Small island biogeography in the Gulf of California: lizards, the subsidized island biogeography hypothesis, and the small island effect

Abstract Aim We used insular lizard communities to test the predictions of two hypotheses that attempt to explain patterns of species richness on small islands. We first address the subsidized island biogeography (SIB) hypothesis, which predicts that spatial subsidies may cause insular species richness to deviate from species–area predictions, especially on small islands. Next, we examine the small island effect (SIE), which suggests small islands may not fit the traditional log‐linear species–area curve. Location Islands with arthropodivorous lizard communities throughout the Gulf of California. Methods To evaluate the SIB hypothesis, we first identified subsidized and unsubsidized islands based on surrogate measures of allochthonous productivity (i.e. island size and bird presence). Subsequently, we created species–area curves from previously published lizard species richness and island area data. We used the residuals and slopes from these analyses to compare species richness on subsidized and unsubsidized islands. To test for an SIE, we used breakpoint regression to model the relationship between lizard species richness and island area. We compared results from this model to results from the log‐linear regression model. Results Subsidized islands had a lower slope than unsubsidized islands, and the difference between these groups was significant when small islands were defined as < 1 km 2 . In addition to comparing slopes, we tested for differences in the magnitude of the residuals (from the species–area regression of all islands) for subsidized vs. unsubsidized islands. We found no significant patterns in the residual values for small vs. large islands, or between islands with and without seabirds. The SIE was found to be a slightly better predictor of lizard species richness than the traditional log‐linear model. Main conclusions Predictions of the SIB hypothesis were partially supported by the data. The absence of a significant SIE may be a result of spatial subsidies as explained by the SIB hypothesis and data presented here. We conclude by suggesting potential scenarios to test for interactions between these two small island hypotheses. Future studies considering factors affecting species richness should examine the possible role of spatial subsidies, an SIE, or a synergistic effect of the two in data sets with small islands.