wuqc'q THE GENETICS OF SEX EXPRESSION AND FRUIT SHAPE, STAMINATE FLOWER INDUCTION, AND f1 HYBRID FEASIBILITY OF A GYNOECIDUS M‘USKMELON Thesis for the Degree of Ph. D. MTCHTGAN STATE UNIVERSITY PHTLUP RAY ROWE 1959 HEIIS LIBRARY Michigan State University This is to certify that the thesis entitled THE GENETICS OF SEX EXPRESSION AND FRUIT SHAPE, STAMINATE FLOWER INDUCTION, AND F1 HYBRID FEASIBILITY OF A GYNOECIOUS MUSKMELON presented by Phillip Ray Rowe has been accepted towards fulfillment of the requirements for Ph.D. degree in Horticulture {K «301w Major professor Date May 9, 1969 0-169 t t . EILEEF PF? [If-fin LN. mLU..vI..ull .t A -.trE. III lull... lit... LP; ..r..t.v.?.r..tw .--...- -,..,.,\t.r.. . . .. : lawful. ..a¢.¢m.v..p. “Hume? . . t. .. x- - 00.2," I; as ...\I-h:...a.......z!t.uwm ABSTRACT THE GENETICS OF SEX EXPRESSION AND FRUIT SHAPE, STAMINATE FLOWER INDUCTION, AND F HYBRID FEASIBILITY OF A GYNOECIOUS MUSKMELON 1 By Phillip Ray Rowe Genetics of Sex Expression The gynoecious sex expression was controlled by modifying genes in addition to the major genotype, Azgg. The recombination of modifiers conditioning gynoecious sex expression, after crossing with monoecious, andro- monoecious and hermaphroditic sex types, resulted in mostly gynomonoecious plants for the £255 genotype in the segregating F2, BCPl and BCP2 populations. The gynomonoe- cious plants had 3/“, l/2 and l/A ovary perfect flowers and staminate flowers with rudimentary stigmas in com- binations or singly depending upon the modifier complement. Different modifiers were apparently responsible for the different partial ovary perfect flower types observed. Heterozygosity for the major gene A did not affect the expression of the modifiers, Reciprocal crosses showed that the maternal parent did not influence the gynoecious sex expression. Phillip Ray Rowe Inheritance of Fruit Shape The oblong fruit shape of the gynoecious line was dominant to the round fruit shape of several andromonoe- cious varietiese Single genes for round fruit shape dominant to the fruit shape of the gyneecious line were found in the andromonoecious Japanese hybrid, Sweetie, and the monoecious line, Morden Monoecious. Staminate Flower Induction Chemical spray treatments and environmental manipula— tions did not induce staminate flowersc Grafting of gynoecious scions onto various oucurbit rootstooks demon— strated differential induction. Pumpkin stocks resulted in the highest number of staminate flowers, The induced flowers were not true staminates, but were mostly staminate flowers with rudimentary stigmas, Different pumpkin stock sizes and scion sizes and ages did not affect the number of staminate flowers induced, Allevia- ) tion f the effe t of the pumpkin roots by using pumpkin ( (fl. interstocks demonstrated that the induction stimulus was from the foliage of the pumpkin, Staminate flowers were not induced on gyneecious cucumber by grafting onto pumpkin, which suggested that different stimuli are responsible for staminate flower induction in gynoecious sex types of these two different species of the genus Cucumis, Phillip Ray Rowe Elfiybrid Feasibility of the Gynoecious Muskmelon Gynoecious sex expression deserves considerable attention for use in the production of hybrid seed. A gynoecious line was developed with the more desirable round fruit shape which is necessary for production of round-fruited hybrids. Grafting onto pumpkin rootstocks induced adequate staminate flowers for increasing gynoecious seed. It is estimated that one hand pollina- tion to increase the gynoecious parent, with subsequent hybrid seed production by bee pollinations, would produce as much hybrid seed as 5,000 hand pollinations with the present method of making hybrids. THE GENETICS OF SEX EXPRESSION AND FRUIT SHAPE, STAMINATE FLOWER INDUCTION, AND Fl HYBRID FEASIBILITY OF A GYNOECIOUS MUSKMELON By Phillip Ray Rowe A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Horticulture 1969 l \i‘\ hgs\ ) m S; 'W T'\ ‘o r») ACKNOWLEDGMENTS The author would like to express great appreciation and give due credit to Dr. C. E. Peterson, who developed the gynoecious line that made this study possible. His foresight into the value of a gynoecious muskmelon, establishment of an inbred gynoecious line and helpful suggestions were vital contributions to this study. Appreciation is given to Dr. D. Markarian for his warm encouragement and guidance in the pursuit of this study. His complete unselfishness and concern for my training have added greatly to my education. To Dr. L. R. Baker, my thanks for his many helpful suggestions and support in the preparation of this thesis. As my major professor for the conclusion of this study, be enriched my graduate training. I wish to thank Dr. Louis Aung for the many sessions of rewarding consultation. Grateful appreciation is given to Dr. S. Honma for his many helpful suggestions in the preparation of the manuscript. ii TABLE OF CONTENTS ACKNOWLEDGMENTS . . . . . . . . . . . . LIST OF TABLES . . . . . . . . . . LIST OF FIGURES . . . . . . . . INTRODUCTION . . . . . . . . . . . . . LITERATURE REVIEW . . . . . . . . . . Superiority of Hybrids . . . . . . . Methods of Hybridizing . . . . . . . Sex Types in Muskmelons . . . . . . Fruit Shapes Associated with Sex Types . Chemical Effects on Sex Expression . . . Grafting Effects on Flowering and Sex Expres Environmental Effects on Sex Expression . Nitrogen Effects on Sex Expression . . . o o m o o o o o H O :3 MATERIALS AND METHODS . . . . . . . . Cultural and Planting Procedures . . . Genetic Study . . . . . . . . . . Parental Material and Crosses . . . . . . Progeny Tests of Different Gynomonoecious F2's Progeny Test of a Deviant Andromonoecious F2 . Progeny Test of a Gynomonoecious Segregate . Genetic Differences Between Hermaphroditic Strains . . . . . . . . . Effect of Selection Pressure on Gynoecious Segregates . . . . . . . . . Classification of Sex Expressions . . . . Fruit Shape Study . . . . . Staminate Flower Induction Studies Chemical Treatments . . . . . Environmental Fluctuations . . . . Graft Studies . . . . . . . . . . . Grafting Procedure . . . . Effects of Different Cucurbit Rootstockso . Effects of Stock Size and Scion Age and Size 9 O O O O O O 0 iii Page ii viii H H FJH UL) R.)l—’\.O\IO\UWJ:‘UU UL) Hta (JUL/U mtasaw o~qxnp 2O Origin of Stimulus for Inducing Staminate Flowers I I I I I I I I I I I o Gynoecious Cucumber Response to Pumpkin Rootstock . . . . . . . . . . . RESULTS AND DISCUSSION . . . . . . . . . Genetic Study . . . . . . . . . . . . Fl Populations o o o o o o o n o o 0 F2 Populations . . . . . . . . . . . Backcrosses to Gynoecious Parent and Cytoplasmic Differences . . . . . . . . . . . Backcrosses to Non-gynoecious Parents . . . Progeny Tests of Different Gynomonoecious F2's Progeny Test of a Deviant Andromonoecious F2 . Progeny Test of a Gynomonoecious Segregate . Use of Genotype, Aagg, as Hybrid Parent . . Effect of Selection Pressure on Gynoecious Segregates . . . . . . . . . Fruit Shape Study . . . . Staminate Flower Induction Studies Chemical Treatments . . . . Environmental Treatments . . Graft Studies . . . . . . Effects of Different Cucurbit Rootstocks . Field Grown vs. Greenhouse Grown Grafted Plants . . . . . . Effects of Stock Size and Scion Age and Size Origin and Description of Induction Stimulus . O C O O 0 Scheduling Staminate Flower Induction . . Gynoecious Cucumber Response to Pumpkin Rootstock . . . . . . . . . . CONCLUSIONS . . . . . . . . . . . . Genetics of Sex Expression . . . . . . Inheritance of Fruit Shape . . . . . . Staminate Flower Induction . . Fl Hybrid Feasibility of Gynoecious Muskmelon LIST OF REFERENCES . . . . . . . . . . iv 0 O 0 O 0 o O O Page Table LIST OF TABLES Frequency distribution of the F1 population from the cross, Gyn l X Morden Monoecious, for sex expression as influenced by the modifier genes . . . . . . . . . . . . . . Frequency distribution of the F1 population from the cross, Gyn l X Hale's Best, for sex expression as influenced by the modifier genes Frequency distribution of the F1 pOpulation from the cross, Gyn l X Polish Hermaphrodite, for sex expression as influenced by the modifier genes . . . . . . . . . . . Frequency distribution of the F population from the cross, Gyn l X Morden Honoecious, for sex expression as influenced by the modifier genes . . . . . . . . . . . . . . Frequency distribution of the F2 population from the cross, Gyn l X Hale's Best, for sex expression as influenced by the modifier genes Frequency distribution of the F2 population from the cross, Gyn l X Polish Hermaphrodite, for sex expression as influenced by the modifier genes . . . . . . . . . . . Frequency distribution of the backcross popula- tion from the cross, Gyn l X (Gyn l X Morden Monoecious F ), for sex expression as influenced by the modif er genes . . . . . . . . . Frequency distribution of the backcross popula- tion from the cross, Gyn l X (Gyn l X Hale's Best F1), for sex expression as influenced by the modifier genes . . . . . . . . . . Frequency distribution of the backcross popula- tion from the cross, (Hale's Best X Gyn 1 F1) X Gyn l, for sex expression as influenced by the modifier genes . . . . . . . . . . Page 3A 3A 3A 36 37 39 A0 A2 Table Page 10. Frequency distribution of the backcross population from the cross, Gyn l X (Gyn l X Polish Hermaphrodite F1), for sex expression as influenced by the modifier genes . . . . AA 11. Frequency distribution of the backcross pOpu- lation from the cross, (Gyn l X Morden Monoe- cious F ) X Morden Monoecious, for sex expres31on as influenced by the modifier genes A6 12. Frequency distribution of the backcross popula- tion from the cross, (Gyn l X Hale's Best Fl ) X Hale's Best, for sex expression as influenced by the modifier genes . . . . . . . . . A6 13. Frequency distribution of the backcross popula- tion from the cross, (Gyn l X Polish Hermaphro- dite F1) X Polish Hermaphrodite, for sex expression as influenced by the modifier genes A7 lA. Frequency distribution of the F3 population from a selfed F2 gynomonoecious plant with the phenotype, predominantly female with a few males with rudimentary stigmas, for sex expres- sion as influenced by the modifier genes . . A9 15. Frequency distribution of the F3 population from a selfed F2 gynomonoecious plant with the phenotype, predominantly female with a few 3/A, 1/2 and l/A ovary perfects and a few males with rudimentary stigmas, for sex expression as influenced by the modifier genes . . . . . 50 16. Frequency distribution of the F3 population from a selfed F gynomonoecious plant with the phenotype, predominantly female with a few 3/A ovary perfects, for sex expression as influenced by the modifier genes . . . . . . . . . 52 17. Frequency distribution of the F3 population from a selfed F andromonoecious deviant plant with the phenotype, andromonoecious with a few perfect flowers in axils of main runner, a few 1/2 ovary perfects and a few males with rudi- mentary stigmas, for sex expression as influenced by the modifier genes . . . . . 53 vi Table Page 18. Breeding behavior of two different hermaphroditic parents, as related to sex expression in the F1 generation, when used as pollen parents to increase the gynoecious line . . . . . . . 55 19. Effects of selection pressure and inbreeding on stabilization of gynoecious sex expression after one and two generations of selfing . . . . . 58 20. Inheritance of fruit shape in the family of Gyn l X RF Mon . . . . . . . . . . . . 6O 21. Inheritance of fruit shape in the family of Gyn l X Morden Monoecious, and the F1 fruit shape in the hybrid of Morden Monoecious with the oblong monoecious line, Monoecious Iroquois . . . . . 60 22. Comparative induction (to anthesis) of staminate flowers on Gyn l scions by different rootstocks in field I I I I I I I I I I I I I I 65 23. Effects of size of stock and size and age of scion on mean number of staminate flowers induced to anthesis in greenhouse . . . . . . . . . 68 2A. Effect of pumpkin rootstock vs. pumpkin inter— stock on mean number of staminate flowers induced, nodal extremities of induction, and mean node of induction of staminate flowers on gynoecious scions in greenhouse . . . . . . . . . . 70 vii Figure LIST OF FIGURES Page Monoecious plant with both pistillate and staminate flowers and gynoecious plant with pistillate flowers only . . . . . . . . 15 Illustration of the range of ovary sizes on the partial ovary perfect flowers that occur on both gynomonoecious and hermaphroditic plants I I I I I I I I I I I I I I 18 Staminate flower with rudimentary stigma (left) and normal staminate flower . . . . . . . 18 Grafting technique . . . . . . . . . . 27 Fruit shape of Morden Monoecious, the gynoe- cious line, the hybrid between the two lines, and typical fruit shape of hybrids between Gyn l and round—fruited andromonoecious varieties . . 62 viii INTRODUCTION Michigan has an average annual production of approxi— mately 3,000 acres of muskmelons (36). Hybrid varieties constitute an estimated 90% of this acreage. Hybrid seed is approximately $100.00 per pound as compared to $A.00 for open-pollinated varieties. This represents an invest— ment of $50.00 per acre in seed for the hybrid melon grower. Hybrid seed is eXpensive because of the laborious process involved in producing it. Most commercial musk— melons are andromonoecious and to be cross-pollinated must be emasculated, hand pollinated and trapped to prevent pollen contamination. Furthermore, Mann (31) reported that hand pollinations were only 60% as successful as bee pollinations, and the resultant fruit were smaller and con- tained fewer seeds. Whitaker and Pryor (59) made field hand pollinations and obtained only 2A-A0% fruit set. Efforts to improve fruit set have had limited success. Burrell and Whitaker (9) and Whitaker and Pryor (59) found that an increase in fruit set following hand pollinations could be obtained by application of one per cent indole- acetic acid or A-chlorophenoxyacetic acid to the ovary at the time of pollination, but fruit set by these methods 1 had fewer seed. Wolf and Hartman (62) increased fruit set of hand pollinations by pruning the fruit and plants. Jones (27) recently facilitated successful hand pollina— tions by applying benzyladenine to the ovaries at the time of pollination. Munger (3A) recognized the superiority of F hybrids l as compared to inbred lines in l9A2. He suggested a labor saving procedure of making hybrids in the afternoon by pollinating immediately after emasculation with male flowers picked in the morning and kept in a cool, moist place. The cost of hybrid muskmelon seed could be lowered if a suitable bee—pollinated seed parent could be developed. Use of the gynoecious sex character is limited by lack of sufficient genetic information, undesirable oblong fruit shape and absence of a chemical treatment that will induce staminate flowers for inbreeding and increasing seed. This study was conducted to determine the genetics of the gynoecious sex type, to ascertain the genetics of fruit shape, to develop a round-fruited gynoecious plant and to determine the effects of chemical treatments, environmental fluctuations and grafting on staminate flower induction. LITERATURE REVIEW Superiority of Hybrids Since Munger's (3A) observation that hybrid musk- melons were superior to varieties, much work has been done to substantiate the desirability of hybrids. The early workers Rosa (A6) and Scott (52) were unable to find hybrid vigor in crosses between inbred muskmelon lines. Munger (3A) reported a 30% increase in yield and greater uniformity in hybrids as compared to the parental strains. Bohn and Davis (3) obtained results that indicated apparent heterosis for earliness which could be controlled in F hybrids by l selecting parental lines for factors affecting fruit matur- ity. Foster (17) made an extensive study comparing selected inbred lines with their Fl hybrids. He reported as much as twice the yield from hybrids as compared to the higher yielding inbred parent. Evidence of heterosis was further substantiated by a group of hybrids with a common parent which outyielded the diverse commercial parents by 8A%. Quality factors such as net density, blossom end thickness, flesh firmness and cavity dryness tended to be improved in the hybrids as compared with the commercial parents. Fruit shape and soluble solids of the hybrids were inter- mediate between those of the parents. Methods of Hybridizing The wide acceptance of commercial hybrid muskmelons prompted investigations into methods of eliminating the tedious and expensive hand emasculations and pollinations. Bohn and Whitaker (6) and Bohn and Principe (A) discovered genetic male steriles as mutants in normally fertile lines. In both cases the sterility was controlled by a single recessive gene. These have the disadvantage of roguing one-half the seed parent plants in the field. Attempts to find linkages of the sterility genes with seedling markers, and thus facilitate thinning of fertile segregates prior to transplanting the male sterile plants to the field, have been unsuccessful (5). Foster (18) grew parental stocks in various patterns in the field to determine the best arrangement to obtain maximum natural cross pollination. Using a glabrous recessive marker (16) as the seed parent, he was able to distinguish the hybrid plants in the seedling stage. Although certain planting arrangements yielded a higher percentage of hybrids than others, none resulted in more than A0% hybrids. In later work (19), he was able to obtain 75% hybrids in open-pollinated pOpulations by using monoecious lines segregating 1 male sterile: 1 male fertile, and planted in close proximity to the pollen parent. Sex Types in Muskmelon The use of a genetically controlled gynoecious sex type as the seed parent has revolutionized cucumber hybrid seed production. Peterson (38) described a gynoecious cucumber line that made production of hybrid cucumbers economically feasible. The major sex types in muskmelon are distinctly different and simply inherited. Rosa (A7) reported that monoecious is dominant to andromonoecious by a single factor. Foster and Bond (20) reported an androecious mutant that is controlled by a single recessive gene. The mutation was not in the sex expression, but was for absence of lateral branches on which pistillate and per- fect flowers occur on monoecious and andromonoecious plants respectively. Poole and Grimball (A2) crossed a hermaphrodite from China with a monoecious, and from the F2 and backcross data proposed a two gene explanation of sex expression in muskmelon: Ag is monoecious, Ag is gynomonoecious, ag is andromonoecious, and ag is hermaphro- ditic. The gynomonoecious plants in Poole and Grimball's observations had mostly pistillate flowers with a few perfect flowers. They had some gynoecious plants in segregating populations, but hypothesized that the gynoecious sex type was not a true genetic type. The gynoecious plants were defined as transitory phenotypes caused by environmental influences on the gynomonoecious genotypes AAgg and Aagg. They concluded that there might be additional genes affecting sex expression, but that two major pairs of genes delineated the four primary sex classes. Peterson (39) proposed that gynoecious muskmelons I could be used in hybrid seed production. Kubicki (29) , a subsequently investigated the possibility of using ; gynoecious plants for seed parents in hybrids, and suggested that the gynoecious seed could be increased by bee— pollinating a field of isolated gynoecious plants with a hermaphroditic sister line. The resultant gynoecious Fl could then be used as the seed parent in commercial hybrid seed production. Both Peterson (39) and Kubicki (29) reported that crosses of gynoecious with andromonoecious and monoecious types yielded 100% monoecious hybrids. Fruit Shapes Associated with Sex Types Monoecious plants are better suited than andro- monoecious plants for making hand pollinations for hybrid seed production since monoecious plants do not require emasculation. Monoecious sex types have not been used as seed parents in hand-pollinated commercial hybrids because, with few exceptions, imperfect pistillate flowers produce oblong fruit. In the Midwest, oblong fruit are not as acceptable to the consumer as round fruit. Rosa (A7) observed this pattern of correlated sex-fruit shape inheritance and suggested that this phenomenon might not be the result of linkage, but of pleiotropic action of the gene determining sex expression. Kubicki (28) reported that pistillate flowers and fruits of monoecious varieties were elongated, whereas the hermaphroditic flowers and fruits of andromonoecious varieties were round. He hypothesized linkage of genes determining fruit shape and sex on the basis of an F2 ratio of three oblong-fruited, monoecious plants to one round-fruited, andromonoecious plant. Wall (57) suggested that fruit shape was regulated by a single gene with incomplete dominance plus minor modifying genes linked in coupling phase with the gene for determination of sex expression. In a related genus, Poole and Grimball (A3) have determined that in water- melons genes for plant sex habit and fruit shape were linked with a crossover value ranging from .1A to .35 in several segregating populations. Chemical Effects on Sex Expression The lack of staminate flowers on gynoecious plants creates a problem in obtaining true breeding lines by self pollinations. Wittwer and Bukovac (60) treated monoecious cucumbers with gibberellin A3 and increased the number of staminate flowers. Subsequently, Peterson and Anhder (A0) induced staminate flowers on gynoecious cucumbers by spraying with gibberellin A Pike and Peterson (Al) 3I sprayed gynoecious cucumbers with gibberellin AHA7 and induced staminate flowers with much lower concentrations than with A3. Peterson (39) and Kubicki (29) were unable to induce staminate flowers on gynoecious muskmelon by treating with gibberellin A3. Gibberellin has been used on other plants to alter sex expression. Shifriss (53) increased the female tendency of monoecious castor beans by spraying with gibberellin A This response is directly opposite that 3. of monoecious cucumbers. Auxins, growth retardants and kinins influence sex expression of many plants. Choudhury and Phatak (10) obtained pistillate flowers on earlier nodes of monoecious cucumbers by treating with maleic hydrazide, naphthaleneace— tic acid, indoleacetic acid and 2,A-dichlorophenoxyacetic acid. The production of staminate flowers on squash was inhibited by maleic hydrazide, and the ratio of staminate to pistillate flowers in cucumber and squash was reduced with naphthalene acetic acid and 2,3,5-triiodobenzoic acid by Wittwer and Hillyer (61). Abdel-Gawad and Ketellapper (l) inhibited appearance of pistillate flowers with N6-benzyladenine and suppressed staminate flowers with 2-chloroethyltrimethylammonium chloride on squash. A report by Prasad and Tyagi (AA) showed an increased number of pistillate flowers on bitter gourds sprayed with maleic hydrazide. Working with grapes, Negi and Olmo (35) changed male clusters to hermaphroditic by applying a synthetic kinin, SD8339. Andromonoecious muskmelons were treated with N,N-dimethylaminosuccinamic acid by Halevy and Rudich (23) with a resultant shift towards femaleness. A new area of research of chemical influence on sex expression is the effect of ethylene evolving chemicals. McMurray and Miller (32) reported a very pronounced change in the sex of monoecious cucumber plants sprayed with 2-chloroethanephosphonic acid. The treated plants had numerous continuous pistillate nodes and upon prolonged treatment approached the gynoecious sex type. Grafting Effects on Flowering and Sex Expression Grafting onto different rootstocks has been effective in changing flowering habits of certain scions. Zeevaart (63) showed that a flowering stimulus can be transferred from stock to scion by using long day and short day plants. Graft combinations between flowering (donor) and non-flowering (receptor) plants were performed and grown under non—inducing daylength for the receptors. He demonstrated that several plants which remain vegetative under certain light regimes can be easily flowered in non-inductive daylengths by grafting onto flowering Specimens of the same or related species. This suggested that one or more factors were transmitted from donors to receptors. Floral stimuli appeared identical in short day and long day plants because donors of one reaction 10 type could cause flower formation in receptors of the other only if they were induced themselves. Habermann and Wallace (22) showed that there was a flowering stimulus transferred from day neutral sunflower. stocks to scions. Scions flowered after very limited vegetative growth when grafted onto flowering plants as compared to when grafted on non-flowering plants. Curtis (12) and Curtis and Hornsey (l3) induced flowering on unvernalized scions of bolting resistant sugarbeets by grafting onto stocks that bolted readily. Graft induced sterility changes are less common than graft induced flowering responses. Frankel (21) obtained transmission of cytoplasmic male sterility into fertile petunia scions by grafting onto sterile stocks. Subse- quently, Corbett and Edwardson (ll) transmitted the sterility factor by grafting from petunia through stems of tobacco plants to petunia with no effect of the factor on the tobacco interstock. Negative results were obtained by Sand (A8) in attempts to transmit cytoplasmic male sterility in tobacco by grafting. He concluded that transmission of cytoplasmic male sterility through a graft union with fertile plants is not a general phenomenon. The induced sex reversal of staminate hop stocks by grafted pistillate scions reported by Limberk (30), prompted Mockaitis and Kivilaan (33) to graft gynoecious muskmelon onto andromonoecious muskmelon and monoecious ll pumpkin. They found that grafting on these rootstocks temporarily induced staminate flowers on the gynoecious scion. Environmental Effects on Sex Expression Sex expression of many plants is altered by environ- mental conditions. Schaffner (A9) reported that hemp plants of known sex produced flowers of the opposite sex in winter greenhouses, but had no sex reversion when grown outside in the summer. He (50, 51) subsequently found that Humulus, Plantago, Arisaema, Thalictrum, Myrica, and Morus also exhibited sex reversions when the photoperiod was changed. Tiedjens (56) observed an increase in the propor- tion of pistillate flowers in cucumber as a result of short day treatment. Nitsch, gg_g;. (37) showed that short days promoted pistillate flowers in gherkin and squash, while long days favored male expression. Conversely, Thompson (5A) promoted female sex expression in spinach with long days. Brantley and Warren (7) stated that long days increased the proportion of perfect flowers in andro- monoecious muskmelon. High temperatures shifted monoecious plants and some pistillate plants toward maleness in experiments with spinach by Janick and Stevenson (26). Bukovac and Wittwer (8) found low temperatures accelerated appearance of pistillate flowers in cucumbers. Heslop-Harrison (25) 12 stated that, in general, low temperatures promoted pistillate and inhibited staminate flowers. Nitrogen Effects on Sex Expression Nitrogen levels have been shown to modify sex expres- sion. Hall (2A), working with gherkin, and Tiedjens (56), using cucumbers, demonstrated that in these monoecious cucurbits high levels of nitrogen appeared to promote female sex expression, whereas low levels favored male expression. Tibeau (55) and Thompson (5A) found that hemp and spinach responded similarly to nitrogen levels. Brantley and Warren (7) reported that medium and high levels of nitrogen increased the total number of flowers and the proportion of perfect flowers in andromonoecious muskmelon. MATERIALS AND METHODS Cultural and Plantinngrocedures All the field plants were germinated in 2-l/A inch peat pots in the greenhouse and transplanted at approxi- mately the two true leaf stage. Black plastic mulch was used and Amiben at two pounds per acre was sprayed between rows of plastic prior to transplanting. Recommended rates of commercial fertilizer were disked under before laying the plastic. Insects were controlled with Dieldrin sprays and the plants remained relatively free of powdery mildew, so fungicides were not applied. To facilitate analysis of large populations at the proper stage of develOpment, the field plantings were made in increments beginning the first week of June. Single plants were hand planted approximately five feet apart in the row. Genetic Study Parental Material and Crosses A gynoecious line selected by Dr. C. E. Peterson was derived from an F2 gynoecious segregate of the cross monoecious X hermaphrodite. This line was used in crosses to determine the genetics of gynoecious sex expression. 13 1A An F single plant was selected and increased by rooting 7 cuttings and inducing staminate flowers for pollinations by grafting onto pumpkin. A progeny test of this seed (henceforth referred to as Gyn 1) showed that it was homozygous for the gynoecious character. 0f fifty-four field plants, only two plants produced a perfect (actually A 3/A ovary perfect) flower as the plants approached senes- . cence. Gynoecious plants characteristically produce a few perfect flowers during senescence. Figure 1 illustrates the gynoecious as compared with monoecious sex type. 5 Crosses were made in the greenhouse between Gyn l and monoecious, andromonoecious and hermaphroditic sex types to study the genetics of gynoecious sex expression. The variety Morden Monoecious (A5) was used as the monoecious parent, Hale's Best Jumbo as the andromonoecious parent, and a hermaphrodite from Poland (29) as the hermaphroditic parent. Fl's, F2's, BCl's and selected F3 grown in the field and classified for sex expression. populations were Progeny Tests of Different Gynomonoecious FQLE Gynomonoecious plants have mostly pistillate flowers and a few perfect flowers. The perfect flowers are not normal perfects, but differ by having various ovary sizes. These variable ovary sized perfect flowers are arbitrarily classified as: (l) 3/A ovary perfect, (2) 1/2 ovary perfect, (3) l/A ovary perfect and (A) male with rudimentary stigma. 15 Figure l. Monoecious plant (left) with both pistillate and staminate flowers and gynoecious plant with pistillate flowers only. l6 I l» S. n. .r» . 4 ~an ..I on. pm 3.. . . a. r. . . .r“ . . r“ .C :1. a a . . 2. . v ‘0 ». we uni l7 Hereafter, these partial ovary perfect flowers will be referred to by these designations. Gynomonoecious plants Inay exhibit these perfect flowers either singly or in any combination. Similar types of perfect flowers occur on 'the main runners of hermaphroditic plants. Figure 2 illustrates the different ovary sizes of perfect flowers (on gynomonoecious and hermaphroditic plants. Figure 3 shows the difference between staminate flowers with rudi- Inentary stigmas and normal staminate flowers. Partial ovary pistillate flowers have never been observed. Three F3 populations from F2 gynomonoecious plants with different partial ovary perfect flowers were grown to determine the segregation patterns and stability of the different forms. The following three F phenotypes were 2 selected from the cross of Gyn l X Hale's Best: (1) pre— dominantly female with a few males with rudimentary stigmas, <2) predominantly female with a few 3/A, 1/2 and l/A ovary perfects and a few males with rudimentary stigmas, and (3) predominantly female with a few 3/A ovary perfects. Progeny Test of a Deviant Andromonoecious F2 A few monoecious and andromonoecious segregates of the cross, Gyn 1 X Hale's Best F were found which 2’ possessed partial ovary perfects. An F2 andromonoecious plant which deviated from normal by having a few perfect flowers in the axils of main runners, a few 1/2 ovary 18 Figure 2. Illustration of the range of ovary sizes on the partial ovary perfect flowers that occur on both gynomonoecious and hermaphroditic plants. Figure 3. Staminate flower with rudimentary stigma (left) and normal staminate flower. l9 20 perfects and a few males with rudimentary stigmas was selected and self—pollinated to observe the F3 popula- tion. Progeny Test of a Gygomonoecious Segregate In a seed increase of 300 plants of Gyn l, a few plants showed perfect flowers before the plants began senescence. A plant with two l/A ovary perfects was self- pollinated to determine if there were genetic differences between this plant and its gynoecious sister plants. Genetic Differences Between Hermaphroditic Strains Hermaphroditic sex types have normal perfect flowers on laterals and partial ovary perfect flowers on main runners. The hermaphrodite from Poland produced only 3/A ovary perfects on the main runners. Whereas, hermaphro- dites segregated in this study from crosses between gynoecious and andromonoecious produced 1/2 and l/A ovary perfects and often staminate flowers with rudimentary stigmas, as well as 3/A ovary perfects on the main runners. Gyn l was crossed with the two different hermaphrodites to observe differences in the F1 populations. Effect of Selection Pressure on Gynoecious Segregates The effectiveness of selection pressure for recovering a homozygous gynoecious plant following an outcross to a 21 ciifferent sex type was tested. A gynoecious segregate A ‘was found in the progeny of a selfed monoecious plant derived from the cross (Gyn 1 X Delicious 51) X Delicious 51. Cuttings were made from this plant, self pollinations were made and the progenies were grown. From this popula- tion a gynoecious B segregate was selected, prOpagated and selfed. Progenies of gynoecious A and B were grown simultaneously and compared for relative gynoeciousness. Classification of Sex Expressions Because of the diversity of sex types in segregating populations, a classification sheet was devised to record the sex expression of individual plants. Sex readings were made when the main runners were from four to six feet long and prior to fruit set. Gynoecious segregates were left until the plant growth ceased to allow maximum time for perfect flowers to develop. All plants were pulled at the time of reading. Fruit Shgpe Study A serious objection to the gynoecious muskmelon is the dominant oblong fruit shape. Crosses were made between fifteen diverse commercial varieties and Gyn l in an attempt to find a round-fruited Fl hybrid. The andro- monoecious varieties used were: Bellgarde, DeConinck Bender, Delicious 51, Florida Honey Dew, Golden Perfection, Granite State, Hale's Best Jumbo, Harvest Queen, Hearts of 22 Gold, Iroquois, Nectarmelon, PMR A5, Seminole, Sungold Casaba, and the Japanese hybrid Sweetie. Ten plants of each were grown. In a collection of breeding lines, it was noticed that Morden Monoecious (A5) differed from other monoecious types by having round instead of oblong fruits. Fl's, backcrosses and F2 populations of crosses between Gyn l and Morden Monoecious were grown to determine the genetics of this round fruit character. Morden Monoecious was also crossed with Monoecious Iroquois, an oblong-fruited type developed by Dr. Henry Munger at Cornell University. Round and oblong fruit shapes of individual plants were recorded by visually observing two to four mature fruit per plant. Previous attempts to decipher fruit shape at anthesis were not as reliable as mature fruit readings. Staminate Flower Induction Studies Chemical Treatments The failure of gibberellin A3 to induce staminate flowers in gynoecious muskmelon (29, 39) prompted investi- gations to find a chemical that would induce staminate flowers. An array of chemicals including growth retardants, auxins, kinins, gibberellins, a morphactin, and an ethylene releasing growth regulator were sprayed on field seedlings at the two true leaf stage. Three applications were made at five day intervals. 23 The concentrations applied were those which had been shown to induce responses on other species of plants and twhich changed the normal growth pattern, but did not :severely stunt or kill the growing point of the plant. (Doncentrations were based on reports found in the litera- tnire for the various chemicals. Growth retardant sprays were: (1) N-dimethylamino- succinamic acid (B-nine), (2) 2-chlorethyltrimethylammonium (chloride (CCC), and (3) maleic hydrazide (MH). Each was 3 Sprayed at 10- molar. MH was dissolved in hot water, and IB-nine and CCC were readily soluble in water at room ‘temperature. The auxin treatments consisted of: (l) indoleacetic acid (IAA), (2) indolebutyric acid (IBA), (3) 2,A—dichloro- ;phenoxyacetic acid (2,A-D), (A) alpha-naphthaleneacetic acid (NAA), (5) beta-naphthoxyacetic acid (NOAA), and (6) .2,3,5-triiodobenzoic acid (TIBA). The 2,A—D and TIBA 5 A 'treatments were 10- molar, and the others were 5 X 10— Inolar. All of the chemicals with the exception of TIBA Ivere brought into solution with 50 ml methanol and diluted ‘to 500 ml with water. A commercial formulation of TIBA Etlready in solution was used. Kinetin and N6-benzyladenine (BA) were sprayed at 5 X 10-“ molar concentrations. Each was dissolved in 25 ml dilute hydrochloric acid and diluted to 500 ml with Water. A commercial solution of synthetic kinin, SD8339, 2A fPOHtShell Development Company was diluted with water and appflied at 5 X 10-Ll molar. A pure acid form of gibberellin AMA from Amdal 7, Chompany, was dissolved in 50 m1 methanol, diluted to 500 5 and 3 x lo'” molar. Inl with water, and applied at 3 X 10— Fkorphactin IT3A56, from Vero Beach Laboratories, Inc., was sirfilarly brought into solution and sprayed at 10—” molar. Tflie ethylene evolving compound, Ethrel, from Amchem Products, 3 Ikic., was diluted with water and applied at 10' molar. Distilled water was used to dilute all the chemicals t;o the proper concentrations. Tween 80 at .05% was added 218 a wetting agent to all the formulations and control :sprays. The two control treatments consisted of plants :sprayed with (l) 50 m1 methanol in A50 ml of water, and (2) 25 ml dilute hydrochloric acid in A75 ml water. The 4 "EI'EI’IJI—‘ul I Hwt—II—Immmwmt-JJ: "U’U'U'TJ’U’U'U'U’U’U’U '11'121'13I-4'1321'11 '12! '12] "13 J: (I) CD Self two of gynoecious segregate N I, x wHH mewwa w INI—‘I—‘I—‘I—‘HI—‘I—‘I—‘HUTHUQH totcrutcrdtotcwrororcwrutu *IJW'IJfiNwI-J'IZIH .t‘: (I) 59 The difficulty in establishing a homozygous gynoe- cious line would not prohibit its use in hybrid seed pro- duction. Perfect flowers on gynomonoecious plants usually occur on terminal ends of runners after the plants have set fruit. In view of the thousands of staminate flowers on a pollen parent, the few perfect flowers on gynomonoecious segregates probably would result in fewer inbreds in the hybrid seed than are now evident in hand- pollinated hybrids. Fruit Shape Study The F fruits of Gyn l crossed with fourteen round- 1 fruited andromonoecious varieties were all oblong. Apparently the genes for fruit shape in these varieties were similar. Three round—fruited recombinants were found among ten Fl plants of a cross between Gyn l and the andromonoecious Japanese hybrid, Sweetie. Open-pollinated seed from one of the round-fruited monoecious plants was saved. A genetic study was made of fruit shape using a homozygous monoecious round-fruited selection (hereafter referred to as RF Mon) from progeny of the open-pollinated seed. Crosses were made with Gyn 1 to determine the inheritance (Table 20) of this dominant round-fruited trait. The genetic data indicated the round fruit char- acter from Sweetie was controlled by a single dominant gene. A round-fruited gynoecious line has been developed using this dominant gene from Sweetie. 60 Table 20. Inheritance of fruit shape in the family of Gyn l X RF Mon. Observed Expected* Cross Round Oblong Round Oblong RF Mon 20 20 Gyn l 20 20 Gyn l X RF Mon Fl 20 20 Gyn l X RF Mon F2 AA 12 A2 1A Gyn 1 X RF Mon Fl BC to RF Mon 2A 2A Gyn l X RF Mon Fl BC to Gyn 1 A3 A5 AA AA N All P Values greater than 0.500. Table 21. Inheritance of fruit shape in the family of Gyn 1 X Morden Monoecious, and the F1 fruit shape in the hybrid of Morden Monoecious with the oblong monoecious line, Monoecious Iroquois. Observed Expected* Cross Round Oblong Round Oblong MM 15 15 Gyn 1 15 15 Gyn 1 x MM F1 A7 A7 Gyn 1 X MM F2 6A 20 63 21 Gyn 1 X MM Fl BC to MM AA AA Gyn l X MM Fl BC to Gyn 1 51 A9 50 50 Monoecious Iroquois X MMIF 32 32 l a All P Values greater than 0.750. 61 Table 21 shows the segregation behavior in crosses made with Gyn 1 to determine the inheritance of fruit shape in the round-fruited Morden Monoecious. The genetic factor for round fruit shape was attributed to a single dominant gene (Figure 5). The round-fruited hybrid of Morden Monoecious crossed with the oblong-fruited monoecious type, Monoecious Iroquois, indicated Morden Monoecious as a parent for developing round-fruited monoecious lines in breeding programs. There are reports of fruit shapes of oblate, spherical, oval, and elongate in the literature (58). Baines and Kang (2) reported that in.a cross of a cylindrical-fruited monoecious plant with a flat-fruited andromonoecious type, the F1 was intermediate. They postulated a single gene without dominance as the govern- ing factor. There are several alleles for fruit shape in muskmelons, and the dominant genes for round fruit from the two different sources in the present study might behave differently in crosses with other fruit shapes. These two sources of dominant round-fruited genes are of‘ value, however, in transferring the round-fruited character to the unacceptable oblong monoecious and gynoecious types. 62 Figure 5. -Fruit shape of Morden Monoecious, the gynoecious line, the hybrid between the two lines, and typical fruit shape of hybrids between Gyn l and round-fruited andromonoe— cious varieties. 63 Marcie n GynXMordosMa-fi MonoeLt 0 (LS EPIC—Q, 67.x Audrmmnnhfi GynoeCIOuS 6A Staminate Flower Induction Studies Chemical Treatments None of the chemicals used induced staminate flowers on Gyn 1. Various growth responses were induced by certain of the chemicals, but none altered the gynoecious sex expression. Environmental Treatments Shock treatments of cold, heat and daylength were also ineffective in the induction of staminate flowers on Gyn 1. These results suggested the gynoecious plants would remain gynoecious if exposed to adverse conditions naturally. This stability is necessary for successful use in hybrid seed production. Graft Studies Effects of different cucurbit rootstocks.-—The grafts of Gyn l scions onto different cucurbits substantiated differences between rootstocks for staminate flower induc— tion. The induced pollen bearing flowers were not normal staminate flowers, but were mostly staminates with rudi- mentary stigmas. These flowers are referred to as staminates as they are bearers of viable pollen equivalent to normal staminate flowers. Table 22 shows the response of Gyn l on the various rootstocks in the field. The pumpkin rootstock was sig- nificantly better than all the other rootstocks with the 65 Table 22. Comparative induction (to anthesis) of staminate flowers on Gyn l scions by different rootstocks in field. 4 Mean No. Rootstock Staminate Flowers* Cucurbita pgpp. 3.6 a Cucurbita moschata 2.A ab Cucumis sativus (monoecious) 1.7 bc Cucumis sativus (gynoecious) 1.6 bc Cucumis melo (hermaphrodite) 1.6 bc Cucumis melo (andromonoecious) 1.5 bc Cucurbita ficifolia 1.3 bc Citrullus vulgaris 0.3 c Gyn l (inverted graft) 0 c Gyn 1 (graft) 0 c Gyn l (ungrafted) 0 c * Means followed by different letters are signifi- cantly different at 5% level. Duncan's Multiple Range Test. 6'6 exception of squash. The absence of staminate flowers on the inverted graft of Gyn 1 demonstrated that the girdling effect alone did not affect staminate flower induction. The few staminate flowers (0.3-3.6) produced indicated that field-grown grafted plants would not produce enough pollen for increase of gynoecious seed by bee pollinations. The lack of sufficient staminate flowers for adequate pollina- tion in the field might be due to abortion of early flowers before anthesis. Abortion is accentuated by wind, less than optimum growing conditions and transplanting shock. Zubov (6A) reported various changes in muskmelon scions grafted onto pumpkin including increased cold and anthracnose resistance, doubled average length of vines and fruit weight, and taste variations depending upon the variety of pumpkin stock. In the present study, there were no visual changes in the gynoecious scions on dif— ferent rootstocks. Gynoecious scions on different root- stocks resembled the Gyn 1/Gyn l and Gyn 1 control plants in all growth and fruit characteristics. Field grown vs. greenhouse grown grafted plants.-— An alternative to growing the grafted plants in the field would be to grow them in the greenhouse and hand pollinate field plants with the greenhouse pollen. By using forceps and detaching each individual anther, approximately four pistillate flowers can be pollinated with one staminate flower. This method of increasing 67 gynoecious seed is not prohibitive compared to hand pollina- tions for hybrid seed production. For each successful pollination of the gynoecious plants with the greenhouse grown pollen, the return would vary from 200 to 500 seeds per fruit. The hybrid seed that could be produced with these gynoecious plants from one hand pollination would approach exponential proportions, assuming a realistic 1000 seeds per plant from bee cross—pollinations. It is estimated that one hand pollination to increase the gynoe- cious parent is equivalent to 5000 hand pollinations for hybrid seed production. The grafting procedure is not limiting, as the grafts are easily and quickly made, and a minimum of 90% scion survival is common. Effects of stock size and scion age and size.-—Since chemical treatments and field grafting did not produce suf- ficient staminate flowers, the most efficient stock size and scion age and size combinations for staminate flower induction in the greenhouse were determined. Table 23 shows the number of staminate flowers induced to anthesis with the different graft combinations. Induc- tion of staminate flowers was not increased by either larger stocks or by the scion size. Seedling scions (S) did not differ from scions taken from laterals (L) on older plants. The check plants, ungrafted Gyn l and Gyn l/Gyn l, were gynoecious with only two of the plants having one perfect flower each as the plants approached 68 !.~.I -Irrfhli .n. .... . .-. ilk :.pcmad wofiuosoam Eonm Hmpopwa u A .wnwaooom u m ncoaooepooeH peeoaeacwan oz .m.z H.mm u q .m.z m.sm u m k .m.z H.mm n mo>moa m .m.z 0.2m u mama H .m.z m.mm u U59 M saw "a T? um m .m.z w.mm :.om m.mm o.mm m.mH 0-0m o.mH m.mm m.mm w.ma mama m HAN ta mém um m .m.z m.:m m.Hm m.om s.mm o.mm o.Hm o.mm m.mH m.HH o.mm mama m m.mm no Tam um M .m.z m.mm m.mm m.:H 5.0m mama o.mm o.mH m.mm m.mm w.mm mama : mom u a 5mm n m M .m.z m.mm m.Hm o.mH m.:m o.mm m.om >.mm m.mm o.mm s.mm mama m .8; M M m .a. M m m M m m mo>moa m mood H U32 omfim xoopm cow< one oufim Gonom .omsoocoopm CH memoopow on ooosozfi mpozoam opmcasmpm mo gonad: some do coaom mo own now omen vow xoOpm no oNHm mo mpoommm .mm canoe p92. 69 senescence. The most suitable stock is one with two true leaves. It is more succulent for grafting and does not require support as do older stocks. Origin and description of induction stimulus.—-An experiment was conducted to learn if the stimulus for staminate flower induction originated from the root or the foliage of the pumpkin stock. Data were taken to determine the duration of induction. Table 2A shows that the stimulus for staminate flower induction came from the foliage of the pumpkin rather than from the roots. The effect of pumpkin roots was easily circumvented by using pumpkin interstocks. There were no differences between mean number of staminate flowers induced, nodal extremities of induction, and mean node of induction between scions on the pumpkin rootstock and the pumpkin interstock. Schedulinggstaminate flower induction.-—In using greenhouse induced staminate flowers for field pollinations to increase gynoecious seed, scheduling the field plants is important. The field plants should be ready for polli- nation when the staminate flowers are at anthesis since the induction stimulus is temporary and is terminated after the twentieth node of the scion. This scheduling could be accomplished by planting the pumpkin rootstocks two to three weeks after the gynoecious plants are planted. Scions from the field plants could be used for grafting. 7O .m.z mm.m H m.oa om.m H 0.0H coaposocfl mo woos cmoz .m.Z mHlm ONIO COHPOSUCH MO mmeHEmfipxm HGUOC Gmmz .m.z m.sm :.Hm poosUCH whosoam opmoaempm no access new: owes a H eso\efixoead\a cso saxoeodxa cam .omsoncoonw ca macaom a saw so mnozoam opwcHEmpm mo coaposoofi so one: some one «coaposocfi mo moHpHEmeuxo Havoc .ooosoca mnozoam opmoHEmpm mo Lopes: some do xoOpmmopofi CHXQESQ .m> xOOpmpoon cHxQESQ mo pommmm .zm canes 71 It is evident from the data that most induced staminate flowers are produced on laterals rather than the main runner. Therefore, laterals on the scion should be encouraged and not pruned as is practiced in growing muskmelons in the greenhouse for seed production. gynoecious cucumber response to pumpkin rootstock.-- Gynoecious cucumber scions did not respond to the pumpkin rootstock. Replicated trials using Gyn l scions and gynoecious cucumber scions on pumpkin stocks resulted in numerous staminate flowers on Gyn l, but no staminate flowers on the gynoecious cucumbers. It is postulated that different stimuli are responsible for staminate flower induction in these two members of thesame genus. This evidence supports the supposition that different stimuli are involved since gibberellin induces staminate flowers readily on gynoecious cucumbers, but has no induc- tion effect on gynoecious-muskmelons. CONCLUSIONS Genetics of Sex Expression Gynoecious sex expression was controlled by multiple modifying genes in addition to the major genotype Azgg. Disruption of the combination of modifiers conditioning gynoecious sex expression by crossing with monoecious, andromonoecious, and hermaphroditic sex types resulted in mostly gynomonoecious plants for the Azgg genotype in the segregating F2, BCPl, and BCP2 populations. The different major sex types had different complements of the modifying genes. The segregation behavior was similar whether the gynoecious type was the seed parent or the pollen parent, and illustrated lack of cytoplasmic differences for gynoecious sex expression. Gynomonoecious plants had 3/A, 1/2 and l/A ovary perfect flowers and staminates with rudimentary stigmas in combinations or singly depending upon the modifier complement. Apparently, different modifiers were responsible for the different types of partial ovary perfect flowers. The highest percentage of gynoecious segregates was recovered from selfed gyno— monoecious plants that had only 3/A ovary perfects. Heterozygosity for the major genes A_and 9 did not affect the expression of the modifiers. Monoecious and 72 73 andromonoecious plants infrequently exhibited the partial ovary perfect flowers in segregating populations, after crossing these two sex types with the gynoecious type. The percentage of gynoecious progeny in crosses of gynoecious, AAgg, with the double recessive hermaphrodite, gggg, varied depending upon the complement of modifier genes of the hermaphrodite. A higher percentage of gynoe- cious progeny was present in theiFl if the hermaphroditic parent had only 3/A ovary perfects. Recovery of a homozygous gynoecious type after the combination of modi- fiers has been changed by crossing to different sex types apparently requires several generations of selfing. Inheritance of Fruit Shape The oblong fruit shape of Gyn l was dominant to the fruit shape of several round—fruited andromonoecious varieties. A gene for round fruit dominant to the fruit shape of Gyn 1 was found in the andromonoecious Japanese hybrid, Sweetie. The round fruit shape of Morden Monoecious was also dominant to the fruit shape of Gyn 1. Genetic studies of these two factors for round fruit shape showed that they were single dominant genes. Staminate Flower Induction Chemical spray treatments with auxins, growth retardants, kinins, gibberellins, a morphactin, and an ethylene releasing growth regulator were ineffective in 7A inducing staminate flowers on Gyn 1. Environmental treat- ments of cold, heat and daylength were equally ineffective. Grafting onto different cucurbit rootstocks demon- strated differential induction. Pumpkin rootstocks resulted in the highest number of staminate flowers on gynoecious scions. The induced flowers were not true staminates, but were mostly staminate flowers with rudi- mentary stigmas. The amount of pollen on the graft- induced flowers was equivalent to normal staminate flowers. No effects other than induction of staminate flowers were observed as responses to the different rootstocks. There were no differences in the numbers of staminate flowers induced due to different pumpkin stock sizes. Scion size did not affect induction, and scions from mature gynoecious plants did not differ from seedling scions. The girdling effect alone of grafting did not influence staminate flower induction. Pumpkin stocks did not differ from pumpkin inter- stocks in either the number of staminate flowers induced, or the nodal extremities of induction, or the mean node of induction. The stimulus from the pumpkin stock is apparently endogenous to the foliage of the pumpkin. Pumpkin stocks were not effective in inducting stami- nate flowers on gynoecious sucumbers. It is postulated that different stimuli are responsible for staminate flower induction in gynoecious muskmelons and cucumbers. 75 El Hybrid Feasibility of Gynoecious Muskmelon Gynoecious sex expression in muskmelon deserves con- siderable attention for use in the production of hybrid seed. A gynoecious line was developed with the more desirable round fruit shape which is necessary for produc- tion of round-fruited hybrids. Two different methods of increasing seed of gynoecious lines may be used. Hand pollinations of gynoecious plants in the field, with pollen induced by grafting in the greenhouse, appears to be more satisfactory than sibbing with a hermaphroditic sister line. Hand pollinations with induced staminate flowers are facilitated by ease of making grafts and the high percentage of scion survival. It is estimated that one hand pollination to increase the gynoecious parent, with subsequent hybrid seed production by bee pollinations, would produce as much hybrid seed as 5,000 hand pollinations with the present method of making hybrids. The difficulty in obtaining the homozygous gynoe— cious type is not prohibitive. Gynomonoecious segregates in the seed parent would contribute a minimum of pollen for contamination, as the perfect flowers usually occur after the plants are of sufficient size to have set fruit. The per cent inbreds in the hybrid seed from gynomonoe- cious pollen probably would not surpass the per cent in- breds observed in hand-pollinated commercial hybrids. LIST OF REFERENCES 10. LIST OF REFERENCES Abdel-Gawad, H..A. and H. J. Ketellapper. 1968. The effect of 2-chloroethyl trimethyl ammonium chloride (CCC) and N6-benzyladenine on growth and flOWering of squash plants. Abstract No. 98, 65th Ann. Meeting Amer. Soc. Hort. Sci., Davis, Calif. Bains, M. S. and U. S. Kang. 1963. Inheritance of some flower and fruit characters in muskmelon. Indian Jour. Genetics 23:101-106. Bohn, G. W. and G. N. Davis. 1957. Earliness in F hybrid muskmelons and their parent varieties. Hilgardia 26:A53-A7l. 1 and J. A. Principe. 196A. A second male sterility gene in the muskmelon. Jour. Heredity 55:211—215. 1968. Independent assortment of young plant characters in muskmelon, Cucumis melo L. Abstract No. A8, 65th Ann. Meeting Amer. Soc. Hort. Sci., Davis, Calif. and T. W. Whitaker. 19A9. A gene for male sterility in muskmelon (Cucumis meloL.).. Proc. Amer. Soc. Hort. Sci. 53:309-31A. Brantley, B. 81 and G. F. Warren. 1960. Sex expres- sion and growth in muskmelon. Plant Physiol. 35: Bukovac, M. J. and S. H. Wittwer. 1961. Gibberellin modification of flower sex expression in Cucumis- sativus L. Adv. Chem. Ser. 28:80-88. Burrell, P. C. and T. W. Whitaker. 1939. The effect of indol-acetic acid on fruit setting in muskmelons. Proc. Amer. Soc. Hort. Sci. 37:829-830. Choudhury, B. and S. C. Phatak. 1959. Sex expression and sex ratio in cucumber (Cucumis sativus L.). Indian Jour. Hort. 16:16A-l69. 77 11. 12. 13. 1A. 15. 16. 17. l8. 19. 20. 21. 22. 23. 78 Corbett, M. K. and J. R. Edwardson. 196A. Inter- graft transmission of cytoplasmic male sterility. Nature 201:8A7-8A8. Curtis, G. J. 196A. Graft transmission of the flowering stimulus from a wild Beta species to a line of beet selected for resisfance to bolting. Nature 203:201—202. Curtis, G. J., K. G. Hornsey, and-G. K. G. Campbell. 196A. Graft transmissible induction of elongation and flowering in scions of sugar beet bred for resis- tance to bolting. Nature 202:1238. Denna, D. W. 1962. A simple grafting technique for cucurbits. Proc. Amer. Soc. Hort. Sci. 81:369-370. de Stigter, H. C. M. 1956. Studies on the nature of the incompatibility in a cucurbitaceous graft. Mededel. Landbouwhogesch Wageningen 56:1-51. Foster, R. E. 1963. Glabrous, a new seedling marker in muskmelon. Jour. Heredity 5A:ll3-115. . 1967: F hybrid muskmelons, I. Superior performance of selected hybrids. Proc. Amer. Soc. Hort. Sci. 91:390-395. . 1968. F1 hybrid muskmelons, III. Field production of hybrid seed. Proc. Amer. Soc. Hort. Sci. 92:A61-A6A. . 1968. F1 hybrid muskmelons, V. Monoecism and male sterility in commercial seed production. Jour. Heredity 59:205-207. . and W. T. Bond. 1967. Abrachiate, an. androecious mutant muskmelon. Jour. Heredity 58: 1.3-1’4 o Frankel, R. 1956. Graft-induced transmission to progeny of cytoplasmic male sterility in petunia. Science l2A:68A-685. Habermann, H. M. and R. H. Wallace. 1958. Transfer of flowering stimulus from stock to scion in grafted Helianthus annuus L. Amer. Jour. Bot. A5:A79-A82. Halevy, A. H. and Y. Rudich. 1967. Modification of sex expression in muskmelon with growth retardant B-995. Physiol. Plantarum 20:1052-1058. 2A. 25. ... 27. 28. 29. 30. 31. 32. 33. 3A. 35. 79 Hall, W. C. 19A9. Effects of photOperiod and nitrogen supply in growth and reproduction in the gherkin. Plant Physiol. 2A:753—769. Heslop-Harrison, J. 1957. Experimental modification of sex expression in flowering plants. Biol. Revs. 32:38-90. Janick, J. and E. C. Stevenson. 1955. Environmental influences on sex expression in monoecious lines of spinach. Proc. Amer. Soc. Hort. Sci. 65:Al6-A22. Jones, C. M. 1965. Effects of benzyladenine on fruit set in muskmelon. Proc. Amer. Soc. Hort. Sci.” 87:335-3A0. Kubicki, B. 1962. Inheritance of some characters in muskmelons (Cucumis melo L.). Genetica Polonica 3:265-27A. . 1966. Genetic basis for obtaining gynoecious muskmelon lines and the possibility of their use for hybrid seed production. Genetica Polonica 7:27-29. Limberk J. 195A. Zmena pohlavi u chmele (Humulus lupulus) [The change of sex in hops (Humulus lupulus)]. Ceskoslovenska Biologie 3:2A3-2A6. Mann, L. K. 1953. Honeybee activity in relation to pollination and fruit set in the cantaloups (Cucumis melo). Amer. Jour. Bot. A0:5A5-553. McMurray, A. L., and C. H. Miller. 1968. Cucumber sex expression modified by 2-chloroethanephosphonic acid. Science 162:1397-1398. Mockaitis, J. M. and A. Kivilaan. ‘196A. Graft- induced sex changes in Cucumis melo L. Nature 202:216. Munger, H. M. 19A2. The possible utilization of first generation muskmelon hybrids and an improved method of hybridization. Proc. Amer. Soc. Hort. Sci. A0:A05-A10. Negi, S. S., and H. P. Olmo. 1966. Sex conversion in a male Vitis vinifers L. by a kinin. Science 152:162A. 36. 37. 38. 39. A0. A1. A2. A3. AA. A5. A6. A7. A8. A9. 80 Nicklow, C. and B. Woolsey. 1967. Summary statistics of fresh market and processing vegetables in the U.S. 196A and 1966. Dept. of Hort., Mich. State Univ. Nitsch, J. P., E. B. Kurtz, J. L. Liverman, and F. W. Went. 1952. The develOpment of sex expression in cucurbit flowers. Amer. Jour. Bot. 39:32-A3. Peterson, C. E. 1960. A gynoecious.inbred line of cucumber. Quar. Bul. Mich. Agri. Exp. Sta. A3:A0-A2. . 1963. Gynoecious muskmelons for hybrid seed production. Abstract No. 332, 60th Ann. Meeting Amer. Soc. Hort. Sci., Amherst, Mass. , and L. D. Anhder. 1960. Induction of stami- nate flowers on gynoecious cucumbers with gibberellin A3. Science 131:1673-167A. Pike, L. M. and C. E. Peterson. 1968. Gibberellin Au/A7 for induction of staminate flowers on the gynoe- cious cucumber. Abstract No. 95, 65th Ann. Meeting Amer. Soc. Hort Sci., Davis, Calif. Poole, C. F. and P. C. Grimball. 1939. Inheritance of new sex forms in Cucumis melo L. Jour. Heredity . l9A5. Interaction of sex, shape and weight genes in watermelon. Jour. Agri. Res. 71:533-552. Prasad, A. and I. D. Tyagi. 1963. Effect of maleic hydrazide on the sex expression and sex ratio in.bitter gourd (Momordica charantia L.) Sci. Cult. 29:605-606. Progress Report of the EXperimental Farm, Morden, Manitoba 1955-1959 (1961): p. A3. Cited from Plant Breed. Abstracts, 1962. 32:528. Art. No. 29A9. Rosa, J. T. 1927. Results of inbreeding melons. Proc. Amer. Soc. Hort. Sci. 2A:79-8A. . 1928. The inheritance of flower types in Cucumis and Citrullus. Hilgardia 3:235-250. Sand, S. A. 1960. Autonomy of cytoplasmic male sterility in grafted scions of tobacco. Science 131: 665. Schaffner, J. H. 1921. Influence of environment on sexual expression in hemp. Bot. Gaz. 71:197-219. 50. 51. 52. 53. 5A. 55. 56. 57. 58. 59. 60. 61. 62. 63. 8l . 1923. Sex reversal in the Japanese hop. Bull. Torrey Bot. 50:73-79. . 1925. Experiments with various plants to produce change of sex in the individual. Bull. Torrey Bot. 52:35-A7. Scott, G. W. 1933. Inbreeding studies with Cucumis melo. Proc. Amer. Soc. Hort. Sci. 29:A85. Shifriss, 0. 1961. Gibberellin as sex regulator in Ricinus communis. Science 133 2061-2062. I Thompson, A. E. 1955. Methods of producing first *3 generation hybrid seed in spinaCh. Cornell Agr. Exp. ‘ Sta. Mem. 336:1-A8. Tibeau, M. E. 1936. Time factor in utilization of minerals by hemp. Plant Physiol. 11:731-7A7. Tiedjens, V. A. 1928. Sex ratios in cucumber flowers as affected by different conditions of soil and light. Jour. Agr. Res. 36:721-7A6. Wall, J. R. 1967. Correlated inheritance of sex expresgion and fruit shape in Cucumis. Euphytica 16: 199-20 . Whitaker, T. W. and G. N. Davis. 1962. Cucurbits. Interscience Publishers, Inc., New York. and D. E. Pryor. 19A6. Effect of plant growth reguIators on the set of fruit from hand pollinated flowers in Cucumis melo L. Proc. Amer. Soc. Hort. Sci. A8:Al7-A22. Wittwer, S. H. and M. J. Bukovac. 1958. The effects of gibberellin on economic crops. Econ. Bot. 12:213- 255. and I. G. Hillyer. 195A. Chemical induction of male sterility in cucurbits. Science 120:893-89A. Wolf, E. A. and J. D. Hartman. 19A2. Plant and fruit pruning as a means of increasing fruit set in musk- melon breeding. Proc. Amer. Soc. Hort. Sci. A0:A15- A20. Zeevaart, J. A. D. 1958. Flower formation as studied by grafting. Mededel. Landbouwhogesch Wageningen 58:1-88. 6A. 82 Zubov, A. A. 1960. Izmenenie dyni pod vliyaniem tykvy-mentora. [Changes in the muskmelon under the influence of a pumpkin stock.] Piull. Nauch. Inform. Tsentral. Genet. Lab. Im. I.V. Michurina 9/10z55-63. N "'IIIIIIIIIIIIJILIIIIIIIGIIIIIIIIIIIIIII“