l Q- U E . . ”3‘,” {‘1 L. UN QM M 1 .1: IQ r -qf. "IF. . aux 12!“..‘01 ; r -. , fi. - ‘ IHESIS I l. lsi. .l.‘ -U".1A1l'lv ‘ ON THE IM'L‘ERITANCE OF FLOYCER COLOR IN ALFALFA ON THE INHERITANCE OF FLOWER COLOR IN ALFALFA THESIS Respectfully submitted in partial fulfillment for a Master of Science degree at the Michigan Agricultural College Brittain B. Robinson ”'3 1924. £3 t1¢a¢ - PLATE I - THE 1922 ALFALFA NURSERY OF THE MICHIGAN EXPERIMENT STATION 1034452; - ACKN OW LED GEICEN T - The writer wishes to express his appreciation to Mr. H. M. Brown for his helpful suggestions and criticism throughout the work. To Mr. E. E. Down credit is due for suggestions in the work. The writer expresses his gratitude to Professor F. A. Spragg for explanations of his experience and crit- icisms of the work and thesis. Thanks are extended to Professor J. F. Cox, Professor F. A. Spragg, and Dr. E. A. Bessey for a final review of this thesis. II III IV VI VII VIII IX XI XII XIII - TABLE OF CONTENTS - P INTRODUCTION ------------------- HISTORICAL l - Origin of Names- - - ------------- 2 - Origin of the Plant- - - -~ ------ - - - - 3 - Classification of Species ----------- CHANGE OF COLOR ------------------ SELECTIOE OF COLOR TO MARK VLRIETIES ------- PREVIOUS WORK ON COLOR INHERITANCE - ------- MATERIAL OBTAINED FOR THE EXPERIMENT ------- FERTILIZATION AND SELF-STERILITY IN MICHIGAN - - - ALLOGAMY vs AUTOGAMY - - ............. l - Evidence for Self-fertilization in Michigan- - 2 - Evidence for Cross-fertilization in Michigan - 3 - Discussion of Fertilization ---------- IKHERITANCE OF ARTHOCYANIN ------------ CONCLUSIONS .................... LITERATURE CITED ----------------- EXPLANATION OF TABLES --------------- TABLES l - 22 16 18 23 24 26 29 34 38 49 1| If I --INTRODUCTION There are several different types of flower color in alfalfa. The color depends upon the inheritance within the plant. The known basic flower colors are purple, white, and yellow, but it is possible to have nearly any shade or combig nation of these colors. Flower color aside from its attractiveness in many plants fills an important position in distinguishing many varieties or strains. This is an important characteristic, that seems well worth ones time to study. The ability to tell varieties apart helps to maintain their purity. Since many states do not have stringent seed laws, adulteration and misrepresentation are practiced. In this way many excellent varieties are lost, because it is impossible to maintain their purity once they are cOmmercialized. Color of flower has been used to distinguish varie- ties of alfalfa. Because the color of the flower does not seem to remain constant this means of distinction is poor. The purpose of this thesis is to determine the Mendelian factors that influence color inheritance, that we may be able _ to obtain strains pure for one color. It is desirable to know what factors influence pollination in Michigan, and to determine whether a color can be used to aid in the keeping of a commercial variety pure. -2- II -- HISTORICAL l - Origin of Names. The evidence indicates that alfalfa is one of the oldest of the cultivated plants. This is confirmed, says Hendry (1923), by two "Arabic words, ratba for green alfalfa, and quatt for alfalfa hay." The word alfalfa is Arabacized Persian, and is derived from.the Iranian word aspo-asti, "to eat," and literally means "horse fodder" (Neldeke 1900). "This word penetrated the Syriac in the form of aspesta or pespesta (the latter in the Geoponic)"(Laufer 1919). That alfalfa was introduced into Arabia at a very early date, seems evident from the characteristic varieties which doubt- less have required centuries to become acclimated to the arid regions of that country. The first reference to alfalfa in literature reported was what the Assyriologists claim is in a Babylonian text of ca. 700 B.C., in which alfalfa is mentioned under its Iranian name aspasti or aspustu (C. Joret). After the defeat of Xerxes in 479 B.C., the Greeks discovered patches of alfalfa growing where the Persian army had been. The Persians had used it as provender for their horses and other animals. As alfalfa was introduced by the Persians from Mespotamia, it was called Medic. Alfalfa was acquired by the Romans about 200 B.C., and reached Spain and Switzerland about 100 AJD. The Spanish -5- and the Portuguese introduced it into South America and Mexico under the name of alfalfa. The plant reached our New England States during the Colonial period, and was called lucern. Clothier says it was introduced into New York from Europe as early as 1820. The earliest importation from the Spanish origin was into California from South America. The California Farmer (1855) mentioned alfalfa growing in "ex- perimental planting" and in Flawns." Doubtless it reached the Southern States earlier as Dr. Phares says it had been grown in the South 50 years before being introduced from California. 2 - Origin of The Plant. The origin of alfalfa is still a disputed question. Trabut (1917) believes, "all cultivated lucernes are of hybrid origin." Medicagq sativa Linn. (purple flowers) does not exist in the wild state. He thinks the g; sativa originated from N. getula Urban and E, tunetana Murbeck. The flowers of these two species vary in color. They may be yellow, blue, pink, or white. He further states that g, falcata (yellow flowers) was not produced by hybridization, as it is found too far north of the species that produced g. sativa. Many writers believe 3;. §_a__t_i_.v_a_ originated from p. falcata. The yellow blossom.fl, falcata was, in other words, -4- the progenitor of the purple blossom M, sativa. Alefeld (1866) received many specimens from a collector in Asia, and he grew these in Germany. These plants showed all the stages from the oldest type of alfalfa to what we have to- day. Under selection and cultivation two main variations have appeared, namely: lst. The plants have become taller and more upright. End. The pods have become longer, and the larger pods more twisted. It was from these differences Alefeld concluded that N, sativa originated from.g, falcata. He believed that he had all the gradations between the two Species. He did not mention the change in the color of the blossom which must have taken place, if this is the true evolution. The above gradations that Alefeld noticed may also be observed in many hybrids between the two species. It would seem that a more detailed explanation is necessary to prove such an origin. It is probably best not to draw conclusions as to whether M. sativa came from g. falcata or vice-versa. It is also possible g. sativa is of hybrid origin but at pres- ent there is lack of sufficient evidence to prove either case. -5- 3 - Classification of Species. The early references to Species are purely botanical. Probably the first mention of M, falcata in literature is by Gesner (1561). It is a brief description under the name of Trifolii genus medica similis. It is described as having yellow blossoms and sickle shape pods. "This reference may, with a resonable degree of certainty, be regarded as the first positive mention of Medicago falcata in literature, since the identity of Book's (1552) plant is somewhat in doubt" (Oakley & Garver 1917). The best early description of N, falcata is given by Clusius in 1585. He gives it the name Medics lutgg_§lg£g. Rivinus (1690) was the first to use the name Medicago falcata and he used it in a generic sense. He published an excellent illustration, which, according to Oakley and Garver (1917), "is the first unmistakable figure of a H, sativa X g, falcata hybrid." Vaillant (1727) put H, falcata in the genus Medicago of Tournefort; however, it remained for Linnaeus to enlarge and define the genus and to assign the name N, falcata to the species. He did this in 1753 in his Species Plantarum. Columella (1745) in addition to giving excellent advise on the culture of alfalfa, describes several Species and the flower colors. The plant, he says, "varies in its flowers, which are blue, violet, purple-blue and blackish, etc." Martyn (1792) gives the first botanical description of, what appears to be, the hybrid. The interesting point in this description is the color change which he has describ- ed for the blossoms. It is as follows: "Medicagg varia (species or rather variety). - Various flowered medick. Yellow medick varies much in the color of its flowers, which are sometimes whitish, quite white, or greenish. The variety here figured is remarkable in having flowers of colors so different as blue and yellow on the same stalk. Casper Bauhin says, it is found in the south of France with whitish yellow, green, blue, purple, black, and variegated flowers, but he does not affirm that these different colors are on the same plant" (Westgate 1910). Persoon (1807) gives the next botanical description of the hybrid as N. media. He says the flowers are pale blue but later may become yellowish. Corbiere (1895) makes somewhat of a broad botanical statement by saying that Medicagg media Pers. is not a hybrid though he does not explain why. III -- CHANGE OF COLOR An important consideration in the study of any char- acter is whether or not it remains constant. We are now to study some characters that change with the maturity of the -7- blossom. This is only true of the hybrids, as the true E, sativa and E, falcata retain their original colors. Westgate (1910) says the flowers in addition to being yel- low or purple may change from violet through blue and green and finally become yellow. Waldron (1911) also says the flowers of the hybrid show the above progressive change even though they contain but a seemingly small percentage of E, falcata parentage, but not as pronouncedly as the first hybrids. Hagem (1919) agrees with westgate and Waldron on the change of color, and denies that it goes in the reverse order. Alefeld (1866) has noticed just the opposite change ,- of color in two sub-species he considers to be under 3, sativa. They are Medicagp sativa versicolor Koch, and H, sativa Kochiana. He says the first is similar to E, sativa but the blossoms sometimes are yellow, later green, and finally turn violet. The latter, Medicago sativa Kochiana, is similar to E, falcata, but the flowers are first yellow, then green, and finally violet. Garcke (1903) calls the cross E, falcata X E, sativa sandlucern. He says the flowers are first yellow, then grass-green, and finally bluish vio- let. Bentham (1865), Bonnier, Corbfere (1893), and Hy (1895) give the range of color from yellOw through green and finally purple. Bonnier also shows a number of colored illustratibns, but his E, media which he said had this range, is colored -8- only one shade throughout, which is yellow. Willis (1910) describes a cross of E, falcata X H, sgtiya as having both blue flowers and yellow flowers on the same plant, and occasionally on the same branch. The recent writers, who are men engaged in practical 'experimental agriculture, have described the range starting with purple and ending with yellow. It seems a little more likely that this vieWpoint is correct, as probably the old school against this idea were men who were concerned chief- ly with the collection of a large number of descriptions for a book on the flora of some country. IV -- SELECTION OF COLOR TO MARK VARIETIES The writer is not the first who has attempted to obtain varieties that had characteristic markings of the blossom. Dr. Giovanni Patane in Italy (1916) writes, "Prof. Strampelli endeavours not only to give the new kinds an indisputable cropping value, but also, when possible, to impart peculiar characters which may distinguish his crea- tions. Thus, for instance, .......he has obtained: ...... lucern with white flowers, with iron gray flowers, greenish flowers, etc." It is not likely that this man has obtained a variety that bred true to greenish flowers. All literature available on alfalfa mentions green blossoms as a stage in the range from purple to yellow, or F1 blossoms that are - PLATE II - ALFALFA GERMINATION PLOTS Each plot was planted with the seed from an indivi- dual plant. A short time after the seed had germinated the seedlings were transplanted as shown in Plate III. -9- nearly green or greenish yellow. Hansen (1917 and 1919) has for several years been selecting a white flowered strain, that was also as hardy and as productive as other varieties. He has obtained a strain known as Hansen's White Flowered Alfalfa, which is 97% pure for white blossoms. Undoubtly other men have tried to obtain a marked variety. As this work was primarily an inheritance study these investigators will be discussed under that heading. Some investigators have probably never published their results. V -- PREVIOUS WORK ON COLOR INHnRITANCE Several experimenters have studied the inheritance of blossom color in alfalfa, but their results are not con- clusive. They agree on the blossom color obtained in an F1 hybrid from.§, falcata X g, sativa, but at present there is no Mendelian explanation for an F2. In all the literature on alfalfa there are only a few tables that show the distri- bution of colors in a variety or in a segregation from a cross. Westgate (1910) presents a table (page 10) in which, "results of counts of the various colors found on the flowers of plants of different varieties or strains of variegated and other alfalfas," are shown. -10- "Aesults of counts of the various colors found on the flowers of plants of different varieties or strains of variegated and other alfalfas." +3: Color of 3 4? Flower ‘ 0 ice" ‘ i f :29.“ Ordinary violet and lavender O Violet,1avender* with dark keels 0 Bark purple 0 Very light lavender O ‘Rluish(usually rusty when old) 0 white tinged with blue or lavender 0 Rhite (Ivory) 0 Wins color 0 Reddish lavender O Blue—black O Blue-green passing into yellow-sreenO Blue or viOlet, with yellow keelsO Old flowers cream or yellow 0 Yellow or creamlOO Total number of plants counted 100 87 583 737 764 *(rarely blue) After 3 0 2| § :1 It -' ‘2 _ E: 'c U?) , H as as 'r U (3 CH If‘ $1.1:3 c :3 r: g '9: c. $4 63 9—} E ,Lpu CH Q; r‘i . , :—-i 0 £2 C3 0 'U 1'” :3 :r (:3 .. (S 3-5:. E ,_, a ' .i'. :3 0r' E. l E‘ '0 C +3 H £%+:vi c 8 .4 n L :3 3-4 :5 (5 CH :44 C U> C' a a ho. ISO. No. No. ho. No. 6 194 307 314 20 17 3 47 84 109 10 0 O 52 60 62 4 8 O 45 52 28 7 O 14 179 137 213 27 9 O 4 1 2 2 l 7 1 l 1 0 O 1 6 16 1 2 O 6 1 11 3 3 9 O 5 3 9 O 2 6 10 10 4 9 1 4 13 20 1 3 O 38 25 31 15 3 2 2 1 4 2 O O Yestgate 1910. Kharkof(Russian)alfslfa No. 43 a o o 0 \ - n1mb1rsk(Russ:anenlfalfs No 0 theeler(KansaS)3 H \O 228 O )4. .rian(hordy fern .Imn o 1‘30 0 9O 49 150 92 321 22 ‘1". O O . fi (J (a C C 9 +3 :3 $4 {'3 r-‘I m +3 C) c... .‘J 0 U) +7 H Ca 2 ‘ e U) (1 F4 ‘4 ’11: , C.) (r: d :5 L i «'1 E4 a L C) >3 r’: $4 0 ‘j 0 s3 0 C: 5-} U2 £1 C 63 C O "—4 0 0H or. :1 '0‘ :2 H 'C': 0r“: 0H O (J "'1 £1 a k -P b C: 0 Ce. H L. 1900 IMO. 1.10. 1300 1‘20 27 223 108 19 814 6 6 3 O 29 3 9 6 2 23 2 4 O O 12 8 4 18 4 3O 4 O 1 O O 2 O O O O 59 249 138 25 908 -11- This table adds but little value to the genetic analysis of color, because the plants in the table do not represent the offspring of single individuals that have been protected from foreign pollen. They are plants obtain- ed from commercial seed which had chances for adulteration and mechanical mixing, as well as crossing. The table, how- ever, does show that E, falcata will breed true for yellow. It shows that the ordinary western grown alfalfa varies only in the degree of purple, which is the E, sativa color. Only true breeding yellows will be considered E, falcata, and true breeding purples as H. sativa in this thesis. The botanical classifications are often not specific in their use of the two terms. Alefeld (1866) believes his sub-species H, sativa intermedia (variegated flowers) is the same as M, intermedia Schultes. However the Index Kewensis (1844) considers Schultes' Medicago intermedia as equal to E, falcata (yellow flowers). Thus we have a variegated flowered species which is supposed to be equal to true E, falcata. Brand (1911) speaks of the color of blossom of the cross E, falcata X;§. sativa as being a characteristic black- ish or smoky-colored flower. He mentions that this is not the common color of the Grimm alfalfa. Waldron (1911) published a table of foreign varieties to indicate the per cent of variegation in different foreign -12- strains. This table of 64 strains, with an average of 37 plants per strain, showed a range of variegation from 72.7% variegation to zero per cent. The average variegation was 24.1%. This seems to help confirm testgate's results that nearly all commercial varieties are variegated, to some extent, in blossom color. Oliver (1913), trying to obtain new varieties for pasture purposes, made a number of crosses which he has described in some detail. Two bear more directly on the subject of blossom color and are cited. "Cross No. 191 (g. falcata S.P.I. #20721 9 x g, satiyg S.P.I. #17698 6)" F1 "The 16 plants looked as if they might have been typical E, sativa; in fact, it was thought that a mistake had been made. However, when the first flowers Opened, this supposition was dispelled. The flower colors of the plants were invariably alike and they resembled the color of the foliage to such an extent that the plants has to be examined closely to find all the flowers." F2 "The colors of the flowers varied exactly as was expected ranrin* from the color of Medicago falcata . , _ _, to that of g. sativa parent." "Cross No. 293 (Grimm F.C.I. #131 O x r. falcata S.P.I. #24455 8)" Results: 293-1 greenish purple 293-3 greenish purple 293-5 greenish purple. ,-13- Southworth (1914) made interesting crosses in Ontario of Li. lupulina X lit. sativa. His Fl results were all varieg- ated, but the shade varied on nearly every plant. The main color was violet with variations into dark purple and green. Yellow did not appear except on the keel, and this was not general. Later Southworth (1918) crossed some F1 plants with sweet clover (probably melilotus §;p§). Piper (1915) gives the origin of sweet clover as the same general region where alfalfa came from. As these two genera will produce a fertile cross it might suggest a means of improvement that may help greatly in producing a desirable forage crOp. Hagem (1919), a Swedish investigator, found that the F1 from.a E, falcata XJM. sativa cross had blossoms var- iegated in color. In the F2, he obtained individuals with flowers which were permanently purple or yellow, and several with categories in between. The article reviewed suggested that he made up several ratios for his segregation, but they were extremely hypothetical, and it may be for this reason they were not included. He encountered self-sterility in his segregations, and was prevented from continuing his work. Witte (1921), also a Swedish investigator, has presented the most recent data upon color inheritance. He gives the following table showing the number of individuals, according to color, that segregated in some of his F2 progenies. Type Type Type Type Type Type Type Type Type Type Type Type Type 10 ll 12 13 -14.. Yellow ("Klincksieck nzr 186") more or less pro- nounced greenish nerve rings, and even a greenish dirty touch; buds somewhat bluish. This type, which includes several sub-types, corresponds more with the blossom color of the F1 ------------- 147 Yellow ("Klincksieck nzr 186-191") This is a more clearfull color than the former with slight nerve coloring ------------------- 19 Yellow with a tendency toward orange but not a bright orange ("Klincksieck nzr 181") as the yel- low lucern ---------------------- 68 Pale yellow ("Klincksieck nzr 196") -------- 10 The younger stage has a touch of violet. It is quite close to the former, and even the follow- ing but more yellow than the latter --------- l9 Pale yellow with a violet touch, and quite marked violet nerved ----------------- 4 Pale dirty yellow with a greenish tinge (The blossoms are violet when half mature) -------- 3 Light dirty violet ---------------- l Pale brownish yellow --------------- 3 Strong yellowish brown -------------- 2 Pale dirty brownish yellow ------------ 2 Light greenish violet- -------------- 5 Pale violet (blue lucern type) ---------- 1 -15- Type 14 Almost white, weakly cream color with pale violet buds ----- - -------------- 1 Type 15 White with no pale violet buds --------- ‘__l_ Sum 285 Yitte obtained 200 and 500 progenies from some of his Fl hybrids. His results based on these progenies are:- Blue X Yellow gives intermediate F1. Yellow is not dominant in the F . l A natural complicated segregation in the F2. Galloway (1907), Konstantinov (1914), Hahn (1915 a 1916), and Moore (1917) have made numerous crosses but their descriptions are inadequate to throw any light on the sub- ject, as they were not working for color inheritance and hence failed to describe the blossom colors. Color inheritance has been studied by crossing E. sativa and E. falcata. A few crosses of other Medicago species give an indication of similarity to E, sativa X M, falcata crosses. Hy (1895) mentions a cross involving other species, but the descriptions are vague and botanical and are of no importance. Purple and yellow are the common basic colors that flowers have been mentioned as appearing in the F2 segregates. Hansen (1919) showed that white will breed true, though we are not sure what degree of whiteness this is. White blos- spms are usually associated with purple buds or purple veins, but this is not always the case. If Hansen's white is homo- zygous, we have three basic colors to work with, namely: purple, white, and yellow. Summarizing the previous investigation, we have the following results: 1st. Recipical crosses of E, sativa and E, falcata give a variegated flower color in the F1. This color was described as green, or greenish yellow. 2nd. The segregation in the 32 appears complicated. The following colors segregate from the F1: purple, yellow, white, and variegations that may include all of these, with shades of green and various intensities of all. It will be necessary to try to discover the F2 Mendelian ratio from the experiment performed in connection with this thesis. VI ~- MATERIAL OBTAINED FOR THE EXPERIMENT The writer was very fortunate, when he began the study of color inheritance in September 1922, in having a chance to take notes on some 11,000 plants growing in an alfalfa nursery at the Michigan Experiment Station. The parent blossom color for every plant in this nursery was known. However, there were no artifically crossed or self- ed plants, and it was too late in the season to make any. -17- The uniformity of the morphological and physiological char- acters, in progenies grown side by side in this nursery, was striking. The mothers of these progenies had been grown in hills surrounded by other plants in a similar nursery. This fact indicates that a great deal of self-fertilization is going on in Michigan, as the strains were known to be very heterozygous when they came to Michigan, only a few genera- tions before. In 1922, 685 individual selections were made. These were threshed separately, but many of them failed to produce enough seed to continue. Ninety-three of these 685 selections were planted in small germination plots in the spring of 1923. When the seedlings were about six inches high they were trans- planted. The progenies were planted in separate rows two feet apart. The plants, in rows, were twelve inches from.cne another. There were a few progenies which required two rows, as there were only 99 individuals in a row. Approximately 10,100 seedlings were transplanted, but only 6,448 produced blossoms. A great many of the seedlings died, because they were improperly planted and had insufficient moisture after transplanting. The first plant to bloom was noticed on July fourth and the plants continued to bloom until October. For this reason note-taking extended over a considerable period. Possibly a few plants blossomed, and the blossoms fell off - PLATE III - THE 1923 ALFALFA NURSERY Showing the method in which the seedling of indivi- dual progenies were set out in separate rows. ~18- without being noticed. While notes were taken on all of them at least once, a number were observed two or more times to study the effect of maturity. In no case observed by the writer among 1,299 in- dividuals have plants changed their flower color from yel- low to purple, but always the reverse. The notes taken at different stages of maturity indicate that yellow is late in showing itself in many cases. In crosses involving yellow X purple, the yellow has shown itself after the purple and green have appeared. It was observed that where the flowers appeared to be a cream color, the blossoms were first white and later the color changed to yellow. It would seem that there is some physiological or Mendelian inhibitor that prevents yellow from expressing itself until the blossom had reached a certain stage of maturity. VII -- FERTILIZATION AND SELF-STERILITY IN MICHIGAN The variability of the 1923 blossoms and the fact that not a single progeny bred true to its parent color, indicates that considerable crossing was taking place, that the parents were quite heterozygous, or that some unknown complications existed. The problem was to obtain some seed that was known to be pure. These pure strains, if secured, could be used to continue the work. There were two alterna- -19- tives. The first was to secure seed from an external source that we knew was pure E, falcata and some that was pure g, sativa. The second was to self-fertilize plants in Mich- igan and to work with known selfed plants. A combination of these two might work best; so some selfing was under- taken. There are differences of opinion as to self-steril- ity of alfalfa, and as to what causes fertilization and tripping. De Candelle (1832), the first to give an expla- nation of this, believed the explosion would take place when the flower reached a certain stage of maturity. Hilde- brand (1866), Henslow (1867 a 1879), muller (1883), Piper (1914), and Hayes and Garber (1921) believed it is possible for plants to self-fertilize and set seed. Other writers, Roberts and Freeman (1907) and Blinn (1920), have found that some plants set seed well when caged while other plants do not. Some writers believe fertilization is only the result of insect tripping (Urban 1873, Kirchner 1905, and Fruwirth 1906). Burkill (1892-‘95) and Oliver (1910) were also unable to obtain seed without tripping, but this did not have to be preformed by insects. Interesting results were obtained by Hagem (1919) who says that sterility in- creases in after generations. The latter is important in inheritance study as it is often necessary to self several generations. there is a great controversy or difference of Opinion. -20- From the previous work on self-sterility we see 3e are assured of the fact that alfalfa is not entirely self- sterile, and in some cases may produce good yields when selfed. are as shown in Table A. The blossoms were not tripped artificially. Plant Number 30576 30957 36525 36905 36971 38042 31836 32413 34032 310928 34920 311204 - Table A - The writer caged twelve plants, and the results Showing number of seed produced from caged plants. Blossom Color fhite White White White White White Yellow Yellow Yellow Yellow Red-purple Variegated No. of Pods Set 2 14 14 74 No. of Seeds 2 12 O 14 72 13 65 26 O 16 161 of Seeds Per Pod 1.000 0.857 0.000 0.269 1.440 1.182 1.048 1.182 0.000 1.600 0.857 2.176 -21- Only two plants in Table A failed to produce seed, while three plants produced quite well. The variegated plant number 311204 produced twice as many seeds as any other plants. This may be due to heterosis, as the blossom color indicates that the plant is a hybrid. This increase in yield is similar to the increase Waldron (1920) found in Fi hay yields. The table gives the number of seeds obtain- ed from each plant, but the writer feels certain that not more than 50% of the seeds will germinate, due to their poor quality. It is interesting to compare Table A with Table B. . Table B shows the weights of threshed seeds from uncaged plants open to natural pollination. It also shows the number of plants with the same color of blossom that were not harvested, because there appeared to be no seed upon them, or only a few seeds which would be lost in the thresh- ing machine. Several of the plants harvested had only a few seeds. The weights of these were calculated by count- ing the number of seeds in 0.5 gram samples. There were found to be 51 seeds in 0.1 gram. This table, B, shows the yellow blossomed plants gave greater seed yields than the white blossomed plants, indicating that the yellow blossomed plants are likely of greater fertility. The average weight of seed per plant is very small, and assuming the plants not harvested had even less seed, we see that the real mean production was -22.. very low and some caged plants in Table A produced well in comparison. - Table B - Showing seed production from plants not caged, but exposed to natural conditions. Color No. of Total Ave.Wt. Seed Approx. No. No. of of Plants Wt. of Per Plant Seeds Per Plants Blossom Harvested Seed Harvested Plant Harv. Not Harv. Whites 64 9.88 0.154 78 57 Yellows 157 37.38 0.238 121 58 These results indicate that selfing in many plants, if caged alfalfa plants result in autogamy, for at least one generation will be possible in our breeding work. This will prove a great help in obtaining desirable segregations in at least the F2 generation. An effort to grow some seed from caged plants in the winter of 1923-'24 resulted in a failure. Three seeds out of 21 germinated, and later they died. One was killed by molds in the germinator and the other two were killed by a frost after they were two inches high. It was thought best not to grow any more seed during the winter, but to wait until spring, even though the results from these selfed plants would have verified or contradicted the conclusions of this thesis. -23- VIII -- ALLOGAMY vs AUTOGAMY On account of the great variegation the genetic analysis of the flower color of these plants seemed an un- determinable quanity. The amount of crossing that took place was unknown. Artificial crossing was therefore post- poned a year, until some idea was determined as to the gen- etic constitution of the plants. It was a general belief, until a few years ago, that alfalfa was self-fertilized but new evidence shows that crossing in many sections is common. Waldron (1919) checker-boarded M, sativa and M, falcata and found the per- centage of hybrids as follows: "2,099 plants coming into bloom certainly having E, sativa for a pistillate parent, 157 (or 7.48%) indicated certain falcata flower color char- acters and were of hybrid origin. 0f the 1,862 blooming plants with E, falcata for a pistillate parent, 795 (or 42.7%) showed sativa flower characters and so were of hybrid origin, ..." This percentage of crossing does not show what took place between plants of the same Species; so we may judge that the real percentage of crossing was much larger than the percentage given, and possibly as high as 85.4% in M. sativa. Witte (1921) says, "the hybrid between the two species is very easy to bring forth, for even without castration of -24- the stamens, cross-pollination gives a higher percent of hybrids than of individuals which have arisen through self- fertilization." 1 - Evidence for Self-fertilization in Michigan. The stand taken by alfalfa workers in Michigan is that alfalfa is considerably self-fertilized. Let us see what they base their judgement upon, and in what way my re- sults work in accordance with this theory. It has been stated earlier that the progeny rows grown side by side at the Michigan Experiment Station, exhibited to general ap- pearence a striking contrast of morphological and physiolog- ical characters. For example, some rows had a pale green color of foliage, while the foliage of the next row had a deeper green color. Plants in a row had an upright habit of growth, while a contrasting row had nearly a recumbent growth. There is a small per cent of exceptions in each progeny row that may break slightly the uniformity of its appearence, but these exceptions can be accounted for by a small per cent of crossing or segregation taking place of some of the more complex characters. It is believed that if a large per cent of crossing was taking place there would not be such contrasting differences of appearences between these rows. It has also been suggested that purple blossoms are -25- expressed by a number of factors as color in the aleurone of §§a_m§y§, and crossing in a large per cent would soon make all the blossoms purple in color, as it is dominant to all the other colors. The alfalfa nursery of the hich- iga Experiment Station is a flower bed in respect to flower color, as nearly all colors and degrees of variegation are found. According to the theory no great amount of crossing can take place here, because nearly all of the white, yellow, and variegated flowers would disappear from the nursery. This would leave the nursery a field of purple flowers, except in the case of the hybrids between E, falcata X g, sativa which are extremely complex. In the West the flowers are all purple, as open pollination takes place there. The results in Table 22 of progenies coming from white individuals has been suggested as showing that self- fertilization has taken place. There are segregations of white and purple flowers in ratios such as, l : 3, l : 4, l : 5, l : 8, 1 : 16, and l : 19. Ho satisfactory explana- tion can be offered for such ratios, but the idea is that progenies of similar ratios are genetically alike and differ from progenies of other ratios. This might happen if there were several factors involved, and groups differed by one factor. - PLATE IV - The 1923 alfalfa nursery showing method that was used in selfing alfalfa plants. -25- 2 - Evidence for Cross-fertilization in Michigan. Evidence was presented earlier to indicate that yellow flowered plants and white flowered plants should breed true under perfect autogamy. From the results obtain- ed in the 1925 nursery, we find that the progenies of yel- low and white flowered plants did not breed true. In the results shown in Table 10 we notice that (in 1923) 18.1% of the progenies of known white flowered mother plants came true to white. We also observe in Table 12 that 20.8% of the progenies from known yellow flowered mother plants came true to yellow (in 1923). It would be hard to estimate, from purple flowered plants, the per cent of crossing taking place, for several reasons. Purple flowered plants and plants with purple in their inheritance greatly outnumber the white and yellow blossomed plants. Hence, there would be more chances of purple crossing upon purple, than upon whites or yellows. Purple is likely due not only to one factor, but to multiple factors, any of which can express purple. If yellow and white flowered plants did cross upon a heterozygous purple, there would be only a slight chance that the zygote would be minus all the factors of purple, and could be distinguish- ed as a cross from the purple blossomed plants. At present we are unalbe to tell if crossing is going on in purple; so we must limit the discussion to yellow and white flowered progenies. -27- Progenies in 1923, from white and yellow flowered mothers, gave nearly the same percentages of individuals like the female parent. This could indicate that they were influenced by similar pollination. If we now contrast these results with another year we would have more evidence to draw conclusions from. In 1922, we find in Table 2 results which are similar to those obtained in 1923. For example, 14.6% of the progeny plants from known white flowered mother plants came true to white, and in Table 8, there were 5.9% of the progenies from known yellow flowered mother plants that came true to yellow. Summarizing this into a table we have: - Table C - Showing the percentage of individuals that came true for yellow and white bloSsoms. mother No. of Individuals No. of Individuals Flower Progenies Like Mother Progenies Like Mother Color 1922 Color 1922 1923 Color 1923 fhites 7 198 or 14.6% 23 353 or 18.1% Yellow 3 17 or 5.9% 14 245 or 20.8% The results obtained in Table C are hard to explain, unless crossing takes place, as we can think of yellow and white only as recessive to purple. The other progenies in the 1923 nursery, also indicate crossing may have taken place. There were 93 progenies of individual plant selec- tions, but not a single progeny bred true to one color. The -28.. history of these selections was traced back to find out when their ancestral stock was introduced into Michigan. It was found that 24 of the progenies had been grown in Michigan for six generations; 62 progenies had been grown in Michigan for five generations, 3 progenies for only three generations, and 4 progenies for just two generations. "Percent of Heterozygous Individuals in Each Selfed Generation when the Number of Allelomorphs Concerned are: l, \ . 50% \1 x s \10\15 \ \ \ 25,952 \ ‘ \ 1 2 :5 4 5 ‘5 '7 s 9 10 Segregating Generations "Graphs showing the reduction of heterozygous in- dividuals and of heterozygous allelomorphic pairs in suc- cessive generations of self-fertilization." After East and Jones 1919. -29.. It would seem probable that if self-fertilization were taking place 86 of the 93 progenies should have shown up more to expectations, because they would have had five or six generations of self-fertilization behind them. If there are 5 main allelomorphic pairs for color we would expect, after five or six generations of perfect autogamy, 85.32 or 92.4? respectively of the population of the progenies to be pure for one color. Undoubtedly, in the rigid selection work that the ancestors have gone through, a higher percentage of heterozygotes have been selected than would be expected. Doubtless, more of the white, yellow,, and purple flowered plants, that may have represented pure lines, have been discarded than would be expected, because the variegated plants seemed to respond better to the re- quirements necessary for high production. If we consider 25% of the progenies pure, we should have obtained some progenies in the 1923 nursery that bred true. As none came true to one color, the results indicate cross-fertilization, or greater complication in the inheritance than we have assumed. 3 - Discussion of Fertilization. The evidence of the West proves that crossing is large, and experimenters in Michigan should disprove that crossing takes place here, rather than to assume it does -30.. not, and require proof for crossing. What factors has the West to influence crossing which do not exist here? It does not seem possible to differentiate the two on different insect life, and say that the common insects here are more active or less active in their work here than there. The Megachile bees (Sladen 1916 & Piper 1914) will cause a large amount of tripping and perhaps crossing but we can not say they are more numerous in the West than in Michigan. The Megachile bees have been found in this state, but even a small difference in their number would not account for the difference between Open-pollination and self-pollination. The same might be said of thrips and many Lepidoptera that visit the blossoms, but so far they have not been proven to do much tripping. We have left an important factor, and that is environ- ment. We know that sunlight will cause a large amount of tripping and perhaps the tripping in the West is largely due to this. There is a chance that crossing may now be accom- plished by the wind or insects, if the tripping did not cause self-fertilization. In Michigan, flowers will also trip automatically as proven by caged plants setting seed, nearly as well as uncaged plants; therefore, we might have the same opportunity here as in the West. The humidity of the season is thought to effect the seed production, and it is also likely to effect the fertilization. As heavier -31- seed yields (Cox 1922) have been obtained in Michigan in dry years, or years of only average rainfall, it may sug- gest that the humidity effects fertilization here in Mich- igan, but, at present, there are no experiments to show if humidity might produce more self-fertilization than cross-fertilization. The effect of environment on seed production is well illustrated by the Idaho and Dakota Grimm seed that is shipped into Michigan. In Michigan this seed produces a hardy plant, but only in rare cases does it make a successful seed crop. It appears that external factors, without scientific proof to disprove them, at present, prob- ably offer the same chances for crossing in Michigan that occur in other sections but environment determines largely the amount of seed produced. Many strains of Grimm alfalfa show marked differ- ences, but when we contrast Grimm grown in the West with Grimm grown in Michigan, we find they are both variegated in flower color. Then, the fact that we have variegation in Michigan is not an adaquate proof of self-fertilization here, because variegation also exists in Grimm varieties where crossing takes place. The first plant to blossom in the 1923 nursery was observed on July fourth, and from then on until heavy frosts in October there were plants blooming and setting seed. It does not seem probable that if Open-pollination takes -32- place in Michigan, the genetic make up of the fertile pol- len would be the same at all times. For example, the major- ity of the yellow and white blossomed plants may not bloom until late in the season or vice-versa in respect to purple. Headden (1896) has mentioned that there is a difference in the maturing of plants. He says, "the deep green, narrow- leaved, red-stemmed plants mostly with deep violet purple flowers, present a very different growth and mature earlier than the light green, large leaved, green stemmed and as a rule lighter flowered plants." It seems likely from the results obtained, in Tables 2, 8, 10, and 12, on the white and yellow flowered progenies, that some grogeny parents blossomed early in the season, and some later, as segrega- tions of the progenies are not alike. This may account for the ratios of white and purple as shown in Table 22, and not indicate autogamy in white flowering plants. Also, white is considered recessive to purple; therefore, why should it segregate more purples than white blossoms if selfing is taking place? There is a probability that there exists an indivi- dual difference in alfalfa plants in their ability to self- fertilize and produce seed. This is especially noticed in caged plants, unless the pollen is carried through the cloth by wind or very small insects. Certain plants, even of allogamous races may have the performance of producing 50% -53.. of their seed by self-fertilization, and other plants may have have the ability to produce seed of which 25% has been self-fertilized. The other 50% and 75% respectively are fertilized by open-pollination. This may also be the result, if there exists a factor in certain plants making them ster- ile to their own pollen. Although definite conclusions can not be drawn from these results, they indicate that crossing probably takes place here in Michigan on white and yellow flowered plants. If there are two different yellow pigments one of which could be a heterozygous yellow and show dominance to purple, it would partially help to explain self-fertilization in hich- igan. The latter is uncertain as anthocyanin inheritance indicates that yellow is always recessive to purple and red. Ho eXplanation can be suggested to disprove the belief in self-fertilization based on the uniformity of the morphological and physiological.characters other than color in the Michigan Experiment Station alfalfa nursery. Hagem (1919) and witte (1921) both working from known Fl have been unable to explain their F2 generations, and it is not likely that the results of one or two years from field selections, of unknown origin except on the female sice, can be explained. The work requires a great deal of further study, and a different method of attack, than has beem followed so far, to obtain the fastest results. Uitte -34- had some progenies in his F2 that had as many as 200 and 300 individuals; therefore, it would seem that working with sufficient numbers we would be able to get a few populations which would not be subjected to large error. IX -- IHHERITANCE 0P AEZTHOCYAI‘IIN-l Anthocyanin pigments are found in a large number of plants. Iolisch (1905) has found the presence of solid and crystalline anthocyanin in the cells of Medicago sativa. Gertz (1906) has pointed out that the anthocyanin is gen- erally subepidermal in the Leguminosae, but it is found in the epidermis of Medicago, and several other genera. It has been noticed as the cause of brown or black Spots on the leaves of ledicago maculata. Anthocyanin inheritance is not simple, and although it has been studied successfully in many plants it has shown marked variations. There are some points in common, and an attempt will be made to show that some of these may be of help in realizing more fully an understanding of alfalfa blossom colors. The presence of anthocyanins may be due to a unit Kendelian factor or complementary factors. It may cause a color in certain organs, but in the absence of the antho— cyanins we will obtain wnite or yellow varieties. Perhaps lReference Wheldale 1916. -35- this is what happened in the white and yellow flowers in alfalfa. The yellow may be due to plastids or to soluble pigments. The inheritance of these two yellows differs, but they are, in many cases studied, dominant to white. In both plastid and soluble yellow pigments there may be a partial loss of the pigment and we obtain a lemon yellow or light yellow. There was variation noticed among yellows in nearly all the progenies studied, and some are called light yellows (lY) and some yellows (Y). There is then a possibility that in alfalfa flowers there may exist this factor that causes a partial dilution of the pigment and produces a lighter yellow. In many plants it seems doubtful if a true albinism is reached. There is what might be called a "partial" albino. Here we may have a tinged effect of purple and pigment pro- duced in other organs. In the note taking in the field many alfalfa blossoms were called white, when they really showed a very light shade of purple in the veins or had purple buds. These blossoms were not true whites, but may belong to this class of "partial” albinism. There is the possi- bility that white may also be dominant due to an inhibitor for anthocyanin color. Barker (1917) describes some morn- ing-glory blossoms that may have the complementary factors for anthocyanin in the simplex condition, and not have any color produced. This blossom is waite in color but is not -36- a true albino or a pure white genotypically. Alfalfa flowers could have a pneuotypic white color but there should cer- tainly be some that were also genotypicly white and breed true. A pigment called cream has been noticed in some plants. This pigment is in the plastics and is orange- yello , but not in very large quantities. This particular plastics in Lendel- .L color is recessive to white or colorless ian inheritance. The writer has called alfalfa blossoms cream in some cases, but in nearly all cases it was uncer- tain whether they were a light yellow or a true cream. In many of these so called cream alfalfa blossoms, white was noticed in the buds, but was replaced by the cream in the open flower. The inheritance of redness and blueness is a great deal more complicated than the albinism. These should be treated separately, but it is not the purpose of this paper to consider all the details when results are without checks as the present ones are. Variations occur in both of these groups, and the color may run into many shades. Blue is usually dominant to red but Karjanus (1912) states that blue is recessive to red in Trifolium pratense, which is "‘9 closely related to the genus Medicago. In alfal a the writer has not observed what might be called either a true blue or true red. Both of these colors seemed masked by greater -57- complication, but it may be possible that they exist. Blue flowers are the result of an alkaline say, and red flowers are produced by an acid anthocyanin sap. In some species there exists a single factor to produce either one or the other of these conditions, while the presence of the two might produce a neutral purple. The theories offered by Theldale (1916) are: "that the blues and purples are more oxidized conditions of the reds, and secondly, that the blues and purples may be formed by the condensation of the red with some other aromatic substances in the cell." In a falfa blossoms the writer has observed purples or shades of blue and red that were darker or lighter than normal. It has been proven that darker colors are often the result of more pigment, and they are different in Iendel- ian inheritance from normal color by the absence of an inhibitor for darkness. The flow rs in which there is not as much pigmentation as normal flowers may be the result of heterosis, where the color intensity acts in a cumulative feet, or they may be homozysous, and differ from normal Pi) e by an inhibitor to produce as much pigmentation as normal. It is possible that the light purple blossom parents in Tables 19 and 20 are separated from true purples by the presence of some inhibitor for further pigmentation. haas and Hill (1915) summarize some of the factors -58- in the inheritance of color in many flowers as follows: "C, a chromogen which is not necessarily coloured, and WLiCh is, in all probability, a glucosidal flavone. "B, an oxioative enzyme, which acts upon C to pro- duce a red colour. "e, another ensyme which acts upon red pigment, due to the action of E, and further changes it to another pigment so that a different colour results. "A, an antioxitase which inhibits the action of E. ”R, a reductase which neutralizes, as it were, the action of E." Yith this knowledge of anthocyanin inheritance it is possible to explain some of the results shown in the tables by hypothetical formulae, but this is not satisfactory for all the cases. X - - COICCLUS IONS The results obtained in Tables 1, 7, 9, and 11 seem to indicate that crossing is taking place upon white and yellow plants here in Michigan. If crossing is taking place on these two types of plants, it is probably occuring on all types of colors. It appears that external factors, without the scien- tific proof at present, offer some chances for crossing in -39- Michigan, but inheritance and environment jointly determine the amount of seed produced. Results indicate that many alfalfa plants are not self-sterile in Michigan. Therefore, we will be able to self to good advantage in our breeding work. No explanation is suggested to disprove the belief in self-fertilization based on the uniformity of the morph- ological and physiological characters other than color in the Michigan Experiment Station alfalfa nursery. The variegated flowers observed by the writer were first purple, then green, and finally yellow. Creamish appearing flowers were first white and later turned a pale cream. It is evident that pure lines of color will rarely be obtained from selections in a nursery or open field. An experiment should be conducted to determine positively the amount of crossing taking place in Michigan. The results obtained by Hansen (1919) indicate that a true breeding white strain may be obtained. It is gener- ally believed that the wild E, falcata breeds true for yel- low. It is also recognized that nearly all southern alfalfas do not contain other than shades of purple. The northern grown hardy strains are usually marked by small per cents of greenish, yellow, and white flowers. The Mendelian factor relations that exist in the -40- inheritance of flower color are still undetermined. A hypothetical formula might be derived for the explanation of some colors based upon the literature of anthocyanin inheritance. The facts will likely be determined when individ- uals have been selfed a number of generations and the progenies carefully studied. The truth may be obtained sooner if crosses are made using pure Er sativa and pure E. falcata as the parents. -41- XI -- LITERATURE CITED Alefeld, F., 1666. Landwirtschaftliche Flora. Berlin. Barker, E. 3., 1917. Heredity studies in the morning-glory (Ipomoea purpurea L. Roth). Cornell Agr. Exp. Sta. Bull. 592. Bentham, G., 1865. Handbook of The British Flora. p. 189. Blinn, ?. K., 1920. Factors that effect alfalfa seed yields. Colo. Agr. Exp. Sta. Bull. 257. Book, H., 1552. De Stirpium .... Commentariorum libri tres. Argentinal. Cited by Oakley and Garver 1917. Bonnier, Gaston, -———. Flore ComplEte Illustrée en Coleurs de France, Suisse, et Belgigue. Tome III. pt. 128. Brand, C. J., 1911. Grimm alfalfa anc its utilization in the Eorthwest. U. S. Dept. ggr. Bull. 209. Burkill, I. H., 1892 - '95. On the fertilization of some of the species of Medicago L. in England. Proceedings of the Cambridge IhiloSOphical Society, Vol. 6: 142-145. Cited by Piper 1914. California Farmer, 1855. Vol. 3. No. 11. p. 82. March 15, 1855. Cited by Hendry 1923. Candolle, A. P. de, 1852. Thysiologie Végétale, t. 2 Paris. -42.. p. 548. Cited by Iiper 1914. Clothier, G. L., ----. Alfalfa. Kansas State Board of Agr. 2th. Biennial Report. fart III. p. 545. Columella, L. J. M., 1745. Husbandry. English translation by Millar. London. Book II. p. 81. Corbiére, L., 1895. Louvelle Flore de Hormandie. pp. 149- 152. Cox, J. F., 1922. Michigan grown alfalfa seed. Mich. Agr. Exp. Sta. guart. Bull. Vol. 5. 30. 1:17. East, E. M. and Jones, D. F., 1919. Inbreeding and Outbreed- ing. p. 90. Fruwirth, 0., 1906. Die Zuchtung der Landwirtschaftlichen Kulturpflanzen. Bd. 3. Berlin. p. 189. Cited by Piper 1914. Galloway, B. T., 1907. U. 3. Dept. Agr. Yearbook. p. 147. Garke, A., 1903. Illustrierte Flora von Leutschland. 19th. Edition. p. 157. Gertz, 0., 1906. Studier bfver Anthocyan, Akademisk Afhandling, Lund. 1906. LXXXVII+ 410 pages. Cited by Vheldale 1916. Gesner, Konrad, 1561. Horti Germaniae ... In Cordus, Valerius. In Hoc Volumine Continentur ... Annotationes in Pedacij. Dioscoridis anazarbei de Medica Materia Libros V ... p. 267. (Argentorati) Cited by Oakley and Garver 191?. -43.. Haas, P. and Hill, T. G., 1913. An Introduction to The Chem- istry of Plant Products. pp. 222-252. Hagem, 0., 1919. Linige F2 und F5 Generationen bei dem Bastard hedicagp sativa X E. falcata. Byt ...—.. Hagazin for Haturvidenskaberne LVI. p. 149- 165. Abstract in Genetica. Vol. 2. p. 555. Hahn, C. S., 1915. Report of the Alaska Agr. Exp. Sta. p. 18. , 1916. Report of the Alaska Agr. Exp. Station. Hansen, N. E., 1917. Hansen's white flowered alfalfa. South Dakota Agr. Exp. Sta. Rpt. p. 36-57. , 1919. Erogress with alfalfa. South Dakota Agr. Exp. Sta. Rpt. p. 29-30. Harrington, H. H., 1892. A study of the composition of grass- es. Texas Agr. Exp. Sta. Bull. 20. pp. 182- 184. Hayes, H. K. and Garber, R. J., 1921. Breeding Crop Plants. p. 207. Heaaden, W. E., 1896. Alfalfa. Colo. Agr. Exp. Sta. Bull. 55. Hendry, G. E., 1925. Alfalfa in history. Jour. amer. Society of Agron. Vol. 15. Lo. 5: 171-176. Henslow, G., 1867. Notes on the structure of Medicago sativa as apparently affording facilities for the intercrossing of distinct flowers. Jour. Linnaean Society, Botany. Vol. 9: 27-329. Cited by Piper 1914. Henslow, G., 1879. On the self-fertilization of plants. Transactions, Linnaean Society, London. Botany 8. 2, Vol. 1. pt. 6. p. 361. Cited by Piper 1914. Hildebrand, E., 1866. Ueber die Vorrichtungen an einigcn Bluthen zur Befruchtung durch Insekten- hulfe Botanische Zeitung, Jahrg. 24, Lo. 10. p. 75. Hooker, J. D. and Jackson, B. D., 1894. Index Kewensis. Plantarum Phanerogamarum Fasciculus III. p. 183. Hy, E., 1895. Observations sur 1e Hedicago media Persoon. Journal de Botanique. Ho. 9: 429-432. Joret, c., ----. Elantes dans 1'antiduité. V01. 2. p. 68. Cited by Laufer 1919. Kajanus, B., 1912. Ueber die Farben der Blfiten und Samen von Trifolium pratense. Fuhlings landw. Ztg. Stuttgart. Vol. 4: 763-776. Cited by Wheldale 1916. Kirchner, 0., 1905. Uber die Wirkung der Selbstbestaubung bei den Tapilionaceen. Haturwissenschaft- liche Zeitschrift fur Land-und Forstwirt- schaft Jahrg. 3. Heft 1. pp. 9—10. Cited by Piper 1914. -45- Konstantinov, P. N., 1914. Selection of yellow alfalfa at the Krasnokutsk Experimental Station. Selsk. Khoz. i Liesov. 246. (Oct.) pp. 173-191. Exp. Sta. Rec. Vol. 33. Ho. 9. p. 831. Laufer, B., 1919. Sino-Iranica. Field museum of Batural History. Vol. 15. Ho. 3. Chicago. L'ficluse, Charles de (Clusius), 1583. Rariorum aliquot stirpium, per lannoniam, Austriam et vicinas quascam prouincias observatarum historia, quatuor libris expressa. pp. 758-760. Antverpiae. Cited by Oakley and Garver 1917. Linnaeus, C., 1753. Species Flantarum. Tomus II. Holmiae . 1753. pp. 778-781. Martyn, T., 1792. Flora Rustica. Vol. 3: 87. Cited by Testgate 1910. Molisch, H., 1905. Ueber amorphes und kristallisiertes Anthokyan. Bot. Ztg. Leipzig. LXIII. pp. 145-162. Cited by Vheldale 1916. Koore, C. T., 1917. Self sterility. Jour. of Heredity. Vol. 8. Ho. 5. pp. 203-207. hfiller, H., 1883. The Fertilization of Flowers. London. p. 179. -45- Keldeke, 1878. Z D M G. V01. 32. p. 408. Regarding some analogous plant-names, see R. V. Stackelberg. 1900. ibid. Vol. 54: 108- 109. Cited by Laufer 1919. Oakley, R. A. and Garver, S. 1917. Hedicagp falcata, a 3e1- 1ow flowering alfalfa. U. S. Dept. Agr. N. 8. Bull. 428. Oliver, G. W., 1910. Eew methods of plant breeding. U. S. B. P. I. Bull. 167. , 1913. Some new alfalfa varieties for pasture. U. S. B. P. I. Bull. 258. Patane, Giovanni, 1916. The selection of cereals in Italy. Internat. Inst. Agr. Rome. Inter. Rev. Sci. and Pract. Agr. 7. Ho. 6. pp. 778-787. Persoon, C. H., 1807. Synopsis Plantarum. V01. 2. p. 356. Piper, C. V., and Co-workers, 1914. Alfalfa seed production; pollination studies. U. S. Dept. H. S. Bull. 75. , 1915. Forage Plants and Their Culture. Phares, ----- Book of Grasses. p. 3. Cited by Harr- ington 1892. Ridgway, R., 1912. Color Standards and Nomenclature. Wash- ington, D. C. Rivinus, A. a., 1690. Ordo Plantarum, guae sunt Flore Irregulari Konopetalo. p. 84. Lipsiae. Cited by Oakley -47- and Garver 1917. Roberts, H. F. and Freeman, G. E., 1907. Alfalfa breeding: material and methods. Kansas Agr. Exp. Sta. Bull. 151. Sladen, F. W. L., 1918. Pollination of alfalfa by bees of the genus Iegachile. Agr. Gazette of Canada. Vol. 5. Feb. pp. 125-126. Southworth, W., 1914. Alfalfa hybridization. Jour. of Her- edity. Vol. 5. No. 10: 448-457. , 1918. Forage crop improvement. Agr. Gazette of Canada. Vol. 5: 158-162. Feb. Trabut, L., 1917. Hybrid origin of cultivated lucerne. Inter. 1 Rev. Sci. and Prac. Vol. 8. Ho. 6: 848- 850. In Comptes Rendus de l'Académie des Science. Vol. 164. lst. Half-year. Ho. 16: 607-609. Paris. April 16, 1917. Urban, 1., 1873. Prodromus einer Honographie der Gattung Medicago L. Verhandlungen Botanisches Vereins, Provinz Brandenburg. Jahrg. 15: 13-17. Cited by Piper 1914. Vaillant, Sébastéen, 1727. Botanicon parisiense ......... p. 124. Leiden. Cited by Oakley and Garver 1917. Waldron, L. R., 1911. Variegation of European alfalfas. Science B. S. Vol. 33: 310-312. Feb. -48- Waldron, L. R., 1919. Cross fertilization in alfalfa. Jour. Amer. Soc. of Agron. Vol. 11. No. 6: , 1920. First generation crosses between two alfalfa species. Jour. Amer. Soc. of Agron. Vol. 12. he. 4: 133-143. Westgate, J. H., 1910. Variegated alfalfa. U. S. Dept. Agr. Bull. 169. Wheldale, E., 1916. The Anthocyanin Pigments of Plants. Cambridge. Willis, 0., 1910. South Dakota Exp. Sta. Report. p. 19. Vitte, H., 1921. Sveriges Utsadesfbr. Tidskr 31. Ho. 5: 185-200. Alfalfa breeding in Sweden.. -49- XII -- EXPLANATION OF TABLES Tables no. 1 to no. 8 inclusive are calculations for progenies grown in 1922. Tables no. 9 to no. 20 inclu- sive are calculations for progenies in the 1923 alfalfa nursery. Tables no. 21 and 22 are abbreviated results of 1922 and 1923. The colors in all the tables are abbreviated for convenience and saving of space. It is therefore necessary to explain the abbreviations. They are as follows: Y -- Yellow (Pt. 4, 23, -.)x G -- Green C -- Cream 1 -- light w -- White d -- dark R -- Red 8 -- smooky B -- Blue V -- variegated P -- Purple All adjectives will be written in small letters, and nouns will be written in capitals. Tables nos. 1, 3, 5, 7, 9, 11, 13, 15, 17, and 19 show distribution of the colors of progenies as the notes were taken in the field. Tables nos. 2, 4, 6, 8, 10, 12, 14, 16, 18, and 20 are groupings made in an attempt to simplify the results. J-Ridgway 1912. -50- XIII -- TABLES - Table 1 - Showing segregation of progenies from seed which was produced on plants having W V 1P or W blossoms (1922). 00047 00801 03160 03403 04050 08042 010458 0 V Y 1 C V le 1 gY V P 1 P V G 2 P V gB l 3 l 3 sP 1 wG V 1P 2 W V le 8 2 27 31 W V 1P 25 34 l 35 32 3 rP l 2 l l 1P V 1B 1 l 4 1 P V B 26 17 32 19 23 48 68 VIP 30 1P 16 20 27 99 52 17 3 P 116 112 125 37 81 61 78 dP 4 9 3 4 6 1 Total 197 195 191 192 197 197 191 grouped classes of Table l (1922). 00047 00801 03160 03403 04050 08042 010458 Total - Table 2 - Showing the number of individuals and per cent of C&Y W 33 34 35 59 34 198 P V 15 P 164 160 188 189 161 133 151 1146 C 87% v.7; P V7. 16.7 17.4 .5 1.6 1.6 17.8 .5 29.9 . 2.0 17.8 5.1 14.6 1.1 83.2 82.1 98.4 98.4 81.7 67.5 79.1 84.3 -52- - Table 3 - Showing the segregation of progenies from seed which was produced on plants that had purple or shades of purple blossoms (1922). 00145 00146 00169 00208 00237 00239 00259 00267 00269 00867 01153 01161 01164 01277 02026 02048 02249 02549 02569 02584 02734 02773 02784 03709 03770 04418 04522 05427 05636 05831 06048 06329 06342 06651 P V 0 {ON 1c p-i ‘14 m H94 >0 (10 >H >4:>:> >> bf'F34 C1: '33:: 6 5 74 5b 1 1 3 3d 3 3 2 1 1 1f 1 1 1 2 1 1 l 5 1? V W r? P 1e 10 Sr- FJP’ no r- 1? V 13 H HpHH a, 50a 8a 1649 05030103011005 km- 22 36 18 36 38 47 19 41 12 16 17 20 48 41 55 53 14 a, H 6 93 UICDI-‘vbOSI-‘OI‘O H (ONO @rbHNNHibNF-‘memtb ##- 07587 07869 07890 08066 08370 08531 08565 09210 09412 09434 01001 01017 01027 01051 01054 01061 01070 01085 TOTAL QOO‘SD I I P V C gY V P 6 2 2 4 9 7 8 1 One is rP Three are P V lY m (5 50 b' > 04 Q4 1 1 6 2 2 2 2 3 11 V B -53- - Table 3 Cont. - W V VIP 15 8 17 110 12 dP V G Two are W V yG One is W V le 7 1P 1P V W ‘.'J V 1e 18' UHNHNsb 7 31 m H a: p. b. ca 257:.- 2 19 2 7O 10 l 7 4 2 41 1 13 5 2 2 4 4 5 4 l3 7 3 51 5 1 44 10 35 8 2 6 25 6 27 2 15 2 2O 1 3 5 3 30 23 5 77 93 2634 666 e -- P V W f -- sP g -- W V P h -- C V 1P 1 -- dP V gR 71 12 80 41 65 78 68 73 31 34 28 47 42 67 59 54 37 54 3555 676 96 7901 -54- - Table 4 - Showing the per cent of segregation from progenies that are shown in Table 3 (1922 . :2. m5 =- 53 c) :> c: In > 53;; t» a: >>4:>:>:>:> mm :> Q4 0. 04 to FM 9+ E: 9:53 H r: 94 F1 04 cu Total 15 8 17 110 12 7 31 77 93 2634 666 3555 676 % .2 .1 .2 1.4 .2 .1 .4 1.0 1.2 33.3 8.4 45.0 8.5 )( 2.6%flvariegated 97.4% Purple -55- - Table 5 - was produced on plants that contained yellow, sometimes green in their blossoms (1922). 09802 #01 H0310 HUNNl—‘varP-x? l‘" NNHHUN LO .p. 09819 H HHHmmHmp 10 15 36 97 09850 21 29 10 pomHHH 94 09910 (000 HNQ Poll-J MN 98 09912 I-’ MOSQl-‘OO‘AH 10 03 09920 09930 F’ +4»: mqmqomqowwum CROP 01¢. 97 purple, 09935 CNN l-’ N CfiCflCflfitOrP-uD-OS (FUJO‘IQ 10 99 and 09944 H N COMM #0305!“ H l'-"l-"I'--'l\3(J1 13 14 97 Showing the segregation of progenies from seed that 010331 19 21 27 96 -56- - Table 6 - Showing the number of individuals and per cent of grouped classes of Table 5 (1922). 8 9 8 E E 8 8 8 2%". 8 E 8 8 8 8 8 8 8 8 8 g 8 Y 2 2 2 7 l 4 3 5 26 lY 2 l 2 3 1 2 11 C 1 2 2 6 W 1 l 29 31 P V 68 20 73 79 55 45 77 41 52 510 P 21 74 19 9 34 47 19 52 36 67 378 Per cent Y 2.1 2.1 2.1 7.1 1.1 4.1 3.0 5.1 2.7 lY 2.1 1.0 2.0 3.2 1.0 2.1 1.1 C 1.1 1.0 2.0 2.1 .6 W 1.0 1.0 30.2 3.2 P V 72.3 20.6 77.7 80.6 59.1 46.4 79.4 41.4 53.6 53.0 P 22.3 76.3 20.2 9.2 36.6 48.5 19.6 52.5 37.1 69.8 39.3 -57- - Table 7 - Showing the segregation of progenies from seed that was produced on plants that had yellow, light yellow, blossoms (1922). Htiflflfifijfi*dfij tSdetd Total a -- P V gB Y 09908 bPJF‘klfiOJ 22 13 21 99 IV Y V C 08639 09832 3 1 4 8 2 4 2 4 1 3 1 1 2 4 12 2 2 2 1 5 l 10 8 2 1a 1 1 1 3 1 7 2 6 9 39 21 12 4 98 95 TOTAL 4 13 2 H Cnfi N H DDHCflNWQth~Q¢3UPJOJ® CRH IOCDh 28 F‘m \JH 292 or cream ~58- - Table 8- Showing the number of individuals and per cent of grouped classes of Table 7 (1922). Y lY Y V C 09908 08639 09832 TOTAL Y 3 1 4 lY 1 4 8 13 C 2 4 6 P V 30 30 35 95 P 68 59 47 174 Per cent Y 3.1 1.1 1.4 lY 1.0 4.1 8.4 4.5 C 2.0 4.2 2.1 P V 30.3 30.6 36.8 32.5 P 68.7 60.2 49.5 59.6 -59- - Table 9 - Showing segregation of progenies from seed that was produced on plants that had W V le or W V 1P blossoms (1923). 22324 22334 22359 22361 22505 22510 22519 22624 22626 22628 22641 22646 22766 22929 23103 23307 23319 23321 23342 23356 23521 23712 25324 27958 Total f:- ,“ 10 4 12 17 13 -8 10 5 14 15 12 l3 l7 3 8 16 21 4 11 la 12 l 1 31 1 3 8 1 2b 17 C) W V lY cw F‘ +4 4 2 4 284 & lY V P W V 1P 1? V W P V W I V P VIP V W rP {OI-‘10 (UPI-4H O) (0 I—’ I0 I-‘CfiUlml—‘l-‘vhbltflwtbid OJ 3 4 3 14 2 3 8 9 13 3 27 17 56 P V B 1 le 1P |'-’ CNI-‘NUlCflxlml-J 01H MM 18 6 55 11 1 35 P (0.513005 ()3 M (\2‘ ...-l H 7 169 1314 39 1952 -50- - Table 10 - Showing the number of individuals and per cent of grouped classes of Table 9 (1923). C&Y vw w P C&X% 77% 0% P7 22324 1 1 10 73 1.2 1.2 11.8 85.9 22334 4 77 4.9 95.1 22359 1 12 71 1.2 14.3 84.5 22361 17 47 26.6 73.4 22505 2 13 61 2.6 17.1 80.3 22510 1 1 8 73 1.2 1.2 9.6 88.0 22519 6 10 51 9.0 14.9 76.1 22624 5 80 5.9 94.1 22626 4 14 63 4.9 17.3 77.8 22628 1 15 59 1.3 20.0 78.7 22641 1 12 71 1.2 14.3 84.5 22646 5 13 73 5.5 14.3 80.2 22766 1 17 43 1.6 27.9 70.5 22929 6 3 77 7.0 3.5 89.5 23103 2 8 55 3.1 12.3 84.6 23307 16 80 16.7 83.3 23319 21 105 16.7 83.3 23321 1 4 41 2.2 8.7 89.1 23342 11 46 19.3 80.7 23356 1 12 56 1.4 17.4 81.2 23521 2 1 31 94 1.6 .8 24.2 73.4 23712 1 26 3 81 .9 23.4 2. 73.0 25324 3 8 63 4.1 10.8 85.1 27958 3 8 17 49 3.9 10.4 22.1 63.6 Total 10 69 284 1589 .5 3.5 14.6 81.4 -51- - Table 11 - Showing segregation of progenies from seed that was produced on plants that had only yellow, light yellow, or cream blossoms (1923). ymmmy 1.7m1m; mmma ( )( ) s) u: o: o: 10 oz r4 e- rt 6: 8: t0 (0 c0 C) 02 rd cu b- 61 a: C) «n to .4 5) co ‘3 '0 b 0‘» H 'd" 05 10 L0 10 £0 10 O 05 05 £0 to 10 I'- b- b- (O b b- tQ 10 b b b- 6: d: 62 c0 02 c0 c0 ()3 21. (O 1.5 17.3 52.3 57.4 68.7 70.7 83.8 62.3 -53- - Table 13 - Showing the segregation of progenies from seed that was produced on plants that had yellow and purple and some- times green in their blossoms (1923). rsvyc 26671 34 10 12 26615 14 9 32 :0 to 0 <5 <1“ 0: cu «w in 02 b' oz 05 as (3 {O to «o (0 IR 02 01 oz 01 cu 2 2 2 l 3 1 2c 3 1 2 1 1 4 l 2 1a 1 2 l 1 1 2 1 5 l l 1 1 3b 8 10 9 3O 11 3 ‘ 2 22 1 1 3 14 10 8 13 5 2 2 70 36 13 40 45 c - 1 is W or C d. - V :17 V b E]? as 8) a) r4 0: c) m2 A) K) d‘ ‘1‘ <5 <5 <14 'd' b- b' b- b- h- 02 02 02 02 or 2 4 1 1 2 1 1 1 1d 1 1 3 l 2 7 8 4 4 4 2 2 2 . 7 4 1 1 1 1 1 3e 1 1 2 2 3 2 1 l 2 l l 3 4 43 8 23 2 8 10 3 4 6 l5 1 5 34 59 44 55 15 4 4 6 1 98 116 134 63 37 e-lisPVgB Table Continued iage 64 ,_. 27503 13 Showing the segregation of progenies from was produced on plants that had yellow and purple - Table -64.. 13 Cont. - times green in their blossoms (1923). o: H In [x m 1Y Y V Y n \J P V C 51 yG g? V P yG V P f V gY P V yG 1 V P T V Y P V G rP V d0 5 IP V gY r? V yG r? V Brown T V 1P ‘I’lr P V11: V W 113 ‘ w P V W E V 13 rP 28 P V B 1 1? 4 P 1 dP Total 39 5301 27718 27738 27741 OH-‘OJC Hume: 16 NH 54 F’ 27758 I""' H0115 15 F’ 27830 UNA? ()3 .5 H COHl—‘O‘: 37 10 to [Q E‘) a) a) b- I» 02 m 2 3 1 1 1 2 5 1 1 1 13 1 1 1 1 1 9 32 11 3O 29 4 6O 91 27936 (CM (‘3 46 *Jo’ 28123 ()7 seed that ans. 8 ome- Cfi 28130 TOTAL Hcv Cow—xxzucoowooa mcm a. \J O CO H+4¢>m +43 F.) a I-‘ml-‘OJOX (t O 41 1261 -55- - Table 14 - Showing the per cent of segregation of progenies shown in Table 13 (192 ). Color Total Po. of Per Cent A11 Progenies Y 53 4.2 1Y 10 .8 Y 5'0 1"" g :3 c 1.1 1P V C 3 .2 gY 17 1.3 yG 4 .3 gY V P 38 3.0 yG V P 21 1.7 P V gY 28 2.2 P V yG 13 1.0 P V 19.5 Y V P l .1 P V Y 15 1.2 P V G 14 1.1 rP V dG 30 2.4 r? V gfl' 23 1.8 rP V yG 27 2.1 rP V Brown 14 1.1 W 2 .2 W V 1P 2 .2 H V P 2 .2 le V W 1 .1 u 1.7 1P V W 7 .6 P V W 5 .4 rP V B 11 .9 rP 350 27.8 P V B 32 2.5 1P 80 6.3 P 72.7 P 400 31.7 dP 44 3.5 Total , 1261 100.0 -66- " 118.1316 15 - Showing the segregation of progenies from seed that was produced on plants that had cream and purple blossoms, and some also had yellow or blue (192 ). o2 on no 44 C) to to (D a) as :4 (3 r4 no N a) c: .4 en E- b- is «o 5- <3 r1 d‘ b- 5- 5- b» b- E4 «0 «o b» b- b- o— o— b- c~ b c) cu cu o2 cu (N or an cu o2 c0 E4 Y 6 l l 5 l 2 16 lY 3 l l 2 2 1 10 Y V W l 1 C 5 1 3 1 10 gY V P l 1 1 2 5 yG V P 1 2 3 P V gY l 1 la 3 P V yG 1 1 Y V P 4b 1c 1 lo 10 8 P V Y 1 l rY V P 2 2 W 15 14 l 1 2d 34 v 19 3 1 l 5 L V P l l 3 l 2 1 1 10 1P V W 1 15e 3f 1 2 1 2 25 P V W 2g 2 6 2h 5 9 l 3 1i 31 IP 13 4 4 8 10 5 3 l 2 8 58 P V B 2 2 4 1P 3 8 14 10 2 3 7 3 9 12 71 P 16 16 56 32 29 49 37 58 35 35 363 dP 1 5 1 l 2 2 12 Total 44 57 129 67 62 72 54 69 57 62 673 a - P V gB e - 1 is le V W i - P V bW b - 3 are C V P f - 1 is 1P V b? c - C V P g 1 is bP W d - l is bW h - 1 is P V wB -57- - Table 16 - Showing the number of individuals and per cent of grouped classes in Table 15 (1923). Y C w PV P Y2 0% w; 272 P% 26602 6 3 2 33 13.6 6.8 4.5 75.0 26719 4 6 19 28 7.0 10.5 33.3 49.1 27035 1 1 39 7 81 .8 .8 30.2 5.4 62.8 27124 2 9 4 5 3.0 13.4 6.0 77.6 27480 7 3 5 5 42 11.3 4.8 8.1 8.1 67.7 27706 2 12 58 2.8 16.7 80.5 27713 1 6 47 1.9 11.1 87.0 27726 1 5 1 62 1.4 7.2 1.4 89.9 27778 2 1 3 3 48 3.5 1.8 5.3 5.3 84.2 27779 4 1 57 6.5 1.6 91.9 TOTAL 26 11 105 23 508 3.9 1.6 15.6 3.4 75.5 -68- - Table 17- Shoming the segregation of progenies from seed that was produced on plants that had red-purple blossoms (1923). i3 :3 S 33 :33 ‘3 58 ‘5 8 2 05 CD 03 O3 on CD 0‘5 Ch 0“ E4 s as s 8 8 8 2:. 2:. s 8 Y 1 1 2 W V lY 1 l gY V P 2 l 3 yGVP 1 1 PVyG 1 1 2 dP V G 1a lb 1c 2d 1 6 P V rY 2 2 dP V rY 1 l rP V rY 3 3 1P V W 2 1 1 1 1 6 rP 63 4O 43 68 131 34 34 56 18e 487 P V B 1 1 3f 1 2 2g 10 dP V B 4h 1 l 1 7 1P 1 - 2 l 2 3 11 v 4 24 P 4 24 19 4 9 17 25 16 16 134 dP 2 5 13 12 7 4 4 47 Total 76 71 75 79 159 71 80 79 46 736 a - dP V gY d - rP V dG g - 1 is rP V B b - P V G e - 2 are er h - P V bW c - P V g” f - 1 is P V b? grouped classes of Table 17 (19 26912 26916 26941 26944 26945 26946 27933 27967 27990 TOTAL Y&W rPV rP 4 63 4O 43 68 3 131 2 34 34 1 56 1 18 3 11 487 - Table 18 - PV P 2 7 6 25 4 28 3 8 l 24 2 33 45 22 1 24 19 216 -69- mm. m ‘4 5). 1.3 Showing the number of individuals rPVfl 5.3 and per cent of r77; KO ()0 ()1 N) (0 37.3 10.1 15.1 46.5 56.3 27.8 52.2 29.1- -70.. - Table 19 - Showing the segregation of progenies from seed that was produced on plants that had very light purple blossoms and some that showed signs of turning white (1923). 53 :3 53 t8 5% £3 £3 13 .6 C5 O5 CO 10 O3 O N CU K“- 28 E3 83 53 E3 33 $3 53 ii N N CG N Y 1 1a 1 Y 7 P 1 w 1 1 2 4 t v le 1 1 r v 1P 2 le v w 3 12 v w 4 1 P v bw 1 19 v B 1 1 P v B 1b 1 2 rP 4 1 1 1 1 19 29 10 24 25 9 33 35 31 49 19 to 9 2 1 2 3 5 7 1 11 P 8 15 7 18 8 21 8 38 7 dP 1 1 3 1 Total 50 3O 34 49 24 72 47 74 72 a - lY b - dP V B Showing the number of individuals and per cent - Table 20 - -71- grouped classes of Table (1923). 26913 27916 29853 210555 28930 28938 210238 210265 210758 TOTAL Y W 7 1 1 l 1 1 5 4 1 1 3 20 PV 1P 31 26 28 14 4O .' 31 61 12 16 20 26 39 2 280 147 77.3 1.7% 14 O O 5 O 3 977; 1P% 62.0 61.9 OSSOK COLOR Y . a OEF SPFTNG F -72... - Table 21 - Showing the totals of grouped classes for all the progenies in the 1922 and 1923 nurseries. rx ‘4 [4'3 (1" N 0} CD 0 (\1 CH , ‘ t") C (\7 r-i O o o o ' ’ . l5 - (\I t4 O“) 00 G2 {0 H r 1 to t‘ LO (Q r‘ ‘0 C a b a.) F C: F N (1- r- i U) H (D O O 0 0 $9 91 v v?! LO N H o H G! N) p. r C. ox ND C‘ ‘9 t J 1‘ H O 0 O G <0 C‘ b- d‘ or (O H H C \1 0‘. H _ (I) b’) 8' U: C H H ‘7 co 5 ' , LC LC) N (‘1 o o o o to H d‘ CL) (C H <‘ 0 us H to K .1 m H O K) H C J o o o 0 CK? H >~ C- a) [—l o o o v Cd r: (3 ‘1‘ C17 N L) P»: D... A N H <0 ‘1‘ 0- 03 O (3.. L\ ~34 u: N o o o r—’ r—1 5'.) (G CO V.) L‘ H L‘ L0 00 N) 04 01 u cu b— 1¢ :0 $4 N F 0 0 9 N 2" Lf “”2 O u:- r—i O 0‘ H H LO 0 o o C... LO H (‘5 H I“) 0‘» {IJ UT) CW (3: H <13 H O (O N .' 01 I") LC) 0 o H ‘2‘ b’) r-l O (C H to r-I . . Cd t“ l‘ 03 co > r 4 5r) . . v m N) k (L. E ‘1. > E‘. E :4 FL. >4 E IN PARENT COLOR 1.8 92.7 .4 4.4 96.5 1.9 1.0 .6 -73- - Table 22 - Showing the ratios between purple and white blossomed plants. These calculations are taken from Tables 1 and 9. RATIO P w P% w% P w 22334 77 4 95.1 4.9 19.25 1 22624 80 5 94.1 5.9 16.00 1 22929 77 9 89.5 10.5 8.55 1 23321 41 5 89.1 10.9 8.20 1 25324 63 11 85.1 14.9 5.73 1 23103 55 10 84.6 15.4 5.50 1 22641 71 13 84.5 15.5 5.46 1 23319 105 21 83.3 16.7 5.00 1 23342 46 11 80.7 19.3 4.18 1 22505 61 15 80.3 19.7 4.07 1 22646 73 18 80.2 19.8 4.06 1 22628 59 16 78.7 21.3 3.69 1 22626 63 18 77.8 22.2 3.50 1 22519 51 16 76.1 23.9 3.19 1 22361 47 17 73.4 26.6 2.76 1 22766 43 18 70.5 29.5 2.39 . 1 RATIO P w 0 P% 7% 0% P w 0 22510 73 9 1 87.9 10.9 1.2 8.1 1 - .11 22324 73 11 1 85.9 12.9 1.2 6.6 1 .09 22359 71 12 1 84.5 14.3 1.2 5.9 1 .08 23356 56 12 1 81.2 17.4 1.4 4.7 1 .08 23521 94 32 2 73.4 25.0 1.6 2.9 1 .06 23712 81 29 1 73.0 26.1 .9 2.8 1 .03 27958 49 25 3 63.6 32.5 3.9 2.0 : 1 : .12 RATIO P w 0 2 w% 07% P w C 00047 164 33 83.2 16.7 4.97 : 1 00801 160 34 1 82.1 17.4 .5 4.71 - 1 . .03 03160 188 3 98.4 1.6 188.00 0 3.00 03403 189 3 98.4 1.6 63.00 1 04050 161 35 1 81.7 17.8 .5 4.60 1 .03 010458 151 34 6 79.1 17.8 3.1 4.40 : 1 .18 08042 134 59 4 68.0 30.0 2.0 2.30 : 1 .07 . .3A ..u a“ \o‘ utwll.elino!r ~ . r z .57.?) TY M77174777171771!11711177731171“