SB U. S. DEPARTMENT OF AGRICULTURE. BUREAU OF PLANT- INDUSTRY— BULLETIN NO. 146. B. T. Galloway, Chief of Bureau. THE SUPERIORITY OF LINE BREEDING OVER NARROW BREEDING. O. F. COOK, BioNOMisT, Bureau of Plant Industry. Issued April 29, 1909. WASHINGTON: GOVERNMENT PRINTING OFFICE, 1909. Pass 8 M I 2 ?j Book 'C 7 Digitized by tine Internet Archive in 2010 with funding from The Library of Congress http://www.archive.org/details/superiorityoflinOOcook U. S. DEPARTMENT OF AGRICULTURE. BUREAU OF PLANT INDUSTRY— BULLETIN NO. 146. B. T. (tALLOWAy, Chief of Bureau. f f THE SUPERIORITY OF LINE BREEDING OVER NARROW BREEDING. / If or F/ COOK, >l BiONOMisT. Bureau of Plant Industry. Issued April 29, 1909. WASHINGTON: GOVERNMENT PRINTING OFFICE, 1 1» 9 . Cn ■' i-i/ ^2> \v BUREAU OF PLANT INDUSTRY. C" PhysiologiKt and PaihoJotjist, and Chief of Bureau, Beverly T. Galloway. Physiologist and Pathologist, and Assistant Chief of Bureau, Albert F. Woods. Laboratory of Plant Pathology, Erwin F. Smith, rathologist in Charge. Fruit Disease Investigations, Merton B. Waite, Pathologist in Charge. Investigations in Forest Pathology, Haven Metcalf, Pathologist in Charge. Cotton and Truck Diseases and Plant Disease iSurrey, William A. Orton, Pathologist in Charge. Pathological Collections and Inspection ^Vork, Flora W. Patterson, Mycologist in Charge. Plant Life History Investigations, Walter T. Swingle, Physiologist in Charge. Cotton Breeding Investigations, Archibald D. Shamel and Daniel N. Shoemaker, Physi- ologists in Charge. Tobacco Investigations, Archibald D. Shamel, Wightman W. (jlarner, and Ernest H. Mathewson, in Charge. Corn Investigations, Charles P. Hartley, Physiologist in Charge. Alkali and Drought Resistant Plant Breeding Investigations, Thomas II. Kearney. Physi- ologist in Charge. Soil Bacteriology and Water Purification Investigations, Karl F. Kellerman, Physiolo- gist in Charge. Bionomic Investigations of Tropical and Subtropical Phints, Orator F. Cook. Bionomist in Charge. Drug and Poisonous Plant and Tea Culture Investigaliuns, Rodney II. True. Physiologist in Charge. Physical Laboratory, Lyman J. Briggs, Physicist in Charge. Agricultural Technology, Nathan A. Cobb, Crop Technologist in Charge. Taxonomic and Range Investigations, Frederick V. Coville, Botanist in Charge. Farm Management, William .T. Spillman, Agriculturist in Cliargc Cfrain Investigations, Mark Alfred Carleton, Cerealist in Charge. Arlington Experimental Farm and Horticultural Investigations, I.oo C. Corliett, Horti- culturist in Charge. Vegetable Testing Gardens, William W. Tracy, sr.. Superintendent. Sugar-Beet Investigations, Charles O. Townsend, Pathologist in Charge. Western Agricultural Extension, Carl S. Scofield, Agriculturist in Charge. Dry-Land Agriculture Investigations, E. Channing Chilcott, Agriculturist in Charge. Po)nological Colleciiuns, Gustavus B. Brackett, Pomologist in Charge. Field Investigations in Pomology, William .\. Taylor and G. Harold P(>\vaman .\. Knapp, Special .Vgent in Charge. Seed Dislrihution (Directed by Cliicf of Bureau). Lisle Morrison. .Assistant in Geneml Charge. Editor, .1. E. Rockwell. Chief Clerk. .Tames E. .Tones. ]{li>.\(i.\lii' l.w i:s iicAiiovs ov T]!iii'ir.u. .\\i> Sriiriicn'HAi. Pi..\.nts. s('1i;nTiI'M.' s r.\ir. (». l'\ Cook, t.ianomist in cinniiv. a. N. Collins anil i'. 1.. l.i'Wlon. XssisUint liolanist.^. H. Pittier, .1. II. Kinslcr. ;nid .\. McLacblan. Si>c; it justify 14G UNNECESSARY UNIFORMITY. 33 US in overlooking the fact that characters of no vahie in themselves may be worthy of careful attention when they are correlated with more iiriportant features. With cotton, for example, there seems to be a very definite correlation between the external form of the boll and the length of the lint. Plants that have longer and more pointed bolls are almost always found to have longer lint than neighboring individuals with more rounded bolls. It is therefore worth while to pay attention to the shape of the boll as well as to its size if the desirable qualitj' of long lint is to be carefully guarded. Similar correlations between unimportant characters and those that have definite utility have been detected by Mr. W. W. Tracy. sr., in varieties of corn, beans, lettuce, and other plants. In the Extra Early Adams variety, corn plants with branched tassels, even though they may be as early themselves as other plants with the un- branched tassels characteristic of this . variety, have been found to yield offspring with a distinct tendency to later maturitj'. Plants of the Golden Wax beans with flowers slightly larger than normal were found to yield offspring with a distinct tendency to j^roduce ordinary green pods instead of yellow. Certain varieties of lettuce which differ in the characters of the cotyledons have also been found to differ in the time of running to seed, even when they appear to be indistinguishably alike in the interval between. The corn plant has been the object of much selection, but this has been directed almost exclusively to the characters of the fruit instead of to the plant as a whole. The result is that even in varieties that give relatively uniform ears and kernels the vegetative characters continue to show wide ranges of diversity in form, color, and hairi- ness, as well as in the tassels and flowers. In a broad-bred stock it is evidently possible for many such fea- tures to continue their natural diversities without detriment. It is only in narrow-bred types that absolute uniformity has appeared a desirability of breeding, and this we can now undej'stand, because the application of such standards has always tended to carry these types toward the superior condition of line breeding. The efforts that have been made to place cross-fertilized types, like Indian corn, on the same basis of uniformity as a line-bred type, like wheat, have not had any corresjDonding measures of success. In spite of long and persistent selection only a small proportion of the plants show any complete uniformity of the fruiting characters. The facts with ref- erence to corn have been summarized by Mr. A. D. Shamel of this Department. It is tlie experience of most corn breeders that it is not possible to produce on an acre more than 5 busliels of luiiform ears even of onr most improved strains. A large majority of the plants pi'odiice ears of small size, irregular 146 34 LINE BREEDING AND NARROW BREEDING. sliape. and light weight, whicli are undesirable. Many of the stalks are barren. Only a small proportion of the plants produce the maximum size and weight of ear.'* Such facts make it evident that there is a very practical distinction between the cross-bred and the self-fertilized types with reference to uniformity. Selection toward uniformity increases the jdeld of self- fertilized types, because their methods of reproduction are well adapted to give regularity in the expression of the desired characters, whereas the cross-fertilized types have not the same power of pro- ducing uniform offspring. With the cross-fertilized plant there is no such direct and necessary connection between uniformity in high yield and uniformity in other characteristics. If corn could be propagated by self-fertilization it might be expected to respond to selection in the same way as the self-fertilized cereals, but no self- fertilized strains of corn have been developed. In the great majority of corn varieties self-fertilization is definitely prevented by the habit of proterandry, the ripening of the pollen of the plant before its silks are ready for fertilization. In an experiment with many varie- ties at Lanham, Md., in the season of 1908, a single variety proved to be an exception to this rule. The stalks produced silks and pollen at the same time, but the variety can hardly be said to be in a normal physiological condition, for the leaves of all the plants were thickly spotted with j^ellow, a sjanptom not shown in any other variety. Several experimenters have reported distinctly unfavorable results from self-fertilization in corn, the self- fertilized progeny being notably inferior to the cross-fertilized. It is possible that strains of corn may be found which will thrive under self-fertilization, but this is rendered somewhat less likely by the fact that the stamens and pistils are produced on different parts of the plant, instead of close together. The internodes of the higher plants have, as botanists know, a certain individuality of their own, the plant being a complex or colony of many of these internode individuals. If we consider the plants from this standpoint of the individuality of the internodes it becomes plain that the pistil and stamen internodes are verj' close relatives in the self- fertilized cereals, while in the corn they are sepa- rated by many generations of internodes, so that their relationship is relatively remote. The self-fertilization of a corn plant does not mean the uniting of cells from adjacent internodes, as on the wheat l^lant, but unites cells which are in some senses as little related to each other as though they were on diffierent plants. a Shnnu'l. A. D. 'I'lic Aii of Seed Si-lccl ion :ni(l I '.reed in. tf. Yoarhook of tlio I>(»l)ar(nu'ii( of .\i:rirni(nr«' foi- l!)(iT. p. T21. 140 THE '^ FIXING OF CHAEACTEKS " BY LINE BREEDING. 85 Thus it appears that the specialized habit of corn to produce its pistils and stamens on different parts of the plant puts it out of the reach of the kind of self-fertilization practiced in wheat. In order to make corn experiments which shall be truly parallel with those of the self-fertilized cereals, it will be necessary to find varieties with hermaphrodite flowers, indications of which are sometimes found on abnormal suckers of our ordinary varieties. In view of these unusual 'obstacles to line breeding it is fortunate that corn has relatively little practical need of the uniformity which is of so much importance in crops like cotton and tobacco. The chief object in a corn crop is a large yield. A large proportion of the corn is not sold at all, but is used on the farms where it is raised. Farmers would be very willing to permit variations, even in the shapes and colors of the grains, if they could get enough more corn per acre. Experiments have shown that increased jdelds can be obtained by crossing varieties of corn. Indeed, the fact seems to have been utilized since prehistoric times among the Indians of the Quezalte- nango region of western Guatemala. Thej?^ follow the practice of planting three different types together for the bulk of their crop, in the belief that larger yields are obtained in this Avay than when the varieties are planted separately. Nevertheless, breeders remain loath to advise mixed plantings of corn, or even to investigate such a possi- bility of increasing the yield, perhaps because this advice is so far out of accord with the methods by which some other cereal crops have been improved. THE " FIXING OP CHARACTERS " BY LINE BREEDING. Recognition of the superiority of line breeding over narrow breed- ing may help to explain the belief of many breeders that inbreeding " fixes " characters. Characters may appear to be fixed because line breeding protects a selected stock from the degenerative diversity which comes with narrow breeding. W^ien such a line-bred stock is crossed with a narrow-bred variety there may be a definite tend- ency for the peculiarities of the superior stock to predominate in the offspring. The mutations of narrow-bred and self- fertilized plants show that mere repetition does not insure stability, but may be followed by wide diversity, even after uniformity has been maintained for many generations. Selection has been supposed to eliminate undesirable characters, but in reality it is only able to postpone changes of ex- pression relations. The characters which the breeder would eliminate continue to be transmitted and may continue to reappear, even in the face of persistent inbreeding. 146 36 LINE BREEDING AND NARROW BREEDING. RECURRENCE OF DIVERSITY IN CROSSES BETWEEN LINE-BRED GROUPS. Further evidence that self-fertilization and in-and-in breeding are to be associated with vegetative propagation rather than with normal conjugation is found in the fact that diversity of expression reap- pears when diversity of descent is permitted to members of line-bred varieties. Normal broad breeding is accompanied by alternative expressions of characters involving continual readjustments of the internal re- lations which govern expression. Line breeding, on the other hand, avoids such readjustments. With respect to the expression of its characters a whole line-bred variety corresponds to a single individual of a broad-bred group. Diiferences between line-bred varieties cor- respond to differences between the individual members of species, not to differences between the species themselves. All the forms of line breeding yield a relatively great uniformity of characters, but there are suggestive differences in the readiness with which the latent diversity reappears. The uniformit}^ secured by vegetative propagation conceals but does not diminish the inherent diversity of expression relations. The individual diversity of seedlings of vegetative varieties is familiar to all propagators of such plants. Self-fertilized types may also show a ready recurrence of diversity of characters when sufficient diversity of descent is supplied to correspond to that of normal in- terbreeding among the diverse members of a species. In strictly self- fertilized types, like wheat, crossing between different individuals of the same variety may bring out diversities as wide as those shown by crosses between different varieties. In plants having a normal aptitude for self-fertilization the uni- formity of varieties is to be thought of as due to this fact of self-fertilization rather than to selection alone. In a strictly self- fertilized tj'^pe, selection can have reference onl^^ to the efficiency with which the characters are maintained under self-fertilization. It ap- pears to have no effect at all upon what may happen when crosses are made, even between individuals of the same variety. The adjustment which suffices for uniformity under self-fertilization would not be sufficient to maintain uniformity under in-and-in breeding. The effect of persistent selection is to conserve those lines of descent which have great uniformity of expressing relations along desired lines, for the others are rejected whenever their expression deviates in any aj^preciablo manner from the standard set by the breeder. Hence the desirability of definitely directing our selection toAvard uniformity of expression, as in the centgener method, which tests the 146 CONCLUSIONS. 37 different lines of descent with reference to the quality of uniform excellence of the progeny rather than with reference to exceptional attainment on the part of particular individuals among the progeny. In types not definitely restricted to self-fertilization selection for uniformity has a more definite influence upon the results of conjuga- tion. With such types diversity comes less readily to the surface, even with crossing, but is still able to show itself in striking forms of mutative variation. Hybrids between two related varieties of Upland cotton commonly show a much smaller range of diversity than the mutations which appear in the pure-bred parental stocks. The diversity between varieties is not great enough to overcome the effects of selection and arouse the latent diversities which appear in the mutations, but when wider crosses are made between unrelated stocks the hybrids offer a wide range of diversities, corresponding to the diversities among the mutations. CONCLUSIONS. Long-standing differences of opinion among breeders regarding the values of crossing and inbreeding can be reconciled by recogniz- ing the fact that there are three primary conditions or methods of reproduction, instead of two. The three conditions may be called broad breeding, narrow breeding, and line breeding. In broad breeding there is no restriction of descent to particular lines. Unions are freely made among large numbers of lines of descent, as in natural species, where lines of descent are united into broad, continuous networks. In narrow breeding descent is restricted to unions among only a few lines, forming a narrow network of descent. In line breeding descent is restricted to simple lines, so that no network of descent is formed. Line breeding is superior to narrow breeding, but can not be con- sidered superior to broad breeding except for special purposes of commercial production requiring a high degree of uniformity. Broad breeding is the condition of normal evolution of species ; nar- row breeding the condition in which degeneration most promptly takes place; and line breeding the more stable and uniform condi- tion desired in many domesticated animals and plants. Broad breeding is constructive, narrow breeding destructive, and line breeding conservative. The uniformity of a group is increased by restricting descent so that all the individuals are' produced from few ancestors instead of from many. Descent from many diverse ancestors maintains the individual diversity of natural species. Diversity lessens as descent is restricted, but there is also a gradual decline in the vigor and 146 38 LINE BEEEDING AND NAREOW BREEDING. fertility of the stock. The practical importance of uniformity should not cause us to overlook the fact that uniformity must be attained at the price of deterioration. We must either avoid the deteriora- tion of varieties or replace them frequently with new varieties. Though all forms of restricted descent lead ultimately to degenera- tion, the decline may be exceedingly slow and gradual if methods of line breeding are followed. Line breeding has practical superiority over narrow breeding when it preserves desirable strains of plants or animals for longer periods of time. The ability of some species and varieties to maintain themselves under line breeding is to be considered as a form of longevity, depending on the power of the protoplasm to continue its activity without new conjugations be-. tween germ cells from different lines of descent. Species and varieties differ in their ability to persist under restricted descent, just as they differ in the longevity of the individual organisms. Four forms of line breeding may be distinguished: Vegetative propagation, parthenogenesis, self-fertilization, and in-and-in breed- ing. All stages and gradations can be found, from the broad breed- ing of natural species, through the narrow breeding of ordinary domesticated varieties, to the strict line breeding of individual strains. The superiority of broad breeding over narrow breeding depends on the factor of normal conjugation between cells derived from different lines of descent, whereas conjugations between cells of too closely related lines often produce weak or abnormal offspring. If conjugations are to take place between unlike germ cells, a con- siderable degree of diA^ersit}'^ of parentage is to be maintained. If conjugations are to be limited to closely related germ cells, diversity of parentage is to be avoided. In broad breeding we imitate the methods of descent in natural sexual species. In line breeding we follow the analogy of vegetative propagation and self-fertilization. A truer idea of the value of other methods of line breeding is gained when we associate them with vegetative propagation than when we consider them as equivalents of the sexual reproduction of broad breeding. That no group of higher plants or animals relies for its reproduction upon any form of restricted descent forbids the assump- tion that domesticated varieties can be permanently maintained under conditions of restricted descent. The superiority of vegetative propagation shows that complete cessation of conjugation preserves varieties better than conjugation without diversity of descent. In parthenogenesis conjugation is likewise in abeyance, tlie vegetative growth of a new organism taking the place of the formation of new sex cells. Self-fertilization is not far removed from parthenogenesis, for the cells that unite have only 146 CONCLUSIONS. 39 recently separated from the same parent and are able to comliine without disturbins: the expression relations of the jDarental char- acters. Evidence of the superiority of line breeding over narrow breeding is also found in nature in the adaptations by which many wild species avoid narrow breeding. Such adaptations are not to be con- sidered as opposed to broad breeding, because of the many cases where the same flower has two kinds of adaptations, some favorable to broad breeding and others to self-fertilization, but both tending to prevent narrow breeding. Thus it appears that vegetative propa- gation and other natural forms of line breeding are to be considered as supplementary to broad breeding rather than as substitutes for broad breeding. The superiority of line breeding over narrow breeding depends on the factor of longevity in the protoplasm, to enable the growth of new individuals to continue without the need of frequent recourse to the physiological stimulus of conjugation. Varieties having the necessary longevity are propagated more successfully b}^ line breeding than by narrow breeding, showing that protoplasmic longevity is able to sustain the vitality of such stocks better than the conjugations which occur under conditions of narrow breeding without adequate diversity of descent. The superiority of line breeding over narrow breeding explains the improvement often wrought by closer selection in narrow-bred groups. Varieties may be improved by more rigid selection for uniformity if they are thus carried from the condition of narrow breeding toward the more favorable condition of line breeding. Varieties may be in- jured by more rigid selection when it carries them from broad breed- ing to narrow breeding, especially those varieties which can not be placed on a basis of line breeding for lack of the necessary longevity. The effect of line breeding is to restrict the expression of characters to a single individual set by suppressing the original diversity of the group. Nevertheless, the suppressed characters have a persistent tendency to return to expression, especially when opportunities are afforded by crossings between different lines of descent or by changes in external conditions. There is no warrant for believing that any method of selection can establish varieties on a stable basis so as to prevent the return of diversity and render further selection unneces- sary. Selection always appears to improve narrow-bred and line- bred varieties, not because it raises them to new standards but because it weeds out those lines of descent which have failed to maintain the old standards. The fact that different physiological principles are involved in the different methods of reproduction practiced in the various kinds of 146 40 LINE BREEDING AND NARROW BREEDING. plants and animals shows that no single generalization can be applied to the whole field of practical breeding. The success of line breeding in some cases does not warrant the advice that line breeding be ap- plied to all cases, nor do particular failures with line breeding justify any general insistence that crossing must be practiced in all varieties. The practical need is to recognize the effects that the different methods of breeding are exerting upon our varieties, so that we may -guard them against deterioration as long as possible and provide other varieties to replace them. 146 INDEX Page. Amitapsis distinguished from partlienogenesis 17 iu Oenotliera 17 Australia, sheep breeding 32 Beans, correlated characters 33 Biotypes originated by mutation ,26 Birds carrying pollen — _ 19 Breeding, balance between broad and narrow 31-32 broad, adaptations 19 and line breeding in same species 19-20 defined 8-37 supplemented by self-fertilization in natural species 18-19 centgener method 21, 36-37 constructive and conservative 25 ci'oss, definition 8 importance of longevity 39 in-and-in, allied to vegetative propagation 14 definition 8-9, 17-18 line, a system of propagation 20-23 adaptations 19 and broad breeding iu same species 19-20 by vegetative propagation 11 definition S-0, 37 improvement of varieties by selection 24 in nature 15 methods 14—15 recurrence of diversity in crosses between groups 36-37 superiority 38 to fix characters 35 vegetative propagation the most effective method 11 methods defined 9-37 essentially diverse *40 narrow, and mass selection 29 avoided in nature 19 definition 8-9,37 physiological principles 39 to fix characters 35 Bud mutations 22 infrequent 23 more frequent in new varieties 23 of coffee 23 Canary Island date palm 19 Carnations, "running out" of varieties 29-30 Cells, simple and double 13 Centgener method of breeding 21,36-37 146 41 42 LINE BREEDING AND NARROW BREEDING. Page. Central America, coffee mutations 23 Cereals, origin of self-fertilization 15 Characters, correlated 33 expression in line-bred groups 36 restricted by line breeding 39 "fixing" by line breeding 35 lettuce, correlated 33 new combinations 25 not result of breeding 25 Coffee, mutations 23 Conclusions of bulletin 37 40 Conjugation as means of rejuvenation 12 prolonged in vegetative propagation 12,13 self-fertilized, not sexual 14 Corn, correlated characters 33 diversity 33 hermaphrodite flowers 1 35 hybrid, increased yield 35 metamerism 34 Corollas specialized to secure cross-fertilization 20 Cotton, correlated characters 33 hybrids, characters 37 diversity 37 Mexican, variation in yield 21 selection in Guatemala 28 Triumph, observations 27 unnecessary uniformity 32 Criminals as examples of degeneration 24 Cross-breeding, definition 8 fertilization contrasted with self-fertilization 16 specialization to secure 20 Crosses between line-bred groups, recurrence of diversity 36-37 Darwin, experiments on self-fertilization 15 on different forms of selection 28 Date palms, vegetative shoots 19 De Vries, doctrines 26 Degeneration as result of breeding 30 in narrow breeding 10,37 Bescent as a network 9,37 in natural species 9-38 restriction, effects 9-11 Diversity in cotton hybrids ' 37 individual crosses 36 maize 33 recurrence in crosses between line-bred groups 36 Domestication of plants 15 East, E. M., on bud mutations 23 Evolution affected by restriction of descent 10 Fertility as standard of superiority 10 Fertilization, cross, contrasted with self-fertilization , 16 six'cialixation to secure 20 Flowers, hermapluodite, in rorn 35 146 INDEX. 43 Page. Gates, R. R., studies of Oenothera 16 Guatemala, method of corn planting 35 Heredity, uniformity not a normal condition 27-2S Hermaphrodite flowers in corn 35 Hybrids, corn, increased yield 35 cotton, characters 37 diversity 37 more uniform than cross-bred individuals 16 vigor 31 Improvement of plants by securing uniformity 11 Inbreeding, definition 8 Insect pollination prevented by unfavorable weather 15 Insects carrying pollen 19 Internodes, individuality 34 Introduction to bulletin 7 Java, sugar-cane experiments 23 Lanham, Md., experiment _J 34 Lettuce, correlated characters 33 Lilies, "running out" of varieties 29 Lockhart, Tex., observations of cotton 27 Longevity a factor in ability to withstand self-fertilization IS importance in breeding IS, 39 in vegetative propagation 11-13 of trees .and herbs 13 protoplasmic ; 32 stages ^ 12 standards . 12 Maize. S'ee Corn. Mebane, A., originator of Triumph cotton 27 Metamerism in corn 34 Mexican cotton, variation in yield 21 Mitapsis absent in Oenothera 17 Mutations evidence of instability 35 of coffee -• 23 vigor 31 Oenothera, absence of mitapsis 17 reproduction 16 Organisms, comparison with clocks _ — 13 Palms, date, vegetative shoots 19 Parthenogenesis S-38 and vegetable propagation 14 compared to amitapsis 17 Plants and animals, wild types, extinction 30 asexual reproduction prevalent in early-blooming species 15 domestication -^ 15 improvement by securing uniformity — 11 line-bred varieties, improvement by selection 24 parthetiogeuetic 15 perennial, first domesticated 15 reproduction, three principal types S-9 superiority, different kinds ' 10-11 146 44 LINE BEEEDING AND NARROW BREEDING. Page. Plants, varieties, " running out " 29-30 seedless, value 10 seedling, stronger 31 Potatoes, " running out " of varieties 29 Propagation, line breeding a system 20-23 vegetative 8 allied to parthenogenesis 14 and longevity 11 as method of line breeding 11 depends on longevity 11-13 relation of self-fertility 15-17 sexuality 12 suppression of characters 36 the most effective method of line breeding 11 Proterandry in maize 34 Quezaltenango, Guatemala, corn culture by Indians 35 Rejuvenescence of varieties 30-31 Reproduction, asexual, prevalent in early-blooming species 15 duality in methods 18 three principal types 8-9 Root crops, tropical, antiquity 11 ■"Running out" of varieties 29 San Antonio, Tex., experiment . 21 Selection a conservative process 24-27 by Indians of Guatemala 28 continued, necessity 27 Darwin's distinctions 28 for earliness in cotton 28 fertility 21 improvement 11 in degenerate varieties 24 " pure lines " 21-22 individual 28-29 limits 21-30 mass, and narrow breeding 29 two forms 28-29 methodical 28 most beneficial in self-fertilized varieties 26 natural 28 of line-bred varieties 24 produces uniformity 36 "pure line," difficulties 22 result mathematical 24 safest vA'heu not too closely restricted 22 suppresses characters 35 unconscious 28 Self-fertility, relation to vegetative propagation 15-17 fertilization 8 advantages 19 allied to vegetative propagation 14 cause of uniformity 36 contrast between corn and wheat 34 146 INDEX. 45 Self-fertilization, contrasted with cross-fertilization 16 developed in cereals 15 exclusive, rare 20 function 20 in maize ^^ natural species 18-19 like vegetative propagation 15 postpones degeneration 16 selection ^^ supplement to broad breeding 18 Sex, method of reproduction 38 Sexuality in self-fertilizing species 20 vegetative propagation 12 Shamel, A, D., statement ^^ tobacco studies 1^ Sheep as example of balanced breeding 32 Species a biological entity ^ elementary "^ extinction. 30 Sterility, biological and agricultural fertility 10 Strawberries, "running out" of varieties 29 Sugar cane bettered by change 23 "running out" of varieties 29 Synapsis in Oenothera 1^ Tobacco, example of self-fertilization 15 Tracy, W. W., sr., statements 33 Uniformity an agricultural improvement H lack, in progeny of single individual 22 leads to deterioration 38 limitations -'' maintenance by selection 24 not a normal condition of heredity 27-28 unnecessary, avoidance 32-35 Varieties, degenerate, of agricultural value 25 selection 24 rejuvenescence 30-ol " running out " 29-30 weakening 13 Vigor as standard of superiority 10 increased in hybrids 31 Vitality and longevity J-* 146 i-uoi iJ9 jj [Continued from page 2 of cover.] No. 80. Agricultural Explorations in Algeria. 1905. Price, 10 cents. 81. Evolution of Cellular Structures. 1905. Price, 5 cents. 82. Grass Lands of the South Alaska Coast. 1905. Price, 10 cents. 83. The Vitality of Buried Seeds. 1905. Price, 5 cents. 84. The Seeds of the Bluegrasses. 1905. Price, 5 cents. 86. Agriculture without Irrigation in the Sahara Desert. 1905. I'rice. 5 cents. 87. Disease Resistance of Potatoes. 1905. Price, 5 cents. v 88. Weevil-Resisting Adaptations of the Cotton Plant. 190G. I'rice, 10 cents. 89. Wild Medicinal Plants of the United States. 1906. Price, 5 cents. 90. Miscellaneous Papers. 1906. Price, 5 cents. 91. Varieties of Tobacco Seed Distributed, etc. 1906. Price, 5 cents. 94. Farm Practice with Forage Crops in Western Oregon, etc. 1906. I'rice, 10 cents. 95. A New Type of Red Clover. 1906. Price, 10 cents. 96. Tobacco Breeding. 1907. Price,, 15 cents. 98. Soy Bean Varieties. 1907. Price, 15 cents. 99. Quick Method for Determination of ^loisture in Grain. 1907. Price, 5 cents. 100. Miscellaneoiis Papers. 1907. Price, 25 cents. 101. Contents of and Index to Bulletins Xos. 1 to 100. 1907. I'rice, 15 cents. 102. Miscellaneous Papers. 1907. Price, 15 cents. 103. Dry Farming in the Great Basin. 1907. Price, 10 cents. 104. The Use of Feldspathic Rocks as Fertilizers. 1907. Price, 5 cents. 105. Relation of Composition of Leaf to Burning Qualities of Tobacco. 1907. Price, 10 cents. ; 106. Seeds and Plants Imported. Inventory No. 12. 1907. I'rice. 15 cents. .107. American Root Drugs. 1907. Price, 15 cents. 108. The Cold Storage of Small Fruits. 1907. Price. 15 cents. 110. Cranberry Diseases. 1907. Price, 20 cents. 111. Miscellaneous Papers. 1907. I'rice. 15 cents. 112. Thfe Use of Suprarenal Glands in the Physiological Testing of Drug Plants. 1907. Price, 10 cents. 113. Comparative Tolerance of ^'arious I'lants for Salts in Alkali Soils. 1907. Price, 5 cents. , 114. Sap-Rot and Other Diseases of the Red Gum. 1907. I'ric9. 15 cents. 115. Disinfection of Sewage Effluents for Protection of I'ublic Water Supplies. 1907. Price, 10 cents. 116.- The Tuna as Food for Man. 1907. Price, 25 cents. 117. The Reseeding of Depleted Range and Native Pastures. 1907. Price, 10 cents. 118. Peruvian Alfalfa. 1907. Price, 10 cents. 119. The Mulberry and Other Silkworm Food Plants. 1907. Price, 10 cents. 120. Production of Easter Lily Bulbs in the United States. "lOU.S. Price. lU cents. 121. Miscellaneous Papers. 1908. Price. 15 cents, 122. Curly-Top, a Disease of Sugar Beets. 190S. Price, 15 cents. 123. The Decay of Oranges in Transit from California. 1908. Price, 20 cents. 124. The Prickly Pear as a Farm Crop. 1908. I'rice. TO <"ents. 125. Dry-Land Olive Cultui-e in Northern Africa." 1908. Price, 10 cents. 126. Nomenclature of the Pear. 1908. Price, 30 cents. 127. The Improvement of Mountain Meadows. 1908. Price, 10 cents. 128. Egyptian Cotton in the Southwestern LTnJted States. 1908. Price, 15 cents. 129. Barium, a Cause of the Loco-Weed Disease. 1908. Price, 10 cents. 130. Dry-Land, Agriculture. 1908. Price, 10 cents. 131. Miscellaneous Papers. 1908. Price, 10 cents. 132. Seeds and Plants Imported. Inventory No. 13. 1908. I'rice, 20 cents. 133. Peach, Apricot, and Prune Kernels as By-Products of the Fruit Industry. 1908. Price, 5 cents. 134. The Influence of a Mixture of Soluble Salts. Principally Sodium Chlorid, upon the Leaf Structure and Transpiration of Wheat, Oat.s, and Barley. 19.08. Price, 5 cents. 135. Orchard Fruits in the Piedmont and Blue Ridge Regions of Mrginia and the South Atlantic States. 1908. Price, 20 cents. 136. Methods and Causes of Evolution. 1908. Price, 10 cents. 137. Seeds and Plants Imported. Inventory No. 14. 190S. Price, 10 cents. 138. The Production of Cigar-Wrapper Tobacco under Shade in the Connecticut Val- ley. 1908. Price, 15 cents. - 139. American Medicinal Barks. [In press.] 140. " Spineless " Prickly Pears. 1S09. Price, 10 cents. 141. Part I. The Relation of Nicotine to the Quality of Tobacco. 1908. Price, 5 cents. Part II. The Granville Tobacco Wilt. 1908. Price, 5 cents. 142. Seeds and Plants Imported. Inventory No. 15. 1909. Price, 10 cents. ' 143. Principles and Practical ^Methods of Curing Tobacco. 1909. Price, 10 cents. 144. Apple Blotch, a Serious Disease *f Southern Orchalrds. 1909. Price, 15 cents. 145. Vegetation Affected by Agriculture in Central America. 1909. Price, — cents. 146