Comparison of harvest index stability between and within four pea genotypes : A dissertation submitted in partial fulfilment of the requirements for Bachelor of Agricultural Science

Abstract

Results are reported on a pea (Pisum sativum L.) experiment carried out at the DSIR, Lincoln, Canterbury on a Templeton silt loam over the season 1986/1987. Four pea cultivars, Birte, PxW, 017 and Rovar, were grown in a randomised block design of three replicates each. The environmental conditions were kept equal for all cultivars. The experimental objectives were to compare the stability of harvest index between and within the cultivars and to define the plant type most suited to growing in crop communities. Throughout growth, regular samplings were taken to obtain individual plant and sample fresh weights. At maturity, samplings also provided parameters such as seed fresh weight. Harvest indices were calculated as the ratios of seed dry weight to total dry weight. Frequency distributions were calculated of individual fresh weight per sample for total and seed fresh weights, and of harvest indices. The shape of the frequency distributions was related to stability. From the flat shape of their frequency distributions it was evident, that Birte and PxW showed poor uniformity, hence poor adaptability to growing in crop communities. 017 showed better adaptability to growing in crop communities by showing normal distributions, and Rovar was the best adapted cultivar. The plant type most suited to growing in a crop situation could be found among Rovar. This plant type would be a weak competitor, allowing efficient distribution of resources to its neighbours and hence the total crop community. Such a crop would reach a maximum and stable yield potential.