INHERITANCE OF PROLIFICACY AND MATURITY IN CROSSES OF SOUTHERN x NORTHERN MAIZE GEPM PLASM By Farrell M. Bagehaw AN ABSTRACT Submitted to Michigan State University in partial fulfillment of the requirements for the degree or DOCTOR OF PHILOSOPHY Department of Farm Crops 1963 74 Approved 2?.Jg:fflfiay444“\a~“t ABSTRACT Four northern and three southern inbred lines of maize were used to study the inheritance of ear number and days to silking. Individual plant data were used to obtain estimates of generation means and variances. Gene numbers, dominance re- lationships and heritability values were calculated. Gene numbers ranged from one to three for ear number. Dominance relationships varied from complete dominance for genes controlling the two-cared characteristic through no dominance to complete dominance for one ear. At several locations previous research with germ plasm adapted to the area had indicated that the two-eared char- acteristic was recessive in nature. This study indicated that the dominance relationships varied with the particular parents used. No generalized statement regarding the domi- nance of the one-eared condition is valid as shown by the lack of dominance in the cross W64A x T115 which represented the least and the most prolific inbreds used in this study. Epistatic effects varied from no epistasis to epistasis for one ear. Heritability values varied from 0.0 to 1.0. averaging 0.439. Previously reported heritability values for ear number have been somewhat lower, perhaps due to the use of pOpulations less genetically diverse. The expression of car number was not significantly af- fected by the competition or lack of competition from adja- cent rows. Fewer rows of each generation per replication and an increased number of replications and locations were recommended for future research. Lack of dominance or partial dominance for fewer days to silking was observed. The number of genes controlling days to silking varied from one to 23. Heritability esti- mates averaged 0.476 and ranged from 0.0 to 1.0. Epistasis for early silking was indicated in some crosses. Independent gene systems for ear number and days to silking were indicated by low, noncsignificant genetic and simple correlations. In certain of the crosses studied, simultaneous selection for early-silking plants bearing two or more cars should lead to the isolation of early-silking, two-cared inbred lines for use in the northern Corn Belt. INHERITANCE OF PROLIFICACY AND MATURITY IN CROSSES OF SOUTHERN x NORTHERN MAIZE GERM FLASH B! Farrell H. Bagshaw A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Farm Crops 1963 INHERITANCE OF PROLIF'ICACY AND MATURITY IN CROSSES OF SOUTHERN x NORTHERN MAIZE GERM FLASH ACKNOWLEDGMENT The author wishes to express his gratitude for the guidance and inspiration of Dr. E. C. Rossman under whose supervision this study was conducted. The author is grateful to Dr. Fred C. Elliott and Dr. M. W. Adams for their suggestions and constructive criticisms. To my wife, Dottie, goes sincere thanks for her assistance and encouragement. TABLE OF CONTENTS Page 0 ha I. IntTOdUCtion . o e e e e e e o e e e e e e e e e 0 Cl II. Review of Literature . . . . . . . . . . . . . . A. Ear NUIDbel‘ e e e e e e O 0 e 0 e o 0 Morphology, Histology and Reproduction . Geography. 0 e e e o e e e e e e 0 Environmental modification of Morphology Effect of Increased Plant Population . Effect Of Shading e e e e e e e e 0 Correlation of Yield . . . . . . . . Inheritance of Ear Number. . . . . . Gene Number and Degree of Dominance. Heritability Estimates . . . . . . . 0000000000 GDGHD ‘Q~eonmcnmuausouu O O O O O 0 O I O O 0 B. Maturity o e e e e e e e e e 0 Environmental and Genetic Factors. . Inheritance. o e e e e e e e e e e e O O O O C. Biometrical Considerations . . . Genotype x Environment Interaction Tests of Scale - Transformations . Partitioning of Variance . . . . . e... 0... e000 0 H (0 III. Materials and Methods. . . . . . . . . . . . . . .15 IV. Experimental RBBUltB e e e e e e e e e e e e e e e21 Ear number 0 e e o o o o 0 O O O O 0 0 e O .21 Dominance Relationships. . . . . . . . . . . .27 Gene Number and Heritability . . . . . . . . .30 correlationsgeeeeoooeeoeeeeO:33 Date of Silking. . . . . . . . . . . . . . .55 Dominance Relationships. . . . . .59 Gene Number and Heritability . . . I I I : I .39 V. Methods of Estimating Generation Variances . . . .42 Expected Genetic Gain. . . . . . . . . . . . . .47 VI. Discussion . . . . . . . . . . . . . . . . . . . .50 VII. Summary. . . . . . . . . . . . . . . . . . . . . .60 VIII. Literatdre Cited 0 . o g Q . o Q Q Q Q Q o o o O 063 INHERITANCE OF PROLIFICACY AND MATURITY IN CROSSES OF SOUTHERN x NORTHERN MAIZE GERM PLASM INTRODUCTION Recent interest in the development of hybrids which consistently produce two-cared plants has resulted from attempts to increase yields with one-eared hybrids at high- er plant pepulations. Some hybrids in current use fail to respond in yield with increases in plant population due to the increased incidence of barren plants. Hybrids differ in their ability to resist barrenness, with those having one or more parents with the two-eared tendency showing some resistance to barrenness under stress. Few inbreds or hy- brids adapted to the northern Corn Belt exhibit the two- eared characteristic. During much of corn breeding in the Corn Belt, se- lection has been against prolificacy and in favor of single- eared varieties. This selection against prolifics dates to the days of hand harvest and continued for some time with the advent of mechanical harvest. In the South, personal preferences and labor economics have not resulted in nearly as much selection against prolifics._ Therefore, there are more good sources of prolific characteristics in southern inbreds, hybrids, and open-pollinated varieties. This study was conducted to obtain information on inheritance of the two-cared characteristic and maturity in crosses of early maturing single-eared inbred lines from the northern Corn Belt with late-maturing prolific inbreds from the South. REVIEW OF LITERATURE A. EAR NUMBER Morphology, Histology, and Reproduction Normal corn plants possess an ear bud in the axil of each leaf below the apical or main ear bud. During the de- velopment of the plant the ear primordia are formed in the exile of the new leaves until the tassel starts to differ- entiate (Hershey, 1954; Bonnett, 1948). At this stage no additional primordia are formed and the last one laid down usually becomes the main ear (Hershey, 1934). The lower or secondary buds may undergo considerable development or may disintegrate. In Corn Belt varieties seldom do more than the first and second ear buds extend beyond the leaf sheath (Kiesselbaoh, 1949). Inselberg (1956) concluded from a morphological study of ear shoot development in the Corn Belt single crosses (VFQ x 0103) and (L317 x R4) that the potential number of ears per plant was uniform but that the ear buds which de- veloped into mature ears began to grow at an increased rate compared to non-developing ear buds approximately one week before silking. Sowell (1959) using the normal and dwarf versions of the inbred Hy at a plant population of 52,000 plants per acre, observed 5} barren plants in the dwarf and 62% in the normal inbred. He explained the ability of the dwarf to resist barrenness as due to the lack of competition between ear development and vegetative elongation following fertili- sation. The dwarf version stopped vegetative growth at flowering and used the products of photosynthesis for ear development. Geography Single-cared plants have been preferred in the northern Corn Belt and selection against the two-cared characteristic continues. However, prolific varieties, which develop an ear at more than one node, have been widely grown in the southern United States. Freeman (1955) described the per- formance of prolific hybrids in the southern United States and concluded that prolific hybrids are better adapted to ex- treme fluctuations in fertility and plant population than single-cared hybrids. Environmental Modification of Morphology The highest percentage of two-cared plants in hybrids reported in the literature is 85 percent (Freeman, 1955; Bauman, 1959). Bauman (1960) in Georgia, found that date of planting greatly affected the expression of the two-cared characteristic. .Effect of Increased Plant Population on Ear Number Comparisons of single-eared and prolific hybrids for yield under various levels of fertility, population, and moisture have been made by workers in several states (Jesephson, 1957; Gibson, 1957; Zuber and Grogan, 1956; Zuber et al., 1960; Bauman, 1960). With increased plant population, ear number was reduced in both single-cared and prolific types. Under stress conditions, single-cared hybrids produced barren plants but no barrenness was observed in the prolific hybrids. Sass and Loffel (1959) correlated ear bud development with barrenness at two planting rates. No difference in de- velopment of first and second ear buds was apparent until 74 days after planting. The period between the 68th and 74th day was critical to ear differentiation and elongation of silks. Experiments using Corn Belt material indicated that hy- brids with the highest expression of the two-cared character- istic at low plant populations tended to resist barrenness at high pepulations (Lang, et al., 1956; Findlay, 1956). Stinson and Moss (1960) studied the effect of reduced sunlight on the yield of eleven hybrids under conditions of high fertility and irrigation. The hybrids were grouped as tolerant and intolerant to thick planting. Shade reduced the yield of both groups. The intolerant group was reduced approximately twice the amount of the tolerant group. Bar- renness increased in both groups in the shade; the intolerant group had six times more barren plants than the tolerant group. They concluded that hybrids differ in their ability to utilize sunlight and that this was the basis for tolerance or intolerance to high plant population. Correlation of Yield with Morphological Characters Numerous cases of correlations of yield with various morphological characters are reported in the literature. The characters showing the highest correlations with yield are usually those which have been termed 'components of yield“. The primary components of grain yield in corn ac- cording to Long (1953) are: (1) number of ears per plant, (2) kernel weight, (3) kernel row number and (4) number of kernels per row. The secondary components are: (1) weight of grain per ear and (2) number of kernels per ear. He studied the effect of heterosis and the degree of dominance of 92 different Fl hybrids and parental lines. The hy- brids were lower in ear number, nearly identical in row number, about 8 percent higher in kernel weight and 42 per- cent greater in number of kernels per row than their re- spective I'top parents“. Inheritance of Ear Number The percentage of total yield from first and second ears of crosses between Mexican, Brazilian, and Corn Belt lines were reported by Griffing and Lindstrom (1954). They were interested in the use of I'exotic" germ plasm and its effect on the combining ability of derived lines and on the modification of Corn Belt agronomic characteristics. They concluded that it was possible to introduce "exotic” germ plasm into a line and to restrict the yield of the inbred and crosses to a predominately single-eared type. Gene Number and Degree of Dominance To the author's knowledge, no estimates of the number of genes affecting car number have been reported. Partial to complete dominance was found for genes controlling ear number estimated from the F2 generations of three prolific single crosses by Robinson, et a1. (1949). Lindsey (1957), working with three Corn Belt Open-pollinated varieties, found the degree of dominance for ear number to be in the partially dominant range. Gardner and Lonnquist (1959), working with the F2 and F8 generations of the single cross (M14 x 187-2) estimated partial dominance for genes control- ling ear number. Heritability_Estima§es Low heritability values have been reported for the two-cared characteristic. Collier (1959), in Texas, ob- tained heritability estimates of 12.5 and 11.2 in the second and third cycle of recurrent selection using two Open-pol- linated varieties. Low heritability for car number may be inferred from the parent-offspring correlations of r I 0.20 3 .09 in 1926 and r a 0.12 t .09 in 1927 by Jenkins (1929). Robinson et a1. (1949) working with the single crosses 01.21 x N07, N016 x N018, and N034 x N045, reported heri- tability values of 32.8, 14.0 and 26.0 for the three hybrids respectively. B. MATURITY Early maturity is essential in hybrids for the northern Corn Belt. Different measures of maturity have been used including hardness of grain, browning of husks, days to silking, physiological maturity, and moisture at harvest. Environmental and Genetic Factors affecting Maturity Shaw and Thom (1951) separated the development of the corn plant into three periods: (1) planting to emergence, (2) emergence to tasseling, and (3) tasseling to silking. Environmental factors were responsible for variations be- tween hybrids for the period from planting to silking which averaged 65 days with a range of 20 days for the years 1921 to 1945. Very early, medium, and late maturing hybrids held their relative position in days from planting to tasseling. The time between tasseling and silking was shortest for the hybrids adapted to the area. Thus, the interval from plant- ing to silking was the most important in determining the time of maturity. In a further study, they found the period from silking to maturity was affected little by weather con- ditions. The period from maturity to safe cribbing moisture, however, was greatly dependent upon environmental conditions. Jones (1952) found genetic differences in days to silk- ing and also from silking to physiological maturity. Some inbred lines are late in silking but reach physiological maturity rapidly. Days from planting to silking would appear to be the logical basis for the corn breeder to adjust genet- ically as well as the period from silking to maturity. Dessureaux and Neal (1948) also found that, in general, hy- brids that flower early also mature more quickly than late flowering types. Inheritance: Dominance Relationships, Gene Numbers and Heritability Estimates Emerson and East (1913) investigated the inheritance of days to flowering in crosses of the early flowering Tom Thumb popcorn variety and a late flowering Missouri variety. The F1 was intermediate with wide segregation in the F2 ranging beyond the means of the parents. .Lindstrom (1943) reported dominance for early flowering in four crosses of corn inbreds. Long's (1951) report of heterotic effects in a single- cross for days from tassel initiation to anthesis were in- terpreted as indicating dominant gene action. Yang (1949), using inbreds of similar maturity found the F1 to be earlier in silking date than the earlier parent. He concluded that two or three independently inherited genes, with dominance for early silking, differentiated the two parents. Jones (1952) observed complete dominance of genes for early date of silking in six different crosses of early x late inbreds. Epistatic effects for the dominant early -10- genes from the inbred R53 were apparent. Gene number esti- mates varied from 5 to 19. Estimates of heritability for days to silking ranged from 11 to 48 percent with a mean of 29 percent. Dominance of three major genes for early silking was reported by Mohamed (1959) using inbreds adapted to Egypt. Giesbrecht (1959) estimated that 3 to 7 genes with partial dominance and epistasis for earliness differentiated the cross of a very early Canadian inbred, V3, and the later maturing inbred 314. The heritability values obtained (59 percent in 1954 and 75.9 percent in 1955) indicated that selection within the segregating generations of the cross should be effective in isolating early maturing lines. Giesbrecht cited estimates of gene numbers for date of silking from Zoebish (1950) who reported at least six factors, Agble (1954) who reported four factor pairs, and Bianchi and Miliani (1954) who estimated at least three and possibly four or five factors exhibiting dominance for early date of silking. C. BIOMETRICAL CONSIDERATIONS Although based on the principle of particulate genes discovered and elaborated by Mendel, characteristics con- trolled by several genetic factors seldom segregate into discrete phenotypic classes so that simple Mendelian ratios can be deduced. The continuous distribution of phenotypes is a result of the segregation of the underlying genetic -11- factors plus the masking effects of environment, the inter- action of genes with each other and the interaction of the genetic factors with the environment. Among the significant contributions to the analysis of complex traits cited by Wright (1952) and Powers (1955) are: (1) Nilsson-Ehle's demonstration of the particulate nature of genetic factors controlling seed color in wheat, (2) Johannsen's evidence of the Joint effects of genotype and environment on the phenotype, and (3) East's and his students' proof that a combination of genetic and environmental variations could produce the continuous variation of quantitative characters and that quantitative characters were indeed inherited ac- cording to the principles of Mendel. The statistical procedures for the separation of total variance into genetic and environmental components are primarily due to Fisher and wright. Other noteworthy con- tributions have been made by Mather (1949); Powers (1955); Lush (1943); Comstock and Robinson (1948): Lerner (1958) and Narner (1952). The analysis of quantitative characters necessitates the use of (1) an adequate genetic design and (2) an ade— quate experimental design. A genetic design utilizing both segregating and non-segregating generations is essential for the estimation of environmental and genotypic variances. An experimental design which facilitates the analysis is one which minimizes or provides sufficient estimates of the genotype x environment variance and reduces the standard errors. -12- A knowledge of the number of genes, the degree of domi- nance, the type of gene action, and heritability allows the plant breeder to predict the results of various breeding schemes, probable rates of change, and to choose the most efficient procedures. Genotype x Environment Interaction The importance of considering the genotype x environ- ment interactions in genetic investigations has been stressed by various writers (Sentz, 1954, and Comstock, 1955). The mistaking of variance due to genotype x environmental inter- action as usable genotypic variance can lead to false con- clusions and wasted effort. Collier (1959) and others, have suggested the use of the mean performance of several loca- tions in an attempt to minimize the obscuring effects of geno- type x environmental interactions upon selection and esti- mates of heritability. Heritability estimates based on one population at one location in one year are of doubtful value at least for certain traits. The need is apparent for a generalized estimate of heritability. Theta of Scale - Transformations In order to facilitate the analysis of quantitative characters an appropriate scale is essential. The first step in an analysis of a quantitative character is a test for a scale on which the effects of the genes concerned are additive. Tests for adequacy of scale and the limitations of these scales were discussed by Mather (1949), Powers (1941, 1950) and Falconer (1960). A satisfactory scale is one on which the action of the genes and non-heritable factors is additive on the average and one which removes the epistatic effects and allows dominance to take its own value on the scale used. Many types of trans- formations are available (Bartlett, 1947). As Falconer points out, transformations should not be made without good reason. For the first purpose of experimental observation is a description of the genetic properties of the popula- tion, and a scale transformation obscures rather than 11- luminates the description. Partitioningof Variance After an adequate scale has been found partitioning of the variance into genetic and non-genetic components follows. If the genotype x environment interaction is negligible, satisfactory separation of non-genetic variance from the total variance of the segregating populations can be accom- plished as illustrated by Mather (1949). The estimation of the environmental variance from the mean variance of the parents or by the use of the F1 variance may be of doubtful value due to the lack of competition between individual plants within the rows of the inbred parents plus the fact that the parents may differ considerably in maturity and be subjected to different climatic conditions at critical stages in their development (Sentz, 1954). The buffering effects -14- of heterozygosity in the F1 may restrict its value as an estimator of environmental variance. To circumvent this difficulty, Warner (1952) suggested the use of the segre- gating generations, F2 and first backcrosses, to estimate the additive genetic portion in estimating heritability. Warner's method assumes that the non-heritable components of variance are equal in the F2 and backcross populations. Burton (1951) used the variance of the F2 minus the variance of the F1 divided by the variance of the F2 as an estimate of maximum heritability. Lush (1943) called this "herita- bility in the broad sense“. This estimate includes both the additive and the non-additive genetic variances. The for- mulae of Mather and Warner each estimate heritability in the “narrow sense“ in which heritability is the ratio of additive genetic variance to the total phenotypic variance. Culp (1960) used both of the above methods in estimating herita- bilities in sesame and stated that if dominance and epistasis were not present both formulae should give estimates which agree closely. -15- MATERIALS AND METHODS Four early maturing 00rn Belt inbreds (W64A, 0h28, Oh5l and R53) and three late maturing Tennessee inbreds (T115, T434 and T490) were used to study inheritance of ear number and maturity. W64A is single-cared, 0h28 and Oh5l exhibit a two-cared tendency, and R53 is intermediate. The Tennessee white inbred T115 and the yellow inbreds T490 and T434 were chosen as parents on the basis of observations taken in a preliminary experiment where they exhibited a high percent- age of two-cared plants. The genetic notation employed in each of the nine crosses was as follows: P1 - the Corn Belt inbred P2 - the Tennessee inbred F1 - the single cross of Corn Belt x Tenn. inbreds F2 - the selfed progeny of the F1 B1 - the first backcross of F1 to P1 B2 - the first backcross of F1 to P2 Crosses used were: 0h5l x T115, W64A x T115, R53 x T490, W64A x T490, 0h5l x T490, 0h28 x T490, R53 x T434, N64A x T434, and Oh51 x T434. A split-plot design with crosses as the main plots and generations as subplots was used to obtain precise estimates of generation variances. Four rows of each generation were -15- used. The crosses were randomized within replications, the generations randomized within crosses with the exception that the inbred parents were planted side by side to minimize the environmental variance between them. Three replications of the crosses (W64A x T115) and (0h51 x T115) were planted in field I and the other crosses were planted in four repli- cations in field 11 of the Crops Farm, East Lansing. Field I was planted on June 2 and field 11 was planted on May 19, 1960. Two seeds were hand planted in hills one foot apart and the plants were thinned in the seedling stage to one plant per hill. Rows were 36 inches apart which gave a plant population of approximately 14,500 plants per acre. Complete fertilizer was applied in the row and nitrogen was added at the second cultivation. The growing season was relatively cool during June and July. Moisture stress was evident in mid-July. Adequate moisture was available during the balance of the growing season. All data were taken on individual plants. Date of. silking observations were made daily by tagging each plant when the main ear shoot exhibited silk one—half inch in ‘1ength. Number of ears per plant was recorded at harvest by examination of each ear for one or more developed kernels. A notation was made to distinguish competitive from non- competitive plants. Plants visibly diseased or damaged were discarded. -17- Date of silking was recorded only for the crosses (W64A x T115), (Oh5l x T115), (R53 x T490) and (WB4A x T490). Silk- ing dates were transformed to days from planting to silking. All plants were used in the analyses of variance of days to silking. Due to the effect of competition on ear number, data from competitive plants only were used in the analysis of variance of ear number. The theoretical means and standard errors were calcu- lated from formulae given by Powers (1955). A significant deviation of the F1 mean from the parental mid-point or av- erage of the parents was interpreted as phenotypic dominance; however, it could be due to genetic epistasis (Figure 1). A non-significant difference between the mean of the F1 and the mean of one of the parents was considered to imply complete phenotypic dominance. A significant difference between the observed and theoretical Fl means and a significant differ- ence between the observed Fl mean and the mean of the near- est parent was interpreted as partial phenotypic dominance. No significant difference between the observed Fl mean and the parental mid-point indicated no phenotypic dominance. Heterosis was implied whenever the mean of the F1 was signif- icantly larger than the mean of the large parent or smaller than the mean of the small parent. If there was no apparent phenotypic dominance exhibited by the F1 population, the mean of the F2 population should be equal to the mean of the F1 population and the means of the first backcross generations should agree closely with ~18- Figure 1 Graphic Illustrations of Dominance, Epistasis, and Heterosis .—v L l ’{1 1 P1 Mid-parent P2 F1 deviates significantly from mid-parent and also from the mean of the nearest parent - indicates partial phenotypic dominance or genetic epistasis. Epistasis in the broad sense includes many unsop- arable effects. I 0'2] L J 1! P1 Mid-parent P2 )— #;‘ F1 does not differ significantly from mean of one of the parents - indicates complete phenotypic dom- inance. 1'52 Be 1 I 1 ‘ Mid-parent P2 Bl R1 L p1 F1 does not differ significantly from mid-parent indicating no phenotypic dominance. F2, 81, and B2 do not differ from theoretical expected values in- dicating no genotypic dominance or epistasis. l p L l J P1 Mid-parent P2 F1 deviates significantly above larger parent or be- low smaller parent - indicates heterosis. their theoretical expected values. The backcross variances should be of similar magnitude also. However, if the F1 popu- lation displayed phenotypic dominance, the observed means of the segregating populations should not differ significantly from their theoretical expected means unless something other than allelic dominance was Operating. The dominance devia- tions from simple additive gene action are considered in all of the formulae used to calculate the expected theoretical means; therefore, any significant difference between the theoretical and observed means of the segregating generations is due to non-allelic interaction or genetic epistasis. With simple allelic dominance the mean of the F2 should fall between the parental mid-point and the F1 mean and the variance of the backcross to the dominant parent should be suppressed due to dominance and should be significantly smaller than the variance of the backcross to the recessive parent. Deviations from the theoretical expected means and variances are considered to imply genetic epistasis or the interaction of non-alleles. The formulas for the calculation of the theoretical means and their standard errors from Powers (1955) are as follows: Theoretical F1 ' 9(51 + P2) Theoretical F2 . {(51 + F: + 2F1) Theoretical Bl fi §(P1 + F1) Theoretical 32 ' 5(P2 + F1) -20- Standard error of theoretical Ti Q J 931312 4» 51.21322 4 .Fjssiifi + 931512 44mm)? 4 sgfiiz + sail? \ 4 SEF12 + ssizz \ 4 -21- EXPERIMENTAL RESULTS Ear Number To determine whether the heritable and non-heritable agents affecting ear number were acting in an additive man- ner, theoretical means were calculated for each segregating population using the arithmetic and logarithmic transformed data as suggested by Powers (1955). The observed and cal- culated arithmetic and geometric means and tests of sig- nificance are shown in Table 1. None of the F values in- dicated a significant difference between the observed means and the theoretical arithmetic or the logarithmic means. The test based on the F2 population is the most sensitive because it includes a larger array of genotypes than either backcross population. If the genes and environmental agents are acting in an additive manner on the average, then the observed and theoretical arithmetic means should not differ significantly. If the effect of a gene substitution on the phenotypic expression of a character under consideration is not additive but multiplicative in action then the trans- formation of the observed data to logarithms is indicated. The non-significance of the F values obtained indicated that either scale would be satisfactory for the genetic analysis of the crosses. -22- Hm.o ma.~ . - oH.o oo.m . u no.0 on.” u - osaup a as.” om.” as.a s«.H we.” ae.~ Ho.a as.” me.H use: as.” am.a om.” we.” os.a es.a He.a as.” am.” vase a Home so.” am.” no.” mm.a mm.” no.H as.” on.a ea.” «nee a «so: ea.a mo.m ae.a ow.” om.a om.H as.” am.” «o.a easy a sum as.” os.a He.H ne.a oe.a Hm.a ma.a we.” ao.a cave a mega no.” ne.a mm.” . on.” so.” as.” Ho.a we.” mm.H cave a ammo as.” mo.~ as.” ea.” ea.” no.” sm.a on.a mm.” case a «we: mo.a we.” se.a mn.a mv.a an.a m¢.a an.a we.” owes a new ma.a vo.m so.m am.” an.” oa.a no.” mo.a ao.m oaan a «as: Hm.« am.m on.» as.” om.a mm.” oo.a vo.m Hm.a oaaa a demo canvas capes ace: odapoa capes use: oaapoa caves use: ndwoobo “WWW-HM debhom IE «85%? dobho a flwuoowo “Maw Umbho u a mono ccox,uQPMHson soapsasmem mm op am «(I coawmwa cm A o» m uneoapea om mfimzbz m ouposzvand seesaw .090.“ 0 naoanm .090 u 0 agenda .090 «Nd eeoao mmommm nm m mam-.4. TABLE 3 NUMBER OF PLANTS, MEANS AND THEIR STANDARD ERRORS, TOTAL AND GENETIC VARIANCES FOR EAR NUMBER Fipu- "No. of: ‘IMean ear S.E. o? TotaI GEnetIc 1gtion _plants number mean variance variance Oh51 x T115 P1 149 1.56 0.043 0.2687 - - Bl 137 1.88 0.038 0.1922 0.0768 F2 123 1.91 0.056 0.3808 0.2654 B2 109 2.30 0.050 0.2693 0.1539 M64A x T115 El 111 1.19 0.036 0.1450 ~0.0607 F1 111 1.64 0.043 0.2058 - - F2 101 2.01 0.074 0.5525 0.3467 B2 89 2.07 0.037 0.1213 ~0.0845 P2 46 2.44 0.098 0.4376 - - R53 3 T499 Pl 187 1.48 0.042 0.5740 - - Bl 148 1.39 0.046 0.5537 0.0652 F2 205 1.42 0.034 0.4795 -0. 116 B2 119 1.67 0.045 0.4881 -0.0032 W64A x T490 P1 233 1.04 0.015 0.0501 - - F1 213 1.30 0.032 0.2139 - - F2 140 1.28 0.038 0.2069 0.0749 82 148 1.55 0.043 0.2694 0.1373 P2 112 1.74 0.044 0.2140 - - -26.. Table 3 continued Pepula- No. of Mean ear S. E. of Total Genetic tion (plants number mean fi_yariance g_ variance Oh51 x T499 Pl 187 1.57 0.034 0.2184 - - B1 219 1.70 0.030 0.1978 -0.0212 F1 177 1.70 0.035 0.2191 - - P2 33 1.82 0.105 0.3608 0.1417 P2 104 1.76 0.044 0.1981 - - Oh28 x T490 Bl 206 1.81 0.028 0.1655 0.0410 F1 136 1.67 0.038 0.1998 - - F2 183 1.65 0.032 0.1837 0.0592 B2 148 1.71 0.036 0.1886 0.0641 P2 64 1.86 0.048 0.1462 - - R53 x T434 Pl 249 1.39 0.029 0.2121 - - Bl 223 1.25 0.028 0.1769 0.0102 F1 112 2.00 0.039 0.1667 - - F2 221 1.54 0.040 0.3557 0.1890 B2 260 1.49 0.033 0.2794 0.1127 P2 206 2.15 0.067 0.9276 - - W64A x T434 P1 215 1.09 0.025 0.1333 - - Bl 241 1.05 0.013 0.0434 -O.2442 F2 250 1.14 0.022 0.1170 -0.l705 B2 228 1.03 0.010 0.0251 -O.2625 P2 195 2.15 0.058 0.6584 - - Oh51 x T434 Pl 123 1.59 0.048 0.2876 - - Bl 185 1.74 0.033 0.2069 0.0519 F1 190 1.80 0.029 0.1550 - - F2 145 1.82 0.038 0.2094 0.0544 32 128 1.80 0.048 0.2960 0.1410 P2 180 1.94 0.046 0.3801 - - ear number of 1.57 ranging from a mean of 1.56 ears per plant in the Oh51 x T115 cross to a mean of 1.59 cars per plant in the Oh51 x T434 cross. The Tennessee inbred T115 possessed the strongest two-eared tendency of the inbreds used in this study. Every plant of the F1 population (Oh51 x T115) had three visible ear buds and developed two or more cars per plant. The relatively large variances shown in Table 3 for the in- bred T115 may have been due in part to the late silking date with cool night temperatures and scarcity of pollen. Dominance Relationshipg A summary of the dominance relationships for ear num- ber is shown in Table 4. The parents in the cross Oh51 x T115 represented the most prolific inbreds of each group. The Tennessee inbred T115 is very late in maturity and una- dapted to the northern Corn Belt. It operated under dif- ferent environmental conditions than the adapted inbreds dur- ing anthesis which may explain the relatively high variance of this population. All of the other populations silked much earlier in the season. The F1 was intermediate in ear num- ber with each plant bearing two or more ears consistently. No dominance and no epistasis was indicated from a study of the various generations. When the least prolific Corn Belt line, W64A, was crossed with the most prolific Tennessee line, T115, no dominance or epistasis was observed. Complete phenotypic and genic domi- nance for the one ear condition was shown by the various swooponno oz oononaaoo canow oz eonondaoo ouquononn oz on; a Sam moo one new oaoopoanu moo one you oaoouoanm goo one you oonondloe ounow opodnaoonH Ado one new oononaaoo ounow oaoansoo ado one new oononaeoo eaazuononn opoaaaoenH ohoo c)» you eononaaoo oanzaononn opoamaoo ones x «we: envy H new poo one new oaoopodzm no. one you eononaaoo canom opoamaeo onoopoamo oz hoe one new odooaopon oonondaoo canom oz oaoapxo onan oononasoo odnzpononn ouonneoo conundaoo oanzpononn oz . mm cm: x mono cove x Hone odoovoamo oz oaoououmo oz no. one new oonondsoo canom oaoanaoonu hoe one you oononaaoo canow oceansoo no. one new eononnnoo canzpononn opoamaoonn no. one new oononaaoo odmzoononn ouodqaoo omen x «em: omen N mom oaoopoano oz oaoopoaoo oz eononaeoe canom oz conundsoo canow oz oononasoo odnzpononn oz oononaaoo odnzaononn oz odds N «em: ease N mmno mmHmszHaon RH on» no pnooauanwam ea .4 ooaoo.o w oooo.es ..ooaoo.o a onoao.oa .eoooo.o a oosoo.on omen s «we: ..ooooo.o a opomo.oa ¢.Hoooo.o s omoao.on oooo.o « osooo.o cove a non .eoaoo.o w ooono.on .eooooo.o « ooaao.oa ..Hoo.o « oommo.ou made a «we: aeo.o u ooooo.oa oHo.o « ooeao.o No.0 a vomoo.o adds a Hone canon oaanadnomog .eoo.H a no.5: ecnoo.o u om.n: .coon.o « m«.Hu cove s «em: .esoo.o u om.oa ..ooo.o a on.na .enuo.o a no.” cove a non ..me.a a oo.oHu ..mmo.o « oo.nn moo.o.unmo.ou mafia s_ onooanpan< ononnm .eum H o oaoanu .oam w m uncanm .oam s 4 cacao ozauuno so mean non mmommm om m 35.9 TABLE 9 MEANS AND THEIR STANDARD ERRORS, TOTAL'VARIANCES, AND GENETIC VARIANCES FOR DAYS TO SILKING Population Mean-days 9.8. of Total Genetic to silking mean variance variance Oh51 x T115 (Logarithmic valueg) P1 1.9647 0.01316 0.025812 - - 81 1.9634 0.0013 0.000222 -0.000108 Fl 1.9601 0.0149 0.000330 - - F2 1.9675 0.0157 0.028821 0.028491 P2 2.0448 0.00165 0.000264 - - I64A x T115 (Arithmetic values) Pl 89.54 0.2074 7.0105 - - 81 87.71 0.8783 16.6686 13.6975 P2 93.80 0.3390 15.8617 12.8902 82 100.42 0.2605 8.7562 5.7848 P2 113.03 0.4625 21.1812 - - 853 x T490 (Arithmetic valueg) Pl 73.46 0.1144 3.4549 - - 81 74.76 0.2361 13.1544 6.5521 Fl 74.52 0.1645 6.6023 - - F2 f 78.18 0.2121 10.6696 4.0673 82 83.84 0.2704 13.0104 6.4081 I64A x T490 (Arithmetic values) P1 77.17 0.1175 3.8080 - - 81 77.77 0.1363 5.0895 -2.ll64 Fl 79.78 0.1628 7.2059 - - F2 81.38 0.2421 12.0714 4.86551 82 85.93 0.2583 12.9480 5.7422 P2 96.05 0.2027 6.4111 - - ‘QQminance Relationships A summary of the dominance relationships for days to silking is shown in Table 10. The various populations of the cross Oh51 x T115 exhibited no phenotypic dominance, no genie dominance, and no epistasis when the logarithmic transformed data were used for the analysis. The crosses W64A x T115 and 853 x T490 exhibited in- complete phenotypic and incomplete genie dominance and possible epistasis for early silking. Incomplete pheno- typic and genie dominance and no epistasis was indicated for the cross WB4A x T490. Gene Number_and Heritability Table 11 shows the estimates of gene numbers and heri- tabilities for silking date. The cross Oh51 x T115 gave rather low estimates for the number of effective factors differentiating the two parental inbreds. The cross W64A x T115 indicated a one-to-five factor difference between the parents for silking date. The heri- tability estimates for this cross varied with the method of calculation used. The crosses 853 x T490 and W64A x T490 gave higher gene number and lower heritability estimates. TABLE 10 SUMMARY'OF DOMINANCE RELATIONSHIPS FOR DAYS TO SILKING Qh51 x T115 No phenotypic dominance No genie dominance No epistasis N64A x T115 Incomplete phenotypic dominance for early silking Incomplete genic dominance for early silking Possible epistasis for early silking R53 x T490 Incomplete phenotypic dominance for early silking Incomplete genie dominance for early silking Possible epistasis for early silking W64A x T490 Incomplete phenotypic dominance for early silking Incomplete genie dominance for early silking No epistasis _41- womoo 03.0 OoNH o.m omvfl K $03 ooo.o Hon.o o.no o.oH cove a non son.o nao.o o.a o.a adds a 44o: coo.” ooo.o o.H o.a mafia x Home nouns: has; unmamxucopnsm unmaasaofipoeo noose unpasamua aoazmmooanom ‘ 1 emcee no nopasz -l‘s-‘ l’b‘ l'li‘ll-l.. MAHm 09 was mom :SHQBE 2 Emma szw b0 mma