Partitioning of dry matter into primary branches and pod initiation on the main inflorescence of Lupinus angustifolius

The hypothesis that pod load on the main inflorescence of Lupins angustifolius (L.) is negatively coupled to the amount of dry matter partitioned into primary branches was tested. Growth rates of main stem and primary branches during the period of pod initiation were measured in a series of field experiments at two different sites. Variation in pod set was generated experimentally by varying sowing date or density and by using four cultivars (Yandee, Danja, Gungurru and Warrah). The cultivars differed in their total growth rate, but not in the way dry matter was partitioned into main stem and primary branches. In contrast, significant variation in dry matter partitioning was observed for cv. Danja tested across site and sowing date. With increased plant density, at the same sowing date, the proportion of dry matter allocated to branches increased consistently, even though it decreased in absolute terms, and the number of pods initiated on the main infloresence also decreased consistently. Thus a negative relationship between pod load and the proportion of dry matter allocated into branches was observed as a result of variation in density. However, there was no consistent relationship when variations in pod load were induced by site, date of sowing or cultivar. These data are interpreted to indicate that increased pod initiation on the main inflorescence is not necessarily coupled with decreased partitioning of dry matter into primary branches.