Does range-restricted evolutionary history predict extinction risk? A case study in lizards

Giovanni Di Virgilio, Shawn W. Laffan, Stuart V. Nielsen, David G. Chapple

Research output: Contribution to journalArticleResearchpeer-review

3 Citations (Scopus)


Aim: Conservation is often prioritized by identifying regional clusters of threatened or endemic species. Another approach is to assess the evolutionary distinctiveness of groups of taxa using phylodiversity measures. However, quantification of evolutionary history has traditionally not accounted for its uneven geographical distribution due to the variation in species ranges. We assess the efficacy of phylogenetic endemism (PE) to predict high extinction risk in comparison to estimates of species range restriction (weighted endemism, WE) and phylogenetic diversity (PD). PE measures the relative range restriction of evolutionary history (lineages), while WE concentrates on the tips of the tree of life, treating all such branches as being of equal length. Location/Methods: Using New Zealand's endemic skinks and geckos, we mapped the geographical variation in their extinction risk, PE, WE and PD and measured the extent to which extinction risk exhibited phylogenetic clustering for each group. Correlations between geographical concentrations of high skink and gecko extinction risk with PE, WE and PD were calculated. Results: PE was predictive of spatial clusters of high extinction risk for geckos (r2 = 0.34, P < 0.001) while WE was markedly less so (r2 = 0.19, P < 0.001). The reverse applied to skinks, with WE most predictive of high risk (r2 = 0.26, P < 0.001). The phylogenetic signal of extinction risk was significantly conserved for geckos, but was weaker and non-significant for skinks. PE and WE were not predictive of low risk. PD was not predictive of risk. Main conclusions: PE and related measures may be predictive of extinction risk when risk is phylogenetically conserved. Mapping the geographical variation in PE could be a useful first assessment of extinction risk for many groups because phylogenies are increasingly available, while full risk status categories are not. These findings might apply to other groups and locations and warrant further investigation.

Original languageEnglish
Pages (from-to)605–614
Number of pages10
JournalJournal of Biogeography
Issue number3
Publication statusPublished - 1 Mar 2017


  • conservation
  • extinction risk
  • geckos
  • macroecology
  • New Zealand
  • phylogenetic endemism
  • skinks
  • spatial analysis
  • species range restriction
  • tree of life

Cite this