Numerous surveys have shown that fish species richness (number of species) is positively correlated with lake pH. However, species richness of fish communities is also correlated with lake size, and low-pH lakes are often small. Thus, conclusions drawn from examination of fish community structure relative to spatial (among- lake) variation in pH have been limited by uncertainties regarding the confounded effects of lake area. The authors used two statistical methods, analysis of covariance and a nonparametric blocked comparison test, to remove effects of lake area and compare fish species richness in low-pH and high-pH lakes. Data from six previous surveys of water chemistry and fish communities in lakes of Ontario and northern Wisconsin were examined. Lakes with low pH ( less than or equal to 6.0) contained significantly fewer fish species than lakes with high pH (> 6.0) when the effect of lake area was considered. A simple probabilistic model showed that the ability to detect differences in species richness is low when lake areas and the pool of potential colonizing species are small. The authors recommend the blocked comparison test for separating the effects of lake area and pH on species richness.
