Cenocimbicinae and the history of Cimbicidae

The Cimbicidae first appears in the fossil record with one fossil of the extinct subfamily Cenocimbicinae in the Selandian, which is dominant and diverse in the Ypresian and disappears after. A single specimen of the extant Pachylostictinae is known in the Ypresian. A small assemblage of the extant subfamilies Abiinae and Cimbicinae appears in the late Priabonian, the family is unknown the Oligocene, and then it reappears in the Miocene showing the modern pattern of dominance by the Cimbicinae and Abiinae (Table 1).

The family might have originated in the Paleocene, but this might also simply reflect the paucity of Paleocene and Late Cretaceous shale insect localities. They—and Symphyta in general—are almost entirely absent in amber for an unknown reason. Their rich record in the Okanagan Highlands could be an artefact of unknown taphonomic factors favouring their preservation there.

......continued on the next page

*Misspelled chromptera by Taeger et al. (2010, #2263)

Alternatively, this record might closely reflect their actual history (see Archibald et al. 2018). Cimbicidae feed on the leaves of angiosperms: Abiinae and Cimbicinae on Betulaceae, Caprifoliaceae, Dipsacaceae, Rosaceae, and Salicaceae; and the Corynidinae on Crassulaceae, Geraniaceae and Papaveraceae (the hosts of Pachylostictinae are little known) (summarized by Vilhelmsen 2019, see references therein). All of these are present in the Okanagan Highlands except the Dipsacaceae and Geraniaceae, and many genera of these families that are important today first appear there (e.g., Crane & Stockey 1987, Wolfe & Wehr 1987; Wehr & Hopkins 1994; Greenwood et al. 2005; Moss et al. 2005; Devore & Pigg 2007, 2010; Mathewes et al. 2016; Pigg & Devore 2016). The temperate montane Okanagan Highlands had a high diversity of insects and plants (woody dicots analysed) associated with climatic factors (Archibald et al. 2010), and may have been the setting for the origin of modern microthermal northern forests that the great majority of extant Cimbicidae prefer which later spread downslope into lowlands in the globally cooler Neogene (Graham 2011).

The only fossil Pachylostictinae known is from the Ypresian megathermal lowland Green River Formation. The Priabonian record of the Cimbicidae (excluding the single Baltic amber larva) is at Florissant, another upper microthermal montane forest, where only modern subfamilies are present. Such montane forest communities are rare in the insect fossil record, which might further bias our knowledge of the history of the family.

The limited morphological information available in known fossils indicates that the Cenocimbicinae forms a clade with the Cimbicinae and Abiinae based on at least two synapomorphies of the forewing: 1, cell 2r together with the pterostigma is distinctly higher than cell 3r (Corynidinae and Pachylostictinae: at most as high as 3r), and 2, crossvein 2r-m joins M distal of 2m-cu (basal in Corynidinae and Pachylostictinae). The strongly bent crossvein 2r-m of Cenocimbicinae separates it from the Cimbicinae + Abiinae, where it is straight or gently curved (we treat rare occurrences of a similar 2r-m in Cimbicinae and Abiinae as convergent).

This implies an unknown presence of Corynidinae and Pachylostictinae at least to the Selandian, and of Cimbicidae originating earlier in the Paleocene or perhaps the Late Cretaceous.