• Body size;
  • flight strength;
  • host plants;
  • life history;
  • Malesia;
  • niche breadth;
  • range size heritability;
  • Sphingidae;
  • wing load


Aim  We tested various species-level traits for their potential to explain species’ range sizes and dispersal abilities.

Location  Southeast Asia and Malay Archipelago.

Methods  We used published maps of geographical distribution estimates for sphingid moths to calculate range areas and classify species according to their dispersion across (present or historical) water straits in the Malay Archipelago. We tested forewing length (FWL), wing load (thorax width/FWL), presence or absence of a functional proboscis (i.e. adult feeding), larval diet breadth and larval diet composition for univariate correlations with range size and inter-island dispersion. We used multivariate, phylogenetically controlled models to test for independent effects of parameters.

Results  Range size correlated strongly with larval diet breadth, a result that was also confirmed in the multivariate model. Adult feeding had a significant impact on range sizes only within the multivariate model, but not in the univariate correlation. Dispersal class also correlated with larval diet breadth, but was additionally influenced by forewing length, wing load and larval diet composition. A univariate effect of adult feeding became non-significant in the multivariate, phylogenetically controlled model.

Main conclusions  Larval diet breadth is the best predictor of range size as well as inter-island dispersion, confirming the importance of niche breadth on the geographic ranges of species. A number of other factors are shown to have additional impact on predictions of range size or inter-island dispersal ability. Our analyses cannot determine the causal mechanisms of these correlations, but may stimulate further research on the adaptive significance of traits affecting range size and dispersal in this system.