1 Several hypotheses have been put forward to explain the phenomenon of masting or mast seeding, i.e. the supra-annual, periodic production of a large number of seeds in long-lived plants. Some of these hypotheses deal with the proximate causes of masting (e.g. the climate hypothesis) but others are concerned mostly with ultimate, evolutionary explanations (e.g. the pollination efficiency hypothesis).
2 The seed production of three tree species, Abies balsamea, Acer saccharum and Betula alleghaniensis, was followed over a 7-year period in an old-growth, cold temperate forest of north-eastern North America. The main objectives were to determine the extent of interannual variations in seed production, to investigate the relationship between viable and potential seed crop and crop efficiency, and to explore the effects of climate on seed production.
3 Potential and viable seed production varied significantly among years for all three species. However, the timing of dispersal remained the same regardless of the level of seed production.
4 Seed rain was spatially less heterogeneous in years of high seed production, suggesting that most trees were reproducing in such years.
5 Over the 7-year period, there was a significant concordance among species in their viable seed crop and crop efficiency, but not in their potential seed crop. Crop efficiency was positively correlated to potential seed crop for Abies and Betula, but not for Acer.
6 High seed production was related to warm, dry conditions in the spring of the previous year (i.e. at reproductive bud initiation) but to a moist summer in the year of seed maturation.
7 Masting in these three species thus appears to be controlled by several factors, including climate and pollination efficiency.