الفهرس 
INTRODUCTIONOnly 14 pages are availabe for public view
 |  | from | 180 |  |  |
| | | from | 180 | | |
Abstract
The current work was carried out at Fac. Agric. Farm, Assiut University during the three summer seasons of 2016-2018. The aim of the study was to improve tolerance and /or resistance of sesame (Sesamum indicum L.) to Macrophomina phaseolina (Tassi) Goid and Fusarium oxysporum f.sp. sesami (Zap) Cast sesami root rot and wilt diseases. The experimental site was artificially infected each year by the two pathogens separately. Two cycles of pedigree selection for eight traits under artificial infection of each pathogen were performed in the field to increase tolerance and/or resistance of sesame. One sesame population stemmed from the cross Shandaweel 3/ Introduction 688 in the F3- generation was used.
The results could be summarized as the following
A. selection for resistance to Macrophomina phaseolina (Tassi) Goid
Description of the base populations; season of 2016 (F3- generation)
1. Families mean squares (185 families) were significant (p≤0.01) for all the studied traits in the population except capsule length and capsule width.
2. Genotypic coefficient of variability was 4.55, 15.84, 16.09, 16.65, 81.58, 39.15, 7.08, 12.66 and 10.27 for days to 50% flowering, plant height, height to the first capsule, length of the fruiting zone, capsules/ plants, seed yield/ plant, seed oil percentage, infection percentage at seedling stage and infection percentage at harvest diseased in the F3-families.
3. Heritability estimates in broad sense in the population were high and ranged from 73.02 % for CW to 99.26% for oil%.
4. The maximum and minimum values in the F3-generation survival families indicated to the feasibility of selection for 50% flowering, PH, HFC. LFZ, NC/P, SY/P, 1000SW and oil%.
5. Family means ranged in days to 50% flowering from 54.00 to 76.667, plant height (from 98.33 to 208.33 cm), height to the first capsule (from 15.00 to 90.00cm to), length of the fruiting zone (from 65.0 to 165.00 cm.), 1000-seed weight (from 4.35 to 6.57g), number of capsules/plant (from 63.50 to 1120.33), seed yield/plant (from 6.79 to 211.68 g), seed oil percentage (from 37.00 to 65.67%).
6. Days to 50% flowering showed weak and negligible correlations with all traits, negative with PH, HCF, NC/P, SY/P, CL, and CW at both of genotypic and phenotypic levels, while it was positive with LFZ, 1000SW and infection% either in seedling stage or at harvest.
7. Plant height showed positive correlations with HFC, LFZ, SY/P, and NC/P at genotypic and phenotypic levels and negative correlations with infection%.
8. Number of capsules /plant gave sizeable genotypic (0.5154) and phenotypic (0.4853) correlations with SY/P.
9. Height to first capsule gave negative weak correlations with LFZ and SY/P, and positive with PH at genotypic and phenotypic levels.
Evaluation of the second cycle selection, season 2018 under artificial infection by M. phaseolina
A-Pedigree selection for days to 50% flowering
1. Mean squares of the selection criterion days to 50% flowering and the other studied traits were significant (p≤ 0.01) after two cycles selection.
2. Heritability in broad sense of days to 50% flowering was very high (97.85%) and unreliable because the evaluation for one season and one location inflated families mean squares by the confounding effects of the interaction of families, years and locations. Otherwise, heritability in narrow sense as estimated by regression of F3 on F2 (0.0085) was very low and increased from F5 on F4 to 0.5471.
3. After the second cycle (F5) selection, the direct genetic response in days to 50% flowering was significant from the mid-parent (-1.74%) followed by positive significant (p≤0.05) correlated gain in HFC of 8.33% from the mid-parent and 13.04 % from the better parent. Positive correlated gains from the mid-parent were also found for NC/P, SY/P and CW. Otherwise; significant (p≤0.05 or p≤0.01) negative correlated gains were recorded for 1000SW, oil% and infection% from both mid- and better-parent.
B-Pedigree selection for days to 50% flowering restricted by seed yield/ plant.
1- Mean squares of the selection criterion (days to 50% flowering restricted by seed yield/ plant) and the other traits was significant (p≤0.05 or p≤0.01).
2- The GCV and PCV of days to 50 flowering after the second cycle were 7.10 and 7.17%, which could be considered moderate variability enough for further cycle of selection. Heritability in broad sense was very high (97.86%). Furthermore, heritability in narrow sense was high and accounted 0.7491 from regression of F4/F3 and increased to 0.9550 from regression of F5/F4- generation.
3- The direct observed genetic gain for 50% flowering was not significant either from the mid- or better-parent. Otherwise, it was significant (p≤0.01) and reached 39.76% from the mid-parent and 34.43% from the better parent. The observed correlated gains from the mid-parent was positive and significant (p≤0.01) for plant height (17.08%), HFC (13.54%), LFZ (19.44%), NC/P (29.46%) and infection percentage at seedling stage, and negative significant (p≤0.05 or p≤0.01) for 1000SW (-7.39%), CL (-2.77%), oil (-1.75%) and infection percentage at harvest.
4- The observed genetic correlated gain from the better parent was positive and significant (p≤0.01) for HFC (18.48%) and infection at seedling (10.09%), and negative for CL, CW, oil, infection% at harvest. It could be noticed that restriction of days to flowering with yield increased PH, LFZ, and NC/P.
C- Pedigree selection for seed yield/ plant
1. Mean squares of the selection criterion SY/P and the other correlated traits were significant (p≤0.01). The GCV and PCV estimates of seed yield/plant after two cycles of selection (Table 4) were high and reached 45.01 and 45.21%, respectively.
2. Heritability in broad sense was very high and unreliable (99.13%), however heritability estimates expressed as regression of F4 on F3 increased from 0.0379 towards homozygosity to 0.3623 for F5/F4.
3. Mean of seed yield/ plant ranged from 9.10 to 39.25g with an average of 24.60 g compared to 15.16 and 16.42g for the two parents.
4. The direct significant (p≤0.01) observed genetic gain in SY/P reached 55.80% from the mid-parent and 49.86% from the better parent.
5. The increase in yield followed by significant decrease in correlated gains in days to 50% flowering (-6.23 and -5.41%), in HFC (-10.42 and -6.52%), in 1000SW (-6.13 and -8.86%), CL (-2.7 and -2.28%), oil% (4.68 and -7.91%) and infection % at harvest (-3.31 and -1.56) from the mid and better parent, respectively.
6. Pedigree selection for SY/P accompanied with favorable decrease in days to 50%flowering, HFC, and infection% at harvest, and adversely affected NC/P, Oil%, 1000SW, and CL. It is the main drawback of single trait selection.
D- Pedigree selection for seed oil percentage
1. Means squares of seed oil percentage and correlated traits are significant (p≤0.05 or p≤0.01).
2. After two cycles of selection, oil% ranged in the selected families from 60.67 to 64.0% with an average of 63.0% compared to 55.00 and 59.00% for the two parents.
3. The genetic direct observed gain in oil% was significant (p≤ 0.01) from the mid-parent (10.53%) and from the better parent (6.78%).
4. selection for oil% accompanied with significant delay in flowering, increase in HFC, decrease in 1000SW, CL and CW. Otherwise, selection for oil% showed significant observed correlated gain from the mid and better parent of 36.67 and 30.16% for PH, 36.11 and 22.50% for LFZ, 38.36 and 13.64% for NC/P, 45.67 and 40.12% for SY/P, -9.87 and -3.16% for infection % at seedling stage, and -15.97 and -14.45% for infection% at harvest, respectively.
E- Pedigree selection for seed oil percentage restricted by seed yield/ plant.
1. Means squares of seed oil%, SY/P and correlated traits are significant (p≤0.01).
2. The GCV (3.28%) and PCV (3.40%) in oil% was very low.
3. Heritability estimates in broad sense was high (93.13%), however, narrow sense heritability was 0.6493 in the F4 and 0.0990 in the F5- generation. Such low variability in the F5 may not allow further cycles of selection.
4. Mean oil% of the selected families ranged from 59.00 to 60.67% with an average of 59.92%, and SY/P ranged from 19.37 to 31.90% with an average of 26.87%.
5. After two cycles of pedigree selection for oil% restricted by seed yield the observed direct genetic gain in oil% from mid- parent was ( 5.12%**), and 70.16%** and 63.67%** for SY/P from the mid- and better parent, respectively, accompanied by significant (p≤0.01) increase of 34.58 and 28.18% for PH,4.79 and 51.09% for HFC, 27.78 and 15.00% for LFZ, 69.54 and 39.24% for NC/P and decrease infection% of -8.48 and -1.67% at seedling stage and -14.53 and -12.98% at harvest from both the mid- and better parent, respectively. selection caused significant (p≤0.01) decrease in 1000SW, CL, and CW.
F- Pedigree selection for length of the fruiting zone
1. Means squares of LFZ and correlated traits were significant (p≤0.01).
2. The GCV and PCV in LFZ were reduced from 16.55 and 14.51% in the F3 to 13.98 and 14.12% in the F5- generation, respectively.
3. Heritability estimates in broad sense was high (97.99%), however, narrow sense heritability was 0.4920 in the F4 and 0.2824 in the F5- generation.
4. Mean LFZ of the selected families ranged from 61.67 to 81.67cm with an average of 72.92cm.
5. The direct observed genetic gain was 21.53%** from the mid-parent and 9.38%** from the better parent. The favorable positive significant correlated gains were 13.33%**and 7.94%** for PH, 9.73%** and 5.55% for SY/P, and favorable decrease in infection% of -9.00** and -7.36**from the mid-and better parent, respectively. Otherwise, selection for LFZ significantly (p≤0.01) decreased 1000SW, CL, CW, and oil%.
G- Pedigree selection for length of the fruiting zone restricted by height to first capsule
1- Mean squares of the selection criteria LFZ and HFC and the other correlated traits are significant (p≤0.01).
2- The GCV and PCV in LFZ were 18.79 and 18.95%, respectively in the F5 and could be considered high for further cycles of selection and were slightly higher than those in the base population.
3- Heritability estimates in broad sense for LFZ was high (98.30%), however, narrow sense heritability was 0.2710 in the F4/F3 and slightly increased towards homozygosity to 0.3723 in the F5/F4 generation.
4- Mean LFZ of the selected families (Table 17) ranged from 71.67 to 100.0cm with an average of 82.50cm, and HFC ranged from 35.00 to 55.00cm with an average of 46.67cm.
5- After two cycles of pedigree selection for LFZ restricted by HCF the observed direct genetic gain in LFZ was 37.50**, and 23.75**, 16.67** and 21.74**for HFC from the mid- and better parent, respectively, accompanied by significant increase of 29.17** and 23.02**% for PH and decrease infection% of -4.74**and 2.35**at seedling stage and -11.59** and -9.99** at harvest from both the mid- and better parent, respectively. selection caused significant (p≤0.01) decrease in 1000SW, CL, CW, and oil%.
H- Pedigree selection for 1000-seed weight
1. Mean squares of the selection criteria 1000SW was significant (p≤0.05), while the other correlated traits were significant (p≤0.01).
2. The GCV (2.04%) and PCV (2.42%) in the F5 were low.
3. Heritability estimates in broad sense for 1000SW was high (71.15%), however, narrow sense heritability was 0.2710 in the F4/F3 and slightly increased towards homozygosity to 0.3723 in the F5/F4 generation.
4. Mean 1000SW of the selected families ranged from 5.30 to 5.54g with an average of 5.43g.
5. After two cycles of pedigree selection for 1000SW the observed direct genetic gain in 1000SW was 0.52**, and -2.40** from the mid- and better parent, respectively, accompanied by significant increase of 44.17 and 7.30**% for PH,4.78** and 5.70**in 50% flowering, 29.17** and 34.78**for HFC, 54.17** and 38.75** for LFZ,113.42** and75.28** for NC/P,62.14** and 55.96**for SY/P, and significant decrease of -4.47** and -4.06** for CL , -5.77** and -6.25** for CW, and -4.14** and 7.49%** for infection % at harvest from the mid and better parent, respectively.
Comparison between the eight selection criteria under artificial infection by Macrophomina phaseolina
1. Two cycles of selection for days to 50% flowering (Table 20) improve earliness from the mid-parent (-1.74*) followed by significantly improve in the correlated traits, HFC (8.33%**), NC/P (20.65%**), SY/P (6.55%**), CW (4.92%**) and INF% at harvest (-6.87%). negative significant decrease was obtained in 1000SW (-7.73%**), oil% (-1.75%*), and Cl (-2.48%**). Otherwise, the best improvement in days to 50% flowering (-6.23 %**) was obtained from selection for SY/P because of the negative genotypic correlation between them in the base population.
2. Selection for days to 50% flowering restricted by SY/P, reduce the improvement of 50% flowering and increase SY/P (39.76%**) compared to selection for days to 50% flowering per se. Mostly, all the characters showed positive correlation with SY/P in the base population were improved. The correlated gains measured from the mid-parent were 17.08%** (PH), 13.54%** (HFC), 19.44%**(LFZ), and 29.46%** (NC/P). Otherwise, negative reduction was obtained in 1000SW (-7.39%**), Cl (-2.77%**), and -1.75% (oil%). This means that incorporation of valuable trait with the selection criterion improved genetic gain.
3. The direct observed genetic gain from selection for SY/P was positive from both of mid-parent (55.80%**) and better parent (49.86%**). The observed correlated gain from the mid-parent was favorable in days to50% flowering (-6.23 %**), (18.33**) in plant height, -10.42%** in HFC, 37.50%** in LFZ. The other correlated traits gave negative unfavorable gains.
4. The direct observed genetic gain from selection for oil% was positive from both of mid-parent (10.53 %**) and better parent (6.78 %**). The observed correlated gain from the mid-parent was unfavorable in days to50% flowering (4.78**), and (37.50**) in HFC, and favorable in PH (36.67 %**), LFZ (36.11 %**), NC/P (38.36 %**) and INF% at harvest (-15.97 %**). The other correlated traits gave negative unfavorable gains.
5. Selection for oil% restricted by SY/P gave direct observed genetic gain from the better parent of 5.12%** (lower than selection for oil% per se), and 70.16%** for SY/P. Favorable correlated gains were obtained in PH (34.58 %**), LFZ (27.78 %**), NC/P (69.54 %**), and INF% at harvest (-14.53 %**). The other correlated traits gave unfavorable gains.
6. Selection for LFZ gave direct observed genetic gain from the mid- parent of 21.53%** and 9.38%** from the better parent. Favorable correlated gains from the mid-parent were obtained in PH (13.33 %**), SY/P (9.73 %**) and INF% at harvest (-9.00 %**). The other correlated traits gave unfavorable gains from selection.
7. Selection for LFZ restricted by HFC increased LFZ by 37.50%** from the mid-parent and 23.75%** from the better parent, followed by unfavorable increase in HFC. selection for LFZ only was better than selection for LFZ restricted by HFC.
8. Direct selection for 1000SW slightly increased seed weight, while it gave sizeable favorable correlated gains from the mid-parent in SY/P (62.34 %**), PH (44.17 %**), LFZ (54.17 %**), NC/P (113.42 %**) and INF% at harvest (-9.14 %**). The other correlated traits were adversely affected.
Conclusion
1. Single trait selection was an efficient method to improve selection criterion, but it had adverse effects on some correlated traits.
2. Selection for days to 50% flowering improved earliness, but it was better if it restricted by SY/P.
3. Selection for oil% restricted by yield gave the best improvement from the mid-parent in SY/P (70.16 %**) followed by selection for 1000SW (62.14 %**), SY/P per se (55.80 %**) and selection for days to 50% flowering restricted by yield (39.76 %**).
4. Inclusions of a trait as independent culling level improve the efficiency of selection. Therefore, selection index incorporating favorable trait is recommended.
5. The correlation coefficients among traits help the breeder to selection for few favorable traits.