LIBRARY 
 
 ANNEX 
 
 ^ jttv *<^ jk|v i»^ jrj% ^^Tv vjv vy* -^ vjv i»^^ *^ « 
 
 COHESIVE CONSTRUCTION 
 
 GUASTAVINO 
 
 sl» ^t^ «A* 'vi^ 'vl" ^>» 'vt^ ^>' ""^t* 'st^ ^t^ *^^ ^X^ ^*t^ *X^ ^^ ^ 
 

 BOUGHT WITH THE INCOME, 
 FROM THE 
 
 SAGE ENDOWMENT FUND 
 
 THE GIFT OP 
 
 Henirg W. Sage 
 
 1S91 
 
 /^.//gJ/., 
 
 ^^. 
 
 ^z"- 
 
G* 
 
 2j0i 
 
 'h 
 
 "ate due 
 
 i^ypfi t9e?Hi^ 
 
 * 
 
 AHK I M M'*** 
 
 PRINTED IN U.S.A. 
 
 Cornell University Library 
 TA 670.G91 1893 
 
 Essajr on the theory and history of cohes 
 3 1924 022 866 440 
 
Cornell University 
 Library 
 
 The original of tiiis book is in 
 the Cornell University Library. 
 
 There are no known copyright restrictions in 
 the United States on the use of the text. 
 
 http://www.archive.org/details/cu31924022866440 
 
ESSAY 
 
 ON 
 
 THE THEORY AND HISTORY 
 
 OF 
 
 Cohesive Construction, 
 
 UPPLIED ESPEGIilLLY to the TIMBREL VilDLT. 
 
 READ BEFORE THE SOCIETY OF ARTS, 
 
 MASSACHUSETTS INSTITUTE OF TECHNOLOGY, 
 
 BOSTON, 
 
 BY 
 
 R. GUASTAVINO, Architect. 
 
 SECOND EDITION. 
 
 BOSTON 
 
 TICKNOR AND COMPANY 
 211 Tremont Street 
 
 1893 
 
^. I (ro^l^ 
 
 Copyright, 1892, 
 By R, Guastavino. 
 
 No -^1^- 
 
 . J. PARKHILL &. CO., PRINTERS, 226 FRANKLIt 
 BOSTON, U.S.A. 
 
DEDICATED 
 
 '' TO THE 
 
 SOCIETY OF ARTS, 
 
 MASSACHUSETTS INSTITUTE OF TECHNOLOGY, 
 
 BOSTON. 
 
CONTENTS. 
 
 I. Introduction ... . . ii 
 
 II. Reskarches and History . , 21 
 
 III. Theory and Coefficients, of Application . 45 
 
 IV. Modern Applications and /Esthetic Impor- 
 
 tance of the Cohesive Construction . 94 
 
PREFACE. 
 
 In October, 1889, at the request of the eminent 
 and well-known architect, Mr. T. M. Clark, the 
 Honorable Society of Arts of Boston paid me the 
 distinguished compliment of inviting me to describe 
 what observation and experience had taught me in 
 regard to the system of arches which at that time I 
 was constructing in the new Public Library of that 
 city, in Copley Square. 
 
 My easiest plan, and the one which I should have 
 followed, perhaps, to avoid the charge of preten- 
 tiousness, would have been to simply explain the 
 practical work which comes in my line of business, 
 and not tread on slippery ground by any analysis 
 and theory. But two reasons obliged me to over- 
 look this, and to enter into theory also. 
 
 The first, and most important reason was, that to 
 build an arch on this system, or build anything on 
 the Cohesive System, presents two problems. One 
 relates to the stability of the structure after being 
 built ; the other, and the main one, to getting the 
 structure built. We may know that a construction 
 on the " Cohesive System " will have stability when 
 set ; but to build it may be an insuperable problem. 
 
PREFACE. 
 
 It is not only to build a wall or arch over a solid 
 centre, where one brick is set over another brick, 
 one stone over another stone, and where, any defect 
 being found, it is a simple thing to correct or change 
 it. For a structure, generally in open space, acting 
 always in inclined or horizontal pressure and thrusts 
 — a structure where the main strength will depend 
 on the cohesion, and which, at the moment of con- 
 struction, has not this cohesion — presents a serious 
 problem. Consequently it was necessary for me to 
 study the problem carefully. This was not so easy, 
 especially when academies and books could teach 
 but little, if anything, on this subject. 
 
 The second reason was, that I remembered that, 
 before my fifteen years of practice as an architect, 
 
 1 passed several years at the students' bench in the 
 institutes, universities and academies, and^my emi- 
 nent professors would have the right to see whether 
 1 had left dust in my books and notes ; and I re- 
 flected also that one cannot explain practical 
 things, or have convictions, without some reasons, 
 and those reasons formed the scientific theory which 
 I was obliged to construct in order to have convic- 
 tions, and also some guaranty that my employes 
 could work safely. Scientific tools were, therefore, 
 necessary, and it was also necessary to have cour- 
 age to confront the difficult situation, and endeavor 
 to explain, in my limited way, the theoretical part. 
 Perhaps these tools are rough, as is usually the' case 
 with new things, yet the perfecting of them is the 
 
PREFACE. 9 
 
 specialty of the professors, to whom I refer this 
 essay for its completion, . . . But whatever knowl- 
 edge I may possess on this subject is due, not 
 so much to my researches and investigations, as to 
 the wisdom of my distinguished professors at the 
 Academy of Barcelona, D. Juan Torras and D. Elias 
 Rogent, who instructed and interested me in the 
 study of the arts and applied sciences, calling special 
 attention to this system of construction in embryo, 
 for which I treasure their memory with gratitude. 
 
 Finally, the disinterested assistance which I have 
 received from the most distinguished architects, due 
 principally to their ready comprehension of the fact 
 that the new construction is the renaissance of an 
 old and noble system, for several centuries unused, 
 but applied now owing to the necessities of the age, 
 has encouraged me to publish in book form the 
 lectures referred to, in the hope that it will have the 
 approval of all interested in the constructive arts. 
 
 R. G. 
 
PART 1. 
 
 INTRODUCTION. 
 
 In Barcelona, there is a family called 
 Muntadas, who are considered genuine repre- 
 sentatives of the aristocracy of the manufac- 
 turers of Catalonia. All the members of this 
 family are, or have been, manufacturers, and, 
 together, in the towns of Barcelona, St. Martin, 
 Sans, Gerona and Ripoll, they employ more 
 than ten thousand people in their bleacheries, 
 manufactories, dyeing and jDrinting buildings. 
 One member of this family owns, in the depart- 
 ment of Zaragoza, a rich and extensive property 
 that for centuries was possessed by the monks, 
 and is called "Monasterio de Piedra" (the Stone 
 Monastery). This land contains about 50,000 
 acres, and the buildings thereon, consisting of 
 churches, convents and the palace of the 
 abbot, of different epochs of Romanesque, 
 Byzantine, Renaissance and modern architect- 
 ure, cover about 200,000 square feet of ground. 
 The owner, Don Frederico Muntadas (who is a 
 great litterateur and pisciculturist), lives there 
 
12 COHESIVE CONSTRUCTION. 
 
 with his family a large part of the year. I was 
 invited by this gentleman, through his uncles 
 Don Jose and Don Ignacio Muntadas, to visit 
 this property, as they intended to convert the 
 immense convent into a summer resort. 
 
 It was in October, 1871, when I made my 
 acquaintance with this estate, which is four 
 miles from the railroad station of Alhama, 
 Aragon, a noted hot-spring resort. 
 
 Here, in that "Monasterio de Piedra," I saw a 
 grotto of immense grandeur, one of the most sub- 
 lime and extraordinary works of nature. Imagine 
 Trinity Church, Boston, covered by an immense 
 natural vault, supported by walls of the same 
 nature, with gigantic stalactites of all kinds of 
 forms and dimensions, like great chandeliers, 
 hanging from above ; the floor a lake, receiving 
 the whole light through an immense ventinel or 
 opening, like a rosette window in a cathedral, 
 covered by the fall of the full mass of water 
 of the river Jalon, its builder, passes over the 
 vault and precipitated more than two hundred 
 feet, taking the form in its fall of a horse's 
 tail, which is the source of its name, " Cola de 
 Caballo." 
 
 I had just left Barcelona, after completing 
 some buildings, among them the large manu- 
 factories of Batllo Brothers, and was under the 
 
COHESIVE CONSTRUCTION. 13 
 
 impression that I liad accomplished something, 
 in these buildings, according to the Cohesive 
 System ; but with this great specimen of nature's 
 architecture before me, I recognized how small 
 and insignificant my work had been. 
 
 The thought entered my mind, while in this 
 immense room, viewing this fall of 'water, that 
 all this colossal space was covered by a single 
 piece, forming a solid mass of walls, foundation 
 and roof, and was constructed with no centres 
 or scaffolding, and especially, without the ne- 
 cessity of carrying pieces of heavy stone, and 
 heavy girders or heavy centres ; all being made 
 of particles set one over the other, as nature 
 had laid them. From that time I became con- 
 vinced that there was much to learn from the 
 immense book called Nature, never enough 
 studied, and that our ordinary system of con- 
 struction was very poor, notwithstanding that we 
 possessed the material for this kind of building 
 that enabled us to imitate nature. Hence I un- 
 derstood why ray distinguished professor of con- 
 struction, D. Juan Terras, said one day : " The 
 architect of the future will construct by imitating 
 nature, because it is the most rational, durable 
 and economical method." This grotto is really 
 a colossal specimen of cohesive construction. 
 Why had we not built on this system ? 
 
14 COHESIVE CONSTRUCTION. 
 
 I knew that in the south of Europe, and in 
 Asia, there existed many buildings on the 
 Cohesive System, erected centuries ago. The 
 types are some of the Roman triumphal arches, 
 the Pantheon, the cupola of the Santa Sophia, 
 the cupola of the Cathedral of Zamora, and 
 others in Asia, and some in Moorish construc- 
 tions of the Middle Ages. The larger part of 
 these constructions are of concrete ; that is, 
 some are built with marble facing or other stone 
 outside and concrete on the interior, while others 
 are concrete throughout. The first attempts 
 made in my enthusiasm for the Cohesive System 
 were carried out in simple concrete. But I soon 
 found that no arch work could be done with 
 concrete — that is, cement combined with broken 
 stone, gravel or sand, to satisfy the needs of 
 the epoch — so well as it could be accomplished 
 with tiles. By this I mean tiles laid in cement, 
 if the material and process are well adjusted. 
 In consequence of this experience the factories 
 of Batllo were projected for tile arches and not 
 for concrete (1869 and 1870). The question, 
 therefore, arose in my mind, which would be the 
 better system — that of the Cohesive Construc- 
 tion based on concrete material, or the tile 
 system, like that used in the floors and ceilings 
 of the Batllo buildings .' 
 
COHESIVE CONSTRUCTION. 15 
 
 Such concreting, in truth, is the imitation of 
 the conglomerates of nature, but, as in nature, 
 requiring great mass and, as a principal factor, 
 TIME, in order to have strength. This process, 
 however, I found too heavy and too slow for 
 this epoch, in which, of course, we appreciate 
 the value of time. But the tile system, on the 
 other hand, was not the desirable one, on 
 account of the excess of plaster required ; 
 besides, there was the irregularity of the Port- 
 land cement of the market to contend with. 
 These two difficulties were rhore than an 
 obstacle, they were an imperfection. 
 
 Under these impressions and conditions I 
 commenced work in Barcelona, beginning with 
 my own private residence, on the corner of 
 Aragon and Lauria Streets. I tried the first 
 experiment on myself, as a physician might 
 try his own medicine, carrying out my ideas by 
 building a construction four stories in height, 
 practically with no beams, using clay and 
 cement. Afterwards I built the bleacheries 
 of Muntadas Aparicio & Co. ; a merino and 
 other woolen-goods factory for Carreras & Sons ; 
 another, also a merino factory, in Villa Franca, 
 for Michans & Co. ; the glass manufactory of 
 Modesto Cossademunt ; the Theatre of Vilsar ; 
 the manufactory of porcelain for Florens & Co. ; 
 
16 COHESIVE CONSTRUCTION. 
 
 and the silk manufactory of Saladriguez. I also 
 made some applications of the system in the 
 private houses of the bankers, D. Victor Blajot 
 and D. A. Anglada, and others. 
 
 But all this work was almost empirical. It 
 had not the right technical sanction, and how 
 was it possible to have it .'' The thickness of 
 the arches was determined by intuition and 
 practice, as a competent blacksmith determines 
 the size of the pieces which he uses, or a com- 
 petent sailor the size of a rope or block. But 
 can he satisfy the sciences with these.'' Can 
 we have any guarantee by these alone ? On the 
 other hand, was it possible to determine any- 
 thing in the embryo state of the manufacture 
 of cements fifteen or twenty years ago, when 
 no manufacturer was able to guarantee his own 
 brand, because of the difficulty in obtaining 
 regularity in it ; when each manufacturer had 
 his own formula, and when, in consequence, the 
 market could not be supplied with a sufficient 
 quantity of each brand to make an average test 
 of strength upon which to base any calculations ; 
 and where there was no certainty of always be- 
 ing able to procure in any market the quality of 
 cement upon which our coefficient was based.' 
 
 That was the condition of affairs in Spain, 
 and in fact, all Europe, fifteen or twenty years 
 
COHESIVE CONSTRUCTION. 17 
 
 ago, and that is still the condition in several 
 countries. Fortunately, the plans of the build- 
 ings and factories referred to were sent to the 
 Philadelphia Centennial, in 1876. The success 
 attained there, and the great Chicago fire, which 
 made an impression on all European minds, con- 
 vinced me that this country was the proper place 
 for the development of the Cohesive System. 
 But I did not succeed in coming here until 1881. 
 I had not been here long before I recognized 
 the necessity of studying American methods, 
 materials and facilities. To this work I devoted 
 five years. It was absolutely essential that I 
 should be well posted, particularly in the matter 
 of the timbrel arches : First, because cement is 
 the essential part ; second, because of the posi- 
 tion of the arches, as failure on their part must 
 endanger the lives of workmen ; third, because 
 the application of the arches being for floors, 
 and requiring speedy work, required also that the 
 floors should in a short time be delivered over for 
 use, and in consequence it was necessary for me 
 to know exactly with what kind of material I 
 was going to work, and under what conditions.* 
 
 *Yet, at the present time, notwithstanding the progress in 
 the manufacture of Portland Cement and the great stock in the 
 market, we must use every precaution as to the quality of the 
 cement we are using. 
 
18 COHESIVE CONSTRUCTION. 
 
 Explanations were given to interest promi- 
 nent architects and builders. But some seemed 
 to take the matter as a dream, or as though I 
 were a visionary ; while others, more benevolent, 
 said it might be beneficial in Spain or Italy, but 
 never in this country, so different in climate, 
 processes and necessities.* 
 
 On the other hand, the manufacture of tile 
 here was almost an impossibility; because, if 
 it was accomplished by hand-work it would be 
 very dear, and if by machinery, the probabili- 
 ties were that it would come out too heavy 
 and useless. Consequently the obstacles and 
 difficulties seemed insuperable, and hope almost 
 left me. 
 
 Fortunately, work and perseverance are two 
 great factors towards success. 
 
 The publication of some artistic works in 
 illustrated papers were received with apprecia- 
 tion, and some successful competitions for semi- 
 public buildings in New York put me in posi- 
 
 *NoTE. — Surely the latter were not aware that in Spain 95 
 per cent, of the architects and 99 per cent, of the builders did 
 not know or may not have heard anything about the system, 
 that the same was apparently true in Italy, and that the fire- 
 proof floors general in both these countries in important 
 buildings were in use long before the flat hollow brick arches 
 and iron beams also used here, tile arches only being used in 
 some states in small arches for common and cheap constructions. 
 
COHESIVE CONSTRUCTION. 19 
 
 tion to begin, with some authority, a series of 
 tests and experiments with imported tiles. 
 
 After ending my connections with my clients, 
 my first work done with this fireproof system in 
 America, was in a four-story private house on 
 78th Street, New York, in 1886, with American 
 tiles. During the same year I commenced to 
 build the interior of the Arion Club, 59th Street, 
 whose building committee accepted my proposi- 
 tion when they ascertained that with my arches 
 they could make a saving of over $5,000, in two 
 floors alone, over the ordinary system of fire- 
 proofing. 
 
 From that time on I have been building in 
 New York, having erected floor arches in sev- 
 eral different structures, among them being the 
 residence of W. Fellows, Esq., in Montclair ; the 
 Corbin Building, corner of John Street and 
 Broadway; the Edison Electric Illuminating 
 Company's Station, 26th and 29th Street, New 
 York, etc. 
 
 Because of these encouraging results, indi- 
 cated in the number of applications for con- 
 tracts, in July, 1889, I put all my affairs in the 
 hands of a corporation, calling itself the Guas- 
 tavino Fireproof Construction Company. But, 
 if the system has become popular, it does not 
 owe its popularity to its name, but to its adjust- 
 
20 COHESIVE CONSTRUCTION. 
 
 ability and its own merits. I have now tlie 
 satisfaction of seeing a system of construction 
 established on a good and satisfactory basis, 
 which only four years ago was considered a 
 dream, and which two years afterwards was 
 noticed in a prominent technical book as a sim- 
 ple curiosity only. It is a great satisfaction to 
 me to be able to say that all the great obstacles 
 which confronted me in my work have at last 
 been overcome. 
 
PART II. 
 
 RESEARCHES AND HISTORY. 
 
 (i) The "Timbrel Arch" is not entirely new. 
 It is as ancient as the " Cohesive System," and 
 may be as old as its opposite, which may be 
 called the "Gravity System." But although 
 the " Cohesive System," including the applica- 
 tion of timbrel arches, was frequently practised 
 by the ancients, after reaching the height of 
 its splendor in the Middle Ages it gradually 
 disappeared, in proportion as modern civilization 
 and the Renaissance approached. 
 
 (2) Was its disappearance due to the fact 
 that after this great constructive age the archi- 
 tects were not builders .■• Or was the disappear- 
 ance of this form of construction in Europe 
 caused by the decadence of the influence of 
 Oriental Architecture, after the great classic 
 era of the Arabs, or rather the Moorish-Spanish 
 Architects, who knew how to decorate con- 
 struction and to construct decoration in the 
 "Cohesive System," as did the Greeks, centuries 
 
22 COHESIVE CONSTRUCTION. 
 
 before, and the Graeco-Romans, in their style? 
 This is exemplified in their system of construc- 
 tion by gravity. 
 
 (3) We have no definite knowledge on these 
 points, but without the mighty fact of the 
 existence of some monuments in the " Cohesive 
 System," which testify to epochs of undoubted 
 progress in constructive art, it would be impos- 
 sible to realize or believe in them. 
 
 Much has been said of late against vaults, and 
 especially "Timbrel Arches" ; in the first place, 
 against their utility and application, and sec- 
 ondly, regarding their origin and ancient use. 
 The most erroneous and contradictory ideas 
 have been emitted in regard to this vaulting, as 
 before occurred with the arch itself, the latter 
 having been credited to the Romans. 
 
 (4) To-day, all that are known are studied b}' 
 exact designs from a great number of antique 
 monuments, some extant and others in ruins ; 
 and we can from these draw the truth and 
 infer history. We can thus say that the general 
 use of vaults of brick, stone or the timbrel, as 
 well as the use of the arch itself, and its origin, 
 is very ancient. They were applied before the 
 days of the Romans, who did nothing but im- 
 prove, making their use general, and giving to 
 them more or less an aesthetic character, which 
 
COHESIVE CONSTRUCTION. 
 
 23 
 
 had not been done before ; because the arch 
 and the vault had hitherto been used solely as 
 a constructive necessity where blocks large 
 enough to cover the space could not be procured. 
 For this reason the arch in the " Gravity Sys- 
 tem" appeared, as well as the "Timbrel Arch" 
 in the " Cohesive Construction." 
 
 No. I. 
 
 EGYPT. 
 
 (5) In a tomb situated in the vicinity of the 
 city known by the name of the City of Sepul- 
 chres, near Thebes, there is an elliptical vault 
 constructed of unburned brick. It is 2.50 m. 
 
24 COHESIVE CONSTRUCTION. 
 
 in length, by 1.42 m. in height, measured from 
 the springing. Among the hieroglyphics which 
 adorn this monument the name of Amenophis 
 can be discerned. It must thus belong to the 
 time of the eighteenth dynasty, dating some 
 seventeen centuries before our era. This is in 
 regard to the brick vault in general. Another 
 specimen in regard to the " Timbrel Vault" — 
 
 (6) In one of the Pyramids of Egypt, at Gizeh, 
 a tomb discovered by Colonel Campbell (see 
 Fig. No. i) forms an arch of unburned bricks. 
 These bricks measure 0.170 m. by 0.126 m. by 
 0.50 m. In order to give them the necessary 
 curve it is understood that they had to be 
 curved before being dried. The construction 
 plainly shows that the flat brick was used with 
 the idea of decreasing the number of pieces, 
 closing the space with the least possible joints. 
 Thus, to give more strength and cohesion to 
 the arch, they are applied in four rows, one 
 above the other, breaking the joints, constituting 
 through this medium an arch without joints. 
 (See " General .History of Architecture," by D. 
 Daniels.) 
 
 (7) It is seen by this specimen that the cohe- 
 sive form, as well as the typical timbrel arch, 
 and the arch in general, was, so to speak, "born," 
 and is not any particular invention. Nor did 
 
COHESIVE CONSTRUCTION. 25 
 
 it originate by any determinate civilization. It 
 was simply the fruit of necessity, a spontaneous 
 resource of the most ancient times. 
 
 (8) This plainly shows that neither brick 
 vaults, stone vaults nor timbrel vaults can be 
 said to belong to any civilization. Similar cir- 
 cumstances necessitated their creation in every 
 country. 
 
 ASSYRIA. 
 
 (g) The Assyrians improved the manufacture 
 of brick. Encamped between the rivers 
 Tigris and Euphrates, and with abundance of 
 clay at their disposal, as well as asphalts and 
 mineral oils, which they used as fuel, they came 
 to the practical idea of burning the clay, and 
 instead of using raw brick, they used burnt. 
 For such purposes ovens were needed; hence 
 the necessity for covering and closing space 
 without lumber or stone, but with bricks and 
 terra-cotta. Thus were the dome and cone 
 shapes developed that they were using in their 
 ovens. 
 
 (lo) The ovens for the manufacture of brick 
 were large domes constructed with bricks or 
 tiles of large dimensions. In some of their 
 experiments the bricks were laid flat, advanc- 
 ing one over the other (Fig. 2), each layer pro- 
 
26 
 
 COHESIVE CONSTRUCTION. 
 
 jecting about an inch, and in this manner form- 
 ing a curve. (See " General History of Archi- 
 tecture," by D. Daniels, already mentioned.) 
 
 r , I 
 
 t . I 
 
 S^ 
 
 ^ 
 
 I. I 
 
 :^ 
 
 Fig. 2. 
 
 (ii) The Assyrians attached great impor- 
 tance to the surface of their bricks, possibly to 
 give them conditions of cohesive strength. For 
 instance, the dimensions of the bricks used in 
 the library of the palace of Khorsabad, and in 
 the palace of Nimrod, were 35 by 32 by 7 centi- 
 metres, or about 14 by 12 by 2 inches. 
 
 The gardens of Semiramis, at Babylon, and 
 the subterranean passages under the Euphrates, 
 were nothing else but vaults built with very 
 large bricks. 
 
 GREECE AND ROME. 
 
 (12) The Greeks and Romans did not use the 
 brick in a better manner than the Assyrians. 
 
COHESIVE CONSTRUCTION. 27 
 
 Their facilities for obtaining clay and fuel were 
 not favorable, and they were therefore more 
 devoted to stone construction. The Romans, 
 in particular, had a marked predilection for stone 
 and concrete, of which they made very good 
 use, not only in triumphal arches and bridges, 
 but also in military and urban constructions, 
 such as sewers, etc. This is illustrated by the 
 sewers which they left in Valencia, Spain, simi- 
 lar to those in Rome, although clay was plen- 
 tiful in that country. [Through these sewers 
 of Valencia a wagon can easily pass. J The 
 aqueduct of Segovia and the city walls of Tar- 
 ragona are other specimens, again showing their 
 ■ predilection for stone and concrete. The former 
 is a wonderfully magnificent structure and a 
 model of static equilibrium. 
 
 (13) We may remark that when the Romans 
 used brick it was generally as a small voussoir, 
 as may be seen in the Flavian amphitheatre, or 
 the Coliseum ; not only in the primitive works 
 in this building, but when rebuilt at different 
 periods, they were used as voussoirs in plain 
 brick arches. 
 
 (14) The only specimens which seem to 
 have existed as vault work, or timbrel, of brick 
 placed, flat in imitation of the specimen shown 
 on page 23 (No. i, Egyptian), are some that 
 
28 COHESIVE CONSTRUCTION. 
 
 were probably in the baths of Antonius Cara- 
 calla. The architect found difficulty in furnish- 
 ing light to the central part, which could only 
 receive it through penetrations in the vault. 
 But for this purpose it was necessary to weaken 
 the arches, and if they were constructed of 
 brick, like those of the Coliseum and others, 
 they had to be given a great thickness and re- 
 quired walls of immense resistance. It seems 
 that the result desired was at last obtained by 
 constructing the vaults with bricks on end, 
 or a timbrel arch, using, I suppose, puzzolana 
 (Pozzuoli) cements, which were slow setting 
 but good, and using, may be, centres which 
 supported the vaults until after the mortar had 
 settled. But of this we cannot be sure, as may 
 be inferred from the following paragraph, taken 
 from the treatise on "Vaults and Bridges," by 
 Samuel Ware. 
 
 (15) " The recollection of the Solar Bath of 
 Antonius Caracalla in the present age, when we 
 assume to ourselves so much credit for the 
 invention of iron bridges, may serve to abate 
 some of our enthusiasm. It was a circular 
 building iii feet in diameter, the roof a dome, 
 composed of copper and brass." 
 
 (16) Spartian says of it: "Reliquit thermas 
 nominis sui eximias quarum cellam solearem 
 
COHESIVE CONSTRUCTION. 29 
 
 archtecti negant posse ulla imitatione qua facta 
 est fieri, namex acre vel cupro, cancelli super- 
 positi esse dicuntur, quibus concameratio tota 
 concredita est ; et tantum est spatii ut idipsum 
 fieri negent potuisse docti mechanici." 
 
 (17) By the foregoing it would appear that 
 "cancelli" were ribs and the "concameratio" 
 
 plates similar to what may be seen in our iron 
 bridges. From this historic description cited 
 by Samuel Ware, it follows that, if the small 
 domes and arches were constructed with " Tim- 
 brel Arches," the large dome was certainly not 
 built in the same way. 
 
 THE MIDDLE AGES. 
 
 (18) The true epoch of the development of the 
 "Cohesive System" and the dome was in the 
 Middle Ages, but no important specimens of 
 the "Timbrel Vault," or with the brick set flat 
 against the centre, are left. We must, however, 
 for several reasons, call attention to the con- 
 struction of arches and domes in the Arabian 
 epoch, and that of the Mussulman in Persia, 
 a country where a new and powerful civilization 
 was already developed, on the spot where the 
 Assyrian left the trail of his ceramic work — ■ 
 a civilization that is dying out under the vast 
 cupola of St. Sophia. 
 
30 COHESIVE CONSTRUCTION. 
 
 (19) The cupola was the dominant line of 
 their monuments (see Coste, Architect, 1840 
 and 1841, "Voyages en Perse "); and as the Ori- 
 ental civilization had great influence in the 
 antique Byzantium, not only did it give to the 
 Byzantines the richness of their colors and 
 decorations, but it gave also the foundation 
 for new ideas in the architectural arts ; to such 
 an extent, that it founded the classical examples 
 of the " Cohesive System." 
 
 (20) The greatest development was in Cordova 
 and Granada, Spain ; but under the influence of 
 the beginning of this civilization the construc- 
 tion of St. Sophia, the grandest and most 
 finished model of the cohesive cupola, was 
 carried out. The cupolas of Persia are all 
 constructed over brick walls, and are the con- 
 tinuation of the same wall with the same 
 material. 
 
 (21) From the building of the cupola of St. 
 Sophia to the period of the Renaissance several 
 cupolas on the cohesive principle were con- 
 structed. The principal of these cupolas were 
 the Mosques of Solyman 11. , Sultan Ahmet, 
 and the Holy Apostles, of Constantinople ; 
 
 •Santa Maria of Ravenna; St. Mark, Venice, 
 and the Cathedral of Zamora, whose cupola 
 is one of the most beautiful in Europe. 
 
COHESIVE CONSTRUCTION. 
 
 31 
 
 (22) After this epoch, in the Renaissance, the 
 most remarkable are those in the Santa Maria 
 del Fiore, and the Medici Chapel and Baptistry 
 of Florence ; St. Augustine's and St. Peter's in 
 the Vatican, Rome ; the Madonna de la Salute, 
 Venice ; Ste. Genevieve, Paris ; St. Paul's, 
 London ; La Real capilla de los Desemparados, 
 Valencia (Fig. 8) and the Dome de los Escol- 
 apios (Fig. 9), Spain. 
 
32 
 
 COHESIVE CONSTRUCTION. 
 
 (23) It may here be appropriate to call atten- 
 tion to one important point. All the cupolas 
 constructed, up to the epoch of Constantine, 
 
 Fig. 5. 
 
 were with brick and concrete, in the Arabic 
 style, following the constructive lines without 
 •altering the aesthetic forms; and in all the cupolas 
 built after the time of Constantine, beginning 
 
COHESIVE CONSTRUCTION. 
 
 33 
 
34 COHESIVE CONSTRUCTION. 
 
 with the period of Brunelleschi in Florence, 
 including the dome of St. Paul's, London, the 
 exteriors are not the representation of the 
 interiors, being successively more prominent. 
 This deviation, in the modern ones, can be 
 seen by comparing the Brunelleschi cupola 
 with St. Peter's of the Vatican and St. Paul's of 
 London (Figs. 3, 6 and 7). Brunelleschi's double 
 cupola, outside, is adapted to the same shape as 
 the inside, giving, apparently, only the hollow 
 space, perhaps to give dry conditions for later 
 decoration, as can be seen by the details of 
 the ribs (Figs. 4 and 5). In the last cupola 
 mentioned (Fig. 7), the interior dome or dec- 
 oration is a hemisphere, the second one is 
 the same shape as a truncated cone, and the 
 third one' is the exterior dome. The whole 
 does not represent the progress of the art of 
 instruction and the way to apply aesthetic 
 forms. 
 
 (24) This anomaly is due to the fact of the 
 disuse of the hydraulic mortars of the Romans, 
 Arabians and Byzantines, because the art of 
 manufacturing these materials, which consti- 
 tuted the basis of their cohesive construction, 
 was lost. In the construction of St. Sophia the 
 Byzantines used baked clay and lava of Vesu- 
 vius, or pumice-stone. 
 
COHESIVE CONSTRUCTION. 
 
 35 
 
 RENAISSANCE. 
 
 (25) The architects of the Renaissance, espe- 
 cially in Italy and Spain, were greatly impressed 
 by the works of the Romans, Byzantines and 
 Arabians, and wished to imitate their bold 
 
 Fig. 7. 
 
 construction; but they did not have at hand 
 either the material or the skilled hands. They 
 therefore used plaster, and thenceforth the 
 timbrel arch was introduced along the coasts of 
 
36 COHESIVE CONSTRUCTION. 
 
 the Mediterranean from Murcia to Valencia, 
 Barcelona and Genoa, and along the Italian coast 
 to Naples. Remains of the timbrel arch will be 
 found in all those parts. This epoch demands 
 great attention because of the many facts it 
 supplies to aid us in our study. 
 
 (26) When the architects of the Pontificate, 
 in order to supply the richness and grandeur 
 called for in this epoch, took for their models 
 the Roman and Byzantine construction, as al- 
 ready stated, they had neither the material nor 
 the skilled labor necessary ; consequently it was 
 impossible for them to imitate when they had 
 only common air-lime and plaster. The first 
 they found impossible to use in constructions 
 similar to St. Sophia, the Cathedral of Zamora 
 or the Arabian cupolas. As to the second, 
 they found that the unlimited expansion of the 
 plaster, which only stops when fully saturated, 
 — that is, when it loses its power of absorption, 
 — compelled the architects, to supply walls of 
 enormous thickness. Besides this disadvantage, 
 when the plaster has arrived at this condition 
 of saturation its strength is gone, loosening the 
 bricks principally where the building is ex- 
 posed to the weather or subject to alternate 
 changes of humidity and dryness. In conse- 
 quence, its use was limited to very heavy walls, 
 
COHESIVE CONSTRUCTION. 37 
 
 and for ceilings having wooden beams and 
 wooden boards, over which were laid Arabian 
 tiles if it was a roof, and mortar and flooring 
 tiles if it was a floor. 
 
 In some cases the timbrel vaults were used 
 as a ceiling and floor, having two or three thick- 
 nesses of tiles with plaster, and the haunches 
 were filled with light pottery ; this pottery was 
 leveled over with rubbish and mortar, finishing 
 with flooring tiles. 
 
 (27) It is necessary to remark that all of this 
 construction was used only in large buildings, 
 such as convents, palaces and churches, where 
 the walls were very thick, amounting to one- 
 third of the full span, and where the character 
 of the building was a guarantee that the ceil- 
 ings would not be abused; otherwise it was 
 necessary to patch and repair every few years. 
 But in general building it was only used in 
 small spans, such as eighteen to twenty inches 
 between beams, three tiles being used to two 
 courses of bricks set flat over the centre ; and in 
 this state it has remained until the present date. 
 
 (28) With this I conclude the review of this 
 form of construction, the antique and Renais- 
 sance, passing to the modern epoch. 
 
 As we may observe, all the timbrel arches 
 of the Renaissance epoch existing in Italy, as 
 
38 
 
 COHESIVE CONSTRUCTION. 
 
 ^jSSSSSSSSl.. 
 
 -^^^^^^.^ 
 
 1Z5 METROS 
 
 O 
 a 
 
 2 
 
 Fig. 8. Dome and Plan 
 
COHESIVE CONSTRUCTION. 
 
 39 
 
 well as in Spain, are constructed with plaster 
 material, which does not meet the exigencies of 
 good construction. Consequently, it is natural 
 that no technical academy in Spain or Italy has 
 
 Fip'. 9. Brick Dome with Iron Rings. 
 
 taken into serious consideration such empiric 
 construction, which has a tendency to lamenta- 
 ble accidents. France and England are not 
 here taken into consideration, because, like the 
 
40 COHESIVE CONSTRUCTION. 
 
 other nations of the north, they have not bricks 
 of dimensions and conditions suitable for the 
 cohesive form ; they have bricks of a small top 
 and bottom surface, that is, four by eight inches, 
 
 Fig. 9. 
 
 when, generally, the types for the cohesive 
 system are the Assyrian bricks, or the bricks 
 of the Orientals, the dimensions of which were 
 about twelve to fourteen inches long, six to 
 eight inches wide and one to two inches thick. 
 
COHESIVE CONSTRUCTION. 41 
 
 (29) In Spain, where this system has been 
 used and is still in use on a larger scale than in 
 any other country, there does not exist any 
 treatise, or a single work, on the theory of this 
 construction, or a single scientific explanation 
 of this manner of building ; not even an 
 empirical explanation to satisfy the curious. 
 
 (30) The time that cements began to be gen- 
 erally used in modern days was from 1845 to 
 1850, and from this date commenced the renais- 
 sance of the "Cohesive Construction." The 
 modern Roman cement that Mr. Parker invented 
 and patented in 1791 and 1796 was so dear, and 
 the conditions of setting were so slow, that its 
 introduction into buildings was much retarded. 
 
 This cement, in the beginning, was called 
 Parker's cement ; its author called it Roman 
 mortar. The other cements called Medina, 
 introduced shortly after, had the same defects. 
 Mr. Aspdin, on the 21st of October, 1824, took 
 out a patent for the formula of the celebrated 
 Portland cement. 
 
 This cement was given the name of Portland 
 by the author, Mr. Aspdin, because, when it is 
 good, and smoothed with the trowel, it is very 
 similar to Portland stone when it is polished. 
 
 (31) Up to the year 1868 the professors 
 of the academy of Barcelona, one of the most 
 
42 COHESIVE CONSTRUCTIO^f. 
 
 illustrious of Europe, and a city where tiles are 
 more in use than in the rest of the world, did 
 not commence to pay any attention to this style 
 of construction until important applications had 
 been made of it. And when at last they did, it 
 was only to comment incidentally on its resist- 
 ance, and its possible utility ; but they did not 
 make a study of it, notwithstanding the fact 
 that they were constantly walking over floors 
 constructed on this system in which plaster was 
 used ; on the other hand, this want of attention 
 is explained by the lack of cements proper in 
 those days for such kind of construction. The 
 want of proper cements, and of an invariable 
 brand on which to base their calculations, was 
 one of the main obstacles which beset the Cata- 
 lonian and Valencian architects. 
 
 (32) The first work of importance of this 
 character constructed in Spain was in 1868 — 
 the manufactory of Batllo in La Corts de Serria. 
 It is a series of buildings where there are 2,000 
 people employed with 1,000 looms and 64,000 
 spindles. Afterwards I built others already 
 mentioned. In some cases the risk and danger 
 caused by the irregularity of the materials was 
 so plain that the workmen were afraid to go 
 ahead, compelling me to remain in the works to 
 inspire confidence and success. 
 
COHESIVE CONSTRUCTION. 43 
 
 (33) The progress verified in Spain, particu- 
 larly in Barcelona, in this special construction, 
 was due principally to the manufacturers, and 
 to the studies and teachings of the professors 
 on these subjects, who were debating for sev- 
 eral years, at the same time, how to improve 
 their respective specialties, and how to obtain 
 new practical systems of construction, knowing, 
 as they did, that the improvement in material 
 required change and progress in construction. 
 But their noble aspirations were restricted, as 
 they had no facilities ; besides which it was nec- 
 essary for them to content themselves by rec- 
 ommending the theories of Vicat about the use 
 of cements, and other applications well based. 
 
 (34) Nothing was done about investigating 
 these structures to which I have referred, and 
 no coefficients were discovered. These only can 
 be obtained when we can depend upon the mate- 
 rials with mathematical regularity, and with 
 powerful apparatus for determining their relia- 
 bility. They can only be obtained, too, in coun- 
 tries where we can find in the market enough 
 guaranteed brands of Portland cement of differ- 
 ent setting degrees ; where clay can be used 
 for these constructions with advantage, and 
 with regularity of manufacture thereafter ; and, 
 finally, after trial in a country where we have 
 
44 COHESIVE CONSTRUCTION. 
 
 powerful apparatus, and where coefficients can 
 be obtained, as has been exemplified by our own 
 work for the past five years. 
 
 (35) On account of these special advantages it 
 seems that all this experimental work culmina- 
 ted in the United States, taking a natural stand 
 in New York and Boston, with specimens that 
 have no rivals in any part of the world for light- 
 ness and resistance. We now see that the 
 movement initiated in England by the unappre- 
 ciated Mr. Parker, with 1791 and 1796 patents 
 — who thought he had discovered the old 
 Roman cements, — after passing the patented 
 improvements of Mr. Aspdin, of October 21, 
 1824, may have culminated in New York and 
 Boston. But not without the valuable assist- 
 ance and confidence of the eminent architects, 
 Messrs. McKim, Mead & White, Buchman & 
 Deisler, R. H. Robertson, F. H. Kimball, T. M. 
 Clark, De Lemos & Cordes, A. H. Pickering, 
 A. F. D'Oench, and others, whose co-opera- 
 tion and support in the annals of constructive 
 art deserve to be held in remembrance. 
 
PART III. 
 
 THEORY AND COEFFICIENTS OF APPLICATION. 
 
 We will divide construction in general into 
 two classes : 
 
 (36) First, "Mechanical Construction," or, 
 construction by gravity. 
 
 Second, "Cohesive Construction," or, con- 
 struction by assimilation. 
 
 {^y) The first is founded in the resistance of 
 any solid to the action of gravity when opposed 
 by another solid. From these conjuctive forces, 
 more or less opposed to one another, results the 
 equilibrium of the total mass, without taking 
 into consideration the cohesive power of the 
 material set between the solids. 
 
 (38) The second has for a basis the properties 
 of cohesion and assimilation of several materials ; 
 which, by a transformation more or less rapid, 
 resemble Nature's work in making conglom- 
 erates. 
 
 (39) We can give another definition more pre- 
 cise and comprehensive for both systems, in 
 
46 COHESIVE CONSTRUCTION. 
 
 saying, that the first, or mechanical, system is 
 that where all the pieces can be separated one 
 by one and then rebuilt in the same or similar 
 manner. To this class belong the pyramids of 
 Egypt and the Greek temples, etc. In "Cohe- 
 sive Construction," on the contrary, the com- 
 ponents cannot be separated without destroying 
 the integral mass. To these belong the Baby- 
 lonian walls of brick with hydraulic mortar; the 
 vaults and cupolas of the Assyrian, Persian, 
 Arabian, Roman and Byzantine — the antique 
 and Middle Age conglomerate construction. 
 
 The structures built by the "Gravity System" 
 can at any time be taken down in the pieces 
 out of which they were formed. Thus, the 
 stone or brick that yesterday formed part of a 
 temple or monument dedicated to the memory 
 of a hero can to-morrow belong to or form a 
 part of the walls of a stable ; while, on the other 
 hand, though man cannot again use the parts 
 of "Cohesive Construction" for modern build- 
 ings, their ruins inspire respect and veneration ; 
 and only Nature, with her slow but sure work 
 of disintegration, can take from this style of 
 building its material for her immense and eter- 
 nal laboratory. 
 
 (40) The materials employed in construction 
 by gravity only require the physical quality of 
 
COHESIVE CONSTRUCTION 47 
 
 hardness. For the "Cohesive Construction" 
 the materials must not only have proper physi- 
 cal conditions, but it is absolutely necessary 
 that the chemical properties of the substances 
 employed should be taken into consideration. 
 The use of the "Cohesive System" was ren- 
 dered impossible to many nations because they 
 had neither the material, nor the knowledge of 
 its use, at their disposal; while on the other 
 hand, all civilizations and all nations could make 
 use of the gravity system. 
 
 (41) The basis of these materials is mortar 
 that does not require exposure to the air for its 
 transformation or setting quality — that is, 
 hydraulic lime and cement ; but for our spe- 
 cialty we must have cements of the quality of 
 Portland. The formula for the manufacture of 
 these kinds of materials, and the manner of 
 their use was, it seems, lost (probably soon after 
 the fall of the Roman Empire), barring some 
 rude practices in the Orient, which soon disap- 
 peared, to be found again in 1791 by Mr. Parker. 
 The two patents taken out by him in 1791 and 
 1796 were not complete, and they came too late 
 to be taken into consideration in the scientific 
 movement, already well advanced, that was 
 giving impulse to the academic and technical 
 schools in the last century, whose text-books in 
 
48 COHESIVE CONSTRUCTION. 
 
 general did not refer to any save the gravity 
 system. These doctrines or ideas prevail yet, 
 and are now the basis of our teachings. 
 
 (42) Nothing had then been written in regard 
 to cohesive construction as applied to the 
 "timbrel arch." This was due to the follow- 
 ing circumstance, which is worthy of remark: 
 The nations who for nearly a century and a 
 half were most advanced in scientific and liter- 
 ary work, and who had written most about the 
 applied sciences, were the English, French and 
 Germans, precisely the people who, on account 
 of the erroneous form of their bricks, and the 
 poor method of using their materials, furnished 
 the frequent spectacle, that, when the walls of 
 any of their buildings were torn down and the 
 bricks taken out, they were so nearly clean, 
 without any mortar adhering to them, that they 
 could be used again in new walls. Could the 
 professors and scientific men of these countries 
 have seriously taken into consideration the cohe- 
 sive strength, although they knew of the exist- 
 ence of some cements.'' Certainly it is not 
 strange that all conscientious professors were 
 inclined to give coefficients of resistance only 
 for the gravity system, and all their books and 
 teachings were on the gravity system, except 
 for tensions of materials working in that way. 
 
COHESIVE CONSTRUCTION. 
 
 49 
 
 Italy and Spain at this epoch had no text-books 
 of their own ; all were translations from the 
 
 French and English works. 
 
 TIMBREL ARCHES. 
 
 (43) We will begin by investigating the way 
 in which this kind of arch works. 
 
 Fig. 10. 
 
 A "Timbrel Vault" of a single thickness of 
 brick or tile (Fig. 10) has no more resistance 
 than an arch or vault built on the " Gravity 
 System " ; because, no matter how good the 
 mortar may be, there is only one vertical joint, 
 and the bricks or tiles are working as voussoirs. 
 Consequently this form of arch belongs to the 
 " Gravity System." But if we put another 
 
 course over the first (Fig. 11), breaking joints, 
 and laid with hydraulic material, we will have 
 the action of cohesive force. In this way the 
 
50 
 
 COHESIVE CONSTRUCTION. 
 
 mortar laid over the first course, or extrados, 
 takes bond with it, and also with the course 
 laid on top. As soon as the cement sets, we 
 will have shearing resistance represented by 
 
 Fig. 12 
 
 17,820 pounds per square foot (Fig. 12, test 
 No. 4873). In this way we introduce a new 
 additional strength to the arch which is a joecu- 
 liarity of the Timbrel Arch System. In the 
 Gravity System (Fig. 10), the strength of 
 gravity alone is the only force keeping the 
 voussoirs in place by pressure against each 
 other in the joints. These joints are not 
 
COHESIVE CONSTRUCTION. 51 
 
 protected, and any reduction in the width of the 
 joints in consequence of pressure, or weight on 
 the arch, coinproiiiiscs the setting of the mortar. 
 For this reason in the " Gravity System " the 
 mortar serves only as a cushion, even if cement 
 mortar, because of bad setting, and adds no 
 strength to the arch. But in our " Cohesive 
 System," with horizontal broken joints, with 
 17,820 pounds per square foot shearing strength, 
 the reduction in the vertical joints is prevented 
 absolutely, as can be proved by the following 
 facts : First, we can build arches of twenty- 
 feet span only three inches thick, using a cen- 
 tre one inch thick, and moving it along as soon 
 as a row of tiles is laid, which usually requires 
 about fifteen minutes. Second, it is common 
 to see the workmen walking over the arch, free 
 from centres of any kind, some hours after it 
 is built ; and third, tvc can run tJie centre under 
 the arch again tvlien it is completed, wliich is 
 tlie most practical illustration that the arch has 
 had the absolnte repose necessary for its settle- 
 ment. 
 
 (44) These three remarkable circumstances 
 are of great value to architects, as the)' can be 
 put in the specifications and can be depended 
 upon as absolute proof of the safety of the con- 
 struction. 
 
62 COHESIVE CONSTRUCTION. 
 
 But this horizontal breaking joint, or new 
 additional strength, is not the only great advan- 
 tage of the system. There are others, the 
 principles of which we will try to explain. 
 
 (4S) It is evident that if we were able to 
 build an arch without joints, it would be the 
 best, as it would have no settlement ; but as the 
 gravity system has only voussoirs of stone or 
 brick a certain number of joints are necessary. 
 
 Let us suppose that we have a brick arch 
 (Fig. 13) of six-feet span. We would have 
 about 26 or 27 joints of common brick. These 
 joints, being one-quarter of an inch thick, repre- 
 sent about seven inches of mortar, which is 
 compelled to set with all the weight of the 
 voussoirs resting on the centres. The centres, 
 contracting, leave the weight or pressure on 
 the mortar, thus preventing a good setting, and 
 raising the centre of pressure of the arch from 
 A to B (Fig. 13); this happens in all the brick 
 arches, more or less. When this arch rises only 
 ten per cent, of the full span it is very dangerous, 
 
COHESIVE CONSTRUCTION. 53 
 
 because the development of the curve measures 
 very little more than its chord C D. The 
 builder or contractor, knowing this, is always 
 afraid when the centres are removed, and, before 
 the architect knows it, he brings dozvn the ccn- 
 h-e of pressure still further, by hammering in 
 little iron wedges or nails in the joints, covering 
 them with mortar so as not to be seen. This 
 is not good practice, for it destroys what cohe- 
 
 Fig. 14 
 
 sion may still be left in the joints, but has the 
 advantage that it prepares the brick for second- 
 hand material by freeing it from the mortar. 
 In our arch, in the same six feet (Fig. 14), we 
 have only 13 joints, one-quarter of an inch each, 
 which is only three and one-quarter inches of 
 mortar; consequently, as we know that the arch 
 with no joints is the best, the one with the 
 least is to be preferred. 
 
 (46) There are other advantages equally im- 
 portant. We know that in every arch the curve 
 of pressure changes according to the position 
 of the load; this means that every arch must be 
 
54 
 
 COHESIVE CONSTRUCTION. 
 
 prepared for work by deflection or tension. 
 Let us suppose an arch laid in brick, in such a 
 manner as to receive a test for tension (Fig 15). 
 
 ^5^^ 
 
 ? 
 
 Illllllllll 
 
 Fig. 15. 
 
 W^' 
 
 V 
 
 The resistance of this tension depends upon 
 the cohesion of the joints, .or the resistance to 
 tension of the mortar. But we have observed 
 that this cohesion in the brick arclies was very 
 unsatisfactory, and that tlie mortar is only a 
 cushion in many cases, but that when these 
 joints have a good settlement the tension will 
 only equal the cohesive strength of the mortar 
 between the bricks, and with good cement mor- 
 
 Fig. 17. 
 
 tar ten days laid, this strength is only from 80 
 to 150 pounds per square inch, while we have 
 for our " Timbrel Arch" tensile strength, test 
 
COHESIVE CONSTRUCTION. 55 
 
 No. 4,875 and 4,876, 287 pounds for ten days, 
 and 1 59 pounds per square inch for seven days 
 (Fig.y). 
 
 This shows that we have for the cohesive 
 construction the following advantages over the 
 brick arch or any arch built by mechanical con- 
 struction : 
 
 (47) First, the protection of the vertical joints, 
 by introducing the new strength coming from 
 the horizontal breaking joints. 
 
 (48) Second, the less number of vertical 
 joints, amounting to only five per cent, of the 
 full span, while the brick arch has ten per cent. 
 
 (49) Third, the resistance to deflection 
 (bending-moment), see page 80, " Analysis of 
 some Peculiarities of the System." 
 
 (50) The result of these advantages is the 
 surprising strength of the "Timbrel Arches," 
 so that no one can at first understand how 1 5 
 or 20-foot arches, three inches thick and with ten 
 per cent, rise, as we said before, can be laid, tak- 
 ing away the centres and giving them over 
 to the uses of the building in a few hours, when 
 an arch of brick with a six-foot span, four 
 inches thick and a ten per cent, rise, requires 
 strong and heavy centres, with several days' 
 repose. Even then this six-foot span, four 
 inches thick and with a ten per cent, rise, is 
 
56 COHESIVE CONSTRUCTION. 
 
 not a safe construction. The span requires 
 eight inches of thickness, as all architects and 
 builders know. 
 
 (51) We may consider, as a safe relation be- 
 tween the brick and "Timbrel Arches," a brick 
 arch four feet six inches in span, ten per cent, 
 in rise, and four inches thick, with cement mor- 
 tar as is usual in buildings, as equivalent to a 
 ten or twelve foot span in a "Timbrel Arch," 
 three inches thick, and with an eight to ten per 
 cent. rise. 
 
 (52) All that we have said about the brick 
 arches in comparison with the " Timbrel Arch " 
 can be applied to the construction of concrete 
 or conglomerate arches, especially in regard to 
 the inconvenience of using heavy centres, and 
 imperfect settlement of the materials. In large 
 arches, the cement cannot be put over the arches 
 quickly enough so that every layer can settle 
 evenly, and the excessive use of the rammer 
 kills part of the cement. That is the reason 
 that for forty years these concrete arches have 
 been tried without success, except in small 
 arches, where the laborers that are generally 
 used in this kind of work can complete the full 
 span of the arch with one single coat having a 
 uniform settlement ; but not without always 
 using more material than necessary. 
 
COHESIVE CONSTRUCTION. 57 
 
 I give an instance of the use of concrete in 
 tlie foundation of tlie manufactory of Batllo 
 Brothers : I specified cement concrete for three 
 feet all along the foundation as a first course, 
 putting on top four courses of bricks (6 by 12 
 by I 1-2 inches) about the same as we are now 
 using. This was in 1869, twenty-three years 
 ago. I gave orders to lay the concrete si.x 
 inches at a time. The cement was slow setting 
 for cement, yet I could see some signs of crys- 
 tallization on the same day. The next day, 
 when inspecting the work of the day before, 
 I found it all a mass of mud. It cost me ten 
 days' labor and many barrels of cement in ex- 
 perimenting with different brands before I as- 
 certained the true cause. It was necessary to 
 adopt a hydraulic mortar composed of two parts 
 lime, two parts sand, and three parts brick-dust, 
 in order to give slow hydraulic mortar, because 
 cement requires repose for a certain length of 
 time in which to set, and this putting it on in a 
 six-inch course and hammering it down so jarred 
 the whole mass that its.rest was disturbed and its 
 crystallizing qualities killed. This can be seen 
 in the use -of our tiles. Two minutes after the 
 tile is bedded in the arch the cement has begun 
 to set, or crystallize, and cannot be disturbed or 
 used again, when the same cement in the mortar 
 bed will remain several hours without setting. 
 
•58 COHESIVE CONSTRUCTION. 
 
 COEFFICIENTS. 
 
 (53) In May, 1877, I commenced a series of 
 experiments in the Department of Tests and 
 Experiments of the Fairbanks' Scale Company, 
 Thomas Street, New York, with the engineer, 
 A. V. Abbott, and I obtained some coefficients. 
 These coefficients are as follows : — 
 No. 4817, May 3, 1887, compression test, 
 
 sq. in., 5 days, 2,277 't)S. 
 
 No. 4818, May 3, 1887, compression test, 
 
 sq. in., s days, 1,624 'bs. 
 
 In the last the heads were not even. 
 No. 4869, June 6, 1887, compression test, 
 
 sq. in., 5 days, 1,43 
 
 No. 4870, June 6, 1887, compression test, 
 
 sq. in., 5 days, 2,911 lbs. 
 
 8,242 lbs. 
 
 = 2,000 lbs. 
 
 4 
 No. 7475, Oct. 22, 1889, compression test, 
 
 sq. in., I year, 3,290 lbs. 
 
 Transverse (Fig. 16). (Page 60.) 
 No. 4871, June 6, 1889. 90 lbs. per sq. in. 
 
 Tension (Fig. 17). 
 Test No. 4875, Jan. 7, 1887. 287 lbs. per sq. in. 
 
 Shearing Stress (Fig. 12). 
 Test No. 4873, June 6, 1887, in Portland cement, 
 8,910 lbs. = 123.7 ^t>s. per sq. in. 
 
COHESIVE CONSTRUCTION. 59 
 
 No. 4872, June 6, 1887, in plaster-of-Paris, 2,450 = 
 34 lbs. per sq. in. 
 
 GENERAL FORMULA FOR SEGMENTAL ARCHES. 
 LS 
 
 (i;4) T C ^ _ — (i) for distributed load. 
 ^ 8 r 
 
 (The explanation of these formulas will be 
 
 given later.) 
 
 L = Load in pounds including material. (L is 
 
 always 12" in length X span and X load in 
 
 lbs. per superficial foot, including material.) 
 R = Resistance in middle of arch, or T X C. 
 C = Coefficient for compression = 2,060 lbs. 
 
 per sq. in., breaking load. 
 C' = Coefficient tension = 300 lbs. per sq. in., 
 
 breaking load. 
 C" = Coefficient transverse = 90 lbs. per sq. in., 
 
 breaking load. 
 T = Area of cross-section, in superficial inches, 
 
 in the middle of the arch. (T will always 
 
 be 12 X thickness, or area represented by 
 
 12 = thickness.) 
 r^ Rise of arch in feet. 
 S = Span in feet. 
 
 (55) We use the formula (i) to get the thick- 
 ness necessary at the centre of the arch with a 
 distributed load, including the weight of the 
 arch itself. After that we find the line of the 
 
60 
 
 COHESIVE CONSTRUCTION. 
 
 extrados of the arch in a graphical manner, 
 derived from the formula given by Dejardin for 
 tracing the equilibrium profile of the extrados 
 for the vaults, giving the section of the arch in 
 the skewbacks, or base of the arch on each side. 
 
 (56) This formula is (2) V == X 
 
 COS. a 
 
 and is 
 
 the general one for any semicircular or seg- 
 mental arch, but making for the first case 
 
 Fig 16, Fig. 18. 
 
 a = 60° for the segment of an arch ; a equals 
 the degrees corresponding to one-third of the 
 segment in which V is the radius vector of the 
 extrados O N (Fig. 18). 
 X is the radius of the intrados O M. 
 c the thickness in the centre, or A B. 
 a the angle that any radius makes with the ver- 
 tical O A. 
 
COHESIVE CONSTRUCTION. 61 
 
 Hence the complete formula is 
 
 ( \ L S 
 
 ^^^ 8rx 12 C 
 which represents the thickness of the centre 
 of the arch in inches, or 
 
 area of i foot arch in length, 
 
 12 
 
 and 
 
 ^4^ 87^X Y^C + (V - (X + B A) for the 
 thickness of the spring of the arch in inches. 
 
 (57) The graphic procedure is as follows 
 (Fig. i8): 
 
 Take the thickness T, or, say A B, and lay 
 off O H equal to it ; draw H' H parallel with 
 the chord O K, draw O N through the point M, 
 which is one-third of B M' — . 
 
 Lay off M N = O G, N gives us one point 
 of the curve of the extrados. See Fig. i8. As 
 N M is the weakest part of the arch, we can 
 safely put the same thickness at the spring ■ 
 which gives us the third point that is necessary 
 to trace our curve. 
 
 (58) With the same formula we can find the 
 thickness of the arch necessary for a single load 
 at the middle, but making 4^ instead of 8 ;-. 
 
 We now come to the problem of a load on 
 any point of an arch. 
 
62 COHESIVE CONSTRUCTION. 
 
 (59) The remark has been made that these 
 kind of arches cannot be used for a moving or 
 concentrated load. We know that if an arch is 
 built with the condition that the curve of pres- 
 sure is inside of the middle third, the arch is safe ; 
 starting with this we apply the general formula 
 for finding the thickness in the centre, and we 
 trace graphically the curve of pressure as shown 
 in Fig. 19, as if the load was resting on point 
 1 1. That gives a lower line of pressure than 
 any other point for one side of the arch, and 
 when the load is on a corresponding position 
 on the opposite side, the same curve reversed 
 gives us the whole arch. 
 
 Now it is necessary to put half the thickness 
 given by the formula on either side of the line 
 of pressure O, O', O". Fig. 19, forming the 
 lines X, X', X", Z, Z', Z", that represents the 
 total thickness of the arch. With this, as we 
 have said, we have the thickness of the arch 
 necessary for the lowest line of pressure re- 
 quired for any position of the load. 
 
 When the load is on 1 1, the pressure is pas- 
 sing through the imaginary lines 1 1 C and 1 1 e ; 
 when the load is on 10, the pressure is passing 
 through the imaginary lines 10 a, 10 e' , etc. 
 Consequently it is inside of this area, between 
 the level of the floor and the line of lowest 
 
COHESIVE CONSTRUCTION. 
 
 63 
 
 lines of the arch (Fig. 19) 
 where the curve of pres- 
 sure rests ; that means that 
 if we fill up solid, or in 
 such a way as to take the 
 place of solid materials, as, 
 for example, in tubular 
 girders, we are sure that 
 the curve of pressure will 
 always be inside of the 
 arch. 
 
 (59) We say, or in such 
 a way as to take the place 
 of solid materials, because 
 in practice it is better to 
 avoid the enormous 
 weight of this unneces- 
 sary mass of material, and, 
 besides, to avoid the con- 
 densation that such a 
 mass of material accumu- 
 lates in the ceiling, by 
 building the lower part of 
 the arch X', X", X'", and 
 Z', Z", Z'", and over that 
 building bridges at a suffi- 
 cient distance (generally 
 two feet apart), and over 
 
 ■•«"«<<! 
 
64 COHESIVE CONSTRUCTION. 
 
 these building again a flat arch of the same 
 thickness as the arch below, forming in all, by 
 this construction, a regular tubular arch, light, 
 dry and well ventilated, and with sufficient 
 strength in every part.* 
 
 (60) Take for example an arch with a 15 ft. 
 span, and a 10% rise, that must support 250 
 lbs. per square foot, including material f and 
 distributing load. 
 
 S = IS feet. 
 
 r =:^ i-,'o- feet. 
 
 C = 2,060 lbs. per sq. in. 
 
 L= 250 lbs. X 15 ^ 3.750 lbs. 
 
 As the load is distributed, 
 
 TC^\75°X '5^ 687.5. and 4:^5 = 2.275 
 8 X 1-5 -21060 
 
 But we are working at 10% breaking load and 
 T = 22.7s" (twenty-two y'/^ superficial inches), 
 or an area 12" X 1.9" or i| tiles; two courses 
 will be used, making 2'' X 12" = 24 square 
 inches. 
 
 But with this we have only the thickness 
 necessary in the middle of the arch. 
 
 * One example o£ this is the section drawn for the projected 
 bridge at Prospect Park, Brooklyn, by the architects, Messrs. 
 McKim, Mead &. White. 
 
 t That weight must be considered as a distributed load. 
 
COHESIVE CONSTRUCTION. 
 
 65 
 
 (6i) To find the thickness for the spring, we 
 shall have to determine the extrados N' A' c 
 (Fig. 20) by the graphic method devised in the 
 formula of Dejardin, or by the formula (4) ; we 
 
 Fig. 20. 
 
 thus find that T in the spring is about 
 12" X 2.26" or, 3 courses, inside of 12" X i-9" 
 and 12" X 2.26' or, 3 courses, or, 3 inches, must 
 be adopted in order to increase the resistance 
 to bending moment * in the haunches of the arch. 
 Two and one-half inches is enough, but it is 
 better to give the half-inch in excess, to be on 
 the safe side, because no thinner tiles are made 
 
 * We give, at the end of this book, a table taking in con- 
 sideration the bending moment (see page 148). 
 
66 COHESIVE CONSTRUCTION. 
 
 to adjust the thickness, and it is preferable to 
 make this allowance, thus increasing the section 
 at the base of the arch, laying three courses in 
 the sides and two in the centre. 
 
 DOMES. 
 
 (62) The dome is the genuine form of the 
 cohesive construction for ceilings, floors and 
 roofs, as well as for timbrel arches. 
 
 (63) We use the following formula for dis- 
 tributed loads : — 
 
 L S 
 
 (S) s — rz 7^-. — foi" l^he thickness on the 
 
 ^^' 8 r X 12 C X 2 
 
 crown. (L = Span X i foot X weight in lbs. 
 per superficial feet including material), and 
 
 <^) 8.Xi2'cX2 + (^-(^ + -^^0^"'- 
 the spring in inches, not taking into considera- 
 tion anything but the pressure. 
 
 (64) Practically we can have the same result 
 by using formula (i) to get the thickness neces- 
 sary at the crown, as in the case of the barrel 
 or segmental arches. Afterwards take from 
 the crown and from the spring half of the 
 thickness given by formula (i), or else deduct- 
 
 . LS 
 
 mg 7; . 
 
COHESIVE CONSTRUCTION. 67 
 
 (65) Example : 
 
 5 = 40. 
 
 r = 4 feet. 
 
 C ^ 2,060 lbs. breaking load. 
 
 L= 250 X 40= 10,000 lbs. 
 As the load is distributed, 
 
 „„ 10,000X40 ,12,500' ^ 
 
 T C = zr ^-- = 1 2, 500, and ■^ — = 60, 
 
 8X4 2,060 
 
 taking 10% of the breaking load, and as we 
 count only a piece of the dome i foot or 12 
 
 inches in length, — :;= ? inches. Now to find 
 12 
 
 the additional thickness, or, to trace the equi- 
 librium profile of the extrados, we will provide 
 the same as in the case of example (60) for the 
 segmental arches. We will find that the in- 
 crease in the spring is about 2" or 7" in all. 
 Now we must take off 
 L S 10,000 X 40 12,500 
 8 ;- or 8X4 or 2,060 = 3 breaking load, 
 22 2 
 
 30 
 so" safe load, and ^^7 = 2 .',''• So we must 
 
 ^ 12" - 
 
 take away 2" from the thickness of the dome. 
 Now as the thickness was 5" in the crown and 
 7" in the spring, taking off 2" will give 3" in the 
 crown and 5" in the spring. 
 
68 
 
 COHESIVE CONSTRUCTION. 
 
 (66) Explanation of the formula : 
 
 We do not pretend to have an absolute math- 
 ematical formula, but a practical one, enough to 
 insure sufficient security for safe construction. 
 
 We may add that these formulas and theo- 
 ries, if not founded absolutely on the known 
 and admitted theories of the " Gravity System," 
 are nevertheless on the principles admitted for 
 all kinds of cast bodies, and I thought it better 
 
 S 't 3 2 
 
 Fia. 21, 
 
 Fig. 21 bis. 
 
 to put them in the form commonly used, to 
 make them more clear, and avoid strange ideas 
 which serve to confuse the mind, before ex- 
 plaining their different forms of working exclu- 
 sively within the theory of the cohesive sys- 
 tem, which we will explain later. 
 
COHESIVE CONSTRUCTION. 69 
 
 {6j) We consider our arch not as an arch 
 with voussoirs, but as a single cast arch, work- 
 ing as a solid piece of arched stone or iron. 
 
 Suppose, now, we make a solid piece of mate- 
 rial of the form i B C (Fig. 21), without taking 
 into consideration its weight, the curve i C, 
 being half of a segment of an arch, and at C 
 a pivot. Applying 10 tons on the point i, we 
 find by the law of mechanics, that, having B C 
 = B I, we need, at the point I, a horizontal 
 pressure of 10 tons for equilibrium. 
 
 (63) Let us pass the 10 tons to point 2, 
 that is, an arch of which the span C - C" (Fig. 
 21) is 4 times the rise ; as B 2 is double B C, the 
 equilibrium of the 10 tons vertically at the point 
 2 must be, 10X2 = 20 tons applied horizontally. 
 
 We now pass the 10 tons to the point 3 ; that 
 is, an arch of which the span is 6 times the rise, 
 and we will find that for equilibrium it will be 
 necessary to multiply 10 X 3 = 30 tons. The 
 same holds good when we pass to No. 4, which 
 will be forty tons, and to No. 5, which will be 50 
 tons : that is, in an arch with a rise of ten per 
 cent, of the span, we have a pressure in the 
 middle five times the weight of the single load. 
 
 If we double the figure, that is, the full arch C 
 A (Fig. 21 bis), and we consider the same load 
 of ID tons, at the point 3 for each half, and if we 
 
70 COHESIVE CONSTRUCTION. 
 
 consider the pivot at the places C and A, as 
 both form a single arch, we will' have the pres- 
 sure in the middle of the arch one-half as much 
 as before, or 30 for either side, and the real 
 pressure, between the two heads, will be the 
 full 30 tons and no more. 
 
 (69) With this we find out, that in any single 
 load over an arch, the thrust is divided equally 
 on both sides ; but in the middle section of the 
 arch we have a pressure one-half of the full 
 load. This shows that the section in any arch, 
 independent of its own weight, and only taking 
 into consideration the weight of the load, will 
 be the same as that of the section on either 
 spring, and the section on the crown will be 
 
 T C = ^ , for concentrated load, 
 
 2X2 
 
 or, T C = — — - for any distributed load 
 
 2X2 ^ 
 
 r X 2 
 replacing B i, B 2, etc., in every case by 
 
 and B C by r, the rise of the arch, and, 
 
 !•£ J -r r- Load X Span X 4 
 
 simplified, T C ^ — 
 
 2 r 
 
 Load X Span L S 
 
 ~ 4 X 2 X r ~ 8~7" 
 
COHESIVE CONSTRUCTION. 71 
 
 L S 
 
 (70) These general expressions, ~^-j , 
 
 which represent, in all cases of distributed load, 
 the pressure in the middle section of an arch in 
 relation to the span and rise, are equal to the 
 area multiplied by the coefficients of resistance 
 of the material of which the arch is built. 
 
 (71) We represent this area by T and the 
 coefficient by C, and T C = resistance. 
 
 Now T = area of the section, and as we 
 always take 12" in length of the arch, or i foot, 
 in order that the load considered may be applied 
 only on each superficial foot all along the span, 
 
 T 
 
 — = thickness of the arch. 
 12 
 
 ^, TC LS T LS 
 
 Now = or — = 5 — — ^ or, 
 
 12 8 ;- 12 12 8 ;■ X 12 C 
 
 thickness = 
 
 8 r X 12 C 
 
 The expression +(V — (X+BA)) comes 
 from the well-known formula of Dejardin, for 
 tracing the equilibrium profile of the extrados 
 
 for vaults to V = X 
 
 COS. a 
 
 (72) If in addition to the expression 
 
 L S 
 
 , we put the thickness to reinforce 
 
 r X 12 C 
 
72 COHESIVE CONSTRUCTION. 
 
 the extrados, we must add to the thickness we 
 already have the difference between the radius 
 of the extrados in the centre, that is, X -f- A B 
 = f A (Fig. 1 8), and the radius vector V:= o N, 
 
 or+(V — (X + B A)), 
 
 or + ( — (X + B A) ). 
 
 ' ^ COS. a ^ ) 
 
 (73) Explanation of the formula for domes. 
 
 LS 
 
 ^^^ 8;- X 12C X 2' 
 
 (^)8y^rF^2+(v-(^x^«^))- 
 
 We must repeat here that wc do not pretend 
 to have an absolutely mathematical formula, 
 but one practical enough to give sufficient 
 security for safe construction. We are here 
 also considering the dome as not one of vous- 
 soirs, but as a simple cast dome working as a 
 single piece. 
 
 (74) The formula (5) is the same as that of 
 
 the barrel arches (i) with the difference that 
 
 (5) formula gives the thickness only and not the 
 
 area X T C ; area X coefficient is transformed 
 
 T C T C L S 
 
 into p^ and -p^ = 7; 7:; and as L 
 
 12 C 12 C 8 r X 12 C 
 
 represents a portion of the arch 12" in length, 
 
 ■ multiplied by the weight in lbs. and also by the 
 
 span, and as the surface of any symmetrical 
 
COHESIVE CONSTRUCTION. 
 
 73 
 
 portion of a dome is just half of the surface of 
 any symmetrical portion of a barrel arch of the 
 same radius and base, L in the dome will be 
 
 just half or — ; 
 
 hence in the dome ^ 8 r X 12 C, 
 
 12 C 
 
 LS 
 
 rX 12 C X 
 
 B 
 
 /<^ 
 
 \' 
 
 ^7^ 
 
 xA 
 
 \ 
 
 / /2^\ 
 
 /^\\ 
 
 \ 
 
 / /V\ 
 
 
 
 //X^2d\ 
 
 'T^-^-^ 
 
 ^ 
 
 ^^^^^ 
 
 ^-"-'^ 
 
 ^ 
 
 ^^^^^■~^^ 
 
 \^/'^/^ 
 
 // 
 
 Vv\\^ 
 
 \^ / / 
 
 / 
 
 \ \\. 16/ 
 
 nP/ 
 
 /iz 
 
 ,Aj^ 
 
 p 
 
 A 
 
 Fig 22. 
 
 (75) Domes in the Cohesive System cannot 
 be considered as a modiiication of the barrel 
 arch but as an absolutely different constructive 
 member for covering spaces. 
 
74 
 
 COHESIVE CONSTRUCTION. 
 
 {76) I. — The surface of a dome is just half 
 the surface of any barrel or segmental arch of 
 the same radius having for its length half of 
 the circumference of the base of this dome, with 
 the peculiarity that the surface is decreasing 
 in direct ratio that it approaches the crown. 
 In effect : 
 
 ■oD -Q- 
 
 Fig, 23. 
 
 Taking the plan of a dome (Fig. 22) and 
 dividing it into small radiating portions I, 2, 3, 
 4, etc., so that each portion can be considered 
 as the smallest expression that can be divided, 
 and afterwards supposing there are built with 
 these portions a segment of arch of the same 
 curve as the dome, but in portions (a^ in Fig. 
 23), having the base of half the portions of the 
 dome in a line B A, and the base of the other 
 
COHESIVE CONSTRUCTION. 
 
 75 
 
 half of the portions in C D, we shall have an 
 
 arch, but with openings X X' X" X'" 
 
 X12 x^s. We will notice that the surface i + 
 24 = Xio, then 2 + 23 = X^, 3 + 22 = X^, etc., 
 and that X + X' + X", etc., = surface (i + 24) 
 + (2 + 23) + 3 4- 22), etc.; but as surface X + 
 
 aT Tc 
 
 Fig. 24. 
 
 Fig. 25. 
 
 X'-|- X" are all open spaces and no load can be 
 
 considered, L of the expression 
 
 LS 
 
 LS 
 
 will be 
 
 il L_S 
 
 2S or 8 ;■ or 
 
 /-X 2 
 
 8 ;- 2 
 
 iT]) 2. — The material of a dome is not only 
 working by compression, but in consequence of 
 
76 COHESIVE CONSTRUCTION. 
 
 its form it is also working by tension, because 
 tlie tliriist depends upon the form and not on the 
 material. Suppose we take a lintel of stone, 
 as in Fig. 24 ; if we put it as in Fig. 24 
 we have practically no thrust. If we take 
 another lintel (Fig. 25), that is like the first 
 one ; but, taking off the material under the 
 curved line a, b, c (Fig. 25), we have some 
 thrust at once. We can see without any 
 demonstration, that in the second case we have 
 thrust and in the first one we have compara- 
 tively none. This does not mean that in the 
 first case it absolutely does not exist, because 
 inside of any lintel we must consider an arch 
 when it is working ; in the same way that in 
 any block of stone, or any block of marble, or 
 any block of wood, we know the most ideal 
 figure exists that the imagination can conceive. 
 The question is to take away the shell that en- 
 compasses it. In the same way, in any piece of 
 wood, or in any block of stone, we have an arch 
 better than any which the most e.xact mathema- 
 tician can define ; and, as soon as we put a 
 lintel to work, this imaginary arch is put into 
 action, and all of the material under this arch 
 is working to take off the thrust, because it is 
 the rod of this imaginary arch. Now, if we 
 take away this material (which is the condition 
 
COHESIVE CONSTRUCTION. 77 
 
 in the second case), as the material which is 
 working as a rod is not there, the arch is more 
 free for weight and thrust. 
 
 (78) But we have a third case, as in Fig. 26. 
 This is not a lintel where we take the lower 
 part of the material, forming, as in the rest, 
 an arch. It is a regular arch, formed by 
 pieces. It is not necessary to demonstrate that 
 this has thrust ; but in this case the thrust is 
 in full, if I may use the term. I mean in full 
 
 Fig. 26. 
 
 because in the second case, the stone lintel, be- 
 ing cut in a curve, is not as free as in the third ; 
 because, to get any signs of thrust, it is neces- 
 sary to break the lintel. Consequently, the 
 effect of the thrust begins when the cohesive 
 resistance of the stone is overcome; whereas, in 
 the third case, this commences at once, because 
 it is in several pieces and there is no cohesion. 
 Our barrel arches are working in the second case. 
 
78 COHESIVE CONSTRUCTION. 
 
 (79) Considering the case of domes, suppose 
 we have (as in Fig. 27) a dome composed of 
 voussoirs. This dome will have thrust, because 
 it is a modification of the segmental arch of 
 voussoirs (Fig. 26). 
 
 s(:'K--±rl%, 
 
 
 
 
 Fig. 27. 
 
 Now suppose we take (see Fig. 28) a big 
 block of stone, say ten feet long and ten feet 
 wide, and one foot or one foot six inches thick. 
 If we support that on the four sides just as a 
 lintel, we have practically no thrust, and if we 
 make a cavity on the under side (Fig. 29), 
 making a curve like a dome, we will have a 
 dome arch, but no thrust. It is not the second 
 case of the lintel, wbere, taking off the material 
 iAai is working as a rod, the thnist commences 
 
COHESIVE CONSTRUCTION. 79 
 
 to act. In the case of the dome it is not so, 
 because the material that is working as a rod, 
 
 ,^,'''- iliilfe 
 
 3l 
 
 / 
 
 - / 
 / 
 
 Fig. 28. 
 
 — OiU-jji' 
 
 
 ^:^ 
 
 ^ ^.^\:\^ \ 
 
 J ills ' 
 
 ' /iij , 
 
 / 
 
 Fig. 29. 
 
 or by tension, is formed into rings which remain 
 as I'ings. This is our case ; if we build the 
 
80 COHESIVE CONSTRUCTION. 
 
 ceilings in the form of domes, and if they are 
 well applied and properly built, we have, prac- 
 tically, no thrust whatever. 
 
 THEORY OF THE COHESIVE CONSTRUCTION, BASED 
 ON THE THEORY OF THE COHESIVE ELEMENTS. 
 
 Analysis of some peculiarities of the system. 
 
 (8q) Before proceeding further, I would like 
 to make a few preliminary remarks, calling at- 
 tention to some essential and peculiar principles 
 of the Cohesive System applied to the supported 
 members. They are not new as technical mat- 
 ters, but they are not taken into consideration 
 to-day in ordinary construction, and cannot be 
 explained by the voussoir theory. 
 
 Fig. 30. 
 
 Fig. 31. 
 
 In any ceiling composed of cohesive barrel 
 arches supported between beams, the less the 
 radius of the curve of the arch, the less the 
 beams have to bear, and consequently the lighter 
 the beams need be. 
 
 (8i) This means that a ceiling composed of 
 cohesive arches built as in Fig. 30 will support 
 
COHESIVE CONSTRUCTION. 81 
 
 more weight than one with the same beams 
 and spans as in Fig. 31, in which the arches 
 have more radius. This principle, at first 
 thought, seems absurd, in consequence of the 
 extra material used, because the arch has 
 greater surface and brings more weight to bear 
 on the beams. That this is true will be shown 
 hereafter. 
 
 Fig. 32. 
 
 Fig. 33. 
 
 S- — 
 
 Suppose Figs. 32 and 33 have an arch with 
 a radius at infinity, or flat, the neutral axis is 
 XX' transverse section. The material situated 
 below the neutral axis will work by tension, and 
 that above by compression, and having in our 
 case, say only about three courses, or say three 
 inches in thickness, the moment for tension will 
 be, one inch and a half, and for compression one 
 inch and a half. 
 
 (82) Suppose Fig. 34 to be an arch whose 
 rise is a foot. The central axis will be X* X^ 
 The part below it, working by tension, will have 
 six inches for its moment, and that above the 
 
82 COHESIVE CONSTRUCTION. 
 
 same. All this part, worked by tension and 
 compression, acts really as a second beam, tak- 
 ing part of the load off the iron beam.* To 
 
 Fig. 34. 
 
 find the extra strength that this form of arch 
 adds to the construction, of which it forms part, 
 — as it depends upon the tensile strength of the 
 material, which is 223 lbs. per square inch, — we 
 can apply the following formula, represented by 
 
 L = Load in pounds. 
 S = Span in feet. 
 ;-^ Rise in feet. 
 
 C = Coefficient 223 pounds breaking load. 
 T = Surface area in square inches. 
 8 r C T 
 
 * It can also be reinforced by means of a counter-arch turned 
 over the beam, which arch will take part of the load of the 
 floor and transfer it against the walls which are tied by the 
 said beams, so that the beams will act in connection with this 
 arch by tension. This disposition has the advantage of reduc- 
 ing the concreting, and in consequence the weight, over the 
 haunches of the arch and beams. 
 
COHESIVE CONSTRUCTION. 
 
 83 
 
 This will be the breaking load ; the beam 
 must be strong enough to add to the strength 
 of the arch sufificiently, so that it can never 
 work up to its breaking load. 
 
 In any ceiling composed of barrel arches con- 
 structed on the cohesive principle of construc- 
 tion, supported between beams, the ends of the 
 beams and the extremity ends of the barrel 
 arch receive the principal load and in conse- 
 quence the beams receive least weight in the 
 middle. 
 
 (83) Suppose Figures 35, 36 and 37 are barrel 
 arches built over two beams, the beams sup- 
 ported by two lateral walls. Let us consider 
 two diagonal arches, a, b, c, d. Any load uni- 
 formly distributed over the arches will affect 
 
84 
 
 COHESIVE CONSTRUCTION. 
 
 ^UHIII>l»/lll>!lt}!)l>mjtHH»t»t»»il»»tn»}iiimi^ii,^„intini,,,„i,}i,nnii>t 
 
 Fig. 36. 
 
 Fig. 37. 
 
 evenly every point of the skewback that is over 
 the flange of the beams ; but as the points a, 
 b, c, d are the most rigid ones, they will be the 
 first to sustain the weight of the arches, conse- 
 quently helping the beams,* and establishing the 
 lines of pressure a and b, c and d. (See Figs. 
 35 and 36.) Similar arches will be formed from 
 the point a to the opposite diagonal c, and from 
 
 *The a, b, c, d arch referred to in Fig. 35 will have on plan 
 the form of a parallelogram limited between the lines of 
 pressure A, O, C, and the similar lines of the next adjoining 
 arch (Figs. 35 and 37). 
 
COHESIVE CONSTRUCTION. 
 
 85 
 
 d to the opposite diagonal c, a tendency that 
 will also be similar to the principles mentioned 
 before. 
 
 (84) Fig. 38 represents a barrel arch broken 
 irregularly and diagonally in two. In prac- 
 tice, the timbrel arch of this form retains its 
 equilibrium, which cannot be explained by the 
 
 Fig. 38 
 
86 COHESIVE CONSTRUCTION. 
 
 voussoir theory. In the voussoirs that begin 
 in the line F, C, Fig. 39, the first ring F, D, has 
 two skewbacks, but the second ring d has but 
 one. The same is the case with rt" a"\ etc., and 
 according to the gravity system the arch is not 
 admissible — it is not safe. But in onr system 
 we know that this arch will stand, because we 
 see it stand every day. All we do is to reinforce 
 the point D, arching in radical form from the 
 point D to the opposite line F, C, or skewback 
 forming the coniform cohesive elements, F D a' , 
 a D a!', a!' D a"' , etc. (Fig. 39), and add it to 
 the other. Thus, the element /;, //, b" , etc. (Fig. 
 40), resting on the skewback C, F, will be ac- 
 cumulated at the point D. 
 
 If we construct in addition to this broken 
 barrel arcli (Fig. 38, plan Fig. 40) the portion 
 E, G, C (Figs. 38 and 40), we shall accumulate 
 the element b, b\ b», b'", b'\ b\ b''^ b''», half to the 
 point D and half with the point E, having a 
 perfectly safe arch, such as we are making in 
 the ground arches, and the sum of the cohesive 
 elements, along the line or skewback F, C, must 
 be equal to the sum of the elements accumu- 
 lated in the point D and the point E. 
 
 (85) It may be remarked that the empty space 
 E, G, D (Fig. 40), not having any load on it, it 
 seems that we can economize the elements 
 
COHESIVE CONSTRUCTION. 
 
 87 
 
 b, b\ b'\ b"\ b'", b", b"", iJVM. But we may notice 
 that the part of the elements radiating from F 
 to C are cut for the opening D, G, E, and they 
 must suffer a deviation pursuing the same devi- 
 
 b 
 
 bl 
 
 bll 
 
 bill 
 
 blV 
 
 bV 
 
 bVI 
 
 bVII 
 
 
 
 r 
 
 
 '----^-^^c----; 
 
 
 bVlllL^r:;:::::::., 
 
 a.11 
 
 2X111 
 
 d.lV 
 
 d.VI 
 d.Vll 
 
 3.VIII 
 
 Fig. 40, 
 
 ation as the elements b, b\ b^\ etc., when they 
 are directed to the point D and E. 
 
 (86) Any of these examples cannot be ex- 
 plained by the voussoir theory and gravity 
 
88 COHESIVE CONSTRUCTION. 
 
 system, and I suppose they are sufficient to 
 illustrate in this essay the incompatibility of 
 these theories with the results of actual expe- 
 rience. I think it is more rational and simple 
 to explain these applications by the cohesive 
 strength of the materials used, and in conse- 
 quence why not explain the theory of the cohe- 
 sive arches by the theory of cohesive elements? 
 
 If we build Figure 41 of stones placed one on 
 top of the other, anchoring the base, the force 
 of gravity alone keeps them in place, and any 
 horizontal pressure, P, must be resisted by the 
 weight of each piece, there being no other 
 force which can help the full construction. 
 
 (87) To illustrate this 
 
 let P = the horizontal pressure 
 
 r = the distance of its point of appli- 
 cation above the joint a a 
 W = total weight above a 
 T = the breadth oi a a' ; 
 then taking moments about a' we get the 
 equation 
 
 T 
 
 P /- = W — 
 2 
 
 (88) But if we construct the figure of cohesive 
 material (Fig. 42), the horizontal pressure, P, 
 will be not only the gravity, but also the cohe- 
 sive strength, of the material to overcome. 
 
COHESIVE CONSTRUCTION. 
 
 89 
 
 This refers to vertical construction on a solid 
 base. 
 
 Making the same investigation with regard to 
 Figure 42 as we did above, we must take into 
 account the cohesive strength of the material 
 
 Fig. 41. Fig. 42. 
 
 which resists tension at a and compression at 
 a!. If the strain developed at each of these 
 
 points = C, then Pr=W-+CT. 
 
90 
 
 COHESIVE CONSTRUCTION. 
 
 (89) For a horizontal construction in space, 
 if we suppose a structure like Figure 43, built 
 with voussoirs, we can readily see that its con- 
 struction would be impossible, as the joints 
 
 Fig. 43. 
 
 have no cohesive strength, and its weight would 
 make it fall at once. But with cohesive mate- 
 rial we can build a structure like Figure 44 in 
 such a way that the cohesive strength of each 
 particle of the structure will resist the force of 
 
COHESIVE CONSTRUCTION. 91 
 
 gravity acting on that particle. To analyze this 
 construction let S = span = B a' and then, as 
 the centre of gravity is nearly one-third of the 
 span from a horizontal distance from the centre 
 of gravity to the section a a', let W = weight 
 of the structure, which in this case tends to pro- 
 duce rupture at a. Taking moments as before, 
 
 S 
 we obtain the expression W — :^ C T. 
 
 3 
 
 In this case we see that the expression 
 
 T 
 
 W X - disappears, but we have the force re- 
 sisting rupture, that is, the cohesive strength of 
 
 s 
 
 the material, or W — = C T. 
 3 
 
 (90) In Figure 41, representing the Gravity 
 System, we build with blocks from the quarry 
 having cohesive strength in themselves, but 
 with no cohesion between them, acting only by 
 gravity. 
 
 (91) In the second case (Fig. 42), we also 
 use material having life in itself, making it 
 into a homogeneous structure, having the same 
 properties throughout as in a monolith, thus 
 imitating the action of nature in the quarry. 
 
 (92) The third case cannot be considered, 
 because such a form cannot be constructed 
 
92 COHESIVE CONSTRUCTION. 
 
 with voussoirs on the gravity system ; and in 
 the fourth we have the plain work of cohesion. 
 
 (93) Comparing these four figures, we see the 
 advantage of working with live material ; that 
 is, with material which has in itself properties 
 of cohesive strength. And we may observe that 
 the minimum strength will be when the material 
 is working only by deflection, or when not all of 
 the section is working by pressure, or else when 
 the position is horizontal (Fig. 44), that is, when 
 ;-^ O. And the maximum will be when the 
 material is working principally by pressure or 
 when 1-^ infinity, that is, when the sides of an 
 arch are practically two walls. 
 
 (94) But at the moment that r (in Fig. 44) 
 commences to have a finite value we will have 
 both tension and compression acting to main- 
 tain stability, and the greater the value of r 
 the more will the compression act and the less 
 the tension be. Consequently we must have 
 two coefficients, one that we can call C for 
 compression and another C for cross-section 
 or modulus of rupture. The formula for the 
 first is : 
 
 LS ^ 
 
 T C ^ and for the second T C ^ 6 . 
 
 2 
 
 r 
 
COHESIVE CONSTRUCTION. 93 
 
 s 
 
 The expression -z. results from the following 
 
 s 
 
 (Fig. 42) : S = the span and the — = \ the 
 
 span, but as we are considering only the weight 
 of the material, we know that its point of appli- 
 
 cation = \ of or -..* 
 2 6 
 
 * I shall not go Into further details about this theory and its 
 universal application in new ideas of construction because it 
 is not within the limits of this preliminary and small book. 
 
PART IV. 
 
 MODERN APPLICATIONS AND ARTISTIC OR 
 ^ESTHETIC IMPORTANCE OF THE COHESIVE 
 CONSTRUCTION. 
 
 (95) It is said, and generally understood, that 
 the art of construction in masonry is as yet in 
 its infancy, and that we are not out of its first 
 rudiments. Perhaps these ideas are based upon 
 the fact that the technical data we possess to-day 
 for this kind of construction, which is being 
 generally taught in our colleges, are little ad- 
 vanced beyond those which were employed in 
 constructing the edifices of 1000 or 2000 years 
 B. C. ; that is, those used by the Egyptians and 
 others. 
 
 (96) It is true that, with the arrival of the 
 romantic epoch, that is, when the architectural 
 art, having gained consciousness of its indepen- 
 dence, was reproducing to infinity its architect- 
 ural theme, we find works that we yet admire 
 as models of construction and art, including 
 those outside of the construction depending on 
 gravity ; but none of the constructive sciences 
 
COHESIVE CONSTRUCTION. 95 
 
 admit as yet that they are but plain general 
 construction, and but few architects are disposed 
 to risk their signatures to them, and those that 
 do only admit the outside lines. Why.' sup- 
 pose an architect intends to build a structure 
 with a combination of domes, as in either the 
 cathedrals of Santa Sophia, in Constantinople, 
 or Zamora, in Spain, and sends plans of it to 
 the Building Department for approval in one of 
 our large cities. He will find it a most difficult 
 matter to obtain a permit to build this structure, 
 and in consequence he will have to make an 
 imitation of the outside and inside artistic lines 
 by a false construction. What does this mean.' 
 Arc we progressing, or is our knowledge in- 
 ferior to that of the Middle Ages .' 
 
 (97) Perhaps this can be explained in the fol- 
 lowing manner: — The builders of the romantic 
 epoch, when they were building, were making 
 architecture; they were builders, architects; 
 they were making plans which they themselves 
 would carry out ; they were, in a word, builders 
 and architects, and could not be in any other 
 way, because those innumerable, great, archi- 
 tectural conceptions of the Middle Ages were 
 not possible to be realized, unless the same 
 genius who designed should build them. So 
 that to-day we do not know which to admire 
 
96 COHESIVE CONSTRUCTION. 
 
 in those monuments most, — the architectural 
 lines or the successful constructive problems 
 solved. 
 
 (98) It is true that they were not restricted 
 by any building laws, being the builders and 
 architects both, or vice-versa. And their full 
 liberty in building was a great advantage to the 
 architectural art in two ways : it was advanta- 
 geous to the progress of construction, and also to 
 the artistic side, because the work of an architect 
 should not be the plans alone, but the building 
 itself ; the plans are the project and the work 
 is the building. Is it then strange that the 
 history of art admits that one of the most 
 brilliant epochs of originality in architecture 
 was in the Middle Ages 1 In their dwelling- 
 houses, palaces, city-halls, chambers of com- 
 merce, churches, convents, fortresses and cathe- 
 drals we cannot surpass their work, either in 
 beauty or engineering skill. We cannot say 
 that of modern times, with the exception of our 
 great iron bridges, depots, etc. 
 
 (99) The architect of the present epoch, 
 especially in this country, for reasons that can- 
 not be discussed here, generally makes plans 
 which are to be built by some one else, and the 
 constructive parts given by the building laws 
 are official data. It is not necessary for 
 
COHESIVE CONSTRUCTION. 97 
 
 him to think to any great extent of problems of 
 construction, and he loses in consequence, day 
 by day, consciousness of the fact that he is a 
 real architect as was formerly understood by 
 this generic word. 
 
 (lOo) On the other hand, the builder, that is, 
 the one who carries out the plans of the archi- 
 tect, is not only, by his character of a simple 
 builder, forbidden to understand anything of 
 the^architect's part, but he cannot modify any 
 part of the constructive problem ; not only 
 as a matter of financial policy to himself, 
 but also from the fact that he is not, as a rule, 
 a man of study, having only some ideas of 
 materials, the general trades, and scaffolding, 
 and enough knowledge, perhaps, for estimating. 
 
 (lOi) The result of this anomalous combina- 
 tion is a very strange one. In the first place, 
 the feeling is, that the work of an architect 
 is to make the plans, and take the direction 
 of the artistic part, and also that the prob- 
 lem of construction must be on a level with 
 the general knowledge of the contractors ; and 
 the architect, in order to be free from difficul- 
 ties, makes the specifications and general 
 conditions of construction in accordance with 
 the official formulas, and according to the intel- 
 lectual abilities of the builders. Now, this 
 
98 COHESIVE CONSTRUCTION. 
 
 being the general modus operandi in nearly all 
 kinds of building, it is evident that, although the 
 so-called artistic part may be in some cases 
 highly studied, if not developed, by the architect 
 as a designer, the constructive progress, the 
 real architectural progress, is obstructed by two 
 fortresses ; namely, building laws, and the con- 
 tractor's financial policy. In consequence, it 
 may be thought, with some sort of reason, that, 
 for making plans for a building, it is not neces- 
 sary to learn construction. 
 
 (102) This puts me in mind of the following 
 anecdote : Rossini one day asked his Professor 
 if it was necessary to learn counterpoint in 
 order to learn opera composition ; and the 
 Professor, thinking only of the modern Italian 
 opera, answered, "no." 
 
 It is to be hoped that some radical change 
 will follow the present state of things, in order 
 to assist in redeeming the art from these igno- 
 ble and servile conditions. 
 
 APPLICATIONS. 
 
 (103) In speaking of the " Cohesive Sys- 
 tem" applied to "Timbrel Arches," of which 
 this book specially treats, I am aware that, 
 even to the initiated in the science of construc- 
 tion, it might occur that this system can only be 
 
COHESIVE CONSTRUCTION. 99 
 
 applied to arches, as it has been applied in the 
 construction of the arching floors of the new 
 Boston Library. This system is not confined 
 to the specialty of the vault, neither is it placed 
 in exclusive competition with brick arches, but 
 it consists of a complete system of construction, 
 including walls, floors, roofs, ceilings, partitions, 
 staircases, columns, etc. 
 
 (104) The great and surprising advantage of 
 the "Timbrel Vault" over the brick arch will 
 be found equally in walls, roofs, partitions, etc. 
 
 FLOORS AND ROOFS. 
 HOW THE MATERIALS WORK IN THEM. 
 
 (105) It is evident that the use of light mate- 
 rials of equal strength, each material used in 
 the way required by its nature, is the basis of 
 building economy. If we put the wood or iron 
 to work under deflection, or submit them to 
 transverse pressure, surely we shall need more 
 material than in using the same wood or iron 
 under tension. If we place them in this posi- 
 tion we will have an economy. The same is 
 true when we put clay and cement to work by 
 pressure ; they then have their greatest effi- 
 ciency, and can replace iron and wood with 
 economy. In a floor in the ordinary system 
 
100 COHESIVE CONSTRUCTION. 
 
 we find wooden or iron beams, and between, 
 wooden boards or brick arches. What are the 
 wooden boards and brick arches doing ? Only 
 bridging between the girders or beams. All the 
 material between is not working at all ; the total 
 weights are supported only by the beams or 
 girders, and these bridges contribute only to 
 the weight. But in the Cohesive System, if 
 well applied, every piece of material is working 
 directly, and just as is necessary; the clay 
 works to support itself, working by pressure, 
 and the iron works as a rod. That is the 
 great economy. All will admit that, with these 
 conditions, it is not strange that this system, 
 although the material itself is dearer, can com- 
 pete with advantage with any system. 
 
 (io6) For many reasons the ceiling or roof 
 is the most difficult part of the project of any 
 building ; and to this the architect generally 
 gives his attention, knowing that any excess or 
 default of material in the ceiling or roof affects 
 not only its stability, strength and economy, 
 but also the walls, columns and foundations — 
 that is, the sustaining parts. 
 
 (107) The structure of a ceiling is based upon 
 the principles of a bridge ; that is, upon the 
 principles of any sustained architectural mem- 
 ber. Nature has given us two fundamental 
 
COHESIVE CONSTRUCTION. 101 
 
 forms of it : First, the suspended form ; second, 
 the arch form, or vault. A third form, nature 
 gives us accidentally, that is, the lintel. 
 
 (io8) If we analyze these three different ge- 
 neric forms of bridging, we will find that the 
 first has a tendency to pull in the sustaining 
 parts, and the second to push them out. The 
 strength of both tendencies is equal. In the 
 first the material works by tension, and in the 
 second by pressure. For the first one, nature 
 always uses fibrous material ; for instance, as in 
 the vine whose branches form a bridge which is 
 suspended across a river. For the second, na- 
 ture uses particles, connected together in cohe- 
 sive work. 
 
 (109) These two opposite tendencies are 
 united in the third form, which nature has also 
 given to us ; that is, the lintel, which is sub- 
 jected at once to both tendencies. But, in the 
 lintel, both tendencies are connected together, 
 that is, the cohesive material which is working 
 by pressure, and the fibrous material that is 
 working by tension. Both thus form a united 
 mass, and the result is that the part working by 
 pressure, which is generally on top, is pressing 
 (in virtue of its intimate connection) the other 
 part that is working by tension, compelling to 
 work by pressure that part which was designed 
 
102 COHESIVE CONSTRUCTION. 
 
 to work only by tension, which is generally at 
 the bottom. . The result is that nearly the whole 
 mass shown in the section of a lintel has a ten- 
 dency to work practically by deflection. Now 
 if to this inconvenience we add that no material 
 has the same coefficient for compression as for 
 tension, the difficulty of securing economy in a 
 lintel or beam is serious, unless they are sep- 
 arated, and the two materials work free and 
 independent of each other, in which case they 
 will not lose any strength. That is the reason 
 that no lintel or beam can be worked alone with 
 economy. A lintel and beam united can be 
 worked with economy when both materials are 
 independent and free to work as is required. 
 But this structure is not a lintel — is not a 
 beam. It is just the combination that we gave 
 in the beginning, that is, the suspended form 
 whose tendency is to pull in the sustaining 
 members or walls, working as tension, and the 
 arch form whose tendency is to push out the 
 sustaining members or walls, working by pres- 
 sure. These two primal forms, when they are 
 connected properly, give the most economical 
 and easy ways to bridge. But in order that 
 these compound bridge forms should work 
 properly, they must have a ver\' flat section, so 
 that all the fibres of the rod may work practi- 
 
COHESIVE CONSTRUCTION. 
 
 103 
 
 cally by tension. One ex- 
 ample of this application is 
 a floor constructed with a 
 continuous barrel arch (A), 
 with some small bent rods 
 (B), (Figs. 45 and 46), with 
 a small partition laid over 
 the rod B, rising to the 
 barrel arch. This arrange- 
 ment will form a contin- 
 uous beam in which the 
 barrel arch, and part of the 
 partitions will work by 
 pressure, while the rods 
 will work by tension. The 
 material in this position 
 works with its true force. 
 
 (no) The rod can be 
 calculated exactly, and in- 
 dependent of the section 
 of the arch, and vice versa, 
 with the additional feature 
 that each material works 
 at its best advantage, and 
 the clay or arch is at the 
 same time supporting the 
 ceiling or roof, and forming 
 a floor or roof, the economy 
 
 
 ^^;^ 
 
 CQ 
 
 tJ) 
 
 OQ 
 
 y/////^///- 
 
 
104 
 
 COHESIVE CONSTRUCTION. 
 
 of which is evident. This is not the case with 
 beams laid parallel next to each other, and 
 wooden planks or brick arches, etc., between, in 
 which the beams are the real support, and the 
 wooden planks and brick arches are only for a 
 bridge between, adding weight. Here the virtual 
 beam or lintel is a continuous one. 
 
 Fig. 46. 
 
 (in) For this reason, under the cohesive sys- 
 tem of construction, if properly, applied, the 
 clay works by pressure, supporting itself, and 
 the iron works only as a rod. If the system is 
 not properly applied, or is misunderstood, no 
 advantage can be obtained. This is the case 
 when girders or beams and barrel arches are 
 combined, as we have already stated. The arch 
 is intended for bridging between girders, and the 
 girders not only work by deflection or dead- 
 weight, which is the most unfavorable condi- 
 tion, but the barrel arch and the weights on it 
 are added to that of the girder, and in conse- 
 quence the clay does not contribute in the least 
 to support any part of the floor, because it 
 
COHESIVE CONSTRUCTION. 105 
 
 depends on the girder. Consequently it must 
 be heavy and expensive. Hence the construc- 
 tion shown in Fig. 45 will always be a supe- 
 rior, more economical and more rational con- 
 struction. 
 
 (112) Better advantages must arise (see Figs. 
 47 and 48) when the rod is in a circular form, 
 making a continuous rod, and the barrel arch is 
 changed to the form of a dome. In that case 
 we have the following advantages : 
 
 1. That the material which is working by 
 pressure, namely, the clay, is not only working, 
 as we stated in the theoretical part, by pressure, 
 but also by tension, when the form of a dome 
 is assumed. 
 
 2. One rod is enough, set just at the begin- 
 ning of the dome in the form of a ring, B, B', 
 B ' (Figs. 47 and 48), and as the material itself 
 is working by tension, counterbalancing in some 
 part the pressure of the dome, this ring or cir- 
 cular rod can be of a smaller section than for 
 any barrel arch of the same surface. 
 
 3. The circumstance that the rod is in the 
 form of a ring, at the base of the dome, gives 
 the great advantage that the iron is better situ- 
 ated to prevent its being heated in case of fire. 
 
 4. The form of the dome, which does not re- 
 quire any rod between the skewbacks, or springs 
 
106 COHESIVE CONSTRUCTION. 
 
 Fig. 48. 
 
COHESIVE CONSTRUCTION. 107 
 
 of the arches, leaves the dome free from any 
 interference, presenting better finished work, 
 and a noble surface for presenting decoration, 
 being itself the principal member of the decora- 
 tion, by its form if not by its material. 
 
 5. This continuous form of the dome, with- 
 out any interference of rods, by its smooth, 
 curved surfaces, affords also the best inside 
 lines for any ceiling, from the point of view of 
 hygiene and ventilation. 
 
 HOLLOW TIMBREL ARCHES FOR FLOORS AND 
 ROOFS. 
 
 (113) In roofs, and also in ceilings where 
 decoration is required, the single timbrel arch 
 has the inconvenience that the roof transmits 
 the temperature, and is cool in winter and very 
 warm in summer ; and in ceilings for decora- 
 tion, if concrete is used for filling, the con- 
 densation of the interior atmosphere causes 
 dampness in the ceiling, changing the colors 
 and destroying the decorations. To avoid this 
 it is necessary to build the ceiling and roof 
 hollow, that is, composed of two series of tim- 
 brel arches, one on the underside, as a ceiling, 
 adding another on top, as a floor or a roof, 
 by means of a rib built between both timbrel 
 arches. The ways to obtain these forms, which 
 
108 COHESIVE CONSTRUCTION. 
 
 we may term "tubular timbrel ceilings or roofs," 
 are infinite. I would here refer to the following 
 examples in this country : — The domed library 
 ceiling of the Arion Club, 59th Street and 
 Park Avenue ; the floor of the extension to the 
 Young Women's Christian Association's build- 
 ing over the boiler-room, on i6th Street, near 
 Fifth Avenue, New York ; the main staircase 
 of the Public Library, Boston ; the large ceiling 
 of the driveway for the same building, and 
 others. 
 
 (114) This hollow arch, or double arch with a 
 separator, has not only an advantage over ribs 
 as an insulator from dampness and heat, but it 
 also increases the strength of the arch, as it 
 adds to the strength of a tubular girder when 
 the moment or radius of gyration is increased. 
 
 (115) For that reason, in any arch that requires 
 more than four courses, it is better, if the space 
 is not limited, to build always in a "tubular" 
 way, that is, with two courses in the bottom, 
 and ribs over as a separator of one, two, three 
 or four inches, according to the space at our 
 disposal, and two courses on top. Again, this 
 "tubular" construction has the advantage that 
 the section of the arch can be increased at will 
 in the haunches, or at any place where the 
 arches require thickness, thus increasing the 
 
COHESIVE CONSTRUCTION. 109 
 
 radius of gyration at a very small cost. Some 
 floors can be built hollow by means of building 
 two series of arches between the beams. One 
 example of this is the parquet floor of the Car- 
 negie Music Hall. The space between the two 
 series of arches was for ventilation purposes, 
 having two courses in the bottom, and three 
 courses on top, between the girders. 
 
 HOLLOW COHESIVE WALLS. 
 
 (ii6) Although cohesive walls hardly belong 
 to the subject of which we are treating, they are 
 so connected with the construction of arches, on 
 account of their being the support and forming 
 part of the skewbacks of the arch, that they 
 require some mention, especially hollow walls. 
 The enormous thickness given to the walls of 
 the present buildings of ten, twelve and more 
 stories, does not seem justified by the present 
 advancement in the character of material ; but 
 it is justified if we take into consideration that 
 in these walls materials are generally used that 
 require several months to set, not only because 
 of the slow-setting nature of the mortar used, 
 but also because it requires the presence of air 
 for its transformation as an element for setting. 
 
 (117) With this kind of mortar, all the walls 
 built with rapidity to a great height cannot be 
 
110 COHESIVE CONSTRUCTION. 
 
 guaranteed, because the mortar receives more 
 weight than it can support without destroying 
 its condition of setting. Why do we not build 
 these tremendously high walls with Portland 
 cement .' As the strength of this material, 
 when set, is similar to the brick, and as the 
 Portland cement has the condition that it does 
 not require to be exposed to the air for setting, 
 the walls could then be built with rapidity with- 
 out compromising the setting conditions of the 
 mortar. 
 
 (ii8) It is true that in this case there is no 
 necessity for a wall of such great thickness, 
 and the radius of gyration for the wall can be 
 reduced ; but that gives the idea of building 
 these walls hollow, that is, of giving to the wall 
 the surface required for its strength, yet in- 
 creasing the radius of gyration, by building the 
 wall with a separation in the' middle ; which 
 means, to make two walls connected together 
 in order to give the same moment as in the 
 thick wall mentioned above, but with less ma- 
 terial. In this way we secure the valuable 
 condition of isolating the inside wall from the 
 outside, which is one of the most important 
 problems of the modern science of construction. 
 
 We have now arrived at the conclusion that 
 hollow walls are a necessity in modern buildings. 
 
COHESIVE CONSTRUCTION. Ill 
 
 (119) Some remarks are necessary here in 
 regard to these hollow walls. First, the full 
 dimensions of the walls would be calculated in 
 the same way as we are calculating now for any 
 hollow bodies, and that is not only to justify 
 the mechanical conditions of the wall, but also 
 to give it the proper artistic effect. The hollow 
 wall works exactly as a hollow column. If we 
 take the superficial inches of ' any cast-iron 
 column, of round or square section, and put 
 all this mass of section solid, not only will the 
 column be of less strength than in the other 
 case when it is hollow, but we also find at a 
 glance that its effect is meagre, poor and with- 
 out any character. The same is the case in 
 any wall. If we build any high wall without 
 giving it the right mechanical dimensions, hol- 
 low or not, the effect will also be poor and it 
 will be without character. Consequently these 
 hollow walls would be regulated by the same 
 principles that we have to apply to find the 
 radius of gyration of any wall, giving to the hol- 
 low space the balance of the thickness necessary 
 to the outside and inside walls, according to the 
 coefficient of pressure of the material used for 
 the same. In that way we secure, first, the 
 exact quantity of material necessary as a sup- 
 port, according to the progress of the manu- 
 
112 COHESIVE CONSTRUCTION. 
 
 facture of constructive materials ; second, we 
 secure the same result of increasing the radius 
 of gyration without increasing the cost of the 
 wall, and the weight upon the foundation ; and 
 third, to have a hollow space always convenient 
 as an isolater for other applications. 
 
 (120) Second, there is a hygienic advantage 
 in the hollow walls, which secure absolute 
 absence of dampness and condensation, thus 
 making the building cooler in summer and 
 warmer in winter. If the floors are also hollow, 
 the hollow walls permit the building to be venti- 
 lated in the corners of the rooms where the im- 
 pure air collects, and permit a greater section of 
 ventilation than can be had in any other system 
 without affecting the solidity of the building. 
 A source of great annoyance to architects when 
 a building requires a great number of flues, is 
 that they have difficulty in finding places on 
 which to lay the beams, making headers after 
 headers, sometimes resting half a door against 
 the head of a single beam. The brain of the 
 architect is thus continually racked to make 
 suitable framing plans. But if all the walls and 
 floors are tubular, every part of the room can be 
 well ventilated in every direction.* 
 
 * Although this is outside of our object, it may be remarked 
 that these tubular walls and tubular ceilings also facilitate the 
 
COHESIVE CONSTRUCTION. 113 
 
 FACTORIES. 
 
 ( 1 2 1 ) There is a country which, like New Eng- 
 land, breathes an industrial activity in cottons, 
 woollens, silks, laces, etc.; a country which, not 
 only because it supplies its own market, but be- 
 cause of its export trade, more especially to 
 South America, East India and North America, 
 is known as the " Spanish Manchester." I refer 
 to Barcelona (Catalonia, Spain), a country where 
 I had for fifteen years my circle of operations as 
 architect and specialist in this kind of building. 
 Hence I can say, without ostentation, that I am 
 not talking about that which is new to me. All 
 that I shall do is to explain what has been done 
 in the last twenty or twenty-five years. The 
 Catalonians now have nearly seventy-five per 
 cent, of their industrial buildings fire-proof. 
 
 How was this done.'' Was it because these 
 industries were backed by larger capital, and 
 
 establishing of a system of pipes, to connect every corner of 
 the room with the fire-box or grate of the boiler-fires, ranges or 
 furnaces, in order to burn the air, passing it through the flames^ 
 which is a good way to transform impure air. Such an appli- 
 ance as this can be used for the ventilation of college buildings, 
 and those who are further interested in this matter may be re- 
 ferred to plans and treatises sent by the author to the Philadel- 
 phia Centennial, on "Improving the Healthfulness of Indus- 
 trial Towns,'' for which a certificate of award was granted, 
 signed by your distinguished fellow-citizen, General Walker, 
 as Chief of the Bureau of Award. 
 
114 COHESIVE CONSTRUCTION. 
 
 had better facilities, or made more profits 
 than those of America? What is the reason 
 that the Catalonian manufacturers have ceased 
 making other than fire-proof buildings? Why 
 do not the American manufacturers do the 
 same? These questions 1 wish to analyze and 
 answer. 
 
 (122) As far back as 1865 it may be said that 
 neither in Barcelona nor in the provinces of 
 Catalonia was there a single factory that was not 
 entirely constructed, as here, of wooden floors, 
 and most of them with wooden columns and 
 girders; they also had, to some extent, the 
 "slow-burning" construction — a combination 
 of wood and iron. I remember when all the 
 factories in the streets of Amalia, Rierretta, 
 Luna, and the districts of San Pablo and San 
 Pedro, were of wood. So old and so saturated 
 with oil were these buildings, that the smell 
 arising from them was ominous of immediate 
 danger, perhaps caused by the knowledge of 
 the scene of horror which would ensue should 
 there be the least neglect, by fire. Nearly all 
 of these buildings have disappeared. Their 
 owners, with wise judgment, have built their 
 factories less closely together and on another 
 system, better suited to their interest — that is, 
 of fire -proof construction. 
 
COHESIVE CONSTRUCTION. 115 
 
 (123) Some may imagine that the manufact- 
 urers of these places were richer or safer in 
 their investments, thus requiring more perma- 
 nent and safer buildings; but this is not so, as 
 they are afforded very little protection, their 
 markets being almost open to the French and 
 English manufacturers, and any slight change 
 of the tariff threatens immediate danger. What 
 was the cause of this change ? 
 
 (124) These manufacturers had learned from 
 experience, that, notwithstanding the insurance 
 money paid, in case of fire, the factory must be 
 stopped for a period, and that customers, if not 
 supplied, will go to some competing firm; so 
 that, when they get started, they practically 
 begin anew. Tlicy also know that the wear and 
 tear and depreciation in a wooden building in 
 five years equals the extra cost of a fire-proof 
 one. A factory costing ^20,000, requiring to 
 be rebuilt in twenty years, means that five per 
 cent, per annum must be laid away to restore 
 it from time to time. This means that in five 
 years the factory costs ^25,000, or as much as 
 a fire-proof one. All these considerations, 
 and the increasing exigencies of the fire in- 
 surance companies rendering them more care- 
 ful in the issuing of policies, compelled the 
 manufacturers to open their eyes to the value 
 
116 COHESIVE CONSTRUCTION. 
 
 of fire-proofing; and under these pressing ne- 
 cessities some of them decided to build their 
 factories fire-proof. 
 
 (125) But twenty-five years ago the Catalo- 
 nian manufacturers were in the same position 
 as those here. They knew only of the English 
 fire-proof factories of iron and small brick arches, 
 expensive and not fire-proof, with the difficul- 
 ties connected with this system regarding the 
 running of shafting, in which any of the fre- 
 quent alterations required in every factory is an 
 enormous task. Realizing this inconvenience, 
 the manufacturers of Catalonia did not at a 
 single bound go from one extreme to the other 
 — that is, from the poorest to the most expen- 
 sive ; they took precautions so that the sudden 
 change might not ruin them, and commenced to 
 study the way to overcome the inconveniences 
 of the English system of fire-proof industrial 
 buildings, and find such a system as was fitted 
 for the manufacturers' convenience, and avoid 
 the heavy losses through stoppage, which the 
 insurance companies could not pay them. No 
 one can blame this prudence, for the capital 
 invested in the factory is the manufacturers' 
 tools, like the hammer and chisel in the hands 
 of the workman ; they cannot, they must not, 
 employ more than just what is necessary, with 
 
COHESIVE CONSTRUCTION. 117 
 
 the least chance of loss. And because of their 
 prudence they now have, as I said, more than 
 seventy-five per cent, of their buildings really 
 fire-proof. 
 
 (126) This system of fire-proofing was a com- 
 bination of clay and wood or clay and iron, with 
 ten feet six inches or more span, according to 
 the bays. I must remark that, for every ten 
 fire-proof factories, eight were constructed of 
 clay and wood, and two of clay and iron. The 
 adopting of this combination of wood and clay, 
 in preference to iron and clay, was not so much 
 due to economy as to the belief that in the way 
 applied it was more fire-proof, besides being 
 more adaptable and convenient for changes and 
 alterations, reducing wood, however, to the 
 lowest quantity, and putting it under absolutely 
 safe conditions. One of these was the firm of 
 Batllo Bros., for whom I drew the first factory 
 plans in this way in Sarria. This was so emi- 
 nently successful that several others were at 
 once erected on the same principle. 
 
 Now, if these men had not been prudent in 
 this matter, and, instead of making this com- 
 bination had been extravagant in adopting the 
 heavy iron and heavy clay as in the English 
 system, they would have secured improvements 
 in name, but not in fact ; for cash is the key to 
 
118 COHESIVE CONSTRUCTION. 
 
 the situation, and, although the manufacturer 
 has noble aspirations for progress and improve- 
 ment, he knows that he must maintain pruden- 
 tial limits. I have enlarged on this point, on 
 account of its importance. 
 
 (127) Thus was established in Catalonia two 
 systems of fire-proof factories : the one type 
 shown in the factory of Vidal Hijos, constructed 
 in 1871 ; the spindle building of Batllo Bros., in 
 1869, and several others ; the other tpye shown 
 in the loom-room of Batllo Bros., the woollen fac- 
 tory of Carreras, etc. The first type consists of 
 wooden girders and tile arches, and the second 
 of tile arches for ribs, with small iron beams 
 with dome between. (See Plates i and 2, at 
 the end of this book.) 
 
 (128) What is the combination — iron and 
 clay, or wood and clay .'' It is very simple, it is 
 only iron or wooden girders, set apart over col- 
 umns at the regular distance of a factory bay, 
 ten feet six inches or more, as has been stated, 
 and between them tile arches, similar to those 
 now to be seen in the Boston Public Library, 
 or the Harcourt Building, or Exeter Chambers, 
 in the same city. 
 
 (129) The combination with iron can be exe- 
 cuted in two ways, as it is used in some of the 
 rooms of the library, acting as a beam or girder, 
 
COHESIVE CONSTRUCTION. 119 
 
 working by deflection ; or, as used in the Colo- 
 rado Telephone Company's Building, Denver, 
 and the Young Womeirs Christian Association 
 Building, New York, working principally by 
 tension. This latter form may also be seen in 
 some of the rooms of the Boston Library. 
 When the combination is of wood, it is work- 
 ing in deflection, as in the case of iron, and 
 in some cases by tension. The construction 
 is cheapest in both materials when working by 
 tension. 
 
 (130) Some may think that fire-proofing in 
 Catalonia is cheaper than here. This, in a gen- 
 eral sense, is not so; because, had they accepted 
 the English system, the relation between the 
 English system and the cohesive system would 
 have been the same there as here. It may be 
 supposed that the cohesive system is dearer 
 here than there. I will show wherein the dif- 
 ference lies. It cannot be in labor, as the same 
 proportion of difference exists in the wages of 
 carpenters for wooden construction as for ma- 
 sons ; and, as the walls are the same in either 
 case, the difference is only in the floors ; and, 
 if any there be, it must be in the material. 
 Wood costs the same there as here. Portland 
 cement, one of the main factors in the construc- 
 tion, costs three dollars a barrel in Spain, against 
 
120 COHESIVE CONSTRUCTION. 
 
 two and one-half dollars here. Plaster costs 
 about the same. The difference is only in the 
 cost of the tile, which in Spain can be purchased 
 at five dollars per thousand, while costing fifteen 
 dollars here ; but taking into consideration the 
 fact that in Spain they use tile five-eighths ot 
 an inch in thickness and here one inch, and 
 that the tile is only one of the components, and 
 that it is only in this special material that the 
 cost is greater, the real difference in the price 
 per foot of that material is only twenty-five per 
 cent. This twenty-five per cent, difference in 
 the cost of tile cannot be sufficient reason for 
 not using the same construction here; for, in a 
 factory one hundred by one hundred feet, or ten 
 thousand feet square, the difference in the floors 
 would only be eight hundred dollars. I do not 
 think this eight hundred dollars would prove an 
 insurmountable obstacle to a New England 
 manufacturer; therefore increased cost cannot 
 be the cause. In the iron there is a little dis- 
 proportion ; but, as the economic state of a coun- 
 try is relative in all things, if the iron and con- 
 struction are cheaper, the production is also 
 cheaper and the income and interest less, so 
 that does not effect the comparison. Conse- 
 quently, the difference in cost there and here 
 between the wooden and fire-proof factories is 
 
COHESIVE CONSTRUCTION. 121 
 
 really the same, and the reason for not adopting 
 the same system here is not due to the extra 
 cost of eight hundred dollars for each ten thou- 
 sand feet of floor. 
 
 (131) The Cohesive System is the most 
 profitable for factories, for the following rea- 
 sons : 
 
 (132) I. Rigidity in the floors, representing 
 an economy in coal : It is evident that all oscil- 
 lating movement in the floor is a loss of power 
 that represents at the end of the year a sum of 
 coal consumed in excess. In the walls and 
 floors of my system the rigidity is absolute, and 
 in a building of great surface, that represents, 
 as I have found by experiments with the old 
 and new factories of Muntadas, in San Martin, 
 before referred to, a net saving of between five 
 and six per cent. 
 
 (133) 2. It is common to see in wooden floors 
 a warping of the wood, caused by the change 
 of temperature or humidity, or currents of air 
 or the proximity of heated bodies, thus chang- 
 ing the level ; consequently the machinery is 
 thrown out of level. Then ensues a loss of 
 power, if they are not re-levelled ; and when 
 this is done, there is no certainty as to their 
 stability, as another change is imminent. The 
 same is true when, in consequence of the bad 
 
122 COHESIVE CONSTRUCTION. 
 
 setting of the beam, caused by no ventilation in 
 the end, dry-rot is precipitated, the more so if 
 care has not been taken in the use of the lime. 
 Very frequently the beams have to be replaced. 
 In my system all the material is permanent, like 
 solid walls. 
 
 (134) 3. We know the necessity for the use 
 of grease and oil in factories, and the dangers 
 incident thereto, especially in cotton mills, be- 
 cause of their extreme inflammability ; and, as 
 all manufacturers require great room surface, 
 and as the distance of many parts of rooms is 
 far from the exits in these buildings, there is a 
 constant menace to safety. 
 
 (135) In factories constructed on the Cohesive 
 System, the floors being laid with tiles, nothing 
 is affected by the oil. The cotton, if it takes 
 fire, has nothing to burn ; no iron work is ex- 
 posed ; all is clay or cement. (See Figures 49, 
 50, 51, 52, 53, 54, 55, Plate No. i.) 
 
 Note. — The walls are constructed of clay tiles, the piers 
 being built hollow and utilized as ventilating-flues. The beams 
 are all covered in the arches and work under tension. 
 
 Hard-burned clay tile floors, etc., are used, as well as fire- 
 proof columns ; no iron is exposed. 
 
 There are on each floor, in addition to space appropriated 
 for manufacturing purposes, offices, store or sample rooms, 
 toilet-rooms and fire-proof stairs. The building is well lighted 
 and ventilated, and is adapted for almost any kind of manufact- 
 
COHESIVE CONSTRUCTION. 123 
 
 STORES AND WAREHOUSES AND COLD STORAGE. 
 
 (136) As these buildings require very strong 
 floors, barrel-arch bridges must be constructed, 
 two feet apart, built to the level of the crown 
 of the arch. These bridges must be of the 
 same material as the arch, and built, if possible, 
 at the same time as the arch. The bridges 
 must be in and against the next haunch of the 
 opposite barrel resting against the wall ; if not, 
 they need a rod on top to tie both ends. The 
 dome is most desirable for this class of build- 
 ings, having the highest strength with the small- 
 est section. The bridges in the dome must be 
 radial, but connected with some rings. The 
 rings must be as high as the ribs. 
 
 DWELLING-HOUSES. 
 
 (137) It is universally believed that the ob- 
 ject of fire-proofing in private houses is to guar- 
 
 ure. The stories are 14', 13', 12/ and 14', respectively, witli 
 bays 25' X 106'. 
 
 Tiie safe load is 150 pounds per square foot, one month 
 after built ; 350 pounds six months after built. 
 
 The building is four stories high, 238 ft. deep and 134 ft. 
 wide, making 26,000 square feet, and the price is based on 
 nothing smaller. 
 
 The cost is 89 cents per superficial foot for each floor, in- 
 cluding walls, floor andiron construction, against 75 or 80 cents 
 per superficial foot of wood floor and girders and walls. 
 
124 COHESIVE CONSTRUCTION. 
 
 antes only against fire, when in fact its main 
 value, which no insurance can protect, is to give 
 a perfectly air-tight floor and partitions between 
 the apartments in any dwelling. With wooden 
 construction, it is practically impossible to avoid 
 cracks in the ceiling and walls and a separa- 
 tion in the joints of the wooden floor; and, in 
 consequence of this, each floor is in communi- 
 cation with the floor below or the adjoining 
 room. This is true in any kind of building with 
 wood, and if it is dangerous to have the air 
 pagsing between two different apartments, it is 
 also dangerous to have it pass between mem- 
 bers of the same family. Medical science has 
 settled that isolation is absolutely necessary in 
 some of the ordinary diseases, to prevent their 
 spreading. 
 
 (138) There is frequently seen in large cities 
 (perhaps compelled by local law), in apartment 
 houses, a notice on the door, that some disease 
 exists within ; in order, perhaps, to prevent the 
 intrusion of callers. But that does not prevent 
 communication through cracks in the floors and 
 partitions. One can often see, through a crack, 
 a light in a room below or adjoining. This 
 crack allows of a circulation of the infected air 
 to the several families of the house. These 
 facts speak for themselves. This circulation 
 
COHESIVE CONSTKUCTION. 
 
 125 
 
COHESIVE CONSTRUCTION. 
 
COHESIVE CONSTRUCTION. 
 
 1-Al 
 
 Fig. 58. 
 
128 
 
 COHESIVE CONSTRUCTION. 
 
 through the walls and 
 floors cannot be pre- 
 vented if wood is used, 
 as every one knows that 
 the natural movement of 
 the beams cracks the 
 cornices and ceilings ; 
 and afterwards, when 
 the floor dries, the joints 
 open. 
 
 (139) With the com- 
 plications we have to-day 
 in ordinary dwellings 
 with steam, water and 
 gas pipes, not to speak 
 of electric wires, which 
 cannot be fitted tightly, 
 the danger of leakage 
 is increased. 
 
 The Figure No. 56 
 represents the section 
 of a dwelling-house, 
 showing a hall entrance 
 and parlor. Figure 57 
 represents the framing 
 iron necessary for a 
 dome for such a parlor 
 and the rough tile con- 
 
COHESIVE CONSTRUCTION. 129 
 
 struction before being decorated ; and Figure 
 
 58 shows this construction decorated. Figure 
 
 59 is the section of ceiling-floor. The section 
 shows two cases : first, one-half the drawing 
 shows the floor filled up level with concrete ; 
 second, on the other side the concrete is left 
 out in order to leave it hollow, and the sleepers 
 are supported and fastened with small anchors 
 with tile piers laid over the arch. 
 
 (140) We recommend the use of domes in 
 private houses. First, because they are stron- 
 ger than barrel arches, and cheaper ; second, 
 because they are of better decorative form. 
 
 (141) To accomplish the object of isolation, 
 coincident with adequate strength, it is suffi- 
 cient to have only two courses of tiles, each one 
 inch thick, for domes from sixteen to twenty 
 feet span, introducing, in some cases, ribs for 
 extra strength. 
 
 COTTAGES. 
 
 (142) One of the frequent demands, especially 
 in New England, is for fire-proof cottages. 
 Some ask for the cellar only to be fire-proof. 
 The fire-proof cottage must be very economical. 
 
 .(143) To make the entire construction of the 
 cottage fire-proof the dome is the cheapest and 
 most appropriate method. 
 
130 COHESIVE CONSTRUCTION. 
 
 (144) All the partitions may be constructed 
 with clay blocks, the advantage of which is that 
 they can always be utilized. These clay parti- 
 tions need cemented door-way frames and win- 
 dow frames, practically requiring only wood for 
 the movable parts in the doors and windows. 
 The roof of the cottage can be of tiles, glazed 
 or salt-glazed, laid with Portland cement. The 
 outside walls may be built with large blocks of 
 tiles with air-spaces between. 
 
 HOSPITALS AND SCHOOLS. 
 
 (145) In hospitals and schools the ceiling 
 could be composed of two parts, the ceiling and 
 the floor, with an air cavity between in such a 
 way that ventilation for the floor above and ceiling 
 below should be sufficiently distributed,, in order 
 to have the greatest number of registers and ven- 
 tilation pipes extend to every corner of the rooms. 
 This condition requires that the spaces between 
 the floor and ceiling be free, for which very small 
 beams are required. Domes in schools and hos- 
 pitals are the best, because the beams may be 
 small enough to work only by tension. 
 
 (146) Barrel arches are not so convenient, 
 because the skewbacks always intercept com- 
 munication with the next arch, and when a flat 
 ceiling is built under the barrel, the ventilation 
 
COHESIVE CONSTRUCTION. 131 
 
 can only be done longitudinally without ex- 
 pense and without weakening the arch. 
 
 OFFICE BUILDINGS. 
 
 (147) The introduction of a great number of 
 stories in modern structures suggested to archi- 
 tects the use of very thin floors, in order to gain 
 in height, and also in lightness.- The use of 
 domes, well combined, has great advantages in 
 this connection. 
 
 STAIRCASES. 
 
 (148) One of the most valuable applications 
 of the timbrel arch is the staircase constructed 
 under our system (Plate III). 
 
 The materials which are in general use to- 
 day for staircases are wood, iron and stone. 
 The first one is not fire-proof. The second is 
 hardly fire-proof, and neither of the first two 
 possesses sufficient architectural character. The 
 stone is good, but it is too heavy and expensive 
 for dwellings and other similar buildings. 
 
 (149) The advantages of the timbrel arch in 
 the staircase are as follows : — ■ 
 
 1. It is absolutely fire-proof. 
 
 2. It is adaptable to any size and in any 
 place where two walls can be disposed of, or 
 one wall and two floors. 
 
132 COHESIVE CONSTRUCTION. 
 
 3. It is a constructive arch and is susceptible 
 of any decoration, and gives infinite richness 
 without great expenditure. Two kinds of stair- 
 cases can be built on this system. One may be 
 called a stair of surmounted arches, and the 
 other a spiral staircase. But when we say 
 spiral staircase it does not only apply to a cir- 
 cular base, but also to a rectangular base. 
 
 (150) The first one, as the name indicates, 
 IS a series of catenarian arches, built, one over 
 the other, in an angular way. The second one 
 is a continuous arch from top to bottom without 
 any intersection except in the walls. 
 
 (151) The way to trace both of these stairs 
 requires some knowledge of the manner in which 
 they work, and it is difficult to explain how to 
 determine the right form of the arch of any of 
 these stairs, and to establish any definite prin- 
 ciples for every case, because the problems 
 change to infinity. The combination of the 
 spring of the arch with the continuous curve 
 needed under the platform in order to adjust 
 the adjoining flight is a question of mechanical 
 appreciation, because it is necessary to take 
 into consideration the space required for the 
 steps, and to give the right continuous curve of 
 pressure to the arch. For this reason no pre- 
 cise rules can be given. 
 
COHESIVE CONSTRUCTION. 13S 
 
 (152) Nevertheless, one of the principal rules 
 is that the first step must be in the lower 
 spring of the arch, which is at the beginning 
 of the flight, and if any flight is longer than 
 the steps required the platform should be 
 placed at the head of the flight, instead of at 
 the beginning. These principles are necessary 
 in order to avoid the results of using brackets, 
 which destroy the character and weaken the 
 stairs. 
 
 (153) Another principle is that the cross- 
 section of any flight is in the form of a bracket ; 
 that is, the arch for the surmounted stairs is 
 formed by two catenarias, one lower, at the in- 
 tersection of the wall, and another higher, at 
 the outside of the steps, under the banister ; and 
 to determine the difference between the two 
 catenarias is a question of appreciation in view 
 of width and the length of the flight, and the 
 length of the adjoining flight. 
 
 (154) The form of bracket for a spiral stair- 
 case must be determined by the two spiral lines, 
 one lower than the other, the lower one being 
 against the wall, and the other against the ban- 
 ister on the end of the steps. 
 
 (155) The problems of staircases built with 
 timbrel arches are complex and varied, and 
 would require a special treatise. 
 
134 COHESIVE CONSTRUCTION. 
 
 (156) The stairs that may be referred to are : 
 the main stairs of the Boston Public Library; 
 the stairs for the American Legion of Honor 
 Building, Huntington Avenue, Boston; the 
 stairs in Exeter Chambers, Boston; the stairs 
 for the private residence No. 122 West 78th 
 Street; four private residences at 154, 156, 158 
 and 160 West 82d Street; the apartments of 
 99th and looth Street (see Plate HI), corner of 
 Columbus Avenue, New York ; and the spiral 
 stairs in the Washington Memorial Arch, New 
 York. 
 
 BRIDGES. 
 
 (157) Among the applications of the timbrel 
 or cohesive arch one of the most valuable is the 
 masonry bridge built on these principles. 
 
 The costly wooden centres required for ma- 
 sonry bridges, built either of brick, concrete or 
 stone, cannot be avoided in order to have the 
 centre rigid enough to receive the weight of the 
 masonry, so as to give to the material rest for a 
 good setting. This takes a great part of the 
 estimate for any bridge, particularly if the spans 
 are large, as is the case in all good bridges. 
 
 (158) In timbrel arches the estimate for the 
 centering is a very small matter, the reason for 
 which is easily understood. As the principle 
 
COHESIVE CONSTRUCTION. 135 
 
 of Cohesive Construction with tiles is to build 
 the arches by coats, one over the other, with 
 broken joints, each coat makes a centre for 
 the next ; so the centre can be considered only 
 for the first few courses, whose weight is very 
 light. This advantage permits of larger spans, 
 gives better conditions for the absolute setting 
 of the material, and is economical. But these 
 advantages, already mentioned, are not the only 
 ones which this system possesses for bridges. 
 
 (159) I. In the cohesive-tile system the 
 bending-moment in any section can be in- 
 creased without adding material, by making the 
 section tubular. 
 
 2. The railing can be constructed in con- 
 nection with the bridge under such conditions 
 that the railing can be counted as one of the 
 members which give strength to the bridge. 
 
 3. The tubular condition, as well as the con- 
 struction of the railing in connection with the 
 bridge, solves in a favorable way the stability 
 of the bridge in regard to moving loads. 
 
 (160) Figures 60, 61, 62, Plate IV, is a plan 
 and section of a bridge. It is not altogether 
 tubular, but only at the ends, as can be seen ; 
 the bending-moment increases at the ends, in 
 order that the course of pressure can be inside 
 for moving loads. 
 
136 COHESIVE CONSTRUCTION. 
 
 These ends are formed by two series of 
 arches. The arch underneath is a continuous 
 dome decreasing the radius of its spheric sur- 
 face at the ends, and over the dome, the second 
 arch (segmental) pursuing just the lines of the 
 roadbed, connecting both arches by ribs. 
 
 In a separate book we hope to treat more 
 extensively on this specialty of bridges. 
 
 CRITICISMS OF THE COHESIVE SYSTEM. 
 
 (i6i) Many discussions have arisen, both in 
 favor of and against the Cohesive System in 
 the past year, as is natural when any new ap- 
 plication is brought forward in tiie arena of 
 scientific discussion. But it has occurred that 
 the greatest friends of the system sometimes 
 go too far in their enthusiasm and favor of the 
 new idea, the result being that in their hands it 
 is disfigured or erroneous. For instance, it is 
 said : — 
 
 (162) I. That the arches under this system 
 have no thrust. It is and it is not so, as I will 
 explain later. 
 
 (163) 2. It is said that the system is more 
 expensive than any other system of fire-proofing, 
 or that there is no economy in it. To this I 
 will answer, that it depends on the way it is 
 adapted. 
 
COHESIVE CONSTRUCTION. 137 
 
 (164) 3. That the ceilings of this construc- 
 tion require greater height than others. Con- 
 sequently, it is necessary to build every room 
 with high ceilings, and that, of course, causes 
 the building to be higher. This impression 
 probably exists because some people may have 
 seen some arches or ceilings that have been 
 constructed specially with a high rise, to give 
 more character or to show more constructive 
 lines ; and they imagine that all ceilings must 
 have a similar rise, when the truth is, that on 
 an average they take less room ; but this de- 
 pends on the wishes of the architect who is 
 applying the system. 
 
 Note to the Reader. — Another consideration lias arisen 
 ill discussion, wliicli I will ask permission of the reader to 
 let me answer; that is, the question in regard to the patents. 
 Now, if the system is patentable, in what does the patent 
 consist, and by what is the patent warranted .'' In the first 
 place, I must say that, allowing tliat in a small portion of Spain 
 and Italy a similar system was applied empirically, and on a 
 smaller scale, it is a fact that it is not applied in any modern 
 public building in either place, because, as I say, it has only an 
 empirical application j nor has any academy a regular system 
 or scientific method for the application of it. 
 
 Improvements have been introduced constantly. Some of 
 them are as follows : — 
 
 I. The custom was to use plaster in the first and second 
 courses; and it was not possible to construct in any other 
 way. This gave an excess of plaster, which was very preju- 
 dicial, as all intelligent builders know, and of which we give 
 
138 COHESIVE CONSTRUCTION. 
 
 (165) In answer to the first I will say that 
 the thrust depends on the form and not on the 
 material, as stated before. (See page 75 in re- 
 gard to barrel arch and domes.) 
 
 (166) The answer to the statement that it is 
 so expensive that it cannot be used with econo- 
 my, is as follows : — 
 
 (167) I. As we have said before, the barrel 
 arch has some thrust, and requires something to 
 counterbalance this, that is, a rod, or rods. That 
 is one of the causes which makes the barrel 
 construction more expensive than the dome. 
 
 (168) 2. The barrel arch requires two sides 
 for the arch to rest on, such as girders or beams, 
 
 full explanations in this book. I pursued 4:his method for 
 several years, with all the attending inconveniences, trying to 
 discover a means of avoiding it. To-day we are using one- 
 tentli of the plaster we used originally. 
 
 2. Before, it frequently happened that, because of the 
 negligence of workmen or others in stepping over or putting 
 heavy weights on the arch before it had set, or for any other 
 cause, some of the tiles of the first course, having become a 
 little separated, were likely to fall. To-day, by means of a 
 flange, the tiles, if any become separated, can never fall, and 
 none the less have the same stability. 
 
 3. Formerly the tiles used to be covered by plaster, leaving 
 that as a rustic form of rough material, purely constructive ; 
 to-day they are employed in a more useful way, the tiles form- 
 ing the construction and decoration. That was one of the 
 constant and noble aspirations of the art of construction; but 
 to arrive at this point with the tile was not an easy problem, 
 
COHESIVE CONSTRUCTION. 139 
 
 or walls. These girders or beams must have 
 the strength to support all the weight that the 
 arch carries, besides the weight of the arch. 
 The prices of the girders or beams must be 
 added to the price of the arch. Consequently 
 we have the same objection as in the regular 
 fire-proofing process, where the arches also rest 
 against the beams, the advantage being that 
 our barrel arches arc lighter, and the span can 
 be greater ; consequently these girders or beams 
 can be lighter and cheaper. But, nevertheless, 
 the fact remains that we must have the girders or 
 beams, and the value of these must be added to 
 the arch. The result is, if we have an economy 
 
 because the decorative tile is the first course, which is the 
 most difficult to have nicely and properly jointed, while the 
 material required for this first course gives no chance for care- 
 fully made right angles and even joints. 
 
 4. In several cases, for industrial, mercantile or special 
 buildings, it is necessary to have flat ceilings, and very light 
 floors, practically deafened. We have these conditions pro- 
 vided for to-day in the cohesive system. 
 
 Now, if the object of the law of patents is to guarantee the 
 intellectual work applied to new applications and improve- 
 ments, is it possible to have no guarantee for all our new 
 applications. But the protection of patent law was invoked, 
 not with a desire of making a monopoly, nor for gain only, 
 with that object. I may remark that, notwithstanding to some 
 architects the system is acceptable, the truth is that the day is 
 very far distant when it can be given to the common use and 
 free practice of all kinds of contractors, while they have not 
 
140 COHESIVE CONSTRUCTION. 
 
 over the ordinary fire-proofing, the barrel arches 
 cannot compete with slow burning construction, 
 in consequence of the heavy girders required. 
 
 (169) 3. But if, instead of using the barrel 
 arches on the Cohesive System, we use the 
 domes on the same system, we have economy, 
 if properly applied, because we have not the 
 objections named — that is, the use of heavy 
 rods between heavy girders, and the use of 
 girders or beams themselves. 
 
 (170) Now, suppose we receive apian from an 
 architect, as is generally the case, with barrel 
 arches and girders, and we quote a price which 
 is more expensive than the combination of the 
 
 yet at their disposal tiie elements necessary, neither o£ mate- 
 rial nor expert hands ; and for this reason the system would be 
 dead, if it was not restricted. For instance, suppose an archi- 
 tect, knowing and believing in the system and convinced of 
 its utility, as there are many to-day, should project a building 
 under this system, as insignificant as it is or appears to be. 
 If he called for competitive estimates, in order to obtain the 
 price, I am sure that neither the architect nor the owner would 
 be certain that the contractor was practically able to erect the 
 building with success in the construction. The architect, not 
 having enough confidence in the contractor, not knowing 
 whether he was practical or not, and realizing that he person- 
 ally is directly responsible, knows that he will be a slave of the 
 building. On the other hand, the owner, knowing that the sys- 
 tem is new and put in the hands of a contractor who cannot 
 give references as to his knowledge of the matter, by his past 
 record, would not have confidence in him, and would pay his 
 
COHESIVE CONSTRUCTION. 141 
 
 dome without the heavy girders and heavy rods. 
 The result is, that it is not as cheap as other 
 combinations that we have already built and are 
 now building. Is it the fault of the system.? 
 Most certainly not. The fault is, that it is not 
 well applied. 
 
 I must remark with great satisfaction that we 
 have commenced to receive plans from archi- 
 tects in every part of the country, some of them 
 showing that this system is commencing to be 
 well understood and appreciated at its just value. 
 
 (171) In regard to the question of height, I 
 can say that, notwithstanding we give ten per 
 cent, rise, as a rule, on all of the arches, it does 
 not mean that it cannot be less or greater; just 
 as, in the ordinary system of beams, we know that 
 the higher the section we give to beams the more 
 economical is the result. For instance, if in a 
 floor framing of beams twenty-five feet in span 
 
 money without a guarantee that the contract was well and 
 safely performed. The contractor, too, would not be in any 
 better position : he could not find the material nor the work- 
 men, under the ordinary conditions that other systems would 
 allow, and consequently everything would be against either 
 success or economy. 
 
 This is the reason that it was necessary to accumulate, year 
 by year, elements of security for architects and owners, and 
 acquire elements for sujjplying the market with material and 
 expert hands, educating able foremen, able masons, and able 
 
142 COHESIVE CONSTRUCTION. 
 
 we use beams eight per cent, of the span, the 
 ceiling would be cheaper than if we give five 
 per cent, for an equal load ; for, in the first 
 case, with less pounds of iron, we have the 
 same resistance. The same is the fact in the 
 arches and domes ; the more rise we give, the 
 more economical they are. This means that 
 we use the same rules as in general bridging ; 
 that we can reduce the rise to eight per cent, in 
 ordinary construction ; and, as the arch's form 
 gives opportunity for greater height to the ceil- 
 ing in the centre, the result is that practically 
 the ceilings are higher in our system than in 
 any other system, taking only, in the crown, six 
 inches of room, but descending on the side, and 
 occupying the place of small corners in all the 
 regular ceilings. Consequently, I cannot see in 
 what way this system takes up more room than 
 any other. 
 
 helpers. The reader will appreciate the cost, the great sacri- 
 fices and the capital expended in order to succeed in ^U this, 
 besides the necessity of placing the system on a scientific 
 basis. All of this could not be done without a patent guar- 
 anty ; and all of this had to be perfected at the commencement 
 of the first year's work, because it was a labor of propagation 
 and evolution. 
 
 It can be seen with these explanations that it was necessary 
 to protect the system, not with a desire of making a monopoly 
 for gain only, but to assure its success. 
 
COHESIVE CONSTRUCTION. 143 
 
 MATERIALS AND IMPROVEMENTS IN THE FUTURE. 
 
 (172) My intention is not to uphold the use 
 of concrete in construction where cohesive 
 strength is required, because it is a slow procr 
 ess. It works practically only by pressure, and 
 there is no shearing strength in the joints, or a 
 chance for perfect setting. It makes a heavy 
 load, especially on the centres during construc- 
 tion. It produces a great load of dampness in 
 the building, and is ruinous for patching or alter- 
 ation. After several years' experience in con- 
 crete construction with no satisfactory results, 
 I have come to the conclusion that the tubular 
 system, as applied in constructing walls and 
 ceilings built with light and well-burned clay, 
 and good Portland cement, is the best, safest, 
 most substantial, economical and most rapid 
 method of construction, on the Cohesive Sys- 
 tem, for dwellings and other kinds of buildings. 
 
 (173) By light and well-burned clay, I mean 
 such as was used by the ancients, which, al- 
 though as strong as the others (from 1,500 to 
 2,500 pounds per square inch), has an average 
 weight of from 80 to 120.16 per cubic foot. 
 
 (174) The mortars used must be of the quality 
 called Hydraulic — mortars that do not need 
 exposure to the air for setting, which, for our 
 especial work, is Portland cement. 
 
144 COHESIVE CONSTRUCTION. 
 
 (175) The bricks must be of as large dimen- 
 sions as a man can easily handle, and of such 
 weight that a man can work with them all day. 
 These conditions produce a brick about four 
 pounds in weight (when the ordinary clay is 
 used), and about an inch thick, and seventy-two 
 square inches in surface, or 6"x 12". These tiles 
 must be a little porous in order to absorb some 
 of the excess of water in the cement. It may be 
 added that the breaking load for a square inch of 
 tile is between two and three thousand pounds. 
 
 (176) It may be remarked that a great deal 
 has still to be accomplished by way of improve- 
 ment, and it is necessary to call for assistance 
 in perfecting our knowledge of the art of build- 
 ing, especially from manufacturers of materials 
 and architects. 
 
 There are three improvements that may be 
 suggested : — 
 
 (177) I. The technical part ought to be put 
 in a treatise, in order that it may be used in 
 the schools of technology for the benefit of the 
 constructive art, 
 
 (178) 2. (This to manufacturers): Our tiles as 
 well as our bricks are too heavy. The cupola 
 of St. Sophia has some rings that were built 
 with lava (pumice stone), and some with brick 
 so light that it would float in water. We have 
 
COHESIVE CONSTRUCTION. 145 
 
 some of this kind of clay in Spain (Barcelona 
 and Valencia; also in Alcora, north of Valencia). 
 Perhaps the people working in the old mines 
 between Peniscola (Valencia) used this kind of 
 clay, but investigations on my part have failed 
 to discover it. The idea of looking for it in the 
 mining districts led me to make a search for it 
 in Colorado, with flattering prospects. I have 
 a specimen from Colorado, but it is heavy ; we 
 are using lighter tiles for ceilings. 
 
 (179) The West is much advanced in the pre- 
 paration of fire-clay, though not as much as in 
 Spain ; and in some parts of Mexico they are bet- 
 ter prepared than here in the East for ceramic 
 work applied for architectural purposes, and they 
 have given some attention to the lighter brick./ 
 
 (180) It is an error to believe that the heavy 
 brick is the best ; the light brick, which I men- 
 tioned, has a breaking strain of 2,200 pounds, 
 and will float in water. 
 
 ECONOMY IN THE FUTURE. 
 
 (181) The clay in the tile we use is about the 
 same as that in the common brick, and the vol- 
 ume is about the same, from 4, 4^ to 5 pounds. 
 Our tile is i x6x 12 inches, or 72 cubic inches, 
 and the brick is 2^ X4X x 8, or 72 cubic inches ; 
 but we have to pay from ^18 to ^20 per thou- 
 
146 COHESIVE CONSTRUCTION. 
 
 sand for the tile,* while brick costs from $y to 
 ^9 per thousand. 
 
 (182) When we consider that tiles are made 
 in blocks of six tiles each, and thus more easily 
 handled than brick at the factory ; that they are 
 thinner, thus drying more rapidly and being 
 easier burned with less fire — it would seem 
 that they ought to be made more cheaply than 
 brick ; and such is the case in Spain, where 
 brick costs from $6 to ^7 per thousand, and 
 tiles from $4 to $<, per thousand. 
 
 (183) This anomaly is perhaps due to the fact 
 that as yet there has not been sufi&cient demand 
 for these tiles here to cheapen their manufact- 
 ure by producing them in large quantities. 
 
 (184) In regard to cement, we now have to pay 
 English manufacturers a large contribution on 
 the Portland cement we use ; when it can safely 
 be made in this country I suppose it ought' to 
 be about twenty per cent, cheaper. We ought 
 from these two sources alone to be able to cheap- 
 en construction from twenty to thirty per cent. 
 
 (185) 3. (This to architects) : In concluding 
 permit me to put before you the following 
 thoughts, which are not my own, but those of 
 one of the eminent English authors of the first 
 part of the century : — 
 
 *The cost now is from $io to $12. 
 
COHESIVE CONSTRUCTION. 14Y 
 
 (i86) "What is the best type of structure — 
 that which for equal periods of duration has 
 more per cent, of its full-covered area occupied 
 by solid walls, or that which has less of its sur- 
 face so occupied by walls .'' 
 
 "In the contemplation of buildings which 
 sJiow their strength by their age, the compar- 
 ative science displayed may be partly estimated 
 by an inverse ratio of the mass of materials to 
 the space covered. 
 
 "The following list of notable buildings, 
 with the per cent, between the areas covered 
 and the wall surface, may be interesting : — 
 
 "The superficial feet of walls of the Church 
 of the Invalides, at Paris, two-sevenths of the 
 whole is solid. 
 
 " In St. Peter's, of Rome, one-fourth. 
 
 "In St. Paul's, of London, two-ninths. 
 
 "The Pantheon, one-fourth. 
 
 " St. Genevidve, at Paris, one-seventh. 
 
 " Salisbury Cathedral, one-fifth. 
 
 "Temple of Peace, one-seventh. 
 
 " Parthenon, two-elevenths. 
 
 " St. Sophia, cohesive system, one-eighth. 
 
 " A great building with few materials, besides 
 the periodical approbation that it will receive 
 from the age, will have an indisputable superi- 
 ority as a rule." 
 
148 
 
 COHESIVE CONSTRUCTION. 
 
 TABLE OF THEORETICAL STRESSES FOB ARCHES 10% RISE 
 WITH UNIFORM LOAD (W) PER SQ. FT. 
 
 c 
 c 
 to 
 
 S 
 
 1 1 
 
 a % 
 
 i % 
 
 a 
 > 
 i 
 
 in 
 
 •B 
 
 d 
 
 a 
 
 ■i. 
 
 S 
 1 
 
 r 
 
 O 
 
 Bending- moment at 
 
 crown. 
 
 Inch pounds 
 
 -a 
 
 liii 
 
 1 
 1 
 
 e X 
 
 i ^ 
 1 " 
 1 
 
 1 
 -eX 
 
 1 
 
 .1 
 " X 
 
 i ^ 
 -i II 
 
 ■s 
 l|x 
 
 1 
 
 ■i 
 
 icX 
 l^ll 
 ■g 
 
 5 
 
 6 2 
 
 24 
 
 8 
 
 .540 
 
 .0675 
 
 6.16 
 
 .25067 
 
 6.073 
 
 278 
 
 .32417 
 
 6 
 
 6 r 
 
 36 
 
 27 
 
 .540 
 
 .03 
 
 6.16 
 
 .17111 
 
 6.673 
 
 .18536 
 
 .20111 
 
 6 
 
 7.2 ; 
 
 36 
 
 27 
 
 .7776 
 
 .0432 
 
 7.392 
 
 .20533 
 
 8.008 
 
 .2227 
 
 24853 
 
 7 
 
 , 8.4 ; 
 
 36 
 
 27 
 
 1.0854 
 
 .0588 
 
 8.624 
 
 .23956 
 
 9.3425 
 
 2595 
 
 29836 
 
 8 
 
 9.6 . 
 
 i 36 
 
 27 
 
 1.3824 
 
 .0768 
 
 9.856 
 
 .27378 
 
 10.677 
 
 2966 
 
 .34468 
 
 9 
 
 10.8 
 
 i 36 
 
 27 
 
 1.7496 
 
 .09719 
 
 11.088 
 
 .30800 
 
 12.013 
 
 .3372 
 
 .40518 
 
 10 
 
 12.0 
 
 5 36 
 
 27 
 
 2.16 
 
 .12 
 
 12.320 
 
 .34222 
 
 13.346 
 
 .3707 
 
 .46222 
 
 11 
 
 13.2 
 
 5 30 
 
 27 
 
 2.6136 
 
 .1453 
 
 13 552 
 
 .37644 
 
 14.598 
 
 .4055 
 
 .52164 
 
 12 
 
 14.4 
 
 3 36 
 
 27 
 
 3.1104 
 
 .1728 
 
 14.784 
 
 .41067 
 
 16.016 
 
 .4449 
 
 .58347 
 
 12 1.44 
 
 1 48 
 
 04 
 
 3.1104 
 
 .0972 
 
 14.784 
 
 .308 
 
 16.016 
 
 .3337 
 
 .4052 
 
 ist 1.5G 
 
 1 48 
 
 64 
 
 3.6504 
 
 .1140 
 
 16.016 
 
 .33367 
 
 17.351 
 
 .3615 
 
 .44767 
 
 14 
 
 16.8 
 
 4 48 
 
 64 
 
 4.2336 
 
 .1323 
 
 17.248 
 
 .35933 
 
 18.685 
 
 .3768 
 
 .49163 
 
 15 
 
 18.0 
 
 4 48 
 
 64 
 
 4.860 
 
 .15187 
 
 18.480 
 
 .38500 
 
 20.02 
 
 .4171 
 
 .63687 
 
 16 
 
 19.2 
 
 4 48 
 
 64 
 
 5.5296 
 
 .1728 
 
 19.712 
 
 .41067 
 
 21.355 
 
 .4449 
 
 .58347 
 
 10 
 
 19.2 
 
 5 60 
 
 126 
 
 5.6296 
 
 .1106 
 
 19.712 
 
 .32853 
 
 21.355 
 
 .3559 
 
 .43913 
 
 17 
 
 20.4 
 
 5 60 
 
 125 
 
 6.424 
 
 .1286 
 
 20.944 
 
 .34907 
 
 22.689 
 
 .3781 
 
 .47757 
 
 18 
 
 21.6 
 
 5 60 
 
 125 
 
 6.9884 
 
 .13977 
 
 22.170 
 
 .36960 
 
 24.024 
 
 .4004 
 
 .50937 
 
 19 
 
 22.8 
 
 5 60 
 
 125 
 
 7.79T6 
 
 .15595 
 
 23.408 
 
 .39013 
 
 25.359 
 
 .4225 
 
 .54608 
 
 20 
 
 24.0 
 
 6 60 
 
 125 
 
 8.64 
 
 .1728 
 
 24.64 
 
 .41067 
 
 26.693 
 
 .4449 
 
 .58347 
 
 20 
 
 24.0 
 
 6 72 
 
 216 
 
 8.64 
 
 .12 
 
 24.04 
 
 .34222 
 
 26.693 
 
 .3707 
 
 .46222 
 
 21 
 
 25.2 
 
 6 72 
 
 216 
 
 9.5256 
 
 .1323 
 
 25.872 
 
 .35933 
 
 28.028 
 
 .3893 
 
 .49163 
 
 22 
 
 26.4 
 
 6 72 
 
 216 
 
 10.4544 
 
 .1452 
 
 27.104 
 
 .37644 
 
 29.363 
 
 .4079 
 
 .52164 
 
 23 
 
 27.6 
 
 6 72 
 
 216 
 
 11.4264 
 
 .1687 
 
 28.336 
 
 .39355 
 
 30.697 
 
 .4263 
 
 .55225 
 
 24 
 
 28.8 
 
 6 72 
 
 21C 
 
 12.441C 
 
 .1728 
 
 29.568 
 
 .41067 
 
 32.932 
 
 .4449 
 
 .58347 
 
 To obtain bending-nioments, stresses, and thrusts in the last seven 
 columns, multiply the figures in column by load per square foot, 
 including the weight of material. 
 
COHESIVE CONSTRUCTION. 149 
 
 The preceding table of stresses is calculated 
 by Gaetano Lanza, Ph. D., Professor of Applied 
 Mechanics at the Massachusetts Institute of 
 Technology. It gives the stresses or strains 
 per square inch of the Timbrel Arches having a 
 rise of ten per cent, of the span and supporting a 
 distributed load of one pound to the square foot. 
 The coefficient was taken from the tests already 
 mentioned (pages 58 and 59) as its ultimate 
 resistance. 
 
 In calculating the safe load which might be 
 put on the arches, we use ten per cent, of these 
 ultimate resistances, thus introducing a much 
 larger factor of safety than is usually employed 
 in such calculations. As a matter of fact the 
 arches may be strained to within one-fifth or 
 one-fourth of the ultimate resistance of the 
 material with perfect safety. 
 
 To determine the safe distributed load per 
 
 square foot in the table of stresses, it is only 
 
 necessary to divide the coefficient 206 by the 
 
 maximum stress at the crown, taken in the line 
 
 of figures at the end of the table ; thus, for a 
 
 , t, 1 ■ f , 206 
 10 span 3 thick, we have — ^ = 4,457 lbs. 
 
 0.4C>222 
 
 per square foot safe load. 
 
Toilet 
 
 u — iL U 
 
 Scale 1-2 in In I (not. 
 
 i 
 
 'M FirfcT &K..-y Floo r lio 
 
 Fig. 49. Plan. Scale 1-16 In. to I foot. 
 
 Section through Piers. 
 PLATE 1. 
 
 Half Elevation of Window. 
 
Fig. 52. 
 
 m 
 
 
 
 Transverse Section of Arches. 
 
 Columns I ft. 4 in. Centre and Centre, 
 
 ^ 
 
 ::] 
 
 n 
 
 i 
 
 Plan of Columns. 
 
 Plan of Struts in Caps 
 of Columns. 
 
 ..J 
 
 Fire-proof Factory, p.rt Front Elevation. Scale 1-8 in. to 1 foot. 
 
 PLATE 
 
St&IfS in Exeter Ch6,rr,hers , Cxetcr 5t., Boston, Mass 
 
 Stairs In I 00th St. ard Columbus Ave., New York. 
 
 r t •■|\ 
 
 
 . 
 
 j-.- , jj. - j!]\^ 
 
 
 ;; 
 
 <r Iri^ 
 
 ' 
 
 >^|^ 
 
 ^ 
 
 yf^i . .|.' . .|. 
 
 
 ^ 
 
 f 
 
 f.yi^l'.^ 
 
 
 ^^^^r. 
 
 n^iludin^l Section- 
 
 Ou.sUv 
 
 J 
 
 5c c cad ftfc 
 TifC-Proof Consfrucrion Co 
 
 i: 
 
 PLATE 111. 
 
Desigfn for Bridge at Lincoln Park, Lake Quinsigamond, Worcester, Mass, 
 
 PLATE IV.