Setting the record straight for fossil flying fishes versus non-flying ones: a comment on Xu et al. (2015)

Fossil flying fishes have been known for 150 years from the Upper Triassic of the Alps [1] and recently from the late Middle Triassic of China [2]. So far, five species belonging to Thoracopterus have been described and included in the Peltopleuriformes family Thoracopteridae [2]. Now, Wushaichthys exquisitus [3], possibly from the Xingyi Fauna (Guizhou Province, China) of Late Ladinian age, has been described as belonging in this family.

The systematic palaeontology in Xu et al. [3] appears at minimum contradictory, as the supposed synapomorphies of Thoracopteridae are actually plesiomorphies widespread among Peltopleuriformes, most of them not considered in their analysis. The original diagnosis of Thoracopteridae [4] was amended [2,5] after careful revisions of all available specimens, leading to the correction of earlier [5,6] anatomical misinterpretations. A new diagnosis was not proposed [3], and the traditional thoracopterid characteristics were not questioned [3,7], so it is curious that the authors have not realized that Wushaichthys does not possess the actual family synapomorphies. Furthermore, the generic diagnosis is totally uninformative and, because of its combination of characteristics, W. exquisitus could well be assigned to already existing Peltopleuridae genera, for whom extensive recent literature is available [5,8,9]. The phylogenetic analysis presented does not include a significant sampling of the order Peltopleuriformes. Thus, its results are biased by the exclusion of taxa that may be significant to the conclusions drawn.

Gradual achievement of the flight capability is also questionable, as all five Thoracopterus species show a very similar ratio in elongation of the paired fins, independently from the scale covering of the body [2,5].

Wushaichthys exhibits sexual dimorphism [3], as do several other Triassic fishes [2,5,8–11]. Modified anal fins do not imply internal fertilization, ovoviviparity and, even less, viviparity, especially if the anal fin is not modified in an actual gonopodium as in extant poeciliid [12], which are ovoviviparous and not viviparous. Wushaichthys, and all other Peltopleuriformes [2,8–10], have no gonopodium-like structure, the anal fin being large, not narrow and elongated, and with hooklets all along the posterior margin, not just at the tip.

The provenience and age of the material described is also problematic. Regarding the origin of the specimens, they are said to come from the Wusha area. Therefore, the only exhaustive excavation that conforms to the scientific standards around Wusha is the one near Nimaigu village [2]: as those specimens are not from that excavation, the outcrop and fossiliferous level/bed they are from, and their exact age, should be specified. The proposed ‘Middle Triassic’ age [3] is too vague, when the precise locality is also obscure.

Finally, Potanichthys xingyiensi [7] has been included as a valid species [3]. However, it has been already synonymized [13] to Thoracopterus wushaensis [2], following the priority rule of the International Code of Zoological Nomenclature, where two names relate to the same species. However, a question remains to be solved about the site and stratigraphic level yielding Potanichthys. When the paper by Xu et al. [7] was published online, after the authors were aware of Thoracopterus specimens from the Nimaigu excavation [2], their specimens were described as being from the Ladinian of Xiemi, Wusha, from where no other fossils have previously been published, not even Wushaichthys, which is much more common than Thoracopterus in the same fossiliferous horizon (personal observation in Nimaigu excavation, 2012).

In conclusion, the morphological features of W. exquisitus are very similar to species of Peltopleurus and Peripeltopleurus, closely related genera of Peltopleuridae [5,8,9]: four of the supposed synapomorphies for thoracopterids sensu [3] are actually widespread characters among all Peltopleuriformes, and the remaining two have not been described in both Wushaichthys and Peripeltopleurus spp.

Data accessibility

No new data have been used for this comment.

Competing interests

The author declares no conflict of interest.

Funding

The author received no funding for this study.