A test of the reproductive economy hypothesis in plants: more offspring per capita come from large (not small) parents in Avena barbata

Under crowded conditions, plant populations typically exhibit L-shaped distributions of size. Many plants remain small and suppressed in such populations, and more offspring are propagated into the next generation from the many smaller plants than from the few large plants. The hypothesis of “reproductive economy” advances that this creates natural selection favouring the ability to reproduce while still small over the ability to become large. We develop a simple model using exponential distributions of size within morphs that differ in their growth potential and size threshold for reproduction. Our model shows that selection consistently favours the morph with greater potential to become large, even if that potential comes at the cost of a larger size threshold for reproduction. We also tested the reproductive economy hypothesis using 4 years of field data at two sites in California, USA from an experiment with the annual grass Avena barbata. Despite strongly skewed size distributions, the genotypes giving larger size always increased in frequency between parents and offspring while those of the smallest size class decreased. Directional selection gradients were also always positive and significant, indicating that natural selection indeed favours the few larger plants over their more numerous smaller neighbours. Neither our empirical nor theoretical findings support the reproductive economy hypothesis. Instead, we argue that selection for large size is a manifestation of underlying selection to acquire resources, resolving any perceived paradox of a preponderance of small individuals under selection for large size.