| Heritability estimation by parent progeny regression
vis-a-vis standard unit method in four wheat crosses I.S. PAWAR, R.S. PARODA* and S. SINGH Department of Plant Breeding, Haryana Agricultural University, Hisar-125004, India Since the effectiveness of selection depends on the additive portion of genetic variance in relation to total variance, heritability in narrow sense is of great importance to the plant breeder. The parent off-spring regression method proposed by LUSH (1940) is being used as one of the most common methods of determining heritability of metric traits in plant populations. In self-fertilizing crops, some of the routinely used parent-progeny combinations are: F1/F2, F2/ F3 and F3/F4. However, as pointed out by SMITH & KINMAN (1965), this method has been frequently misused for estimating heritability in autogams since the previous inbreeding of the parents which can cause an upward bias in heritability estimate is not usually considered. The reliability of heritability estimate thus can be increased by adjusting regression estimate for previous inbreeding of the parents. FREY & HORNER (1957) suggested a modification in the parent progeny regression procedure for calculating heritability in standard units. This modification was suggested to reduce the scaling effects of genotype x environment interaction. The present study was undertaken to compare heritability estimates obtained through the above mentioned two methods in segregating generations of four wheat crosses. Materials and Methods Visual selections were carried out from large, space planted F2 and F3 populations of four wheat crosses, namely, WL 711 x HD 2122 (cross 1), HD 2122 x Sonalika (cross 2), WL 711 x HD 1981 (cross 3) and WG 377 x HD 1925 (cross 4) during rabi 1982 - 83 and 47 plants were selected from each of these two populations in each cross. The progenies (F3 and F4) of these plants were grown in a randomized block design with three replications in single row plots of 2.5 m length having row to row distance of 23 cm and plant to plant 10 cm at Haryana Agricultural University research farm during rabi 1983 - 84. Observations were recorded on five randomly selected competitive plants per row per replication from each of the four crosses for number of effective tillers per plant, number of grains per spike, 1000 grain weight (g) and grain yield per plant (g). Heritability (narrow sense) was calculated as per methods suggested by SMITH & KINMAN (1965) and by FREY & HORNER (1957). The expected values of genetic advance were estimated at 5 percent level of selection intensity using standard procedure. Realized genetic gains were also calculated using F2 and F3 data only. Finally, the values showing differences between expected and realized genetic gains were calculated to determine the relative reliability of the two methods used to estimate heritability. |
| * Present address: Director, National Bureau of Plant Genetic Resources, Pusa Complex, New Delhi-110012. |
| --> Next |