FACTORS INFLUENCING PHENOTYPIC EXPRESSIONS OF CYTOPLASMIC MALE-STERILITY IiM THE SUGAR BEET (Beta vulgaris L.) By George U . Hogaboam A THESIS Submitted to the School of Graduate Studies of Michigan State University of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Farm Crops 1956 ProQuest Number: 10008660 All rights reserved INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion. uest ProQuest 10008660 Published by ProQuest LLC (2016). Copyright of the Dissertation is held by the Author. All rights reserved. This work is protected against unauthorized copying under Title 17, United States Code Microform Edition © ProQuest LLC. ProQuest LLC. 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 4 8 1 0 6 - 1346 0 A CKNOWLEDGMEiNTS The author expresses his sincere appreciation for the valuable advice and guidance given by Dr, E. E. Down, professor of Farm Crops; Dr, S, T, Dexter, professor of Farm Crops; Dr. G. B. "Wilson, professor of Botany and Plant Pathology; and Dr. A. L. Andersen, associate professor of Botany and Plant Pathology. The author is especially grateful to his associates, Dr. F. W. Snyder and Mr. B. W. Bockstahler, for their consultations and for their advice during preparation of the manuscript. The counsel of Dr. F. V. Owen, Sugar Crops Section, United States Department of Agriculture, Salt Lake City, Utah, was of great value in this investi­ gation. This study was made possible through the generous cooperation of personnel in the Sugar Crops Section of the United States Department of Agriculture. ii FACTORS INFLUENCING PHENOTYPIC EXPRESSIONS OF CYTOPLASMIC MALE-STERILITY IN THE SUGAR SEET (beta vulgaris L.) By George J . Hogaboam an abstract Submitted to the School of Graduate Studies of Michigan State University of Agriculture and Applied Science In partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Farm Crops Year Approved C As a type 0 plant. In our search for type 0 plants, many plants were found to be segregating for the pollen restorer gene or genes but no true breeding type 0 plants were found. The present investigation was undertaken to evaluate some of the factors influencing male-sterile phenotypes. The effects of U6-50 degrees Fahrenheit temperatures applied dur­ ing microsporogenesis were studied without successfully altering the phenotypes of the progeny SL9U60M. An attempt to alter the degree of pollen restoration through changes in the auxin concentration within the plant was without success for the concentrations of indole acetic acid and maleic hydrazide used in the experiment. A study of the Sx progeny of a plant, containing male-sterile cytoplasm yet producing pollen, indicated that a gene for pollen restoration and the gene for monogermness were carried on the same chromosome. Furthermore, this progeny revealed the indefinite pheno­ typic expression of the heterozygous pollen restorer genotype. Breeding experiments with another progeny lead to the discovery of a mutation which would complement the restoring action of the iv pollen restoring gene to the extent that the misclassification of the heterozygous pollen restoring genotypes -was improbable. The dominant complementary gene (Sh) was found to be lacking in pollen restoring ability when the dominant pollen restorer gene (X) was absent in male sterile cytoplasm. References 1. Owen, F. V. Cytoplasmically inherited male sterility in sugar beets. Jour. Agric. Res. 71s U23-UUO. 19^5. 2. 3* Owen, F. V. Utilisation of male sterility in breeding superioryielding sugar beets, Proc. Amer, Soc. Sugar Beet Tech. 157-lol. /?*?. Owen, F. V. The sugar beet breeder's problem of establishing male—sterile populations for hybridisation purposes. Proc. Amer. Soc. Sugar Beet Tech. I9I-I3U. 1950. t v TABLE OF CONTENTS Page INTRODUCTION.......................................... 1 literature review...................................... 2 Nuclear Gene Action............................. Cytoplasm-Nuclear Gene Interaction ................ Cytoplasmic Action........... 3 3 3 EXPERIMENTAL METHODS................................... 7 Experiment Experiment Experiment Experiment Experiment 1 .................................... 2........................................ 3........................................ h ........................................ 5........................................ 11 13 17 GENERAL SUMMARY AND CONCLUSIONS.......................... 27 BIBLIOGRAPHY.......................................... 29 vi 8 9 LIST OF TABLES TABLE Page 1. Assignment of Treatments and Schedules to Groups....... 10 2. Schedule of treatments by plant number for branch tips to be submersed in Auxin, Anti-Auxin, and Water........... 12 3. Number of plants in each phenotype................... lU li.. Chi-Square tests of 13H15-169 progeny.............. 16 5. Anther classification by dates, notes on set of selfed seed, and genotypes (where known) of the 128 plants in progeny 15-1-722................................... 19 6. Chi-Square tests of 15-1-722 progeny by dates of classifi­ cation fit to a 3 »1*h ratio....... 21 7• Derivation of Ratios Expected from Sib-Pollination of White Anthered Plants in I5-1-7E2 Progeny with the Yellow Dehiscing Anthered Plants in the same Progeny.......... 22 8. Chi-Square tests of open pollinated progenies from white anthered plants.................................... 23 9. Results of Chi-Square Tests to Determine Genotypes From the Selfed Progeny of the Yellow Anthered Plants in the I5-1-7E2 Progeny................................... 25 vii 1 INTRODUCTION The small, perfect flowers and indeterminate flowering habit of the sugar beet make it impossible to undertake large scale controlled cross-pollinations -without the cumbersome use of marker genes for identification of true hybrids. Although self-sterility has been used to enforce cross- and sib-pollinations, this character complicates inbreeding procedures and does not result in an all hybrid progeny. Large scale production of truly hybrid seed appeared possible with Owen*s {h9) discovery of cytoplasmically inherited raale-sterility. Ideally, this made it possible to develop emasculated equivalents of the desired female parent of a cross by repeatedly backcrossing the desired hermaphroditic plant to one with male-sterile cytoplasm. However, Owen reported partial pollen restoration through the action of nuclear genes. X and/or He assumed two dominant pollen restoring genes, in this type of cytoplasmic male-sterility. Therefore, it is also necessary to have the pollen parent of the emasculated equiva­ lent (female) line recessive for the pollen restorer gene or genes. ,L % 'I0) Type 0 plants^ then, have the cytoplasmic-genetic formula of iM xx or JMxxzz and owe their pollen production ability to the N (normal) cyto­ plasm as contrasted to the S (male-sterile) cytoplasm. In our search for type 0 plants, many plants were found to be segregating for the pollen restorer gene or genes but no true breeding type 0 plants were found. Conversations with other sugar beet breeders have revealed similar difficulties. The present investigation was undertaken to evaluate the factors influencing male-sterile phenotypes. 2 LITERATURE REVIEW The complete or partial failure of microsporogenesis in plants may be caused by several factors, of which some operate externally and other internally. Failure due to external factors have been reported by several investigators. Takenaka (.66) reported a temperature effect on failure of microsporogenesis in Colchicum autumnale. Miyazawa (li3) found pollen sterility in scarlet runner beans (Phaseolus multiflorus) growing at abnormally high temperature. Jones (32) reported sex reversals to neutral tassels of Zea mays by decreasing photoperiod. Pollen abortion as a result of photoperiod abnormalities was observed in soy beans (Soja max) by hielsen (U7). An entire series of failures in microsporogenesis was observed by Madsen (39) with an experimental group of Cosmos sulphureus CAV in which the photoinductive cycles and post inductive photoperiods were altered, waylor and Davis (U5) found that dilute concentrations of maleic hydrazide (0.025$) applied at the time of microsporogenesis caused sterility of the staminate inflorescences in maize while the pistillate flowers remained fertile. Rehm (52) states " . . . treatments with growth regulating substances has resulted in male sterility in cucurbitaceous plants and the tomato for periods of one to several weeks," WLttwer and hillyer (68), work­ ing with cucurbits, were able to produce plants with the usual number of pistillate flowers and no staminate flowers by spraying with maleic hydrazide (250-350 PPM) at the time the first true leaf was expanding. 3 The internal causes of failure in microsporogenesis may be con­ sidered under three categories, namely: nuclear gene action, cytoplasm-nuclear gene interaction, and cytoplasmic action* Nuclear Gene Action Male-sterility due to recessive mutants has been reported in mary plants, a partial listing is as follows: Alfalfa-Childers (10); cayenne pepper - Martin and Crawford (l;l); field corn - Easter (Id,19, 20), Beadle (6,7,8,9), Singleton and Jones (63), Emerson, Beadle, and Fraser (1/); rape - Morinaga and Kuriyama (UU); sorghum - Stephens (65), Hadjinov (26); sugar beets - Owen (50); Tomato - Lesley and Lesley (36), Rick (55,56), Mancini (Uo) . Male sterility due to dominant genes has been reported in alfalfa by Childers (10), and in Origanum vulgare by Lewis and Crowe (38). Cytoplasm-Nuclear Gene Interaction According to Jones (30) cytoplasmic pollen abortion was first discovered by Correns in 1908. Bateson and Gairdner (5) in 1921 reported two types of segregation for male-sterility in flax. In 1929, Gairdner (21;) defined cytoplasmic male-sterility in flax which was accompanied by pollen restorer genes, agreed with that of Gairdner. von Wettstein*s (67) report Rhoades (53) first described cytoplasmic male-sterility in corn in 1933. Jones (29) in 1951 demonstrated that nuclear genes would restore pollen production in cytoplasmic malesterile corn. Gabelman (23) partially characterizes the cytoplasmic factor (plasmagene) causing male-sterility in corn. He states h “The reproduction and distribution of this particle are quite similar to the reproduction and distribution of chromosomes. in reproduction holds both abrmeiosis and mitosis. The similarity The similarity of distribution holds quite well at mitosis but is completely random at meiosis. It must, therefore, have many characters in common with the chromosome." Newlin (1*6), working with maize, found that the nuclear genes for pollen restoration in male-sterile cytoplasm may vary in their effect under different environmental conditions. Rhoades {Sh) found that the iojap (ij) gene in corn caused cytoplasmic mutations, one of -which was cytoplasmic male-sterility. ne stated "Evidence of the particulate nature of this cytoplasmic factor is the occurrence of partially male-sterile plants with diverse percentages of pollen abortion. It is believed that these partially sterile plants arise from egg cells with a cytoplasm containing both normal and mutated particles. Completely male-sterile plants presumably possess only mutated particles while male-fertile individuals have non-mutated particles Cytoplasmic male-sterility in onions, described by Jones and Clarks (3l) in \9 h 3 as modified by the action of nuclear genes, may also be modified by temperature during the period from the tetrad stage to the first post m eiotic division according to Barham and Hunger (It) . In 19U5, Owen (U9) reported cytoplasmically interited male-steril­ ity in sugar beets in which there was a partial restoration of pollen production due to the action of nuclear genes. s Lamm (35) indicated a male-sterility in potatoes which was prob­ ably due to interaction between the cytoplasm of S, tuberosum and one or more genes of S. acaule. Shams-Ul-Islam Khan (3h) working with potatoes concluded: mPollen sterility is a complex phenomenon, involv­ ing chromosomal, genic, physiological and environmental factors.11 Michaelis (1|2) in experiments with Epilobium concluded that nuclear genes and plasmic units react together to form a complicated genetic system. Cytoplasmic Action Clayton (ll) in 19509 working with hybrids whose genome complement consisted of strikingly non-homologous genomes and replacing (by success­ ive back crosses) the female genomes of the hybrid with the male genomes, obtained cytoplasmic male-sterile tobacco from the cross hicotiana debenyi by h. tabacum. Clayton stated, ’’Male-sterility appeared in B C ^ increased in BC2 and was complete in BC3.*— wThe gradual development suggested progressive incompatibility between cytoplasmic and nuclear constituents.’1 Fukasawa (22) was able to do essentially the same thing to make a cytoplasmic male-sterile durum wheat in his studies on restoration and substitution of nucleus in Aegilotricum (a Aegilops ovata by Triticum durum hybrid). he found that where extra chromosomes of Ae. ovata occurred, in addition to the lit durum bivalents, only partial sterility resulted. After reviewing the literature, it appears that a delicate balance must be maintained between the plasmagenes in the cytoplasm and the 6 chromogenes in the nucleus in order to have normal microsporogenesis. Although either plasmagenes or chromogenes may mutate to cause malesterility, sooner or later a pollen restorer mutant uill be found. External environment may alter the balance sufficiently to cause malesterility or pollen-restoration, as the case may be. Apparently, only when the plasmagenes come from a widely divergent source from that of the chromogenes may one expect a lasting type of cytoplasmic malesterility (one without pollen res^b^ation). 7 EXPEftIMENTAL METHODS The following experiments were conducted to determine the role of various factors in pollen production in sugar beets containing malesterile cytoplasm: Experiment 1 — Aceto-carmine slides were made of various anther types to determine possible microscopic characteristics which could differentiate anther types. Experiment 2 — Cytoplasmic male-sterile annual plants were sub­ jected to varying lengths of cold treatments at different stages of seed stalk development to determine effects of cold temperature on pollen production. Experiment 3 — Branches of cytoplasmic male-sterile plants were treated in different ways with auxins and anti-auxins to determine possible auxin relationships to pollen production. Experiment 1; — A plant containing male-sterile cytoplasm, yet producing pollen, was selfed to observe the genetic segregation within the same cytoplasm and to observe possible linkages with the genes controlling monogermness. Experiment 5 — Observations were made on a number of single plant crosses to white anthered cytoplasmic male-sterile plants in an attempt to discover additional gene actions modifying the phenotypic expression of the major pollen restorer genes. reported in the progeny I5-1-7E2. Such an action is 8 B:xperiment 1 Materials and. Methods Several anthers -were collected from each of several plants known to contain male-sterile cytoplasm. A description as to color, full­ ness of anther, and physiological age of each anther was recorded by plant number. An aceto-carmine smear was made of each freshly collected anther and immediate microscopic examination was made. Notes were made on abundance, size, shape, and sculpturing of pollen grains. Results l/foite anthered plants. Some anthers contained no pollen grains, although normal tetrads were observed. Other anthers contained sculptured pollen grains which were empty, with collapsed walls, and smaller than normal pollen grains which were unstained. Yellow anthered plants. The anther contents ranged from small, empty, collapsed, unstained, sculptured pollen grains to apparently normal viable pollen grains. Conclusions Since some empty, small, sculptured pollen grains with collapsed walls were found in anthers from nearly every white anthered plant examined, it was felt that the presence or absence of pollen grains was not a safe criterion for differentiating genotypes in male-sterile cytoplasm. For this reason, microscopic examination of anthers was discontinued and a search for ways to develop greater visual 9 differences between white and yellow anthers was continued. Experiment 2 Male-sterile plants had been observed -which were thought to be white anthered only to be classified as yellow anthered plants with a subsequent observation. Based on Owen's suggestion of the possible influence of cold on pollen restoration an experiment was initiated to study the effect of cold temperature on the production of pollen in male-sterile cytoplasm. Materials and Methods Twenty-four containers were planted with seed lot 5L9U60M, an annual cytoplasmic male sterile seed lot which Owen considered to be homozygous recessive for pollen restorer genes. To control nutritional variability nutrient solution and sand culture was used. Supplemental incandescent lights from 6pm to 6am were used from seedling emergence , December 18, 195>3, to the completion of the experiment, March 26, The beets were thinned to ten per container. 19$h. The cold treatment was started on February 9, 195U» when seed stalks ranged in height up to six inches. In the cold room, continuous incandescent lighting was maintained over the plants. The temperature at the crown of the beets was maintained between U6 and 50 degrees Fahrenheit. The variables in this experiment were: the stage of development of seed stalk at time of exposure and duration of cold treatment. The schedule of treatments is shown in Table 1. 10 Results The opening flowers were observed daily to determine any sectors of yellow anthers on the plants which could be related to the period of cold exposure. A few scattered yellow anthers were observed on some plants in almost every treatment including the control which had no cold exposure. Neither the yellow anthers nor the white anthers were confined to any specific sectors on the plant. Table 1. Assignment of Treatments and Schedules to Groups (Containers) . Group JUo. Type of Treatment Treatment Started 1 1*8 hrs. cold 1*8 hrs. cold 1*8 hrs. cold 1*8 hrs. cold 1*8 hrs. cold 1*8 hrs. cold 1*8 hrs. cold 1*8 hrs. cold 1*8 hrs. cold 1*8 hrs. cold 1*8 hrs. cold 1*8 hrs. cold 2-9-5* 2-11-5* 2-13-5* 2-15-5* 2-17-5* 2-19-5* 2-21-5* 2-23-5* 2-25-5* 2-27-5* 3-1-5* 3-3-5* 2 3 h 5 6 7 8 9 10 11 12 Group IMO . 13 11* 15 16 17 18 19 20 21 22 23 21* Type of Treatment Treatment Started 96 hrs. cold 2-9-5* 96 hrs. cold 2-13-5* 96 hrs. cold 2-17-5* 96 hrs. cold 2-21-5* 96 hrs. cold 2-25-5* 96 hrs. cold 3-1-5* controly no cold treatment control. no cold treatment 288 hrs. cold 2-9-5* 288 hrs. cold 2-9-5* ll*l* hrs. cold 2-9-5* ll*l* hrs. cold 2-21-5* Conclusions Contrary to expectations, this population was not true breeding for all white anthered plants. The lack of any consistent pattern of anther color on the treated plants as well as the presence of yellow anthers on the check plants indicates that the cold treatments given in this experiment were not effective in altering the anther color of 11 the plants in the population SL9^60M. Although this experiment does not rule out the possibility that cold temperature may influence pollen restoration in plants having male-sterile cytoplasm, it can be said that a temperature of U6° to 50° F does not have a pronounced effect on pollen restoration. Experiment 3 A branch of a plant, containing S cytoplasm but having both white and yellow anthers, was observed to have a preponderance of yellow, full, and dehiscing anthers near the tip which had reassumed an upright position following accidental inversion. A temporary change in the auxin concentration during development of the anthers was postulated to account for the apparent shift to yellow, full, dehiscing anthers. This experiment was designed to test the possible role of auxin in pollen restoration. Materials and Methods Indole acetic acid (IAA) was used as the auxin source at a concen­ tration of 1,000 parts per million. Maleic hydrazide (MH) at a con­ centration of 200 parts per million was used as a growth inhibitor. Two methods of application were used. In the first method the ends of branches were submerged in the treatment solutions for varying lengths of time, one set of branches had the tips removed while another set of branches was left intact. Eight plants were used to test the effects of submersion of branch tips in IAA and MH. is shown in Table 2. The design of experiment 12 The second method of application involved spraying the terminal portions of the branches. Branches of white anthered and yellow anthered plants were sprayed with IAA while other branches of the same plants were sprayed with MH. These marked branches were observed during their subsequent blooming period. The tips of branches on many plants of both white and mixed white and yellow anther types were tied in an inverted position in an effort to substantiate the original observation. Table 2. Schedule of treatments by plant number for branch tips to be submersed in Auxin, Anti-Auxin, and Water. Hours Submersion MH 1.5 3 5, l, 10 k 2, 9 8, 2k kQ 72 IAA 5, l, 29 8, Water 3 10 5, l, 2) 8, k 9 3 10 k 9 Results JMecrosis of the tips of the branches occurred after 2k hours sub­ mersion in the IAA solution, hence all IAA submersion treatments were discontinued after 2k hours. The internodes of the IAA-treated plants elongated considerably, probably because of an increase in auxin content of the cells. IAA and MH had no visible effect on anther color. iMo alteration in anther color was noted from the spray treatments as compared to unsprayed branches. The tips of seed branches tied in the inverted position responsed geotropically but without an alteration in anther color. 13 Tips of seed stalk branches were removed from several white and several yellow anthered plants to alter the auxin concentration. effects were noted from this treatment. ho Perhaps leaflets also should have been removed to further alter the auxin relationship. Conclusion Although these experiments do not exclude the possibility of auxins having an effect on anther color the evidence fails to indicate that they play a major role. Experiment U Several cytoplasmic semi-male-sterile plants considered to be heterozygous for monogermness, Mm, were selfed to study the anther types of the segregates and possible linkages between the pollen restorer gene or genes and the gene controlling monogermness. Only one of the selfed plants, 13Hl£-l69, set enough seed to permit a thorough study of its progeny. The progeny from this plant were classified for monogerm and multi-germ type plants and also for anther types according to the following eight classess Y^ -all anthers yellow and dehiscing. Y2 -all anthers yellow, no dehiscing observed. Y3 -most anthers yellow and dehiscing but some anthers, at least, were not definitely yellow. Y4 -most anthers yellow, no dehiscing observed, some anthers "white”. Ill Ys -most anthers nwhite” but some anthers yellow and partially full. Y6 -most anthers ’’white” but some anthers yellow and shrunken, -anthers dark and shrunken with no yellow visible or anthers a light buff color not typical of white anthers. 'W3 -anthers white with no yellow coloration observed. The number of plants in each phenotypic class in the selfed progeny of 13H15-169 are presented in Table 3. The data indicate that the plant 13H15-169 was heterozygous at the locus for anther color and at another locus for monogermness. Table 3. *1 mono germ s 6 double germ : 37 total ii3 JNumber of plants in each phenotype. *3 : 0 : 2 2 *3 *4 Anther Types, ; *5 : 2 : 1 : 0 : 3 : ll| : U : 10 : 3 16 9 10 6 wa Total : h i 19 33 : 93 : Ii9 : 138 19 86 187 t If the cytoplasmic-genetic constitution of plant 13H19-169 for anther color and monogermness is assumed to be S XxMm, then there were far too many white anthered plants in the progeny to permit an explana­ tion that one dominant X gene will always result in a yellow anthered plant. However, if one postulates: first, that the homozygous dominant (XX) genes in male sterile cytoplasm (S) produce plants which have all yellow full anthers; second, that the homozygous recessive (xx) genes in S cytoplasm give all white anthers; and third, that the 15 heterozygous (Xx) plants have anther type classifications which range from the Y3 category (most anthers yellow full) through the -white category, then the homozygous dominant Y^^ and Y2 categories (all yellow, full anther phenotype) can be tested versus the combination of the other phenotypes using a 1:3 ratio. Such a chi-square test gave a good fit, with a probability figure of 70$ to 80$ (Table k ). Another possible explanation for anther color would be a duplicate recessive epistasis ratio, in which xx is epistatie to 2, and zz is epistatic to X. to 7 white. Such a gene action would give a ratio of 9 yellow The fit of this ratio to the data is dependent upon the genotype of the anther color phenotype. If the Wx anther color type is classed as yellow, the chi-square test gives a fit of p * 50-70$, however if the anther color is considered as actually white then the chi-square test gives a fit of P = less than 1$. Other data to be given in experiment 5 concur with the postulates of the first explanation of the inheritance of anther color. The chi-square test in Table h indicates P value of 70$ for one set of alleles controlling the monogerm character, which is in agree­ ment with the findings of Savitsky (58). The dominant multigerm all­ ele involved in this progeny was probably the M 1 allele described by Savitsky (59) since no more than tw© flowers per cluster were observed on the parent plant or its progeny. "When the two factors, anther color and monogermness, were tested together in a chi-square test the P value was reduced from 70$ for each independent ratio to 10-20$ for the combined independent 16 Table U. Chi-Square tests of 13H15-169 S^ progeny. Class Observed Calculated (o-c) (o-c) 2 c 1*3 segregation all yellow full anthers v.s. "white anthers" Yellow Cyx-y2) "White" (y3-w) 15 lh 2 18? U6.75 iUo.25 1B7T00 -1.75 1.75 0.00 0.0771 0.0218 =0.0989 P=70- Multigermness v.s. Monogermness 3:1 segregation Multigerm Monogerm 138 h9 187 1U0.25 1+6.75 187.00 -2.25 2.25 0.00 0.036 0.108 x2=07lU+ V=70% Combined germ type and anther color (independent inheritance) Yellow multiwVJhiteH multi Yellow mono•'White" mono- 39 99 6 1+3 187 35.07 105.21 11.69 35.07 187 .01* 3.93 6.21 -5.69 7.93 - 0 .1*1*01* 0.3665 2.7696 1.7931 x2=5 !3696 P=10-20$ Combined germ type and anther color (Linkage in coupling phase with 37.5$ crossover) Yellow multi"White" multi Yellow mono*White11 mono 39 99 6 1+3 IFF 1*0.18 100.08 6.57 1*0.17 187.00 -1.18 -1.08 -0.57 2.83 0.00 0 .031*7 0.0116 0.01*95 0 .1991* X 3 =0.2952 17 inhertiance ratio. If a linkage In the coupling phase, with 37.5$ crossover, is assumed, then the P value is increased to 95-98$. Pleiotropic action of the monogerm gene in the recessive mm genotypes to give more "white" phenotypes would be another possible explanation for the reduction of the P values upon combining the independent ratios. The linkage theory is preferred. Experiment 5 The anther color segregation was checked for a number of two plant (one normal cytoplasm hermaphroditic male plant and one malesterile cytoplasm, white anthered female plant) crosses in an attempt to discover more about the action of the pollen restorer genes, An unusual gene action was noted in the progeny I5-1-7E2. The 128 plants in this progeny were given permanent numbers to identify them in sub­ sequent studies. The plants were individually classified at four different dates as to anther color, fullness of anthers, and dehiscence. They were first classified in the field June 29, 195U when they were past full bloom. These same stecklings sent out a second growth of seed stalks as the first crop of seed was maturing. The anthers were classified again August 27 which was approaching the end of the second blooming period. The plants were potted in the fall and moved to the greenhouse where the third reading was made February 2h, 1955, just previous to full bloom and the fourth, March 3 as the terminal bloom­ ing period approached. The following anther classification was used: ly -yellow, full, dehiscing. 2y -yellow, full, delayed dehiscence. 18 3y —■ anthers of earliest blooms lacking definite yellow color­ ations and shrunken in appearance, however, in later blooms yellow partially full and dehiscing anthers appear. hy -same as 3y except with delayed dehiscence. £y -yellow, shrunken, no dehiscence, anthers of earliest blooms may lack definite yellow coloration. 6W -anthers white, shrunken, sometimes darkened before blooming. Four branches of each plant moved to the greenhouse were bagged just prior to blooming. Of the 128 plants, U died after the first reading of anther type; 23 of the 128 were always judged to be ly and of these, b set no selfed seed, 2 set seed with imperfect embryos, and the other 17 set normal selfed seed but only 10 produced enough for progeny testing; 28 of the 128 had yellow full anthers in at least one of the classification dates and also fell within the categories 3y to 6W at least one time; 1? of the 128 had at least one classifi­ cation of 5Y and either a 5V or 6W the other times, 10 of these were 5_ at all classifications; while the remaining 56 of the 128 were y always observed to be in the 6 category. The complete classification of all plants on the different dates are given in Table 5 along with the identified genotypes from Table 9. From Table 5 it is apparent that anther classifications varied considerably between different reading dates for several of the individual plants. A complementary factor controlling the expression of fullness of anthers was indicated during the June 29, 195^ reading and re-evaluated at subsequent readings. This new gene locus has been designated the Sh locus for the generally shrunken anthers of the homozygous recessive phenotype. Depending on the genotypes of the parents, the 19 73 Table 5. Anther classification by dates, notes on set of selfed seed, and genotypes (where known) of the 128 plants in progeny I$-1-7E2. CD •H © Os •H ^ © —1 *£3 fOt i X CD g c 5 ft 73 ©boO ft ,0 to oa X CO - P -C-t ~G -G pCj rM CO CO CO ^^ COCOCO r-G iX X* X l X X 03 CO X X .-G G COCO CO CO CO CO COCO co 03 03 03 CQ X X to 1111 CO CO CO CO CO CO CO CO CO CO ( I rH i—1 H r l H r l H H c ^ H r l H C A-P s Q MD vO M3 M 3 sO i I r* i—I i—I i—I (—I i—1 rH i—1— zj H rH i—I CA LA vO vO vO SO MD H H La LA sO sO sO sO sO sO rH i—1 r—1 LA H LA LA so sO sO M3 MO MD P G i —1 o H O E rH rH rl H H H r l r l H H r ^ iH r l H r l rH rH rH > ft 7©3 -P |H <» © © ro ra rH O Oh s; 3 to © G -p © P X Q S rH bD bO b Q b O b O b O b O b O b O b O i—I i—1 LA CM XA I A H G [LA p G rH c d ftj -p 03 G 0 \C " - C— fA C "-O JX A C O V O C A rH - f t XA LA CO 30 On O rH (O O N -Z flA _ G -G O C''—s O O cA_ft i —1i —I _=t XAsO CO rH i—! (—I i—i i—I CM i—1 i—I i 1 i—I i— I CM CA-G LT\ CM CM CM CM CM (—I I—I rH I—I i—I O f t f - G XAMO L A X A M O MO sO CO o H J s o sO A - C— £"— C*— CO cAsO Os O I— 03 co CO Os CM Os H E— 73 Os Os O O O rH rH rH I—1— G sO £— CO rH fA CO Os O rH i—I i—I rH CM CM CA LAsO Os CM CM CM CM CM O CM sO Os rH CA CA CA CA-G G G P £ P LA CO CA_G C\J CM CA-ZjLA rH S g E E P so c5s r— H - X LA SO co o H ra X U3 p ctt £ X> rH O ft rH © CO OJrH H O i —!i —I l a cm O O bDbO -P ra © -p G ft ft bO bO bO g KO h 73 73 ca £ s p C G P tA rH ^ fA C 1 "—CO oo os e a a a s 03 PEP i— I Os sO LA C"- O CM OS Os Os O O O rH rH rH rH rH CM CALA LA 73 rH LA LA LA CA i —1i—Ii—1 r-H rH 73 73 - G psO LAP H sO LA X A - p _G CA 73 73 LA LA LA LA 73 - p sO LAsO _ p (A i—I rH XA rH — P rH i—! sO H LA XA H XA rH 73 XA CA rH casO > -p s O CA CA CA C A -P rH LA LA rH sO CA 73 LA C A -P C A fA (A CA m 3 _ P C A -P rH CM CA LAsO LA LA rH sO CALA CA rH O Ph ¥>a 331 P rH O G 03 O EI^H ft rH © CO 73 © -P © © 03 03 9 L A C— Os E G g E G g E E E E S g G g G - p A - , l _ p 'irs CM CM CA CA CA 03 o LAC0 03 < A _ p -P -p L A g 73 P E P 9 rH O P-. 3 CM f A O p s O rH rH rH W P P G © biO-p O ra G © ftft 73 © © 73 P E OS CM CA A—CM LA '•ft sO sO A- 73 G o © •H Ph ft o G r7> P 1—1 O G O © ft © bO •H > © X O ft •H 73 bD G ra P ft 03 P O G G P ra © O rH O o ft > a ft G G O 73 © Pi C © © X E 03 IS © P P -P X 73 © G P •H p i 73 f t © rH S « © © O 03 CO W o 73 rH © 73 73 ,P © © O 73 03 f t O P f t g rH bjO © O © 73 G 73 ra X b£) f t f t © f t 73 P G G rH o O P P © o G i—1 i—1 03 bO © f t f t U II II P g bO 73 II n > 20 presence of the Sh gene may change the expected backcross ratio of 1:1 for the locus to a ratio of 3 yellow full: 1 yellow shrunkenl li. white or a ratio of 1 yellow fullt 1 yellow shrunken: 2 white. example, S xxShsh by N XxShsh would give a For ratio because Sh fails to act in the presence of xx in S cytoplasm and for the same reason S xxShsh by Xxshsh gives a 1:1:2 ratio. The data in Table 6 show a good fit to a 3*1*^- ratio indicating the genotype of the female parent of I5-1-7E2 to be S xxShsh and that of the male parent to be XxShsh. To further test the action of the Sh gene, the progenies from open pollinated white anthered plants (classified 6W at all classifi­ cations) in the isolation plot containing the I5-1-7E2 progeny were indexed in the spring of 1955. Since the only source of pollen in this isolation plot was from the yellow dehiscing anthers, the geno­ types of the white anthered females could be determined, assuming even distribution of pollen, from the segregation ratios observed in plants grown from the open pollinated seed. The derivations of the ratios for each genotype are given in Table 7. Since the size of progeny required to differentiate the expected ratios was large, classification was completed for only six of the open pollinated white anthered progenies. Of these six progenies, four fit the 5 si*6 ratio for S xxShsh mother plants and two fit the 2:1:3 ratio for S xxshsh mother plants. these ratios are given in Table 8. The chi-square tests for 21 Table 6. Chi-Square tests of 15-1-7E2 progeny by dates of classification fit to a 3*1:1; ratio ) Date Class Observed 6-29-51+ l-Uy 5y 6W lt7 19 62 12B hQ 16 6h 12H -1 3 -2 U2 19 60 121 145.38 15.13 60.50 121.01 -3.38 hi 17 52 110 1+1.25 13.75 55.oo 110.00 -0 .2 5 3.25 -3 .0 0 U-27-51+ 1-lly 6w 2-23-55 l-Uy *y Calculated o-c 0 (o-c)‘ c 0.020b 0.5625 0.0625 x2=o.61^56 P =70-80% 0.2517 0.9899 O.OOiil -0.01 x?=l .2557 P =50-70% 3 .8 7 -0 .5 0 0.0152 0.7662 0.1636 0 .0 0 x^O .9U70 P =50-70% 3-3-55 l-Uy Sy 6w U6 19 53 HE UI+.25 ill.75 59.00 116.00 0.0692 1.75 I. 22 I 46 U.25 0.6102 -6.00 0.00 x^i^oUo P =30-50% Probable True Phenotype i-Uy 5 6w 51 17 56 12U 1+6.5 15.5 62.0 12I+.0 U.5 O.U355 0.lJi52 1.5 -6 .0 0.5806 0.0 X *3=1.1613 -s~\l P =50-70% 22 vs LA LA CM ft t© O T©3 *H -P £ ■H £ ft W a pT I rH •• Ad & X ft *H rH ,,--CO © += © l © I— I © Sr! CO »« X! CO « CM rH •• »• Xl co CM -_s rH X! CO Ixl rH -— / CM © i—1 xi •H tuO *H rH W) © £ o •H P o ft H LA CM CM XI w X & © tuO o CM ft XI ra o ■H P £ £ *H O cq ra pt ft £ PH £ © rH i—1 O rH XJ ra M £ T3 O I Vs ft r—v X3 B >* Pf I rH^ Pf £ © O o p £ o r^-H CO W p •H cd P £ © cO pT T £ © o £ © pT TJ ra © p © © rH £ S. X3 I 3 x; ra CO R a -P O ra -p £ © £ © rH s rH x: X3 5 CO M X3 m ><3 CO XJ CO H CO C O X3 i —f t-A >> LA co 0" £ MD R £ w cd m § © ft ft © Pt I X3 vO* >s la LA 23 Table 6. Plant ho. 1*3 Chi-Square tests of open pollinated progenies from white anthered plants. Best fitting ratio Observed Calculated 5 (W US 6 «£) 8 52 Io5 105 10U 5.585 0.917 -6.U98 5 (ly-lty) 1 (5y) U5 7 38 90 37.5 7.5 U5.0 90.0 7.5 -0.5 -7.0 0.0 1.500 0.033 1.089 x2=2.622 5 (ly-Uy) 1 (5y) 6 (6£) 53 9 U2.5 8.5 51.0 102.0 10.5 .5 -11.0 0.0 2 .59)4 0.029 2.373 x 2=U.996 2 (Xy—Uy) 1 3 (6w> 1*6 0.036 0 .06U 0 .0h8 x2=0 .lii.8 I1O.U15 8.083 U8.U98 96.996 UO 102 37 165". oo 1.25 -.75 -.50 0.00 (o-c)2 c us 9 U2 97 5 (1V-1*J l C5y) 7 6 ( 6„) 6 ( 6£) Ii9 U3.75 8.75 52.5 o-c 2 (ly-Uy) 1 u l) 3 (£) 31 15 50 “93 20 22 53 "55 32 16 1*8 T3 31.668 15.83U U7.502 95.0014 O.OOh -1 -1 2 0 -11.668 6.166 5.U98 - 0 . 00U 0.772 O.lOli 0.871 x2=1 .714? 0.031 O .063 0.083 x2=0.177 h.299 2.1*01 0.636 x2=7.336 * Plant US occurred in the isolation plot next to plant U7 later identified as a SXx Sh sh genotype, which, when crossed to a Sxx sh sh plant, would give a l!l:2 ratio very close to the observed "ratio. 2h An additional test of the interactions of the X-x and Sh-sh alleles wag obtained when the selfed progenies of the yellow anthered I5-1-7B2 plants were classified in the greenhouse during the winter of 1955-56. Sufficient plants were obtained from 15 of the 36 lots of selfed seed to index them for anther type. The results of this classification are given in Table 9. Three of the progenies had parents of the S XxShSh genotype; ten of them had parents of the S XxShsh genotype; and two progenies had parents (plants 12 and 6l) of the S Xxshsh genotype. The anther classification of plant 12 was 5y (yellow shrunken) in the first two readings, and 3 in the last two. (yellow "full”) Plant 61 had an anther classification of 5V in the V first three readings and li in the fourth. This evidence corroborates