| Association of seed protein with grain weight and size
in winter and spring wheat crosses. N.M. SHAHANI and N.N. SAULESCU* Department of Plant Breeding and Genetics, Sind Agriculture University Tandojam, Pakistan Selection of high grain protein content and improved quantitative traits in cereals is faced with a problem to increase protein content without sacrificing high levels of grain yield of the parental lines. The main limitations are the negative correlation between yield v/s grain protein percentage and grain boldness v/s protein percentage (JOHNSON et al. 1971 ; SHAHANI 1980). Complex genetic control of protein content (COWLEY & Wells 1980 ; KERTESZ et al. 1980 : SHAHANIet al. 1983), influence of the environmental factors on grain protein content (BABYAKIN & PISCUNOVA 1979 ; SHAHANIet al. 1983) and the effect of fertilizers on grain protein percentage (COCHRAN et al. 1978 ; DUBETZ et al. 1979) are other major problems to evolve improved varieties with high grain protein percentage. The present study was, therefore, intended to study the relationship between grain protein percentage and other important quantitative characters in genetically diverse gene pools of winter and spring wheat F1 and F2 crosses, in order to improve the theoretical basis by this new wheat breeding approach. Materials and Methods Two high yielding winter wheat lines, (F310C3-4 and F21-76) bred at Research Institute for Cereals and Industrial Crops, Fundulea, Romania, were crossed direct and reciprocal with two semidwarf spring wheat varieties (Pak-70 and Tandojam-75) procured from Agricultural Research Institute, Tandoja, Pakistan. Seeds of F0 hybrids were sown immediately after harvest in the last week of June 1978, in Phytotron. Parents, F1 and F2 populations were sown in second fortnight of October, 1978 and March, 1979, using randomized complete block design with three replications in order to study the biological material in two different contrasting environments. Protein content was determined by microkjeldahl method as crude nitrogen times 5.7. Data for coefficient of correlation (r) values, coefficient of determination (r2) and coefficient of regression (b) values were determined after Snedecor (1956). Results and Discussions 1000-Grain Weight and Grain Protein Percentage. Correlation coefficient (r) values of parents and F1 populations (Table 1) are small and nonsignificant, taking in consideration that practically the parents and F1 populations are generally uniform. This means that the variation, which was caused by the environmental factors in grain weight, was not correlated with the corresponding variation of grain protein percentage. In F2 generations most of the combinations, except a few, have small and nonsignificant correlation coefficient (r) values (Table 2). These results suggest that grain protein content was highly heterogenous in all the parental forms and parents are genetically different from each other, hence most of the variability for grain protein content was independent to the variation of 1000 grain weight. The results are in confirmation with MCNEAL et al. (1972) and JAIN et al. (1976). Frequency distribution (Fig. 2) shows transgressive segregates of more grain weight and high protein percentage. This sort of transgressive segregations may provide the chances for selection of superior plants. |
| * Research Institute for Cereals and Industrial Crops, Fundulea, Romania. |
| --> Next |