TESTING HISTORICAL HYPOTHESES OF MORPHOLOGICAL CHANGE: BIOMECHANICAL DECOUPLING IN LORICARIOID CATFISHES.

The "decoupling hypothesis" has been proposed as a mechanistic basis for the evolution of novel structure and function. Decoupling derives from the release of functional constraints via loss of linkages and/or repetition of individual elements as redundant design components, followed by specialization of one or more elements. Examples of apomorphic decoupling have been suggested for several groups of organisms, however there have been few empirical tests of explicit statements concerning functional and morphological consequences of decoupling. Using the loricarioid catfishes, we tested one particular consequence of decoupling, the prediction that clades possessing decoupled systems having increased biomechanical complexity will exhibit greater morphological variability of associated structures than outgroups having no such decoupled systems. Morphometric procedures based on interlandmark distances were used to quantify morphological variance at three levels of design at successive nodes in the loricarioid cladogram. Additional landmark-based procedures were used to localize major patterns of shape change between clades. We report significantly greater within-group morphometric variance at all three morphological levels in those lineages associated with decoupling events, confirming our predictions under the decoupling hypothesis. Two of 12 comparisons, however, yielded significant variance effects where none were predicted. Localization of the major patterns of shape change suggests that disassociation between morphological and functional evolution may contribute to the lack of fit between variance predictions and decoupling in these two comparisons.