Data from: Drosophila wing modularity revisited through a quantitative genetic approach

To predict the response of complex morphological structures to selection
it is necessary to know how the covariation among its different parts is
organized. Two key features of covariation are modularity and integration.
The Drosophila wing is currently considered a fully integrated structure.
Here, we study the patterns of integration of the Drosophila wing and test
the hypothesis of the wing being divided into two modules along the
proximo-distal axis, as suggested by developmental, biomechanical, and
evolutionary evidence. To achieve these goals we perform a multilevel
analysis of covariation combining the techniques of geometric
morphometrics and quantitative genetics. Our results indicate that the
Drosophila wing is indeed organized into two main modules, the wing base
and the wing blade. The patterns of integration and modularity were highly
concordant at the phenotypic, genetic, environmental, and developmental
levels. Besides, we found that modularity at the developmental level was
considerably higher than modularity at other levels, suggesting that in
the Drosophila wing direct developmental interactions are major
contributors to total phenotypic shape variation. We propose that the
precise time at which covariance-generating developmental processes occur
and/or the magnitude of variation that they produce favor proximo-distal,
rather than anterior-posterior, modularity in the Drosophila wing.