A molecular phylogeny of the families and orders of Collembola (Arthropoda: Hexapoda) using the 18S rDNA gene

In this study I investigate the phylogenetic relationships of the families and orders of Collembola, with special reference to six groups that traditionally have been problematic, or that have become problematic with the implementation of the cladistic philosophy to the systematics of the group. Some authors have questioned the monophyly of the order Poduromorpha, as well as the supraordinal categories Arthropleona and Symphypleona (s.l.). In addition the specific phylogenetic affinities of Poduridae, Actaletidae, and Tomoceridae are unresolved.; To try to clarify these relationships I sequenced the complete 18S rDNA gene from 21 species of Collembola representing all four orders and 10 of the 13 families commonly recognized. I also sequenced the 18S gene from a jumping bristletail (Microcoryphia: Machillidae), and the house centipede (Chillopoda: Scutigeromorpha). The sequences from these last two species and the sequences from the brine shrimp, Artemia salina, and the beetle Tenebrio molitor, (obtained from Genbank) were used as outgroup in the phylogenetic analyses. Phylogenetic analyses were performed using parsimony and distance (Neighbor-Joining) methods. For the parsimony analysis the data set was subdivided into eight subsets, following suggestions by Smith (1994).; Only four taxa were supported by all parsimony analyses. I have attributed the failure of 18S to support the monophyly of groups well supported by morphological characters to differences in rates of evolution among lineages, and to low stemminess in the most parsimonious trees, presumably resulting from bursts of radiation in relatively short periods of time. Implementation of the suggestions of Smith (1994) give an indication of the relative strength of specific nodes but can result in contradictory hypotheses of relationships. In the Neighbor-Joining analyses choice of outgroup did not affect the topology of the tree, but it did affect support for some nodes as measured by bootstrap analysis.; The monophyly of Poduromorpha was supported by all analyses. Support for a monophyletic Arthropleona was ambiguous and cannot be resolved at this time. A monophyletic Symphypleona (s.l.) was never supported. The placement of Poduridae in Poduromorpha was always supported, but its affinities within the order could not be unambiguously resolved. All of the iconoclastic propositions of the phylogenetic affinities of Actaletidae were rejected, but the question remains of whether this family is the sister group of Isotomidae or basal to all other Entomobryomorpha. The proper placement of Tomoceridae could not be resolved because in most analyses this taxon clustered with different groups.