Archaeognatha, Euzygentoma

Zrzavý, Jan, 2008, Four chapters about the monophyly of insect ‘ orders’: A review of recent phylogenetic contributions, Acta Entomologica Musei Nationalis Pragae 48 (2), pp. 217-232 : 218-220

Archaeognatha, Euzygentoma , and Tricholepidion : one more insect ‘order’?

As both groups of the primarily wingless true insects, the bristletails ( Archaeognatha ) and the silverfish ( Zygentoma ), are superficially similar morphologically, they were united in one group (‘Thysanura’). It has long been known, however, that the bristletails are in general more basal and that the silverfish are more closely related to the winged insects (together forming the Dicondylia) than to the bristletails.

It may be that things are not that simple. The apparently primitive zygentoman family Lepidotrichidae was originally described from the mid-Eocene Baltic amber genus Lepidotrix Menge, 1854 , and is now represented by a single living species, Tricholepidion gertschi Wygodzinsky, 1961 , from northern California. The family Lepidotrichidae itself may not be monophyletic, since Tricholepidion Wygodzinsky, 1961 (the only zygentoman with distinct ocelli) may be more primitive than Lepidotrix and less closely related to the Euzygentoma (= Nicoletiidae s. lat. + Lepismatidae + Maindroniidae ). The Lepidotrichidae possess large abdominal sterna with posteriorly attached coxopodites, and a large number of pregenital abdominal styli and eversible sacs. Five characters have been proposed to support intrazygentoman position of Tricholepidion (see GRIMALDI & ENGEL (2005) and references therein). They include (i) unique sensillar structures on the terminal filament (shared by Tricholepidion and the Nicoletiidae ), (ii) a widened apical segment of the labial palp, (iii) obliteration of the superlingua, (iv) mating behaviour (the male deposits a spermatophore on a web which he has spun during the final phase of foreplay; the female picks up the spermatophore using her ovipositor; see STURM (1997)), and (v) sperm conjugation (putatively shared by Tricholepidion and the Lepismatidae ). However, even in the nicoletiids ( Atelura Heyden, 1855 ), sperm bundles are stored in the proximal part of the testes, intermingled with dense granules and forming the so called ‘spermatolophids’. Sperm aggregation in the Zygentoma is, therefore, quite diverse ( DALLAI et al. 2001a, b, 2002) and its phylogenetic interpretation is uncertain.

Tricholepidion has five-segmented tarsi, which is the presumed plesiomorphic condition for Pterygota, while both the Archaeognatha and the Euzygentoma have two- or three-segmented tarsi. Moreover, Tricholepidion ovaries are clearly primitive and archaeognath-like, being composed of seven metamerically arranged ovarioles (compare three or five non-metameric ovarioles in the Euzygentoma and the highly variable number and arrangement of ovarioles but no ovary metamerism in the Pterygota; ŠTYS et al. 1993) and including opaque granules in early vitellogenic oocytes ( SZKLARZEWICZ et al. 2004).

A possible sister group relationship between Tricholepidion and the Euzygentoma + Pterygota has been raised by KRISTENSEN (1981), based on the retention of the intergnathal connective ligament in the Archaeognatha and Tricholepidion . A detailed analysis of the mandibles and mandibular musculature also corroborated the view that Tricholepidion is a basal dicondylian ( STANICZEK 2000). BITSCH & BITSCH (1998, 2000, 2004) in their numerical analyses of morphological characters did not find any resolution of the representatives of the Zygentoma and of the Pterygota. BEUTEL & GORB (2001) performed a numerical analysis of 115 morphological characters and concluded that Tricholepidion represents a sister group of the Euzygentoma-Pterygota clade, the latter supported by a reduced size of the postocciput and reduced pleural folds, reduced ligamentous head endoskeleton, reduced transverse mandibular apodeme, presence of musculus mandibulo-hypopharyngalis, shortened maxillary palps, and reduced or absent pregenital vesicles and styli.

In molecular analyses, the position of Tricholepidion was highly unstable, ranging from conventional placement as a sister group of the Euzygentoma ( MISOF et al. (2007): 18S ribosomal DNA) to an unorthodox intra-pterygote position next to the Odonata ( KJER (2004), using the same molecular marker). Mitochondrial genomes have not yet been conclusive, evidently owing to poor taxon sampling, but they seem to support zygentoman monophyly (see COOK et al. 2005, CAMERON et al. 2006, CARAPELLI et al. 2006). GIRIBET et al. (2004) analyzed relationships among basal hexapods on the basis of a cladistic analysis of five genes and 189 morphological characters in a simultaneous analysis. Morphological characters solely corroborated monophyletic Zygentoma ( Tricholepidion vs. Euzygentoma) but with very low support; the same applied to the multigene molecular analysis. Using a sensitivity analysis approach and testing for stability within the combined morphological-molecular analysis, the most congruent parameters resolved Tricholepidion as a sister group to the remaining Dicondylia (with very low support again), whereas most suboptimal parameter sets grouped Tricholepidion with the Euzygentoma. Finally, in the combined analysis by KJER et al. (2006), using eight gene sequences and 170 morphological characters, Tricholepidion appeared as a sister group of the Euzygentoma.

The position of the enigmatic Tricholepidion remains controversial as none of the data sets provides substantial support for either of the two competing hypotheses (i.e. monophyletic or paraphyletic Zygentoma ). It seems only clear that a sister-group relationship between Tricholepidion and the Nicoletiidae has not yet been supported by morphology and molecular analyses. By all means, Tricholepidion is the most promising candidate for a new insect ‘order’ (or, in other words, the only insect species whose ordinal classification remains uncertain), either as a sister group of the Euzygentoma, or of the Euzygentoma-Pterygota superclade.