DID TRILOBITES SHOW DECLINING INTRASPECIFIC VARIABILITY?

Several physical and biological factors have been proposed to explain the unique magnitude of evolutionary innovation achieved during the Cambrian radiation. Biological factors hypothesized to account for the subsequent decline in metazoan evolutionary lability include a progressive increase in developmental integration/canalization (the genomic hypothesis) or in ecological competition (the ecospace hypothesis). Assuming that developmental and ecological constraints exert a stabilizing influence on the mature phenotype, both hypotheses predict a general decline in intraspecific morphological variability following the Cambrian radiation (which may [ecospace] or may not [genomic] be relaxed in the aftermath of major extinctions).

The complex morphology, abundance, diversity, and high preservation potential of trilobites make them an ideal group to test hypotheses of decreasing morphological variability over time. Drawing upon the results of previous studies plus new analyses presented here, there is limited support for the predicted trend. Although morphometric analyses of shape fail to find a consistent trend in the degree of intraspecific variability over short (<5 million years) or long (>100 m.y.) intervals, the extent to which this is relevant to the potential for significant macroevolutionary innovation is uncertain. Intraspecific polymorphism in qualitative morphological features is perhaps more pertinent to the issue, and here trilobites offer less equivocal results: the well-documented decline in intraspecific variation in thoracic segment number seems irrefutable; and examination of over 40 character matrices reveals that intraspecific polymorphisms are coded significantly less frequently in cladistic analyses of post-Cambrian trilobites than in analyses of Cambrian trilobites.

However, optimism must be guarded: few analyses adequately control for potential intersample differences in taphonomic, ecophenotypic, or phylogenetic components of variability; variation might represent tightly controlled alternative phenotypes rather than genuine “lack of constraint”; and meta-analyses are fraught with assumptions regarding equality of studies. Such issues should be addressed before the existence and nature of any trend is accepted.