Novotny et al. 2007

Novotny V, Miller SE, Hulcr J, Drew RAI, Basset Y, Janda M, Setliff GP, Darrow K, Stewart AJA, Auga J, Isua B, Molem K, Manumbor M, Tamtiai E, Mogia M, Weiblen GD. 2007. Low beta diversity of herbivorous insects in tropical forests. Nature 448: 692-697.

These authors examined 370 species of caterpillars and about 130 species of ambrosia beetles (Scolytinae and Platypodinae) and fruitflies (Tephritidae) in a large contiguous patch of rainforest in Papua New Guinea. While total species richness was high, as expected for the tropics, beta diversity or turnover in species composition through space, was surprisingly low. The authors attribute this surprising finding primarily to the genus-specific rather than species-specific diets of many of the Lepidoptera. '

Beta diversity can be overestimated by inadequate sampling of many rare species, or underestimated by relying on taxonomically known species that tend to be the most widespread and abundant species. In this study, these issues were addressed by studying relatively well-described insect taxa on a limited subset of the available plant hosts, in this case four genera of trees including 175 species in the study area. Study sites were evenly distributed across a large area that was nearly homogeneous for altitude, climate, soil, vegetation, and other factors. In general, herbivore diversity tracked the diversity (alpha and beta) of their plant hosts, which in itself is a less than surprising finding. Other studies that have found high beta diversity of tropical insects have followed topological or climatic gradients, where plant host beta diversity is also high.'

This paper appears in the same issue of Nature as Dyer et al. (2007), and a “News & Views” commentary by Stork (2007). Dyer et al. (2007) found high beta diversity of tropical insects, but their study is different in many ways from that of Novotny et al. (2007).