Get access

The role of geography and environment in species turnover: phytophagous arthropods on a Neotropical legume


Correspondence and current address: Karen L. Bell, Royal Botanic Gardens Melbourne, Birdwood Avenue, South Yarra, Vic. 3141, Australia.




Broad-scale patterns in the distribution of phytophagous arthropods have been explained by species turnover due to host plant, environment, geographical distance and historical biogeography. In this study, variability due to plant species turnover was eliminated by focusing on the arthropods utilizing a single widely distributed plant species, Parkinsonia aculeata L. (Leguminosae: Caesalpinioideae). This enabled us to examine the effects of geography and climate on arthropod species turnover.


Neotropics (USA to Argentina).


A statistical approach was used to determine whether arthropod assemblages across the range of P. aculeata differed significantly between biogeographical areas. We determined whether host-specific arthropods have more restricted ranges, contributing to higher species turnover. Finally, generalized dissimilarity modelling (GDM) was used to correlate significant differences in assemblages between biogeographical provinces with environmental differences and geographical separation.


Phytophagous arthropod assemblages differed significantly between biogeographical areas. Most herbivores had a small geographical distribution. We did not detect higher species turnover for host-specific species. The GDM model explained 21.8% of the variance in species assemblages. Of the variables examined, climate and geographical distance were both better predictors of variation in phytophagous arthropod assemblages on P. aculeata than biogeographical areas, with climate the better predictor of the two.

Main conclusions

Observed patterns in turnover of the phytophagous insect fauna of P. aculeata are consistent with expected broad-scale responses to historical processes and climatic variation. Climate and geography explain similar proportions of compositional dissimilarity, and most species, regardless of host-specificity, have restricted geographical ranges. The statistically significant differences observed between arthropod assemblages in different biogeographical provinces are likely to be caused by a combination of niche-based processes and to some extent dispersal limitation. Although variability due to host plant has been eliminated in the current study, there remains no simple explanation for broad-scale patterns in the distribution of phytophagous arthropods.