Demographic consequences of delayed germination in two annual grasses from two locations of contrasting aridity

Delayed seed germination is considered to be a bet-hedging strategy, but experimental evidence of its adaptive role as an inherited trait is still lacking. In each of two co-occuring annual grass species, populations of Mediterranean and desert origin were studied during three consecutive years for population demography and seed germination in the reciprocally introduced experimental soil seed banks. The two environments strikingly differed in productivity (annual rainfall) and predictability (variation in amount and timing of annual rainfall). The two species exhibited highly similar pattern of seed size and dormancy across the two environments. In both species, a higher proportion of dormant seeds was observed at the desert location and for the seeds of desert origin, consistent with bet-hedging buffering against unpredictability of rainfall and high probability of drought in this environment. In addition, in both species seed mass was significantly less in plants of desert origin than in plants of Mediterranean origin. The two environments differed in demographic consequences of temporal variation in precipitation. In the Mediterranean population, even in the year of least precipitation, adults grew to maturity and seeds were produced. These seeds served to maintain population size. In contrast, in the desert population, in the year of least rainfall no seedlings survived to maturity and the soil seed bank was the only source of population persistence. Altogether, the results concur with predicted by adaptive bet hedging importance of delayed germination under marginal precipitation.