Priority effects in competitive communities allow the order of colonization and/or initial densities to influence the long-term outcome of species interactions by giving a short-term advantage to one of the interacting species. Some have argued that because competition should only influence species populations at high densities where communities are at equilibrium and resources are limiting, priority effects should not be observable at low densities, and therefore, should not affect populations in the exponential phase of growth. We experimentally tested this hypothesis by growing each species in monoculture, and also varying the initial densities of two strains of marine pico-cyanobacteria (a red vs. green pigmented Synechococcus) in polycultures that had equal or highly skewed initial densities of each of the two strains. Species were maintained in semi-continuous batch culture for 30 days with samples taken every 2-3 days to assess cell densities by flow cytometry. A hierarchical Bayesian approach using a Gibbs sampler was applied to estimate competitive effects on growth rates, which allowed us to assess how competition varied as a function of initial densities throughout the growth phase.

Results/Conclusions

We found that the green strain of Synechococcus ultimately dominated all polycultures regardless of initial densities; thus, there was no evidence of priority effects in final assemblages that had reach maximal densities. However, contrary to our initial hypothesis, we found evidence of priority effects operating during the exponential growth phases of population growth. Specifically, competition among strains with equal initial densities led to a reduced exponential growth rate of the red strain, but had no influence on the green strain. Both species had significantly lower growth rates in the treatment skewed towards high initial densities of the green strain, but neither strain had an altered growth rate in the treatment skewed towards high initial densities of the red strain. These data indicate that asymmetric interspecific competition occurred at low densities during the exponential phases of growth. As a result, the green species was able to achieve a higher specific growth rate compared to its competitor when it was given an initial numerical advantage. Our study suggests that the effects of competition are observable even in the exponential phases of population growth, and that numerical advantage can still change initial growth rates even when priority effects are not observable at equilibrium.