Chaetophractus vellerosus Gray, 1865

Chaetophractus vellerosus Gray, 1865 View in CoL

Referred material: Five subcomplete vertebrae in connection with three bony plates (= three bands) of which seven osteoderms of the movable bands from the dorsal shield are preserved: PIMUZ A/ V 423 ( Fig. 3 View Fig ).

Comment: PIMUZ A/V 423 was originally assigned to Dasypus proximus Lydekker, 1895 (Additional file 1: Table S1). In his work from 1887, Ameghino named the species Proeuphractus proximus Ameghino, 1887 , later reassigned to D. proximus by Lydekker (1895). I could not find any taxonomic work mentioning this species. Yet Castro (2015) defined as valid species of the Dasypodini several genera and species, four of which occur in the Pleistocene Pampean region, one extinct, P. sulcatus (see above), and the others still extant Dasypus hybridus Desmarest, 1804 , Dasypus novemcinctus Linnaeus, 1758 , (or Dasypus sabanicola Mondolfi, 1968 , see Feijó et al., 2019), and Dasypus septemcinctus Linnaeus, 1758 , whose fossil record is unknown (Feijó, 2020). Te vertebrae of PIMUZ A/V 423 are particularly broken and are not diagnostic. In contrast, the osteoderms are of small size and exhibit a large and smooth cranial portion, a well-marked transverse depression, and a caudal portion bearing two main parallel sulci (separating the central figure from the peripheral figures) with four foramina on each for the most complete osteoderm. None of the members of the Dasypodini mentioned above have parallel main sulci. Tese groove orientations often meet in the anteriormost part of the caudal portion in this Dasypodini (e.g., Castro, 2015). Tis feature is attributable to an Euphractinae , either belonging to the genus Chaetophractus or Zaedyus Ameghino, 1889 , which is in agreement with their occurrence in the Ensenadan of Buenos Aires province (Soibelzon et al., 2010). Given the small size of the specimen, I favor a reattribution to C. vellerosus or Z. pichiy Desmarest, 1804 (Carlini et al., 2016; Superina & Abba, 2014). Te only difference mentioned for a distinction of these two species on the osteoderms of the mobile bands is a less "sharp" shape in C. vellerosus (Carlini et al., 2016; Superina & Abba, 2014). As the specimen presents relatively flat and rectangular osteoderms, I favor its attribution to C. vellerosus but I note that this reattribution has to be taken with caution with regard to the little difference that exists with the species Z. pichiy . It is also to be noted that the recent phylogenetic analyses tend to discuss the relevance of differentiating these two species (e.g., Abba et al., 2015; Gibb et al., 2016). A point of interest brought by this specimen is the contribution of a partial axial skeleton in connection with the dorsal shield. Te armadillo vertebral column has been studied mainly in D. novemcinctus for functional aspects (e.g., Gaudin & Biewener, 1992), or across extant armadillos in a study of the evolution of their axial skeleton (e.g., Galliari et al., 2010), but fossil specimens have rarely been the subject of comparative studies.