TITLE: Identification of Potential Combiners Through Reciprocal
Selection in Segregating Generations of Cotton
(Gossypium hirsutum L.)
DESCIPLINE: Breeding and Genetics
AUTHORS: Dr.Somashekhar (Corresponding Author)
Research Associate
ARS,Dharwad Farm –Dharwad580 007
Phone08362447874
Emailsomuvd@rediffmail.com
Dr.S.A.Patil
Director, IARI,
New Delhi
Dr.S.S.Patil
Principal Scientist,and Head
ARS,Dharwad Farm –Dharwad580 007
Dr.Rajesh .Patil
Asst.proffessor(GPB)
ARS,Dharwad Farm –Dharwad580 007
Abstract: TwoF_{1} hybrids (RAHH102 and RAHH136), which are distinct, were identified through their predicted double cross performance as potential sources of inbred lines for hybrid cotton cultivars. F_{4} lines were derived from these crosses and utilized in a study on variability for combining ability. Sets of 26 lines each from the two crosses were crossed in a reciprocal fashion to the F_{1} parent as a tester for combining ability. The improvements in performance of F_{1} hybrids derived by crossing the best performing F_{4 }lines as predicted by their reciprocal test cross performance indicated that progress could be made for gain in combining ability through a breeding procedure similar to reciprocal recurrent selection in cross pollinated crops.
Cotton improvement programmes that concentrate on the development of hybrids have contributed to improving cotton productivity (Dagaonkar and Malkandale,1993). However, genetic gain in yield potential of hybrids appears to be approaching stagnation. In breeding programmes aimed at improving productivity of pure lines, i.e., not hybrids, variability is created and exploited by practicing selecting for yield during segregating generations. However, improving the performance of hybrids requires that scientists consider the combining ability of potential parental material (Patil and Patil, 2003). In cross pollinated crops like maize, hybrid breeding programmes are supplemented by regular systematic programmes aimed at improving combining ability (Patil and Pandit, 1991). Systematic attempts have not been practiced in cotton to create variability for combining ability, i.e., combining ability was not considered as a trait for improvement in hybrid breeding programmes. Reciprocal recurrent selection schemes for improving combining ability have been an integral part of hybrid breeding programmes in cross pollinated crops and such programmes have contributed to success of hybrid maize. The procedures of improving combining ability in cross pollinated species can not be followed in cotton without suitable modification. Hence, there is a need for defining procedures of improving combining ability to serve as a prerequisite in hybrid breeding in cotton. It is possible to recombine two, four or more lines (selected for combining ability) by single, double or multiple crossing or simulated intermating. Generally, individual plants in the F_{4} generation are selfed and crossed with a tester line to initiate the selection of improved inbreds. The objective of this research was to determine the combining ability among segregant F_{4} lines within two diverse populations.
Following analysis (data not shown) of a large set of single crosses, two single cross hybrids RAHH 102 (RAH10 RA100) and RAHH 136 (RAH20 RAH200) were selected for this study based on their predicted double cross performance(Patil and Patil, 2003). Plants within each population were advanced to the F_{4} generation. Twentysix, i.e., single plants, from each cross were selected randomly and crossed to the reciprocal F_{1} hybrid as the tester parent. Thus, F_{4} plants from RAHH 102 were crossed with RAHH 136 F_{1} and random F_{4} plants from RAHH 136 were crossed to RAHH 102 F_{1} to establish two sets of reciprocal hybrids. A field evaluation was conducted for two set of hybrids in Randomized complete block design with three replication having two rows of 5 m length. The whole experiment i.e., Crossing and evaluation of the hybrids, was conducted at Dharwad which receives an annual rainfall of 750 mm. Proper pest and disease control measure was taken to avoid economic loss during 200405 for the characterization of the combining ability status of the 26 F_{4} lines of each set. This was determined based on the performance of the crosses (seedcotton yield) compared with the F_{1} reciprocal testers. Each F_{4} line was assigned to one of four classes based on the overall mean of all crosses. These classes were: 1 (greater than (single cross parental_{ }mean + 1 sd unit)) , 2 (equal to the (single cross parental mean + 1 sd unit) , 3 (equal to the single cross mean – 1 sd unit) , and 4 (less than (single cross parental mean – 1 sd unit)) as suggested by Patil (1995). Thus, for lines of RAHH 102,, four classes of combining ability status were defined as E1, E2, E3, and E4, respectively. Similarly, F1, F2, F3, and F4 classes were defined representing the decreasing order to superiority of the crosses for the lines of RAHH 136.
Percent improvement in performance of reciprocal test cross hybrids over the mean of the reciprocal hybrid parents was calculated as an estimate of the combining ability of the each population. Hybrids were developed by crossing the best combining inbred lines in all possible combinations in the following season. Performance of these hybrids, F_{5} x F_{5}, were determined with three replications of hybrids in a RBD having three row x 5 m plots during 2006 in the same location . Hybrids were compared with a commercial cultivar, Bunny, and the two original single crosses.
Four lines of RAHH 102 lines with F_{1} RAHH 136 hybrids exceeded the mean of all of the 26 test cross hybrids by more than one standard deviation unit (Table 1). These were developed from lines R18 (102), R25 (102), R22 (102), and R26 (102) and yielded 2930, 2804, 2591, and 2582 kg ha^{1}, respectively, compared with the mean of all 26 hybrids of 2173 kg ha^{1}. Twelve additional lines of RAHH 102 with reciprocal hybrid yielded within one standard deviation above the overall mean while eight hybrids performed within one standard deviation below the mean, and only two R (102) hybrids yielded more than one standard deviation below the overall mean. The highest yielding hybrid, R18 (102) x RAHH 136 F_{1}, produced 49% more seedcotton than the average of two straight crosses, while R2 (102) x RAHH 136 F_{1 }yielded 45 % less seedcotton. Four lines of RAHH 136 with RAHH 102 F_{1 }as a tester produced hybrids that exceeded the mean of two straight crosses by more than one standard deviation unit (Table 2). These were R2 (136), R14 (136), R16 (136), and R15 (136), which yielded 2817, 2397, 2392, and 2389 kg ha^{1}, respectively, compared with the mean of two straight crosses of 1930 kg ha^{1}. Seventeen of these 26 hybrids were in the F2 or F3 categories, i.e., within one sd unit of the overall mean, while five hybrids yielded more than one sd below the overall mean. The superior F_{4} lines from RAHH 102(four) and from RAHH 136(four) that produced the superior reciprocal hybrids were subsequently crossed to produce all possible F_{1} hybrids (Table 3). All hybrids except R26 (102) x R15 (136) exceeded (p=0.05) the yield of Bunny, which was not different than the mean yield of RAHH 102 and RAHH 136 in this trial. The numerically highest yielding hybrid was R25 (102) x R2 (136) at 3593 kg ha^{1}, which was 51% higher than Bunny. In conclusion, in the reciprocal recurrent selection scheme proposed herein for cotton, the elite high combiner plants obtained from the reciprocal populations represent gain obtained from practicing selection for combining ability. We propose that such elite lines of the corresponding population can be intermated to start the next cycle of recurrent selection. In this study, the elite lines per se produced highly productive hybrids, indicating the magnitude of improving combining ability achieved through selection practiced in one cycle of reciprocal selection.
Table 1: Performance of reciprocal crosses derived from the lines RAHH102 crossed with tester RAHH136 F_{1} during 2004.
 Crosses  Seed cotton yield (kg ha^{1})  % Improvement  Ranking 
R18 (102)  R18 (102) x RAHH 136 F_{1}  2929.89  49.01  E1 
R25 (102)  R25 (102) x RAHH 136 F_{1}  2803.73  42.59  E1 
R22 (102)  R22 (102) x RAHH 136 F_{1}  2590.61  31.75  E1 
R26 (102)  R26 (102) x RAHH 136 F_{1}  2582.01  31.32  E1 
R5 (102)  R5 (102) x RAHH 136 F_{1}  2488.1  26.54  E2 
R8 (102)  R8 (102) x RAHH 136 F_{1}  2420.63  23.11  E2 
R11 (102)  R11 (102) x RAHH 136 F_{1}  2366.4  20.35  E2 
R21 (102)  R21 (102) x RAHH 136 F_{1}  2355.82  19.81  E2 
R17 (102)  R17 (102) x RAHH 136 F_{1}  2351.85  19.61  E2 
R7 (102)  R7 (102) x RAHH 136 F_{1}  2316.14  17.79  E2 
R20 (102)  R20 (102) x RAHH 136 F_{1}  2314.81  17.73  E2 
R14 (102)  R14 (102) x RAHH 136 F_{1}  2220.9  12.95  E2 
R15 (102)  R15 (102) x RAHH 136 F_{1}  2210.98  12.45  E2 
R4 (102)  R4 (102) x RAHH 136 F_{1}  2191.14  11.44  E2 
R13 (102)  R13 (102) x RAHH 136 F_{1}  2167.99  10.26  E2 
R9 (102)  R9 (102) x RAHH 136 F_{1}  2135.58  8.61  E2 
R1 (102)  R1 (102) x RAHH 136 F_{1}  2115.74  7.60  E3 
R16 (102)  R16 (102) x RAHH 136 F_{1}  2030.42  3.26  E3 
R10 (102)  R10 (102) x RAHH 136 F_{1}  2021.83  2.83  E3 
R12 (102)  R12 (102) x RAHH 136 F_{1}  2005.29  1.98  E3 
R23 (102)  R23 (102) x RAHH 136 F_{1}  1917.99  2.46  E3 
R19 (102)  R19 (102) x RAHH 136 F_{1}  1903.44  3.20  E3 
R3 (102)  R3 (102) x RAHH 136 F_{1}  1883.6  4.20  E3 
R6 (102)  R6 (102) x RAHH 136 F_{1}  1832.01  6.83  E3 
R24 (102)  R24 (102) x RAHH 136 F_{1}  1256.61  36.09  E4 
R2 (102)  R2 (102) x RAHH 136 F_{1}  1086.64  44.74  E4 
Mean  2173.083 

 
Standard deviation (sd)  402.6988 

 
Single cross parents 


 
 RAHH102  2140.21 


 RAHH 136  1792.33 


Mean of single crosses  1966.27 


Table 2: Performance of reciprocal crosses derived from the lines of RAHH136 crossed with tester RAHH102 (F_{1}) during 2004.
F_{4} line No..  Crosses  Seed cotton yield (kg ha^{1})  % Improvement  Ranking 
R2(136)  R2(136) x RAHH 102 F_{1}  2817.46  45.96  F1 
R14(136)  R14(136) x RAHH 102 F_{1}  2396.83  24.17  F1 
R16(136)  R16(136) x RAHH 102 F_{1}  2391.53  23.90  F1 
R15(136)  R15(136) x RAHH 102 F_{1}  2388.89  23.76  F1 
R21(136)  R21(136) x RAHH 102 F_{1}  2294.31  18.86  F2 
R20(136)  R20(136) x RAHH 102 F_{1}  2246.03  16.36  F2 
R12(136)  R12(136) x RAHH 102 F_{1}  2240.08  16.05  F2 
R11(136)  R11(136) x RAHH 102 F_{1}  2236.77  15.88  F2 
R5(136)  R5(136) x RAHH 102 F_{1}  2206.35  14.30  F2 
R19(136)  R19(136) x RAHH 102 F_{1}  2200.53  14.00  F2 
R23(136)  R23(136) x RAHH 102 F_{1}  2170.63  12.45  F2 
R6(136)  R6(136) x RAHH 102 F_{1}  2132.94  10.50  F2 
R1(136)  R1(136) x RAHH 102 F_{1}  2095.24  8.55  F2 
R8(136)  R8(136) x RAHH 102 F_{1}  2076.72  7.59  F2 
R3(136)  R3(136) x RAHH 102 F_{1}  2035.05  5.43  F2 
R17(136)  R17(136) x RAHH 102 F_{1}  1964.29  1.76  F3 
R18(136)  R18(136) x RAHH 102 F_{1}  1941.8  0.60  F3 
R7(136)  R7(136) x RAHH 102 F_{1}  1906.75  1.22  F3 
R10(136)  R10(136) x RAHH 102 F_{1}  1851.85  4.06  F3 
R9(136)  R9(136) x RAHH 102 F_{1}  1780.42  7.76  F3 
R26(136)  R26(136) x RAHH 102 F_{1}  1710.32  11.39  F3 
R13(136)  R13(136) x RAHH 102 F_{1}  1583.33  17.97  F4 
R24(136)  R24(136) x RAHH 102 F_{1}  1539.68  20.23  F4 
R4(136)  R4(136) x RAHH 102 F_{1}  1536.38  20.41  F4 
R22(136)  R22(136) x RAHH 102 F_{1}  1414.02  26.74  F4 
R25(136)  R25(136) x RAHH 102 F_{1}  1148.38  40.51  F4 
Mean  2011.792 

 
Standard deviation (sd)  367.1382 

 
Single cross parents 


 
 RAHH136  2110.21 


 RAHH 102  1750.33 


Mean of single cross parents  1930.27 


Table 3:Performance of elite crosses involving best combiners extracted from opposite population
SL No  Pedigrees of F_{1} hybrids from superior F_{4} lines  Seed cotton yield (kg ha^{1})  % Improvement over single cross parents  % Improvement over commercial check 
1  R25 (102) x R2(136)  3592.7  56.2  51.0 
2  R18 (102) x R2(136)  3412.9  48.4  43.5 
3  R22 (102) x R2(136)  3225.7  40.2  35.6 
4  R25 (102) x R14(136)  3183.4  38.4  33.8 
5  R25 (102) x R16(136)  3076.4  33.8  29.3 
6  R26 (102) x R14(136)  3051.5  32.7  28.3 
7  R18 (102) x R14(136)  3047.3  32.5  28.1 
8  R26 (102) x R2(136)  3027.3  31.6  27.3 
9  R22 (102) x R14(136)  2994.7  30.2  25.9 
10  R25 (102) x R15(136)  2945.8  28.1  23.8 
11  R26 (102) x R16(136)  2944.7  28.0  23.8 
12  R22 (102) x R16(136)  2896.4  25.9  21.8 
13  R18 (102) x R16(136)  2865.2  24.6  20.4 
14  R22 (102) x R15(136)  2856.6  24.2  20.1 
15  R18 (102) x R15(136)  2756.9  19.9  15.9 
16  R26 (102) x R15(136)  2692.7  17.1  13.2 
MEAN  3035.6  32.0  27.6  
Bunny  2378.9 

 
Mean of single cross parents  2300.0 

 
 CD @5%  267.1 


REFERENCE:
Dagaonkar, V.S., and J.D. Malkandale. 1993. Diallel analysis using fixed effect model in upland cotton. Journal of Indian Society for Cotton Improvement, 18: 5053.
Patil, S.S., and V.V. Pandit. 1991. Use of B R cross for improving combining ability in sorghum. Golden Jubilee Symposium on Research and Education : Current Trends and the Next Fifty Years, held at New Delhi, p. 397.
Patil, S. S. 1995. Report on the Work Done in the Area of Hybrid Research CIMMYT. The International maize and Wheat Improvement centre El Batan Mexico.
Patil, S. S., and S.A. Patil. 2003. Role of improving combining ability in increasing performance of cotton hybrids. Third World Cotton Research Conference, 913 March 2003, held at Cape Town South Africa, pp. 234238.