Wheat is the most important food cereal of our country. The attainment of maximum crop yield is an important objective in most of the breeding programmes. Our wheat production has reached to the level of self sufficiency but still per acre yield is less than the other wheat growing countries. Previously exploitation of heterotic effects for more grain yield were largely attributed to cross pollinated crops, but evidence is now available to confirm presence of such effects in self pollinated crops like wheat. Briggle and Sajnani suggested the possibility of heterotic effects in wheat as well.
Khan and Khan reported that maximum heterosis was obtained for tillers per plant (31.91%) followed by grain yield per plant (19.41%), 1000-grain weight (17.32%), number of grains per spike (11.37%) and plant height (5.23%). Chowdhry et al. observed heterosis for traits like plant height, number of tillers per plant, 1000-grain weight and grain yield per plant. Akhter et al. studied heterosis for characters like plant height and grain yield per plant.
Yield of grain is very important for the farmers and we can get good combinations through heterosis which can provide high grain yield. Standing ability is another important character which we can improve through this technique. Lodging resistance and short stature varieties developed through heterosis increase the grain yield by reducing the risk of lodging due of strong winds. Similarly we can improve characters like disease resistance, insect resistance, gain quality and early maturity through heterosis.
MATERIALS AND METHODS
The research work was carried out in the experimental area of the Department of Plant Breeding and Genetics, University of Agriculture, Faisalabad. Five wheat varieties/lines viz., PBW 222, LU26S, Uqab 2000, 8952 and 8961 were crossed in a diallel fashion. The F1s including direct crosses and their reciprocals were attempted during crop season 2001-2002.
These crosses were field planted on November, 8, 2003 to evaluate their performance as compared to their parents.
The seeds were sown by using dibble. Two seeds per hole were planted, later thinned to one seed per site, keeping plant to plant and line to line distance i.e. 15 and 30 cm, respectively with three replications. All the agronomic practices were performed uniformly. At maturity ten guarded plants were selected at random and data were recorded for the following yield and yield related parameters.
To estimate significant differences among parents and hybrids, the data were subjected to statistical analysis by using the analysis of variance technique Steel and Torrie. Significant differences were further subjected to Duncan's new multiple range test (DMR).
The percent increase or decrease of F1 hybrids over mid parent as well as better parent value was calculated to estimate possible heterotic effects for above mentioned parameters following Fonseca and Patterson.
The t value for heterosis was computed following the formula reported by Wynne et al..
RESULTS AND DISCUSSION
The analysis of variance revealed highly significant differences among all the wheat genotypes for all the traits under study (Table 1).
Plant height (cm): The individual comparison of means of all genotypes (Table 2), indicated that among parents genotype 8952 had maximum plant height (115.84 cm), while the parent PBW 222 had minimum plant height (81.53 cm). Among F1 hybrids 8952xUqab 2000 had maximum plant height (116.38 cm), while cross PBW 222xUqab 2000 had minimum plant height (91.35 cm). The heterotic effects in Table 3, indicated that seven out of twenty F1 hybrids showed increase in the plant height over their mid parental values and their heterosis ranged from -7.00 (8961x8952) to 4.17% (8952xUqab 2000). Eight crosses were highly significant in which six had negative values, while two with positive estimates and only one cross was significant but showed positive estimate. Eleven crosses were non-significant in which six showed negative estimates ranged from -0.35 (Uqab 2000x8961) to 1.59% (PBW 222xLU26S).
As for the heterobeltiosis is concerned two out of twenty F1 hybrids exhibited the positive values ranged from 0.47 (8952xUqab 2000) to 2.17% (LU26xUqab 2000). Thirteen out of twenty crosses were highly significant. All of thirteen estimates were negative. Three crosses were significant and four were non-significant. Negative heterobeltiosis ranged from -0.39 (8961xUqab 2000) to -18.51% (8952xPBW 222). The negative estimates of heterosis and heterobeltiosis for plant height are preferred over their mid and better parents in wheat breeding because dwarfness is a desirable character because of its responsiveness to fertilizer and resistance to lodging. The hybrid 8952xPBW 222 with 98.69 cm height is considered to be the best for further use in wheat breeding programme due to -4.35% decrease in height from the mid parent and -718.51% decrease in height from better parent. The results are in agreement with the research findings of Chowdhry et al..
Number of tillers per plant: A perusal of Table 2 indicated the highly significant differences among treatment means for this trait. The average number of tillers per plant ranged from 8.33 (8952xLU26S) to 12.97 (PBW 222). All the crosses showed negative heterotic effect ranged from -0.45 (8952xUqab 2000) to -31.22% (8952xLU26S). Nineteen out of twenty crosses were highly significant but showed negative estimates. Only one cross (8952xUqab 2000) showed non-significant but negative results.
As for the heterobeltiosis is concerned all the crosses showed negative estimates ranged from -6.23 (8952xUqab 2000) to -33.16% (8952xLU26S). All the results were highly significant but showed negative estimates. The negative estimates of heterosis and heterobeltiosis are not desirable because less number of tillers per plant means low yield. Similar results have been reported by Chakraborty and Tiwari.
Number of grains per spike: It is evident from Table 2 that marked differences were found for number of grains per spike among parents and hybrids which varied from 72.10 (Uqab 2000) to 46.30 (LU26Sx8961). Heterotic effects for number of grains per spike of F1 over their respective mid and better parents are given in Table 3.
Positive and highly significant heterosis was observed in nine crosses which surpassed mid parental value ranging from 2.86 (LU26SxPBW 222) to 13.97% (LU26SxUqab 2000). One cross PBW 222x8952 showed positive but significant heterosis. Two crosses PBW 222xLU26S and Uqab 2000x8961 showed positive but non-significant heterosis. Five crosses viz., PBW 222xUqab 2000, LU26Sx8961, Uqab 2000xPBW 222, 8961xLU26S and 8961xUqab 2000 showed negative and highly significant heterosis. Three crosses like Uqab 2000x8952, 8952xLU26S and 8952x8961 showed negative but non-significant heterosis. Heterobeltiosis estimates for grains per spike were also exhibited in Table 3. Positive and highly significant values were observed in six crosses, ranging from 2.68 (in two cross combinations LU26Sx8952 and 8952xPBW 222) to 9.25% (8961xPBW 222) (Table 3). Two crosses i.e. PBW 222xLU26S and PBW 222x8952 showed positive but non-significant heterobeltiosis. The maximum negative heterobeltiosis (-20.71%) was shown by the cross PBW 222xUqab 2000. Two crosses 8952xLU26S and 8952x8961 showed negative but non-significant heterobeltiosis. Presence of hybrid vigour in number of grains per spike had already been reported by Larik et al..
1000-grain weight (g): Highly significant differences were found among parents and F1 hybrids (Table 1). The perusal of Table 2 revealed that maximum grain weight (47.77 g) was recorded from the parent genotype 8952, while the minimum grain weight (34.00 g) was recorded from the parent genotype PBW 222. Among the crosses, minimum 1000-grain weight was exhibited by the cross i.e. Uqab 2000xPBW 222 (41.23 g) while the maximum weight was shown by the cross Uqab 2000xLU26S (49.00 g) closely followed by the cross PBW 222xLU26S (48.40 g). Thirteen out of twenty crosses showed positive and highly significant heterosis over their mid parents ranging from 5.24 (Uqab 2000x8952) to 20.96% (PBW 222xUqab 2000).
|| Analysis of variance (mean square values) for different traits
|| Mean performance and statistical significance for different
traits in wheat
|Means having the same letters do not differ significantly
at 0.05 level of probability by Duncans multiple range test
|| Estimation of percent heterosis (Hb%) and heterobeltiosis
(Hbt%) for different parameters
Two cross combinations 8961xLU26S and 8952xLU26S showed positive but non-significant heterosis (Table 3). The hybrids LU26Sx8952 and 8961x8952 showed highly significant but negative heterosis. Three cross combinations LU26Sx8961, 8961xUqab 2000 and 8952x8961 showed negative but non-significant heterosis. For heterobeltiosis, three hybrids like PBW 222xUqab 2000, Uqab 2000xPBW 222 and Uqab 2000xLU26S showed positive and highly significant values. Three hybrids PBW 222xLU26S, 8961xLU26S and 8952xLU26S showed positive but non-significant heterosis over their better parents. Eight crosses showed highly significant but negative estimates ranging from -2.87 (LU26SxUqab 2000) to -9.44% (8961xUqab 2000). Only one cross Uqab 2000x8961 showed significant but negative heterobeltiosis. Five cross combinations like PBW 222x8952, LU26SxPBW 222, LU26Sx8961, 8952xUqab 2000 and 8952x8961 showed negative and non-significant results. Similar results were reported by Atta and Khan.
Grain yield per plant (g): The analysis of variance for grain yield per plant is presented in Table 1. The individual comparison of means of all the genotypes is given in Table 2. The estimate of percent heterosis and heterobeltiosis for grain yield per plant of twenty wheat hybrids were given in Table 3. These estimates exhibited that five cross combinations PBW 222x8961, PBW 222x8952, LU26SxUqab 2000, Uqab 2000xLU26S and 8952xUqab 2000 showed positive and highly significant results. Three crosses like PBW 222xUqab 2000, 8961xPBW 222 and 8952xPBW 222 showed positive but non-significant estimates. Twelve crosses revealed negative values ranging from -0.58 (PBW 222xLU26S) to -23.70% (8952xLU26S). Out of these twelve crosses, eleven were highly significant and one cross i.e. PBW 222xLU26S was non-significant. For heterobeltiosis, four crosses viz., PBW 222x8952, LU26SxUqab 2000, Uqab 2000xLU26S and 8952xUqab 2000 showed positive and highly significant results. Sixteen out of twenty hybrids showed negative estimates ranging from -7.25 (PBW 222x8961) to -25.39% (8952xLU26S). All the negative estimates were highly significant. The results obtained also supported by the earlier observations on heterotic effects in grain yield, reported by Atta and Khan.
A review of results made it clear that most of crosses exhibited remarkable heterosis over mid and better parents for various characters. However, crosses PBW 222xUqab 2000, 8961x8952, Uqab 2000x8952 and Uqab 2000xLU26S may be considered for selection as hybrid or pure line wheat varieties after achieving desired homozygosity.