Research Article
Response to Selection in Sorghum for Resistance to Head Bug (Eurystylus oldi Poppius)
Not Available
S.O. Alabi
Not Available
P.E. Olorunju
Not Available
O. Ajayi
Not Available
Eurystylus oldi Poppius (Hemiptera:Miridae) or Sorghum head bug is an important insect pest of Sorghum (Sorghum bicolor (L.) Moench) in Asia and Africa. In West Africa head bug have gained importance in recent years due to the introduction of elite, early flowering cultivars with compact panicles. E. oldi damages Sorghum plants by its feeding and oviposition activities on the panicles, leaves and within the plant (Gahukar, 1991; Steck et al., 1989). Head bug damage spoils the grain quality by rendering the grain soft, which leads to high incidence of grain mould, thus resulting in poor seed germination and renders the grain unfit for human consumption (Sharma and Lopez, 1990; 1991) Search for resistant sources has yielded some cultivars with varying levels of resistance (Sharma et al., 1994; Showemimo, 1998). Inheritance studies have revealed resistance to be recessive without maternal effect, preponderantly non-additive gene action in control, while additive gene action is also important in some cases.
A pedigree selection procedure, that utilizes quantitatively inherited traits, which are characterized by off spring (s)-desired parents resemblance (Falconer, 1981; House, 1995) was employed to measure the response of Sorghum resistant to E. oldi. Therefore, this study reports the progress obtained after two generations of the selection scheme.
F2 populations of crosses between 3 E. oldi resistant cultivars (HRhb 94001, HRhb 94002 from ICRISAT and Malisor 84-7 from IER, Mali) and 3 E. oldi susceptible cultivar (HRhb 94001S from IAR, Samaru, HRhb 94002S from Ghana and ICSV 112 from ICRISAT). The experiment was conducted for four cropping seasons (1995-1998) in the research field of Institute for Agricultural Research (IAR) Samaru, Zaria, Nigeria located in 11°11N; 07°38E at 686 m above sea level. Factorial mating system was used to produce 9 F1 hybrids; these were selfed and advanced to F2 populations. To achieve early generation selection and improvements, the F2 population was considered base or source population. Ten percent selection intensity was exerted on the F2 population due to high variability and segregation at this generation. Selection was biased towards desired resistance traits.
The F2 seeds from each population were planted 2 to 3 weeks after infector rows were planted. Screening and selection were done by artificially infesting each Sorghum head at 50% flowering or half anthesis stage with 10 pairs of E. oldi adult using the No choice method of head cages as recommended by Sharma et al. (1992). The selection criteria were based on the following desired resistance traits; number of head bugs/5 panicles (counted 20 days after infestation); grain damage rating (evaluated at physiological maturity on a 1-9 scale) according to Sharma et al. (1992); 1000-grain weight (the weight of 1000-grain from the 5 randomly tagged, caged and infested panicles after threshing); grain yield (the weight of threshed grains in kg ha-1). The procedure of resistant and 3 susceptible cultivars) were evaluated for desired resistance traits. The trial was arranged in randomized complete block design with three replications. screening and selection was repeated in 1996 for the F3 populations, (thus F2 derived lines in F3 (F2:3) and in 1997 for the F2 derived lines in F4 (F2:4).
The base population (F2), F2:3, F2:4 and six checks (The plot size was 2 rows each, 5 m long with 0.75 m and 0.25 m inter and intra row spacing, respectively to give approximately 88,888 plants per hectare. Prior to ridging and sowing, 32 kg ha-1 of P2O5 as single superphosphate fertilizer was applied as basal dressing. Split application of nitrogen fertilizer (Urea) was done: 32 kg ha-1 was applied as basal dressing while 30 kg ha-1 was applied 3 weeks after thinning as top dressing. A mixture of Gramoxone and Sorghoprim A at the rate of 300 mL/20 L of water was sprayed as pre-emergence herbicide. Weeds were controlled afterwards by hoe weeding at 3 and 6 weeks after sowing. Vetox 85 was sprayed to control stem borer.
Analysis of variance was computed for all data recorded, orthogonal comparison was made among means of entries; selection differential was used to estimate gains/generation. Phenotypic and genotypic correlation coefficient and correlated responses were computed as used by Singh and Chaudhary (1985). Genetic advance was estimated using the method proposed by Obilana and Fakorede (1981).
Means squares and significant levels of F-test for four E. oldi resistance traits are presented in Table 1. The mean squares show highly significant effect among F2, F2:3 and F2:4 for all the resistance traits except 1000-grain weight among F2:4 population. Table 2 presents the means for each of the derived lines, resistant and susceptible checks, estimated gain and percentage gain/generation for 4 desired resistance traits. There were decreases as the generation of selections progresses for number of head bugs/5 panicle and grain damage rating, while there were increases as the generation of selection progresses for 1000-grain weight and grain yield. The F2:4 gave better result compared to the resistant check for all the traits. Gain/generation revealed a considerable loss in terms of the number of head bugs and grain damage rating (F2:3 = -178, F2:4= -407 and F2:3= -0.90, F2:4= -2.20, respectively). A gain/generation of 3.47 and 7.01 for F2:3 and F2:4 for 1000- grain weight, similar gain/generation of 272 and 405 for F2:3 and F2:4 for grain yield, respectively was obtained. The corresponding percentage gain/generation was also presented (Table 2).
Phenotypic (rp) and genotypic (rg) correlation coefficients (Table 3) were negative and highly significant between number of head bugs and grain yield (rp = -0.6102 and rg = -0.5917). Grain damage rating was also negative and highly significantly correlated with grain yield (rp= - 0.6492) while significant at p = 0.05 at genotypic correlation. Positive and significant phenotypic correlation was obtained between 1000-grain weight and grain yield. Negative and significant phenotypic correlation was obtained between number of head bugs and grain damage rating and 1000-grain weight. The correlated response revealed negative percentage for number of head bugs and grain damage rating in association with grain yield while positive percentage was recorded for 1000-grain weight (10.71%) in Table 3.
Expected genetic gain and percentage genetic gain for all the traits in two generations of selection are presented in Table 4.
Table 1: | Mean squares and significance levels of F-test for four E. oldi resistance traits in Sorghum |
*, * * = Significant at 0.05 and 0.01 probability levels, respectively |
Table 2: | Mean of F2, F2:3, F2:4, resistant and susceptible checks for resistance to E. oldi |
+Expressed as a proportion of F2 generation |
Table 3: | Phenotypic and genotypic (in bracket) correlations among for E. oldi resistance traits and their correlated responses (CR %) |
*,* * denote significant at 0.05, 0.01 level of probability, respectively |
Table 4: | Expected genetic gain (ΔG) and percentage genetic gain (%ΔG) of F2, F2:3 and F2:4 for four resistance traits of E. oldi |
+ Expressed as a proportion of proceeding generating. -No percent gain over base population |
Decrease in number of head bugs and grain damage rating was recorded as the generation of selection progresses, an increase in genetic gain was recorded for 1000-grain weight and grain yield. There was an increase percentage genetic gain for all the traits in the two generation of selection except for 1000-grain weight.
The reduction in number of head bugs and damage rating, further evidenced by the gain/generation of selection and percentage gain/generation of selection, revealed that selection for head bug resistance is feasible and probably more so after two or more generation of selection. The corresponding increase in 1000-grain weight and grain yield as the generation of selection progresses attest to this fact. The large gain/generation of selection in this study suggests that the gene (s) of resistance is being fixed. According to Barry et al. (1983). changes in gene frequencies as the cycle of selection increases leads to gene fixation.
Phenotypic and genotypic correlation between traits in any population is important if progress is to be made in selecting superior genotypes/cultivars/lines. Grain yield was negatively but significantly correlated with all other traits except 1000-grain, this implies that selecting genotypes with higher number of head bugs and grain damage rating will lead to significant decrease in grain yield, meanwhile, grain yield will improve by selecting genotypes with lower number of head bugs and grain damage rating. The positive and significant phenotypic correlation between 1000-grain weight and grain yield shows that selecting genotypes for resistance vis-à-vis grain yield and grain weight are important. The negative and significant phenotypic correlation and negative genotypic correlation between number of head bugs and grain damage rating and 1000-grain weight mean that selection of higher number of head bugs could increase grain damage rating and lower the grain weight. This result agrees with those of Beyo (1996), Showemimo (1998) and Showemimo et al. (2000). Correlated response measure changes in a trait due to indirect selection on an associated trait. Therefore, this study revealed a minimum of 16 and 19% reduction in number of head bugs and grain damage thus, a meaningful improvement in resistance along with increase grain yield is achievable through selection via these traits. Approximately 11% correlated response is expected in improving resistance vis-à-vis grain yield by selecting for healthy and heavier grains.
The reduction in genetic gain of number of head bugs and grain damage rating from F2 to F2:3 and F2:4 indicate that the fewer the number of head bugs the lower the grain damage rating, however, the higher the grain weight and grain yield as the generation advances, thus, the better the predictability of lines towards head bug resistance. Similar results were reported by Obilana and El-Rouby (1980) and Whan et al. (1982).
In conclusion, selection and response to selection for resistance to Sorghum head bug (Eurystylus oldi (Poppius)) in early generation using pedigree selection method is achievable and advantageous in term of shortening time, money and space in the development of improved resistant genotypes. The improved populations will readily serve as superior or sources of genotypes resistant to E. oldi.