• Crustaceans;
  • direct development;
  • diversity;
  • latitudinal gradient;
  • lecithotrophic;
  • molluscs;
  • planktotrophic;
  • species richness;
  • temperature



It has been proposed that the mode of larval development (MLD) could explain the exceptions to the latitudinal gradient of species richness exhibited by several groups of marine organisms. We evaluate the generality of the latitudinal patterns of species richness for different MLDs, and analyse the effects of environmental variables on species richness – explicitly considering feeding type during the larval phase – to identify the relevance of the dichotomies between feeding and non-feeding larvae and between pelagic and non-pelagic development.


South America, from 10° to 54° S in the south-western Atlantic and from 18° to 54° S in the south-eastern Pacific.


Using molluscs and crustaceans in two oceans, we deconstructed latitudinal patterns of species richness for three MLDs (planktotrophic, lecithotrophic and direct development). The effects of sea water temperature, chlorophyll a concentration and shelf area on the spatial variation in species richness were analysed for each MLD.


We found remarkably similar trends in species richness deconstructed by MLD. Planktotrophic species richness decreased polewards, and the converse pattern was observed in direct developers. In the majority of groups with lecithotrophic development, species richness increased from middle to higher latitudes. The main component explaining the variation in species richness was the spatial structure of sea water temperature. Temperature showed positive effects on planktotrophic species and negative effects in almost all lecithotrophic or direct-developing groups in both oceans and both taxa.

Main conclusions

The similar relationship of temperature with lecithotrophic and direct developers, which contrasts with planktotrophic species, suggests that food availability during planktonic larval development is a critical factor determining species distribution of marine invertebrates. We ruled out the cost of brooding as a single causal factor explaining the poleward increase in diversity of direct developers. Simple linear models can describe changes in the proportion of species with temperature within each MLD.