO 
 
 
 * 
 
 4J 
 
 TO 
 
 
 tn 
 
 
 , 
 
 0) 
 
 CO 
 
 . 
 
 
 >> 
 
 XI (C 
 
 a. 
 
 ■r-l 
 
 
 
 Cu 
 
 
 
 
 , 
 
 
 
 
 r— 1 
 
 , 
 
 
 
 O-l 
 
 Q 
 
 X 
 
 
 H 
 
 LU 
 
 .—( 
 
 CO 
 
 4_» 
 
 (fc 
 
 ^ ' — ' 
 
 
 
 ■H 
 
 DQ 
 
 
 
 
 
 
 
 CQ 
 
 
 
 
 
 
 u 
 
 U-l 
 
 
 pi 
 
 
 
 C 
 
 
 
 
 ?2 TO 
 
 4-1 
 
 14-I 
 
 
 c 
 
 4-1 
 
 U) 
 
 
 c 
 
 
 M 
 
 
 
 
 •< 
 
 i-i 
 
 TO 11 
 
 • 
 
 1 — I 
 
 
 v-r a 
 
 
 ■r^ 
 
 A 
 
 0) 
 
 
 t— 1 
 
 <K 
 
 01 
 
 'c r 
 
 il 
 
 rH 
 
 
 5 
 
 3 
 
 x; 
 
 •-> 
 
 r~l 
 
 x: 
 
 •r-l 
 
 I/) 
 
 ■a 
 
 •^ 
 
 •r-l 
 
 U) 
 
 
 ■rA 
 
 ■r-l 
 
 0) 
 
 V 
 
 •r^ 
 
 >- 
 
 4-) 
 
 
 < 
 
 « 4-) 
 
 
 3 
 
 U) 
 
 C 
 
 
 
 
 
 . 1 
 
 
 
 a 
 
 
 . — 1 
 
 
 
 —1 4-1 
 
 X 
 
 
 
 IJ 
 
 ^ Vj 
 
 
 
 (0 
 
 
 •* 
 
 CO 
 
 •H 
 
 —1 x: 
 
 X 
 
 • - 
 
 . 1 
 
 CO 
 
 X 
 
 , " 
 
 r— * 
 
 W 
 
 4-1 TO 
 
 tx, 
 
 
 ■r-l 
 
 X C 
 
 
 C 
 
 3 
 
 c 
 
 >. 
 
 
 u. 
 
 TO (J 
 
 
 
 
 'J 
 
 
 
 
 7) V) 
 
 i> 
 
 
 TO U 
 
 _y 
 
 • 
 
 . > 
 
 3 CX 
 
 
 •rH 
 
 4_) 
 
 •r-l 
 
 ^ 
 
 >! 
 
 
 "T" "J 
 
 TO 
 
 
 
 a 
 
 -0 
 
 TO 
 
 V c 
 
 c 
 
 T3 
 
 J-" 
 
 rT- 
 
 V4 
 
 •r-l 
 
 J 
 
 
 V 
 
 
 
 
 3 
 
 u 
 
 tc 
 
 H 
 
 
 
 ■r-l 
 
 ■r-l 
 
 G 
 
 ^ 
 
 4J -rJ 
 
 ■^ 
 
 
 
 —1 C 
 
 
 (U 
 
 . X 
 
 a: 
 
 
 u 
 
 '^ 
 
 ■0 
 
 
 
 
 
 c 
 
 
 >-l 
 
 rr 
 
 a 
 
 
 
 t-i 
 
 TO 
 
 ex. 
 
 
 
 — ' 1; 
 
 , 
 
 e 
 
 •r.* 
 
 
 
 
 
 -0 
 
 .— ( 
 
 i^ 
 
 x: 
 
 •r^ 
 
 CO < 
 
 Q- 
 
 >^ 
 
 CO 
 
 U. 
 
 Q- 
 
 r-( 4-1 
 
 CO 
 
 Uu 
 
 C- u 
 
 Q 
 
 < 
 
 in 
 
 J CO 
 
 
 p 
 
 
 
 il 
 
 
 w 
 
 -^ 
 
 < 
 
 a 
 
 N 
 
 
 
 D. 
 
 
 
 
 
 a. c 
 
 
 
 TO 
 
 fa 
 
 
 ^ 
 
 —1 
 
 • 
 
 •r-4 
 
 TO 
 
 • 
 
 < 
 
 c 
 
 . 
 
 , 
 
 <C 
 
 ;^ i 
 
 , 
 
 . 
 
 < 
 
 , 
 
 • 
 
 CJ 
 
 £ ^: 
 
 
 " 
 
 >- 
 
 CQ 
 
 U. 
 
 s 
 
 :c '-s 
 
 " 
 
 UJ 
 
 <-> 
 
 •-) 
 
 " 
 
 •n 
 
 •-) 
 
 - 
 
 < 
 
 n 
 
 fa 
 
 CD 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ac 
 
 r* 
 
 
 
 00 
 
 
 CT> 
 
 
 
 
 
 
 
 
 
 i—t 
 
 
 
 CM 
 
 
 rJ 
 
 <r 
 
 
 
 <r 
 
 
 <f 
 
 
 
 
 in 
 
 
 
 
 iri 
 
 
 
 LO 
 
 
 «x: 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CJ 
 
 UJ 
 
 
 
 w 
 
 
 UJ 
 
 
 
 
 w 
 
 
 
 
 U 
 
 
 
 UJ 
 
 
 II. 5-8 
 

 CO 
 
 
 w 
 
 >> 
 
 CO 
 
 
 CO 
 
 a. 
 
 
 «« 
 
 o 
 
 o:: 
 
 r< 
 
 ej3 
 
 Pu| 
 
 CD 
 
 
 _j 
 
 o 
 
 CO 
 
 ca 
 
 H 
 
 CO 
 
 ^ 
 
 Ou 
 
 O 
 
 iJh. 
 
 « 
 
 t— 
 
 
 
 
 
 
 
 
 
 
 
 
 
 s 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 X 
 
 
 X 
 
 X 
 
 
 X 
 
 
 
 
 
 
 ty> 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 
 
 
 5 
 
 
 
 
 
 
 
 
 
 
 
 
 
 a 
 
 a: 
 
 
 X 
 
 
 X 
 
 X 
 
 
 X 
 
 
 
 
 
 X 
 
 tJ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Z 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 <u 
 
 0) 
 
 
 <u 
 
 
 0) 
 
 
 eu 
 
 O 
 
 1-4 
 
 
 
 
 
 o 
 
 u 
 
 
 o 
 
 
 o 
 
 
 o 
 
 
 
 
 
 t^ 
 
 T-l 
 
 
 1-1 
 
 
 1-1 
 
 
 1-t 
 
 
 
 
 X 
 
 
 <M 
 
 U4 
 
 
 <*4 
 
 
 ^ X-) 
 
 
 ^ <4-l 
 
 
 
 u 
 
 
 «H 
 
 <4-l 
 
 
 c*-t 
 
 
 I-l <4-l 
 
 
 U >4-l 
 
 CO O 
 
 CLi 
 
 C3 
 
 
 03 
 
 0bl 
 
 
 o 
 
 o 
 
 
 o 
 
 
 03 O 
 
 
 CO o 
 
 t— 
 
 
 
 
 c 
 
 e 
 
 
 C 
 
 
 c 
 
 
 c 
 
 0) O 
 
 60 -T* 
 
 l-t T-t 
 
 r^ > 
 
 O -r* 
 CJ Q 
 
 «< 
 
 
 (U 
 
 
 <u o 
 
 o 
 
 
 0) o 
 
 
 <u o 
 
 
 eu O 
 
 tj 
 
 
 CiO 
 
 
 r-< .r^ 
 
 tH 
 
 
 1-4 -1-1 
 
 
 60 1-1 
 
 
 CO 1-1 
 
 o 
 
 
 <U 
 
 
 4J en 
 
 en 
 
 
 4J CO 
 
 
 0) <a 
 
 
 eu CO 
 
 ^J 
 
 
 i-H 
 
 
 4J 1-t 
 
 1-< 
 
 
 -U •r^ 
 
 
 •-I 1-t 
 
 
 I-l 1-t 
 
 
 
 i-l 
 
 
 ca > 
 
 > 
 
 
 CO > 
 
 
 r-l > 
 
 
 1-1 > 
 
 
 
 o 
 
 
 0) -r^ 
 
 •r4 
 
 
 0) 1-1 
 
 
 O 1-t 
 
 
 O 1-t 
 
 
 
 CJ 
 
 
 CO Q 
 
 Q 
 
 
 CO Q 
 
 
 U Q 
 
 
 O Q 
 
 
 
 
 
 
 
 
 « 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 3 
 
 
 
 
 
 
 
 
 
 
 
 
 .1 
 
 
 O 
 
 
 
 
 
 
 
 a: 
 
 
 
 
 
 <V- 
 
 
 « 
 
 
 
 
 
 
 
 C3 
 
 
 
 
 
 c 
 
 
 
 3 
 
 
 
 
 
 
 
 »— 
 
 
 
 
 
 o 
 
 •f4 
 
 
 
 ^ 
 
 
 
 
 
 
 
 C£ 
 
 
 
 
 
 4J 
 
 n) 
 
 
 
 4) 
 
 
 
 
 
 
 CO 
 
 CX3 
 
 
 
 
 
 4J 
 
 o 
 
 
 
 5 
 
 
 
 
 
 
 O 
 CM 
 
 
 a. 
 
 
 
 
 
 
 
 > 
 
 
 
 
 
 
 4J 
 
 
 
 
 
 
 
 
 
 
 .O 
 
 
 
 
 
 
 u 
 
 
 
 
 
 
 
 
 
 
 (U 
 
 
 
 
 
 
 <U 
 
 
 
 
 
 
 
 6 
 
 
 
 
 
 
 
 
 
 o 
 
 1-4 
 
 P-I 
 
 
 
 
 /^ 
 
 
 
 
 u 
 
 
 
 
 
 
 
 
 
 
 
 
 
 < 
 
 
 
 
 PM 
 
 
 
 « 
 
 
 
 
 
 /-> 
 
 
 
 
 
 
 
 PM 
 
 
 
 
 b 
 
 
 
 ^ 
 
 
 
 
 
 < 
 
 
 o 
 
 
 
 
 M 
 
 
 
 
 c 
 
 
 
 CO 
 
 
 
 
 
 & 
 
 
 u 
 x: 
 
 
 
 
 C 
 
 
 
 
 0) 
 
 
 
 
 cu 
 
 
 
 
 
 
 
 
 
 ^*-v 
 
 •r-l 
 
 
 
 
 "O 
 
 
 
 to 
 
 V4 
 
 
 
 
 
 C 
 
 
 j_i 
 
 
 
 TS 
 
 c 
 
 
 
 
 n 
 
 
 
 CO 
 
 (U 
 
 3 
 
 4J 
 
 
 
 
 
 v4 
 
 
 o 
 
 
 
 4J 
 
 <U 
 
 
 
 
 c 
 
 
 
 Vi 
 
 O 
 
 
 
 
 
 s 
 
 
 
 
 c 
 o 
 
 T3 
 
 
 
 
 iJ 
 
 
 
 P4 
 
 CO 
 
 
 
 
 
 4) 
 
 
 o 
 
 J-1 
 
 
 
 (0 
 
 
 
 
 s 
 
 
 
 
 >4-l 
 
 
 
 
 
 •g 
 
 
 
 
 
 
 
 
 
 0) 
 
 
 en 
 
 ■-4-1 
 
 o 
 
 
 
 
 
 
 HI 
 
 
 4J 
 
 
 co" 
 
 
 <u 
 
 
 
 
 ^-^ 
 
 
 r^ 
 
 
 CO 
 
 
 
 
 
 C 
 
 
 c 
 
 eu 
 
 
 
 
 S 
 
 
 
 
 o 
 
 x: 
 
 
 ON 
 
 (0 
 
 S 
 
 
 
 
 
 S 
 
 
 
 3= 
 
 
 
 
 
 en 
 
 
 ts 
 
 
 
 4-1 
 
 S£ 
 
 
 
 
 
 ic 
 
 
 0) 
 f— 4 
 
 
 =5 
 
 
 <u 
 
 
 
 0) 
 
 
 e 
 
 
 
 >J 
 
 fc 
 
 
 
 
 
 CO 
 
 
 o. 
 
 D 
 
 3 
 m 
 
 
 «C 
 
 
 n 
 
 
 
 •I-l 
 
 
 o 
 
 
 
 0) 
 
 
 
 
 
 
 4) 
 
 
 
 
 
 b 
 
 
 
 "O 
 
 
 ^J 
 
 
 
 a 
 
 c 
 
 
 
 
 
 h 
 
 
 
 
 
 
 
 
 D 
 
 
 IM 
 
 
 
 o 
 
 1-4 
 
 
 
 
 
 U* 
 
 
 
 
 
 
 
 
 4J 
 
 
 
 
 
 u 
 
 
 
 
 4J 
 
 
 
 >-> 
 
 
 ^ 1 
 
 
 
 > 
 
 
 
 CO 
 
 
 m 
 
 
 
 (U 
 
 (U 
 
 
 
 C 
 
 
 
 
 
 1 
 
 
 
 T3 
 
 
 
 
 
 <u 
 
 
 
 
 CO 
 
 
 
 <U 
 
 
 
 4J 
 
 
 0) 
 CO 
 
 
 
 
 D 
 
 
 
 0) 
 
 
 I-l 
 
 
 
 0) 
 
 3 
 
 
 
 e 
 
 
 
 c 
 
 
 
 
 
 4J 
 
 
 
 6£ 
 
 
 CJ 
 
 
 
 bC 
 
 
 
 
 1-1 
 
 
 
 <u 
 
 
 
 
 
 CO 
 
 
 
 CO 
 
 
 1-1 
 
 
 
 CO 
 
 I-l 
 
 
 
 u 
 
 
 
 s 
 
 
 
 
 
 
 
 
 u 
 
 
 u 
 
 
 
 u 
 
 O 
 
 
 
 0) 
 
 
 
 t4 
 
 
 o 
 
 4J 
 
 
 
 
 <u 
 
 
 
 o 
 
 /-N 
 
 u 
 
 
 
 o 
 
 >4-l 
 
 
 
 a 
 
 
 
 h 
 
 
 
 
 
 bC 
 
 
 
 4J 
 
 CO 
 
 CD 
 
 
 
 4-1 
 
 
 /~\ 
 
 
 X 
 
 
 
 0) 
 
 
 CO 
 
 
 
 
 CO 
 
 ^-\ 
 
 
 CO 
 
 & 
 g 
 
 P^ 
 
 •-N 
 
 
 CO 
 
 tn 
 
 CO 
 
 
 w 
 
 
 
 Ci 
 
 
 
 
 
 
 n 
 
 CO 
 
 
 
 
 CO 
 
 
 
 en 
 
 pn 
 
 
 
 •— \ 
 
 
 X 
 
 
 c: 
 
 0) 
 
 
 
 
 o 
 
 1 
 
 
 c 
 
 <U 
 
 & 
 g 
 
 
 -o 
 
 0) 
 
 ^ 
 
 
 6C 
 
 CO 
 
 
 w 
 
 ^~\ 
 
 
 uu 
 
 
 CO 
 
 
 
 N_^ 
 
 c 
 o 
 
 
 c 
 
 CO 
 
 u 
 
 
 
 N«^ 
 
 
 c 
 
 •r^ 
 
 § 
 
 
 f^ 
 
 CO 
 
 CO 
 
 
 f=i 
 
 
 
 
 
 ct) 
 
 
 pa 
 
 v_^ 
 
 
 
 I-l 
 
 
 
 c 
 
 ^l4 
 
 
 1-1 
 
 1 ^J 
 
 x; 
 c 
 
 0) 
 
 
 
 I-l 
 
 
 u 
 
 x 
 
 •H 4-1 
 
 
 
 4J 
 
 c 
 
 Oi 
 
 1-1 
 
 4J 
 
 o 
 
 ^— ' 
 
 4J 
 
 p 
 
 
 F= 
 
 
 01 
 
 
 V4 
 
 c 
 
 I-l u 
 
 bC 
 
 
 U 
 
 o 
 
 
 I-l 
 
 I-l 
 
 XI 
 
 
 U 
 
 3 
 
 ^-^ Vj 
 
 
 
 
 > 
 
 3 
 
 O 
 
 CJ 
 
 I-l o 
 
 c 
 
 5 
 
 O 
 
 •rl 
 
 01 
 
 I-l 
 
 o 
 
 0) 
 
 
 o 
 
 1— 1 
 
 o 
 
 iJ 
 
 
 
 (1) 
 
 o 
 
 cx 
 
 ^ 
 
 (U C- 
 
 •I-l 
 
 o 
 
 a. 
 
 4-1 
 
 u 
 
 0) 
 
 Q. 
 
 Q 
 
 iJ 
 
 a. 
 
 CO 
 
 4J Oi 
 
 CO 
 
 
 
 ^J 
 
 (-1 
 
 01 
 
 
 C <U 
 
 4-J 
 
 V-l 
 
 <u 
 
 to 
 
 « 
 
 c 
 
 q; 
 
 
 en 
 
 0) 
 
 
 en G) 
 
 c 
 
 OJ 
 N 
 
 
 
 
 
 XI 
 
 ck; 
 
 c 
 
 •r^ Pd 
 
 r: 
 
 J3 
 
 OS 
 
 >-l 
 
 
 •H 
 
 ci 
 
 C 
 
 C 
 
 a: 
 
 X) 
 
 C OS 
 
 
 
 ^ 
 
 3 
 
 
 GJ 
 
 D. 
 
 M 
 
 3 
 
 
 <0 
 
 "o 
 
 tx 
 
 
 01 
 
 u 
 
 
 0) 
 
 V4 
 
 W C/5 
 
 
 
 u 
 
 Q 
 
 CO 
 
 N 
 
 CO CO 
 
 r3 
 
 o 
 
 CO 
 
 p- 
 
 y) 
 
 CO 
 
 CO 
 
 (U 
 
 M 
 
 CO 
 
 4J 
 
 tj CO 
 
 O 
 U 
 
 
 
 
 c 
 
 • 
 
 b 
 
 ^ 
 
 o 
 I-l 
 
 
 C- 
 O 
 CO 
 
 Q 
 
 1 
 
 Zi 
 
 u 
 
 CM 
 
 O 
 
 A 
 
 ^ 
 
 >^4 
 
 o 
 
 CO 
 
 (ui 
 
 1 
 
 CO 
 
 0) 
 
 aj g 
 
 fij 2 
 
 <=> 
 
 
 
 
 
 
 
 
 
 
 
 
 
 a: 
 
 
 
 
 
 
 
 
 
 
 
 
 
 rd 
 
 
 CO 
 
 
 <t 
 
 «r» 
 
 
 «o 
 
 
 r* 
 
 
 00 
 
 
 «« 
 
 
 m 
 
 
 m 
 
 m 
 
 
 \n 
 
 
 <r> 
 
 
 !Q 
 
 o 
 
 w 
 
 CJ 
 
 
 o 
 
 
 o 
 
 o 
 
 
 o 
 
 
 o 
 
 
 o 
 
 
 
 u 
 
 
 w 
 
 w 
 
 
 UJ 
 
 
 UJ 
 
 
 w 
 
 II. 5-9 
 
3= 
 
 -a: 
 oc: 
 
 t3 2 
 
 LA- Q 
 
 1^ 
 
 1 — 
 
 
 ' 
 
 
 
 
 
 
 
 CJ 
 
 
 
 
 
 
 
 
 
 "X 
 
 
 
 
 
 
 
 
 
 ot: 
 
 
 
 
 
 
 
 
 
 t— 
 
 
 
 
 
 
 
 
 
 oo 
 
 
 
 
 
 
 
 
 
 
 X 
 
 
 x; 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 J 
 
 
 
 
 
 
 
 
 
 uZ 
 
 
 
 
 
 
 
 
 
 C3 
 
 
 
 
 
 
 
 
 
 Crt 
 
 
 
 
 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 i= 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Il> 
 
 
 
 ^ 
 
 
 A! 
 
 
 ^ 
 
 
 ^ 
 
 1-1 
 in 
 
 re 
 
 CD 
 
 a. 
 
 
 
 
 C14 
 
 
 P4 
 
 
 
 H^ 
 
 
 
 
 
 
 
 
 c 
 
 «=: 
 
 0) 
 
 
 0) 
 
 
 a> 
 
 
 u 
 
 
 
 C5 
 
 00 
 
 
 OO 
 
 
 60 
 
 
 00 
 
 1-1 
 
 CD 
 
 1 
 
 a) 
 
 
 (U 
 
 
 0) 
 
 
 « 
 
 to 
 
 
 i-H 
 
 
 r-l 
 
 
 i-H 
 
 
 r-) 
 
 •H 
 
 
 cH 
 
 
 1-1 
 
 
 rH 
 
 
 •-I 
 
 > 
 
 
 O 
 
 
 o 
 
 
 
 
 
 
 
 •1-1 
 
 
 o 
 
 
 o 
 
 
 CJ 
 
 
 u 
 
 Q 
 
 
 
 
 
 
 
 
 c 
 
 
 
 
 
 
 4J 
 
 
 
 
 
 
 
 
 
 
 
 
 c 
 
 
 
 
 
 •H 
 
 
 c 
 
 
 
 
 
 CO 
 
 
 
 
 
 ■4J 
 
 
 
 
 
 
 
 
 r-« 
 
 
 
 
 
 CO 
 
 
 4-1 
 
 
 
 
 
 (14 
 
 
 
 
 
 4J 
 
 
 6£ 
 
 c 
 
 
 
 u 
 
 
 o 
 
 
 
 
 
 to 
 
 
 •H 
 
 
 
 o 
 
 
 PL4 
 
 
 
 
 
 c 
 
 
 X 
 
 
 
 <4-l 
 
 
 PLI 
 
 
 
 
 
 
 
 
 CO 
 
 
 3C 
 
 
 
 
 
 
 
 
 a 
 
 
 to 
 
 
 C3 
 
 0) 
 
 
 >> 
 
 
 
 
 
 (U 
 
 
 s 
 
 
 h- 
 
 4J 
 
 
 <0 
 
 
 
 
 
 Q 
 
 
 
 
 ■u: 
 
 to 
 
 
 o 
 
 
 
 
 
 1 
 
 
 
 
 «_5 
 
 4-1 
 
 
 M 
 
 
 
 
 
 r-l 
 
 CD 
 
 
 c 
 
 a; 
 
 
 
 c 
 <u 
 
 
 (U 
 
 
 
 
 
 4J 
 
 
 OJ 
 
 •0 
 
 
 eu 
 
 o 
 c 
 o 
 o 
 
 c 
 
 •t-l 
 
 0) 
 
 4-1 
 
 o 
 u 
 
 Ph 
 
 
 c 
 o 
 H 
 
 o 
 
 CO 
 
 o 
 
 U-l 
 
 a, 
 
 VI 
 
 o 
 u 
 
 >. 
 
 u 
 
 <u 
 (-; 
 
 
 6( 
 
 c 
 •I-l 
 
 (0 
 (0 
 (U 
 
 
 
 u 
 
 
 
 0) 
 
 •S 
 
 n 
 
 a) 
 
 a 
 
 X 
 
 w 
 
 
 to 
 
 u 
 
 4J 
 
 c 
 to 
 1—1 
 Ph 
 
 4J 
 
 
 
 
 
 CO 
 
 
 <u 
 
 T^ 
 
 
 •H 
 
 
 
 
 
 V4 
 
 1-1 
 
 
 
 T^ 
 
 
 4J 
 
 x: 
 
 
 pt4 
 
 
 
 
 CU 
 
 T-l 
 
 
 
 u* 
 
 
 TO 
 
 en E 
 
 o 
 
 to 
 
 
 
 
 
 
 c 
 
 
 >> 
 
 P^ 
 
 
 
 o 
 
 
 1 4J 
 \0 CO 
 
 •a 
 
 
 4J 
 
 u 
 
 
 
 •r-l 
 4-1 
 
 to 
 
 to 
 
 6 
 
 1 
 
 
 . 
 
 c 
 
 
 u 
 
 o 
 
 
 
 
 
 
 
 
 » 
 
 
 
 
 <yi 
 
 CO 
 
 o 
 
 
 u-i 
 
 
 o 
 
 vD 
 
 iJ 
 
 
 1^ 
 
 
 
 
 
 Ph 
 
 
 Cfl 
 
 VD 
 
 -1-1 
 
 
 -•^ vD 
 
 4-1 
 
 fu 
 
 vO 
 
 to 
 
 
 vO 
 
 >A 
 
 
 fl4 
 
 
 C3 
 
 ON 
 
 4-1 
 
 
 O C3> 
 
 CO 
 
 
 tj> 
 
 V4 
 
 
 t3N 
 
 
 • 
 
 
 
 ^ 
 
 1-1 
 
 o 
 
 
 i-H F-< 
 
 o 
 
 CO 
 
 f-l 
 
 CO 
 
 
 1-1 
 
 u 
 
 Q 
 
 1 
 
 
 3 
 
 
 •^ o 
 
 0) 
 
 
 a 
 
 
 
 
 
 
 
 
 =) 
 
 T3 
 
 
 /-v in 
 
 •r^ 
 
 
 1) 
 
 
 
 l*-l 
 
 . •» 
 
 4.) 
 
 
 «< 
 
 o 
 
 
 CO tN ID 
 
 u 
 
 
 C/2 
 
 
 
 
 r-t 
 
 u 
 
 
 
 u 
 
 
 ^ c 
 
 o 
 
 
 
 
 
 
 CO 
 
 A 
 
 
 
 
 
 p^ 
 
 
 
 CD 
 
 iJ 
 
 
 
 
 
 
 
 to 
 
 3 
 
 a 
 
 
 
 
 
 g s" 
 
 
 to 
 
 
 u 
 
 
 w 
 
 OJ 
 
 •-5 
 
 (U 
 
 
 
 (U 
 
 
 v-^ 0) 
 
 B 
 
 u 
 
 
 
 
 s 
 
 u 
 
 
 ct: 
 
 
 
 j:: 
 
 
 •t-l 
 
 o 
 
 
 12 
 
 
 CO 
 
 < 
 
 • 
 
 
 
 
 4J 
 
 
 •a > 
 
 ■r-l 
 
 J3 
 
 
 C/D 
 
 
 
 
 Oi, 
 
 4J 
 
 y— \ 
 
 
 
 
 C (U 
 
 CO 
 
 to 
 
 
 
 y-N 
 
 
 
 CM 
 
 
 c 
 
 • 
 
 
 c 
 
 
 CO pcj 
 
 OJ 
 
 hJ 
 
 
 (U 
 
 a 
 
 u 
 
 
 
 " 4J 
 
 (1) 
 
 
 
 
 o 
 
 
 r- 1 
 
 o 
 
 
 
 X 
 
 
 
 
 c a, 
 
 E 
 
 c • 
 
 
 
 
 TJ CO 
 
 
 60 
 
 
 4-1 
 
 «4I 
 
 4-1 
 
 c 
 
 Q) 1-1 
 
 Dh 
 
 H 
 
 
 CO 
 
 
 O 0) 
 
 (0 
 
 c 
 
 
 
 
 u 
 
 
 
 1-1 K 
 
 
 
 
 
 0) 
 
 3 
 
 ^ 
 
 CO 
 
 •H 
 
 
 14-1 
 
 < 
 
 
 
 •■-1 
 
 I-l tJ 
 
 1-1 
 
 « 
 
 
 •H 
 
 c 
 
 . u 
 
 OJ 
 
 V4 
 
 
 
 
 
 a. 
 
 *J 
 
 < CO 
 
 a 
 
 CO 
 
 
 •a 
 
 o 
 
 U 0) 
 
 o 
 
 0) 
 
 
 
 CO 
 
 OJ 
 
 to 
 
 D 
 
 > 
 
 w 
 
 
 D 
 
 ■ri 
 
 X 
 
 o 
 
 <u 
 
 u 
 
 ^ 
 
 (0 
 
 OS 
 
 3 
 
 • c 
 
 0) 
 
 0) 4J 
 
 LU 
 
 4-) 
 
 4-1 
 
 - U) 
 
 u 
 
 c 
 
 M 
 
 
 
 X 
 
 
 I-H 
 
 ffi to 
 
 
 
 4-> V4 
 
 
 CO 
 
 D- 
 
 U -r-l 
 
 eu 
 
 •H 
 
 o 
 
 tA 
 
 0) 
 
 u 
 
 to 
 
 ^ 2: 
 
 
 CO 
 
 
 
 <U fa 
 
 
 60 
 
 a. 
 
 4_) 
 
 H 
 
 
 
 > 
 
 
 •a 
 
 lU -Oi 
 
 Ix, 
 
 3 
 
 CO 
 
 >, 
 
 c 
 
 4) 
 
 CO 
 
 ^^-' 
 
 u 
 
 w 
 
 A x) 
 
 c 
 
 > <u 
 
 
 u 
 
 CO 
 
 •H i-l 
 
 u 
 
 w 
 
 oi 
 
 
 
 
 
 
 
 4-1 (J 
 
 to 
 
 u (A 
 
 
 o 
 
 c 
 
 Vj (0 
 
 nj 
 
 
 
 •rl 
 
 0) 
 
 (0 
 
 c 
 
 ■r< x: 
 
 
 to 
 
 
 M 
 
 o 
 
 <0 -r^ 
 
 c 
 
 .-( 
 
 4J 
 
 1~< 
 
 ^ 
 
 Vl 
 
 
 
 E to 
 
 x: 
 
 rd 4J 
 
 
 
 u 
 
 CM o 
 
 •H 
 
 CO 
 
 o 
 
 a 
 
 c 
 
 4-1 
 
 
 CO -f-l 
 
 
 
 y 
 
 
 
 u 
 
 E 
 
 Vj 
 
 to 
 
 ex 
 
 •r-l 
 
 c 
 
 4-1 
 
 l-< 
 
 >-■ 
 
 C to 
 
 
 1 
 
 c 
 
 • 01 
 
 -w 
 
 4-1 
 
 M 
 
 < 
 
 OJ 
 
 
 
 u 
 
 4-1 
 
 • XI 
 
 to 
 
 to Vj 
 
 
 H 
 
 r3 
 
 cc: 1 
 
 r-l 
 
 c 
 
 4-1 
 
 
 s 
 
 
 
 
 
 c 
 
 <: 3 
 
 (U 
 
 IV 4J 
 
 
 M 
 
 
 
 <u 
 
 <u 
 
 c 
 
 1) 
 
 
 .0 
 
 D- 
 
 to 
 
 D. 
 
 CO 
 
 c 
 
 
 £^ 
 
 5 
 
 • o 
 
 u 
 
 o 
 
 o 
 
 r^. 
 
 • 
 
 a 
 
 
 
 1—1 
 
 • c 
 
 
 
 
 
 
 s: 
 
 W O 
 
 (U 
 
 ■ — ^ 
 
 o 
 
 H 
 
 :2 
 
 CO 
 
 k: 
 
 IX, 
 
 :^ 3. 
 
 Pi 
 
 ^-' a 
 
 CD 
 
 
 
 
 
 
 
 
 
 ac 
 
 
 
 
 
 
 
 
 
 ri 
 
 
 
 
 
 
 
 
 
 «c 
 
 l-< 
 
 
 e>^ 
 
 
 <n 
 
 
 <t 
 
 10 
 
 CJ 
 
 o 
 
 
 o 
 
 
 
 
 
 
 
 
 
 
 o 
 
 
 o 
 
 
 
 
 
 
 
 
 
 
 04 
 
 
 ^1 
 
 
 fa 
 
 
 Ci-, 
 
 fa 
 
 II. 6-1 
 
I- 
 m 
 
 X 
 
 
 C3 
 
 0) 
 CO 
 
 <u 
 
 o 
 o 
 
 
 0) 
 
 
 u 
 
 
 •r( 
 
 ^ 
 
 <M 
 
 V4 
 
 M-l 
 
 ca 
 
 O 
 
 Oi 
 
 
 
 c 
 
 <u 
 
 o 
 
 00 
 
 -r-l 
 
 01 
 
 (0 
 
 1— t 
 
 •r-l 
 
 i-i 
 
 > 
 
 o 
 
 •H 
 
 o 
 
 Q 
 
 0) 
 
 &0 
 0) 
 
 o 
 o 
 
 15 
 
 0) 
 
 o 
 o 
 
 CJ 
 
 •H 
 IM 
 
 o 
 
 c 
 o 
 
 •1-1 
 
 > 
 •H 
 
 Q 
 
 0) 
 
 o 
 
 •w 
 
 ^4-1 
 
 o 
 
 c 
 o 
 
 tH 
 U) 
 •H 
 
 > 
 
 •r-l 
 
 Q 
 
 a. 
 
 -a: 
 
 CO 
 
 S 
 
 (Z) 
 
 w 
 
 V) 
 CO 
 
 w 
 
 CJ 
 
 § 
 
 z 
 
 o 
 
 M 
 
 P4 
 
 —i 
 ex. 
 
 en 
 
 ►— 
 
 
 c 
 o 
 
 u 
 u 
 
 3 
 
 o 
 
 M 
 
 u 
 
 u 
 C 
 0) 
 
 u 
 
 5 
 o 
 
 4-) 
 
 o 
 u 
 
 CO 
 •H 
 
 o 
 
 CO 
 
 I 
 
 lO 
 
 fO 
 
 tn 
 
 a 
 
 CD 
 
 O 
 
 > 
 
 o 
 
 CO o 
 
 y-i z 
 
 (0 
 
 ^ u 
 
 D 
 
 • -a 
 o c 
 
 M 
 
 
 &0 N 
 O u 
 
 o 
 
 O CJ 
 
 • CJ 
 
 Q Q 
 
 O 
 
 o 
 
 c« 
 
 I 
 
 C 
 
 o 
 
 (/) )-l 
 
 c o 
 
 V-1 Cu 
 
 O CO 
 
 CM 
 
 o 
 
 cu 0) 
 
 O CO 
 
 oi 
 
 fu 
 
 u ■ 
 
 o 
 
 a 
 
 O) 
 
 u 
 o 
 
 C3 
 
 vO 
 O 
 
 o 
 ft, 
 
 1^ 
 
 eo 
 
 o 
 
 o 
 
 o 
 
 o 
 
 (t4 
 
 b 
 
 <J\ 
 
 o 
 
 1-4 
 
 o 
 
 1-1 
 
 i-l 
 
 o 
 
 o 
 
 o 
 
 ta 
 
 ri< 
 
 In 
 
 eg 
 
 o 
 
 II. 6-2 
 
u 
 
 CO 
 
 u 
 o 
 
 § 
 
 z 
 
 o 
 
 H 
 O 
 
 u_ a: 
 
 C9 
 
 CO 
 
 ►— 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CJ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 «i: 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 »< 
 
 
 
 
 
 
 t<o 
 
 
 
 
 
 
 
 
 
 
 
 
 
 oo 
 
 
 
 
 
 
 
 
 
 
 
 
 
 «t 
 
 
 
 
 
 
 
 
 
 
 
 
 
 i 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Li- 
 es 
 
 
 
 
 X 
 
 
 
 
 
 
 
 
 
 oe: 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CJ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 £ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 0) 
 
 
 01 
 
 
 
 
 
 
 
 
 
 
 
 u 
 
 
 u 
 
 
 
 
 
 
 
 
 
 
 
 1-< 
 
 
 <r4 
 
 ^ 
 
 ^ 
 
 X 
 
 .ai 
 
 
 
 J>t 
 
 
 
 
 <4-l 
 
 
 >4-l 
 
 u 
 
 14 
 
 u 
 
 M 
 
 
 
 u 
 
 
 
 
 U-t 
 
 
 <W 
 
 ca 
 
 CO 
 
 (0 
 
 <a 
 
 
 
 9 
 
 
 
 as 
 
 C3 
 
 o 
 
 
 o 
 
 P4 
 
 ti* 
 
 P4 
 
 Pi 
 
 
 
 P4 
 
 
 
 ►— 
 
 c 
 
 
 c 
 
 <u 
 
 « 
 
 0) 
 
 4J 
 
 
 0) 
 
 a> 
 
 
 
 «t 
 
 o 
 
 
 o 
 
 60 
 
 bO 
 
 60 
 
 60 
 
 
 f-l 
 
 60 
 
 
 
 ca 
 
 •r^ 
 
 
 1-1 
 
 0) 
 
 0) 
 
 0) 
 
 0) 
 
 
 4J 
 
 <u 
 
 
 
 CO 
 
 U 
 
 
 OS 
 
 r-» 
 
 r-l 
 
 iH 
 
 I-l 
 
 
 4J 
 
 f-l 
 
 
 
 ^J 
 
 •H 
 
 
 •H 
 
 f-l 
 
 r-l 
 
 r-l 
 
 r-* 
 
 
 CO 
 
 f-l 
 
 
 
 
 > 
 
 
 > 
 
 o 
 
 O 
 
 O 
 
 O 
 
 
 0) 
 
 o 
 
 
 
 
 •r^ 
 
 
 t4 
 
 u 
 
 O 
 
 o 
 
 o 
 
 
 CO 
 
 CJ 
 
 
 
 
 Q 
 
 
 Q 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 o 
 
 
 
 
 
 
 ID 
 
 
 
 
 
 
 
 
 ;= 
 
 
 
 
 
 
 c 
 o 
 1-1 
 
 4J 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 c 
 
 
 
 
 _J 
 
 
 
 
 
 
 a 
 
 
 
 
 
 o 
 
 
 
 
 CO 
 
 
 
 
 
 
 V4 
 
 
 
 
 
 •H 
 
 
 
 
 CL. 
 
 
 
 #-v 
 
 
 
 0) 
 
 
 
 
 
 4J 
 
 
 
 
 
 
 
 « 
 
 
 
 
 •r4 
 
 
 
 
 
 u 
 
 
 
 
 
 
 
 C 
 
 
 
 
 CO 
 
 
 
 
 
 Q 
 
 
 
 
 
 
 
 •rt 
 
 
 
 
 c 
 
 
 
 
 
 d 
 
 
 
 
 
 
 
 h 
 
 
 
 
 o 
 
 
 
 
 
 CU 
 
 
 
 
 
 
 
 0) 
 
 
 
 
 u 
 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 Q> 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 e 
 
 
 
 
 ex 
 
 
 
 
 
 n 
 
 
 
 
 
 
 
 •H 
 
 
 
 
 c 
 
 
 
 
 
 TS 
 
 
 
 
 
 
 
 6C 
 
 
 
 
 •r4 
 
 
 
 
 
 1-4 
 
 
 
 
 
 
 
 
 e 
 
 
 
 
 CO 
 
 
 6C 
 
 
 
 
 1-4 
 
 
 
 
 
 T3 
 
 
 
 H 
 
 
 
 
 n 
 
 
 c 
 
 
 
 
 o 
 
 
 
 
 
 
 
 
 
 
 
 
 v 
 
 
 fl 
 
 
 
 
 w 
 
 
 
 
 
 4J 
 
 
 
 •o 
 
 
 
 
 u 
 
 
 n 
 
 
 
 
 t 
 
 
 
 
 
 c 
 
 
 
 o 
 
 
 
 
 o 
 
 
 CO 
 
 
 
 
 T3 
 
 
 
 
 
 o 
 
 
 
 o 
 
 
 
 
 h 
 
 
 V 
 
 ^~\ 
 
 
 
 •H 
 
 
 
 
 
 u 
 
 
 
 h 
 
 
 
 
 P4 
 
 
 a 
 
 •o 
 
 
 
 3 
 
 
 
 
 
 V-' 
 
 
 
 
 
 
 
 
 o 
 
 c . 
 
 
 
 a 
 
 
 
 
 
 
 
 
 o- 
 
 
 m 
 
 
 J= 
 
 
 u 
 
 to 
 
 
 
 •H 
 
 
 
 
 
 iH 
 
 
 
 4J 
 
 
 c 
 
 
 m 
 
 
 fa 
 
 t-i 
 
 
 c 
 
 
 iJ 
 
 
 
 
 
 r~ 
 
 
 
 " 
 
 
 o 
 
 
 1-1 
 
 
 
 >> 
 
 
 E 
 
 
 
 
 
 
 */i 
 
 o^ 
 
 
 
 B u 
 
 
 ^ 
 
 
 fa 
 
 
 X 
 
 u 
 
 
 3 
 
 
 h 
 
 
 
 
 
 r-l 
 
 
 
 W O 
 
 
 4J 
 
 
 
 
 CO 
 
 CO 
 
 
 I-l 
 
 
 e 
 
 
 
 
 
 
 
 (U 4J 
 
 
 fd 
 
 
 >■ 
 
 
 •H 
 
 X 
 
 
 O 
 
 
 U-l 
 
 
 
 
 
 
 
 ^4 lH 
 
 
 u 
 
 
 4J 
 
 
 fa 
 
 
 
 u 
 
 
 
 
 
 
 ^ 
 
 
 4J -o O 
 
 
 •r* 
 
 
 iJ 
 
 
 
 U-l 
 
 
 
 
 <u 
 
 
 
 
 ^3 
 
 u 
 
 
 M W r^ 
 
 
 M-l 
 
 
 a 
 
 
 > 
 
 o 
 
 
 c 
 
 
 6£ 
 
 
 
 
 «« 
 
 o 
 u 
 o 
 
 
 1 
 C M 00 
 O C vO 
 
 
 •rl 
 O 
 
 
 fa 
 
 0) 
 
 
 4J 
 CO 
 
 4J 
 
 
 o 
 1-1 
 
 4-> 
 
 
 3 
 
 C4-I 
 f4 
 
 
 
 
 
 4J 
 
 (U 
 
 
 •<-l 
 
 0] 
 
 s 
 
 
 u 
 
 
 fa 
 
 1-1 
 
 
 CO 
 
 
 u 
 
 
 
 
 
 l-l 
 
 (X 
 
 
 4J 4J •• 
 
 (U 
 
 ^^ 
 
 
 3 
 
 
 
 CO 
 
 
 1-4 
 
 
 4J 
 
 
 o 
 
 
 
 o 
 
 10 
 
 
 C 3 /-^ 
 
 c 
 
 P4 
 
 
 4J 
 
 
 >^ 
 
 u 
 
 
 I-l 
 
 
 . c 
 
 /^ 
 
 CJ 
 
 
 
 C- 
 
 c 
 
 
 (0 .C 1-1 
 
 1-1 
 
 g 
 
 
 3 
 
 
 O 
 
 5 
 
 
 1-1 
 
 
 CU 
 
 s 
 
 S 
 
 
 
 0) 
 
 M 
 
 
 r-l tH I-l 
 
 T-l 
 
 
 fa 
 
 
 <4-l 
 
 > 
 
 
 4-t 
 
 
 u 
 
 
 3 
 
 
 
 ai 
 
 O 
 
 
 PM M ^ 
 4J ro 
 
 <u 
 
 T3 
 
 P 
 
 
 O 
 
 
 fa 
 
 iH 
 
 c 
 
 
 CO 
 
 1-4 
 
 
 CO 
 
 v3 
 
 CO 
 
 
 
 m 
 
 Q 
 
 
 C c >3- 
 
 1-1 
 
 o 
 
 /-\ 
 
 4J 
 
 ^-\ 
 
 Q 
 
 3 
 
 
 Q 
 
 
 3 
 
 < 
 
 O 
 
 
 
 en 
 
 CO 
 
 u 
 
 O O ^ 
 
 3 
 
 «-l 
 
 OT 
 
 
 CO 
 
 U 
 
 v—" 
 
 
 
 
 m 
 
 
 4J 
 
 
 
 <U 
 
 O 
 
 u 
 
 •-^ O M 
 
 o 
 
 
 ptj 
 
 •o 
 
 Exj 
 
 
 
 
 o 
 
 
 M 
 
 en 
 
 
 
 
 >j 
 
 N— • 
 
 o 
 
 u -^ C 
 
 
 m 
 
 § 
 
 (1) 
 
 § 
 
 IM 
 
 o 
 
 4-1 
 
 4J 
 
 
 (U 
 
 ca 
 
 4J 
 
 
 
 M 
 
 
 a. 
 
 CO 1-1 
 
 c 
 
 c 
 
 g 
 
 4J 
 
 2 
 
 O 
 
 f-l 
 
 U 
 
 
 ^N 
 
 > 
 
 X 
 
 U 
 
 
 
 O 
 
 c 
 
 <u 
 
 U 4J h 
 
 o 
 
 o 
 
 >— / 
 
 CO 
 
 V^ 
 
 
 i-H 
 
 o 
 
 c 
 
 to 
 
 
 CU 
 
 O 
 
 
 toi 
 
 V4 
 
 o 
 
 aJ 
 
 U CO <U 
 
 1-1 
 
 1-1 
 
 
 t-< 
 
 
 c 
 
 (U 
 
 a, 
 
 o 
 
 2 n 
 
 CO 
 
 H 
 
 fa 
 
 
 g 
 
 ^ 
 
 U) 
 
 
 to Vl OJ 
 
 4J 
 
 4-1 -U 
 
 J-l 4J 
 
 CD 
 
 J-l u 
 
 o 
 
 j: 
 
 CU 
 
 1-1 
 
 CO 
 
 s^ 
 
 (U 
 
 
 
 i-l 
 
 c 
 
 c o c 
 
 CO 
 
 U CO 
 
 CO )-l 
 
 a; 
 
 CO U 
 
 1-1 
 
 o 
 
 Pi 
 
 4J 
 
 ■ <u 
 
 
 Di 
 
 
 4J 
 
 3 
 
 o 
 
 O jr 1-1 
 
 3 
 
 O T3 
 
 c o 
 
 
 C O 
 
 4J 
 
 u 
 
 
 1-1 
 
 v-^ o 
 
 u 
 
 a> 
 
 
 
 
 c 
 
 O 
 
 tA 
 
 g 60 M 
 
 1—1 
 
 O- C 
 
 )-i ex 
 
 CO 
 
 Vj O- 
 
 CO 
 
 CO 
 
 j-i 
 
 T3 
 
 a. 
 
 fa 
 
 ^ 
 
 4J 
 
 
 
 d 
 
 o 
 
 U 
 
 0) -W C 
 
 CO 
 
 0) <U 
 
 U <1) 
 
 e 
 
 W 0) 
 
 O 
 
 s 
 
 o 
 
 -a 
 
 U V 
 
 1 
 
 c 
 
 O 
 
 
 
 <-l 
 
 
 o 
 
 o > w 
 
 > 
 
 y—\ Oi, 6 
 
 od 
 
 (U 
 
 ftj 
 
 •7-i 
 
 
 CO 
 
 < 
 
 CU o:: 
 
 c 
 
 ■H 
 
 CO 
 
 
 
 Ph 
 
 a 
 
 (U 
 
 CO 
 
 u 
 
 C/5 fc 
 
 • 
 
 rH 
 
 • 
 
 U-l 
 
 • 
 
 U 
 
 
 >^ 
 
 (U 
 
 0) 
 
 u 
 
 
 
 
 Q 
 
 o. 
 
 CJ 5C T3 
 
 
 § fa " 
 
 a w 
 
 J3 
 
 CJ en 
 
 1-1 
 
 S 
 
 4J 
 
 •a 
 
 Q CO 
 
 a; 
 
 S 
 
 4-1 
 
 
 
 CJ 
 
 
 m 
 
 Ph O 
 
 p^ 
 
 ^ 
 
 O 
 
 Ph 
 
 ■a 
 
 
 c 
 
 1-1 
 
 pL^ 
 
 V4 
 
 
 c 
 
 
 
 Ph 
 
 • 
 
 C 
 
 ^ . O 
 
 Q 
 
 • ^ 
 
 l>l 
 
 • sl 
 
 o 
 
 • 
 
 o 
 
 o 
 
 • s 
 
 o 
 
 • 
 
 o 
 
 
 
 ►< 
 
 o 
 
 M 
 
 W 
 
 ^2 (S 
 
 Pi 2 
 
 fa 
 
 Pli g 
 
 S 
 
 O 
 
 CJ 
 
 < 
 
 •-) z 
 
 CO 
 
 3 
 
 CJ 
 
 
 "T 
 
 
 
 
 
 
 
 
 
 
 
 
 
 3B 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 <*1 
 
 
 «* 
 
 ir» 
 
 « 
 
 n. 
 
 00 
 
 
 a> 
 
 o 
 
 
 
 ^ 
 
 i-l 
 
 
 f-« 
 
 t-i 
 
 1-4 
 
 1-1 
 
 f-l 
 
 
 f-< 
 
 CM 
 
 
 
 o 
 
 
 o 
 
 o 
 
 o 
 
 o 
 
 o 
 
 
 o 
 
 o 
 
 
 CJ 
 
 PUI 
 
 
 (^ 
 
 ft* 
 
 b 
 
 fa 
 
 fa 
 
 
 fa 
 
 fa 
 
 
 
 II. 6-3 
 

 C/1 
 
 >- 
 
 UO 
 
 rn 
 
 C/2 
 
 ex. 
 
 -a: 
 
 C/3 
 
 cr: 
 
 CJ 
 
 rr: 
 
 C3 
 
 a. 
 
 
 
 — 1 
 
 5 
 
 
 
 no 
 
 1— 1 
 
 
 H 
 
 CJ 
 
 U 
 
 Q. 
 
 UJ 
 
 l-k. 
 
 o 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 <_> 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 «< 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 cc 
 
 
 
 X 
 
 
 
 
 
 
 
 
 
 
 X 
 
 
 
 
 oo 
 
 
 
 
 
 
 
 
 
 
 
 
 
 1 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ■»»; 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 2 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 — 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 LA. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 cz> 
 a: 
 
 X 
 
 
 X 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 C_3 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 3E 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 <u 
 
 
 
 aj 
 
 
 0) 
 
 
 
 ly 
 
 CJ 
 
 
 
 
 
 
 
 
 w 
 
 
 
 o 
 
 
 o 
 
 
 
 o 
 
 o 
 
 
 
 
 
 
 
 
 ■H 
 
 
 
 •H 
 
 
 1-1 
 
 
 
 1-1 
 
 •r-l 
 
 
 
 
 
 ^ 
 
 
 ^ 
 
 U-l 
 
 ^ 
 
 
 U-l 
 
 
 iw 
 
 
 
 u-l 
 
 U4 
 
 
 
 
 Z 
 
 ^1 
 
 
 n 
 
 <« 
 
 u 
 
 
 H-l 
 
 
 u-l 
 
 
 
 u-l 
 
 <w 
 
 
 
 
 CD 
 
 CO 
 
 
 CO 
 
 O 
 
 CO 
 
 
 o 
 
 
 o« 
 
 
 
 O 
 
 O 
 
 
 
 
 »— 
 
 PU 
 
 
 PLJ 
 
 
 (U 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 C 
 
 
 
 c 
 
 
 c 
 
 
 
 c 
 
 C 
 
 
 
 
 0) 
 
 
 aj 
 
 O 
 
 <u 
 
 
 (U o 
 
 
 0) o 
 
 
 aj 
 
 o 
 
 o 
 
 
 
 
 C^ 
 
 M 
 
 
 M 
 
 •t-( 
 
 00 
 
 
 r-l -r^ 
 
 
 >-> -rA 
 
 
 1-1 
 
 •H 
 
 ••-( 
 
 
 
 
 o 
 
 ^ 
 
 
 (U 
 
 U) 
 
 0) 
 
 
 u en 
 
 
 U (0 
 
 
 ■U 
 
 to 
 
 CO 
 
 
 
 
 ~^ 
 
 
 I-l 
 
 •r-l 
 
 r-l 
 
 
 U -M 
 
 
 ■U -I-l 
 
 
 4J 
 
 •I-l 
 
 •rl 
 
 
 
 
 
 i-t 
 
 
 1-4 
 
 > 
 
 I-l 
 
 
 to > 
 
 
 (0 > 
 
 
 to 
 
 > 
 
 > 
 
 
 
 
 
 O 
 
 
 O 
 
 ■r-l 
 
 o 
 
 
 0) -r-l 
 
 
 0) -H 
 
 
 CJ 
 
 •r^ 
 
 •^^ 
 
 
 
 
 
 CJ 
 
 
 CJ 
 
 Q 
 
 o 
 
 
 W Q 
 
 
 00 Q 
 
 
 CO 
 
 Q 
 
 Q 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 e 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 3 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 •1-1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 o 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 o. 
 
 
 
 as 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 %. 
 
 
 
 CD 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CO 
 
 
 
 ^_ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 «£ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 f-l 
 
 
 
 f_> 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 4.J 
 
 
 
 
 
 
 
 
 c 
 
 
 
 —J 
 
 
 
 
 
 
 
 c 
 
 
 
 
 
 
 
 
 o 
 
 
 
 CO 
 
 
 
 
 
 
 
 n 
 
 
 
 
 
 
 
 
 •r-l 
 
 
 
 rD 
 
 
 
 
 
 
 
 t-l 
 
 
 
 
 
 
 
 
 4J 
 
 
 
 o- 
 
 
 
 
 
 
 
 c 
 
 
 
 
 
 
 
 
 to 
 
 c 
 
 >4 
 
 
 
 
 
 o> 
 
 
 
 
 
 o 
 
 
 
 c 
 
 
 
 
 
 
 0) 
 
 
 
 
 
 :5 
 
 
 
 
 
 •r-l 
 4J 
 
 to 
 
 
 
 o 
 
 
 
 
 
 
 4J 
 
 c 
 
 M 
 
 
 
 
 ^" 
 
 
 <t 
 
 
 
 
 
 (0 
 
 
 
 j:: 
 
 
 
 
 
 
 r~l 
 
 
 
 
 
 
 
 
 
 
 
 
 
 tn 
 
 
 
 
 
 
 CJ^ 
 
 
 
 
 c 
 
 
 
 ; 
 
 * 
 
 
 
 
 
 
 
 to 
 
 
 
 
 
 
 I— 1 
 
 
 
 
 U) 
 
 .^>. 
 
 
 
 
 
 g 
 
 
 
 S 
 
 
 
 
 
 
 
 
 
 
 <u 
 
 <»■ 
 
 
 
 
 
 (U 
 
 
 
 
 
 
 
 
 
 0) 
 
 
 
 
 •H 
 
 -w 
 
 
 
 
 
 n 
 
 
 
 * 
 
 
 
 
 
 
 j: 
 
 
 
 
 CJ 
 
 >^ 
 
 O 
 
 
 
 
 
 
 
 
 c 
 
 
 
 
 
 
 ij 
 
 
 
 
 
 u 
 
 1-1 
 
 
 
 u 
 
 
 
 1-4 
 
 
 
 OJ 
 
 
 
 
 
 
 
 
 
 
 
 , — 1 
 
 3 
 
 
 
 
 c 
 
 
 
 to 
 
 
 
 <u 
 
 
 
 
 
 
 u-l 
 
 
 
 oo 
 
 f*^ 
 
 i-l 
 
 4-J 
 
 
 
 (U 
 
 
 
 4J 
 
 
 
 ■a 
 
 
 
 
 
 
 O 
 
 
 
 
 CJN 
 
 w 
 
 C 
 
 
 
 B 
 
 
 
 c 
 
 
 
 >-i 
 
 
 
 
 <M 
 
 
 
 
 
 S" 
 
 
 Q) 
 
 
 
 s 
 
 
 
 0) 
 
 
 
 at 
 
 
 
 
 t^ 
 
 
 "L 
 
 
 
 CD 
 
 *"* 
 
 -a 
 
 E 
 
 
 
 u 
 
 
 
 g 
 
 
 
 X) 
 
 
 
 
 <T> 
 
 
 ttf 
 
 
 
 ZJZ 
 
 •H 
 
 a 
 
 
 
 to 
 
 
 
 t^ 
 
 
 
 < 
 
 
 
 
 1-1 
 
 
 c 
 
 
 
 >— 
 
 D 
 
 --N o 
 
 
 
 J) 
 
 
 
 u 
 
 
 
 
 
 
 
 
 
 •r-l 
 
 
 
 =5 
 
 cr 
 
 CO '-' 
 
 
 
 ■< 
 
 
 
 <u 
 
 
 
 r 
 
 
 
 
 
 
 -a 
 
 
 
 -a: 
 
 1 
 
 •r-l 
 .—I 
 O 
 
 s > 
 
 ^-^ Q 
 
 G -a 
 3 c 
 
 O nJ 
 
 01 
 
 c 
 
 '-N 
 
 c 
 o 
 1-1 
 *J 
 
 D 
 
 r-l 
 
 1— 1 
 
 
 
 o 
 
 <U 
 
 x: 
 
 
 
 
 
 0) 
 
 x: 
 
 
 
 en 
 
 to 
 
 to 
 
 
 J* 
 
 CU 
 (U 
 
 u 
 o 
 u 
 
 
 
 
 
 U 
 
 •r-l 
 
 o 
 
 o 
 
 
 
 ij 
 
 
 
 u-l 
 
 
 
 0) 
 
 XJ 
 
 u 
 
 x: 
 
 
 
 
 C 
 
 ca C 
 
 Sj 
 
 c 
 
 Oi 
 
 
 
 to 
 
 
 
 O 
 
 
 
 o 
 
 CL 
 
 a- 
 
 u 
 
 
 
 
 •H 
 
 w 
 
 CQ 
 
 »— 1 
 
 
 
 
 
 
 
 
 
 
 Q 
 
 •H 
 
 *t-i 
 
 
 
 
 
 
 . -r-l 
 
 
 
 u 
 
 •-N 
 
 
 i_) 
 
 
 
 i-t 
 
 y^ 
 
 
 u 
 
 U 
 
 u 
 
 C 
 
 
 
 
 C 
 
 ►J w 
 
 C 
 
 ^ 
 
 o 
 
 to 
 
 
 c 
 
 /^■\ 
 
 
 C 
 
 CO 
 
 
 PLI 
 
 ^x O 
 
 o 
 
 •H 
 
 
 
 
 o 
 
 O 
 
 •r^ 
 
 w 
 
 U-l 
 
 1 
 
 
 o 
 
 to 
 
 
 ty 
 
 
 
 
 CO to 
 
 (0 
 
 
 
 
 
 •^^ 
 
 • Q 
 
 
 u 
 
 
 
 e 
 
 fel 
 
 
 E 
 
 
 U 
 
 b 2 
 
 3 
 
 ■o 
 
 
 
 
 u 
 
 7Z 
 
 a 
 
 il »-> 
 
 7) 
 
 
 o 
 
 
 
 (U 
 
 k; 
 
 
 C 
 
 ^ CO 
 
 : -^ c 
 
 t) 
 
 
 
 
 a 
 
 <f) 
 
 ^. 
 
 u Vj 
 
 c 
 
 v,^ 
 
 
 0-1 
 
 ?: 
 
 
 AJ 
 
 ^— ^ 
 
 
 CJ 
 
 to to to 
 
 JT 
 
 • 
 
 
 kAj 
 
 g 
 
 .« en 
 
 ro 
 
 ■j^ o 
 
 o 
 
 
 
 to 
 
 ^v.^ 
 
 
 ro 
 
 
 
 > 
 
 ^—^ ^ 
 
 CJ u, 'C 
 
 ro 
 
 '■">■ 
 
 
 t 
 
 3 
 
 Q) a 
 
 — ' 
 
 V O. 
 
 •rJ 
 
 M 
 
 
 J2 
 
 
 
 X 
 
 S 
 
 
 .—1 
 
 
 to y. ■ 
 
 •r^ 
 
 H 
 
 
 1 — 
 
 cn 
 
 a o 
 
 
 ■> <u 
 
 u 
 
 Vj 
 
 u 
 
 <; 
 
 •H 
 
 LJ 
 
 < 
 
 c 
 
 i_» 
 
 o 
 
 •r-l -O 
 
 to 5^. -a 
 
 •—1 
 
 M 
 
 
 K* 
 
 c 
 
 ^ o 
 
 U-, 
 
 u a: 
 
 to 
 
 a 
 
 u 
 
 
 --I 
 
 u 
 
 
 to 
 
 Vj 
 
 to 
 
 ,-1 0) 
 
 a v-' o 
 
 X 
 
 xt 
 
 
 
 o 
 
 PL, u 
 
 O 
 
 to 
 
 -a 
 
 r\ 
 
 
 
 ^ 
 
 —1 
 
 o 
 
 C 
 
 v) 
 
 o 
 
 1 
 
 r-l jr 
 
 O X 
 
 3 
 
 ■ — ^ 
 
 
 
 o 
 
 o eu 
 
 
 X ^ 
 
 
 c 
 
 a. 
 
 o 
 
 o 
 
 D. 
 
 O 
 
 to 
 
 D. 
 
 to 
 
 0) to 
 
 O J= to 
 
 C 
 
 
 
 
 
 "y^ 
 
 O 
 
 u 
 
 o 
 
 •r^ 
 
 o 
 
 -,-< 
 
 c 
 
 CJ 
 
 ■rA 
 
 Q 
 
 CJ 
 
 3 
 
 C —1 
 
 Ij O •r^ 
 
 
 U 
 
 
 
 u 
 
 u 
 
 ts 
 
 C CO 
 
 
 
 a 
 
 Di 
 
 u 
 
 'H 
 
 erf 
 
 U 
 
 
 a: 
 
 o 
 
 ■r^ .—1 
 
 Cu r: -^ 
 
 o 
 
 a. 
 
 
 
 <u 
 
 . CiJ 
 
 to 
 
 (0 l-> 
 
 J-) 
 
 
 3 
 
 CU 
 
 
 D 
 
 • 
 
 
 CJ 
 
 D. X 
 
 •r^ XI 
 
 ^ 
 
 Um 
 
 
 
 3 
 
 Q 
 
 ).l 
 
 V -LJ 
 
 o 
 
 to 
 
 to 
 
 r— J 
 
 to 
 
 to 
 
 1— 1 
 
 < 
 
 to 
 
 3 
 
 CO 3 
 
 U Pu 3 
 
 
 
 
 
 O 
 
 ^ 
 
 o 
 
 ■o c 
 
 u 
 
 
 fa 
 
 5^: 
 
 f— 1 
 
 
 b 
 Ci 
 
 .—1 
 
 
 u, 
 
 cr 
 
 a. 
 
 CU D. 
 
 o 
 
 c 
 
 
 
 ^ 
 
 
 l-> 
 
 o o 
 
 o 
 
 • 
 
 O 
 
 • 
 
 
 
 • 
 
 g 
 
 < 
 
 • c 
 
 lii • C 
 
 ^ 
 
 o 
 
 
 
 < M 
 
 LO 
 
 w O 
 
 Qi 
 
 ^ 
 
 PJ 
 
 —> 
 
 cw 
 
 •-) 
 
 ' 
 
 •-) r) 
 
 : Qi r) 
 
 
 
 
 CD 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ar 
 
 t-< 
 
 
 f<J 
 
 
 en 
 
 
 »* 
 
 
 m 
 
 
 vO 
 
 
 f^ 
 
 
 
 
 1 
 
 <M 
 
 
 CM 
 
 
 f^J 
 
 
 C>J 
 
 
 CM 
 
 
 «M 
 
 
 OJ 
 
 
 
 
 . — 1 
 
 . O 
 
 
 O 
 
 
 o 
 
 
 o 
 
 
 o 
 
 
 O 
 
 
 o 
 
 
 
 
 «s 
 
 (ii 
 
 
 in 
 
 
 (M 
 
 
 u* 
 
 
 Ct4 
 
 
 h 
 
 
 Pu 
 
 
 
 
 CD 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 , 
 
 
 
 
 
 
 
 
 
 
 
 . 
 
 
 
 
 
 
 II.6-A 
 
-X 
 Cxi, 
 t3 
 (=3 
 
 00 
 
 w 
 
 to 
 w 
 o 
 o 
 
 O 
 
 CO ^ 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CJ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 «« 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 X 
 
 
 
 
 
 
 
 
 
 X 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 
 
 
 o<r> 
 
 
 
 
 
 
 
 
 
 
 
 
 
 «« 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 =5 
 
 
 
 
 
 
 
 
 
 
 
 
 
 — 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 L^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CD 
 
 
 
 
 
 
 
 
 
 
 
 
 
 o; 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CJ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 5£ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 4J 
 
 
 <U 
 
 
 <u 
 
 <u 
 
 
 (U 
 
 
 
 
 
 
 CJ 
 
 
 u 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 •H 
 
 
 1-1 
 
 
 •rl 
 
 1-( 
 
 
 •1-1 
 
 
 
 
 ^ 
 
 
 ■W 
 
 
 <4-l 
 
 
 4-1 
 
 <M 
 
 
 14-1 ^ 
 
 
 ^ 
 
 2= 
 
 (-1 
 
 
 <M 
 
 
 14-1 
 
 
 M-l 
 
 M^ 
 
 
 14-1 V^ 
 
 
 w 
 
 <=> 
 
 CL4 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CO 
 fa 
 
 
 fa 
 
 C3 
 
 
 
 c 
 
 
 C 
 
 
 c 
 
 C3 
 
 
 c 
 
 
 
 (U 
 
 
 
 
 
 
 
 
 
 
 
 
 
 <u 
 
 
 *L 
 
 00 
 
 
 •rl 
 
 
 •H 
 
 
 •r4 
 
 t4 
 
 
 1-1 00 
 
 
 00 
 
 0) 
 
 
 (0 
 
 
 CO 
 
 
 CO 
 
 (0 
 
 
 CO (U 
 
 
 V 
 
 
 1-1 
 
 
 •H 
 
 
 •M 
 
 
 •rl 
 
 •t-l 
 
 
 1-1 r-( 
 
 
 f-4 
 
 
 .-1 
 
 
 > 
 
 
 > 
 
 
 > 
 
 > 
 
 
 > r-l 
 
 
 r-l 
 
 
 o 
 
 
 1-1 
 
 
 1-1 
 
 
 •H 
 
 •M 
 
 
 •rl 
 
 
 
 
 
 CJ 
 
 
 a 
 
 
 Q 
 
 
 Q 
 
 Q 
 
 
 Q 
 
 
 
 
 
 
 
 
 
 ^->> 
 
 
 
 
 
 
 
 
 
 
 
 
 
 U 
 
 
 
 
 
 
 
 
 
 
 
 
 
 P- 
 
 
 
 
 
 
 
 
 
 
 >% 
 
 
 
 fe 
 
 
 
 
 
 
 
 
 
 
 60 
 
 
 
 
 
 
 
 
 
 
 
 
 
 O 
 
 
 
 
 
 
 
 
 
 
 
 
 
 r-l 
 
 
 
 0) 
 
 
 
 
 
 
 
 
 
 
 O 
 
 
 
 4J 
 
 
 
 
 
 
 
 
 
 e 
 
 
 T^ 
 
 
 
 CB 
 
 
 
 
 
 
 
 
 
 o 
 
 
 ^ 
 
 
 
 H 
 
 
 
 
 
 
 
 
 
 >-i 
 
 
 o 
 
 
 
 4J 
 
 
 
 
 
 
 
 
 as 
 
 <4-( 
 
 
 M 
 
 
 
 c 
 
 
 
 
 
 to 
 
 
 
 
 C3 
 
 
 
 o 
 
 
 
 ai 
 
 
 
 
 
 e 
 
 
 
 
 1 
 
 c 
 
 
 •w 
 
 
 
 
 
 
 
 
 
 (U 
 
 
 
 
 <S 
 
 o 
 
 
 s 
 
 
 
 c 
 
 
 
 
 
 1-1 
 
 
 
 
 CJ 
 
 •H 
 
 
 
 
 
 
 
 
 
 
 
 ,0 
 
 
 
 
 
 4J 
 
 
 r-l 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Zj 
 
 O 
 
 
 CO 
 
 
 
 
 
 
 
 
 I-l 
 
 
 
 
 CO 
 
 3 
 
 
 T-l 
 
 
 
 c 
 
 
 
 
 
 fa 
 
 
 
 
 =D 
 
 X) 
 
 
 M 
 
 
 
 1-1 
 
 
 
 
 
 
 
 
 
 a. 
 
 O 
 
 
 4-1 
 
 CO 
 
 3 
 
 c 
 
 
 
 0) 
 
 
 n 
 cu 
 
 
 
 
 
 c 
 
 1-1 
 
 4-J 
 
 to 
 
 
 
 
 
 4-) 
 
 
 H 
 
 
 
 
 
 
 
 C 
 
 
 
 
 u 
 
 
 
 
 
 ro 
 
 
 
 PM 
 
 
 
 x: 
 
 
 CO 
 
 
 
 
 cu 
 
 
 
 
 
 ^■ 
 
 
 ^ 
 
 
 
 
 M 
 
 
 i-I 
 
 
 
 
 C- 
 
 
 
 
 
 -o 
 
 4-1 
 
 4J 
 
 
 O 
 
 
 
 
 •r-I 
 
 
 PL4 
 
 
 
 
 
 
 
 
 
 
 XJ 
 
 c 
 
 ^ 
 
 
 14-1 
 
 Vj 
 
 
 
 lii 
 
 
 
 
 
 
 
 
 
 
 
 
 a) 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 •0 
 
 
 
 
 
 U 
 
 o 
 
 r-l 
 
 
 >. 
 
 14-1 
 
 
 
 U 
 
 
 CM 
 
 
 
 
 c 
 
 
 
 
 
 c 
 
 C 
 
 P4 
 
 
 4-1 
 
 
 
 
 
 
 
 fa 
 
 
 > 
 
 
 
 to 
 
 
 
 
 
 o 
 
 o 
 
 
 
 CU 
 
 4-1 
 
 
 
 >4-l 
 
 
 
 
 13 
 
 
 
 
 
 
 
 
 
 CJ) 
 
 c 
 
 c 
 o 
 
 
 1-1 
 CJ 
 
 O 
 
 C 
 CO 
 1-1 
 
 
 
 4J 
 
 c 
 
 
 
 
 4-1 
 C 
 
 
 3 
 U 
 CO 
 
 
 
 1-1 
 
 
 
 
 
 ^ 
 
 CNl 
 
 •H 
 
 4J 
 
 
 CO 
 
 P~t 
 
 
 
 CD 
 
 
 1-1 
 
 
 
 
 
 to 
 
 •rl 
 
 
 
 oo 
 
 r^ 
 
 cu 
 
 en 
 
 
 
 
 
 
 r-l 
 
 
 
 
 S 
 
 
 
 (U 
 
 4-1 
 
 
 
 
 cr* 
 
 4-1 
 
 vj 
 
 
 <4.4 
 
 c 
 
 
 
 CM 
 
 
 CO 
 
 
 <u 
 
 
 
 
 
 to 
 
 
 
 en 
 
 , — 1 
 
 o 
 
 4-) 
 
 
 o 
 
 
 
 ^s 
 
 
 
 
 4-1 
 
 
 1-1 
 
 
 
 
 
 
 
 
 CD 
 
 
 !-l 
 
 CO 
 
 -— X 
 
 
 1-1 
 
 Vj 
 
 
 c 
 
 
 3 
 
 
 > 
 
 
 
 4J 
 
 •rl 
 
 
 
 or 
 
 P-i 
 
 C 
 
 CO 
 
 60 
 
 4-) 
 
 
 
 
 
 
 
 G 
 
 
 0) 
 
 
 
 c 
 
 <4-l 
 
 
 
 ^~ 
 
 
 o 
 
 taij 
 
 C 
 
 CO 
 
 4-1 
 
 
 1-1 
 
 
 C 
 
 
 oi 
 
 
 
 CO 
 
 1-1 
 
 
 
 =5 
 
 £1 
 
 E 
 
 ^ 
 
 •1-1 
 
 U 
 
 CJ 
 
 
 4-1 
 
 
 M 
 
 
 
 
 
 r-l 
 
 T) 
 
 
 
 «« 
 
 tfl 
 
 a; 
 
 2 
 
 4-1 
 
 u 
 
 cu 
 
 
 CO 
 
 
 
 
 6£ 
 
 
 
 fa 
 
 
 
 
 
 
 •r-l 
 
 o 
 
 >w/ 
 
 0) 
 
 C/) 
 
 a. 
 
 
 u 
 
 
 ?- 
 
 
 C 
 
 
 
 
 S 
 
 
 
 
 l^ 
 
 
 
 cu 
 
 c 
 
 w 
 
 
 u 
 
 
 U 
 
 
 1-1 
 
 
 
 CM 
 
 
 
 
 
 
 c 
 
 N 
 
 2: 
 
 
 
 C 
 
 
 w 
 
 
 
 
 
 J-i 
 
 
 
 
 
 M 
 
 
 
 
 U-( 
 
 •1-1 
 
 4J 
 
 
 e 
 
 H 
 
 
 c 
 
 
 4-1 
 
 
 (U 
 
 
 
 
 
 
 
 
 
 O 
 
 
 C 
 
 1-1 
 
 cu 
 
 
 
 
 
 
 CO 
 
 
 <u 
 
 
 
 s 
 
 U-l 
 
 
 fa 
 
 
 
 
 4J 
 
 •r^ 
 
 CO 
 
 
 
 U 
 
 
 E 
 
 
 M 
 
 
 c 
 
 
 
 <u 
 
 
 
 e 
 
 
 
 U3 
 
 c 
 
 s 
 
 D 
 
 
 Q 
 
 
 cu 
 
 
 
 
 
 1-( 
 
 
 
 iH 
 
 C 
 
 
 
 
 4-1 
 
 Zi 
 
 T) 
 
 c: 
 
 ^4 
 
 CO 
 
 
 Q 
 
 4J 
 
 JD 
 
 
 « 
 
 
 
 > 
 
 
 
 
 
 
 
 O 
 
 B 
 
 r-l 
 
 c 
 
 
 
 3 
 
 
 
 &. 
 
 CO 
 
 
 c 
 
 
 
 <u 
 
 iH 
 
 
 <4-l 
 
 
 
 <U 
 
 ■r-l 
 
 O 
 
 < 
 
 14-1 
 
 s.-' 
 
 
 ■0 
 
 •rl 
 
 rJ 
 
 
 w 
 
 
 
 Pi 
 
 4-1 
 
 
 
 
 
 
 a 
 
 H 
 
 CJ 
 
 
 
 
 
 C 
 
 U 
 
 
 
 
 
 
 
 to 
 
 
 
 
 
 CO 
 
 (U 
 
 
 
 
 > 
 
 • 
 
 
 CO 
 
 ^^ 
 
 .r 
 
 
 fa 
 
 
 
 T3 
 
 u 
 
 ^-\ 
 
 3 
 
 
 
 < 
 
 c 
 
 • 
 
 4-1 
 
 u 
 
 u 
 
 
 
 CO CO 
 
 
 
 
 
 
 
 
 C 
 
 <u 
 
 to 
 
 CU 
 
 
 
 
 X 
 
 Q 
 
 
 CO 
 
 CO 
 
 
 ■ 4J 
 
 h 3 
 
 U 
 
 
 
 
 to 
 
 -0 
 
 fa 
 
 1-1 
 
 
 
 r-i 
 
 CjJ 
 
 
 •u 
 
 c 
 
 
 
 C 
 
 § c 
 
 CO 
 
 
 
 /■^ 
 
 e 
 
 
 •rl 
 
 2j 
 
 > 
 
 *->. 
 
 
 CO 
 
 
 M 
 
 cu 
 
 p 
 
 •% 
 
 
 0) 
 
 g CO 
 
 (U 
 
 ^1v 
 
 u 
 
 CO 
 
 3 
 
 C 
 
 to 
 
 ^ 
 
 CU 
 
 CO. 
 
 UJ 
 
 O 
 
 01 
 
 I-l 
 
 4J 
 
 3 
 
 c 
 
 
 E 
 
 ^ S 
 
 CO 
 
 CO 
 
 
 
 & 
 
 •T3 
 
 
 
 C 
 
 v—' 
 
 fa 
 
 ^ 
 
 —J 
 
 •H 
 
 1-1 
 
 r-l 
 
 c 
 
 00 
 
 
 
 
 •rl 
 
 
 (U 
 
 
 C4-I 
 
 C 
 
 •rl 
 
 
 
 
 
 2j 
 
 
 tl. 
 
 CT3 
 
 (U 
 
 cu 
 
 
 01 
 
 u 
 
 J-l 
 
 >. -0 
 
 Qi 
 
 
 s 
 
 CO 
 
 CO 
 
 
 
 4-1 4J 
 
 6C 
 
 S 4J 
 
 y^ 
 
 o 
 
 o 
 
 ^ 
 
 tn 
 
 c 
 
 1—1 
 
 M 
 
 CU 
 
 rH (U 
 
 
 01 
 
 N.-' 
 
 H 
 
 CO 
 
 
 CO )-l 
 
 c 
 
 ^-^ w 
 
 .—1 
 
 01 
 
 U3 
 
 <U 
 
 
 
 3 
 
 
 
 a 
 
 r-l X 
 
 CO 
 
 '-' 
 
 CU 
 
 
 
 
 3 
 
 61 
 
 -c 
 
 •rl 
 
 
 
 
 o 
 
 1 
 
 CO 
 
 u 
 
 •1-1 
 
 
 
 Q. 
 
 X 
 
 •rl CO 
 
 
 
 ex. 
 
 c 
 
 >. 
 
 e 
 
 
 
 C 
 
 ^ a 
 
 U 
 
 4-1 0. 
 
 
 •r^ 
 
 o 
 
 PM 
 
 0- 
 
 4-1 
 
 
 
 CU 
 
 w 
 
 « •rl 
 
 •rl 
 
 1-1 cu 
 
 •rl 
 
 f— 1 
 
 cu 
 
 to 
 
 •r-l 
 
 tj m 
 
 CU 
 
 CO 0) 
 
 
 u 
 
 u 
 
 
 
 
 
 
 Qi 
 
 
 r-l 
 
 E 
 
 1-^ ffi 
 
 r-l 
 
 r-4 
 
 S 
 
 •1-1 
 
 u 
 
 pi; 
 
 cu 
 
 c ai 
 
 
 0) 
 
 u 
 
 • 
 
 u 
 
 CU 
 
 • 
 
 
 
 • x> 
 
 
 
 cu 
 
 
 
 •H 
 
 
 Q 
 
 •rl 
 
 • 
 
 c 
 
 u 
 
 
 iJ 
 
 CO 
 
 J 
 
 cu 
 
 Ci 
 
 h4 
 
 C/3 
 
 CO 
 
 '-> 3 
 
 c 
 
 CQ CO 
 
 -a 
 
 ca 
 
 00 
 
 
 3 
 
 00 
 
 iH 
 
 fa CO 
 
 
 a 
 
 h4 
 
 
 D- 
 
 t/l 
 
 
 
 u. 
 
 Cu 
 
 
 
 fa 6 
 
 •rl 
 
 
 fa 
 
 fa 
 
 cr 
 
 Ea 
 
 w: 
 
 Bd 
 
 
 CT3 
 
 CO 
 
 • 
 CO 
 
 CO 
 
 PLI 
 
 c 
 I-l 
 
 • 
 
 
 
 i 
 
 • C 
 
 
 
 3 
 
 
 H 
 
 ^ 
 
 e 
 
 <U 
 Pi 
 
 Oi* g 
 
 c 
 
 fa 
 
 Pci* ^ 
 
 <=> 
 
 
 
 
 
 
 
 
 
 
 
 
 
 a: 
 
 00 
 
 
 o» 
 
 
 
 
 
 f-4 
 
 e»« 
 
 
 n 
 
 
 •* 
 
 « 
 
 e>4 
 
 
 CM 
 
 
 ro 
 
 
 fO 
 
 CO 
 
 
 ro 
 
 
 *n 
 
 » 
 
 O 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 -« 
 
 fu 
 
 
 fa 
 
 
 fa 
 
 
 fa 
 
 fa 
 
 
 fa 
 
 
 b 
 
 t_> 
 
 
 
 
 
 
 
 
 
 
 
 
 
 II. 6-5 
 
4- 
 
 X 
 
 > 
 
 Q 
 
 o 
 
 > 
 
 a 
 
 X 
 
 X 
 
 0) 
 
 o 
 
 o 
 
 
 <u 
 
 
 u 
 
 
 •i-i 
 
 
 14-1 
 
 
 IM 
 
 
 O 
 
 
 C 
 
 <u 
 
 
 
 1—1 
 
 •H 
 
 u 
 
 05 
 
 u 
 
 •i-t 
 
 (0 
 
 > 
 
 0) 
 
 ■r-l 
 
 w 
 
 Q 
 
 u 
 Id 
 
 o 
 o 
 
 <u 
 o 
 
 •H 
 
 o 
 
 o 
 
 > 
 •1-1 
 
 Q 
 
 o 
 
 <M 
 
 o 
 
 c 
 o 
 
 •H 
 00 
 
 > 
 
 •1-1 
 Q 
 
 Ql- 
 
 cc: 
 
 C3 
 
 
 CO 
 
 w 
 
 00 
 V) 
 
 w 
 o 
 
 I-I 
 H 
 O 
 
 Q 
 O 
 
 d 
 o 
 o 
 
 
 0) 
 
 CO 
 
 o 
 
 CO 
 
 
 
 
 tH 
 
 
 
 
 
 
 
 
 
 
 
 
 6C 
 
 
 
 
 
 
 
 
 
 
 4-1 
 
 
 
 <> 
 
 
 
 
 
 
 
 CD 
 
 
 
 
 
 
 
 
 CO 
 
 
 
 1-^ 
 
 
 
 
 
 
 
 I-I 
 
 
 
 
 
 
 
 
 g 
 
 
 
 o 
 
 
 
 
 
 
 
 O 
 
 
 
 
 
 
 
 
 >-( 
 
 
 
 o 
 
 
 
 
 
 
 
 4J 
 
 
 
 
 
 
 
 
 o 
 
 
 
 1 
 
 
 
 « 
 
 
 
 
 
 U) 
 
 
 
 
 
 
 
 
 IM 
 
 
 
 M 
 
 
 
 c 
 
 
 
 
 
 
 
 
 
 
 
 
 
 c 
 
 
 
 K 
 
 
 
 •r^ 
 
 
 
 
 
 X 
 
 
 
 
 
 
 
 
 H 
 
 
 
 O 
 
 
 
 r-4 
 
 
 
 
 
 CO 
 ■H 
 
 
 
 
 
 
 
 
 w 
 
 
 
 
 
 
 c 
 
 
 
 
 
 b 
 
 
 
 
 
 
 
 
 c 
 
 
 
 1-4 
 
 
 
 to 
 
 
 
 
 
 
 
 
 
 
 
 
 
 t4 
 
 
 
 f^ 
 
 
 
 w 
 
 
 
 
 
 < 
 
 
 
 
 
 
 
 
 U 
 
 
 
 CT^ 
 
 
 
 
 
 
 
 
 PM 
 
 
 
 I-I 
 
 
 
 
 
 
 o 
 
 
 
 t-t 
 
 
 
 4-1 
 
 
 
 
 
 H 
 
 
 
 to 
 
 
 
 
 p-l 
 
 
 
 <u 
 
 
 
 
 
 
 o 
 
 
 
 
 
 
 
 
 en 
 
 
 
 
 
 
 
 
 c 
 
 
 
 c 
 
 
 
 3 
 
 
 
 
 
 •(•4 
 
 
 
 O 
 
 •^ 
 
 
 
 
 
 •H 
 
 
 
 3 
 
 
 
 TJ 
 
 
 
 01 
 
 
 
 O 
 
 
 
 a 
 
 CO 
 
 
 
 
 
 
 U 
 
 
 
 erf 
 
 
 
 o 
 
 
 
 tu 
 
 
 
 
 
 
 o 
 
 Cxj 
 
 
 
 14J 
 
 
 
 C 
 
 
 
 
 
 
 u 
 
 
 
 •H 
 
 
 
 CO 
 
 
 
 ^4 
 
 g 
 
 
 
 o 
 
 
 
 U 
 
 
 
 CD 
 
 
 
 ^ 
 
 
 
 •a 
 
 3 
 
 
 
 (U 
 to 
 
 
 
 &< 
 
 N— ' 
 
 
 
 >■ 
 
 
 
 M 
 
 
 
 CO 
 
 /-N 
 
 
 > 
 
 
 
 4J 
 
 
 
 > 
 
 
 
 > 
 
 
 
 
 T) 
 
 
 
 O 
 
 
 
 K 
 
 ■ • 
 
 
 pq 
 
 
 
 w 
 
 
 
 J-l 
 
 
 
 -0 
 
 c 
 
 
 
 D 
 
 
 
 >4-l 
 
 
 
 
 o 
 
 
 
 
 
 
 
 
 to 
 
 
 4J 
 
 3 
 
 o 
 
 
 u 
 
 i-J 
 
 
 
 
 
 
 a) 
 
 C 
 
 
 r 
 
 
 
 tu 
 
 
 
 c 
 
 
 PU 
 
 4-1 
 
 •1-4 
 
 
 Oi 
 
 CO 
 
 o 
 
 
 to 
 
 4J 
 
 
 
 4-1 
 
 M 
 
 
 c 
 o 
 
 
 
 to 
 
 
 
 <: 
 
 
 •H 
 V4 
 
 CO 
 
 c 
 c 
 
 
 1-1 
 
 5 
 
 CJ 
 
 
 c 
 
 
 
 
 «« 
 
 
 •r4 
 
 
 
 V4 
 
 ^-\ 
 
 
 g 
 
 
 O 
 
 c 
 
 CO 
 
 
 o 
 
 o 
 
 Q 
 
 
 0) 
 
 /— \ 
 
 
 U-l 
 
 CO 
 
 
 4-1 
 
 
 
 o 
 
 CO 
 
 
 a) 
 
 
 tn 
 
 o 
 
 S 
 
 
 to 
 
 •1-1 
 
 
 4J 
 
 CO 
 
 
 o 
 
 M 
 
 
 CU 
 
 
 
 4J 
 
 1 
 
 
 4-1 
 
 
 3 
 
 T-l 
 
 
 
 to 
 
 3 
 
 > 
 
 Pa 
 
 l~) 
 
 c 
 
 b 
 g 
 
 
 
 0) 
 
 4-1 
 
 .-1 
 
 
 
 CO 
 
 
 CO 
 
 ^•N 
 
 c: 
 
 4-1 
 
 • 
 
 
 g 
 
 ^ c: 
 
 o 
 
 N--* 
 
 u 
 
 
 
 
 4.) 
 
 4-1 
 
 M 
 
 CU 
 
 /--s 
 
 
 
 
 >1 
 
 CO 
 
 to 
 
 to 
 
 •n 
 
 
 CU 
 
 CO to 
 
 Dd 
 
 
 o 
 
 ■u 
 
 
 u 
 
 CO 
 
 o 
 
 B 
 
 CO 
 
 
 t 
 
 s_x 
 
 
 CO 
 
 
 s 
 
 O 
 
 
 
 ' 1-1 
 
 b 2: 
 
 
 c 
 
 a. 
 
 
 \_^ 
 
 
 o 
 
 01 
 
 CL 
 
 c 
 
 u, 
 
 
 
 
 
 
 1-1 
 
 •o 
 
 
 XI 
 
 JS 
 
 u 
 
 <TI 
 
 OJ 
 
 TJ 
 
 
 IJ 
 
 a 
 
 > 
 
 dJ 
 
 u 
 
 1 
 
 4-1 
 
 o 
 
 •r4 
 
 AJ 
 
 >, 
 
 X) 
 
 U-l 
 
 c 
 
 4J 
 
 o 
 
 g T3 
 
 c 
 
 
 ai 
 
 <u 
 
 c 
 
 i-l 
 
 a 
 
 u 
 
 Di 
 
 
 u 
 
 SI 
 
 r-( 
 
 IJ 
 
 u 
 
 ^— ^ 
 
 (U 
 
 .--1 
 
 CO 
 
 U 
 
 u 
 
 v^ (U 
 
 •H 
 
 
 ■o 
 
 o 
 
 o 
 
 erf 
 
 CO 
 
 
 u-l 
 
 
 o 
 
 o 
 
 1—1 
 
 o 
 
 tt 
 
 
 x; 
 
 T3 
 
 
 
 
 0< 
 
 x: 
 
 )-l 
 
 o 
 
 LJ 
 
 c 
 
 •-4 
 
 o. 
 
 
 p^ 
 
 iJ 
 
 
 
 
 CU 
 
 c 
 
 <u 
 
 ex 
 
 c 
 
 4J 
 
 CO 
 
 O 
 
 4-1 
 
 0- 
 
 
 u to 
 
 o 
 
 -d 
 
 o 
 
 o 
 
 c 
 
 o 
 
 > 
 
 
 o 
 
 
 u 
 
 0) 
 
 < 
 
 c 
 
 (1) 
 
 •r4 
 
 CO 
 
 •H 
 
 ^ 
 
 CO 
 
 tu 
 
 w 
 
 to 1-1 
 
 <u 
 
 c 
 
 CO 
 
 g 
 
 c 
 
 cc: 
 
 u 
 
 c 
 
 CO 
 
 XJ 
 
 o 
 
 k: 
 
 
 •r4 
 
 ce: 
 
 C 
 
 C 
 
 1-1 
 
 
 C 
 
 erf 
 
 c 
 
 C: r-4 
 
 c 
 
 •H 
 
 I-I 
 
 CZ 
 
 jO 
 
 
 CO 
 
 CO 
 
 U 
 
 )-i 
 
 >, 
 
 
 x: 
 
 ti 
 
 
 •H 
 
 M 
 
 XI 
 
 1-4 
 
 M 
 
 
 -H 
 
 Vj XI 
 
 1-1 
 
 _] 
 
 u 
 
 O 
 
 ^ , 
 
 CO 
 
 
 OJ 
 
 i_i 
 
 O 
 
 Q 
 
 C/1 
 
 c« 
 
 CO 
 
 C/3 
 
 1-1 
 
 tj 
 
 3 
 
 f-l 
 
 w 
 
 CO 
 
 •-' 
 
 LJ 3 
 
 Ml 
 
 
 c 
 
 O 
 
 
 
 g 
 
 o 
 
 c 
 
 c 
 
 
 b 
 g 
 
 cy 
 
 
 b 
 2 
 
 OJ 
 
 
 CI. 
 
 1-1 
 
 
 b 
 
 1—1 
 
 a. 
 
 C 
 
 u 
 
 o 
 
 
 • 
 
 CO 
 
 o 
 
 o 
 u 
 
 o 
 
 Pi 
 
 • 
 
 
 •-) 
 
 
 erf 
 
 c 
 
 U 
 CO 
 
 • 
 
 erf 
 
 1 
 
 1-1 
 
 • c 
 
 in 
 
 vO 
 
 r» 
 
 t» 
 
 CJN 
 
 
 
 I-I 
 
 CM 
 
 ro 
 
 en 
 
 r> 
 
 n 
 
 ro 
 
 St 
 
 vl- 
 
 <t 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ftl 
 
 Cb 
 
 b 
 
 fa 
 
 fa 
 
 fa 
 
 fa 
 
 fa 
 
 II. 6-6 
 
a. 
 — I 
 
 00 
 
 '-J 
 
 o 
 o 
 
 7^ 
 
 ►— 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 <-> 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 •«« 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 o: 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 X 
 
 
 t/> 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 OQ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 •<£ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 SE 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 _j 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Ul. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CD 
 
 
 
 
 X 
 
 
 X 
 
 
 
 X 
 
 
 
 X 
 
 
 c_> 
 
 
 
 
 
 
 
 
 
 
 
 
 
 • 
 
 z 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 y 
 
 
 QJ 
 
 0> 
 
 <u 
 
 4) 
 
 
 
 
 
 
 a* 
 
 
 
 
 u 
 
 
 
 
 u 
 
 
 
 
 
 
 u 
 
 
 
 u 
 
 
 
 
 •H 
 
 
 f-j 
 
 •1-4 
 
 •w 
 
 •r-l 
 
 
 •r4 
 
 
 
 •I-l 
 
 
 
 
 U-l 
 
 
 <tl 
 
 U4 
 
 u-l 
 
 U-l 
 
 
 UJ 
 
 rii: 
 
 
 u-l 
 
 
 ^c 
 
 
 U-l 
 
 
 U-( 
 
 «H 
 
 Vi-I 
 
 >4-| 
 
 
 «M 
 
 Vj 
 
 
 u-l 
 
 
 C3 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 m 
 
 
 
 
 
 ►— 
 
 
 
 
 
 
 
 
 
 
 ELi 
 
 
 
 
 «« 
 
 
 C 
 
 
 c 
 
 C 
 
 C 
 
 C 
 
 
 c 
 
 
 
 C 
 
 
 C_> 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 0) 
 
 
 
 
 
 C3 
 
 
 •H 
 
 
 •r^ 
 
 •rl 
 
 -H 
 
 •t-l 
 
 
 •r4 
 
 60 
 
 
 ■H 
 
 
 
 
 w 
 
 
 05 
 
 CO 
 
 CO 
 
 CO 
 
 
 (0 
 
 V 
 
 
 (0 
 
 
 
 
 •r-< 
 
 
 •r-( 
 
 •rl 
 
 •H 
 
 •H 
 
 
 •H 
 
 i-l 
 
 
 •H 
 
 
 
 
 > 
 
 
 > 
 
 > 
 
 > 
 
 > 
 
 
 > 
 
 ■-I 
 
 
 > 
 
 
 
 
 -r-l 
 
 
 •M 
 
 T-l 
 
 •H 
 
 •r-l 
 
 
 •i-l 
 
 
 
 
 •r-l 
 
 
 
 
 Q 
 
 
 Q 
 
 Q 
 
 Q 
 
 Q 
 
 
 Q 
 
 
 
 
 Q 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 .U 
 
 
 
 u 
 
 c 
 
 (U 
 
 a> 
 
 UJ 
 
 c 
 
 
 2: 
 
 
 
 
 
 
 
 
 
 
 
 
 C 
 
 
 
 < — » 
 
 
 
 
 
 
 
 
 
 
 
 
 Q) 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 3 
 
 I-l 
 
 »— 
 
 
 
 
 
 
 
 
 
 
 
 
 •-I 
 
 CO 
 
 •ce 
 
 
 
 
 
 
 
 
 
 
 
 
 u-l 
 
 u 
 
 C3 
 
 
 C 
 
 
 
 
 
 
 
 
 
 
 
 u-l 
 
 •H 
 
 3 
 
 
 
 
 
 
 
 
 
 
 
 
 w 
 
 60 
 
 •-I 
 
 
 c 
 
 CO 
 0. 
 
 
 
 
 0. 
 
 § 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 u 
 
 
 
 
 
 
 
 
 
 
 
 u 
 
 u 
 
 
 
 iJ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 u 
 
 
 
 4-1 
 
 
 
 
 
 
 
 
 
 
 
 u-l 
 
 H 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CJ 
 
 
 
 
 
 
 
 
 
 
 
 V4 
 
 CO 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 
 
 
 0) 
 
 
 ^— ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 u 
 
 •r! 
 
 
 "3 
 
 t£ 
 
 
 
 
 
 
 
 
 
 
 
 n 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CN 
 
 
 
 Vj 
 
 a) 
 
 
 4-1 
 
 
 
 
 
 
 
 
 
 
 
 r~- 
 
 
 
 CD 
 
 x: 
 
 
 c 
 
 )-i 
 
 
 
 
 
 
 
 
 
 ON 
 
 
 
 a 
 
 to 
 
 
 
 
 iJ 
 
 
 
 
 
 
 
 
 
 r-l 
 
 
 
 CJ 
 
 •r-l 
 
 
 u 
 
 C/2 
 
 
 
 
 
 
 
 
 
 -1 
 
 
 
 in 
 
 Cu 
 
 
 
 x: 
 
 
 
 
 
 
 >-. 
 
 
 
 
 •r-l 
 
 
 
 c 
 
 y 
 
 . ^ 
 
 CM 
 
 4J 
 
 
 
 
 
 
 U 
 
 
 
 
 Vj 
 
 
 
 
 
 •^^ 
 
 I/O 
 
 r^ 
 
 .r-4 
 
 
 
 
 
 
 (U 
 
 
 
 
 a 
 
 
 
 •rJ 
 
 .-N jj 
 
 ^ 
 
 ON 
 
 "3- 
 
 
 
 
 
 
 > 
 
 
 
 
 
 
 < 
 
 
 
 i-> 
 CO 
 
 CO c 
 
 £3 CO 
 
 
 
 
 c 
 
 
 
 
 
 
 
 u 
 
 
 
 
 c 
 
 
 
 JJ 
 
 S ""^ 
 
 >— 
 
 
 
 
 
 
 
 0) 
 
 
 
 
 •r^ 
 
 
 
 
 .-J 
 
 Z J-) 
 
 ^ < 
 
 1Z3 
 
 
 1-) 
 
 
 
 
 
 
 
 
 
 
 CU 
 
 
 
 Cu 
 
 
 •«x 
 
 
 TO 
 
 r— I 
 
 
 
 > 
 
 D 
 
 
 
 
 u 
 
 c 
 
 lU 
 
 > 
 
 l-l 
 
 
 
 
 s 
 
 0) 
 
 x: 
 
 
 
 VJ 
 
 ■r^ 
 < 
 
 • .-1 
 Vj to 
 *^ 3 
 
 c 
 
 
 
 c 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 <n 
 
 
 i-l 
 
 
 
 -a 
 
 
 
 
 
 
 
 
 
 
 
 CO 
 
 
 
 <u 
 
 
 
 
 
 
 
 
 
 CJ 
 
 
 
 c 
 
 
 
 
 
 
 
 > 
 
 
 c 
 
 
 
 > 
 
 
 
 
 
 
 
 
 
 
 
 
 •0 
 
 
 iJ 
 
 ■r-l 
 
 ■U 
 
 •r-l 
 
 
 
 r-J 
 
 
 
 a 
 
 
 
 •r^ 
 
 
 
 
 <u 
 
 
 Cu 
 
 u 
 
 D. 
 
 
 
 
 
 
 
 
 c 
 
 ^^ 
 
 
 i-i 
 
 
 
 
 to 
 
 
 •r-l 
 
 r3 
 
 •r^ 
 
 1/1 
 
 
 
 (0 
 
 r^ 
 
 
 
 •r4 
 
 
 
 
 f3 
 
 
 
 
 
 
 
 Vj 
 
 C 
 
 Vj 
 
 c 
 
 ''^ 
 
 
 to 
 
 * 
 
 
 
 >J 
 
 u 
 
 
 N 
 
 
 
 R 
 
 
 & 
 
 
 CJ 
 
 w 
 
 CJ 
 
 J^ 
 
 ^"j 
 
 
 •r^ 
 
 sJ 4-» 
 
 
 
 *r-( 
 
 9 
 
 
 •a 
 
 
 
 
 
 
 
 
 
 10 
 
 
 
 to 
 
 
 
 fa 
 
 
 Q 
 
 f3 
 
 
 
 AJ 
 
 
 J-1 
 
 
 
 •r-t 
 
 
 u 
 
 
 a 
 
 u 
 
 D 
 
 ■0 
 
 /■-» 
 
 
 
 
 Ci -0 
 
 
 
 C 
 
 Oi 
 
 
 c 
 
 
 W 
 
 ^ 
 
 U 
 
 
 PU 
 
 /-^ 
 
 c 
 
 r-l 
 
 ^- C 
 
 .:^ 
 
 
 Cj 
 
 4-1 
 
 
 
 HI 
 
 N-^ 
 
 
 <U 
 
 ^->. 
 
 c 
 
 CO 
 
 
 
 ^-N 
 
 
 CO 
 
 2. 
 
 < 
 
 CO 
 
 TO 
 
 ^— /' 
 
 
 
 r^ 
 
 CJ 
 
 t 
 
 
 > 
 
 r-l 
 
 
 c 
 
 1.1 
 
 
 
 > 
 
 
 
 CO 
 
 1". 
 
 •H 
 N 
 •r-l 
 
 ta 
 6 u 
 
 0) 
 f— 1 
 01 
 
 CO 
 
 u-l 
 
 
 
 
 ■a 
 
 n 
 
 
 
 CJ 
 
 CJ 
 
 4-1 
 
 u-l 
 
 
 G 
 
 4_t 
 
 •• 4-1 
 
 a c 
 
 CJ 
 
 t 
 
 
 •J: 
 
 G 
 
 c 
 
 V) 
 
 CO 
 
 v./ U 
 
 CO 
 
 4-J 
 
 ^— ^ 
 
 
 
 ■r-l 
 
 •w^ CJ 
 
 
 JZ 
 
 Vj 
 
 c 
 
 
 
 JZ ■■T. 
 
 
 
 O 
 
 Vj 
 
 
 
 0) 
 
 -~^ • 
 
 
 
 
 
 v-- 
 
 c 
 
 
 •" 
 
 i-) 
 
 -C 
 
 4-t 
 
 cu 
 
 
 
 
 
 CJ 
 
 Oj 
 
 
 
 C 
 
 
 
 Qi 
 
 Q 
 
 c 
 
 1-1 
 
 ij n. 
 
 >-l 
 
 c- 
 
 CJ 
 
 u 
 
 </> 
 
 c 
 
 iJ 'J) 
 
 c 
 
 
 
 C 
 
 ■ ^ 
 
 Vj 
 
 U Vj 
 
 
 
 1 
 
 -0 
 
 
 
 Sj 
 
 
 
 U) o 
 
 o 
 
 1-1 OJ 
 
 t= 
 
 (0 
 
 ■~> 
 
 f-~* 
 
 0) —1 
 
 a 
 
 yC 
 
 CJ 
 
 4J 
 
 H 
 
 T. c- 
 
 
 
 tc 
 
 c 
 
 l-i 
 
 cr 
 
 a oi 
 
 J-J 
 
 C Dd 
 
 4_1 
 
 cc: 
 
 -X 
 
 C 
 
 f— 1 
 
 •-> 
 
 C ■-I 
 
 G 
 
 
 Qi 
 
 cc 
 
 
 
 
 
 c 
 
 — i 
 
 CJ 
 
 C 
 
 >> 
 
 c 
 
 >-i 
 
 c. 
 
 >. 
 
 '/I 
 
 1-1 
 
 .c 
 
 •r^ 
 
 Vj S: 
 
 c 
 
 • 
 
 
 D 
 
 Vj 
 
 • Vj 
 
 
 
 ■ri 
 
 J 
 
 i_t 
 
 •r-l 
 
 Q CO 
 
 nj 
 
 UJ CO 
 
 m 
 
 Q t/5 
 
 
 
 u 
 
 D 
 
 l-i 
 
 la 3 
 
 •^ 
 
 Q 
 
 CO 
 
 .— 1 
 
 O 
 
 a o 
 
 
 
 u 
 Q 
 
 u 
 
 4-1 
 
 >7 
 
 
 
 
 
 Q 
 
 . fa 
 
 u 
 
 
 Q 
 
 . fa 
 
 < 
 
 cti 
 
 C 
 
 c 
 :3 
 
 •r^ 
 Q 
 
 a. 
 • c 
 
 
 < 
 
 
 > 
 
 UJ 
 
 4-J 
 tn 
 
 • CO 
 < Oj 
 
 C3 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ■JC 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 1 
 
 
 r-l 
 
 
 <f 
 
 «r» 
 
 vD 
 
 r^ 
 
 
 00 
 
 Ov 
 
 
 
 
 
 1 
 
 
 <r 
 
 
 vj- 
 
 •* 
 
 <f 
 
 •J- 
 
 
 •* 
 
 <f 
 
 
 in 
 
 
 «« 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 (_> 
 
 
 Ij_ 
 
 
 Uh 
 
 ti. 
 
 \x* 
 
 k< 
 
 
 Uj 
 
 U, 
 
 
 Cu 
 
 
 II. 6-7 
 
re 
 ex. 
 
 «3: 
 cc 
 
 CD 
 
 cp 
 
 1 
 
 CQ 
 
 ca 
 a. 
 
 CO 
 
 to 
 
 w 
 o 
 o 
 
 z 
 
 o 
 
 l-l 
 H 
 
 u 
 
 Q 
 O 
 
 CO 
 
 -<< 
 
 CO 
 
 U 
 
 O 
 
 »< 
 
 
 01 
 
 
 o 
 
 
 •H 
 
 X 
 
 U-( 
 
 u 
 
 VM 
 
 ro 
 
 O 
 
 P4 
 
 
 
 c 
 
 <u 
 
 o 
 
 JJO 
 
 •H 
 
 a) 
 
 to 
 
 1-1 
 
 •r^ 
 
 .-1 
 
 > 
 
 o 
 
 •H 
 
 o 
 
 Q 
 
 0) 
 
 
 
 
 
 
 
 e 
 
 
 
 
 
 
 
 (0 
 
 
 
 
 
 
 
 <u 
 
 
 
 
 
 
 
 i-l 
 
 
 
 
 
 
 
 u 
 
 
 
 
 
 
 
 Vi 
 
 
 
 
 
 
 
 W) 
 
 
 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 (1) 
 
 
 
 
 
 
 
 o 
 
 
 
 
 
 
 
 o 
 
 
 
 
 
 
 
 ^4 
 
 
 
 
 
 
 
 (i< 
 
 
 
 
 
 
 
 w 
 
 
 
 
 
 
 
 c 
 
 
 
 
 
 
 
 T-l 
 
 
 
 
 
 
 
 XJ 
 
 
 
 
 
 
 
 « 
 
 
 
 
 
 
 
 (1) 
 
 
 
 
 
 
 
 u 
 
 
 
 
 
 
 
 H 
 
 
 
 
 rH 
 
 
 
 U-l 
 
 
 
 
 CO 
 
 
 
 o 
 
 
 
 
 U 
 •r4 
 
 
 
 (U 
 
 
 
 
 dC 
 
 
 
 to 
 
 
 
 
 o 
 
 
 
 /\ 
 
 
 
 
 t-l 
 
 
 
 a 
 
 
 
 
 o 
 
 
 
 u 
 
 
 
 
 •H 
 
 
 
 •^ 
 
 
 
 
 i^ 
 
 
 
 ru 
 
 
 
 
 4J 
 
 
 
 0) 
 
 
 
 
 O 
 
 
 
 £. 
 
 
 
 
 CO 
 
 
 
 u 
 
 
 
 
 CQ 
 
 
 
 U 
 
 
 
 
 
 
 
 O 
 
 
 
 
 CO 
 
 
 
 >4-< 
 
 
 
 
 
 
 
 U 
 
 
 
 
 13 
 
 
 
 M 
 
 
 
 
 3 
 
 
 
 « 
 
 
 
 
 4J 
 
 
 
 x: 
 
 
 
 
 W 
 
 
 
 o 
 
 
 
 
 0) 
 
 
 
 5 
 
 
 
 
 M 
 
 
 
 o 
 
 
 
 
 3 
 
 
 
 i-i 
 
 
 
 
 ■O 
 
 
 
 fe 
 
 
 
 
 tU 
 
 o 
 
 
 
 X) 
 
 
 
 
 o 
 
 
 
 c 
 
 
 
 
 u 
 
 
 
 to 
 
 
 /'N 
 
 
 CM 
 
 /^\ 
 
 
 *J 
 
 •U 
 
 • 
 
 
 i-< 
 
 V) 
 
 
 3 
 
 e 
 
 o 
 
 4J 
 
 ttj 
 
 1 
 
 
 o 
 
 CtJ 
 
 o 
 
 U 
 
 o 
 
 
 1 
 
 1— 1 
 
 
 o 
 
 •H 
 
 
 >- 
 
 CM 
 
 b 
 
 D. 
 
 U 
 
 ^^ 
 
 
 (U 
 
 
 3 
 
 <U 
 
 >, 
 
 
 4-1 
 
 ►4 
 
 C) 
 
 -a 
 
 ecj 
 
 i-l 
 
 >— 1 
 
 M 
 
 
 Ph 
 
 c 
 
 
 tU 
 
 .-1 
 
 O 
 
 T3 
 
 lu 
 
 D 
 
 u 
 
 c 
 
 0) 
 
 D- 
 
 OJ 
 
 
 u 
 
 o 
 
 < 
 
 J^ 
 
 O 
 
 to 
 
 (U 
 
 o 
 
 ca 
 
 
 to 
 
 Oi 
 
 
 
 SI 
 
 JD 
 
 vi 
 
 T3 
 
 to 
 
 
 c 
 
 iJ 
 
 M 
 
 vi 
 
 •H 
 
 CM 
 
 Ol 
 
 o 
 
 
 to 
 
 c 
 
 a 
 
 
 b 
 ^ 
 
 u 
 
 c 
 
 o 
 
 o 
 
 to 
 
 • 
 
 Ph 
 
 T-l 
 
 ^-^ 
 
 u 
 
 cd 
 
 to 
 
 
 o 
 
 
 o 
 
 
 •r-l 
 
 ^ 
 
 li^ 
 
 n 
 
 IM 
 
 to 
 
 O 
 
 Ph 
 
 
 
 c 
 
 tU 
 
 o 
 
 bO 
 
 1-1 
 
 tU 
 
 to 
 
 I-l 
 
 •r^ 
 
 I-l 
 
 > 
 
 O 
 
 T-l 
 
 O 
 
 Q 
 
 CO 
 
 tH 
 
 r^ 
 
 !-J 
 
 o 
 
 to 
 
 1-1 
 
 > 
 
 o 
 
 iH 
 
 4J 
 
 T-4 
 
 to 
 
 O 
 
 a 
 E 
 o 
 
 o 
 
 a, 
 o 
 ci 
 
 to 
 
 tu 
 o 
 •1-1 
 tw 
 
 •4-1 
 
 o 
 
 c 
 o 
 ■1-1 
 
 to 
 
 •H 
 
 > 
 
 Q 
 
 X 
 
 X 
 
 o 
 
 •r^ 
 
 U-( 
 U-l 
 
 O 
 
 C 
 
 o 
 
 •H 
 to 
 
 •1-1 
 > 
 
 •H 
 
 Q 
 
 tU 
 U 
 ■H 
 CM 
 VM 
 O 
 
 c 
 o 
 
 •r* 
 
 cn 
 
 •rt 
 > 
 
 T-l 
 
 Q 
 
 U 
 
 •f-l 
 
 <4-l 
 IM 
 O 
 
 c 
 o 
 
 T-l 
 
 to 
 •1-1 
 > 
 
 T-l 
 
 Q 
 
 u 
 
 o 
 
 D- 
 1) 
 
 CO 
 
 C CO 
 
 tu 
 a 
 o 
 
 I-l 
 
 CJ u 
 
 ■U U 
 
 U) O 
 
 CU Cu 
 
 > OJ 
 
 M a: 
 to 
 
 ~ -u 
 o 
 
 C to 
 
 CO U 
 
 u c 
 
 o o 
 
 »-l 
 
 0. *-) 
 
 4J W 
 
 CO O 
 
 u oi 
 to 
 
 m 
 
 O 
 
 O 
 
 I 
 
 X 
 
 <u 
 u 
 •rl 
 
 U-l 
 U-) 
 
 O 
 
 c 
 o 
 
 T-l 
 
 CO 
 
 T-l 
 
 > 
 
 T-l 
 
 Q 
 
 to 
 >-i 
 
 to O 
 
 CJ ex. 
 
 > <y 
 
 u cd 
 
 to 
 
 H 
 
 •a 
 
 3 
 
 5 
 
 tu 
 
 T-l 
 > 
 
 &=: 
 
 tc 
 c 
 T^ 
 
 M 
 
 ■QJ 
 
 tu 
 
 c 
 
 c 
 (id u 
 
 O 
 
 - CU 
 
 Vi tu 
 
 o ed 
 
 Q CO 
 
 o 
 
 m 
 o 
 
 to 
 
 o 
 
 m 
 
 o 
 
 u-i 
 
 to 
 o 
 
 m 
 o 
 
 tM 
 
 o 
 
 CO 
 
 o 
 
 II. 6-8 
 
w 
 >- w 
 
 -I- M 
 
 < 8 
 
 CD 0-. 
 
 o 
 d 
 
 M 
 H 
 O 
 
 Q 
 O 
 
 CO 
 
 a. 
 
 ► - 
 
 
 
 ■ 
 
 
 CJ 
 
 
 
 
 
 «* 
 
 
 
 
 
 ex: 
 
 
 
 
 
 1 — 
 
 
 X 
 
 X 
 
 
 C/O 
 
 
 
 
 
 CO 
 
 
 
 
 
 <x 
 
 
 
 
 
 
 
 
 
 
 2E 
 
 
 
 
 
 1 
 
 
 
 
 
 ka- 
 
 
 
 
 
 
 
 ><; 
 
 
 
 CJ 
 
 
 
 
 
 ^. 
 
 
 
 
 
 •^ 
 
 
 
 
 
 
 
 o 
 o 
 
 o 
 
 •H 
 U-l 
 
 
 z: 
 
 
 U-l 
 
 U-i 
 
 
 CD 
 
 
 O 
 
 O 
 
 
 
 
 
 
 
 ►— 
 
 
 c 
 
 c 
 
 
 «« 
 
 
 o 
 
 o 
 
 
 C_3 
 
 
 •H 
 
 •H 
 
 
 O 
 
 
 </) 
 
 to 
 
 
 
 
 •■-1 
 > 
 
 Q 
 
 •H 
 > 
 
 •H 
 
 Q 
 
 
 
 
 
 T) 
 
 
 
 
 
 
 c 
 
 
 
 
 
 
 « 
 
 
 
 2= 
 
 
 
 rH 
 
 
 
 C3 
 
 
 
 CO 
 
 
 
 ^« 
 
 
 
 4J 
 
 
 
 ►— 
 
 
 
 c: 
 
 
 
 -a: 
 
 
 
 <u 
 
 
 
 CJ 
 
 
 
 e 
 
 
 
 Hj 
 
 
 
 •H 
 
 
 
 CO 
 
 
 
 )-l 
 
 
 
 13 
 
 
 
 0) 
 
 
 
 a. 
 
 X) 
 4-1 
 
 c 
 o 
 u 
 
 
 P- 
 X 
 
 w 
 
 0) 
 4-1 
 
 o 
 
 ■4-1 
 
 c 
 
 CO 
 
 x: 
 
 c 
 
 ■ 
 
 
 - — .. 
 
 ro 
 
 
 <u 
 
 
 
 t/O 
 
 0^ 
 
 
 s 
 
 
 
 3C 
 
 1— ( 
 
 
 
 
 
 «< 
 
 
 
 o 
 )-l 
 
 
 
 
 
 
 
 
 
 
 
 
 vO 
 
 
 u-l 
 
 
 
 
 
 o 
 
 
 o 
 
 ^-^ 
 
 
 
 
 o 
 
 1 
 
 
 0£ 
 
 CO 
 
 
 
 
 M 
 
 
 c 
 
 >^ 
 
 
 
 
 X 
 
 >-% 
 
 •H 
 
 ^ 
 
 
 
 
 o 
 
 • 
 
 c 
 
 ^*-^ 
 
 
 
 
 v-^ 
 
 H 
 
 o 
 
 43 
 
 c 
 
 
 
 
 u 
 
 M 
 
 0) 
 
 <0 
 
 
 
 
 ^< 
 
 
 o 
 
 tJ 
 
 e 
 
 
 
 
 o 
 
 •» 
 
 
 c 
 
 j= 
 
 
 
 
 D 
 
 tn 
 
 T3 
 
 (0 
 
 0) 
 
 
 
 
 U 
 
 ^ 
 
 C 
 
 .-1 
 
 hJ 
 
 
 
 
 Qi 
 
 4) 4J 
 
 <0 
 
 C14 
 
 
 
 UJ 
 
 
 
 •U )-l 
 
 
 
 , 
 
 
 — ij 
 
 
 iJ 
 
 tf) o 
 
 oc 
 
 c 
 
 J 
 
 
 ►^ 
 
 
 a 
 
 O O. 
 
 c 
 
 
 
 
 
 H^ 
 
 
 01 
 
 > o 
 
 •rH 
 
 •H 
 
 •a AJ 
 
 
 
 
 
 M cd 
 
 3 
 
 4-1 
 
 c u 
 
 
 
 
 o 
 
 (B 
 
 o 
 
 <0 
 
 <o 
 
 
 
 
 ^ 
 
 S •u 
 
 ^ 
 
 )-l 
 
 D. 
 
 
 
 
 C nj 
 
 H 
 
 4-> 
 
 U OJ 
 
 
 
 
 i—t 
 
 (13 Vj 
 
 TD 
 
 c 
 
 >. 
 
 
 
 
 (fl 
 
 0; u 
 
 •H 
 
 
 
 Q to 
 
 
 
 
 c 
 
 O C 
 
 D 
 
 e 
 
 •fe 
 
 
 
 
 ^r4 
 
 O O 
 
 to 
 
 (U 
 
 
 
 
 U^ 
 
 V-' u 
 
 OilQ 
 
 -) Z 
 
 
 C3 
 
 
 
 
 
 z: 
 
 
 a\ 
 
 O 
 
 
 1 
 
 
 u-i 
 
 vO 
 
 
 * 
 
 
 o 
 
 o 
 
 
 -< 
 
 
 u. 
 
 u. 
 
 
 c_> 
 
 
 
 
 
 II. 6-9 
 
v\ f*>J 
 
 T 
 
 LECTE© AdSTKACT 
 
 ■vi *] r Ti 
 
Secti©&i A 
 
 GEr^E^AL 
 
 III. 1-1 
 
A0I3 
 
 Marine Protein Concentrate 
 Fishery Leaflet 584 
 
 BCF, Fish and Wildlife Service, U.S. Department of Interior 
 
 ABSTRACT 
 
 General information was given on a method to convert red 
 hake into a high-quality marine protein concentrate for human 
 consumption. Plans for and estimates of processing and product 
 costs were shown for a proposed commercial process. Data were 
 presented on the physical, chemical, nutritional, bacteriological, 
 and toxicological studies on marine protein concentrate, produced 
 by a method closely approximating the proposed commercial process. 
 
 III. 1-2 
 
A014 
 
 Proceedings, Conference on Fish Protein Concentrate 
 Federal-Provincial Atlantic Fisheries Committee, Canada 
 
 ABSTRACT 
 
 Session 1: The World Protein Situation 
 
 World Fisheries and the Protein Gap - R. Hamlisch, R. Kreuzer 
 The Constitution and Objectives of the FAO/WHO/UNICEF Protein Advisory 
 Group (PAG) - Z. I. Kertesz. The Role of the Canadian Aid Program .... - 
 L, D. Hudon. 
 
 Session 2: Canadian Raw Material Resources 
 
 The Potential Use of Canadian Raw Material Resources for the 
 Manufacture of FPC - Brian Meagher. Fish Resources and their Potential 
 Utilization for FPC - W. R. Martin. Maximum Utilization of Resources: 
 the Fishing Industries View - E. L. Harrison. Fillet Trimmings as a 
 Potential Raw Material for FPC - A. Cunningham. 
 
 Session 3: Processing Technology 
 
 The Development and Scope of the Halifax Process - D. R. Idler 
 Nutritional Characteristics and Projections on Production Costs of 
 Products, prepared by the Halifax Process - H. E. Power. FPC Processing 
 Methods - E. R. Pariser. Research Progress on FPC - D. G. Snyder. 
 
 Session 4: Standards of FPC 
 
 Wholesomeness of FPC - Leo Friedman. Safety of FPC - A, B. Morrison, 
 The Purpose and Need for Quality Standards for FPC - H. V. Dempsey. 
 Canadian Food and Drug Regulations as They May Relate to FPC - D. G. Chapman 
 Poisonous Fishes and Marine Biotoxins - R. F. Nigrelli. 
 
 Session 5: Development of the Industry 
 
 The Development of a Canadian FPC Industry; Possible Approaches - 
 
 B. Weinberg. An Economic Assessment of a Commercial FPC Operation - 
 
 C. D. Holder, K. H. Kidd , W. B. Jagyar, D. S. Ross. Some Engineering 
 Aspects of Commercial FPC Production - A Carsten. 
 
 III. 1-3 
 
Session 6: Markets for FPC 
 
 Potential Utilization by the Cereal Products Industry - J. Holme, 
 Potential Markets for FPC in the Pharmaceutical and Diet Food Industry 
 D. L. Simmons. Fish - the Underdeveloped Resource for Food, Feed, and 
 Chemicals - E. Levin. The Development of the FPC Industry in North 
 America - W. M. Chapman. 
 
 Addenda 
 
 FPC in Foods - V. D. Sidwell. Feeding Trials on Undernourished 
 Children. . . - N. L. Lahiry, M. N. Moorjani, M. Swaminathan. 
 
 III. 1-4 
 
A017 
 
 Fish Protein Concentrate Program 
 National Marine Fisheries Service 
 
 ABSTRACT 
 
 This paper presents one approach for using the potential 
 of the sea to provide the animal protein that man so greatly needs, 
 It emphasizes an accelerated program for development of a protein 
 supplement — fish protein concentrate — employing the techni- 
 cal and international capabilities of the Agency for International 
 Development (AID) and the Department of the Interior, Bureau of 
 Commercial Fisheries . Present and future f ood-f rom-the-sea 
 goals of AID and BCF are discussed in terms of such considerations 
 as interactions with industry; costs, personnel, and facilities; 
 alternatives and constraints. 
 
 III. 1-5 
 
An23 
 
 The Fish Protein Concentrate Story, Part 5: Bringing the Product Home 
 
 Bureau of Commercial F.isheries Program 
 
 D. G. Snyder 
 
 ABSTRACT 
 
 Research on fish protein concentrate (FPC) by the Bureau 
 of Commercial Fisheries of the Department of the Interior began 
 in 1961. The general aim of the program v/as to study the manu- 
 facture and use of FPC as a major deterrent against global pro'- 
 tein malnutrition and as a potential economic stimulus to the 
 American fishing industry. The program as initially conceived 
 encompassed several research areas. At the urging of the Secre- 
 tary of the Interior, the program was later accelerated to effect, 
 as rapidly as possible, participation between government and in- 
 dustry in getting FPC to the people who need it, in a form they 
 want and can use, and at a price they can afford. 
 
 III. 1-6 
 
A033 
 
 Fish Protein for Human Foods 
 R. Finch 
 
 ABSTRACT 
 
 This article reviewed the growth of world fishery supplies, 
 their present contribution to human protein needs, the potential 
 for increasing their harvest and the efficiency of their utiliza- 
 tion in the human diet, and the factors which may limit their 
 further development. The major part of the review was devoted 
 to fish protein concentrate (FPC) since the production, safety, 
 nutrition, and use of various forms of this product, and their 
 potential for increasing the availability of fish protein, had 
 been the subject of a number of scientific publications in re- 
 cent years . 
 
 III. 1-7 
 
A034 
 
 Fish Protein Concentrate - Its Status Today 
 V. Sidwell; B. Stillings; G. Knobl, Jr. (NMFS) 
 
 ABSTRACT 
 
 The FPCs produced were, in general, bland tasting and 
 varied in color from vrhite to dark tan. They contained between 
 75 and 95% high-quality protein and they exhibited limited func- 
 tional properties according to standards set by industry for high- 
 protein foodstuffs. The characteristic of limited functional 
 properties, far from being a drawback, was in many circumstances 
 advantageous since FPC could be added to existing food products 
 to improve markedly the nutritional quality without significantly 
 altering other characteristics. Also, in some instances, the 
 addition of FPC appears to improve the shelf life of the final 
 baked products. 
 
 All solvent extracted FPCs or even all isopropyl alcohol 
 extracted FPCs were not completely alike. Significant differ- 
 ences in odor, lipid content, stability, taste, nutritional value, 
 and functional properties were obtained depending upon the process- 
 ing conditions and the species of fish used. 
 
 Experimental work was conducted to produce FPC with various 
 solvents, and with enzymes, microorganisms, or combinations of 
 enzymes and solvents. The functional properties of p-^oducts 
 resulting from these processes were improved. Some of these prod- 
 ucts appeared to be particularly promising for use in certain 
 food because of their functional attributes. 
 
 Although many problems remained to be solved and addition- 
 al research was required to show how FPC can be utilized more 
 efficiently, an FPC industry had been born and was growing. 
 
 III. 1-8 
 
A035 
 
 Fish Protein Concentrate - The State of the Art 
 
 G. M. Knobl, Jr. (NMFS) 
 
 ABSTRACT 
 
 This report covered the background of the efforts to develop 
 fish protein concentrate; the work being done to establish FPC 
 production plants was briefly outlined. Most of these plants were 
 producing FPC by solvent extraction procedures (usually isopropyl 
 alcohol) . 
 
 The FPCs were, in general, bland tasting and varied in color 
 from white to dark tan. They contained between 75 and 95% high- 
 quality protein and exhibited limited functional properties accord- 
 ing to standards set by industry for high protein foodstuffs. The 
 report outlined the experimental work to produce FPC with various 
 solvents and with enzymes, microorganisms, or combinations of en- 
 zymes and solvents. The products resulting from these processes 
 showed improved functional properties. Some of these appeared 
 promising for use in certain foods. 
 
 Although many problems remained to be solved and additional 
 research to show how FPC could be utilized more efficiently was 
 indicated, the conclusion was that an FPC industry had been es- 
 tablished. 
 
 III. 1-9 
 
A037 
 
 Fish Protein Concentrate 
 (FPC) Program Review 
 
 Northrop Services, Inc. 
 
 ABSTRACT 
 
 This was the first part of a two-part document on FPC 
 by the contractor. The purpose of this part of the report was 
 to provide for the Congress and the American public a compre- 
 hensive and objective review of the first eleven years of the NMFS 
 program to develop fish protein concentrate. 
 
 The report covered the FPC Program from its undertaking 
 in 1961 by the Bureau of Commercial Fisheries, defined the 
 magnitude of the problem, and provided background to the 
 existing program. Important milestones of the program were 
 detailed, giving information regarding budgetary history, pro- 
 gram management, program interfaces, and activities of related 
 governmental agencies. Laboratory and contract research on 
 FPC was covered, and the NMFS pilot experiment and demonstra- 
 tion plant, built to prove the feasibility of commercial manu- 
 facture, was described. 
 
 Appendices to the review included a bibliography of 
 references on the product. 
 
 HI. 1-10 
 
A039 
 
 FPC Program Plan 
 
 Northrop Services, Inc. 
 
 ABSTRACT 
 
 This dociiraent constituted Part II of the Northrop 
 Services, Inc., contract report. The purpose of this part 
 of the report was to provide for NMFS : 
 
 (1) a logical, effective plan of work to fulfill 
 the objectives of an FPC program, and 
 
 (2) a management plan to enable efficient and effective 
 implementation of the plan of work. 
 
 The management plan developed by the contractor delineated 
 the roles of the FPC Program Manager and other key personnel and 
 their interrelationships. Also covered were the interfaces among 
 organizations outside of NMFS. Project planning and 
 implementation methods were developed, ^including schedules <and 
 techniques for status monitoring and reporting and program 
 evaluation. 
 
 The report defined an integrated program plan and summarized 
 by project area the necessary work to respond to the conclusions 
 and recommendations of the FPC Program Review prepared by th6 
 contractor. 
 
 III. 1-11 
 
A04l\ 
 
 The Preparation and Properties of Fish Protein Isolates 
 J. Rasekh, B, R, Stillings, and A. Metz 
 ABSTRACT 
 
 The paper described the process for preparing a fish protein 
 isolate with desirable functional properties. The fish used through- 
 out these studies was either frozen whole or frozen headed and gutted 
 red hake ( Urophycis chus s) . The resulting product was a light-colored 
 functional protein with only a slight fishy flavor. The viscosity of 
 the foam was still quite stable after standing at room temperature for 
 
 120 minutes. It is most dispersible in distilled water at pH and 
 
 3 
 pH . The proximate composition of the isolate ranged from 91.6% to 
 
 93.8% protein; 4.4% to 5.2% moisture; 8% to 2.77c, ash, and .4% to .5% 
 
 fat. The amino acid composition did not deviate markedly from that of 
 
 raw fish. The nutritional quality of the fish isolate was 62% of casein, 
 
 This product could be used in a high protein beverage or as a binder in 
 
 meat emulsion products such as bolognas, frankfurters, and luncheon 
 
 meats . 
 
 III. 1-12 
 
A042 
 
 FPC Alternate Plan Development 
 
 Northrop Services, Inc. 
 
 ABSTRACT 
 
 At the direction of the FPC Program Manager, all efforts 
 to develop an alternate plan for the conclusion of the NMFS 
 FPC Program were terminated on January 2, 1973. The purpose 
 of this contract report was to provide a summary of the follow- 
 ing: 
 
 (1) events and efforts leading up to the development 
 of the alternate plan 
 
 (2) the work which had been planned and the proposed 
 budget 
 
 (3) reasons for suspension of the effort 
 
 The latter pare of the report briefly detailed what the 
 contractor felt to be work necessary to conclude the program 
 effectively under the established budget and time constraints. 
 
 III. 1-13 
 
ecti® 
 
 III. 2-1 
 
BOOl 
 
 Effect of Several Processing Variables on Protein Content 
 and Quality of Fish Flour 
 
 Co H. Kurtzman, D. G. Snyder, L. E. Ousterhout, and T. F. Brancher 
 
 ABSTRACT 
 
 In determining the quantity of protein, a precise chemical 
 assay was available, but in determining quality, the biological 
 assay was needed. Results of a biological assay, however, were 
 dependent, in part, on the experimental animal used; two common 
 ones were, of course, the rat and the chick. It was desirable to 
 have x^esults with both for comparative purposes. 
 
 The objective of the research reported in this communica- 
 tion was to determine by use of chemical and biological (rat and 
 chick) assays, whether the variables of method of processing or 
 of the fish utilized in processing affected the content or the 
 quality of the protein of fish flour. 
 
 III. 2-2 
 
B002 
 
 Differences in Quality and Quantity of Protein of Various Fish Flours 
 in Comparisons with Other Protein Supplements 
 
 C. H. Kurtzman, D. G. Snyder, L. E. Ousterhout, F. T, Piskur and T. F. Brancher 
 
 ABSTRACT 
 
 Chemical analyses and rat and chick assays were conducted 
 to determine (1) whether there is variability in content and 
 nutritive value of the protein of fish flours currently or po- 
 tentially available on the world market and (2) how the quality 
 of these fish flours compares with that of protein supplements 
 more commonly utilized. 
 
 III. 2-3 
 
B004 
 
 Possible Interference of Fats, Carbohydrates and 
 .Salts in Amino Acid DftteTininations in. Fish Meals, 
 Fish Protein Concentrates and Mixed Animal Feeds 
 
 P. Smith, Jr., M. E. Ambrose, G. M. Knobl, Jr. 
 
 ABSTRACT 
 
 In the preparation of protein hydrolyzates for amino acid 
 determination, the presence of nonprotein component's may result 
 in the destruction of some amino acids. This study was to 
 determine if destruction would be caused by the fats, carbohydrates, 
 and salts present in fish meals, fish protein concentrates, or 
 mixed animal feeds. In generarl, lipids, salts, and most levels of 
 carbohydrate had no effect. High levels of carbohydrate, however, 
 interfered in the case of some amino acids, particularly proline 
 and tyrosine. 
 
 III. 2-4 
 
B006 
 
 Increasing the Protein Content of Fishery 
 Materials by Fermentation 
 
 P. R. Burkholder, 0. A. Roels, and A. Chu 
 
 ABSTRACT 
 
 A study was undertaken of the fermentation, with nitrogen- 
 fixing microorganisms, of certain fatty fish and of certain 
 other marine organisms that are not usually included in the 
 human diet :(1) to increase the protein content and the biologi- 
 cal value of these products and (2) to enhance their palata- 
 bility. 
 
 After testing a number of mesophilic and thermophilic 
 fungi and bacteria, some species were found to be promising. 
 When homogenized fatty fish was fermented aerobically for brief 
 periods of time with microorganisms selected from our screening 
 program, reductions of over 30% in the fat content and concomi- 
 tant increases of 10% in protein content were achieved. The 
 ,use of some of these organisms masked the fish odor of the fer- 
 mented products and induced flavors into the product that are 
 characteristic of certain foods well accepted in Western society. 
 
 III. 2-5 
 
B007 
 
 Effects of Isopropyl Alcohol Extraction Procedures 
 on Biochemical Characteristics of Fish 
 
 0. A. Hammerle and E. R. Pariser 
 
 ABSTRACT 
 
 Solvent-extraction techniques applied to freshly ground 
 raw whole fish effectively reduced the water and lipid content 
 of the raw material to levels commensurate with improved stabil- 
 ity and keeping quality without adversely affecting the nutri- 
 tive quality. 
 
 Specifically, the use of isopropyl alcohol as the extrac- 
 tant had been shown, with both lean and fatty fish, to result 
 in powdered products having a protein content of about 80%, a 
 residual lipid content of 0.5% or less, and a total volatiles 
 content (moisture and isopropyl alcohol) of about 5%. In gener- 
 al, the protein quality of the extracted products reflected 
 only a very small, if any, reduction in the protein efficiency 
 ratio for these products as compared with the raw material; yet 
 the processing procedure reduced taste and odor components of 
 the fish to amounts less than those discernible by the senses. 
 
 III. 2-6 
 
BOOS 
 
 Nutritional Evaluation of Fish Protein Concentrate 
 
 B. R. Stillings 
 
 ABSTRACT 
 
 Several years ago, the U. S. Bureau of Commercial Fish- 
 eries embarked on a program to develop an inexpensive, stable, 
 wholesome protein concentrate of high nutritive quality from 
 fresh fish. The program has resulted in the development of a 
 fish protein concentrate that is being comprehensively evaluated. 
 This report discussed its nutritional evaluation. In addition, 
 the nutritional value of fish protein concentrate in the treat- 
 ment of protein malnutrition and the problems involved in eval- 
 uating its effectiveness were covered. 
 
 III. 2-7 
 
B009 
 
 The Solubilization of Fish Protein Concentrate 
 M. G. Ahearn 
 
 ABSTRACT 
 
 A problem associated with utilization of fish protein 
 concentrate (FPC) in some food products is its relative insolu- 
 bility. Two procedures for solubilizing FPC were therefore 
 investigated: (a) treatment with sodium hydroxide in the region 
 of pJS 12 at 95°C. and (b) hydrolysis with proteolytic enzymes. 
 
 Optimum time for alkali solubilization was established as 
 being not over 20 minutes; at this time, most of FPC protein was 
 in solution. Denaturation, as indicated by the reduced viscosity 
 of the protein solution, occurred during the first 10 minutes, 
 followed by degradation of the protein molecules; little hydrol- 
 ysis occurred throughout the solubilization, as indicated by 
 the ratio of amino to total nitrogen. Racemization of all amino 
 acids occurred: average conversion to D-isomers for all amino 
 acids was approximately 11% after 10 minutes and approximately 
 14% after 20 minutes, as indicated by optical rotation studies, 
 Racemization of methionine was studied by microbiological assay 
 of the laevo form, in conjunction with determination of total 
 methionine: conversion to D-methionine v/as approximately 20% after 10 
 minutes and approximately 25% after 20 minutes. The jotal methio- 
 nine and total tryptophan content of FPC protein remained constant 
 throughout the solubilization. Evolution of ammonia was measured 
 for a similar process at 100°C; after 20 minutes of the process 
 approximately 3% of FPC nitrogen had been evolved. 
 
 Collagen, as estimated by the enzyme collagenase, appeared 
 to constitute 23% of FPC protein. 
 
 Enzymatic solubilization of FPC by Pronase, pepsin, plant 
 and fungal proteases was investigated. The plant and fungal 
 enzymes were required in rather high concentrations to substan- 
 tially solubilize FPC and did not achieve complete solubilization. 
 In the case of Pronase and pepsin, 95% solubilization of FPC 
 nitrogen could be achieved; these enzymes were effective at rela- 
 tively low concentrations . The degree of hydrolysis achieved 
 by Pronase was relatively high; that achieved by pepsin was re- 
 latively low. The Pronase hydrolyzate was meaty in flavor, while 
 the pepsin hydrolyzate was bitter. The Pronase hydrolyzate con- 
 tained a significant quantity of sodium chloride resulting from 
 neutralization; the pepsin hydrolyzate contained a much greater 
 quantity of this substance. 
 
 III. 2-8 
 
BCf2l 
 
 The Determination of Fluoride in Fish Protein 
 Concentrate by a Direct Non-Isolation Fluoride Electrode Method 
 
 R. B. Klemm, E. G. Zook, and G. M. Knobl 
 
 ABSTRACT 
 
 A rapid method for the determination of fluoride in fish 
 protein concentrate (FPC) has been developed. The fluoride in 
 FPC was solubilized by digesting samples in an aqueous buffer 
 solution containing EDTA. Following this simple treatment, and 
 xd-thout attempting further isolation of the fluoride, the con- 
 centration of fluoride in the sample solution was determined 
 directly by the method of known additions, utilizing a fluoride 
 ion-selective electrode and a "specific ion" meter. The function 
 of the EDTA was to chelate with the calcium in the bone portion 
 af the sample, thereby releasing the bound fluoride for detec- 
 tion by the electrode. A solid residue was recovered from the 
 digestion and then reanalyzed. This residue was found to be 
 essentially free of fluoride. The conditions of the digestion 
 procedure were optimized by evaluating the results of a study of 
 the effects of the amount of EDTA used, the time and temperature 
 of the solubilization step and the particle size distribution 
 in the FPC samples. The method yielded reproducible and quanti- 
 tative recoveries (about 100 percent) of standard sodium fluoride 
 added to FPC samples. The results from this new method were in 
 close agreement with those obtained using the AOAC procedure. 
 In addition, our new method eliminated ashing and the dangerous 
 perchloric acid distillation called for by the AOAC. 
 
 III. 2-9 
 
B022 
 
 Determination of Isopropyl Alcohol in Solid Fish Protein 
 Concentrate by Gas-Liquid Chromatography 
 
 P. Smith, Jr. and N, L. Brown 
 
 ABSTRACT 
 
 A method for the determination of isopropyl alcohol in 
 fish protein concentrate by gas-liquid chromatography 
 consists of heating the FPC to 180°C in a closed tube for 20 
 minutes to release the isopropyl alcohol and then injecting the 
 alcohol directly into the system. The method is also applicable 
 to the determination of similar volatile components in other 
 solid materials. 
 
 III. 2-10 
 
B023 
 
 Sequence of Limiting Amino Acids in Fish Protein 
 Concentrate Produced by Isopropyl Alcohol 
 Extraction of Red Hake ( Urophycis chuss ) 
 
 B. R. Stillings, 0. A. Hammerle, and D. G. Snyder 
 
 ABSTRACT 
 
 Experiments were conducted to determine the sequence of 
 limitation of the essential amino acids in fish protein con- 
 centrate produced by isopropyl alcohol extraction of whole 
 red hake ( Urophycis chuss) . Diets were prepared containing 
 1.28% nitrogen from FPC and 0.32% nitrogen from various combina- 
 tions of amino acids. The diets were fed to weanling rats for 
 four weeks; weight gain, feed intake and protein efficiency 
 ratio were determined. The studies indicated that, as a non- 
 essential nitrogen source, glutamic acid was utilized as effi- 
 ciently as was a mixture of nonessential amino acids. From the 
 results of these studies, the amino acids in FPC were grouped 
 according to their limitation, from greatest to least, as follows: 
 1) methionine; 2) histidine, tryptophan and threonine; 3) valine, 
 isoleucine and phenylalanine; and 4) leucine, lysine and arginine, 
 
 III.2-U 
 
B03I 
 
 The Chemistry of Flavor in Fish Protein Concentrates 
 and Chemical Evaluations of Solvent 
 Purity Required for FPC Production 
 
 E. Wick 
 
 ABSTRACT 
 
 The goal of these investigations was to gain maximum 
 understanding of the non-nutritive quality factors of fish 
 protein concentrates and of the relationships between these 
 factors and the methods used for FPC production. 
 
 III. 2-12 
 
B032 
 
 Effect of Heat on the Chemical and Nutritive 
 Stability of Fish Protein Concentrate (FPC) 
 
 D. L. Dubrow and B. R. Stillings 
 
 ABSTRACT 
 
 Experiments were conducted to determine the, heat stability 
 of FPC (fish protein concentrate) . FPC was subjected to one of 
 two forms of heat: (1) dry air (oven) and (2) moist (autoclave). 
 Temperatures of dry heat were 100°C, 120°C, or 150°C. Tempera- 
 tures of 100° or 200°C were used for autoclaving. Exposure 
 times for both treatments were 0, 30, 60, 120, or 240 minutes. 
 Proximate composition, amino acids and pH were determined. NPN 
 was also determined in autoclaved samples. PER values were ob- 
 tained for protein quality. Results of dry heat showed little 
 change to occur in samples heated to 120°C for up to four hours. 
 Samples heated to 150° C for 60 minutes or longer showed a de- 
 crease in lysine, arginine, available lysine and PER. Moist 
 heating produced an increase in NPN, a decrease in pH and little 
 change in amino acid concentration. The protein quality was 
 decreased after four hours at 120°C. 
 
 III. 2-13 
 
B034 
 
 Lipid Composition of Fish Protein Concentrate 
 H. Olcott 
 
 ABSTRACT 
 
 Studies were conducted on pout, Atlantic herring, Pacific 
 anchovy, Gulf menhaden (lot nos. 60-400, 69-404, and 68-402), 
 alewife, and Atlantic menhaden FPC meals. Samples of pout, Gulf 
 menhaden (60-400), and alewife FPC were stored at 31°C and at 
 50°C to determine what changes occur in lipid composition under 
 these conditions. The percentages of the more polar lipids 
 and of the shorter chain (14 to 18-carbon atom) fatty acids 
 increased with storage at the higher temperatures. 
 
 Samples of Atlantic menhaden and Gulf menhaden (68-402 
 and 69-404) were stored at 37°C and 50°C. They are now being 
 analyzed. 
 
 III. 2-14 
 
B037 
 
 Biological Availability of the Fluoride of Fish 
 Protein Concentrate in the Rat 
 
 I, Zipkin, S. M. Zucas and B, R, Stillings 
 
 ABSTRACT 
 
 The purpose of this study was to compare the availability of 
 fluoride of fish protein concentrate (FPC) with that of sodium 
 fluoride added to casein diets in the rat. The FPC in this study 
 was prepared by isopropyl alcohol extraction of red hake ( Urophy - 
 cis chuss) and it contained 8A% protein, 2 76 ppm fluoride, and 
 13% ash. Three diets contained approximately 20, 50, and 80 ppm 
 fluoride furnished by the incorporation of approximately 10%, 20% 
 and 30% FPC in the diet. Three other diets contained similar 
 amounts of fluoride as sodium fluoride and protein from casein. 
 The 10% casein diet without added fluoride acted as the control. 
 At the end of 28 days, femurs, mandibles, molars and incisors of 
 five animals \^eve analyzed for fluoride. The individual carcasses 
 of the remaining five rats of each group were also analyzed for 
 fluoride. Tlie fluoride in the diets containing 20, bO and 80 ppm 
 fluoride as FPC was A2 to 52%, 39 to 49%, and 25 to 35% as avail- 
 able as similar concentrations of fluoride added to the casein 
 diets in the form of sodium fluoride. No incisor stirations were 
 seen in any of the groups. 
 
 III. 2-15 
 
B040 
 
 Fish Protein Concentrate: a New Source ot Dietary Protein 
 
 B. R. Stillings and G. M. Knobl, Jr. 
 
 ABSTRACT 
 
 The concept of fish protein concentrate (FPC) is based on 
 the more efficient use of our fishery resources by converting 
 underutilized fish to acceptable products for human consumption. 
 FPC is not a single product; it is rather a family of products 
 produced by different processes. Each member in the family of 
 products has different characteristics and can be used for dif- 
 ferent purposes. Although a variety of processes has been de- 
 veloped, most have been based on solvent extraction of whole fish 
 to remove lipids and water. FPC produced by solvent extraction 
 of fish of several different species contains between 75% and 
 95% protein, which is particularly high in quality. This type 
 of product can be used in foods to improve markedly their nutri- 
 tional quality without significantly changing their other charac- 
 teristics. FPC produced by some solvent extraction methods, 
 however, has limited functional properties. New processes are 
 being developed which use enzymes, various solvents, or a com- 
 bination of enzymes and solvents. Products from some of these 
 processes have iiiipiuved fuuctional properties and these FPCs 
 appear to be .particularly promising for use in foods for their 
 functional attributes. Several problems still remain to be 
 solved and research is needed to determine hov7 FPCs can be pro- 
 duced and utilized most efficiently and effectively. There is, 
 however, a commercial industry emerging, and indications are 
 that FPCs will find a significant place in the market. 
 
 III. 2-16 
 
B04I 
 
 Comparative Utilization of Casein,' Fish Protein 
 Concentrate and Isolated Soybean Protein 
 In Liquid Diets for Growth of Baby Pigs 
 
 W. G. Pond, W. Snyder, E. F. Walker, Jr. 
 B. R. Stillings and V. Sldwell 
 
 ABSTRACT 
 
 Milk substitutes were needed for use in feeding early- 
 weaned pigs and in human infant feeding. The use of vegetable 
 proteins in infant feeding was reviewed by Tepley and Gyorgy 
 (1962) . Soybean protein had been most thoroughly studied and . 
 was probably in most common use (Forman, 1959; Forman et al., 
 1963; Omans, Leuterer and Gyorgy, 1963; Schneider and Sarett, 
 1969) . The inferior utilization of soybean protein as compared 
 with milk protein by baby pigs^ (Maner, Pond and Loosli, 1961; 
 Schneider and Sarett, 1969) suggested a need for alternative 
 protein sources for use in feeding the neonate. 
 
 A fish protein concentrate (FPC) developed by the U. S. 
 Bureau of Commercial Fisheries by isopropyl alcohol extraction 
 of raw whole red hake (Urophycis chuss) was described. This 
 FPC contained 80 to 85% crude protein of high biological value 
 as measured in weanling raLs, An FPC derived from isopropyl 
 alcohol-extracted, steam-dried herring meal and containing 80 
 to 85% crude protein was also available. 
 
 The purpose of the work was to compare the adequacy of 
 casein, the above FPCs and isolated soybean protein (ISP) for 
 growth and their effects on hematology of baby pigs. 
 
 III. 2-17 
 
B047 
 
 Enzymatic Solubilization of Fish Frotein Concentrate: 
 Batch Studies Applicable to Continuous Enzyrae Recycling Pi^-ocesses 
 
 C. Cheftel, M. Ahern, D. I. C. Wang and S. R. Tannenbaum 
 
 ABSTRACT 
 
 Batch experiments were performed to obtain information for 
 a continuous solubilization process of fish protein concentrate 
 by proteolytic enzymes. The batch kinetic data were gathered 
 with the intention of converting this system into a contin- 
 uous reactor capable of enzyme reusage through the use of 
 an ultrafiltration membrane. Methods were adapted in order to 
 assay proteolytic activity, particularly in FPC hydrolyzates . 
 Among the relative activities of various proteases for FPC 
 solubilization, pepsin and Pronase were particularly effective. 
 The effects of pH, temperature, substrate, and enzyme concen- 
 trations on the rate and extent of FPC proteolysis and sol- 
 ubilization by Pronase were tested. The stability of Pronase 
 
 was determined during FPC hydrolysis. After 8 hours of proteolysis 
 
 _ . o 
 at j4 C, Prona.se degradation was approxyua Lely 30% dud FroriAse 
 
 inhibition by the products of proteolysis was 60%. Gel chro- 
 matography showed the molecular weights of the soluble peptides 
 to be less than 2000; those of the proteolytic enzymes of 
 Pronase were close to 20,000. 
 
 III. 2-18 
 
B045 
 
 Moisture Adsorption of Fish Protein Concentrate at 
 Various Relative Humidities and Temperatures 
 
 J. G. Rasekh, B. R. Stillings and D. L. Dubrow 
 
 ABSTRACT 
 
 A simple method was used to determine moisture adsorption 
 isotherms of FPC (fish protein concentrate) made from whole fish 
 (red hake and Gulf menhaden) by isopropyl alcohol extraction. 
 The hake was ground to two particle sizes and either nons teamed 
 or steam stripped to remove residual alcohol. Moisture adsorp- 
 tion isotherms were determined at 25°, 35° and 42°C in constant 
 relative humidities ranging from 11-86%. Equilibrium moisture 
 content ranged from about 5% at 11% RH to about 16% and 86% RH. 
 Particle size of the hake FPC did not appear to affect the mois- 
 ture adsorption. Steam-stripped samples adsorbed slightly more 
 moisture at low relative humidities and slightly less moisture 
 at high relative humidities than did nons teamed samples. Men- 
 haden FPC adsorbed slightly less moisture than comparable hake- 
 FPC samples. This difference was due to differences in quanti- 
 tative chemical composition of the two types of FPCs. 
 
 III. 2-19 
 
B049 
 
 Availability of Amino Acids in Sarcoplasmic Fish 
 Proteins Complexed with Sodium Hexametaphosphate 
 
 G, A. Pelroy and J. Spinelli 
 
 ABSTRACT 
 
 Comparisons were made on the availability (enzyme digestion) 
 of four amino acids (lysine, methionine, tryptophan, and threonine) in 
 sarcoplasmic fish proteins and sarcoplasmic fish proteins complexed 
 with sodium hexametaphosphate. The protein complexes were formed 
 by reacting 1% solutions of sarcoplasmic fish proteins at pH 
 values ranging from 2-4 and at phosphate concentrations ranging 
 from 0.005-0.05 M. Uncomplexed sarcoplasmic protein was obtained 
 by precipitation from 70% isopropanol. The samples were digested 
 with the enzyme Pronase and the available amino acids were measured 
 with a bio-assay system. No significant differences in the 
 availability (enzyme digestion) of amino acids were found between 
 complexed and noncomplexed samples. Assays of samples dried at 
 75 C showed that nearly 100% of the methionine, threonine and 
 tryptophan and 85% of the lysine were available. l-Zhen the 
 samples were subjected to moist heat (50% moisture, 100 C for 
 up to 24 hr . )availability of lysine methionine, tryptophan and 
 threonine was reduced by 35, 22, 39 and 347o respectively. Results 
 of this study indicated that the nutritional characteristics of 
 fish proteins were unimpaired when they were complexed with sodium 
 hexame taphosphate . 
 
 HI. 2-20 
 
B05I 
 
 Survival of Microorganisms in Fish Protein Concentirate 
 Stored Under Controlled Conditions 
 
 D. Goldmintz 
 
 ABSTRACT 
 
 Fish protein concentrate, inoculated with Salmonella 
 montevideo , Stap h ylococcus aureus , and a Bacillus sp . , was 
 stored at 35°C for four months at constant relative humidities, 
 The numbers of S. montevideo and S. aureus dropped sharply as 
 the moisture increased whereas the Bacillus sp . decreased 
 slowly over the study period with little change at different 
 humidities. The initial extraneous mold count was near zero, 
 but proliferation occurred rapidly at high humidity. This was 
 in direct contrast to the survival pattern of the bacteria. 
 
 III. 2-21 
 
B052 
 
 Mercury Levels in FPC 
 
 W. E. Fox and E. G. Zook 
 
 ABSTRACT 
 
 Data are given on the mercury content of a number of 
 samples of raw, whole Pacific hake (Me rluccius productus ) and 
 of the fish protein concentrates prepared from them by extrac- 
 tion with isopropyl alcohol. Mercury concentration ratio was 
 about four in contrast to a solids increase of seven. Mercury 
 contents of several FPCs prepared from anchovy (Engr aulis 
 mordax ) , Pacific herring (Clu pea harengus pallasi ) , Pacific 
 hake ( Merluccius productus ) , Atlantic menhaden ( Brevoortia 
 tyrannus ) and red hake ( Urophycis chuss ) are given. Mercury 
 levels were found to be below 0.5 ppm. 
 
 III. 2-22 
 
B053 
 
 Simultaneous Determination of Isopropyl Alcohol and Water 
 In Fish Protein Concentrate by Gas Chromatography 
 
 V. White and E. Zook 
 
 ABSTRACT 
 
 Using a 10 ft. X 1/4 in. stainless steel column packed with 80-100 mesh 
 Porapak Q in a Hewlett-Packard llodel 810 gas chromatograph equipped with both 
 flame ionization and thermal conductivity detectors a method was developed for 
 the simultaneous determination of isopropyl alcohol (IPA) and water in fish 
 protein concentrate (FPC) . It was compared to several other procedures for 
 determining IPA and was found to be the fastest, to use the least equipment 
 for sample preparation, to be reproducible, and to give results comparable 
 with the other procedures. 
 
 III. 2-23 
 
B054 
 
 Further Studies on the Availability of Fluoride 
 In Fish Protein Concentrate 
 
 B. Stillings, H. R. Lagally, E. G. Zook and I. Zipkin 
 
 ABSTRACT 
 
 Previous studies showed that the biological availability 
 of FPC-F (fish protein concentrate-fluoride) was: (1) only 
 25-52 percent of that of NaF in weanling female rats (Zipkin 
 et al.j J. Nutr. 100:293, 1970) and (2) similar to that of NaF in 
 adult male humans (Spencer et al.^ Fed.Proc. 27 :423, 1968). In 
 the present studies, attempts were made to determine the reasons 
 for the differences between the previous two studies. Weanling 
 and adult male rats were placed in metabolism cages and fed 
 diets containing F from either FPC or NaF for four weeks. In 
 the first study, when diets contained 32 ppm F, the ratios of 
 FPC-F/NaF in the feces and carcasses of weanling rats were 1.86 
 and 0.65. For adult rats, corresponding ratios were 1.63 and 
 0.69. When diets contained 70 ppm F, ratios for weanling rats 
 were 2.44 and 0.30 and for adult rats they were 1.75 and 0.73. 
 These results were similar to those previously found with rats. 
 In the second study,, diets contained 7 ppm F and the F intakes 
 per unit of body weight were similar to those in the human study. 
 The ratios of FPC-F/NaF in the feces of weanling and adult rats 
 were 1.08 and 1.10. In the third study, diets were reduced to 
 levels closer to those used in the human study. Results were 
 close to those previously found with humans. Differences in 
 results from the previous two studies appear to be partially 
 due to differences in the amounts of F, Ca, and P used in the 
 diets. 
 
 III. 2-24 
 
B055 
 
 The Chemistry of Flavor in Fish Protein Concentrate 
 and Chemical Evaluations of Solvent Purity 
 Required for FPC Production 
 
 E. Wick 
 
 ABSTRACT 
 
 A gradual increase in the levels of amine hydrochlorides and 
 continued bland aromas in all samples were absorbed as the storage 
 period progressed. Gas chromatographic analyses, which must be 
 recognized as qualitative rather than quantitative in nature, 
 suggest that ammonia is produced throughout the storage period, 
 particularly in samples stored at high humidity. This degradation 
 of the FPC appears to have no ill effect on the overall quality 
 of the product, at least as viewed from the limits of this study. 
 
 Bacteriologically , there was no change in the bacteria plate 
 count. The mold count increased at the higher temperatures and 
 humidities. 
 
 Bread and cookies were made v/ith the stored FPC at the close 
 of each storage period. There was no change in the performance 
 of the FPC. There were no significant differences between flavor, 
 texture or appearance of the bread, cookies or in the "grit test." 
 
 III. 2-25 
 
B036 
 
 Effect of Hydrogen Peroxide on the Color, 
 Composition and Nutritive Quality of FPC 
 (Fish Protein Concentrate) 
 
 J. Rasekh, B. R. Stillings and V. Sidwell 
 
 ABSTRACT 
 
 Various levels of hydrogen peroxide were used to determine 
 the lev.el needed to lighten the color of FPC made from Atlantic 
 menhaden ( Brevoortis tyrannus ) by countercurrent extraction with 
 91% isopropyl alcohol and still maintain^a high nutritive quality. 
 At the 5% level of H2O2 the dry FPC, wet paste, and the cookies 
 containing 10% FPC had the lightest color. Panelists found no 
 difference in the texture and flavor of the cookies containing 
 107o FPC bleached with varying levels of H„0„. The cookies with 
 the FPC bleached with 2.57o or 5.0% H2O2 were significantly lighter 
 than the cookies containing the FPC that had not been treated. The 
 6% level of H2O2 had no effect on the amino acid composition or the 
 nutritive quality of the FPC. The higher levels tended to lower the 
 nutritive quality of the FPC. This might have been due to the oxida- 
 tion of the sulfur amino acids. 
 
 III. 2-26 
 
BC57 
 
 Effects of Processing Parameters on the Chemical 
 and Functional Characteristics of FPC 
 
 D. L. Dubrow 
 
 ABSTRACT 
 
 Information in the literature on solvent extraction of 
 
 fish showed that, in most cases, the extraction temperature was 
 usually at or near the boiling point of the solvent used. In 
 the case of isopropyl alcohol (IPA) , extractions were made at 
 close to 80*^. Since most proteins underwent changes in solubil- 
 ity characteristics at about 40-60° C, the FPCs produced lacked 
 the desired functional properties of the native proteins. 
 
 Even without elevated temperatures, protein denaturation 
 occured in the presence of various alcohols, including IPA. 
 Several reports have been published with respect to changes in 
 protein characteristics when in contact with IPA and other alco- 
 hols. It was pointed out that the rate of thermal denaturation 
 ol sei.uiii glouuliu was increased by ethanol and the rate increased 
 as its concentration increased. 
 
 Results of experiments conducted at the Fishery Products 
 Technology Laboratory, College Park, Maryland, to determine the 
 lipid and protein characteristics of FPC when extracted at 20°, 
 40 and 50° C clearly showed the effects of temperature on the 
 two fractions. Similar experiments were conducted to determine 
 the effects of drying temperature and steam desolventizing on 
 the chemical and functional characteristics of FPC. 
 
 III. 2-27 
 
B058 
 
 Effect of Spray Drying, Freeze Drying, and 
 Stabilizers on the Functional Properties of 
 Fish Protein Concentrate (FPC) 
 
 J. Rasekh and B. R. Stillings 
 
 ABSTRACT 
 
 The purpose of this study was to find ways to improve the 
 functional properties of fish protein concentrate (FPC). The 
 spray drying or freeze drying with or without the addition of 
 a stabilizer made no major change in pH, suspended solids iri a 
 water and FPC mixture, or emulsion capacity. The spray drying 
 or freeze drying did change the hydrophobic character of the 
 FPC; therefore, it was much more wettable. The addition of 
 Tween 80 increased the rate of wetting. The treatments lowered 
 the bulk density and increased the water swelling and water binding 
 index. The amount (2-3%) of soluble protein did not change with 
 the various treatments, but the addition of Mynoplex 600 did increase 
 the solubility to 7.1%. 
 
 III. 2-28 
 
B059 
 
 Functionality Modifications of Fish Protein Concentrate 
 
 A. F. Anglemier 
 
 ABSTRACT 
 
 The objective of the work done under contract with Dr. Anglemier, 
 Department of Food Science and Technology. Oregon State University, 
 Corvallis, Oregon, was to improve the functional properties of FPC 
 made by the isopropanol extraction method by chemical modification. 
 
 Of all the methods investigated, the alkaline hydrolysate was 
 the most successful in modifying the solvent extracted FPC. The re- 
 sulting product showed foaming properties comparable to egg albumin. 
 The product did have a bitter flavor and the nutritional quality was 
 extremely poor. 
 
 The use of surfactants did not improve appreciably the wettability 
 or solubility of FPC in water. Agglomeration of the FPC did reduce the 
 powder iiie as auu iuiiJLuve Liie uegiee Ol wet tctui-jLity , uut it di<a not produce 
 other more desirable functional qualities. 
 
 The alcohol extracted FPC could be used successfully in sausage 
 making. The FPC tended to indicate that it had water holding and fat 
 emulsion properties. The food technologist found that FPC can be in- 
 corporated successfully in extruded french fries, snack foods, and candy 
 with very little difficulties. 
 
 III. 2-29 
 
B063 
 
 Chronic Exposure of Rats to Methyl Mercury in Fish Protein 
 P. Newberne, 0. Glaser, and L. Friedman 
 
 ABSTEIACT 
 
 Fish protein concentrate (FPC) prepared by the National Marine Fisheries 
 Service from whole red hake (Urophycis chuss) by extraction with isopropyl 
 alcohol was fed to 5 generations of rats. The FPC was incorporated into a 
 semi-synthetic diet at a level to supply 20% protein to the diet. Casein 
 incorporated into the same basal diet and a commercial rat chow served as 
 control diets. The mercury content of the FPC used in these studies was 
 0.22 ppm Hg of which more than 907o was found to be methyl mercury. The casein 
 and chow diets averaged less than 0.01 ppm and 0.06 ppm respectively. 
 
 Congenital anomalies or abnormalities were not observed in any litters 
 of any generation and at no time were any significant clinical signs or 
 symptoms noted among rats fed the 3 diets, although the rats on the FPC 
 diet did collect mercury in their tissues. 
 
 III. 2-30 
 
B068 
 
 Safety Evaluation of Fish Protein Concentrate 
 Over Five Generations of Rats 
 
 P. M. Newberne; 0. Glaser; L. Friedman and B. Stilllngs 
 
 ABSTRACT 
 
 Fish protein concentrate (FPC) prepared from whole red 
 hake ( Urophycis chuss ) was fed as the sole source of protein to 
 five generations of rats; observations on growth, reproduction 
 and lactation of these groups were compared to control groups 
 fed either laboratory chow or a semisynthetic diet containing 
 casein as the protein source. Selected biochemical parameters 
 were also evaluated and tissues from all groups were assessed 
 for histopathologic alterations. Major differences noted were 
 better reproduction and decreased food efficiency in the chow 
 group, compared to either FPC or casein groups; toxicologic and 
 pathologic data were essentially the same for all groups. Based 
 on this evidence, FPC prepared according to methods developed 
 by the Bureau of Commercial Fisheries can be considered nutritious 
 safe and wholesome. 
 
 III. 2-31 
 
BOZO 
 
 Effect of Drying and Desolventizing on the Functional Properties 
 of Fish Protein Concentrate (FPC) 
 
 D, Dub row 
 
 ABSTRACT 
 
 Experiments were performed to determine the effects of 
 drying and steam desolventizing on the functional properties of 
 fish protein concentrate (FPC) . The FPCs were produced by a 
 room temperature extraction of either red hake or menhaden with 
 azeotropic isopropyl alcohol (IPA) . FPCs thus produced con- 
 tained about 36% soluble protein and when dried at ambient tem- 
 perature and pressure showed very little loss in protein solu- 
 bility. Drying the extracted wet solids at 40^ to 50°C, 60° to 
 70°C; 90^ to 100°C, or 140^ to 150°C for 30 or 120 minutes produced 
 decreased protein solubility, i.e., 30.7% (40'' to 50°C) to 12.5% 
 (lOO'^to 120 C) . Emulsion stability of an FPC-water-oil system 
 was satisfactory with all samples except those dried at 140" to 
 150°C. 
 
 Desolventizing dry solids or alcohol wet solids by steam 
 stripping produced a dramatic loss in soluble protein and emul- 
 sion stability. There was also a significant darkening in color 
 of the FPCs desolventized as wet solids as compared to FPCs 
 desolventized as dry solids. 
 
 III. 2-32 
 
B07I 
 
 Reduction of Mercury with Cysteine in 
 Comminuted Halibut ana Hake Fish frotein Concentrate 
 
 J. Spinnelli, M. Steinberg, R. Miller, A. Hall, L. Lehman (KMFS) 
 
 ABSTRACT 
 
 A study was made to determine the effectiveness of 
 cysteine in reducing the mercury content of comminuted fish 
 and fish protein concentrate (FPC). Comminuted fish tissue 
 was extracted with 0.1 M NaCl containing cysteine HCI-H in 
 concentrations ranging from to 0.5%. The amount of Hg that 
 could be extracted from the fish was related to the concentrations 
 of cysteine and the pH of the tissue, but the relation was not 
 linear. Fish blocks were prepared from the cysteine-extracted 
 comminuted tissue and did not develop off-flavor during 
 storage. When cysteine was added to isopropyl alcohol to 
 prepare FPC, the Hg in the FPC was reduced up to 50%. 
 
 III. 2-33 
 
T-^ 
 
 i)©e 
 
 .1 y © ill 
 
 
 CrtT fiBC?.!;. 
 
 ^ y ^4j^ 
 
 III. 3-1 
 
COOl 
 
 Fish Flour for Human Consumption 
 J. H. Olden 
 
 Abstract 
 
 This leaflet gives a review of the scientific literature 
 on fish flour; references to the original literature are 
 appended. 
 
 III. 3-2 
 
C009 
 
 Nutritional Quality of Wheat Flour and Bread Supplemented 
 With Fish Protein Concentrate or Lysine 
 
 .W. H. Kwee; V. Sidwell; R. C. Wiley; and 0. A. Hammerle 
 
 ABSTRACT 
 
 Studies were made to determine the relative supplemental 
 value of fish protein concentrate (FPC) and of lysine when added 
 to wheat flour. The nutritive quality of the supplemented wheat 
 flour was determined before and after processing into bread. 
 Mixtures were prepared containing wheat flour and either to 
 257o FPC or 0.1 to 1.07o lysine, l^en unprocessed mixtures were 
 added directly to rat diets at a 1.67o nitrogen level, 157o FPC 
 and 0.2 to 0.47o lysine produced the maximum weight gain, pro- 
 tein efficiency ratio (PER), net protein utilization, and total 
 protein and fat in the carcasses. Maximum responses obtained 
 with FPC were greater than those obtained with lysine. When 
 bread was added to diets at a 1.67o nitrogen level, 257. FPC and 
 0.47o lysine produced maximum weight gain and PER. The protein 
 ratings for bread buppleiaeated with ,10 to 257> FPC were 35 to 
 1507o higher than those for bread supplemented with lysine. 
 When bread was included in diets on the basis of 807. by weight, 
 the highest weight gains were produced with either 107o FPC or 
 0.47. lysine. With 107. FPC the weight gain was 367. higher than 
 with 0.47. lysine. 
 
 III. 3-3 
 
C017 
 
 Quality and Nutritive Value of Pasta Made from Rice, Corn, 
 Soya, and Tapioca Enriched with Fish Protein Concentrate 
 
 W. H, Kwee; V. Sidwell; R, C. Wiley; and 0. A. Hammerle 
 
 ABSTRACT 
 
 Fish protein concentrate supplemented to flours of 
 rice, corn, soya, and tapioca could contribute significantly to 
 the protein intake of the population of developing countries. 
 As pasta may well be considered a universal food, it was chosen 
 as a carrier for FPC. Pasta were evaluated organoleptically 
 and in animal feeding studies and were also evaluated obiec- 
 tively. Both 10 and 20% FPC additions were efficient in increas- 
 ing the protein content and nutritional value of pasta. From 
 sensory and objective evaluations, rice pasta appeared to be 
 the most acceptable. Of corn, soya, and tapioca pasta, tapioca 
 seemed most promising for further investigation; it promoted a 
 more attractive color and counteracted grittiness in the FPC, 
 but its texture became too soft during cooking. 
 
 III. 3-4 
 
C020 
 
 Phosphate Complexes of Soluble Fish Proteins 
 Their Formation and Possible Uses 
 
 J. Spinelli and B. Koury 
 
 ABSTRACT 
 
 Sarcoplasmic fish proteins in dilute aqueous concentrations 
 were quantitatively precipitated from solution as condensed phos- 
 phate complexes. The ratio of phosphorus to nitrogen in the com- 
 plexes was dependent on the concentration of the condensed phos- 
 phate present in the solution, the type of phosphate, and the 
 pH of the solution. Available lysine and feeding data indicated 
 that the nutritional characteristics of the complexed proteins 
 were not impaired. The work suggested that the reaction product 
 between condensed phosphates and soluble proteins could be used 
 to recover waste proteins from industrial effluents. 
 
 III. 3-5 
 
C02I 
 
 Changes in Physical and Sensory Characteristics of 
 Doughs and of Bread Containing Various 
 Amounts of Fish Protein 
 Concentrate and Lysine 
 
 V. D. Sidweil and 0. A. Hammerle 
 
 ABSTRACT 
 
 A study was made of the physical and sensory characteris- 
 tics of bread fortified with various amounts of either fish 
 protein concentrate (FPC) or lysine. Doughs were prepared from 
 mixtures containing wheat flour and 0, 5, 10, 15, 20, and 25% FPC 
 or 0.2, 0.^, 0.6, 0.8, and 1.0% lysine. Standard methods were 
 used to evaluate the rheology of doughs and characteristics of 
 bread. When 20% FPC was added to wheat flour at a ratio of 20% 
 or less, more water was absorbed by the dough than without FPC. 
 FPC increased farinograph development time and stability, and 
 in the extensigraph test it caused a decrease in extensibility 
 and an increase in resistance to extension. Fortification of 
 wheat flour with various amounts of lysine did not change the 
 characteristics of dough appreciably. Addition of FPC decreased 
 the volume of loaves, and the crumb became darker, coarser, and 
 more compact.. Addition of lycinc to bread had little effect on 
 either loaf volume or appearance. Judges accepted the texture 
 and flavor of bread with 5 or 10% FPC about as well as that with 
 no FPC. Bread containing higher amounts of FPC was less accepta- 
 ble. Acceptability studies revealed no appreciable differences 
 between breads containing various amounts of lysine. 
 
 III. 3-6 
 
C024 
 
 Nutritive Quality of Wheat Flour and Bread Supplemented 
 with Either Fish Protein Concentrate or Lysine 
 
 B. R. Stillings, V. D. Sidwell, 0. A. Hammerle 
 
 ABSTRACT 
 
 Studies were made to determine the relative supplemental 
 value of fish protein concentrate (FPC) and of lysine when added 
 to wheat flour. The nutritive quality of the supplemented wheat 
 flour was determined before and after processing into bread. 
 Mixtures were prepared containing wheat flour and either to 
 25% FPC or 0.1 to 1.0% lysine. When unprocessed mixtures were 
 added directly to rat diets at a 1.6% nitrogen level, 15% FPC 
 and 0.2 to 0.4% lysine produced the maximum weight gain, pro- 
 tein efficiency ratio (PER) , net protein utilization, and total 
 protein and fat in the carcasses. Maximum responses obtained 
 with FPC were greater than those obtained with lysine. I'Then 
 bread was added to diets at a 1.6% nitrogen level, 25% FPC and 
 0.4% lysine produced maximum weight gain and PER. The protein 
 ratings for bread supplemented witti 10 to 25% FPC were 35 to 
 150% higher than those for bread supplemented with lysine. 
 When bread was included in diets on the basis of 80% by weight, 
 the highest weight gains were produced with either 10% FPC or 
 0.4% lysine. With 10% FPC the weight gain was 36% higher than 
 with 0.4% lysine. 
 
 III. 3-7 
 
C025 
 
 Evidence Indicates, that a Premix of FPC and Wheat 
 Flour can be Made and Transported 
 
 V. D. Sidwell, B. R. Stillings and G. M. Knobl, Jr. 
 
 ABSTEAGT 
 
 The authors conducted a study to determine if mixtures 
 of wheat flour and fish protein concentrate (FPC) would tend 
 to separate during the agitation of a mechanical shaker. 
 Mixtures of 90% wheat flour and 10% FPC were placed on a 
 shaker for 168 hours. Despite differences in particle size 
 between the wheat flour and the FPCs, there was no evidence 
 of separation. 
 
 III. 3-8 
 
C028 
 
 FPC Institutional Recipes 
 R. G. Kerr and B. J. Gerson 
 
 ABSTRACT 
 
 The home economist in the Market Research and Services 
 Division of NOAA developed 22 recipes, including breads, desserts, 
 main dishes, sandwich filling, sauces and soups. These foods con- 
 tained as one of their ingredients 5-12% FPC; the percent of FPC 
 was based on the weight of the dry ingredients. The recipes were 
 considered acceptable in texture, flavor, and appearance when the 
 panelists rated it 90% or better. The inclusion of FPC in the 
 formulation increased the cost to one half cent or less per serving, 
 and increased the protein content two to three times over that of 
 a similar food product with no FPC. 
 
 III. 3-9 
 
C030 
 
 FPC Consumer Recipes 
 R. Kerr and B. Gerson 
 
 ABSTRACT 
 
 Fish protein concentrate (FPC) prepared from red 'hake ( Urophycis 
 chuss ) by the isopropanol extraction method was one of the ingredients 
 in a number of family-size recipes for breads, desserts, main dishes 
 or casseroles, sandwich filling, and soups. The recipes were modified 
 until the final products received a 95% acceptability by a panel of 
 10-12 judges. (Note: No decision was made for the publication of 
 these recipes.) 
 
 Hi. 3-10 
 
C031 
 
 Utilization of FPC and Acceptability 
 V. D. Sidwell 
 
 ABSTRACT 
 
 This review paper described the work that laboratories through- 
 out the world had done in the utilization of FPC. The work done at 
 the NMFS laboratory concentrated on the use of FPCs with varying in- 
 tensities of flavor and a comparison of the functional qualities of 
 FPC with so_y flour and dried milk solids. 
 
 Menhaden FPC was made under different conditions so that re- 
 sulting products would have varying kinds of physical characteristics. 
 Differences in flavor diminished when the FPCs were incorporated at the 
 107o level in bread and plain cookies. There was no difference in texture 
 when chocolate was added to the cookie. Because the menhaden-FPC v;as 
 naturally dark, the final products were dark, but even that characteristic 
 improved when incorporated into a food at the 10% level. The loaf volume 
 of the breads containing 10% FPC made from presscake or fish meals were 
 significantly greater than the breads made from regular FPC, probably 
 due to the residual fate in the FPCs that may act as a svrfaf" taTi*" - Tt was 
 found that a higher percentage of protein supplement could be added when 
 FPC was the supplement (9%) than if dried milk solids or soy flour (3-6%) 
 were used. Pasta fortified with 3 or 6% protein from FPC, 3oy flour, dried 
 milk solids, and .2% or .4% lysine was evaluated for its nutritional value 
 and its physical properties. 
 
 The suggested fortification level is 6.8% FPC of the total dry 
 ingredients. Indications were that when comparable amounts of FPC, soy 
 flour or milk solids were added to a formulation, FPC posed the fewest 
 problems in the modification of an established formulation or process. 
 
 III. 3-11 
 
C032 
 
 Crackers Fortified with Fish Protein Concentrate (FPC) 
 V. Sidwell and B. Stillings 
 
 ABSTRACT 
 
 In recent years considerable attention has been given 
 to the development and use of protein supplements. Among 
 these, fish protein concentrate (FPC) has been shown to be 
 especially high in nutritional quality and relatively low in 
 cost. 
 
 To be effective in providing protein for people, how- 
 ever, protein supplements must be in a form they will accept. 
 Obviously, people eat "foods" and not "nutrients"; thus it is 
 important that the nutrients be supplied in foods they enjoy 
 eating. Snacks are popular and provide an excellent vehicle 
 for supplying required nutrients. 
 
 In these studies, crackers containing varying amounts 
 of FPC were e'^'alu^te'^ ^^'^i^ nll^r^ ti nnal aualitv. sensory factors 
 and physical characteristics. 
 
 III. 3-12 
 
C033 
 
 Candy Fortified with Fish Protein Concentrate (FPC) 
 
 V. Sidwell 
 
 ABSTRACT 
 
 Fish protein concentrate (FPC) made by the isopropanol method 
 could be used to fortify candy with very few technological problems, 
 especially if the FPC was added at the end of the processing pro- 
 cedure. Up to 12% of the total weight of the ingredients was added 
 to caramels, 6% to cream fillings. The color of the final product 
 was the biggest difference between the fortified and the non-fortified 
 candies. The changes in texture and flavor were not significant. The 
 effect of color could be minimized by the addition of colorings, nuts, 
 or fruits. 
 
 III. 3-13 
 
C035 
 
 Effect of Varying Amounts of Fat and Dough Conditioner 
 on Loaf Volume and Rheology of Doughs 
 
 V. Sidwell and B. Stillings 
 
 ABSTRACT 
 
 This study was made to observe the effect of varying levels of fat 
 
 (lard) and dough conditions on the loaf volume. Three different dough 
 
 conditions were used--sucrose monotallowate (SMT) , sodium stearoly-2- 
 
 lactylate (SSL) , and a blend of mono- and diglyceride and polysorbate 
 
 60 (MDP) . The bread was prepared by the sponge method. Fish protein 
 
 concentrate was added at the rate of 10% of the flour. One percent of 
 
 SMT replaced 8% of the lard in che formulation and produced a comparable 
 
 size loaf of bread. The maximum size loaf was obtained when 3% SMT and 
 
 2 or 37c fat (lard) was used. The addition of .5% SSL, along with 1 or 2% 
 
 fat (lard) made the largest loaf of bread. It appeared that SSL had a 
 
 sparing effect on the use of fat in the formulation. MDP could replace 
 
 the fat that is used in the formulation, i.e., 1.5% MDP and 3% fat. 
 
 The best results were obtained when a small amount was added to improve 
 ^1 — 1 ji,'_„ „\ 
 
 
 The amylo/visco/graph showed that the addition of SMT changed the 
 viscosity of the mixture most markedly. It took more time to reach peak 
 viscosity with each addition of SSL, but in no case did it change the 
 density of the gel. Each level of MDP did not make a marked change in 
 the viscosity of the flour and FPC mixture. 
 
 The farinograph curves showed that with each addition of SMT, more 
 tim.e was needed to develop the dough. At the .5%. of SSL there was an 
 improvement in the stability of the dough over that with no or more SSL. 
 The use of MDP improved the stability of the dough, but no change was 
 very marked. 
 
 HI. 3-14 
 
^ 
 
 ©Stl©i1 
 
 ^1 
 
 PdlSTE^IAL 
 
 S^sKDil'ilC ASPECT 
 
 Hi. 4-1 
 
D007 
 
 FPC 
 An Economic Model for a New Industry 
 
 S. Wright (SWECO, Inc.) 
 
 ABSTRACT 
 
 The data in this brochure have been prepared from actual 
 tests and represent a definitive economic analysis of FPC plant 
 construction and operation. A basic modular design for an FPC 
 production plant was developed, and economic models are present- 
 ed for plants with capacities of 200, 400, and 600 tons of raw 
 fish per day. Graphs were developed for these models that show 
 the effects on gross profit (before selling expenses, general 
 and administrative expenses, and corporate taxes) of variations 
 in fishing season, cost of raw fish, and selling price of the 
 finished FPC. The data in these graphs were condensed from the 
 results of extensive computer analyses in which a number of 
 fixed costs were integrated with more than 50 combinations of 
 these major variables. 
 
 III. 4-2 
 
D098 
 
 Commercial Feasibility of Fish Protein 
 Concentrate in Developing Countries 
 
 Volume I - The Protein Situation in 
 Korea and the Potential Role for Fish 
 Protein Concentrate 
 
 General Oceanology, Inc. 
 
 ABSTRACT 
 
 This report established that protein malnutrition exists 
 in certain target groups in Korea, and that fish protein 
 concentrate might contribute to ameliorating the protein short- 
 age. Economics and raw materials shortages, however, mitigated 
 against a self-priming commercial FPC operation in Korea. Some 
 alternative paths for protein relief via FPC were, however, 
 possible. 
 
 The elements of protein nutrition and fortification were 
 first reviewed. The Korean protein system was described, and 
 the "protein gap" was calculated so that the need of nutrition- 
 ally deprived target groups could be determined. Alternatives 
 were proposed for meeting the TCnrppn prntein np.eds. including 
 the use of FPC. The cost/effectiveness of protein fortifica- 
 tion was determined. The feasibility of producing FPC locally 
 was examined, and alternatives were sought for obtaining sup- 
 plies of FPC in the future. The implications of a protein sup- 
 plement program on national nutrition policies were examined, 
 including the possible role of FPC, and the cost of closing 
 the "protein gap". 
 
 III. 4-3 
 
DOll 
 
 Fish Protein Concentrate: The Growth of an Industry 
 George M. Knobl, Jr. 
 
 ABSTRACT 
 
 This was a six page paper defining several types of FPCs, 
 some of their uses and characteristics, approved fish, active 
 or emminent commercial or experimental FPC plants with status 
 and descriptions of processes, some market and price considera- 
 tions, and present research efforts. 
 
 In summary, an FPC industry had been established and was 
 growing. Several plants were either in operation or under con- 
 struction. Isopropyl alcohol was used, in each instance, as the 
 solvent to remove water and lipids . The IPA-FPC produced was 
 essentially a "non-functional" product, but had a place on the 
 world market. FPCs with improved functional properties were 
 desirable and research was underway to develop them. Barring 
 unfoi-eseeu circums Lances , the FPC industry chculd have continued 
 to grow. 
 
 III. 4-4 
 
D012 
 
 The Economics of Fish Protein Concentrate 
 A. H. Kyle (MIT) 
 
 ABSTRACT 
 
 This contract report reviewed the dimensions of the world protein 
 situation and explored the possibility of applying cost benefit analyses 
 to investment in protein supplement programs. The conclusion was that 
 at the time of the report it was not useful to attempt to quantify the 
 benefits. FPC was compared with its alternatives, with the conclusion 
 that at contemporary (1970) cpsts, FPC would rarely be the most efficient 
 source for a developing country. Technical problems in evaluating costs 
 associated with balance of payments considerations, unemployment, and 
 the cost of capital were also discussed. Cost and demand data on FPC 
 for the specific case of Chile were developed. The cost in Chile was 
 estimated to be 25 cents per pound, and at this price FPC at best would 
 be marginally competitive with other protein sources. For the United^ 
 States, FPC's nutritional qualities would not necessarily guarantee it 
 a market where overnutrition is as big a problem as undernutrition. 
 With improvement in ^functional properties, FPC could begin to compete 
 with meat products. The public-good argument would support a well-funded 
 government research and development program in FPC. 
 
 III. 4-5 
 
D0I4 
 
 Utilization of Protein Ingredients in the 
 U.S. Food Industry 
 
 Part I 
 The Current Market for Protein Ingredients 
 
 Part II 
 The Future Market for Protein Ingredients 
 Timothy M. Hammonds and David L. Call 
 
 The first part of this contract report reviewed and 
 evaluated the existing market for new protein sources, specifi- 
 cally FPC. A substantial current market potential for protein 
 was identified and potentials of various proteins were estab- 
 lished. It was concluded, however, that the protein market was 
 not an easy one into which to introduce FPC. 
 
 Part II of the report discussed the vulnerability of each 
 food category investigated in Part I. A market guide was 
 developed, presenting a summary of major foods using protein, 
 the properties required to penetrate each category, the competi- 
 tive price range needed (1969) , and an estimate of protein 
 volume potential. The canned and processed meat industry was 
 identified as having the largest growth potential. The other 
 fastest growing markets, in order of growth, were instant 
 breakfast, snack food, and pet food. Canned and processed meats, 
 baked goods, and breakfast products were certified as the prime targets 
 
 for new proteins. Dairy products may be secondary targets 
 because of problems of penetration, and pet food because of the 
 very low cost required. The ingredients with which a new 
 protein may expect to compete with were identified as the soy pro- 
 teins, nonfat dry milk, and dry whole milk. Dry whole milk has 
 the lowest protein content and is often included for its fat, not 
 its protein content. This leaves soy and nonfat dry milk as the 
 principal targets. Elimination of disagreeable flavor was determined 
 to be the single most important factor for stimulating sales of new 
 protein ingredients. 
 
 III. 4-6 
 
D016 
 
 Preparation of Fish Protein 
 Concentrate and Fish Meal 
 
 John Spinelli 
 
 ABSTRACT 
 
 Raw, ground, whole or deboned fish was contacted with an 
 acidic aqueous solution of a condensed inorganic phosphate, e.g., 
 sodium hexametaphosphate, to insolubilize the protein fractions 
 of the tissue components. The insolubilized protein could then be 
 water washed to removed oil and other undesired tissue components. 
 Either fish meal, suitable for animal feed, fertilizer, etc., or 
 high protein concentrate, suitable for human food use, was recov- 
 ered. 
 
 III. 4-7 
 
D017 
 
 Cost Comparison of IPA and Presscake - IPA Process 
 L. Durilla, K. Almenas, and G. Gentry 
 
 ABSTRACT 
 
 A cost comparison of the standard IPA and Presscake-IPA 
 process for the production of fish protein concentrate (FPC) was 
 made. The major advantages of preprocessing the fish before it 
 is fed into the IPA processing plant appeared to be: 
 
 1. A fattier fish could be used as the raw material. The 
 cooking and pressing operations used in producing the presscake 
 feed also removed the major portion of the oil associated with the 
 fish. 
 
 2. Cooked presscake could be stored readily. This could 
 be taken advantage of in extending the operation time of the IPA 
 processing units. 
 
 3. Presscake was a more concentrated feed, thus processing 
 capacity for an equivalent amount of output product can be reduced, 
 
 III. 4-8 
 
D022 
 
 The Economics of Fish Protein Concentrate 
 J. Crutchfield and R. Deacon 
 
 ABSTRACT 
 
 Research conducted on fish protein concentrate in the 
 past has been concerned primarily with processing technology 
 and nutritional evaluation. Efforts to evaluate the economic 
 role of FPC in world markets have been few by comparison. Al- 
 though such research has been of generally high quality, it has 
 often been limited in scope. Some analyses have focused upon 
 single locations for production and consumption and have exam- 
 ined FPC in light of a single specific use; others have looked 
 at potential markets with little or no analysis of economic as- 
 pects of production; most reports have evaluated FPC on the ba- 
 sis of present technology and have not tried to assess potential 
 gains from further research. A more global approach to the prob- 
 lem, considering production, marketing and the allocation of 
 research effort is needed. An outline of such an approach and 
 some preliminary ccnclusicnc from this type of analysis are 
 provided in this paper. 
 
 III. 4-9 
 
D023 
 
 Engineering Economic Model 
 for Fish Protein Concentration Processes 
 
 K. K. Almenas, J. W. Gentry, and J. M. Marchello 
 
 ABSTRACT 
 
 A process engineering economic model for fish protein 
 concentration processes has been developed. The model predicts 
 the construction and operating costs for fish meal plants and 
 for plants producing fish protein concentrate (FPC) by iso- 
 propyl alcohol extraction, biological, and presscake isopropyl 
 alcohol extraction. 
 
 Typical process flow sheets and a computer program were 
 developed to be used in the design and cost computations. The 
 model provided for each process to be studied both internally 
 and externally in comparison with alternate processes. The pro- 
 gram and model were prepared in such a way that changes and up- 
 dating may be accomplished quite readily as new information be- 
 comes available. This report contains directions for users and 
 descriptions' of the processes. A listing of the computer pro- 
 gram and example calculations for each process are presented in 
 the appendix to guide the user and to illustrate the nature of 
 the model output. 
 
 While the model did develop construction, operating, and 
 production aspects of the processes, it did not deal with the 
 economics of selling the products and the resulting profit and 
 return on investment. Problems of allocation of costs and mar- 
 keting arrangements were not covered in this report, but must be 
 considered in the final decisions relating to a complete evalua- 
 tion of alternatives. 
 
 III. 4-10 
 
AT©i^Y 
 
 OC 
 
 s 
 
 III. 5-1 
 
E005 
 
 Design and Construction of a Fish Protein Concentrate 
 Processing Unit Utilizing Solvent Extraction Procedures 
 
 J. Connally 
 
 ABSTRACT 
 
 A model-scale plant for processing raw fish by a variety 
 of solvent extraction procedures was designed, constructed, 
 tested, and delivered to the College Park Laboratory of the Bu- 
 reau of Commercial Fisheries of the U.S. Department of the Inter- 
 ior. Built largely of stainless steel, with sanitary fittings 
 In all important sections, the unit permits the processing of 
 over 500 pounds of raw fish per day using the following process- 
 ing steps: 
 
 1. Batch extraction 
 
 2. Continuous liquid-solid separation 
 
 3. Solvent recovery by distillation 
 
 4. Vacuum drying of extracted fish proteins. 
 
 Experimental runs were made using different tjrpes of fish. 
 The lowest fat content was achieved by an isopropanol extraction 
 operation developed by BCF personnel: the process utilized two 
 batch extractions and a continuous percolation of solvent through 
 a fish slurry. This produced a fish protein concentrate product 
 containing approximately 70% protein with a fat content of 0.3%. 
 
 III. 5-2 
 
E008 
 
 Design and Evaluation of a Process 
 for the Manufacture of Fish Protein Concentrate 
 
 IONICS, Inc. 
 
 ABSTRACT 
 
 For the past several years the U. S. Bureau of Commercial 
 Fisheries has concerned itself with the development of a high 
 quality fish protein concentrate which would be suitable for 
 worldwide use in human diets. The major emphasis of their 
 work was directed toward producing samples of fish protein 
 concentrate and evaluating them from the standpoint of nutri- 
 tional value. These samples proved to be excellent sources of 
 animal protein. Therefore, it now remained to be demonstrated 
 whether this process could yield high quality FPC at a low cost. 
 The analyses performed were concerned with demonstrating the 
 technical feasibility and estimating the manufacturing costs 
 of FPC plants capable of handling either 5 or 50 tons per day 
 of lean, raw fish. IONICS investigated the variables associated 
 with the dehydration and extraction steps such as time of contact, 
 temperature, solvent composition and solvent to fish ratio, as 
 well as the general area of recovery of valuable components from 
 an FPC miscella. 
 
 III. 5-3 
 
E013 
 
 Odor Removal from Reclaimed 
 AIPA in Marine- Protein Concentrate Process 
 
 (Report No. I) 
 S. C. Spalding 
 
 ABSTRACT 
 
 Azeotropic isopropanol recovered from the process for 
 preparing a food grade marine protein concentrate was shown 
 to be contaminated by amines which were volatile and which gave 
 the alcohol a fishy to ammoniacal odor overtone character. To 
 remove the amines and deodorize this product so that it may be 
 reused in the process it was recommended that the hydrogen form 
 of a strong acid type (Dowex SOW x 8) ion exchange resin be used 
 to collect these amine compounds. The remaining overtone odors 
 which then become noticeable are removed by one pass through an 
 absorbant carbon (Bamebey Cheney RA grade carbon) column there- 
 by completing the regeneration of the solvent isopropanol for 
 reuse in the process. 
 
 III. 5-4 
 
 # 
 
E015 
 
 Odor Removal from Reclaimed 
 AIPA in Marine Protein Concentrate Process 
 
 (Report No. II) 
 
 S. C. Spalding 
 
 ABSTRACT 
 
 A series of odor panel samples were submitted to BCF to 
 demonstrate certain process sequences which yield an uncontam- 
 inating (odor-wise) constant boiling isopropanol-water product 
 judged as satisfactory by an odor evaluation panel. 
 
 Three deodorization processes were studied in greater 
 detail to be certain that the several unit operations employed 
 would provide reasonable engineering solutions. The product of 
 each of the three procedures passed an odor panel trial and 
 
 1. The present cross flow process should be changed over 
 to a counter current process so that proper simulation of product 
 streams may be obtained for study. 
 
 2. Of the three processes found to be capable of prodncnng 
 an acceptable product, the two step fractional distillation — 
 flash evaporation scheme should be applied to the 600///day pilot 
 plant runs contemplated under item 1 above. 
 
 III. 5-5 
 
.E016 
 
 A Study of Centrifugal Methods for Fluorine Reduction 
 
 In Fish Protein Concentrate and Fish Protein Recovery 
 
 From Isopropanol Suspension 
 
 C. Y. Lee, G. F. Schneider, W. A. Ruemell 
 
 ABSTRACT 
 
 It has been shown that reduction of fluorine in red hake 
 protein concentrate to less than 100 ppm is possible by centri- 
 fugal deboning in an isopropanol-water slurry under certain con- 
 trolled conditions. Yields of 80 to 85% dry solids input are 
 indicated at rates up to 30 GPM with the P-660 Super-D-Canter. 
 The importance of comminution conditions upon yield and fluorine 
 removal was confirmed; fluorine removal appears surprisingly 
 dependent upon isopropanol concentration as well. 
 
 Recovery of red hake protein concentrate from isopropanol 
 suspension was effected at 0.3 to 0.7 GPM with the P-660 Super- 
 D-Canter. Insoluble solids removals varied from 97 to 85% at 
 0.3 and 0.7 GPM, respectively. 
 
 The use of centrifugal equipment for both fluorine removal 
 and protein recovery appeared practicable. However, protein 
 recovery, particularly after the first stage extraction, would 
 require the use of a decanter and disc centrifuge in series if 
 practical recoveries are to be achieved. 
 
 III. 5-6 
 
E020 
 
 A Mechanism of Mass Transfer in a Red Kake 
 (Uyophycis chuss) - Azeotropic Isopropanol System 
 
 D. Harris, R. M. Heck, and H, G. Blocker 
 
 ABSTRACT 
 
 The Bureau of Commercial Fisheries embarked on a program 
 to design a countercurrent extraction process for the production 
 of fish protein concentrate from red hake ( Urophycis chuss ) . 
 The extraction process involved contacting azeotropic isopropyl 
 al'cohol (AIPA) with ground fish to removed lipids and water from 
 the fish solids. The wet extracted solids were then dried and 
 steam stripped to remove residual alcohol. The resulting dried 
 product was then milled and bagged. The FPC product was approved 
 by the Food and Drug Administration for human consumption. 
 
 The design of the extraction system for this process re- 
 quired fundamental data on the mechanism of the extraction of 
 the lipids and the rate of extraction as a function of tempera- 
 ture, impeller speed, and power input. The lipids needed to be 
 extracted because they were a class of compounds that contribute 
 to spoixage oj. j-'ivj uncier storagE, 
 
 There was no known work in the literature concerning mass 
 transfer, extraction rate, approach to equilibrium, of mass 
 transfer temperature dependence concerning red hake extraction 
 processes. The purpose of the work reported here was to obtain 
 such information for use in process design. 
 
 III. 5-7 
 
E022 
 
 Bone Particulate Separation — Fish Protein Concentrate 
 Barkeley and Dexter Laboratories, Inc. 
 
 ABSTRACT 
 
 The objectives of this study were the separation of bone 
 fraction from FPC. Methods tested: cascading/air slit, cyclone/ 
 vortex, electrostatic, vibration/flotation separation techniques. 
 Based on the assumption that FPC may contain as high as 2000 ppm 
 F, fillets contain 45 ppm and it was determined desirable to re- 
 duce fluoride content to 100 ppm in final FPC fraction. 
 
 Reliable analysis was a major problem. An x-ray method of 
 control proved inadequate. Ash contents were indicative but not 
 accurate. Fluoride assays and ash analyses were also made by 
 Skinner and SherTn?in; Tnr , . Newton Mass.. Barkley and Dexter used 
 an Orion specific ion meter for fluoride assays. 
 
 The cascade/air slit method applied to milled FPC initially 
 containing 360 ppm F resulted in fractions containing as high as 
 830 ppm F and as low as 130 ppm. The cyclone/vortex method re- 
 sulted in fractions as low as 99 ppm F. Electrostatic separation 
 of unmilled FPC indicated no significant separation. Vibration 
 separation tests resulted in fractions from 350 ppm F down to 
 100 or 150 ppm F. It was concluded that vibration/separation 
 can recover 84% with 200 ppm F. 
 
 The contractor recommended the construction and evaluation of 
 a pilot plant unit combining methods of bone separation. A recovery 
 of 52% containing 100 to 150 ppm F was predicted. 
 
 Conclusion: Fluoride content cannot be reduced effectively 
 to less than 100 ppm by mechanical separation in view of fact that 
 milled fillet is of the same order F content. 
 
 III. 5-8 
 
E037 
 
 Controlling the Fluoride Level of Fish Protein Concentrate 
 By Selective Removal of Bones 
 
 L, Beiremain and W. rfamperson 
 
 ABSTRACT 
 
 The report consisted of five single spaced, typed pages 
 of data diagrams and a graph relating ash content to fluoride 
 content. The objective was to recover a low-fluoride fraction 
 from FPC [by utilizing milling/screening/air classification] and 
 to verify the assumption that the fluoride level would be rough- 
 ly proportional to bone content and whether lead was concentrated 
 in bone. 
 
 Unmilled hake FPC was subjected to various combinations 
 and permutations of coarse milling, air classification, and 
 screening with 20, 24, 32 and 48-mesh screens. Previous work 
 on an earlier sample had indicated good possibilities for these 
 techniques. It was not possible to achieve separation of a 
 satisfactory fraction with less than 100 ppm F by any sequence 
 applied. It was questioned whether variations in F content^ of 
 fish or methods of processing FPC (raw fish grinding, mixing, 
 extracting or temperatures) have resulted in the solids having 
 different physical or chemical composition. Samples separated 
 by floating in chloroform separated bone fractions containing 
 1200 ppm F (ash, 58.0%) and light fraction containing 148 ppm F 
 (ash, 8.2%) from a mixture containing 268 ppm F (ash, 19.5%). 
 A sample of sea bass, bones separated by hand, resulted in bones 
 containing ash, 48% (dry wt.) and flesh with ash, 3.1%. Hake 
 flesh fraction separated by chloroform contained 6.7% ash (158 
 ppm F) indicating the hake particles of flesh fraction contained 
 considerable bone. 
 
 The main conclusion drawn was that the two samples used 
 did not represent a material well suited to mechanical separa- 
 tion of the bone and meat portions and would not have resulted 
 in a product with fluoride reliably below 100 ppm. Further 
 study of FPC process was recommended to determine what conditions 
 affect the amount and distribution of fluorides, also coarse 
 grinding of fish prior to extraction. 
 
 III. 5-9 
 
E038 
 
 Use of Solvents in the Manufacture of 
 Fish Protein Concentrates 
 
 N, L, Brown, E. R. Pariser, and Jo A, Price 
 
 ABSTRACT 
 
 This survey was made under contract by Esso on potential candidates 
 for use in FPC manufacture; the report discussed the criteria used for 
 selection and presented a survey of some seven solvents which could be 
 used in FPC manufacture. 
 
 Technical criteria, categories, functions and history of solvent 
 use and selection relative to FPC were discussed. Unlike more commercial 
 solvent extractions, which were for recovery of the oil, FPC extraction 
 was for the purpose of completely removing the lipids from the substrate 
 and therefore further restricting the characteristics and function of the 
 solvent as to its effects on the FPC product. Isopropyl alcohol, ethanol, 
 methanol, butanols, n-hexane, n-heptane, acetone, trichloroethane, carbon 
 disulfide, cellosolve, dioxane, 1, 2-dichloroethane and ethylacetate were 
 some of the solvents that have been used singly or in combination for 
 FPC extraction. Few have been used in commercial productions. The 
 apnendix contained the Esso table of solvents and an pvnl^na^i on of 
 the table was given. The following were given in the list whenever 
 available: boiling point, molecular weight, refractive index, freezing 
 point, density, surface tension, water interfacial tension, dielective 
 constant, dipole moment, viscosity, vapor pressure, flash point, flamma- 
 bility limit, specific heat, heat of vaporization, Trouton's constant, 
 solubility parameter, degree of hydrogen bonding, water solubility, lipids 
 extracted, azeotrope with water, health hazard, availability and cost. 
 
 III. 5-10 
 
En4i 
 
 Using Enzymes to Make Fish Protein Concentrates 
 
 M. B. Hale 
 
 ABSTRACT 
 
 Fish protein concentrates having desirable functional 
 properties could be prepared by using selected enzymes to solu- 
 bilize protein and release lipids. Research by the National 
 Marine Fisheries Service into enzymatic processes was described 
 and published in the form of a Special Scientific Report. 
 
 The relative activities of 23 commercially available pro- 
 teolytic enzyme preparations acting on a fish protein substrate 
 were measured. Pancreatin, pepsin and papain had highest ac- 
 tivity per unit cost of enzyme. 
 
 Soluble hydrolysates were prepared from red hake (Uro- 
 phycis chuss ) using a variety of enzyme and digestion conditions. 
 Concentrations of the amino acids tryptohan and histidine in the 
 soluble products were critical nutritionally and varied with the 
 pH of hydrolysis. A soluble FPC having a protein efficiency ra- 
 tio equal to that of casein was prepared with an alkaline bacter- 
 
 -f - T „ „j. ^u Q C 
 
 -LclX ClliC^lUC CXI_ ^J1.L \J , J , 
 
 Production costs were estimated for a soluble product pre- 
 pared from whole fish and a partially soluble product prepared 
 from press cake through the use of enzymes. 
 
 III. 5-11 
 
E043 
 
 Effect of Ice Storage on 
 the Chemical and Nutritive Properties of 
 Solvent-Extracted Whole Fish - Red Hake, Urophycis chuss 
 
 D. L, Dubrow, N. L, Brown, E. R. Pariser, n. Miller, Jr., and V. Sidwell 
 
 ABSTRACT 
 
 Because red hake that are to be used in the future pro- 
 duction of fish protein concentrate will be caught in quantity, 
 the preservation of the hake during periods of glut will present 
 a problem that possibly can be solved by storage of the hake in 
 ice. 
 
 Whole red hake were held in ice for 2, 6, 8, and 11 days. 
 Organoleptic tests on the fresh fish showed that they were edible 
 on the eighth day but were not edible on the eleventh day. 
 
 Samples of fish were removed during each period of storage 
 and were processed (1) by freeze drying to produce a reference sam- 
 ple and (2) by solvent extraction with isopropyl alcohol to pro- 
 duce a fish protein concentrate. Proximate composition, amino 
 acid composition, and nutritive quality were determined compara- 
 tively on both of these two kinds nf nrocessfd samDles , 
 
 From the data obtained, it was concluded that red hake 
 stored in ice for 8 days were suitable for use in the production 
 of fish protein concentrate and that they would be suitable for 
 this use up to the point of spoilage of the fish, which occurred 
 sometime between 8 and 11 days. 
 
 III. 5-12 
 
E044 
 
 Solubilization of Fish Protein Concentrate 
 S. R. Tannenbaum, D. Wang, and M. C. Archer 
 
 ABSTRACT 
 
 The solubilization of FPC by Bacillus subtilis protease 
 (Monzyme) has been investigated. The conditions vjhich eliminate 
 the problems of microbial contamination and salt accumulation 
 were defined. A kinetic treatment revealed that enzyme was 
 absorbed to the surface of the substrate, the initial rate of 
 reaction being proportional to the surface area of substrate 
 exposed to aqueous phase. The overall kinetics were described 
 by a sequence of first order process - an initial, fast re- 
 action in which loosely bound polypeptide chains were cleared 
 from an insoluble protein particle, and a second, slower re- 
 action in which a more compacted core of the protein is digested. 
 
 Further studies showed that the solubilization of the de- 
 solventized or non-desolventized FPC did not change the physical 
 characcer of the £iiial piuduut. Meuhaden presscake studies shewed 
 that 50% of the total lipids was extracted into the soluble fraction 
 of the substrate. 
 
 The continuous production of the solubilized FPC in a pilot 
 reactor capable of producing many kilograms per day showed the 
 effects of FPC concentration, residence time and reactor temperature 
 correlated with the batch data obtained in former studies. 
 
 III. 5-13 
 
E046 
 
 Studies on the Use of Carbon Dioxide Dissolved in 
 Refrigerated Brine for the Preservation of Whole Fish 
 
 H. Barnett, R. W. Nelson, P. J. Hunter, S. Bauer, and H. Groniger 
 
 ABSTRACT 
 
 Although storing fish in refrigerated seawater had many 
 advantages over storing them in ice, the use of refrigerated 
 seawater also had several disadvantages, one of which was the 
 difficulty in controlling the growth of spoilage bacteria in the 
 fish. Reported here is the effect on the growth of bacteria in 
 rockfish and chum salmon of dissolving carbon dioxide in brine. 
 Storing the fish in the refrigerated brine treated with carbon 
 dioxide inhibited the growth of the bacteria, retarded the rate 
 at which the fish decrease in quality, and increased their stor- 
 age life by at least 1 week. 
 
 III. 5-14 
 
E047 
 
 Protein Autolysis Rates at Various pH's and Temperatures* 
 
 in Hake, Merluccius productus , and Pacific Herring, 
 
 Clupea harengus pallasi , and "Their Effect on 
 
 Yield in the Preparation of Fish 
 
 Protein Concentrate 
 
 B, Koury, J. Spinelli, D. Wieg 
 
 ABSTRACT 
 
 The rate of protein autolysis at temperatures ranging from 
 30° to 80° C and at pH's ranging from 3.0 to 7.0 was determined 
 on hake and Pacific herring. Autolysis rates were generally 
 greatest at acidic pH's and began to decrease after temperatures 
 exceeded 50°C. Autolysis rates were much greater in hake than 
 in Pacific herring. The yield of fish protein concentrate pre- 
 pared from hake showed a close inverse correlation to the degree 
 of autolysis. 
 
 III. 5-15 
 
E043 
 
 Making Fish Protein Concentrate by Enzymatic Hydrolysis 
 
 M. B. Hale 
 
 ABSTRACT 
 
 Research into biological methods for fish protein concen- 
 trate (FPC) preparation carried out v/ithin the National Marine 
 Fisheries Service is summarized. The effects of various process- 
 ing conditions and commercially available proteolytic enzymes 
 on yields and characteristics of water-soluble fish protein 
 hydrolysates are presented. Soluble FPC prepared from red hake 
 ( Urophycis chuss ) tended to be deficient in either tryptophan 
 or histidine, depending on the pH of hydrolysis. Hydrolysis of 
 raw fish with an alkaline protease of Bacilus subtilis at pH 
 8.5 or above gave the best balance of essential amino acids and 
 a high yield of soluble product. Pancreatin also gave very good 
 results at pH 8.5. The protein efficiency ratio (PER) of a to- 
 tally soluble FPC prepared from alewife ( Alosa pseudoharengus) 
 was equivalent to that of casein. Soluble products prepared 
 from hake were equivalent to casein as a wheat supplement but 
 not as a sole source of protein. Process outlines and prelimi- 
 nary cost estimates are presented for the production of two types 
 of fish protein hydrolysatp.s. Possible food uses and the flavor 
 problem are discussed. This report includes a literature survey 
 of fish protein modifications by fermentation and selected chem- 
 ical hydrolysis methods, as well as by enz3nnatic hydrolysis pro- 
 cesses. 
 
 III. 5-16 
 
E049 
 
 Approaches to the Utilization of Fish for the Preparation 
 
 of Protein Isolates 
 Enzymic Modifications of Myofibrillar Fish Protein 
 
 J. Spinelli, B. Koury, and R. Miller 
 
 ABSTRACT 
 
 It was concluded that the sensitivity of fish proteins to 
 physical and chemical alterations precluded their use as func- 
 tional ingredients in foods unless they could be modified so 
 that desirable functional characteristics could be preserved 
 during processing and subsequent storage. 
 
 The use of proteolytic enzymes to produce soluble protein 
 hydrolysates from fish proteins has been studied by several 
 groups. Cheftel et al. (1971) described the use of proteolytic 
 enzymes to solubilize fish protein concentrate. The use of raw 
 fish as a substrate for proteolytic enz>Tne hydrolysis was stud- 
 ied by Sen et al. (1962) and Srlpathy et al. (1964) . Some of 
 the problems associated with such procedures were the prolonged 
 times required for the complete solubilization of the proteins 
 and the development of bitter flavors due to the formation of 
 
 III. 5-17 
 
E050 
 
 Approaches to the Utilization of Fish 
 for the Preparation of Protein Isolates 
 
 Isolation and Properties of Myofibrillar 
 and Sarcoplasmic Fish Proteins 
 
 J. Spinelli, B. Koury, and Px.. Miller 
 
 ABSTRACT 
 
 Recently Meinke et al (1972) discussed some of the factors 
 that influence the production of protein isolates from fish. In 
 this work, data on the solubility characteristics and the recovery 
 of isolates from whole raw fish and frozen fish at various pH's, 
 salt concentrations and temperatures were presented. No attempts, 
 however, were made to separate the basic protein components of 
 fish muscle, and factors that influence the functional and organo- 
 leptic characteristics of isolates when subjected to various pro- 
 cessing procedures were not reported. The work dealt in these 
 latter areas and was divided into two parts: 
 
 1. To separate the two major protein fractions of fish 
 muscle (myofibrillar and sarcoplasmic) and to determine some of 
 the physical chemical and or'^anols'^tic '•^ro'^erties (^f pach before 
 and after drying. 
 
 2. To evaluate the use of some proteolytic enzymes as an 
 aid in modifying the physical and chemical characteristics of the 
 separated protein fractions. 
 
 111.5-1^ 
 
E053 
 
 Bench Level Storage Study (Fresh Menhaden in IPA) 
 
 D. Dubrow 
 
 ABSTRACT 
 
 Bench level experiments were conducted to examine the 
 effects of several levels of IPA on the storage stability of 
 fresh, raw menhaden, cooked menhaden, and cooked and pressed 
 menhaden for storage periods of up to sixty days. 
 
 This experiment consisted of two parts as follows: 
 
 (1) Unprocessed stored material was analyzed for aerobic/ 
 anaerobic bacteria, soluble nitrogen, and lipid oxida- 
 tion at one-week intervals for the first month, then 
 at 45 days, and at 60 days. 
 
 (2) FPC samples from three series of bench level extrac- 
 tions made at (control), 30 days, and 60 days were 
 analyzed for yield, organoleptic qualities, proteins, 
 lipids, ash and volatiles. Variables included 
 three different levels of IPA and raw, cooked, and 
 cooked and pressed menhaden. As the experiment pro- 
 gressed, IPA level variables decreased. 
 
 In addition, peroxide was added during extraction and com- 
 pared to samples with no peroxide content. From the results 
 reported, the following conclusions were drawn: 
 
 1. Raw fish stored with 40% or less alcohol, by weight of 
 the water content, did not produce a satisfactory FPC. 
 
 2. Raw fish could be stored at a level of 160% or 1:1 sol- 
 vent to wet solid ratio for 3 to 5 weeks and produced a satisfac- 
 tory FPC. 
 
 3. Raw fish could probably be stored at an 80% IPA level 
 (0.5 to 1 solvent to solid ratio) for a similar period of time. 
 
 4. Cooked fish, could be stored at a minimum level of IPA 
 of 40% (0.25 to 1 solvent to solid ratio) or 160% (21:1) for 3 
 to 5 weeks and produced a satisfactory FPC. 
 
 5. Cooked-pressed fish was not recommended for storage for 
 FPC at this time. 
 
 III. 5-19 
 
E054 
 
 Frozen Menhaden Storage Studies 
 
 J. Spinneli 
 
 ABSTRACT 
 
 The objective of this project was to identify at the 
 laboratory level critical constraints and/or problems associated 
 with the processing for FPC of frozen menhaden that have been 
 held in storage for periods of 30 and 60 days. 
 
 The results of 30-day studies, conducted at both bench 
 level and scaled-up unit, indicated that organoleptic qualities 
 and yields of FPC were acceptable. Sixty-day storage studies at 
 bench level indicated that a borderline acceptable product could 
 be made. 
 
 These studies had a twofold significance: 
 
 (1) Frozen menhaden could be used successfully for 
 
 (2) A storage period of up to sixty days for frozen 
 
 menhaden could increase substantially the operating 
 time for a commercial plant. 
 
 III. 5-20 
 
E055 
 
 Separating Bone Particles 
 From Whole Menhaden FPC 
 (Air-Flotation) 
 
 D. Dub row 
 
 ABSTRACT 
 
 The objective of this study was to determine the feasibility 
 of separating bone matter from whole menhaden FPC in order to 
 increase the total protein content. FPCs processed from fresh 
 fish and frozen fish were tested on a V135A Gravity Separator 
 Table. After the FPCs had been passed over this device, they were 
 analyzed to determine fractions of ash, protein, and volatile 
 content. Unmilled, coarsely milled, and finely milled FPCs were 
 tested, with the resulting determination that the coarsely milled 
 samples were best suited for this method of separation. The use 
 of the air flotation method appeared to be technically promising 
 for the removal of bone particles from FPC; it may not have been 
 economically feasible on a commercial scale. 
 
 III. 5-21 
 
E056 
 
 Preparation and Storage Properties of Presscake Made 
 by the Aqueous Phosphate Process for Use in FPC Manufacture 
 
 J. Spinelli 
 
 ABSTRACT 
 
 Several methods of preparing FPC have been specifically 
 investigated to offset the production costs that are related to 
 a fluctuating raw material supply. In reviewing the status and 
 suggested methods for preparing FPC, Finch (1970) cites six 
 methods where the starting material for preparing FPC is not 
 raw fish, but rather fish meal. Two methods use a cooked press - 
 cake instead of raw fish (Lahiry et al., 1972; Whaley and Moshy, 
 1966). 
 
 Although it has been adequately demonstrated that high 
 quality FPC can be prepared from presscake (Lawler, 1970; Spi- 
 nelli et al. , 1971), the storage characteristics of presscake 
 and quality of FPC prepared from presscake stored for protract- 
 ed periods of time has not been reported. 
 
 It was the purpose of this work to determine the process 
 parameters necessary to produce a stable presscake (made by the 
 aqueous phosphate process) that could be stored at ambient tem- 
 perature and remain suitable for the preparation of a high quali- 
 ty FPC. The specific objectives were as follows: 
 
 1. Evaluation of the conditions of preparation and the additives 
 necessary to prevent microbiological and oxidative deterioration 
 
 of the presscake during storage. 
 
 2. Evaluation of the organoleptic and nutritional properties 
 of FPC made from stored presscake. 
 
 III. 5-22 
 
Section F 
 
 P^OI^yCTIOI^ PROCESSES 
 
 III. 6-1 
 
FOOl 
 
 MIT-UNICEF Studies on the Production of 
 Fish Protein Concentrate for Human Consumption 
 
 E. R. Pariser and E. Odland 
 
 ABSTRACT 
 
 Among procedures suggested for the preparation of fish 
 protein concentrates, solvent defatting and solvent deodorizing 
 showed promising results and were further investigated. Process- 
 ing methods based upon these operations vary mainly in the 
 method of dehydration, choice of solvent or solvents, and condi- 
 tions of extraction. In an effort to clarify the effects of 
 these variables on the composition, protein quality, and taste 
 and smell characteristics of the final product, the United Nations 
 International Children's Emergency Fund (UNICEF) , in consultation 
 with the Food and Agriculture Organization (FAO) , arranged for 
 a study of those problems by the Department of Food Technology of 
 the Massachusetts Institute of Technology (MIT) , This study was 
 completed in 1961. The purpose of this paper was to summarize 
 the results. 
 
 This summary reported on the following: the raw material 
 selected; the processing methods and equipment used? and the 
 effect of the processing variables on the final product. 
 
 III. 6-2 
 
F002 
 
 Preliminary Process Design 
 
 and Cost Estimate for a 50 Ton Per Day 
 
 FPC Plant 
 
 Central Engineering Laboratories Food Machinery Corp. 
 
 ABSTRACT 
 
 The preliminary plant design and cost estimate for a plant 
 to convert 50 tons per day of a low oil content fish into fish 
 protein concentrate (FPC) was presented in this report. The 
 process used is extraction of fresh fish with an isopropyl 
 alcohol-water mixture to remove the oil and water. The insolu- 
 ble solids remaining consist of 80 to 85% protein. 
 
 The capital required to build a plant of this size was esti- 
 mated at 1.3 million dollars. This figure was arrived at by the 
 relatively standard method of estimating the cost of major pro- 
 cess equipment items and applying factors for such items as 
 installation, piping, utilities, buildings, etc. The accuracy 
 of this type estimate is about ±25%. Based on this capital in- 
 vestment, the product cost was estimated at about 26c/lb FPC, 
 when oTT^Tc^ tr1 nc ^50 dflvs/vear. 
 
 The amount of fish processed in a solvent extraction plant 
 such as this depended quite strongly on the oil content of 'the 
 fresh, whole fish. This design and cost estimate was based on 
 an oil content of 2.5%wt, typical for red hake caught off the 
 coast of Massachusetts. Fish with higher oil content will re- 
 quire a greater amount of solvent, and, as a result, each extractor 
 will contain less fish in each batch. Thus, for a fish such as 
 this, the capacity of the plant described in this report would 
 decrease and the operating costs per pound of FPC would increase 
 correspondingly . 
 
 This preliminary design and cost estimate indicated 
 the relative contribution of several areas of the proposed pro- 
 cess to the overall cost. The results permitted attention to be 
 focused on those areas which are likely to have the greatest 
 effect on the cost of producing FPC. 
 
 III. 6-3 
 
F009 
 
 FPC; The NMFS Experiment and Demonstration Plant Process 
 
 R. C. Ernst 
 
 ABSTRACT 
 
 The purpose and objectives of the National Marine Fisheries 
 Services semi-works FPC plant in Aberdeen, Washington, are de- 
 scribed. The design, construction and operation of the plant are 
 discussed and the results of operations to date presented. 
 These results are then discussed, conclusions drawn, and recom- 
 mendations are made for further plant modifications and operations 
 
 I 
 
 III. 6-4 
 
F0I8 
 
 Modification of EDP for Fatty Fish Processing 
 G. M. Marchello 
 
 ABSTRACT 
 
 This eight page memo was Dr. Marchello 's report on recom- 
 mendations for modifications to the EDP as a result of his chair- 
 ing a technical review and discussion meeting in Aberdeen, May 
 24-25, 1971, among Ocean Harvesters, Sweco, and NMFS personnel. 
 Attendees were listed; highlights, recommendations, diagram and 
 cost estimates were included. These recommendations were later 
 modified by NMFS and Ocean Harvesters. [Unofficial question 
 was "what can be done for $150,000".] 
 
 This meeting was for the purpose of technical review and 
 to recommend plant modifications for fatty fish processing. Prob- 
 lems and possible solutions were discussed. Four-stage extraction, 
 IPA losses, oil recovery, steam sparge to still, second oil cen- 
 trifuge addition, sludge/oil inversion, and vacuum evaporator were 
 all lueuLloned. Changes rcccmmcndcd were: 1) conversion of present 
 oil centrifuge to three-phase, 2) installation of film evaporator, 
 3) addition of acid to still or mono-sodium acid phosphate solu- 
 tion, and 4) installation of spare slurry pump and deboner parts. 
 Total estimate $140,500. [note: Recommendations were determined 
 reasonably valid at the time, but were later determined obsolete.] 
 
 III. 6-5 
 
F022 
 
 storage of Hake in Brine 
 Ocean Harvesters, Inc. 
 
 ABSTRACT 
 
 A one-week, short "loss" experiment was reported on storage 
 of hake in refrigerated brine at the EDP. Trend curves of data 
 indicated: protein - loss 0.30%/day, 0.39%/day; volatile matter - 
 loss 0.14%/day, gain 0.14%/day; fat - loss 0.16%/day, 0.21%/day; 
 ash - gain 0.34%/day, 0.39%/day from whole and broken fish respec- 
 tively stored in refrigerated brine. 
 
 1. A yield of 10 to 10.5% deboned hake-FPC (96% protein) 
 was determined realistic for good quality hake. 
 
 2. A yield of 8 to 9% was determined to be expected of 
 fish stored 3 or 4 days in brine and was achieved at Aberdeen. 
 
 3. Screen problems at Aberdeen were being intensified by 
 the poor quality fish. This was especially true cf extraction 
 stage one, because the IPA could not harden the fish protein 
 which had been altered by the brine and by autolysis - even at 
 SST. 
 
 III. 6-6 
 
F027 
 
 FPC Processes 
 R. Finch 
 
 ABSTRACT 
 
 The general and technical requirements for a satisfactory 
 process for making fish protein concentrate were investigated. 
 Solvent extraction processes, especially those using isopropyl 
 alcohol (IPA) , were examined most widely. Stages in the extrac- 
 tion of fish with IPA as a prototype method were reviewed. Vari- 
 ation in solvent processes, modifications to simplify them 
 and -to retain functionality in the end products, and non-solvent 
 methods, including enzymic digestion were considered. Future 
 developments were explored. 
 
 III. 6-7 
 
F02a 
 
 Bacteriological Aspects of Fish Protein Concentrate Production 
 from a Large-Scale Experiment in Demonstration Plant 
 
 S. L. Paskell and D. Goldmintz 
 
 ABSTRACT 
 
 A fish protein concentrate Experiment and Demonstra- 
 tion Plant capable of handling 100,000 pounds per day of fish 
 has been built by the National Marine Fisheries Service in Aber- 
 deen, Washington, Although FPC is now being produced which meets 
 microbiological specifications, several difficulties were en- 
 countered during start-up. In the production of FPC from this 
 plant microbiological studies included locating possible sources 
 of contamination by evaluating operational areas of fish-holding, 
 intra-plant transport of intermediate and dried product, air 
 quality, and equipment cleaning as well as analyses of product 
 -from various stages. 
 
 Food grade Pacific hake (Merluccius productus ) held in 5% 
 circulating brine at 0° to 2°C were used for FPC production. 
 While Flavobacterium , Achromobacter , Pseudomonas , and other or- 
 ganisms were introduced with fish into the brine, by 8 days when 
 spoilage was detected, as much as 80% of the psychrophilic popu- 
 lation were pseudomonads. Acceptable fish with counts up to 
 500,000 per gram yielded FPC with less than 300 per gram and a 
 few thermaduric Bacillus . S taphylccoccus aureus , enterics, and 
 Clostridia were occasionally found. When FPC with higher counts 
 was encountered, growth in residual solids within the conveyors 
 of the dry transport system was the usual cause. During process- 
 ing, material left the driers at 70°C with only 6% moisture. As 
 the product cooled, moisture increased due to condensation within 
 the system and drier malfunction. Temperatures were optimum for 
 mesophilic growth throughout much of the conveyor system. Con- 
 tamination was traced to difficulty in sanitizing conveyors 
 and removing residual product. Fallout, human contact and 
 Bacillus in the system provided the chief flora. By redesign 
 of offending equipment and emphasis on stringent sanitation, 
 the plant can operate with alleviation of microbiological 
 problems. 
 
 III. 6-8 
 
F033 
 
 EDP Discussion and Review of Plant Design and Operation Problems 
 Requiring Consideration for Modification 
 
 R. C. Ernst, Jr. 
 
 ABSTRACT 
 
 The purpose of the dissertation was to give an engineering con- 
 tractor additional understanding of the basic process, the problems with 
 the plant and suggested solutions. The EDP process was described with 
 the recommended changes. No specific dat?, sizes, models or rates were 
 given. This document preceded the P/E Development Company's report but 
 contained some preferred alternatives. 
 
 Frozen menhaden were tempered, removed from boxes, chopped and ground 
 as whole fish or optionally thawed in a heated screw and deboned with a 
 Bibun or Beehive separator. Screens and presses were replaced with tentri- 
 fugal decanters for four heated stages of extraction. Desolventizing was 
 modified. Air conveying was used from fourth stage to milling room. Re- 
 ducing noise level, sanitation, and storage of unmilled FPC, for analysis 
 prior to milling were emphasized. Other changes in the plant were to 
 be made to permit adequate control, sampling, and measurements for ob- 
 taining adequate design data for commercial plant design. 
 
 III. 6-9 
 
F035 
 
 Engineering Review Study of EDP 
 C. Linderman 
 
 (: 
 
 ABSTRACT 
 
 The objective of this study was to specify modifica- 
 tions for the EDP so that it would be capable of producing 
 specification grade FPC while running continuously (100 hours 
 or more) at design conditions and feed rates. Work included 
 a review of the FPC process as it now exists in the EDP, in- 
 vestigation of problems experienced with the present equipment, 
 processing and waste disposal problems experienced during the 
 hake run of 1971 and subsequent trials on menhaden and anchovy, 
 and finished product quality. Solutions to the existing prob- 
 lems were developed and associated costs were estimated. As 
 the result of the problem review, proposed solutions and cost 
 estimates, several alternate plans were added to the report. 
 These alternates include three waste treating systems, two 
 control room locations, two methods of providing additional 
 office space, and the effect on capital costs of leasing cer- 
 tain pieces of process equipment. 
 
 III. 6-10 
 
F037 
 
 Summary Report of the Hake Run 1971 (OHI-001) 
 Ocean Harvesters, Inc. 
 
 ABSTRA.CT 
 
 This summary report outlines the activities and accomplish- 
 ments of the 1971 hake operation at the National Marine Fisheries 
 Service (NMFS) fish protein concentrate (FPG) Experiment and 
 Demonstration Plant at Aberdeen, Washington. 
 
 III. 6-11 
 
F039 
 
 Fresh Anchovy Studies 
 J. Spinelli 
 
 ABSTRACT 
 
 Bench level studies assessed the effect df two methods of 
 preservation and storage time upon the processing of fresh an- 
 chovy into FPC. The scaled-up study simulated on a smaller 
 scale a hypothetical operation with the EDP. This was necessary 
 to enable design input for storage and handling facility modi- 
 fications which in all probability would be required at the EDP. 
 
 The studies completed in the foregoing experiments demonstra- 
 ted the feasibility of shipping commercial quantities of fresh 
 anchovy from San Pedro to the EDP for the purpose of manufactur- 
 ing FPC. 
 
 III. 6-12 
 
F050 
 
 Evaluation of a Dissolved Air Flotation 
 Separator for EDP Waste Water Treatment 
 
 A. McPhee and R. Ernst 
 
 ABSTRACT 
 
 The disposal of waste water used or produced during the 
 production of fish protein concentrate at the NMFS Experiment 
 and Demonstration Plant in Aberdeen, Washington, has been a 
 water pollution problem of severe magnitude. To combat this 
 problem, all plant effluent streams were routed through a 
 gravity skimming tank and then through a carborundum unit before 
 being discharged into the Chehalis River. An evaluation of 
 these water pollution control facilities and analysis of 
 individual waste water streams is presented. The carborundum 
 unit did not perform as well as originally expected. Typical 
 operating problems for the Carborundum Flotation Separator were 
 also defined as well as results to be expected for different 
 waste water sources. Secondary waste water treatment was also 
 dealt with briefly. 
 
 III. 6-13 
 
F054 
 
 Start Up Critique (OHl-002) 
 
 Ocean Harvesters, Inc. 
 
 ABSTRACT 
 
 This contract report covered the experiences, procedures, 
 problems, solutions, and modifications encountered during 
 start up operations with the Experiment and Demonstration 
 Plant at Aberdeen. 
 
 I 
 
 III. 6-14 
 
F055 
 
 Operations Manual (OHl-003) 
 
 Ocean Harvesters, Inc. 
 
 ABSTRACT 
 
 This contract report presented descriptions of the 
 process and equipment (with flow sheets attached) for the 
 Experiment and Demonstration Plant at Aberdeen. The 
 report included details of operator duties, checklists, 
 and procedures for start-up, operation, shut-down, emergen- 
 cies, and cleaning. 
 
 111.6-15 
 
F056 
 
 Lean Fish Operations (OHl-004) 
 
 Ocean Harvesters, Inc. 
 
 ABSTRACT 
 
 This contract report covered production, operation, 
 process, equipment, and product evaluation for lean fish 
 operations at the Experiment and Demonstration Plant at 
 Aberdeen. Flow sheets, material balances, equipment 
 specifications, analytical procedures, and recommendations 
 are included for the FPC processing of hake. 
 
 III. 6-16 
 

 F057 
 
 Fatty Fish Operations (OHl-005) 
 
 Ocean Hai^vesters, Inc. 
 
 ABSTRACT 
 
 This contract report reviewed the limited fatty fish 
 runs on frozen anchovy and menhaden which were made at the 
 Experiment and Demonstration Plant at Aberdeen. Production 
 descriptions, problems, operational modes, and analyses 
 were covered. 
 
 III. 6-17 
 
F059 
 
 Final Project Report (OHI-006) 
 
 Ocean Harvesters, Inc. 
 
 ABSTRACT 
 
 This contract report reviewed in entirety the contrac- 
 tor's role in the Experiment and Demonstration ^lant at 
 Aberdeen. Described and evaluated were research, design, 
 construction, modifications, and operations, including the 
 contractor's conclusions and recommendations. The report 
 included as appendices research reports, drawings, specifi- 
 cations, operating reports, analyses, and evaluation reports. 
 
 III. 6-18 
 
F060 
 
 Rapid Thawing and Deboning of Frozen Menhaden 
 for the Experimental and Demonstration Plant 
 
 J. Dyer and L. Lehman, Division Office 
 ABSTRACT 
 
 To develop a quick, economical method for thawing of frozen 
 menhaden for future operations at the Experiment and Demonstra- 
 tion Plant (EDP) by using an active-surface heat exchanger, sever- 
 al combinations of disintegrator, extructor and thermascrew were 
 used. 
 
 The feasibility of operating the EDP on frozen fish with 
 from 15 minutes to 1 hour of plant running time required to clear 
 the thawing system was proven. There are some improvements in 
 this thawing system still to be made and tested. These are: 
 
 1. The case around the disintegrator should be jacketed 
 with warm water and designed for more efficient discharge. This 
 should prevent the ground material from freezing solid and plug- 
 ging the space outside the screen. This blockage probably con- 
 tributed greatly to the over-reduction in particle size. The 
 frozen rish chips weie helJ in coutact with the hammers toe long. 
 
 2. The thermascrew should be a twin-screw, counter-rotating 
 unit with the flights intermeshed. This configuration would knead 
 the frozen agglomerates into complete annihilation resulting in 
 uniform and more complete thawing of the fish particles. 
 
 III. 6-19 
 
PhWf l¥ 
 
 SELECTED POeyR/'aEP^TATIOPI 
 
S@£ti© 
 
 A004 
 
 AOlO 
 
 AOll 
 
 A020 
 
 A021 
 
 A022 
 
 A023 
 
 A026 
 
 A028 
 
 A031 
 
 "Fish Protein Concentrate - A High Quality Animal Protein" 
 E. R. Pariser (NMFS) 
 
 "Fish Protein Concentrate" 
 (NMFS) 
 
 "Fish Protein Concentrate" 
 
 (NMFS) 
 
 "The Fish Protein Concentrate Story - 1. Enter: FPC" 
 
 D. G. Snyder (NMFS) 
 
 "The Fish Protein Concentrate Story - 2. The Deep Yarn of FPC 
 
 E. R. Pariser (NMFS). 3. pish Catch for FPC - W. M. Chapman 
 
 "The FPC Story - Here and There: Dragging the Net" - 4. World 
 Efforts Towards. FPC - G. M. Knobl, Jr. 
 
 "The FPC Story; BrinBing the Product Home - 5. Bureau of 
 Commercial Fisheries Program" 
 D. G. Snyder (NMFS) 
 
 "Fish Protein Concentrate" 
 
 NMFS, Hearings, House of Representatives Committee 
 
 "The U.S. Fish Protein Concentrate Program" 
 R. Finch (NMFS) 
 
 "The Fish Protein Concentrate Story - 10. U. S. Bureau of 
 Commercial Fisheries FPCs: Nutritional Quality and Use in Foods" 
 V. Sidwell, B. Stillings, G Knobl, Jr. (NMFS) 
 
 IV. 1-1 
 
SECTION A (Cont'd) 
 
 A038 FPC Program Executive Summary 
 
 (Northrop Services, Inc.) 
 
 A040 "Fish Protein Concentrate" 
 
 Hearings, House of Representatives Committee on Merchant 
 Marine and Fisheries 
 
 IV. 1-2 
 
Sectaooi 
 
 F[H^O0)5JCT CMA^ACTElF^iSTiC 
 
 B004 "Possible Interference of Fats, Carbohydrates, and Salts in 
 
 Amino Acid Determinations in Fish Meal, Fish Protein Concen- 
 trates and Mixed Animal Feeds" - Smith, Ambrose, Knobl (NMFS) 
 
 BOOS ''Nutritional ^valuation of Fish Protein Concentrate" 
 
 B. R. Stillings, (NMFS) 
 
 BOll The Chemi stry of Flavor Reversion in Fish Protein Concentrate 
 
 W. Wick (MIT) 
 
 B012 "Nature of the Residual Lipids in Fish Protein Concentrate (FPC)" 
 
 B. F. Medwadowski, J. Van Der Veen, H, S. Olcott (University 
 of California) 
 
 B013 "Volatile Constituents of Fish Protein Concentrate" 
 
 E. Wick, E. Underriner, E. Paneras (MIT) 
 
 B016 "Nature of Residual Lipids in Menhaden Fish Protein Concentrate" 
 
 B. Medwadoski, J. Van Der Veen, H. S. Olcott (University of 
 California) 
 
 BO 18 The Chemistry of Flavor Reversion in Fish Protein Concentrate 
 
 E. Wick (MIT) 
 
 B022 "Determination of Isopropyl Alcohol in Solid Fish Protein 
 
 Concentrate by Gas-Liquid Chroma togrphy" 
 P. Smith, Jr., N. L. Brown (NMFS) 
 
 B023 "Sequence of Limiting Amino Acids in Fish Protein Concentrate 
 
 Produced by Isopropyl Alcohol Extraction of Red Hake" 
 B. R. Stillings, 0. A. Hammerle,'D. G. Snyder (NMFS) 
 
 B024 Ch emical Evaluation of Solvent Purity Required fo r Fish 
 
 Protein Concentrate Production Using Recovered IPA Azeotrope 
 
 and the Chemistry of Flavor in Fish Protein Concentrate 
 E. Wick (MIT) 
 
 B027 "FPC's Quality Virtually the Same as Its Raw Material's Quality" 
 
 D. L. Dubrow, E. R. Pariser, N. L. Brown, H. Miller, Jr. (N>tFS) 
 
 B030 "Solubilization of Fish Protein Concentrate - 1. An Alkaline 
 
 Process: 
 S. R. Tannenbaum, M. Ahern (MIT), R. P. Bates (NMFS) 
 
 IV. 2-1 
 
SECTION B (Cont'd) 
 
 B032 "Effect of Heat on the Chemical and Nutritive Stability of 
 
 Fish Protein Concentrate (FPC)" 
 
 D. L. Dubrow, B. R. Stillings (NMFS) 
 
 B034 Lipid Composition of Fish Protein Concentrate 
 
 H. Olcott (University of California) 
 
 B036 "Adaptation of the AOAC Fluoride Method for Determining 
 
 Fluoride in Fish Protein Concentrate" 
 R. B. Klemm, M. E. Ambrose (NMFS) 
 
 B037 "Biological Availability of the Fluoride of Fish Protein 
 
 Concentrate in the Rat" - I. Zipkin, S.M. Zucas , B.R. Stillings 
 
 B040 "Fish Protein Concentrate; A New Source of Dietary Protein" 
 
 B. R. Stillings, G. M. Knobl, Jr. (NMFS) 
 
 B043 "Relative Utilization of Casein, Fish Protein Concentrate and 
 
 Isolated Soybean Protein for Growth and Pancreatic Enzyme 
 Regeneation of the Protein Calorie Malnourished Baby Pig" 
 W, G. Pond (Cornell Univ.), W. Snyder, J. T. Snook, E. F. 
 Walker, Jr., D. A. McNeill, and B. R. Stillings (NMFS) 
 
 B044 "Effect of Storage on Lipids of Fish Protein Concentrate" 
 
 B. Medwadoski, A. Hanley, J. Van Der Veen, H. S. Olcott 
 (University of California) 
 
 B047 "Enzymatic Solubilization of Fish Protein Concentrate: Batch 
 
 Studies Applicable to Continuous Enzyme Recycling Processes" 
 
 C. Cheftel, M. Ahern, D. I. C. Wang, S. R. Tannenbaum (MIT) 
 
 B048 "Moisture Adsorption of Fish Protein Concentrate at Various 
 
 Relative Humidities and Temperatures" 
 J. G. Rasekh, B. R. Stillings, D. L. Dubrow (NMFS) 
 
 B049 "Availability of Amino Acids in Sarcoplasmic Fish Proteins 
 
 Complexed with Sodium Exametaphosphate" 
 G. A. Pelroy, J. Spinelli (NMFS) 
 
 B050 "The Wholesomeness of Fish Protein Concentrate: A New Approach 
 
 to the Evaluation of Food Safety" 
 L. Friedman, 0. G. Glaser, N. L. Brown, E. R. Pariser (MIT) 
 
 B055 The Chemistry of Flavor in Fish Protein Concentrate and Chemical 
 
 Evaluations of Solvent Purity Required for FPC Production 
 
 E. Wick (MIT) 
 
 IV. 2-2 
 
SECTION B (Cont'd) 
 
 B056 "Effect of Hydrogen Peroxide on the Color, Composition and 
 
 Nutritive Quality of FPC (Fish Protein Concentrate)" 
 J. Rasekh, B. R. Stillings, V. Sidwell (NMFS) 
 
 B057 Effects of Processing Parameters on the Chemical and 
 
 Functional Characteristics of FPC 
 D. L. Dubrow (NMFS) 
 
 B061 Use of Fish Protein Concentrate in the Treatment of Kwashiorkor 
 
 and Its Protein Value in Pre-School Age Children 
 N..S. Scrimshaw (MIT) 
 
 B064 Nutritional and Safety Characteristics of FPC 
 
 B. R. Stillings (NMFS) 
 
 B069 The Use of Hydrogen Peroxide in FPC Extraction 
 
 R. Ernst (NMFS) 
 
 B071 "Reduction of Mercury with Cysteine in Comminuted Halibut 
 
 and Hake Fish Protein Concentrate" 
 J. Spinelli, M. Steinberg, R. Miller, A. Hall, L. Leman (NMFS) 
 
 IV. 2-3 
 
ti© 
 
 CT USE 
 
 C007 
 
 "FPC in Foods" 
 V. Sidwell (NMFS) 
 
 C015 
 C017 
 
 C020 
 
 C023 
 C024 
 
 C025 
 
 C027 
 C029 
 C031 
 
 "Fatty Fish Petition to FDA" 
 P. Roedel (NMFS) 
 
 "Quality and Nutritive Value of Pasta Made from Rice, Corn, 
 Soya, and Tapioca Enriched with Fish Protein Concentrate" 
 W. H. Kwee (University of Maryland), V. Sidwell, R. C. Wiley, 
 0. A. Hammerle (NMFS) 
 
 "Phosphate Complexes of Soluble Fish Proteins; Their Formation 
 
 and Possible Uses" 
 
 J. Spinelli, B. Koury (NMFS) 
 
 "Anchovy Petition to FDA" 
 P. Roedel (NMFS) 
 
 "Nutritive Quality of Wheat Flour and Bread Supplemented 
 
 with Either Fish Protein Concentrate or Lysine" 
 
 B. R. Stillings, V. D. Sidwell, 0. A. Hammerle (NMFS) 
 
 "Evidence Indicates that -- A Premix of FPC and Wheat Flour 
 
 Can Be Made and Transported" 
 
 V. D. Sidwell, B. R. Stillings, G. M. Knobl, Jr. -(NMFS) 
 
 "Increased Residual IPA Level Petition to FDA" 
 P. Roedel (NMFS) 
 
 "Petition for Approval of FPC in >fenufa ctured Foods" 
 J. Culbertson (National Fishmeal and Oil Association) 
 
 Utilization of FPC and Acceptability 
 V. D. Sidwell (NMFS) 
 
 IV. 3-1 
 
S®cti©Bi ED 
 
 irxIPySTKSAL ECO^JO^l'^iC ASPECT 
 
 D014 Utilization of Protein Ingredients in the U. S. Food Industry 
 
 Part I - The Current Market for Protein Ingredients. Part II 
 The Future l^Iarket for Protein Ingredients 
 P. M Hammonds, D. L. Call (Cornell U.) 
 
 D016 "Preparation of Fish Protein Concentrate and Fish Meal 
 
 John Spinelli (NMFS) 
 
 D021 Consumer Attitudes Toward Alternative Protein Ingredients 
 
 D. L. Call (Cornell U.) 
 
 D022 The Economics of Fish Protein Concentrate 
 
 J. Crutchfield and R. Deacon (U. of Washington) 
 
 IV. 4-1 
 
@cti©ii 
 
 
 ^) 
 
 E003 
 
 "Literature Study of Solvents for Fish Extraction' 
 
 J. A. Price 
 
 (Esso Research and Engineering Company) 
 
 E017 
 
 E022 
 
 E027 
 
 E030 
 
 E031 
 
 E032 
 
 E033 
 
 E038 
 
 E042 
 
 "Fish Fermentation". 
 
 L. Burkholder, P. R. Burkholder, A. Chu, N. Kostyl, and 
 
 0. A. Roels (Columbia University, N. Y.) 
 
 Bone Particulate Separation -- Fish Protein Concentrate 
 (Barkeley and Dexter Laboratories, Inc.) 
 
 Studies for the Purification of Isopropyl Alcohol 
 
 C . Cinrins ^F.sso Rpsparrh and F.ncn' np><=>-ri no n^>m^lan■^7^ 
 ». N -- -— - ---o o r yx 
 
 "Experimental Production of Fish Protein Concentrate 
 (FPC) From Mediterranean Sardines" 
 N. L. Brown, H. Miller (NMFS) 
 
 "Relative Activities of Commercially Available Enzymes 
 in the Hydrolysis of Fish Protein" 
 M. B. Hale (NMFS) 
 
 ''Holding Raw Fish (Red Hake) in Isopropyl Alcohol for FPC 
 (Fish Protein Concentrate) Production" 
 D. Dubrow, 0. Hammerle (NMFS) 
 
 Studies for the Purification of Isopropyl Alcohol 
 G. Ciprios (Esso Research and Engineering Company) 
 
 Use of Solvents in the Manufacture of Fish Protein Concentrate 
 N. L. Brown, E. R. Pariser (MIT) J. A. Price (Esso Research 
 and Engineering Company) 
 
 "Chemical and Nutritional Characteristics of Fish Protein 
 
 Concentrate Processed from Heated Whole Red Hake , ( Urophyc is 
 
 chuss )" 
 
 D. L. Dubrow; B. R. Stillings (NMFS) 
 
 IV. 5-1 
 
SECTION E (Cont'd) 
 
 E045 
 
 E047 
 
 E048 
 E051 
 E053 
 E05A 
 
 E055 
 
 E056 
 E059 
 
 "The Fish Protein Concentrate Story-13. Aqueous Phosphate 
 
 Processing" 
 
 J. Spinelli, J. Dyer, L. Lehman, D. Wieg (NMFS) 
 
 "Protein Autolysis Rates at Various pH's and Temperatures in 
 Hake, Merluccius productus , and Pacific Herring, Clupea 
 harengus pallasi, and Their Effect on Yield in the Prep- 
 aration of Fish Protein Concentrate" 
 B. Koury, J. Spinelli, D. Wieg (NMFS) 
 
 Making Fish Protein Concentrate by Enzymatic Hydrolysis 
 M. B. Hale (NMFS) 
 
 "Application of Ternary Equilibrium Data to the Production 
 of Fish Protein Concentrate"- McPhee, Dubrow, Henderson 
 
 Bench Level Storage Study (Fresh Menhaden. in IPA) 
 D. Dubrow (NMFS) 
 
 Frozen Menhaden Storage Studies 
 J. Spinelli (NMFS) 
 
 Separating Bone Particles from Whole Menhaden FPC (Air- 
 Flotation) 
 D. Dubrow (NMFS) 
 
 Preparation and Storage Properties of Press 
 
 Aqueous Phosphate Process for Use in FPC Manufacture 
 
 J. Spinelli (NMFS) 
 
 Frozen Menhaden Storage 
 R. Ernst (NMFS) 
 
 II 
 
 IV. 5-2 
 
Sectaon F 
 
 RODOCTiOi^ PROCESSES 
 
 F009 FPC : The KMFS Experiment and Demonstration Plant Process 
 
 R. C. Ernst, Jr. (NMFS) 
 
 F015 EDP Evaluation Guidelines 
 
 (NMFS) 
 
 F021 Power Consumption in Solid-Liquid Slurries 
 
 A. D. McPhee, N. L. Brown (NMFS) 
 
 F035 Engineering Review Study of EDP 
 
 C. Linderman (PEDCO) 
 
 F037 Summary Report of the Hake Run 1971 (OHI-001) 
 
 (Ocean Harvesters, Inc.) 
 
 F039 Fresh Anchow Storape Studies 
 
 J. Spinelli (NMFS) 
 
 F04A Drying Desolventization Study 
 
 J. Dyer (NT^FS) 
 
 F0A6 Decanter Selection 
 
 J. Dyer (NMFS) 
 
 F048 Distillation Alternatives 
 
 R. Ernst (NllFS) 
 
 F050 Evaluation of a Dissolved Air Flotation Separator for FPC 
 
 Effluent Water Treatment 
 A. D. McPhee, R. C. Ernst, Jr. (NT^FS) 
 
 F052 Rapid Analytical Procedure Studies (Bacteriological) 
 
 S. Paskell (N"MFS) 
 
 F053 Seasonal Composition (Menhaden) Variations 
 
 D. Pubrow (NT-IFS) 
 
 F058 Review of Engineering Review Study 
 
 J. Dyer, R. Ernst (NMFS) 
 
 F059 Final Project Report (OH 1-006) 
 
 (Ocean Harvesters, Inc.) 
 
 IV. 6-1