الفهرس 
AcnoledgementOnly 14 pages are availabe for public view
 |  | from | 202 |  |  |
| | | from | 202 | | |
Abstract
SUMMARY
The aim of the present investigation was to compare the relative effeiciency of S1 lines, D-I and D-II (Comstock and Robinson, 1948) and diallel cross designs (Griffing, 1956-Method - 1 - Model - II) for estimation the genetic variance. Also, to estimate the magnitude of the genetic variance components and their interactions with years, heritability and expected genetic advance of yield and other agronomic traits (grain yield / plant, ear length, ear diameter, number of rows / ear, number of kernels! / row, plant height, ear height, tasseling date and silking date for the two maize populations, Corn Belt (C.B.) and Tuxpeno C - 17 (C-17).
In 1985 season, 500 ears from each population were planted and selfed to produce S1 lines for making the different mating designs, i.e., S1 design, Design-1, Design-11 and diallel cross mating design in 1986 season. Approximatly equal parents and progenies were produced from each desing.
All designs in both populations were tested in randomized complete block
design with three replications in both successive seaons 1987 and 1988 at the Agriculture Research Station of the Agricultural Collage at Shebin El-Kom, Menoufiya University. Usual practices were applied during the growing seasons. The obtained results could be summarized as follows :-
A- Mean performance :
1 - Mean performance of the two populations was higher in the first season than in the second one for most traits. The Corn Belt population exhibited the highest mean values in most studied traits in all mating designs.
2 - Coefficient of variability estimates were higher in the second growing season than those in the first one in most traits for both populations in all designs.
B - Genetic variance components
1 - In S1 lines design, genetic variances (a 2G) in the two maize populations for grain yield / plant were higher in the second growing season than those in the first one . Also, the amount of genetic variance components in C-17 population was higher than that of C.B.
2 - In S1 lines design the interaction of a G x Y for Grain yield / plant was non - significant for the two maize populations. The estimates of a 2 G x Y were significant for ear length and ear height in C-17; and for plant height and two flwering dates in the two populations.
3 - In design- I mating system (D-I), additive genetic variance was significant for grain yield / plant, in the combined data, plant and ear height in the two growing seasons; for silking and tasseling dates in their three estimates in C.B. population. On the other side, it was significant in C-17 for ear yield / plant, number kernels / row and silking date in the second year; for ear length in the second year and the comined data plant and ear heights in their three estimates; and for tasseling date in the two growing seasons.
4 - In design - I, the dominance variance was significant only for ear diameter in combined data for C.B. population. Most traits had negative values and may be deduced to zero.
5 - In design-I, significant additive genetic variance x year interaction was obtained for plant and ear heights, tasseling and silking dates in C.B. population while a 23 x Y interaction was significant only for grain yield / plant and number of kernels /row in C.B. population and ear length in C-17 population.
6 - In design - II mating system (D-II), the additive genetic variance components was significant in C-17 population for plant height in the first year and
silking date in the second season and the combined analysis. Corn Belt maize population had the highest amount of additive variance for some traits.
7 - In design - II, the dominance genetic variance was significant for; grain yield / plant, ear length and tasseling date in the combined analysis of the two populations; ear diameter in the second year and the combined data, number of rows / ear and number of kernels / row in the combined data in C-17 population and ear height, and silking date in C.B. population. Also, most of the estimates of interaction between genetic variance components a 2G, a 2A and a were
were
insignificant in all traits in both populations.
In general, dominance genetic variance was predominant component and it was controlled in the inheritance of most studied traits. It was evident from the ratio of a 2D / a 2A except for few cases.
8 - In diallel crosses design, the general combining ability (G. C.A.) variances component and the additive variances (a 2A) were significant only for number of rows/ ear in the first season for C.B. population and number of kernels / row for C-17 population . Consequently, most cases had negative values and may be deduced to zero, while the specific combining ability (S.C.A) variances components and the dominance variances (a 2D) were significant for almost traits. Also, the C.B. population had the highest amounts of S.C.A. for all traita. In most traits a 2 S.C.A. was the predominance genetic components. Over dominance or partial dominance were obtained from the ratio of a 2D / a 2A for the two populations. The estimates of G.C.A. and S.C.A. or additive and dominance genetic variance interaction with year for all traits were not significant.
C- Comparison of the genetic variance components resulted from different mating designs :
1 - The lowest estimates of mean performances were obtained from the Si lines design, in all cases except the two flowering traits. The two mating designs, i.e. D-I and D-II ranked firstly followed by diallel cross mating design for mean performance for most studied traits.
2 - SI lines design and diallel cross system exhibited lowest environmental error estimates followed by the two North Carolina mating designs in two populations for grain yield / plant. For the rest studied traits low estimates were obtained with diallel cross design followed by D-I mating in both populations. Relative to coefficients of variability, diallel cross mating design gave the lowest estimates, while Si lines design gave the highest estimates.
3 - The lowest values of additive genetic components from grain yield / plant were detected from the diallel crosses followed by Si lines design. North carolina D-I mating gave the highest values of additive genetic variance followed by D-II mating.
4 - The diallel cross design exhibited the highest estimates of dominance genetic variance followed by North caroling mating design - II and then D-I.
5 - The highest estimates of additive genetic x year interaction resulted from S1 lines and D-II in both populations and D-I in C-17 population. Meanwhile, the highest values of dominance x year interaction resulted from DI, DII and diallel cross design of C.B. and diallel cross design of C-17 population. Generally, values of interaction between genetic components and years differed from one population to another and from design to design.
D - Heritability :
1 - In S1 lines design, in both years low to maderate estimates of h2 in narrow sense were obtained for grain yield / plant, moderate values for yield components and from moderate to high values for other traits in the two populations . For yield, most estimates of h2 were high in the second year than those of the first one. Invers results for the remaining traits were obtained. In comined data, most traits gave low h2 estimates except in few cases, where grain yield / plant gave zero estimates in the two populations.
2 - In D-I in both years high to moderate heritability values were obtained for most traits in both populations. Also, heritability values in narrow sense were high to moderate in magnitude and nearly equal to their corresponding values in brood sense for most traits. In the combined analysis, moderate to low values in broad sense were obtained for most traits in both populations. Also, heritability values in narrow sense were nearly equal to their corresponding values in broad sense for most traits.
3 - In D-II, in the two years, high heritability values in narrow and broad senses were detected in ,most traits. Also, these values were nearly equal to their corresponding values in broad sense for most traits. In the combined analysis,nearly similar values for both broad and narrow sense heritability were obtained for silking and tasseling dates in C-17 population and number of kernels / row in C.B. population.
4 - In diallel cross design, in the first year, high heritability values in broad sense were detected for all traits. However, low heritaibility values in narrow sense were obtained for most studied traits. In the second year and the combined analysis, highly to moderate values in broad sense were detected for all traits. Low or zero h2 values in narrow sense were recorded for all the studied traits.
E - Expected genetic advance from selection :
1 - S1 Lines family, test cross (recurrent selection - R.S.) and full sib family selection methods were the most superior methods proposed to improve the two maize populations, where they gave the highest expected gains (A g %) amongest the other selection methods.
2 - The most suitable selection method for improving the two populations at the same time would be the reciprocal recurrent selection method to utilize the amounts of 6 and a -D existing in n the two maize populations.