Abstract
Animal body size has important evolutionary implications. The wolf spider genus Schizocosa Chamberlin, 1904 has developed as a model for studies on courtship, with visual and vibratory signals receiving attention; however, body size has never been carefully evaluated. Although species of Schizocosa can be distinguished from their close relatives by differences in genitalic structures, male ornamentation, and behavior, some species are morphologically similar, making diagnosis, and identification difficult. Evaluation of species boundaries using genetic data across Schizocosa is limited. The similar species S. maxima Dondale & Redner, 1978 and S. mccooki (Montgomery, 1904) are separated predominantly on the basis of size differences, with S. maxima being larger. We evaluate the evolution of size in these two Schizocosa species distributed in western North America, where gigantism of S. maxima is hypothesized to occur, particularly in California. We sampled subgenomic data (RADseq) and inferred the phylogeny of S. mccooki, S. maxima, and relatives. We apply a variational autoencoder machine learning approach to visualize population structuring within widespread S. mccooki and evaluate size within the context of a comparative phylogenetic framework to test the hypotheses related to genetic clustering of populations and gigantism. Our data show S. mccooki populations are not genealogically exclusive with respect to S. maxima. Likewise, S. maxima individuals are not recovered as a lineage and do not form an isolated genetic cluster, suggesting that the observed differences in size cannot be used to accurately delimit species. The cause of gigantism in S. maxima remains unexplained, but provides a framework for future studies of size variation and speciation.
Investigation of life history traits (seasonal fluctuations of body size and clutch size, and the breeding season) of the wolf spider, Pardosa astrigera in the plain area of Mie Prefecture, Japan
