What determines whether a species of insect is described? Evidence from a study of tropical forest beetles

1. The rainforest canopy has been called 'the last biological frontier', and if this is true, there should be more undescribed species in this stratum than the ground stratum. 2. Here, we test this and other hypotheses regarding traits of described and undescribed species by a sub-sample of 156 species into 96 described and 60 undescribed species from a beetle assemblage of 1473 species collected from the canopy and ground in an Australian lowland rainforest. 3. We show that described species are significantly more likely to be in the canopy, are more likely to be larger and, if they are large, are more likely to have been described earlier. 4. Undescribed species are just as likely to be found near the ground as in the canopy and are more likely to be smaller. 5. After the first year of sampling, 'new' described and undescribed species not previously encountered continued to appear in each of three further years of trapping. 6. These data show that the canopy fauna is in fact relatively 'well known', and that the undescribed species to be found in both strata are likely to be smaller than described species and are less likely to be plant feeders.