الفهرس 
INTRODUCTION
Description of the base populations (season 2009).Only 14 pages are availabe for public view
 |  | from | 194 |  |  |
| | | from | 194 | | |
Abstract
The present study was carried out at Assiut Univ. Exp. Farm during the three summer seasons of 2009-2011, the main objectives of this work were study the comparative importance of two breeding methods i.e., single trait and multiple traits selection (selection indices) using pedigree selection method to improve yielding ability and earliness in two F6 populations of Egyptian cotton at late planting and study the possibility of selecting superior lines adapted to late planting date (May), after early winter crops. The basic materials consisted of two F6- populations. Population I (pop.I) stemmed from the cross between ((Giza 80 x Pima S 6) x Giza 91) and Population II (pop.II) from the cross between (Giza 80 x Dandara). The two populations were subjected to pedigree selection for two cycles. The selection procedures were single trait selection for earliness index and lint yield/plant, in addition to eleven selection indices. The index selection proposed by Pesek and Baker (1970) was used with different combinations of characters. In season 2009 the two populations in the F6 generation were planted in two separate experiments on May 1st in a randomized complete block design. Population I was represented by fourty families, and population II by thirty-eight families. At the end of the season twenty families were selected for each selection procedure. Reselection in the F7-generation was applied in the two populations, and ten families were selected from each population for each selection procedure. In season 2011 (F8-generation) the ten selected families for each selection procedure were evaluated in two replicated experiment, one for each population. The results could be summarized as follows:
A- Description of the base populations, season 2009
1. Mean squares of the families for all the studied traits of the two populations were highly significant under late planting condition.
2. The population means were 39.35 and 40.92g for seed cotton yield/plant, 14.98 and 15.97g for lint yield/plant, 38.10 and 38.94 for lint percentage, 2.50 and 2.72g for boll weight, 15.74 and 15.28 for number of bolls/plant, 10.27 and 9.36g for seed index, 6.33 and 5.98g for lint index, 0.82 and 0.85 for earliness index and 57.70 and 63.75 for days to first flower for population I and II, respectively.
3. High gcv estimates were observed in seed cotton yield/plant (23.61 and 26.59%), lint yield/plant (23.70 and 27.60%) and number of bolls/plant (21.61 and 29.78%) in populations I and II, respectively.
4. Estimates of broad sense heritability were high and unreliable, and accounted for 94.17 and 96.30% for seed cotton yield/plant, 93.78 and 94.88% for lint yield/plant, 74.16 and 66.00% for lint percentage, 98.46 and 99.02% for boll weight, 92.89 and 96.60% for number of bolls/plant, 95.00 and 81.63% for seed index, 86.21 and 75.16% for lint index, 98.35 and 98.11% for earliness index and 77.07 and 80.30% for days to first flower for population I and II, respectively.
5. High estimates of heritability in broad sense coupled with high estimates of gcv gave unreliable expected genetic advance in percentage of the mean of 23.61 and 26.59% for seed cotton yield/plant, 23.70 and 27.60% for lint yield/plant, 21.61 and 29.78% for number of bolls/plant, 11.18 and 10.41% for lint index and 13.35 and 14.69% for earliness index for population I and II, respectively.
B- Pedigree selection for earliness index:
1. Mean squares of the selected families for earliness index and the other traits were significant (P<0.01) after two cycles of selection in the two populations.
2. The pcv and gcv of earliness index were 8.28 and 8.12% for pop. I, compared to 10.68 and 10.60% for pop. II, respectively.
3. The realized heritability of earliness index in pop. I was 0.2500 for cycle 1 and 0.4550% for cycle 2. Likewise, narrow sense heritability as calculated from regression of offspring on parents was 0.1514 for cycle 1 and 0.1127% for cycle 2.
4. In pop. II, the realized heritability estimates and parent-offspring regression were low compared to the very high estimates (98.58%) of broad sense heritability of earliness index.
5. Mean earliness index ranged from 0.77 to 0.96 with an average of 0.89 for pop. I, and from 0.62 to 0.94 with an average of 0.82 for pop. II.
6. The direct observed gain from the unselected bulk sample in pop. I was significant (P<0.01) for seven families, the observed gain from the bulk sample in pop. II indicated that seven families exceeded significantly (P<0.05 to P<0.01) the bulk sample in earliness index.
7. In pop I, the correlated gain in seed cotton yield/plant as calculated from the bulk sample was significant (P<0.01) for three families i.e., family No. 2, No. 145 and No. 175, and ranged from 12.25 to 70.41%, two families from them showed significant (P<0.01) observed gain from the better parent Giza 91. Also, three families for lint yield/plant, three and six for lint percentage, three for boll weight, two and one for number of bolls/plant, seven and five for seed index, three and six for lint index and four and three families for days to first flower showed significant correlated observed gains from the bulk ample and the better parent; respectively.
8. In pop II, the significant average direct gain in earliness index from the bulk sample (3.80%) accompanied with significant (p<0.05 to p<0.01) correlated gains in seed cotton yield/plant (15.94%), lint yield/plant (8.52%) and days to first flower (-6.01) from the bulk sample. Furthermore, significant (p<0.01) correlated gains from the better were observed for seed cotton yield/plant (7.89%), lint yield/plant (12.90%), lint percentage (5.34%) and lint index (14.59%). However, not all the families which were superior in earliness index showed increase in the correlated traits; especially yield. Only two families; No. 11 and No. 151 could be considered promising, and showed significant direct gain in earliness index and most of the correlated traits.
C- Pedigree selection for lint yield/plant:
1. The families mean squares of the selection criterion; lint yield/plant and the other studied traits were significant (P<0.01) in both of pop. I and pop. II.
2. The gcv and pcv% were high in the two populations and accounted for 22.59 and 22.91% for pop. I and 21.50 and 21.81% for pop. II, respectively.
3. The realized heritability estimates for lint yield/plant were 0.3140 and 0.4128 for cycle 1 and cycle 2 in population I; respectively. Also, heritability estimates from the parent-offspring regression were 0.1013 and 0.1528 for cycle 1 and cycle 2 of selection, respectively. Likewise, in population II, the estimates of realized heritability were 0.2771 for cycle 1, and 0.3970 for cycle 2, compared to parent offspring regression of 0.1137 and 0.1459 for the two respective cycles of selection.
4. Mean lint yield/plant in pop. I ranged from 7.87 for family No. 82 to 20.77 for family No. 184 with an average of 15.48 g, while it ranged from 11.89 for family No. 78 to 22.63 for family No. 73 with an average of 16.07g in pop. II.
5. The average direct observed gain from the bulk sample was significant (P<0.05) in pop. I and reached 6.68%. Six out of the 10 selected families showed significant (P<0.01 – P<0.05) direct gain ranged 16.20% for family No. 140 to 43.14% for family No. 184. The direct significant (p<0.01) observed gain in lint yield/plant from the better parent; Giza 91 ranged from 17.49 for family No. 140 to 44.74% for family No. 184 with a significant (p<0.01) average of 7.82%. Six out of the 10 selected families showed significant (P<0.01) direct gain from the better parent in pop. I. The most superior promising families in pop. I, were families No. 128 and No. 184.
6. In pop. II, four selected families which showed significant (P<0.01) direct gain in lint yield/plant ranged from 11.29 for family No. 16 to 48.49 for family No. 73 with an average of 5.45% from the bulk sample. Furthermore, five families showed significant (P<0.01) direct gain in pop. II from the better parent Giza 80, which ranged from 9.49 for family No. 182 to 54.47 for family No. 73 with an average of 9.69%.
7. Two cycles of pedigree selection for lint yield/plant in pop. I accompanied with favorable significant (P<0.05) correlated gains of 6.31 for seed cotton yield/plant and 2.53% for earliness index from the bulk sample. Likewise, the correlated gain from the better parent was significant (P<0.01) for lint percentage (7.51%), lint index (15.62%) and earliness index (6.17%).
8. In pop II, selection for lint yield/plant for two cycles showed significant (P<0.01) increase in seed cotton yield/plant of 19.37, boll weight of 23.64, seed index of 8.54 and earliness index of 4.00% from the bulk sample, and decrease in days to first flower to -5.82%. Four of the 10 selected families were superior, and showed significant or highly significant correlated gains for seed cotton yield/plant, earliness index and most of the other traits. These superior families; No. 73, No. 151, No. 160 and No. 182. However, only two families (No. 151 and No. 160) showed significant (P<0.01) correlated gains from the better parent for seed cotton yield/plant, lint percentage, lint index and earliness index. These two families could be considered promising superior families.
D- Selection indices.
The analysis of variance of the selected families of the eleven indices indicate significant (p<0.01) mean squares of families in all indices in the two populations except for families mean squares of boll weight in pop I which was significant (p<0.05).
D.1. The first population (pop I).
1. The gcv ranged from 18.0 to 27.35% for seed cotton yield/plant, 17.98 to 27.35% for lint yield/plant, 5.95 to 7.81% for lint percentage, 5.79 to 11.07% for boll weight, 15.90 to 30.79% for number of bolls/plant, 3.86 to 8.54% for seed index, 9.70 to 13.39% for lint index, 7.07 to 8.75% for earliness index and from 2.38 to 4.29% for days to first flower.
2. Narrow sense heritability estimates as calculated by parent-offspring regression for all traits of the eleven indices were low in general (<0.4), and intermediate for some cases.
3. Selection index 1 resulted in four superior promising families; No. 2, 85, 128 and No. 175. These families show superiority in yield and earliness index. For example, family No. 2 significant (p<0.05- <0.01) observed gain from the better parent of 61.54, 52.06, 23.94, 28.50, 7.87, 17.28 and -5.83% for seed cotton yield/plant, lint yield/plant, boll weight, number of bolls/plant, earliness index and days to first flower.
4. Both of index 2 and index 4 failed to improve yield and earliness except for family No. 175.
5. Selection index 3 resulted in three superior families No. 2, 58 and No. 128 which significant (p<0.01) out-yielded both of the bulk sample and the better parent. Family No. 2 showed significant (p<0.01) observed gains from the better parent of 61.54, 52.06, 23.94, 28.50, 7.87, 17.28 and -5.83% for seed cotton yield/plant, lint yield/plant, boll weight, number of bolls/plant, seed index, earliness index and days to first flower; respectively. Families No. 58 and No. 128 showed favorable significant (p<0.01) observed gains for seed cotton yield/plant, lint yield/plant and earliness index.
6. Selection index 5 resulted in two superior promising families; No. 128 and No. 175, which showed significant observed gains in yields and earliness index from both of the bulk sample and the better parent. The observed gains from the better parent were significant (p<0.01) and accounted for 26.83, 49.97, 18.16, 9.87, 43.17 and 13.58% for seed cotton yield/plant, lint yield/plant, lint percentage, seed index, lint index and earliness index; respectively.
7. According to index 6, only two selected families; No. 92 and No. 175 showed significant observed gain for seed cotton and lint yields and earliness index from the better parent. Family No. 175 out-yielded significantly (p<0.05-<0.01) the better parent by 26.83, 49.97, 18.16, 14.01, 9.87, 43.17 and 13.58% for seed cotton yield/plant, lint yield/plant, lint percentage, number of bolls/plant, seed index, lint index and earliness index; respectively.
8. Selection index 7 resulted in two superior promising families; No. 35 and No. 58 showed significant (p<0.01) observed gain in yields and earliness index from both of the bulk sample and the better parent. Family No. 92 showed significant (p<0.01) observed gain from the better parent in yields and earliness index. Family No. 35 showed significant (p<0.01) observed gain from the better parent of 42.84, 42.51, 32.43 and 17.28% for seed cotton yield/plant, lint yield/plant, boll weight and earliness index, respectively.
9. Selection index 8 resulted in five selected families; No. 35, 58, 85, 92 and No. 128 were the superior families. The best promising family was No. 35, which gave significant (p<0.01) observed gain from the better parent of 42.84, 42.51, 32.43 and 17.28% for seed cotton yield/plant, lint yield/plant, boll weight and earliness index; respectively.
10. Selection index 9 showed that family No. 175 was the best family and showed significant observed gains for yields, earliness index and most of the other traits, which accounted for 26.83, 49.97, 18.16, 14.01, 9.87, 43.17 and 13.58% from the better parent for seed cotton yield/plant, lint yield/plant, lint percentage, number of bolls/plant, seed index, lint index and earliness index; respectively.
11. The two superior families (No. 92 and No. 175) selected by index9 were also selected in index 10 and index11, and showed significant observed gain from the better parent in yields and earliness index.
D.2. The second population (pop II).
1. The range of genotypic coefficient of variability was high for seed cotton yield/plant (15.94-22.26%), for lint yield/plant (19.34-23.29%), number of bolls/plant (13.43-24.19%), lint index (10.71-14.64%) and earliness index (7.44-12.23), while it was moderate for the other traits except for days to first flower which was narrow (3.29-4.29%).
2. The parent-offspring regression indicated low narrow sense heritability for all traits in most of the selection indices.
3. Index 1 resulted in three superior early high yielding families No.138, No.151 and No.179. The best one; No.151 showed significant (p<0.05-<0.01) observed gain from the bulk sample of 32.31, 36.09, 29.57, 5.30, 29.21, 16.46 and -5.28% for seed cotton yield/plant, lint yield/plant, boll weight, seed index, lint index, earliness index and days to first flower; respectively.
4. Selection index 2 and index 4 resulted in four superior families; No. 11, 47, 138 and No.151 showed significant observed gain from the bulk sample and the better parent for yields, earliness index and some traits. The best superior family No. 47 showed significant (p<0.01) observed gain from the better parent of 41.61, 52.01, 7.21, 14.71, 17.29, 19.25, 37.21 and 6.02% for seed cotton yield/plant, lint yield/plant, lint percentage, boll weight, number of bolls/plant, seed index, lint index and earliness index; respectively.
5. Selection index 3 resulted in three superior promising families; No.47, 138 and No.179 in both yields and earliness index. The best one; family No. 47 showed significant (p<0.01) observed gain from the better parent for all traits except days to first flower, which accounted for 41.61% for seed cotton yield/plant, 52.01% for lint yield/plant, 7.21% for lint percentage, 14.71% for boll weight, 17.29% for number of bolls/plant, 19.25% for seed index, 37.21% for lint index and 6.02% for earliness index.
6. Selection index 5 detected four superior promising families; No.47, 138, 160 and No.179 both in yields and earliness index. The best family No.138 which showed significant observed gain from the bulk sample of 24.99, 19.75, 11.67, 13.79, 5.06 and -7.38% for seed cotton yield/plant, lint yield/plant, boll weight, lint index, earliness index and days to first flower, respectively.
7. Selection index 6 indicated to four superior families; No.11, 47, 138 and No.151. The best superior and promising family No.47 showed significant (p<0.01) observed gain from the better parent for all traits except days to first flower, followed by family No.151 which showed significant observed gain (p<0.01) from the better parent of 23.12, 41.57, 14.90, 22.43, 34.67 and 10.84% for seed cotton yield/plant, lint yield/plant, lint percentage, boll weight, lint index and earliness index.
8. Selection index 7 detected four superior families; No. 47, 73, 138 and No.160. The four families showed significant (p<0.01) observed gain in yields from both of the bulk sample and the better parent, and significant observed gain for earliness index from the bulk sample. The best family No.73 showed significant observed gains from the bulk sample for yields/plant, boll weight, number of bolls/plant, seed index and earliness index.
9. Selection index 8 detected four superior families; No. 47, 73, 138 and No.179. These four families showed significant (p<0.05-<0.01) observed gain from both of the bulk sample and the better parent in yields and earliness index except for family No.138 from the better parent; respect to earliness index.
10. Both of selection indices 9, 10 and 11 detected five superior families; No.11, 47, 138, 151 and No.179, which showed significant observed gain from the bulk sample in yields and earliness index and some traits. The best selected family No.47 showed significant (p<0.01) observed gain from the better parent for eight traits. The second promising family No.151 showed significant (p<0.01) observed gains from the better parent accounted for 23.12, 41.57, 14.90, 22.43, 34.67 and 10.84% for seed cotton yield/plant, lint yield/plant, lint percentage, boll weight, lint index and earliness index; respectively.
E. The relative merits of the selection procedures.
1. In population I; all the selection procedures, either single trait selection (selection for earliness index and selection for lint yield/plant) or multiple trait selection; selection index, detected nine superior promising families; No.2, 35, 58, 85, 92, 128, 147, 175 and No.184. The superiority of a selected family depends mainly upon its performance in lint yield/plant and earliness index under late planting. All selection procedures ranked from 1 to 9 according to their efficiency to improve both lint yield/plant and earliness index. Single trait selection for earliness index ranked the fourth, while selection for lint yield/plant per se was inferior and ranked the eighth. Therefore, it could be concluded that selection index in general was better than single trait selection in improving lint yield/plant and earliness index in pop I.
2. In population II; all the selection procedures resulted in seven superior promising families; No.11, 47, 73, 138, 151, 160 and 179. Firstly, selection index 2 and 4 shared the same families, and index 9, 10 and 11 shared the same families. The efficiency of the different selection procedures in detecting the superior families differed greatly from their efficiency in pop I. single trait selection was inferior respect to selection index in improving lint yield/plant and earliness index. Selection for earliness index ranked the sixth, and selection for lint yield/plant ranked the ninth.