The systematics of carnivorous sponges.

Carnivorous sponges are characterized by their unique method of capturing mesoplanktonic prey coupled with the complete or partial reduction of the aquiferous system characteristic of the phylum Porifera. Current systematics place the vast majority of carnivorous sponges within Cladorhizidae, with certain species assigned to Guitarridae and Esperiopsidae. Morphological characters have not been able to show whether this classification is evolutionary accurate, and whether carnivory has evolved once or in several lineages. In the present paper we present the first comprehensive molecular phylogeny of the carnivorous sponges, interpret these results in conjunction with morphological characters, and propose a revised classification of the group. Molecular phylogenies were inferred using 18S rDNA and a combined dataset of partial 28S rDNA, COI and ALG11 sequences. The results recovered carnivorous sponges as a clade closely related to the families Mycalidae and Guitarridae, showing family Cladorhizidae to be monophyletic and also including carnivorous species currently placed in other families. The genus Lycopodina is resurrected for species currently placed in the paraphyletic subgenus Asbestopluma (Asbestopluma) featuring forceps spicules and lacking sigmas or sigmancistras. The genera Chondrocladia and Cladorhiza are found to be monophyletic. However, results indicate that the subgenus Chondrocladia is polyphyletic with respect to the subgenera Meliiderma and Symmetrocladia. Euchelipluma, formerly Guitarridae, is retained, but transferred to Cladorhizidae. The four known carnivorous species currently in Esperiopsis are transferred to Abyssocladia. Neocladia is a junior homonym and is here renamed Koltunicladia. Our results provide strong evidence in support of the hypothesis that carnivory in sponges has evolved only once. While spicule characters mostly reflect monophyletic groups at the generic level, differences between genera represent evolution within family Cladorhizidae rather than evolution of carnivory in separate lineages. Conflicting spicule characters can be reinterpreted to support the inclusion of all carnivorous sponges within Cladorhizidae, and a carnivorous habit should thus be considered the main diagnostic character in systematic classification.