Digitized by the Internet Archive in 2011 with funding from LYRASIS members and Sloan Foundation http://www.archive.org/details/improvementofnatOOjone S Bulletin 266 March, 1925 ^2 QJunnrrttrut A^rtrultural ^£xpnmmt ^tattu« '^^ta Haopn, (Cottn^rttrut The Improvement of Naturally Cross- Pollinated Plants by Selection in Self-Fertilized Lines I. THE PRODUCTION OF INBRED STRAINS OF CORN D. F. Jones P. C. Mangelsdorf The Bulletins of this Station are mailed free to citizens of Connecticut who apply for them, and to other applicants as far as the editions permit. CONNECTICUT AGRICULTURAL EXPERIMENT STATION OFFICERS AND SIAFF March 1925. BOARD OF CONTROL. His Excellency, John H. Trumbull, ex-officio, President. Charles R. Treat, Vice President Orange George A. Hopson, Secretary Mount Carmel Wm. L. Slate, Jr., Director and Treasure r New Haven Joseph W. Alsop Avon Elijah Rogers Southington Edward C. Schneider Middletown Francis F. Lincoln Cheshire STAFF. E. H. Jenkins, Ph.D., Director Emeritus. Administration. Chemistry. Analytical Laboratory. Wm. L. Slate, Jr., B.Sc, Director and Treasurer. Miss L. M. Brautlecht, BooRkeeper and Librarian. Miss J. V. Berger, Stenographer and Bookkeeper. Miss Mary E. Bradley, Secretary. William Veitch, In charge of Buildings and Grounds. E, M. Bailey, Ph.D., Chemist in Charge. R. E. Andrew, M.A. 1 C. E. Shepard I Owen L. Nolan [ Assistant Chemists. Harry J. Fisher, A.B. W. T. Mathis J Frank C. Sheldon, Laboratory Assistant. V. L. Churchill, Sampling Agent. Miss Mabel Bacon, Stenographer. Biochemical Laboratory. Botany. Entomology*. T. B. Osborne, Ph.D., Sc.D., Chemist in Charge. G. P. Clinton, Sc.D., Botanist in Charge. E. M. Stoddard, B.S., Pomologist. Miss Florence A. McCormick-, Ph.D., Pathologist. Willis R. Hunt, M.S., Graduate Assistant. G. E. Graham, General Assistant. Mrs. W. W. Kelsey, Secretary. W. E. Britton, Ph.D., Entomologist in Charge; State Ento- mologist. B. H. Walden, B.Agr. M. P. Zappe, B.S. Philip Garman, Ph.D. Roger B. Friend, B.S., Graduate Assistant. John T. Ashworth, Deputy in Charge of Gipsy Moth Work, R. C. Botsford, Deputy in Charge of Mosquito Elimination. Miss Gladys M. Finley, Stenographer. Assistant Entomologists. Forestry. Walter O. Filley, Forester in Charge. A. E. Moss, M.F., Assistant Forester. H. W. HiCOCK, M.F., Assistant Forester. Miss Pauline A. Merchant, Stenographer. Plant Breeding. Donald F. Jones, S.D., Geneticist in Charge. P. C. Mangelsdorf, M.S., Graduate Assistant. Soil Research. M. F. Morgan, M.S.. Investigator George C. Scarseth, B.S., Graduate Assistant. Tobacco Sub-station at Windsor -, In Churge. N. T. Nelson, Ph.D.. Plant Physiologist. The Wilson H. Lee Co. CONTENTS. Page The effect of inbreeding upon corn 353 Result of crossing 361 An interpretation of hybrid vigor 364 The transitory nature of hybrid vigor 369 Inbreeding after crossing 371 The attainment of complete homozygosity 374 Mutations in corn 375 The value of inbreeding 377 Possibility of obtaining vigorous inbred strains 380 Selection in self-fertilized lines 3S2 Method of pollination 385 Selection of ears for planting 385 Elimination of self-fertilized lines 386 The production of abnormalities 390 The approach to uniformity and constanc}^ 399 Differences in the selected lines 401 Susceptibility to disease 404 Criterions of selection 410 Classification of selected lines 411 Correlation between the first and last generations 412 Limiting factors 415 Conclusion 417 SUMMARY. The results of previous investigations on inbreeding corn are reviewed to show the development of the method of selection in self -fertilized lines. Four varieties of corn have been self -fertilized and selected for five generations. Eighty-six lines were started and twenty of these were lost or discarded. The method of procedure was to grow three progenies in each line and self -pollinate five of the most desirable appearing plants in the best progeny each year. A large number of clear-cut recessive abnormalities appeared during the course of the inbreeding. In all except one case these were eliminated by the fifth generation. No significant difference in yield was found between segregating and non-segregating progenies in lines showing recessive abnormal- ities in the previous generation. Also lines having recessive abnormalities at the start showed no greater reduction in yield during the five generations than lines that were free from them throughout the experiment. All lines showed a marked reduction in yield and a slowing down of the rate of growth. Although great differences were shown, no lines were as productive as the original variety. No appreciable correlation was found between the characters of the seed ear, weight of seed, size of seedling, or the appearance of the plants at pollinating time and the production of grain in the same genera- tion. Some correlation in certain characters was found between the first and last generations, particularly in height of plant and in per cent, of moldy ears. Less association was shown in amount of tillering and in smut infection, while in productiveness practically no relation was found, showing that good and poor yielding strains may come from productive or unproductive plants at the start. THE IMPROVEMENT OF NATURALLY CROSS-POLLI- NATED PLANTS BY SELECTION IN SELF-FER- TILIZED LINES. I. The Production of Inbred Strains of Corn. D. F. JONES and p. c. mangelsdorf The improvement of naturally self -fertilized plants, particularly the small grains, has gone steadily forward following the develop- ment of effective methods of procedure. In contrast to the older methods of mass selection based upon appearances, stands the system of individual plant selections chosen on the basis of the performance of their progeny, as worked out by Louis de Vilmorin in 1856 and later appHed by Hjalmer Nilsson in 1891 at Svalof in Sweden and by W. H. Hays at the Minnesota Agricultural Experi- ment Station in 1892. Although the early methods of applying the progeny performance test involved much unnecessary^ effort, the principle was sound and its extensive application has resulted in a large number of valuable new varieties of important crop plants, notably wheat and cotton. The theoretical soundness of this procedure, first applied in an empirical way, was later fully established by the re-discovery and demonstration of Mendel's Law, which postulates that a large part of inherited variability is due to the recombination of stable units. This led directly to Johannsen's genotype conception of organisms which appear alike but breed differently and those which are themselves diverse but give similar offspring. The improvement of naturally cross-fertilized plants, reproduced by seeds, is in no such satisfactory situation. The variation brought about by Mendelian recombination makes it very difficult to have any adequate control over the heredity when inter- pollination is continually going on. Moreover, intensive selection for particular characters often results in decreasing the niimber of hybrid combinations and this, like all other forms of inbreeding, brings about a reduction in vigor. Any advantage which might come about from the concentration of desirable germplasm is offset by the loss of growth due to consanguinity. Com, a monoecious plant and wind pollinated, is almost com- pletely cross-fertilized in every generation. This mode of pollina- tion has brought about a condition in which a continuation of the same degree of germinal heterogeniety is necessary to maintain full vigor. The experimental results of inbreeding and crossing and their theoretical interpretation show clearly why the methods aimed at the improvement of com in the past have been largely fruitless. Formerly the selection practiced with this plant was largely based upon the appearance of the mature ear. Investiga- tion has shown that com has now been brought to such a high plane of development that the correlation between the appearance (349) 350 CONNECTICUT EXPERIMENT STATION BULLETIN 266. of the seed and the productiveness of the crop grown from that seed is very low; so low in fact that it is often possible to get as good results from planting the poorest looking ears to be found in a field as from the choicest specimens. This is due to the fact that hybrid combinations of hereditary factors which make possible high production can not be transmitted intact and there- fore the offspring of any exceptional individual can not all be equally productive. An early appreciation of this situation following the application of experimental methods to the study of com breeding led to the ear-to-row system in which selection was based on the performance Figure 16. The seed from these large and small ears yielded the same. Their difference in size is due, not to heredity, but to the place where the plants that produced them happened to grow, one lot in a good, the other in a poor situation. This shows the complete lack of correlation in this case between the appearance of the seed ears and their performance. of the progeny instead of the appearance of the seed parents. Although the progenies differed markedly in yield those above the average failed to maintain their high production in later genera- tions. In 1908 G. H. ShuU outlined a method of com breeding radically different from any previously followed. In this he called attention to the large number of germinally different types which exist in every field of com and suggested that these cotild be separated out INTRODUCTION 351 by inbreeding. Although vigor was lost by this process this was to be regained by crossing inbred strains and utilizing only the first following generation in which hybrid vigor is at its maximum. East also advocated the same method and reached the same con- clusions as to the importance of hybrid vigor, as the result of independent observations on the effects of inbreeding and hybrid- ization. The crossing of different varieties of com had been advocated long before this by Beal at the Michigan Agricultural Experiment Station, and Morrow, Gardner and McCluer at Illinois. Two important contributions to methods for com im- provement were made by Shull and East. One was making clear the complex germinal constitution of a variety in a cross-fertilized plant such as com and the way in which the composition of any particular individual is masked by hybrid vigor. The other was in showing that the maximiim degree of hybrid vigor could be secured by first reducing the plants to homozygosity and then crossing, thereby bringing about the greatest number of hybrid combinations of hereditary units. Both East and Shull con- sidered hybrid vigor as a physiological stimulus resulting from the condition of hybridity itself, differing from the specific action of individual hereditary factors. For this reason they stressed the importance of securing the maximum effect of hybrid vigor. The more important service of inbreeding in automatically eliminating abnormalities and serious weaknesses and in making possible the detection and isolation of the potentially most valuable germ- plasm was not fully appreciated at first by those who attempted to apply this method to com improvement. For that reason the full utilization of the pure line principle was delayed until hybrid vigor was shown to be merely the expression of dominant hereditary factors. This brought out clearh^ and forcefully the great value of inbreeding as a means of obtaining the finest hereditary material existing in a cross-fertilized plant like com by controlling the inheritance through the pollen parent as well as through the seed parent, and fixing this in such a way that it would not be lost. Following up this line of attack a method of corn improvement was outlined in 1920 under the general title of "Selection in Self- fertilized Lines.".* It is here proposed to review the results of inbreeding and crossing which have led to the development of this method and show how inbreeding can best be applied to the im- provement of com and other naturally cross-fertilized plants. As the application of this method is still in progress the plan is to publish the results in a series under the general heading of "The Improvement of Naturally Cross-PoUinated Plants by Selection in Self -fertilized Lines." The first of this series, submitted in the following pages, deals only with the detection and isolation of desirable hereditary qualities in com, that is, the production of inbred strains which possess either in visible expression or in *Jour. Agronomy, 12:77-100. 352 CONNECTICUT EXPERIMENT STATION BULLETIN 266. potential power those valued characters that make for increased production. Later publications are planned to deal with the test- ing and utilization of inbred strains of com and the application of the same principle and method to other cross-fertilized plants. Figure 17. Two inbred strains from the same variety that have been grown side by side for eighteen years. The difference in abiHty to stand erect is inherited. THE EFFECT OF INBREEDING UPON CORN 353 The Effect of Inbreeding Upon Corn. All of the main types of com such as dent, flint, sweet, pop and flour corn have been inbred by self-fertilization for several succes- sive generations. The results have been the same in general for all types. Particular attention has been given to several strains resulting from a variety of Learning grown originally in central Illinois. Inbreeding was started by Dr. E. M. East in 1905. Four lines descending from three individual plants at the start have been continued to the present time under the direction of Dr. H. K. Hayes and later by the writers, and in 1923 they had been inbred by seventeen successive self-fertilizations. The results obtained have been reported from time to time. Particular reference is made to "Inbreeding in Com" and the "Distinction between Development and Heredity in Inbreeding" by East, published in the report of the Connecticut Agricultural Station and in the American Naturalist, and "Heterozygosis in Evolution and in Plant Breeding" by East and Hayes in a Bureau of Plant Industry Bulletin. Later results are given in a bulletin of the Connecticut Agricultural Station under the title of "The Effects of Inbreeding and Crossbreeding on Development" and the "Attainment of Homozygosity in Inbred Strains of Maize" in Genetics by the senior writer. As the method of selection in self-fertilized lines Table I. Yield and Height of Four Inbred Learning Strains of Corn Self-Fertilized Seventeen Generations. Strain A Strain B Strain C Strain D No. of Yield Height Yield Height Yield Height Yield Height Gen. Bu. Bu. Bu. Bu. Selfed per Acre Inches per Acre Inches per Acre Inches per Acre Inches 74.7 117.3 74.7 117.3 74.7 117.3 74.7 117.3 1 42.3 60.9 60.9 59.1 2 51.7 59.3 59.3 95.2 3 35.4 46.0 59.7 57.9 4 47.7 63.2 68.1 80.0 5 26.0 76.5 25.4 81.1 41.3 96.5 27.7 86 '.7 6 38.9 7 45.4 85.0 39.4 41.8 8 21.6 47.2 83.5 58.5 88 ".6 78.8 96.0 9 30.6 78 '.7 24.8. 25.5 10 31.8 82.4 32.7 84.9 19.2 86.9 32.8 97.7 11 35.1 79.7 42.3 78.6 37.6 83.8 46.2 103.7 12 24.5 77.0 27.2 80.3 20.4 85.2 49.6 100.4 13 26.9 85.5 29.0 83.7 25.1 80.6 25.8 85.3 14 23.6 87.3 38.3 86.9 36.3 87.8 35.2 94.0 15 21.1 85.4 33.4 89.9 30.0 98.2 33.6 99.6 16 17.6 76.1 24.6 89.1 25.3 94.6 29.8 97.7 17 27.8 91.7 16.9 88.9 19.8 88.4 354 CONNECTICUT EXPERIMENT STATION BULLETIN 266. has been the direct outgrowth of these investigations as to the effects of inbreeding, a brief restime of the results obtained to date will be given here. The method of inbreeding followed in the earlier experiments was to self -pollinate a ntmiber of plants at random and use one of these as the progenitor for the following generation. Such a family descending from a single self -fertilized plant in each genera- tion is called a line or strain. The yield of grain and height of plant of four hnes from Learning during seventeen successive self- fertilized generations compared to the non-inbred variety are given in Table I. The four lines A, B, C, D, were derived at the start from three different plants. One of these was separated in the third generation into two lines, B and C. These have been continued separately since. Other lines were started from the same variety but have since been lost on account of failure to secure self -polli- nated seed. In some cases this loss has been accidental, but for the most part these strains were maintained previous to their extinction with great difficulty and showed a much greater reduc- tion in growth and vigor than the other strains which survived. Although there is wide variation in yield of grain and height of plant from year to year the general direction is downward. After the ninth generation size and productiveness have remained on about the same level. The original variety yielded at the rate of eighty-eight bushels per acre the year it was first self-fertilized. In 1916 seed of the same variety was obtained from the original source and grown in comparison with these strains, then in the ninth or tenth generation. On account of its change to a new location under conditions to which it was not as well adapted as the inbred strains, which had been grown there for many years, no strict comparison can be made. In spite of their possible ad- vantage the inbred strains were only from one-half to one-third as productive and were also noticeably reduced in height. This decrease in yield which results from a reduction in size of all parts of the plant and a lessening of the growth rate has so far been the universal result of inbreeding com as far as known to the writers. Several hundred self-fertilized strains have been grown long enough to bring this out clearly. Accompanying the lessening of productiveness and growth vigor there has been a reduction in variability. From a variety that showed the usual variation in height, color of silks, glimies and leaf sheaths, number of ears, position of the ear and other details in all parts of the plants there resulted in the four self -fertilized lines a marked uniformity among all of the plants within each line. This similarity in type became noticeable in the earlier generations of inbreeding, and after seven or eight successive self-fertilizations every plant in any one line was as much like every other plant in that line as any two plants in a naturally self-fertilized species, such as wheat or tobacco, from seed from the same individual. In other words, the vari- THE EFFECT OF INBREEDING UPON CORN 355 ability that resulted from the recombination of hereditary factors was in time eliminated. Figure 18. Two inbred strains from the same variety of flint corn, one with many tillers and the other without any. 356 CONNECTICUT EXPERIMENT STATION BULLETIN 266. Where the original variety had some plants with colored silks and others with uncolored, some of the lines now have all their plants with red silks while in others all the silks are green. In some lines the foliage on all the plants is a bright glossy green, in others a dull bluish green. All the plants of one of the lines re- main green and stand firmly erect throughout the season while in other lines the foliage turns yellow towards the end of the growing season and in still another the plants frequently go down on account of a weak root system. Differences in susceptibility to smut are shown in these four strains as brought out in table II. In every detail of structure of the plant, including tassel and ears, all the individuals of one line are remarkably alike and noticeably different from the other lines. Some of these differences are shown in the accompanying illustrations, figures 17, 18 and 19. The uni- formity within the line and the differences between the several lines are brought out statistically in tables III to VI, which show the height of plant, length of ear, n amber of nodes and rows of grain on the ear for the original variety and the four strains derived from this variety. Table II. Per cent, of Plants Showing Smut Infestation in Fonr Inbred Learning Strains. Strain 1917 1918 1919 1920 1921 1922 1923 Ave A .3 .7 1.9 14.3 15.2 3.0 .0 5.1 B 9.8 25.9 8.6 32.8 50.0 27.3 69.0 31.9 C .5 9.1 4.1 6.0 13.8 17.5 52.7 14.8 D .0 1.0 1.4 25.0 4.1 2.2 .7 4.9 During the early generations of self-fertilization various forms of abnormalities appeared. The most frequent of these are seedlings wholly or partially lacking in chloroph3dl, various types of striped plants, golden plants, dwarfs, plants with ears showing many poorly developed and aborted seeds, and others with sterile tassels and ears. These are a few of the more strikingly aberrant types. Some of these are able to produce seed and when self-fertilized come true to their abnormal condition. Others are wholly in- capable of reproduction and are eliminated, but the inbred strains in which they appear may continue to produce them regularly as part of their offspring in the following generations. After several generations these abnormalities are usually no longer produced and the remaining plants are all normal in type although reduced in size and in rapidity of growth. Many of the abnormal forms which appear in large numbers in the inbred families are occasion- ally seen in fields of com which have never been artificially self- fertilized. Obviously, inbreeding is not responsible for their creation. They are recessive in mode of inheritance; that is, when crossed with other plants the following generation is all normal but the abnormality reappears in the subsequent generations. THE EFFECT OF INBREEDING UPON CORN 357 K^^^^Hii^ Figure 19. Differences in height of two inbred strains from the same variety self-fertilized four generations and selected for vigor and produc- tiveness but not for height. 358 CONNECTICUT EXPERIMENT STATION BULLETIN 266. td ^ O hJ PL, ffi Q "3 '^. +j 00 'S '^ OS ^ 4^ 00 t^ (M '^ as 00 CD o O) lO 00 CO CO i-H CO 41 -« 41 4^ 41 c^ O 1—1 OS T— 1 CD 00 I— 1 CD (M ■> < m O Q +^ a c C c •r^ m rt 03 03 a tH Vh Vi 1-1 +J +j > CO M 02 W Q '-' o < .2 H F3 Ph 41 41 41 41 -H CD -t^ -* 00 lO O o o o o < 41 41 41 41 41 lO o (M (M CO I> •o TJH lO o 00 Th ,_, LO a> 2; C<1 I— 1 y-t GO c^ ""^ (M IM lO I^ 00 CD t^ >1 < m O a a c c c OJ 03 o3 03 03 Vh t- u -u > C/3 C/D (n a: THE EFFECT OF INBREEDING UPON CORN 359 1—1 CO CO CO CM > o ^ t^ t^ o lO 00 00 CO CO o to o o o O < i\ 4^ o 4^ o lO 41 •* CO (M CO CO CO O CO s l2 C5 CD CD to 05 CO 1—1 CO c o CO CO - lO lO CO CO lO 00 o (M - (M 1—1 05 (M 1—1 T-l 00 1—1 >^ < m O G 'v-i > S-i I- .S o w o 1— 1 > d nl o ca < M-l H o f^ pa I-f D CD o 1—1 CO 00 CO > 4 CM CM 41 -H 00 CO 41 CO CM 41 o 1—1 00 05 cri o C<1 CO o o 1—1 c << 4^ 41 lO 41 as CM 00 1—1 CO T-l CM 1—1 i-O z C<1 CM 1—1 1—1 CD o CM lO 00 lO CM to '^i CD • t^ CO 1—1 00 1) J2 O c o ■5 ^ CM Id CO lO -' CM o "o ^2 lO C<1 Ci < PQ o Q 'u > c CO +-> 360 CONNECTICUT EXPERIMENT STATION BULLETIN 266. In ordinary fields of com they are generally kept out of sight by continual crossing with normal types which are dominant. Plants carrying such factors for abnormality, when self -fertilized, produce them in approximately one-fourth of their progeny. Some of the normal plants in the same progeny carry the abnormality and some do not. Sooner or later, progenitors are used which do not carry any of these striking abnormalities, after which they cease to appear. The rate at which reduction in growth takes place and the final size and productiveness of the several lines, after the reduction comes to an end, vary in different lines. Of the four Learning strains the D line has regularly been taller and larger and has yielded more than the others. The rate of reduction has been nearly alike in all of the four lines although A was reduced in yield somewhat more quickly than any of the others. The attainment Figure 20. Comparative production of a variety of Learning corn, two inbred strains derived from this variety, and their first generation hybrid. Grown in adjoining rows, they yielded 96, 32, 20 and 115 bushels per acre respectively. of uniformity may also proceed at a different rate, depending upon the degree of heterozygosity of the plant chosen as progenitor. Some strains remain variable for many generations while others become uniform in nearly every feature after a few generations of self-fertilization. From the foregoing facts it is obvious that inbreeding is a process of sorting out. From a mixture of many genetically different individuals all varying in hereditary composition and in heterozy- gosity any number of homozygous lines can be ultimately obtained, each differing to a greater or less degree from every other. A naturally cross-fertilized species is thus changed into an artifically self -fertilized species. In uniformity and constancy these artific- ially inbred plants are quite comparable to naturally self -fertilized species, with the important difference that in com they are mark- edly reduced in size and vigor. result of crossing Result of Crossing. 361 The vigor which is lost by inbreeding is at once restored when two self-fertiHzed Hnes descending from different plants at the start Figure 21. Two inbred strains and their first generation hybrid show- ing differences in time of flowering. are crossed. This is shown in figure 20. Here the ears produced by the original non-inbred variety are shown in comparison with the ears produced by two Hnes self -fertilized 12 generations and the 362 CONNECTICUT EXPERIMENT STATION BULLETIN 266. first generation hybrid between these two Hnes. An equal number of plants of the four lots were grown in adjoining rows and yielded 96, 32, 20 and 115 bushels per acre respectively. A comparison of a large number of first generation crosses between inbred strains derived from the same variety showed that the yield of the hybrids was increased 180 per cent., height of plant 27, length of ear 29, number of nodes 6, and rows of grain on the ear 5 per cent, above the average of their inbred parents.* From this it is seen that size characters such as height of plant and length of ear are affected more noticeably by hybrid vigor than the number of parts, such as nodes and rows of grain on the ear, while yield, which stmis up Figure 22. Representative ears of three inbred strains of dent corn and two first generation hybrids resulting from the crossing of the two adjoining types, harvested at the same time to show the difference in maturity. the entire growing capacity of the plant, is increased more than anything else. In other words hybrid vigor has much the same effect as favorable environmental factors. Fertile soil, good season and careful cultivation influence the growth of the com plant. Under these conditions corn grows taller, the ears are larger and the production of grain is much greater than under the less favorable conditions, while the number of nodes or the rows of grain on the ear are not so much changed. *"The effects of inbreeding and crossbreeding upon development." Connecticut Agric. Exper. Station Bull. 207. RESULT OF CROSSING 363 Another noticeable effect of crossing inbred strains of com is that of hastening the time of flowering and maturing. Figure 21 shows two inbred strains in which the tassels are just beginning to appear. No silks are out. The first generation hybrid of these two strains in the center is shedding pollen from nearly all of the tassels and the silks are well out on many of the plants. Representative ears of three inbred strains and first generation hybrid ears resulting from the cross of the two adjacent strains are pictured in figure 22. All were picked at the same time and show the greater maturity of the hybrid ears. All of the combinations of inbred strains have shown increased Figure 23. A first generation hybrid showing the uniformity in height and in tassel type. The two inbred parental strains are in the adjoining rows at the left. growth and yield whether the parental strains come from the same original variety or from different varieties. Some combinations have yielded more than others. A few have been better than others in many respects. Crosses between strains from different varieties have not been conspicuously better than crosses within the variety although no extensive test of this point has been made. Furthermore, no reliable comparison of the yield of the hybrids with the original variety can be made because this variety is not well adapted to the local conditions in which the self-fertilized 364 CONNECTICUT EXPERIMENT STATION BULLETIN 266. lines have been grown for many years. Kiesselbach reports the average yield of seven first generation hybrids tested two years as 52 bushels per acre in comparison with 42 bushels for the original variety. This is an increase of 24 per cent. The highest yielding hybrid produced 59 bushels or an increase of 40 per cent. The most noticeable and important feature of the first generation hybrids between fixed inbred strains is the even gro^vth, similarity in size and structural details and uniform production of all plants where the growing conditions are equal. This is shown for height of plant and tassel type in figure 23. Barring accident every plant is like every other plant. They grow to the same height. All ears are borne usually at the same node. The tassels and silks appear at the same time and the plants all ripen within a few days of each other. The fact that every plant produces a good ear is a most important factor in making crosses between strains so productive. In ability to yield from every plant and in uniformity of ripening, these first generation com hybrids are equal to any naturally self- fertilized crop such as wheat and tobacco or any vegetatively propagated plant as potatoes and sugar cane. Since com is very susceptible to damage by unfavorable weather at pollinating time, the uniformity in flowering may be undesirable particularly in those regions where hot dry weather is a frequent occurrence at this critical time. For that reason some other method of utilizing inbred strains may prove to be more practicable. This will be considered more fully in later publications. It is sufficient here to point out that in these first generation hybrids we have a new kind of corn which in many important respects is radically different from the mixtures of hybrids of varying degrees of heterozygosity now constituting an ordinary field of com. An Interpretation of Hybrid Vigor. The observations of gardeners and animal husbandmen have led to a general conviction that crossing somewhat different but related plants or animals usually results in a greater gro^\'th. Many instances of this phenomenon of hybrid vigor, in which the offspring excel both parents have been noted in the higher plants and in mammals, birds, insects and some of the lower forms of animals. Largei size or more rapid growth usually results when the parents are visibly different in some respects but are sufficiently related to produce fertile offspring. Many notable cases of hybrid vigoi' also occur in wider crosses where the offspring are partially or wholly sterile. This is well illustrated by the mule, which is sterile. A similar wide cross in plants is the combination of the radish and cabbage in which the hybrid makes a luxuriant gro^^^h but sets no seed. Some species crosses show no increased vigor but on the other hand may be extremely weak. East and Hayes have given several illustrations of tobacco hybrids which are barely able to live and make only a weak growth. Many crosses AN INTERPRETATION OF HYBRID VIGOR 365 of different species in animals and plants do not develop normally. Hybrid weakness as well as hybrid vigor must be taken into con- sideration although this is not be to expected in crosses that are fertile. Afier the limits of physiological compatibility are reached cross-fertilization cannot be accomplished. A series can therefore be arranged as follows: (1) Crosses between organisms which are so nearly alike in germinal constitution that no increased growth .<^>N. / .^^i^i^-^ Figure 24. Crossed corn showing vigorous growth. results. (2) Crosses between germinally diverse but closely related organisms that grow to a larger size and at a more rapid rate and are fully fertile. (3) Sterile crosses between more distantly related organisms which are extremely vigorous. (4) Sterile crosses which are weak and often abnormal. (5) Crosses which cannot be made on account of the germinal difference in the forms united. H^^brid vigor in domestic animals and cultivated plants most frequently results when breeds or varieties of different type are brought together. Thus it is a common practice to cross the 366 CONNECTICUT EXPERIMENT STATION BULLETIN 266. bacon and lard types of hogs or the mutton and wool breeds of sheep to secure some of the advantages of both parental races. Dent and flint varieties of com when crossed usually give greater increases in yield than crosses within either type. In these diverse crosses many of the desirable features of both parental races are brought together. How this works is well illustrated in the cross of a "golden" type of com which is deficient in chlorophyll with a "dwarf" as shown in figure 26. The plants resulting from this cross are tall, normally green and quite vigorous and productive. In this particular case one parent contributes normal stature and the other normal chlorophyll. Both these characters are dominant over the recessive condition so that all the hybrid plants Figure 25. It is the uniform production of a good ear on every plan- that makes the first generation hybrids between inbred strains so product five. are alike in their tall stature and green color. Another case is shown in figure 27 of two dwarfs which are genetically different and which, when crossed, give a tall, vigorous hybrid. One of the dwarfs lacks something essential to normal height and all the plants are alike as long as they are not out-crossed. The other dwarf is lacking in some other essential factor present in normal When these two small plants are combined each type com. supplies what the other lacks so that the result is normal stature in all the hybrid plants the first year after crossing. These illus- trations of the result of crossing are extreme cases which show how conspicuous abnormalities are suppressed by crossing so that the hybrid offspring are able to make a greater growth than either parent. The same situation in principle exists in all crosses from AN INTERPRETATION OF HYBRID VIGOR 367 which hybrid vigor ensues. Different organisms possess different hereditary quahties. When brought together there is always a tendency for the hereditary factors which make for greater growth vigor to dominate the factors for lesser growth. The bringing Figure 26. The result of crossing a golden, liguleless type, on the left, with a green dwarf on the right. The hybrid, in the center, has tall stature, normal foliage and green chloropliyll due to dominant factors contributed by each parent. together of the best of both parents in this way gives the hybrid offspring a temporarv^ advantage over either parent in the first generation following the cross. Recessive weaknesses are con- 368 CONNECTICUT EXPERIMENT STATION BULLETIN 266. tinually occurrinj^ as mutations as shown by the many controlled observations on the fruit fly and other forms of life. In cross- fertihzed organisms, and particularly in domesticated animals^ and plants, crossing keeps these covered over and out of sight by combining them with normal factors. Many of these recessive weaknesses are not distinct and visible characters as are the Figure 27. Two genetically different dwarf types give tall plants when crossed, due to the fact that the normal growth factor which each lacks is supplied by the other. chlorophyll deficiency or dwarfness in com but nevertheless they weaken the organism in some way. When such crossbred races are inbred, the heterozygous combinations are reduced and the resulting individuals which are homozygous to a greater and THE TRANSITORY NATURE OF HYBRID VIGOR 369 greater degree, as the inbreeding is continued, show the recessive weaknesses and are either unable to reproduce themselves or are reduced in size and rate of growth to a point below that of the original stock. The inbred individuals each receive some of the hereditary factors for vigorous growth. Some receive more than others as a chance allotment and are therefore better able to sur- vive the inbreeding process. Others are so weakened that they perish. On account of the way in which the hereditary mechanism operates it is extremely improbable that any one individual will receive all the more favorable growth factors, and in actual practice inbred strains of com are all reduced by inbreeding. It is theo- retically possible to obtain individuals which possess an unusually large share of the more favorable growth factors or even all of them and for that reason show no reduction from inbreeding. Darwin obtained self-fertilized races of Iponiea and Mimulus which were more vigorous than the naturally cross-fertiHzed variety at the start. Cummings reports self -fertilized strains of squash that are as productive as the original variety and much more uniform in type. King has obtained inbred rats after long-continued brother and sister mating that are fully as vigorous as the material with which she started. The fact that no such result has been ob- tained with com shows how dependent this plant has become upon cross-fertilization to maintain production. The Transitory Nature of Hybrid Vigor. The increased growi;h resulting from crossing is quickly lost in the following generations when the h^-brid individuals are bred among themselves or again inbred. In other words, hybrid vigor is a temporan,' manifestation which ordinarily cannot be fixed and made permanent in sexually reproduced offspring. The reason for this is readily appreciated when the illustrations previously given are followed into the later generations. The cross of the golden and dwarf com gives all normal tall green plants in the first hybrid generation. Seed from these h^^brid plants, either selfed or inter-crossed, always gives in the next generation aU the possible combinations of characters that went into the cross. In this particular case the golden plants also lacked the ligule which is the small extension of the leaf sheath surrounding the stalk above the leaf blade. Liguleless plants hold their leaves in a characteris- tically upright position close to the stalk. In the second generation of this cross of liguleless golden by dwarf, eight different kinds of plants are produced. These are shown in figure 28. Due to the re- combination of Mendelian units, this generation is extremely variable, and while some of the tall, green, liguled plants may be as vigorous and productive as the first crossed plants this generation as a whole averages much less productive. By further inbreeding, eight distinct pure-breeding combinations of these three characters 370 CONNECTICUT EXPERIMENT STATION BULLETIN 266. can be obtained and within each type still further minor differences could be established. Crossing any two of these types gives increased growth and restores the normal condition provided the factors for normal growth are all present in one or the other type. In the same way the vigorous and productive crosses between inbred strains of com fall off in size and yield in the second genera- tion and are much more variable. This always results whether the first crossed plants are self -fertilized or are inter-crossed among themselves. If the inbred strains are uniform and fixed in their type the first generation hybrid plants are germinally all alike so Figure 28. The second generation offspring from the crossing of golden Uguleless by dwarf. Eight different combinations of these three characters are obtained by Mendehan segregation and recombination. that it is easily understood why self-fertilization and inter-crossing give the same result. To test this out two inbred strains were crossed after 14 generations of self-fertilization. A number of the hybrid plants were self -fertilized and an equal number were inter- pollinated. The seed of these two lots was planted in alternate rows, replicated three times. The self -fertilized plants averaged 76. 2 ±.57 inches in height in comparison with the intercrossed plants which averaged 73. 8 ±.70. In production of grain they stood respectively 22. 2 ±1.2 and 22.0 ±2.4 bushels per acre. In neither case are the differences significant. inbreeding after crossing 371 Inbreeding after Crossing. When the second generation plants are allowed to intercross naturally no further reduction in vigor is expected. Variability and yield should remain at the same level thereafter until natural or artificial selection eliminates certain strains. But when the second generation plants are self -fertilized there is a further reduction in size, and if the inbreeding is continued the decline in size and vigor and in variability proceeds in approximately the same way as when the parental strains were first inbred. This is shown in figures 29, 30 and 31. In this demonstration of inbreeding after crossing, two inbred strains, self-fertilized for eight generations, were crossed and the first generation plants again self -fertilized. In the second genera- tion a single plant was again chosen as the progenitor and polli- nated in the same way, and this was continued for eight successive Figure 29. The result of inbreeding after crossing. Two inbred strains at the left, their first generation hybrid adjoining, followed by seven suc- cessive generations self-fertilized. generations. Seed was saved from each year's selfing up to the fifth generation. Since com seed will not retain its germination satis- factorily for more than six years, single plants were again self- fertilized the fifth year in each generation and this seed was used from then on. All eight inbred generations were growm in 1923 along with the two parental strains as shounti in the accompanying illustrations. This demonstration has been gro\\Ti each year since the original cross was made and the yields obtained in the different years are given in table VII. Production has varied rather widely from season to season and from generation to genera- tion. This is due in part to the character of the individual plants chosen for progenitors. A ver}^ noticeable drop takes place from the first to the second generation amotuiting to over 30 per cent, as an average of the six years. Kiesselbach tested the first and second generations of eight hybrid combinations of different strains during two seasons and obtained an average of 52.2 and 27.8 bushels per acre respectively for the two generations, to be com- 372 CONNECTICUT EXPERIMENT STATION BULLETIN 266. pared with 41.7 bushels for the original com from which the inbred strains were obtained. He secured his seed for the second generation by pollinating several first generation plants with composite pollen from 15 sib plants. The reduction from the first to the second generation of nearly 50 per cent, is even greater than in our case where the plants were self -fertilized. Kiesselbach also grew a third generation from seed of interpollinated plants. The comparative yields obtained for the first, second and third genera- tions were 51.5, 29.4, and 25.6 bushels per acre. The reduction from the second to the third as would be expected from this mode of pollination is small compared with the drop from the first to the second. Continued inter-pollination should cause no further decrease in yield unless particularly unfavorable strains are isolated. The average height of these successive self -fertilized genera- tions compared with the first generation hybrid and the parental strains is shown graphically in figure 32. There is a continued Table VII. The production of grain in bushels per acre, of two inbred strains of corn and their hybrid and the Fi to the Fs generations successively self- fertilized. Year Generati ons Grown Pa Pb Fi F2 F, F4 Fi Fs F7 Fs 1917 22 6 65 56 1918 27 24 121 128 is' 1920 16 28 128 48 35 '29' io' 1921 20 13- 73 55 49 33 15 '23" 1922 20 26 160 83 74 68 49 36 '2.3' 1923 13 21 61 45 41 47 16 23 26 27" Ave. 20 20 101 69 43 44 23 27 25 27 reduction in each generation, but the decrease is much less during the last three generations than in the first four. From the first to the fifth generation there is a decline of 27.2 inches in stature and from the fifth to the eighth 8.6 inches. The rate of growth as measured by the daily gain in height is also steadily reduced as shown in figure 33, the decline being greater during the first stage of inbreeding than in the last. The dift'erences between the last two generations in all measurable characters, including yield, height, length of ear and rate of growth, are so small that it seems evident that the reduction in size and vigor is rapidly approaching an end. The last two generations are so similar in appearance that they cannot be distinguished in the field. In tassel type, foliage character, position of the ear on the stalk, and in the size and conformation of the ears these two generations are practically identical. The reduction in variability from the first to the eighth genera- tion was very noticeable in the field. One of the parent strains has green silks, the other red. The first generation hybrid plants INBREEDING AFTER CROSSING 373 all had red silks. The second, third and fourth generations segre- gated for this color while the remaining generations were all uniformly colored. Height of plant, position of the ear on the stalk, form of tassel and all structural details were noticeably uni- form in the parents and the first hybrid generation. The plants in the generations from the second to the fifth were quite variable but later became more and more uniform until in the last two generations they showed as little variation as either of the parental strains. The inbred strain which resulted from this second period of self- fertilization differs from both parental strains. In tassel, ear, and character of the foliage it is quite unlike either but is noticeably susceptible to smut like one of the parents. In other words, Figure 30. Inbreeding after crossing generations shown in figure 29. Representative plants from the Mendelian recombination has taken place so that the details of structure are altered. Apparently this inbred strain has about the same nimiber of favorable growth factors, and for that reason it is no better or no worse than the parental stocks that went into the vigorous and productive hybrid from which the new strain was derived a' few generations before. For all practical purposes the reducing effect of self-fertilization in this particular case has ceased at the sixth inbred generation. This closely parallels the course of events when the parental strains were first inbred. Theoretically the loss of vigor follows the rule of halving the remaining dift'erence in each generation. If we take an individual heterozygous for a single Mendelian pair of factors such as Aa we expect in the next generation fifty per 374 CONNECTICUT EXPERIMENT STATION BULLETIN 266. cent, of the plants homozygous for this pair of factors and having the composition A A or aa; the other fifty per cent, will on the average still be heterozygous for this factor pair; i. e. Aa in composition. In choosing a single self -fertilized individual for the progenitor the chances are even that it will be homozygous or heterozygous. This holds for any number of factor pairs and since each pair when once alike must remain so thereafter in self- fertilization the niunber of mixed pairs is steadily reduced by half in each generation. Starting with an individual 100 per cent, heterozygous, the following generations would be on the average 50, 25, 12.5, 6.25, 3.125, 1.5625, etc. Naturally the progeny of any heterozygous individual will vary greatly in composition. Some will be nearly or completely homo- zygous while others will be nearly or completely heterozygous with respect to all factor pairs. For that reason the result of any process of inbreeding depends entirely upon the composition of the individual plants which are chosen as progenitors. It is theoreti- cally possible to obtain individuals in each generation which are as heterozygous as their parents and others that are completely homozygous. For that reason inbreeding may cause no reduction in size, vigor or variability, or complete reduction may take place in a single generation. The chances that such a result will be obtained, however, are extremely remote. Actually the reduction follows the rule of halving the remaining difference very closely so that it is evident that a very large number of factors play a part in hybrid vigor. How many such factors there are, we have no way of estimating at the present. Many factors which bring about visible differences possibly have no effect upon vigor but apparently the number of them which are essential to normal development in com is exceedingly great. The Attainment of Complete Homozygosity. Whether complete fixity of type, absolute homozygosity, is possible of attainment by continuous self-fertilization has been •previously discussed. (Jones 1924.) The experimental results show that small germinal differences may remain after many generations of inbreeding. Two lines separated from one in the third generation and then continued separately for several genera- tions gave a marked increase in size when crossed, although not as' great as in the case of lines separated at the beginning, showing that two self-fertilizations had not produced much uniformity in germinal constitution. The four original Learning strains were continued as single lines up to the eighth generation. At that time they were all remarkably uniform and apparently fixed in their type. Then each line was separated into two lines which were continued separately thereafter for eight or more additional generations. At that time two of the paired lines had remained exactly alike. No visible differences in any respect could be seen. MUTATIONS IN CORN 375 One of the paired lines differed only in color of the seeds, one being noticeably brighter in color in some seasons. As the growing conditions were alike for all plants this slight difference can not be accounted for in any other way than as an heritable difference. The other paired line differed noticeably in many respects. One of the members was taller, the leaves were broader and lighter colored and the ears were larger, the seeds broader and duller in color. Crossing these paired lines gave significant increases in all measurable characters in the one strain whose paired lines were visibly different. The other strains all showed slight but appar- ently significant increases in some characters. The two strains whose paired lines showed no visible differences were again tested after fourteen generations of self-fertilization in the following way. The two strains which were distinct from the beginning were cross- ed and gave the usual vigorous and uniform hybrid plants. A Figure 31. Inbreeding after crossing. The production of grain from the plants shown in figure 29. number of these were self-fertilized and an equal number were inter-pollinated by sib plants. A careful test failed to show any differences in size or productiveness in the plants grown from these two lots of seed. If the parental strains were not germinally alike within themselves, intercrossing the first generation hybrid plants would not cause such a decrease in heterozygosity as self-fertiliza- tion. The fact that no diff'erence was shown indicates that the parental strains were completely homozygous for all factors which influence gro^vth vigor. However, this test is not a ver\' delicate one and final proof awaits the crossing of the paired lines which have been separated in the seventeenth generation and will be carried along for several additional generations. Mutations in Corn. Complete homozygosity may be impossible to attain because of spontaneous variations, mutations, occurring from time to time. 376 CONNECTICUT EXPERIMENT STATION BULLETIN 266. During the seventeen years in which the four inbred Learning strains have been under observation only two apparent germinal changes have been recorded. Until a fairly high degree of uniform- ity was reached, after six generations, various abnormalities occurred singly or in greater numbers in the rather small progenies that were grown. Presumably these were, at least in the great majority of cases, merely segregations from a heterozygous com- plex. But new characters appearing after uniformity is obtained which have not been noted previously have every indication of being mutations. Two such have been observed in different lines. One produced in the thirteenth generation a single self -pollinated ear segregating for defective seeds. All of the lines had been examined for the new character during three previous generations, without noting anj^thing of this kind, and since the character Figure 32. Graph showing the height of the two parental strains and the generations from the Fi to Fs. segregated as a single Mendelian recessive when out-crossed, there is every reason to assimie that a germinal change took place shortly before its appearance. Among approximately a thousand plants of another line, self -fertilized more than ten generations, which has always produced white cobs, four ears were found with light red cobs. The cobs of this strain are flattened and the plants are otherwise easily identified. The red cob plants were examined at harvest and noted to be typical for the strain in all respects except cob color. Neither of these changes could have been due to out-crossing. Stray pollen from any outside source immediately results in vigorous plants twice as large as the inbred plants ever grow and the crossed plants are completely changed in type. Since the mutant plants were in other respects typical plants of the strain and were no larger they could not have resulted from out-crossing. Two additional changes have occurred in other inbred material THE VALUE OF INBREEDING 377 such that they have every indication of being recent germinal alterations. One strain after five generations produced for the first time striped, variegated plants which bore no pollen or seed. They occured in later generations in about 25 per cent, of the off- spring from normal plants. Another strain after nine generations gave small narrow-leaved dwarf plants which were quite distinct from the normal plants. They produced a small amount of pollen and when out-crossed to normal plants they reappeared in later generations showing that the change was heritable. These four apparent mutations are all that have been noted in a large number of uniform strains which have been under obsen-ation for many years. Hayes and Brewbaker record the production of chlorophyll deficient seedlings in four lines out of 953 which had Figure 33. Graphs showing rate of growth (average daily gain in height) for the same generations as in the preceding ilhistrations. not shown such abnormalities previousl^^ In these cases the appearance of the abnormalities may have been due to delayed segregation, since the lines had not been reduced to uniformity and constancy. While it is evident that com does mutate, the fre- quency of these changes is so low that inbred strains, when once reduced to uniformity, are stable for all practical purposes. Some care will be needed to maintain self -fertilized lines true to type, and when recessive abnormalities appear those progenies which show them will have to be discarded. The Value of Inbreeding. This review of the effects of inbreeding and crossing upon com has been given in considerable detail because the facts learned from 378 CONNECTICUT EXPERIMENT STATION BULLETIN 266. these investigations form the basis for the method of ini])rovement by selection in self-fertihzed Hnes. In the inbreeding experiments just described no selection of superior individuals to perpetuate the strain was made. The aim was to take normal plants at random and note the outcome. Nevertheless a great deal of natural selection has taken place. All abnormalities which interfere with or markedly reduce reproductive ability have been automatically eliminated. In this way many chlorophyll deficiencies, endosperm abnormalities and inherited sterility in tassels and ears, unfavor- able conditions almost always present in every cross-pollinated 75 Figure 34. A diagrammatic representation of the actual and theoretical results of inbreeding corn. The solid lines represent strains which have already been obtained, the dotted lines those which may be expected when corn is worked with more extensively. variety of corn, have been cleaned out. But this outcome of in- breeding, valuable as it may be, is less important than the control over the heredity made possible by hand pollination and the result- ing fixity of type. In common practice, selection with nearly all cross-fertilized plants has been based on the appearances of the plant or upon the performance of the progeny, and no adequate control of the heredity brought in from the pollen parent has been possible. As generally practised, corn breeding has been similar to a system of animal breeding in which selection is carried on only with the dams paying no attention whatever to the sires. The disastrous THE VALUE OF INBREEDING 379 result that such a system would have upon purebred live-stock can readily be appreciated. With all cross-fertilized plants it would be theoretically possible to follow the method now used in animal breeding. Certain desirable individuals could be chosen as seed parents and others as pollen parents. Pollination could be made by hand and the progenies compared on the basis of their performance. There is no doubt that this system followed up as carefully as it is in mating farm animals would give equal results. But such a method is wholly impracticable on account of the small value of the individual plant. The time spent on selecting the Figure 35. Self-pollinated ears grown on selected plants of Burweirs Yellow Flint, No. 40. Each ear is the starting point of a selected Hne. These are numbered 1 to 9, top row, and 10 to IS, bottom row, left to right. parents and on polHnating each generation would not be repaid by the possible gains. Furthermore, with com, selection is greatly handicapped due to the fact that the principal objective, pro- duction of grain, is not visible until after pollination. A new method of attack, which will make possible a control of the heredity transmitted thru the pollen as well as thru the egg, is needed for all naturally cross-fertilized plants. Since inbreeding is a sorting-out process, selection carried on dtrring the time the plants are being reduced to uniformitv and constancv makes 380 CONNECTICUT EXPERIMENT STATION BULLETIN 266. it possible to look for desirable qualities with a certainty of being able to hold them, when once secured, that has never before been possible. From this viewpoint inbreeding is not so important as a method of gaining the maximum effect of hybrid vigor when the inbred strains are crossed as it is of separating out and making visible the very best hereditary qualities that may exist in a heterozygous stock. Strains when once reduced to fixity remain the same indefinitely, barring mutations. With due regard to seasonal variation, crosses between inbred strains give the same result whenever the same combination is made. The uniform production of the first generation hybrids between homozygous strains is an important feature. In this respect cross-fertilized plants are equal to self -fertilized plants in uniformity and fixity of type and have the added advantage of crossing to bring together and use in the first generation the desirable qualities within the species, which in a self -fertilized organism can be used only when recombined and fixed in a homozygous condition. It should there- fore be clearly understood that the crossing of inbred strains as such is without particular value and that the opportunity afforded to find and to fix the very best hereditary qualities possessed by a cross-bred race is the more important function of inbreeding. Crossing is merely a means of utilizing this good heredity by giving it maximum vigor. It is to be expected that many inbred strains will have only medium value and give no improvement over the original variety when crossed. The bulk of the germplasm in every population is mediocre. Of necessity only the exceptionally few will give outstanding results. For these reasons the outcome of selection in self-fertilized lines depends upon how extensively and skillfully it is applied. Possibility of Obtaining Vigorous Inbred Strains. Most of the inbred strains of corn so far produced have been reduced to about fifty per cent, or less of the production of the original cross-bred varieties. Some strains have failed to repro- duce after one generation of self-fertilization. Others have per- sisted in a weakened condition for several generations and then perished. Still other strains are able to survive, but are continued only with the greatest difficulty. The majority of the self- fertilized lines, when uniformity and fixity of type are reached, are about one-third as productive as at the start. A few are exception- ally good. They grow more vigorously and yield more than the rest and are equally uniform and fixed in their type. But even the best of these are still below the original variety in amount or quality of grain produced. On the basis of hybrid vigor being due to dominance of the more favorable factors it is theoretically possible to secure inbred strains that will show little or no reduc- tion in vigor, and a few may sometime be obtained that are even OBTAINING VIGOROUS INBRED STRAINS 381 m(3re vigorous and productive than the cross-bred variety. This is deduced from the fact that most heterozygous combinations of factors are less effective than the homozygous combinations of the same factors. Thus the cross of yellow and white corn gives a lighter color than pure yellow. The cross between a determinate gro\\i;h type of tobacco with an indeterminate growth type (Jones, 1921) which involves a single factor, differs from either parent in size of plant and number of leaves. Dominance is seldom perfect and while there is little direct evidence in this respect for characters Figure .36. Self-pollinated ears grown on selected plants of Gold Nugget, No. 105. Each ear is the starting point of a selected line. These are numbered 2 to 10, top row, and 11 to 20 bottom row, left to right. (Ear 1 was shelled before photographing. It was similar to No. 2.) which directly affect vigor there is every reason to expect that a homoz^'gous combination of all the more favorable dominant growth factors will make possible a greater development than the heterozygous combinations of the same factors with weaker allelo- morphs. However, as just noted, certain results are obtained from heterozygous combinations that can not be obtained from either factor alone. If there are many of these that play a part in growth vigor, then heterozygosity may be indispensable to maximum development. Moreover, recombinations of large nimiber of 382 CONNECTICUT EXPERIMENT STATION BULLETIN 266. factors are extremely difficult to obtain and since favorable and unfavorable growth factors are distributed indiscriminately throughout the hereditary mechanism the chances of securing self-fertilized strains of com which equal the cross-bred varieties are so exceedingly small that there is little hope of obtaining them. The most that can reasonable be expected are inbred strains which are appreciably better than any that have so far been produced. The results that have already been obtained from self-fertilizing corn, and the theoretical possibilities, some of which may be attain- ed in the future, are shown diagrammatically in figure 34. Figure 37. Self-pollinated ears grown on selected plants of Century Dent, No. 110. Each ear is the starting point of a selected line. These are numbered 1 to 9 top row and 10 to 18, bottom row, left to right. Selection in Self-Fertilized Lines. To demonstrate the value of inbreeding as a means of isolating good heredity a system of selection in self -fertilized lines was begun in 1918. Four varieties of com were chosen as material with which to work. These varieties have been grown in Connecticut for many years and are well adapted. In a variety test of long duration they have proven to be among the best in production of grain and in other qualities. The four varieties are as follows : Burwell's Yellow Fhnt, No. 30 and No. 40. An eight rowed yellow com of the Canada Flint type. The ears are medium in size, one or two on the stalk. The plants are medium in maturity. SELECTION IN SELF-FERTILIZED LINES 383 Gold Nugget, No. 105. An eight rowed yellow flint com with large ears, broad kernels and heavy cobs. The stalks are large with few suckers. The plants mature late in the season. Century Dent, No. 110. A light yellow dent corn with broad, smooth, shallow dented kernels. The ears are medium in size and have from 14 to 18 rows. The plants are medium in size and mature well in practically every season. Beardsley's Learning, No. 112. A yellow dent corn with taper- ing ears with 16 to 22 rows and small, shallow kernels. The stalks are large. This variety is later in maturing than Century Dent and is usually more productive. Figure 38. Self-pollinated ears grown on selected plants of Beardsley's Learning, No. 112. Each ear is the starting point of a selected line. These are nunAered 1 to 8, top row, and 9 to 16 bottom row. The plan of procedure was to self -fertilize a niunber of the best plants in each of these four varieties and to use each of these plants as the starting point of an inbred line. These lines were to be continued by self-pollination of the best plants in each generation until uniformity and constancy were reached. Accordingly from about 60 plants each of the four ^^arieties gro\\Ti from a general mixed lot of seed, 20 plants of each variety were selected at pollinating time and self-fertilized. These four lots of ears are shown in figures 35 to 38. Some of the seh'-pollinated plants 384 CONNECTICUT EXPERIMENT STATION BULLETIN 266. failed to set seed but all of the ears that had enough seed to work with were planted. The original hand-pollinated ears were ranked according to their appearance in size, form of the ear and quality of the seed. Ear number one represents the best, number two the next best and so on down. The ear numbers became the numbers for the self -fertilized lines derived from them. Therefore, the number of the line shows how its original progenitor was classi- fied. It is of considerable interest to note to what extent good strains can be obtained from unpromising ears at the start. Each self -pollinated ear was planted in a row the following year and five plants of each were again selected at pollinating time as the most desirable and were self -fertilized. It was noted that the best appearing plants at the tirrie of pollination were not always ••PLANTS oj : ORIGINAL VARIETY LINE A I LINE C i o o Figure 39. Diagram of a method of selection in self-fertilized lines. An individiial plant becomes the starting point of each inbred strain. Three progenies are grown but only one is selected to continue the line. the most productive at maturity. For this reason more plants were self -pollinated than there were progenies planted, thus allow- ing for some failures of pollination and also to permit of some selection among the hand pollinated ears. Also, in order to base selection upon progeny performance rather than upon the appear- ance of the seed ear, three progenies from each line were grown each year. At pollinating time the best appearing progeny was chosen and five plants were again self-fertilized, the other two progenies being discarded. This method of carrying on selection is shown diagrammatically in figure 39. About thirty plants were grown in each progeny. From three to five times this niimber of seeds was planted and the poorest SELECTION OF EARS FOR PLANTING 385 seedlings pulled out after they were well started, leaving the tallest and most vigorous plants. An even stand was obtained in most cases. The end plants in each row were usually avoided in selecting the plants for hand-pollination as these are nearly always larger and better developed than the others on account of their better opportunity to grow. METHOD OF POLLINATION. The plants were pollinated by hand as shown in figures 40 and 41 . The general method used is as follows:* A three pound manila grocer's bag is placed over the ear shoot before the silks appear. The tassels are covered with an eight or ten pound bag as soon as they are above the upper leaves. When the silks are about three- fourths out, pollen is dusted over them and the tassel bag placed over the ear. Care is taken not to touch the silks or the inside of the tassel bags with the hands in order to "avoid contamination with foreign pollen. If the silks extend more than three or four inches beyond the tip of the ear the}^ are cut back with a knife sterilized in alcohol. After the first generation or two, out-crossed plants can be easily noted by their much greater size and darker green color so that contaminating pollen is not a cause for great concern. Effort is made to pollinate as rapidly as possible. Only one application of pollen is made. If sufficient seed does not result from this application the ears are not used. Some good plants are lost because all the pollen has been shed and has lost its viability before any silks appear. This tendency to protandry is accentuated in some inbred lines. Such strains could be main- tained by sib-crossing but since this method of inbreeding is much less effective than self-fertilization in bringing about homozygosity the latter system has been rigidly adhered to. In this way sterility and recessive abnormalities of all kinds are most quickly eliminated. SELECTION OF EARS FOR PLANTING. Each hand-pollinated plant is tagged with a printed form upon which notes as to the character of the plants in the field and the hand-pollinated ears when mature are entered as follows: Pedigree number Color and markings of foliage Field plot number Infection on plant Height to ear-bearing node Smut on ear Height to first branch on tassel Mold on ear Number of ears containing seed Number of rows of grain on ear, Number of leaves regularity of rows, and length of Number of tillers ' ear Posture, whether erect, leaning, Color and general character of seeds bent, broken or fallen Color and shape of cob. * A method of pollinating proposed by Jenkins and known as the "bottle method" was also tried. Under our conditions it did not prove as satisfactory as the procedure described here. 386 CONNECTICUT EXPERIMENT STATION BULLETIN 266. At harvest these tags are transferred to the hand-pollinated ears. In choosing the three ears for planting in each line, from the five ears pollinated, the characters of the plants in the field as well as the size and appearance of the ears are taken into consideration, chief attention being given to ability to stand erect, color of foliage, freedom from smut and other infection on the plant and ear and absence of mold on>the ears. ELIMINATION OF SELF-FERTILIZED LINES. In all, 86 self-fertilized lines were started, distributed among Figure 40. Plant bagged for hand pollination. Small bags can be used over the ear shoot and the tassel bag placed on the ear when polli- nated. Wire clips are now used to hold the bags on the ear and tassel. the four varieties as follows: From Burwell's Yellow Flint number 40 there were 18 ears self -pollinated in 1918, ranked and numbered from 1 to 18 in order of their excellence as shown in figure 35. In addition to these there were 14 ears of the same variety which had been self -fertilized in 1914 for another purpose and not used. These were included among the Bur^vell strains with the variety ntunber 30 to distinguish them from the other strains which were ranked according to their appearance. The fact that these ears had been held five years before planting has interest in connection with the possible elimination of abnormal- ELIMINATION OF SELF-FERTILIZED LINES 387 ties due to the age of the seed, as will be noted later. From the Gold Nugget variety, number 105, twenty lines were started (figure 36); from Century Dent, number 110, eighteen lines (figure 37), and from Beardsley's Leaming, number 112, sixteen lines were started (figure 38). The once self-pollinated ears beginning these 86 lines were planted in 1919 and hand-pollinated ears were obtained from all lines except one in Gold Nugget and two in Century Dent. These failures to produce seed in all five pollinations in each line may have been due to delayed pollination and unfavorable weather conditions. But since good ears were obtained in the other lines Figure 41. Pollinating corn. Only one man is necessary' for this opera- tion. Care is taken not to touch the silks or the inside of tassel bag. If the silks are more than three inches long they are cut back to about one inch with a knife sterilized in alcohol. it is fair to assume that these lines were less vigorous or for some reason were not as able to reproduce under this method of polhna- tion. In the second generation two more lines were lost because no self -pollinated seed was obtained. In the third generation four lines were discontinued. In two of these no hand-pollinated ears were obtained, and the other two were so badly damaged by ■mold that they were discarded. In the fourth generation eleven lines were eliminated. Nine -were discarded because they were so \'ery poor and unpromising 388 CONNECTICUT EXPERIMENT STATION BULLETIN 266. that it was thought advisable not to carry them further. Some of these failed to produce any seed on any plants. All of the hand- pollinated ears of two lines proved to be out-crossed, due possibly to the fact that the bags covering the ears of the previous genera- tion were broken and allowed foreign pollen to enter. By the fifth generation practically all of the lines had- become uniform and stable. All that had survived up to this point gave promise of being able to continue indefinitely if sufficient effort was put forth and provided the season was not too unfavorable. During the course of the five-year selection period the following lines were eliminated for various reasons : Figure 42. Self-fertilized ears showino; defective or aborted seeds. In No. 30, line 2 was accidentally lost. In No. 40, lines 2, 5, 11, 12, 17, 18 were discarded. In No. 105, lines 1, 4, 5, 12, 19 were lost or discarded. In No. 110, lines 8, 12, 13, 14 were lost or discarded. In No. 112, lines 2, 5, 11, 13 were discarded. In all, 20 lines were not continued to the end of the fifth genera- tion. Three of these were accidentally lost thru no fault of their own. The others were too poor to be carried along. An examina- tion of the original ears from which these lines came (figures 35 to 38) shows no marked relation between their poor behavior and their appearance when first pollinated. Dividing each lot of ears into two equal groups and not counting the three lines that were ELIMINATION OF SELF-FERTILIZED LINES 389 accidentally lost, we find that seven from the best appearing lines at the start were discarded and ten from the poorest. The original plan was to keep all lines that could be successfully propagated even though they became extremely poor. It was fully appreciated that inbred strains may themselves be very Figure 43. Seedlings lacking chlorophyll are common hereditary- variations in corn. undesirable and still have potentially great value when crossed with other strains. For this reason no lines were discarded unless the amount of seed produced was so small that enough plants to permit satisfactory measurements could not be grown. Many lines were continued which were extremely weak, unproductive and showed markedly undesirable characters. They were continued 390 CONNECTICUT EXPERIMENT STATION BULLETIN 266. to compare them in crossing with other strains. The results of these comparisons will be reported in a later pubhcation. It should be emphasized here that the 20 lines, or 23 per cent, of the original number, which were lost or discarded, represent for the most part extremely poor and undesirable material that would probably be lost in any selection experiment. By growing a larger number of plants in order to give a greater opportunity for selection and by hand-pollinating a larger number of individuals it would probably have been possible to continue many of these lines and some might even have turned out to be good strains in the end. Whether it is worth while to work more intensively with a few lines or expend the same amount of time on a larger number of strains less intensively selected is one of the most important problems to be considered. Figure. 44 Various tj^pes of chlorophyll deficiencies found in inbred strains of corn. THE PRODUCTION OF ABNORMALITIES An examination of the original ears after the first self-fertilization (figures 35 to 38) showed eight that were segregating tor small, dull colored seeds that were clearly abnormal. These recessive seeds varied on different ears from almost entirely empty pericarps to seeds nearly normal in size but shriveled and opaque in appear- ance, as shown in figure 42. These aborted seeds, in most cases, failed to grow and those which did germinate, produced abnormal seedlings none of which reached maturity. The normal seeds from ears showing defectives when planted produced segregating ears on some of the plants in the following generation. In addition, five ears which were not clearly segregating in the first generation produced some ears with abnormal seeds in their second generation progenies. It has since been found that this defective seed condi- tion is due to a large number of lethal or semi-lethal factors which are hereditarily distinct. They are wideh' distributed in all kinds of corn. In cross-pollinated plants only a few of these abortive PRODUCTION OF ABNORMAfilTIES 391 Figure 45. A chlorophyll-deficient dwarf compared to a normal plant in the same family. 392 CONNECTICU*r EXPERIMENT STATION BULLETIN 266. seeds are seen on any ears and these are not conspicuous. It is quite possible that the plants carrying these factors in a hetero- zygous condition may be seriously weakened by them and for that reason the elimination of these lethal endosperm factors is probably important. When the first generation self -fertilized ears were grown, chloro- phyll-deficient seedlings appeared as Mendelian recessives in fifteen lines. Eight of these segregated for white seedlings, one yellow, three yellowish green and three light green. These abnor- mal seedlings were quite distinct and most of them died as soon as the food stored in the seeds was exhausted. Several distinct types of striped and variegated plants which represent various forms of chlorophyll deficiency were observed and are shown in figure 44. Other clear-cut abnormalities which appeared in the first genera- tion as recessive segregates were golden plants in four lines, various forms of dwarfs in three lines, sterile tassels which pro- Figure 46. Various types of dwarfs found in inbred strains of corn. duced no pollen in five lines. Barren plants without ears and which had the appearance of being simple Mendelian recessives were found in three lines but the inheritance of such steiility factors has not been definitely proven. In addition to these common abnormalities some new characters were found which had not been observed in other material. A few plants of one line bore ears with no silks and such plants were therefore entirely sterile in the pistillate parts as shown in figure 47. Good pollen was produced and when crossed on to normal plants the silkless ears reappeared in later generations. This character was not found until the second generation. It may have occurred the first year and been overlooked. Another strain produced square cobs and another had ears with many silks in place of one for each seed. This latter character failed to reappear in later generations and apparently was not inherited, or at least not as a simple recessive. IVIany other variations from normal occurred. They differed in degree of abnormality, some affecting the plants much more seriously than others. PRODUCTION OF ABNORMALITIES 393 In twelve lines no abnormalities were noted in the first two generations, but in the third or fourth generation, various types appeared, in the form of chlorophyll-deficient seedlings, striped and variegated plants, dwarfs, seedlings with tube leaves instead of normally flat, and plants with only the mid-ribs in place of normal leaves. In some of these cases recessive segregates may not have appeared in the first generations on account of elimination due to poor germination or they may have been thinned out with Figure 47, family. Silkless ears compared to normal specimens from the same the weaker seedlings. In some cases, however, there seems to be no question that they are due either to original mutations or to delayed segregation resulting from some complicated mode of inheritance. A good illustration can be given in the production of the narrow-leaved plants shown in figure 48. Such a striking variation as this could not be easily overlooked. All the selected lines were carefully examined for abnormalities throughout the season, beginning with the early seedling stage. Narrow-leafed plants were first observed in the third generation in lines 112-13 394 CONNECTICUT EXPERIMENT STATION BULLETIN 266. Figure 48. Plants with narrow leaves occurred in two inbred lines. PRODUCTION OF ABNORMALITIES 395 and 112-14. All three progenies of 112-13 produced some abnormal plants; two, nine and eleven narrow-leafed individuals appearing in the different progenies in a total of about 25 plants in each. This line had been segregating previously for dwarfs, golden plants, yellowish seedlings and striped dwarfs. Line 112-14 produced one narrow-leafed plant in the third generation. Though only normal plants were self -pollinated in the third generation, all of the fourth generation plants in line 112-13 were abnormal, being short and with streaked and wrinkled leaves varying in width from a mere mid-rib to nearly full width. The plants were so poor that no self -pollinated ears were obtained and the line was lost. Line 112-14 produced no narrow leaves in the fourth generation. All the plants were described as uniform, leafy but short in stature. In the fifth generation three progenies, all from ears borne on normal plants in the fourth generation, were grown. No plants were obtained from one and only a few in the other two. All of these had typical narrow leaves and were badly stunted. They made a feeble growth and produced no ears. Pollen from typical narrow-leafed plants of the third generation out-crossed on to normal plants failed to show any abnormal plants in either the first or the second generation. Five self- fertilized progenies of the third generation were grown and in about 30 plants one narrow-leafed plant was found. The inheri- tance of this abnormality is not understood. In the fourteen lines of Burwell's Flint which came from ears self -pollinated in 1914 and not planted until 1919 no abnormalities of any kind were noted in the first two generations. In the third and fourth a few chlorophyll-deficient seedlings, striped plants and tube leaves appeared. In contrast to this are the 18 lines of the same variety self-pollinated in 1918 and planted the following year which segregated the first generation for defective seeds, dwarf plants and chlorophyll-deficient seedlings in five lines. Five other lines of this lot were so poor they were discarded, while none of the 1914 lot were eliminated. Though the number of lines is too few to be conclusive it seems that the delay of five years in planting may have eliminated many abnormalities by the death of the seeds carrying them. A germination test of these ears, made in 1919, showed a viability ranging from 10 to 100 per cent. Eight of the 14 ears germinated 90 per cent, or less. None of the one year old self -pollinated ears of the same variety germinated less than 85 per cent, and only two were less than 95 per cent. There was clearly an elimination of- seeds in the five-year resting period and this could easily have been selective, the seeds carrying the recessive abnormalities being less viable. If this is proven to be the case, some method of destroying the less viable seeds such as exposure to high temperature, alternate germinating and dr}'-ing or similar harsh treatment may be an effective means of weeding out defective germplasm. Many of these recessive abnormalities after they once appeared. 396 CONNECTICUT EXPERIMENT STATION BULLETIN 266. ^ ^ J- J .-"if Figure 49. Representative plants of three flint lines; from top to bottom they are 40-4, 105-10, and 105-20. PRODUCTION OF ABNORMALITIES 397 kept reappearing in the following generations, but were finally eliminated, in every case except one, by the fifth generation. One line which was vigorous and productive and quite uniform in the fifth generation has segregated for white seedlings in CA^ery genera- tion. Selection of progenies has usually been based upon produc- tiveness and general appearances of the plants without regard to whether they were segregating for abnormalities or not. Out of the original 86 lines only 32 lines or 37 per cent, showed no clear-cut recessive abnormalities during the five generations they were self -fertilized. As stated before, 13 lines or 15 pei cent, segregated for defective seeds, and 15 lines or 17 per cent, for chlorophyll-deficient seedlings. Many of the lines had several types of abnormality. In a lot of 575 self-fertilized ears from six varieties of white fiint com in another selection experiment there were found 19 ears or a little more than 3 per cent, segregating for defective seeds. Of these, 441 were grown and 40 lines or 9 per cent, were found to be segregating for chlorophyll-deficient seedlings. Hutchison self -fertilized 2,110 ears from a large number of different varieties of corn common!}^ grown in various parts of the country and found 3 per cent, segregating for defective seeds and 36 per cent, for various seedling characters, of which the greater nxrmber were chlorophyll deficiencies. The widespread occurrence of these recessive abnormalities is fully established. In normally cross-pollinated plants they are comparatively rare in appearance since they are present as reces- sives in the heterozygous condition. To what extent, if any, they reduce growth in the heterozygous condition has not been estab- lished. Lindstrom (1920) suggests that in eliminating these recessive abnormalities many desirable factors with which they are linked may also be taken out. Since these recessives are presumably scattered throughout the chromosomes many other factors both good and poor will be taken out A^ath them. It has been argued that the recessive abnormalities tend to be eliminated by natural selection except in those cases where they happen to be closely linked with exceptionally favorable growth factors, in which case they would be preserv^ed, and in weeding them out the factors which promote growth woiild be lost with them. The only answer to such an argument is to see what the facts are. Twenty-five lines segregating for clear-cut abnormalities gave progenies in the following generation, some with and some without the recessives. The 25 progenies which still carried the recessives averaged 50.8 bushels per acre yield in comparison with 50.4 bushels for the 25 progenies grown in the adjoining rows, and from which the abnormalities had been eliminated. An equally good stand was obtained in each case, as an excess of seed was planted and the recessive abnormalities thinned out. The differ- ence in yield in the two lots is not significant. If there are favorable gro\\^h factors in the segregating progenies which are not present 398 CONNECTICUT EXPERIMENT STATION BULLETIN 266 S^nipsisppp ' % /V^t; / X „ ' "!/' Figure 50. Representative plants of three earl}^ dent lines; from top to bottom they are 110-4, 6, 10. UNIFORMITY AND CONSTANCY 399 in the non-segregating progenies from the same grand-parental plant they have no more effect than to counterbalance an}^ weaken- ing influence that the recessive abnormalities may have in the heterozygous condition. Another comparison is made by finding the average per cent, reduction in yield of all segregating lines from the first generation to the fifth generation, by which time the abnormalities were eliminated. This reduction was found to be 57.1 per cent, com- pared to the reduction of 58.1 for all lines which were free from abnormalities at the start. If any favorable groA^ith factors were lost when the recessive characters were weeded out, their departure caused no greater reduction in yield than took place in the other material from which no abnormalities were removed. From this it seems evident that the chances are no greater for good factors to be eliminated than poor ones and with other things being equal it seems highly desirable to take out these clear-cut recessive abnormalities. In fact it is necessary, in most cases, to eliminate all lethal and semi-lethal factors, in order to bring the strains to uniformity. THE APPROACH TO UNIFORMITY AND CONSTANCY. As expected, the first and second generations were quite variable but in the third generation, after three successive self- fertilizations, a number of lines became fairly uniform in height of plant, color of foliage and in general characteristics. In the fourth generation the majority of the lines had become well fixed in their type, and after five generations all of the selected lines, with a few exceptions, were alike within themselves. This uniformity was apparent in the plants of each progeny and in the similarity among the several progenies of the same line. A few lines remained variable throughout the five generations. As a rule the lines that showed uniformity in the third generation de- clined somewhat in size and yield in the two subsequent generations. Practically all of the best strains can be picked in the fifth genera- tion. Many of them can be recognized in the fourth and a few in the third. However, it is necessary to have a record of their performance during two and preferably three seasons after uni- formity is reached in order to be sure that they are fixed in their type. Several strains that were considered to be ^^ery promising in the third generation declined so in vigor and productiveness in the two following generations that they were much inferior to strains that had, earlier, been far less promising. On the other hand a few of the most vigorous and productive lines in the fourth and fifth generations were not noted as being promising in the third. While it cannot be asserted positively that strains which are uniform and good in appearance during the fourth and fifth generations will maintain themselves without further reduction the evidence from the older inbreeding experiments indicates that 400 CONNECTICUT EXPERIMENT STATION BULLETIN 266. ^-^l^ ,^ .^ '/% \ 7.,1 \-%h Figure 51. Representative plants of three late dent lines; from top to bottom they are: 112-1, 4, 9. DIFFERENCES IN SELECTED LINES 401 they can be expected to maintain their level of vigor without much loss. Therefore in carr^dng out a selection process of this kind the fourth and fifth generations are the most important in affording an opportunity to pick the best-appearing self -fertilized strains. The selection process was carried out with the aim of securing the most vigorous and productive inbred strains, uniform and fixed in their t3^pe so that their good qualities could be maintained indefinitely. For this purpose five generations of self-fertilization are necessary in most cases. Differences in the Selected Lines. In the fourth generation all of the selected lines had become strikingly differentiated. Differences in height, color of foliage, size and shape of ears made each line distinct from every other line. In the Burwell Flint lines differences in average height ranged from 51 to 98 inches, in the Gold Nugget lines from 44 to 84, in the Century Dent from 44 to 76 and in the Beardsley's Learning lines from 54 to 100. Color of foliage varied from ver\^ dark bluish green, through all gradations in shade to light green and yellowish green. In some lines the leaves were streaked with alternate rows of light and dark tissue. Various forms of fine and coarse flecking and mottHng of the leaves were a regular feature of some strains while others were entirely free from this ph}'siological irreg- ularity of the chlorophyll. The flint strains were most noticeably different in number of tillers. A number produced no large tillers and some had only a very few inconspicuous shoots from the base of the plants. Others branched very freely, producing many large branches on every plant. Alany of these were as large as the main stalk and bore ears. Some strains regularly produced seeds in the tassels on nearly all plants while others never did this. The ability to stand erect throughout the season is one feature that has been carefully selected for in all lines. Marked differ- ences in this respect were sho'WTi, being greater in some seasons than in others. Certain lines regularly went down sometime during the latter part of the season while others stood stiffly erect up to maturity. Equally pronounced differences in time of flowering are sho^^^l by the lines derived from the same variety. Most of the lines matured satisfactorily every season while others were so late as to be barely able to ripen seed. The weakening effect of inbreeding delays maturity in all lines but in spite of this some were earlier in ripening than the variety from which they were derived. Along with these diff'erences in maturity were great dissimilarities in character of the grain. The seeds of some were hard, translucent and bright colored; others were soft, dull colored and in some lines regularly moldy. 402 CONNECTICUT EXPERIMENT STATION BULLETIN 266. Figure 52. Representative ears of four productive Burwell Flint lines; from top to bottom they are: 30-19, 40-1, 7, S. DIFFERENCES IN SELECTED LINES 403 Figure 53. Representative ears of four unpro- ductive Burwell Flint lines; from top to bottom they are: 30-5, 6, 40-15, 16. 404 CONNECTICUT EXPERIMENT STATION BULLETIN 266. The features named are the more striking ones. Differences in structural details are brought out in the accompanying illustra- tions showing the plants and ears of some of the selected lines in the fourth generation (figures 49 to 57). In details of structure and arrangement of parts the lines are so distinct that they can usually be easily recognized in the field and after harvest. In a few features certain strains may be alike. Some strains have similar plants but differ decidedly in ear structure. In others the ears are somewhat similar but are borne on markedly different plants. For the most part the differences are far more obvious than the similarities. Susceptibility to Disease. The most common diseases with which com has to contend in Connecticut are smut (Ustilago) , leaf blight (Helminthosporium) , and various -root, stalk and ear roots (Diplodea, Gibberella and other forms of Fusarium). Marked differences in smut infection were shown. Two lines 105-14 and 110-17 showed no smut infection on any plant in any progeny during the five generations they were grown. Eleven strains had no more than one plant affected in any one year throughout the same period. The place on the plant where the smut balls appeared was usually quite characteristic, some strains having them on the basal nodes, others at the ear node, still others on the leaves or tassels. In some lines numerous light infections on the plant or ears were shown which apparent^ did not do any serious damage. Other strains had many plants badly injured and sometimes killed outright during mid-season. The most striking case of segregation of suscepti- bility to parasitism by the smut fungus occured in line 110-3. In the first generation four per cent, of the plants were smutted. In the second three progenies were grown having twelve plants in each. In one progeny none of the plants had any indication of smut infection. In another all of the plants were smutted and most of them were killed during the middle of the summer. In the third progeny 27 per cent, of the plants were attacked. The original seed of the two strikingly different progenies was planted again the following year with the result that out of 57 plants of the resistant progeny, only one plant o^ 1.7 per cent, was infected. The smutted lot had 14 plants infected out of 31 grown, or 45.2 per cent. In the next generation no smutted plants were seen in the one line and 65.6 per cent, in the other. Marked differences were shown in the seeds of the two lines. Plants of the susceptible line were extremely weak but the seeds were normal in appearance. However the germination of these seeds was poor and in the fifth generation no plants were obtained. The resistant line produced more vigorous plants having a noticeably darker green color. All of the seeds produced on these plants were distinctly abnormal . When dry they were shriveled and discolored although not showing SUSCEPTIBILITY TO DISEASE 405 Figure 54. Representative ears of four Century Dent lines: from top to bottom they are: 110-3, 4, 5, 10. 406 CONNECTICUT EXPERIMENT STATION BULLETIN 266. any of the usual molds. Ears of this line are shown in figure 54. In spite of their unfavorable appearance some of the seeds germ- inate and the plants produced are about as good as the average inbred strain of the same variety. None of the smut-free lines were outstandingly good in other respects and some of the most vigorous and productive strains now regularly show a high percentage of smut infection. The smut -free or low-smut strains may have value in crossing with other strains which have good qualities but are lacking in smut resistance. The growing season of 1922 was unusually wet and the selected lines then in the fourth generation showed very pronounced differences in the amount and severity of infection of Helmintho- sporitun. This organism, which is seldom injurious to ordinary cross-pollinated corn, readily attacks many inbred plants and on some completely kills the leaves after seed formation begins. Leaf blighting due to this organism had been noted each year in some lines but in the wet season of 1922 it was particularly injurious. Seventeen of the eighty-six lines showed heavy infection. Some of them lost all their foliage prematurely and the ears were badly stunted, the grains being small and poorly developed. Some of the most vigorous and productive strains in former years were so injured in this way as to give them a very low rating. The follow- ing year was unusually dry. Very little damage from this cause was seen, but the effect of the drought on different strains was very striking. Some strains which had always before produced green luxirriant foliage had their leaves killed at the sides and tips by the dry heat and were unproductive for that reason. Most of the strains which had been badly injured by leaf infection in the wet season were beautifully green throughout the dry period of 1923 and were among the best appearing and most promising of all the selected lines. These marked differences in different seasons makes it extremely difficult to judge the value of inbred strains and makes it necessary to test them during several years after they have become uniform and fixed in type. The investigations of Hoffer, Holbert and others have empha- sized the importance of the root, stalk and ear rot organisms attacking corn. The results of the earlier inbreeding experiments indicated that marked differences would be found among inbred plants to resist infection. Throughout the selection experiment great importance was placed on the ability of the plants to stand erect throughout the season and have the ears free from any in- dication of mold. Fallen plants or moldy ears were avoided when- ever possible. The most outstanding differences in ability to stand erect and in freedom from mold on the ears, were seen in the third and later generations. In 1922, a wet season, four lines (30-6, 105-20, 110-2, 110-15) had all the plants of all three progenies erect throughout the season. This same vear twelve lines (30-8, 30-9, 105-3, 105-18, 110-1, 110-2, 110-6^ 110-7, 110-18, 112-6, SUSCEPTIBILITY TO DISEASE 407 Figure 55. Representative ears of four Gold Nugget flint lines; from top to bottom they are: 105-3, 10,-17, 20. 408 CONNECTICUT EXPERIMENT STATION BULLETIN 266. Figure 56. Representative ears of four produc- tive Beardsley's Learning lines; from top to bottom they are: 112-1,4, 6, 9. SUSCEPTIBILITY TO DISEASE 409 1 = w / ^C!S< < • I*- , 1? Figure 57. Representative ears of four unpro- ductive Beardslev's Learning lines; from top to bottom: 112-3, 10, 14, 15. 410 CONNECTICUT EXPERIMENT STATION BULLETIN 266. 112-7, 112-8) produced no moldy ears. Only one line 110-2 had all plants erect with ears free from mold. In the first generation this line had four per cent, of moldy ears and ten per cent, of fallen plants but no smut. In the second generation there were ten per cent, moldy ears, ten per cent, fallen plants and no plants showing smut infection. In the third, fourth and fifth generations there was no mold, smut or fallen plants on the three progenies grown each year. This strain is also productive for the variety, although surpassed in this respect by several other strains. The seeds are hard and bright but very pale yellow in color and almost white on top. In contrast to this is line 105-20 with all the plants erect in the second, third and foirrth generations but with 29, 17 and 44 per cent, of moldy ears in the same years. On the other hand, 40-8 had all of the plants fallen in two progenies of the fourth generation and no moldy ears. To complete the combinations 112-11 had 87 per cent, of the plants on the ground in 1922 and 67 per cent, of ears moldy. Criterions of Selection. At the beginning of the selection experiment the plan as previous- ly stated was to self -pollinate five plants in each line and to select three of the best self -fertilized ears for planting the following year. Even when these ears differed greatly in appearance no consistent differences were noted in the progenies grown from them. The coefficient of association between the appearance of the ear and yield of the different progenies within several lines is- — .18. This indicates that self -pollinating a large number of ears in order to make more extensive selection of desirable looking ears is of doubt- ful value. Of the three progenies grown only one was to be chosen to continue the line, the other two not being pollinated. It was soon noted, however, that there was very little relation between the appearance of the progenies at the time of bagging and their pro- duction of grain and the general appearance of their plants at harvest. The coefficient of association between the appearance of the plants at pollinating time and the yield of the different progenies within the several lines is — -.28. Seedlings were groum in the greenhouse and their weight and height after thirty days of growth were compared with the yield of the same progenies in the field. The third and fourth generations showed that those prog- enies that had the tallest seedlings yielded 1.6 bushels per acre more than the other progenies in the same lines. This difference is hardly enough to make a selection of the progenies on this basis worth while. Since there is no appreciable correlation between the characters of the seed- ear, weight of seed, size of the seedling, or the appear- ance of the plants at pollinating time and production of grain the only selection of progenies that can be made with any degree of CLASSIFICATION OF SELECTED LINES 411 effectiveness is at maturity. Here also yield is highly influenced by the amount of heterozygosity remaining. In some lines there are more homozygous combinations than in others and they are correspondingly less vigorous and productive although they may be potentially more desirable. For this reason final judgment must be left until the plants are reduced to uniformity and constancy. Hence it is interesting to note what resemblance the resulting inbred strains, when finally reduced to imiformity and fixity of type, have to the same strains in the first generations of inbreeding. Classification of Selected Lines. Taking into consideration all features of these selected lines as they grow in the field and after han^est in the fourth and fifth generations and giving most importance to the production of bright sound grain, the four outstanding good and poor strains in each variety are listed as follows, with their yields in bushels per acre in the fifth generation compared with that of the original variety grown the same ^^ear: Bur-well's Flint 51.2 Good Lines Poor Lines Number Yield Number Yield 30-5 12.2 30-10 44.2 30-19 15.3 30-18 18.3 40-4 33.6 40-3 35.1 40-8 25.9 40-16 24.4 Gold Nugget 54.0 Good Lines Poor Lines Number Yield Number Yield 105-11 29.0 105-3 9.2 105-15 33.6 105-8 13.7 105-17 22.9 105-13 7.6 105-20 10.7 105-16 29.0 Century Dent 48.3 Good Lines Poor Lines Number Yield Number Yield 110-2 15.3 110-1 28.9 110-4 16.8 110-9 4.6 110-5 10.7 110-15 1.5 110-10 19.8 110-17 12.2 Beardsley' s Leaming 49.5 Good Lines Poor Lines Number Yield Number Yield 112-1 42.7 112-3 10.7 112-6 27.5 112-7 21.4 112-9 33.6 112-14 .0 112-12 12 2 . 112-16 9.2 412 CONNECTICUT EXPERIMENT STATION BULLETIN 266. This is purely an arbitrary classification based upon the general appearance of the plants and ears. Some of the poor lines yielded more than the good lines but produced a very poor quality of grain. The original ears from which these lines descended (figures 35 to 38) show that there is no relation between the good and poor strains after uniformity was attained and the appearance of the seed ears from which they came. Low and high numbers are represented about equally in the good and poor strains. Correlation Betwee^t the First and Last Generations. In order to find out whether the elimination of the poor lines at the beginning of the inbreeding period is advisable, the correlation Table VIII, CoeflEicients of association between early and later generations of self- fertilized corn. Generations Compared 1-4 1-4 1-5 2-5 1-4 Variety Height Mold Tillers Smut Yield Burwell's Flint 60 .89 .64 —.08 Gold Nugget 35 .38 .38 .14 Century Dent 80 .80 —.72 .72 .28 Beardsley's Learning 95 .38 .50 .20 .50 Ave. Flints 50 .65 .55 .10 .05 Ave. Dents 89 .63 .17 .52 .38 Average 71 .64 .27 .27 .19 between the behavior of the plants in the first generation and the last generation has been worked out for the most important characters. In Table VIII are shown the coefficients of association ^ 1 — I ■ I I i r FIRST FOURTH HEIGHT FIRST FOURTH MOLD FIRST FIFTH TILLERS SFCCND FIFTH SMUT FIRST FOURTH YIETLD Figure 58. Diagram representing the average of the upper and lower groups in the first generations and the average of the same lines in the last generations, based on the data in Table IX. between the first or second inbred generation and the fourth or fifth for height of plant, per cent, moldy ears, number of tillers, per cent smutted plants, and yield of grain. The fifth generation, grown in 1923, was so variable on account of the extremely dry season afTecting different parts of the field unevenly that the coefficients for height and yield are based on the first and fourth generations. There was very little smut infection in the first CORRELATION BETWEEN GENERATIONS 413 generation and practically no mold in the fifth so that the coefficient for per cent, smut is based upon the second and fifth generations and for per cent, mold upon the first and fourth. The figures show a fairly high association for height of plant and moldy ears. This means that by selecting the highest lines in the first generation the resulting inbred strains in the fourth generation would tend to be taller than the average. Similarly, by selecting lines at the start that were free from mold, inbred strains could GENERATIONS Figure 59. Graph showing the behavior of two lines with respect to height during four generations of self-fertihzation, selected for vigor and productiveness but not for height. From one to three progenies are grown in each generation. finally be attained that would on the average be freer from moldy ears than other strains which showed more mold at the start. This relation does not hold so well for the other characters ; number of tillers and per cent. smut. For these the coefficients are low and in two of the varieties a negative correlation is shown. This means that lines without tillers and showing low smut infection may be obtained from plants at the start which have tillers and are susceptible to smut infection. 414 CONNECTICUT EXPERIMENT STATION BULLETIN 266. Another method of bringing out the relation between the several lines at the start and at the end of the selection period is to separate all the lines of each variety into the upper and lower halves, with respect to the characters studied, in the first generation and then compare the average of these two groups with the averages of the same lines after being inbred for four or five generations. This has been done in Table IX, making the separation within each variety into equal sized groups in the first generation. Thus the basis for separating the groups is the median instead of the mean. The results are stmimed up graphically in Figure 58. It will be seen that the relative position of the upper and lower halves remains nearly the same at the end of the period of selection as at the Table IX. The relative position of the same self-fertiUzed Unes at the and at the end of the period of selection. ; beginning Characters measured Groups No. of Strains First Last Average First Last Relative First Last Height of plant in inches in the first and fourth generations High Low 36 36 94 96 81 70 70 61 100 87 100 87 Per cent, of moldy ears in the first and fourth generations High Low 36 35 88 95 IS 5 15 8 100 29 100 57 Number of tillers per plant in the first and fifth generations High Low 32 33 91 94 .9 .3 1.2 .8 100 36 100 67 Per cent, of plants with smut in the second and fifth generations High Low 32 33 90 93 14 1 12 7 100 9 100 59 Yield of grain in bu. per acre in the first and fourth generations High Low 31 31 84 82 81 52 44 41 100 64 100 93 beginning for such characters as height of plant, number of tillers, per cent, of moldy ears and smutted plants although the difi^erences are generally less at the close than at the start. This tendency to change during the period of inbreeding is most marked for 3deld of grain. In this respect the high and low groups are ver}^ nearly ahke at the end of the selection period in spite of the fact that all along attention has been given to productiveness. These results indicate that it is unwise to eliminate the unproductive strains in the first generations, as from them lines may be obtained that are as productive as those from high yielders at the start. Other char- acters can apparently be somewhat more surely selected for at the beginning of the inbreeding period. If such characters as freedom from mold and smut are of chief importance it might be advisable to eliminate those lines which show much mold and smut in the first inbred y:enerations. LIMITING FACTORS 415 The general tendency for some of the lines to hold the same relative position throughout the process of selection is illustrated by the height of plant of two lines shown graphically in figure 59. In the first inbred generation the two lines averaged 69 and 77 inches in height. In the second generation two progenies over- lapped but from then on they were clearly distinct, the difference in height increasing until the end of the selection period. The same result is shown in the average number of tillers per plant of two other lines as brought out in figure 60. Differing at the start the two lines remained distinctly different in all their progenies throughout the period of inbreeding. In marked contrast to this is the result shown graphically in figure 61. Two fines differing noticeably in their nimiber of tillers changed positions so that in 3 GENERATIONS Figure 60. Graph showing the behavior of two lines with respect to tillering during five generations of self-fertilization, selected for vigor and productiveness but not for the number of tillers. The relative position of these two lines remained the same. the end the few tiller strain at the start averaged more tillers on all progenies than did the many tiller strain. Similarly two strains which were alike in this respect at the start became extremely different as uniformity and constancy was reached, as shown in figure 62. Limiting Factors. In planning and carrying out a selection program the best procedure will depend upon the number of plants which can be grown and the number of hand pollinations which can be made in a season. Where the facilities available for artificial pollination is the limiting factor, and this is usuall}' the case, the best procedure 416 CONNECTICUT EXPERIMENT STATION BULLETIN 266. is to self-pollinate just enough plants to continue as many lines as possible until a reasonable degree of homozygosity is reached. If the amount of land available to grow the plants is the limiting factor it would be better to pollinate a larger number of plants within each line, although extensive selection within a progeny has been shown to have little value, as the better individuals are almost certain to be more heterozygous, making it difhcult to arrive at their true value. More attention should be paid to increasing the number of progenies within the more desirable appearing Hnes, basing selection on their behavior throughout the season and their uniformity and productiveness at maturity. The method now being used at this station is to grow three progenies in each line and to pollinate two plants in each progeny. On the basis of the general appearance of the plants in the field and CrNCRATIOINS Figure 61. Graph showing two lines which differed in the amount of tillering and which changed positions during the five generations of self- fertilization. their productiveness at maturity the best and second best progenies are noted where there is an appreciable difference. Two ears from the best progeny and one from the second best are used for planting the following year. If no differences are shown, one ear from each of the three is planted. This procedure is based upon the results in the five-year selection experiment described above in which no reliable criterions of selection were found which could be used before the time of pollination. It is still provisional and will be modified as future experience justifies. It is possible that better results can be obtained by paying still less attention to selection during the reduction period than the method outlined. By expend- ing the same amount of time and effort on more lines, growing only one progeny in each generation and pollinating only enough plants to insure the perpetuation of the strain until uniformity and constancy are reached, more diverse material would be available CONCLUSION 417 from which to select the best inbred strains. In this procedure there would be the possibility, and even probability, of missing altogether valuable material which might exist in some lines. However, since it has been shown that many of the lines change greatly during the reduction process, selection during this period will always be somewhat ineffective. From a theoretical stand- point the best method is the one which will produce the largest number of fixed strains from which to choose the ones best suited to the purpose for which they are to be used. In this connection one further point should be mentioned. When- ever any particularly outstandingly good strain has been obtained there is the possibility that still better material may exist in that strain in the earlier generations. This would indicate that it GENERATIONS Figure 62. Graph showing two lines which showed the same amount of tillering at the start but differed widely at the end. might be well worth while to go back to the earlier generations and grow as much of this material as possible from the remaining seed in order to obtain the very best gemiplasm available in this strain. In fact, this procedure has already been followed with several of the more promising lines and it has been possible to isolate new strains which are distinctly superior in some respects to the old ones. Conclusion. The one fact that stands out from the results secured in this selection experiment is that there is no single criterion by which high-yielding strains can be obtained. During the process of inbreeding, with the resulting segregation and recombination and the automatic elimination of heterozygous combinations of factors, selection for particular characters is somewhat effective. By 418 CONNECTICUT EXPERIMENT STATION BULLETIN 266. choosing tall plants as progenitors in each generation tall strains can be produced. By selecting plants free from tillers, strains with , few tillers can be obtained. Similarly, freedom from disease in- fection, as far as resistance is inherited, can be expected by selecting during the reduction period only those plants which show no infection in fields where infection is present. Even with these characters the association is far from complete. But productive- ness, yield of grain, which sums up the plant's entire energies shows no such simple relation. High yielding strains may come, and have come, from plants which are poor producers. Promising strains during the first generations may be very unproductive or undesirable in some respect when finally reduced to uniformity and constancy. This emphasizes the fact that effective selection must be based upon the performance of the plants after homozygosity is attained. LITERATURE CITED. Cummings, M. B., and Stone, W. C, 1921. Yield and quality in Hubbard squash. Vermont A. E. S. Bull. 222. East, E. M., 1908. Inbreeding in corn. Report Conn. A. E. S. 1907- 1908. 1909. The distinction between development and heredity in inbreeding. Amer. Nat. 43: 173-181. East, E. M., and Hayes, H. K., 1912. Heterozygosis in evolution and in plant breeding. U. S. Dept. Agr., Bur. P. I. Bull. 243. Eyster, W. H. 1924. A primitive sporophyte in maize. Amer. Jour. Bot. 11:7-14. Hayes, H. K.,and Brewbaker, H. E. 1924. Frequency of mutations for chlorophyll-deficient seedlings in maize. Jour. Her., 15: 497-502. Hoffer, G. N. and Holbert, J. R. 1918. Selection of disease-free seed corn. Ind. A. E. S. Bull. 224. Hutchison, C. B., 1922. Heritable variations in maize. Jour. Agron., 14: 73-78. Jenkins, M. T., 1923. A new method of self -pollinating corn. Jour. Her., 14: 41-44. Jones, D. F. 1918. The effects of inbreeding and crossbreeding upon development. Conn. A. E. S. Bull. 207. 1920. Heritable characters of maize, IV. A lethal factor- defective seeds. Jour. Her. 11: 160-167. 1921. The indeterminate growth factor in tobacco and its effect upon development. Genetics, 6: 433-444. 1924. The attainment of homozygosity in inbred strains of maize. Genetics, 9: 405-418. Kiesselbach, T. A., 1922. Corn investigations. Neb. A. E. S. Res. Bull. 20. King, H. D., 1918. Studies on inbreeding, I-III. Jour. Exper. Zool. 26. Lindstrom, E. W., 1920. Chlorophyll factors of maize. Jour. Her. 11:269-277. 1923. Heritable characters of maize XIII. Endo- sperm defects: sweet defective and flint defective. Jour. Her., 14: 127-135. Mangelsdorf, P. C, 1923. The inheritance of defective seeds in maize- Jour. Her., 14: 119-125. v„; >^ -J •-.• 'u ^ University of Connecticut Libraries 39153028954701