United States
£ GBP
A phylogeny for the Notodontidae was constructed based on cladistic relationships among 52 notodontid species. The study included taxa from all previously recognized family group categories and all biogeographic regions. Twelve species, chosen from the rest of the Noctuoidea, served as an outgroup. The data set consisted of 174 morphological characters (100 from adults and 74 from final instar larvae) comprising 437 character states.
Cladistic analyses using the Hennig86 parsimony program produced eight equally parsimonious trees, each with a length of 853 steps, a consistency index of 0.30, and a retention index of 0.66. All morphological characters are described and all character states are figured using line drawings, photographs, and scanning electron micrographs. Adult and larval habitus photographs for taxa used in the analysis are also included. The Notodontidae, as here defined, is a monophyletic group. Six adult and five larval synapomorphies support monophyly, but none occur universally. The genus Doa Neumoegen and Dyar (Doidae) is hypothesized to be the sister-group of the Notodontidae.
Oenosandra Walker, an Australian genus formerly referred to the notodontid subfamily Thaumetopoeinae, is instead placed as the sister group of Doidae + Notodontidae. The Oenosandridae, new family, is proposed. In a reclassification of the Notodontidae, nine monophyletic subfamilies are recognized: the Thaumetopoeinae, Pygaerinae, Platychasmatinae, Notodontinae, Phalerinae, Dudusinae, Heterocampinae, Nystaleinae, and Dioptinae. The Hemiceratini of previous authors is of uncertain status, and placement of the genus Lirimiris requires further research. The new tribe Scranciini is placed in the Dudusinae. A key to notodontid subfamilies is presented, and apomorphic character states, as well as diagnoses, are listed for each group. General topics related to phylogeny of the Notodontidae are discussed. First, the question of whether taxon addition will affect the proposed classification is addressed. Platychasma virgo Butler, previously placed in its own notodontid subfamily, the Platychasmatinae, is used as a test case. When added to the analysis this species arises as a basal lineage in the Notodontidae. Topology in the rest of the cladogram is unaffected. The subfamily Platychasmatinae is retained.
The proposed hypothesis of relationships among subfamilies is roughly similar to previous concepts of notodontid phylogeny. For example, the current scheme supports placement of the Thaumetopoeinae and Pygaerinae as basal groups and placement of the Dioptinae and Nystaleinae as derived sister groups. However, in most other respects this new phylogeny differs from previous hypotheses. Larval and adult data were also analyzed separately. Consensus trees for larvae and for adults show poor resolution compared to the tree based on all data. Monophyly of some subfamilies (e.g. the Dudusinae) is supported exclusively by adult synapomorphies, and of others (e.g., the Heterocampinae) by larval characters. These results suggest that constructing a stable classification for the Notodontidae requires data from both life stages. As a group, the Notodontidae do not show obvious larval host association patterns. Nevertheless, cladistic analyses suggest that almost all Salicaceae-feeders belong in a monophyletic group and that, for the most part, Fagaceae specialists are restricted to the Heterocampinae. The Dioptinae, unique among Notodontidae in being diurnal and in typically feeding on toxic plants, appear to have evolved from a nocturnal ancestor that utilized nontoxic hosts.
