New results concerning the morphology of the most ancient dragonflies (Insecta: Odonatoptera) from the Namurian of Hagen-Vorhalle (Germany)

The holotype specimen of the 'protodonate' Erasipteroides valentini (Brauckmann in Brauckmann et al., 1985) and the paratype specimen K-13 of the giant 'protodonate' Namurotypus sippeli Brauckmann and Zessin, 1989 from the Upper Carboniferous (Namurian B) of Hagen-Vorhalle (Germany) are redescribed, and a new specimen of Erasipteroides cf. valentini is described. The new evidence is used to refine the groundplan reconstruction of Odonatoptera and the reconstruction of odonatoid phylogeny. Prothoracic winglets for Erasipteroides and the absence of an archaedictyon are documented. Furthermore, a very long and sclerotized ovipositor with gonangulum is described from the female holotype specimen of Erasipteroides valentini, and it is proposed that it was not used for endophytic but for endosubstratic oviposition, The record of prothoracic winglets in early, odonatoids, and their presence in fossil Palaeodictyoptera and 'protorthopteres', indicates that the groundplan of Pterygota indeed included three pairs of wings. A phylogenetic analysis suggests that the Palaeozoic giant Meganisoptera and "higher" odonatoids (incl. crowngroup Odonata) together form a monophyletic group which is here named Euodonatoptera. Erasipteroides and the other 'Erasipteridae' are shown to be more closely related to Euodonatoptera than to Eugeropteridae, The description of the male primary genital structures of Namurotypus sippeli is emended and a new interpretation is proposed. including new hypotheses concerning their function. The males of Namurotypus had a paired penis with a pair of lateral parameres, and a pair of leaf-like, but still segmented, gonopods. Segmented leg-like male gonopods are considered as a groundplan character of insects, while a paired penis is regarded as a putative synapomorphy of the palaeopterous insect orders Palaeodictyopteroida, Ephemeroptera, and Odonatoptera. It is proposed that Namurotypus did not mate by direct copulation but retained the archaic deposition of external spermatophores, just like the primarily wingless insects, The sigmoidal male cerci may have been placed behind the female head and used to drag the female over the spermatophore, which is remotely similar to the mating behaviour of some extant arachnids (e.g. Amblypygi). Three hypothetical scenarios regarding the evolution of secondary copulation in modern Odonata are proposed.