Annual species' experimental germination responses to light and temperature do not correspond with their microhabitat associations in the field

Questions Annual species have evolved sets of germination cues that are thought to be predictive of the post‐germination environment. In naturally patchy environments, germination microsites often vary considerably in the amount of light they receive and in the diurnal temperature fluctuations they experience. However, whether species' differential germination responses to light and temperature are associated with their spatial patterns of occurrence remains largely untested. Location Mediterranean‐climate woodlands in Southwest Western Australia. Methods We surveyed species' occurrences in annual plant communities in 150 quadrats across gradients of canopy cover and litter cover. Nineteen species recorded in this survey were then included in a germination experiment that manipulated (1) Light vs Dark (12 h light or continuous dark) approximating seeds near the soil surface vs those covered by litter and (2) Cold vs Warm temperature regimes (7/18°C and 7/24°C) approximating diurnal fluctuations experienced in shaded vs sun‐exposed microsites, respectively. Results In the germination experiment, six species had highest germination probabilities in the Light treatment (regardless of temperature), five in Cold + Light, one in Warm + Light, two were indifferent to the treatments, and four did not germinate at all. Binomial linear mixed‐effects models showed that species' maximum responses to light and temperature did not explain their spatial distributions along canopy cover and litter cover gradients, contrary to theoretical expectations of germination being a strong driver of species' occurrences. Conclusions Despite variation in species' responses to experimental treatments, no association was found with their field microsite associations. Germination strategies in our system were wider than expected for Mediterranean systems. Our results support that germination cues are not strong drivers of microhabitat associations in this system.