CORDULIIDAE, Kirby, 1890

Kohli, Manpreet Kaur, Ware, Jessica L. & Bechly, Günter, 2015, How to date a dragonfly: Fossil calibrations for odonates, Palaeontologia Electronica 19, pp. 1-14 : 8-9

CROWN CORDULIIDAE

Node Calibrated: (9) Crown Corduliidae . This node represents the crown Corduliidae .

Species. Croatocordulia platyptera ( Corduliidae ) de Charpentier, 1843

Specimen. The holotype specimen (described without collection number), a male dragonfly specimen preserving three wings and body fragments, was loaned to Charpentier by the famous German botanist Heinrich Göppert (Breslau). It is lost according to Kiauta (1969). Our inquiry showed that the fossil is not preserved at the University of Wrocław in Poland, where most of the remaining specimens of Göppert’s private fossil collection are deposited in the Geological Museum and the Institute of Evolutionary Biology and Ecology). It is also not in the Radoboj collection of Oswald Heer at the ETH Zürich ( Switzerland), even though Heer (1849) had examined and redescribed this fossil specimen. A figure of this fossil can be found in Heer (1849), on page 5 and figure number 3b.

Phylogenetic Justification. Kiauta (1969) suggested a close relationship of this fossil genus with the recent Corduliid genera Cordulia , Dorocordulia , and Somatochlora , based on a comparison of wing venational characters. This attribution was followed by Nel and Paicheler (1994). Several synapomorphies confirm an attribution to Corduliidae + Libellulidae , like the hind wings with triangle recessed to arculus and a boot-shaped anal loop with midrib. The presence of an anal triangle and anal angle excludes a position in Libellulidae . The presence of only one crossvein beneath the pterostigma, as well as the phenetic similarity of the venation pattern in both pairs of wings, confirms an attribution to Corduliidae .

Minimum Age. 12.7 Ma.

Age Justification. Found in Radoboj (Radoboy) near Krapina, north of Zagreb in Croatia, late middle Miocene, Serravalian/early Sarmatian, lacustrine limestone. According to Mlíkovský (1997), the Radoboj site was specified as Sarmatian, which corresponds to the younger part of the middle Miocene, or to the micromammal Neogene zones MN 7–8. This author mentioned that micropaleontological research of the middle Miocene deposits around Krapina dated two of the localities (Lopatica and Gornja Šemnica) as early Sarmatian, and one (Frug) in the nannoplankton zones 6–7. Mlíkovský (1997) considered it as probable from this context that the classical Radoboj site belongs in micromammal zone MN 7, which lasts from 14.8–12.7 Ma according to Steininger et al. (1996).

Discussion. Two older alleged crown group Corduliidae have been described: Stenogomphus ? scudderi was described by Cockerell (1921) from the middle Eocene Green River Formation from Green River, Wyoming ( USA), and Stenogomphus carletoni was described by Scudder (1892) from the late Eocene Florissant Formation of Colorado ( USA). Unfortunately, the attribution of Stenogomphus to Corduliidae is dubious and would need a thorough revision of the type material ( Nel and Paicheler, 1994). Ris (1910) considered Stenogomphus carletoni to be closely related to the recent genera Neurocordulia , Platycordulia , and Aeschnosoma , which was considered by Nel and Paicheler (1994) as the most probable hypothesis. If this attribution could be confirmed by a revision of the holotype, it would extend the minimum age of crown group Corduliidae to the Oligocene.

We did not use Molercordulia karinae from the Palaeocene Moler Formation of Denmark as a calibration point because its attribution to Corduliidae by Bechly (2005) was based on symplesiomorphies. It cannot be ruled out that Molercordulia could be a stem group representative of Corduliidae , a stem group representative of Libellulidae , or even an advanced stem group representative of the clade Corduliidae + Libelulidae.