Time and the Number of Herbivore Species: The Pests of Sugarcane

Abstract

Previous analyses have established that host species range in the most important determinant of plant— and animal—associated parasite species richness, and that the age of a host species within a region is, at most, much less important in determining this richness. However, these previous analyses were not able to completely rule out age as a cause, because the available host age estimations were crude and few host groups were available for comparison. This paper examines the arthropod pests of sugarcane (Saccharum officinarum) in 51 regions for which dates of introduction, areas in cultivation, and associated pest data are available. The sugarcane data are more complete and extensive than those for previously investigated host—parasite associations, and they make possible a more detailed analysis of the causes of species richness. Multiple regression analyses of these data cannot identify time to have any measurable influence upon the species richness of sugarcane pests. This implies pest species accumulation to reach an asymptote within, at most, a few hundred years. The asymptote is set by the geographic range of host species. The results of this study are consistent with those of four other analyses of host—parasite associations. We suggest sufficient evidence has accumulated to justify abandoning nonasymptotic species accumulation models (the "time) and "stability—time" hypotheses), at least for parasite communities. We caution against invoking "time" hypotheses for the explanation of diversity variation in other ecological systems where the influence of area and immigration rates cannot be dealt with. The "time" hypothesis also often involves virtually untestable assumptions about habitat constancy over very long periods of time.