الفهرس 
INTRODUCTIONOnly 14 pages are availabe for public view
 |  | from | 51 |  |  |
| | | from | 51 | | |
Abstract
The present experiment was carried out at the Experimental Farm of Gemmeiza Agriculture Research Station, Egypt during the four successive seasons i.e. 2009 / 2010, 2010 / 2011, 2011 / 2012 and 2012/2013. Six genotypes of bread wheat were chosen as Gemmeiza 11 (Drought sensitive), Line 1 (Drought tolerant), Line 4 (Drought tolerant), Line 5 (Drought tolerant), Gemmeiza 9 (Drought sensitive) and Misr 1 (Drought sensitive). Three crosses derived from the above parents have been chosen as follows: Cross 1 = Gemmeiza 11 × Line 1 Cross 2 = Line 4 × Line 5 Cross 3 = Gemmeiza 9 × Misr 1 In 2009/2010 season, the crossing between six parents pair crosses were made to obtained the F1 hybrid seeds. In 2010/2011 season, the F1 plants of each cross were selfed to produce F2 grains. In 2011/2012 season, the obtained materials F1, F2 and the parental genotypes of each cross were sown. Twenty random F2 plants of each cross were crossed, as males, back to its respective parents P1, P2 and F1 (P1 × P2) to produce L1i (P1 × F2i), L2i (P2 × F2i) and L3i (F1i × F2i) respectively. In 2012/2013 season, Triple test cross (TTC) families, P1, P2 and F1 were evaluated in two separate irrigation regime. The first treatment normal (N) was irrigated three times after planting irrigation i.e. four irrigations were given through the whole season. The second treatment water stress treatment (S) was given one surface-irrigation 50 days after planting irrigation i.e. two irrigations were given through the whole season. Objectives of study: The objectives of this investigation were to study (i) The role of non-allelic interaction (epistasis) in the inheritance of grain yield and its components using triple test cross given by Kearsey and Jinks (1968),(ii) The detection and estimation of additive (D) and dominance (H) components of genetic variation (iii) Determination of drought tolerance index, yield reduction ratio and drought susceptibility index for grain yield/plant, (iv) computing the genetic correlation among various traits and partitioning it to additive, dominance and epistasis correlations, (v) Predicting the properties of recombinant lines and (iv) Estimation of the genetic diversity of TTC families. Traits studied: Data were recorded on eleven traits i.e. number of days to heading, flag leaf area, number of days to maturity, number of spikes per plant, plant height, main spike length, number of spikelets per main spike, number of kernels per main spike, main spike yield, grain yield per plant and 1000- kernels weight. Statistical Procedure: The concept of triple test cross (TTC) analysis was proposed by Kearsey and Jinks in (1968). The mean squares for deviations (L1i + L2i – 2L3i) was used for detection of epistasis. Epistasis, additive and dominance correlation coefficients were computed from deviations ( L1i + L2i – 2 3i L ), ( 1i L + 2i L + 3i L ) and ( 1i L - 2i L ) respectively. The proportion of inbreds superior or outperform parental range its equal to the probability integral corresponding to the value [d] / D , while, the range of inbreds is given by m ±2 D ( Pooni and Jinks , 168 Summary 1978). The proportion of recombinant lines corresponding to the probability level were obtained using Fisher and Yates (1963) Tables. Principal components analysis was performed, on the correlation matrix of contributed characters for all genotypes. The principle components were expressed as eigen value, latent root, and manifested in eigen vector for all studied traits in each principal component axis (Hair et al. 1987). Hierarchical clustering procedure using ward’s minimum variance method, which minimize within group sum of squares across all partitions, was applied to determine the genetic diversity and distance as outlined by Johnson and Wichern (1988). The obtained results can be summarized as follows: Mean performance of TTC families: The results obtained showed that the water stress affected in all traits studied under three crosses. These results were confirmed by the significant differences among all TTC families and between L1 + L2 + L3 families as well as L1 + L2 families for all traits under different conditions. Genetic behavior of triple test cross families: Analysis of variance revealed highly significant differences among L1, L2 and L3 TTC families for all traits studied in the three crosses under normal and stress conditions revealing presence of fair amount of genetic variability. Test of epistasis: Significant overall epistasis was detected for all traits studied in the three crosses under both conditions except main spike length and number of kernels per main spike under stress in the first cross (TTC1), main spike length under normal condition in the second 169 Summary cross (TTC2) and main spike length and number of spikelets per main spike in both conditions in the third cross (TTC3). Partitioning of total epistasis to its component parts revealed significant and highly significant mean squares for additive × additive (I) epistatic type for all traits studied in the three crosses under both conditions except plant height, main spike length, number of spikelets per main spike and kernels per spike under normal condition and number of spikes per plant, main spike length, number of spikelets per main spike and kernels per main spike under stress condition in TTC1, Number of days to heading and main spike length under normal case in TTC2 and main spike length, number of spikelets per main spike and number of kernels per main spike under normal condition and number of spikes per plant, main spike yield and grain yield per plant under stress condition in TTC3. The rest of epistatic components i.e. additive x dominance and dominance x dominance (J+ L) types were highly significant for all traits studied in the three crosses except number of days to heading under normal condition and main spike length and Number of kernels per main spike under stress condition in TTC1, main spike length under stress condition inTTC2 and main spike length and number of spikelets per main spike under both conditions in TTC3. Individual epistasis: The individual epistasis deviation of each F2 male were generally variable each of magnitudes and sign for all traits studied in the three crosses under both conditions, indicating how epistasis differed among crosses. 170 Summary Detection and estimation of additive and dominance: The mean square estimates due to sums, additive, ( L1i + L2i + 3i L ) were found to be significant and highly significant for all traits studied in the crosses except main spike length and number of spikelets per main spike in both conditions and number of kernels per main spike under stress condition in TTC1, number of spikes per plant in both conditions and grain yield per plant under normal condition in TTC2 and number of days to maturity, number of spikelets per main spike and main spike yield under normal condition and main spike length, number of spikelets per main spike and number of kernels per main spike under stress condition in TTC3. The mean square estimates due to differences, dominance, ( 1i L - 2i L ) were found to be significant and highly significant for all traits studied in the crosses studied except number of days to maturity, main spike length, number of spikelets per main spike and number of kernels per main spike under normal condition and number of spikes per plant, number of spikelets per main spike, number of kernels per main spike and main spike yield under stress condition in TTC1, number of days to heading, main spike length, number of spikelets per main spike and grain yield per plant under normal condition and number of days to heading, number of spikes per plant and number of spikelets per main spike under stress condition in TTC2 and number of days to maturity, number of spikes per plant, main spike length, number of spikelets per main spike and main spike yield under normal and number of spikes per plant, main spike length, number of spikelets per main 171 Summary spike, number of kernels per main spike and grain yield per plant under stress condition in TTC3. The dominance genetic variance (H) was found to be much larger in magnitude than the corresponding additive (D) for most traits studied in TTC1 under both condition resulting in (H/D) 0.5 more than one and reflecting over dominance for these traits. Whereas, in TTC2 the results revealed that additive genetic component were larger in magnitude than those of dominance ones for most traits under normal condition resulting in (H/D) 0.5 less than one. While, under stress condition results revealed that dominance component was larger in magnitude than the corresponding additive ones for most traits studied resulting in (H/D) 0.5 more than one. With respect to TTC3 the additive genetic components were larger in magnitude than those of dominance for most traits under both conditions resulting in (H/D) 0.5 less than one reflecting the role of partial dominance in the inheritance of most traits studied. Direction of dominance: The direction of dominance showed insignificant values for most traits studied in three crosses under both conditions, reflecting ambidirectional dominance. While the remaining traits have significant (F^) values positive or negative for flag leaf area, number of days to maturity under normal condition in the first cross , main spike length under stress in the first cross, plant height, main spike length , number of kernels per main spike and main spike yield under normal condition in the second cross, no. of days to maturity, plant height, number of spikelets per main spike under stress in the second cross and main spike length and number of kernels per main spike under normal in the third cross revealing that, the dominance was unidirectional . Evidences of water stress tolerance, The results of Tolerance index (TOL), Yield reduction ratio (YR) and Drought Susceptibility index (DSI) indicated that larger values represent relatively more sensitivity to water stress, thus a smaller values of (TOL), (YR) and (DSI) were favored. Selection based on (TOL), (YR) and (DSI) favors genotypes with low yield potential under non stress treatment and high yield under stress treatment. Genetic correlation Partitioning of the total genetic correlation to its components of epistasis, additive and dominance were computed among TTC families. The results of genetic correlations revealed that additive and dominance gene effects controlling yield and its components were significantly associated with each other, suggesting common genetic pool, pleiotropy or linkage. Prediction of superior recombinants The results of prediction revealing that the highest proportions of recombinants which outperform parental range for number of spikes per plant (44.43%), followed by number of days to maturity (40.51%), grain yield per plant (34.45%), flag leaf area (28.43%) and plant height (17.10%) under normal condition. While, under stress recorded for main spike yield (98.52%) followed by grain yield per plant (33.72%), 1000-kernels weight (32.22%), flag leaf area (26.10%) and number of days to maturity (13.13%). In TTC2 recorded for plant height (42.07%) followed by number of days to heading (36.69%), flag leaf area (24.19%), 1000-kernels weight (6.55%) and grain yield per plant (4.74%) under normal. While, under stress recorded by number of days to heading (48.40%) 173 Summary followed by grain yield per plant (36.31%), flag leaf area (10.56%), number of days to maturity (9.60%) and plant height (9.13%). In TTC3 recorded by 1000-kernels weight (98.52%) followed by grain yield per plant (35.56%), number of days to heading (30.85%), number of spikes per plant (29.46%) and flag leaf area (28.77%) under normal. While, under stress recoded by number of days to heading (34.82%) followed by number of spikes per plant (32.63%), main spike yield (31.56%), number of days to maturity (28.77%) and flag leaf area (25.46%). Estimate the interrelationships among TTC families The principal components analysis in TTC1 showed that first four PC axes accounted for about 70.5 % and 72.8 % of the total variation under normal and stress conditions respectively. While in TTC2 showed that first four PC axes accounted for about 80.0 % and 78.1 % of the total variation under normal and stress conditions respectively and in TTC3 showed that first four PC axes accounted for about 73.2 % and 69.8 % of the total variation under normal and stress conditions respectively. The genetic divergence based on Euclidean distances among TTC families for crosses studied under both condition is graphically illustrated as dendrogram (tree diagram) according to hierarchical clustering analysis based on the relative dissimilarity among 60 families and 11 agronomic traits. The sixty families of TTC1 were grouped into thirteen major clusters under normal condition, while it clustered into fifteen major groups under stress condition. In TTC2 their grouped into sixteen major clusters under normal condition, while it clustered into eight major clusters under stress. In TTC3 their grouped into 174 Summary nineteen major clusters under normal condition, while it clustered into fifteen major clusters under stress. The promising families selected from diverse clusters in TTC1 under normal condition detected that cluster 5 contained two families (4 and 56); cluster 4 contained one family (60) and cluster 6 contained one family (3). While under stress condition cluster 10 contained two families (56 and 57); cluster 9 contained one family (3) and cluster 8 contained four families (17, 58, 59 and 60). The promising families selected from diverse clusters in TTC2 under normal condition detected that cluster 12 contained one family (26); cluster 8 contained one family (48) and cluster 7 contained one family (50). While the under stress condition are cluster 7 contained six families (32,36,37,38,39 and 40); cluster 1 contained twelve families (5,14,17,21,23,24,42,43,44,49,51and 52) and cluster 6 contained six families (26,28,29,30,31 and 33). The promising families selected from diverse clusters in TTC3 under normal condition detected that cluster 5 contained one family (48); cluster 13 contained one family (53) and cluster 10 contained one family (55). While the under stress condition are cluster 12 contained one family (48); cluster 3 contained five families (12,15,16,25 and 26) and cluster 6 contained seven families (1,3,5,6,14,18 and 19). The results obtained recorded that selected families (3, 56 and 60) in TTC1, family (26) in TTC2 and family (48) in TTC3 had good performance and superiority under normal and stress conditions indicated that these families have broad adaptability. These results confirm the superiority of F2 plants, thus selection of these families might produce promising lines that could be used as a good genotypes in wheat breeding program.