ONTOGENY AS A CONSTRAINT IN THE EVOLUTION OF THE TRILOBITE GLABELLA (Invited Presentation)

With a rich fossil record spanning 270 million years, trilobites offer an excellent opportunity to study patterns and processes of phenotypic change over micro- to macroevolutionary timescales. Integrative approaches that combine ontogenetic, morphometric, and phylogenetic data allow investigation of whether and how disparification was constrained by development, and of the timescale over which any developmental constraints operated. With adequate sampling of appropriately preserved material, it is possible to rigorously test hypotheses regarding constraint imposed by the degree of static trait variance (e.g., canalization), the structure of static trait covariance (e.g., integration/modularity), and by the structure of dynamic trait covariance (ontogenetic allometry). Here, I explore disparification of the shape of the glabella—a structure of ecological and phylogenetic relevance—and test whether the pattern of ontogenetic shape change served as a line of least resistance (LOLR) for evolutionary change. A dataset consisting of exemplars from most trilobite clades depicts the broad pattern of disparification across Trilobita, revealing the extent to which particular shapes of glabella—perhaps reflecting distinct trophic guilds—characterize particular groups or were convergently evolved. More detailed phylomorphospace- and phyloallometryspace-based analyses of numerous olenelloid, corynexochoid, and “ptychoparioid” trilobites find that individual species underwent considerable ontogenetic shape change, often exceeding the phenetic distance between clades. Over short phylogenetic distances it is common to find a relatively conserved pattern of ontogenetic shape change and a reasonably close parallelism between ontogeny and phylogeny; evolutionary change often occurs by allometric scaling (heterochrony) or by lateral transposition (a modification of starting form). However, such constraints rarely predict longer-term evolutionary trends, over which it is common to find substantial changes in the pattern of ontogenetic shape change and thus no constraining LOLR. Such findings shed light on issues such as trait (in)dependence and the timescale of decay of developmental constraints, and have broad implications for the selection of appropriate evolutionary models.