was discovered that the internal area of the post-hole corer was 0.09 m 2 but not before 
twelve samples were taken with other methods that had an actual area of 0.1 m 2 . 
Additionally, the volume of residue retained on the 1.0 mesh sieve at several sites 
exceeded several liters, which was impractical to process. This volume of residue 
necessitated sub-sampling the residue in 78 of the 217 samples (36%). The samples 
were subsampled to the minimum practical extent. The occurrence of 0.1 -m 2 samples 
and the subsampling resulted in twelve functional sample sizes, ranging from 0.0028 to 
0.1 m 2 , a 36-fold difference in sample area. Because the majority (56%) of the 
samples were taken with the post-hole sampler, all benthic abundances were 
normalized to 0.09 m 2 for analysis. Species richness does not scale linearly with area 
so no attempt was made to normalize the number of taxa per sample. Accordingly, we 
did not analyze species richness with the entire benthic data set. 
7 
