Nonasymptotic species richness models and the insects of british trees

Nonasymptotic models of species diversity are those that do not consider negative feedback between number of species in a biota and the net rate of species addition. These models propose species richness differences to be primarily the product of geologic age differences among biotas. Nonasymptotic explanations are traditional for various diversity difference spectra, including latitudinal diversity gradients and the greater species richness of ancient compared to young lakes. I review new evidence that renders these nonasymptotic explanations doubtful.The only uncontested evidence in favor of nonasymptotic species accumulation is Southwood's correlation between the number of insect species associated with British tree taxa and the number of Quaternary fossil records of these taxa. I show that nonasymptotic explanations of species richness variation are unacceptable for this system also. The variation in insect species richness among these taxa is well accounted for by a species-area relationship. The species richness asymptote is attained within a few hundred years, and the influence of cumulative host taxon abundance (number of fossil records) upon insect species richness is insignificant after the statistical influence of present-day host plant range is removed. I conclude Southwood's correlation to lack causality; it probably derives from a correlation between the number of fossil records of the host taxa and their present range.