Pod and seed numbers as a function of photothermal quotient during the seed set period of field pea (Pisum sativum) crops

The effects of radiation and temperature during the seed set period (SSP) on pod number per square metre (PN m(-2)) and seed number per square metre (SN m(-2)) and those of temperature during grain filling on unit seed weight (USW milligram per seed) of field pea (Pisum sativium L.) were examined in experiments involving irrigated crops of three or more cultivars of concrastiug maturity sown on two or more dates per year from 1996 to 1998 at Buenos Aires, Amgentina. The duration of the seed-setting phase was estimated from records of the progress of lowering on he main stem and an estimate (obtained using an optimisation procedure) of the thermal time from flowering at which the uppermost reproductive node reached the final stage of seed abortion (FSSA). The FSSA at a particular node was assumed to be achieved 200degreesC day (T-b = 4degreesC) after flowering at the same node. The grain-filling phase was assumed to run from the achievement of FSSA at the first reproductive node through to 200degreesC day (T-b 0degreesC) after the date of achievement of the FSSA by the second flowering node. The treatments (cultivar, sowing date, year) produced important ranges of above-ground biomass (AGB) at maturity (2111782 m(-2)). seed yield (SY, 119-331 g m(-2)), SN (1062-3698 seeds m(-2)) and USW (67-150 mg seed(-1)). Seed yield was strongly correlated with SN, and there was full compensation between SN and USW in large-seed cultivars in the high SO ranee. but not at lower values of SN or in small-seeded cultivars. Both PN (r = 0.83) and SN (r = 0-81, P < 0.0005) were strongly correlated with the mean daily value of the photothermal quotient (PQ = incident radiation/(mean temperature - base temperature)) for the seed-setting phase. Large- and small-seeded cultivars had PN/PQ and SN/PQ relationships with slopes which did not differ among categories but with significantly different intercepts. When the effects of low temperatures during flowering and early Train growth were allowed for, outliers on the PN/PQ and SN/PQ relationships for unstressed crops fell within the confidence limits of the respective linear regressions. Unit seed weight showed a negative response to mean temperature during the grain-filling phase in large- and small-seeded cultivars. We conclude that the relationships established in these experiments, taken together with previous work by other authors, constitute a robust basis for modelling the yield of unstressed field pea crops. (C) 2004 Elsevier B.V. All rights reserved.