Send to

Choose Destination
J Evol Biol. 2008 Sep;21(5):1201-19. doi: 10.1111/j.1420-9101.2008.01573.x. Epub 2008 Jul 24.

Measuring and comparing evolvability and constraint in multivariate characters.

Author information

Department of Biology, Center for Ecological and Evolutionary Synthesis, University of Oslo, Oslo, Norway.

Erratum in

  • J Evol Biol. 2009 Apr;22(4):913-5.


The Lande equation forms the basis for our understanding of the short-term evolution of quantitative traits in a multivariate context. It predicts the response to selection as the product of an additive genetic variance matrix and a selection gradient. The selection gradient approximates the force and direction of selection, and the genetic variance matrix quantifies the role of the genetic system in evolution. Attempts to understand the evolutionary significance of the genetic variance matrix are hampered by the fact that the majority of the methods used to characterize and compare variance matrices have not been derived in an explicit theoretical context. We use the Lande equation to derive new measures of the ability of a variance matrix to allow or constrain evolution in any direction in phenotype space. Evolvability captures the ability of a population to evolve in the direction of selection when stabilizing selection is absent. Conditional evolvability captures the ability of a population to respond to directional selection in the presence of stabilizing selection on other trait combinations. We then derive measures of character autonomy and integration from these evolvabilities. We study the properties of these measures and show how they can be used to interpret and compare variance matrices. As an illustration, we show that divergence of wing shape in the dipteran family Drosophilidae has proceeded in directions that have relatively high evolvabilities.

[Indexed for MEDLINE]
Free full text

Supplemental Content

Full text links

Icon for Wiley
Loading ...
Support Center