Mioquerquedula soporata (Kurochkin, 1976), comb. nov.
Anas soporata: Kurochkin, 1976, p. 61, text-fig. 8; 1985, p. 43, text-fig. 18, pl. 5, figs. 1–6 (part.); Zelenkov and Kurochkin, 2012, p. 92, text-figs. 1b, 1c, pl. 17, fig. 3; 2015, p. 171, pl. 15, figs. 23–25.
Dendrocygna soporata: Mlíkovský and Švec, 1986, p. 262. Anatinae gen. indet.: Zelenkov and Kurochkin, 2012, p. 94.
Anas velox (part.): Cheneval, 1987, p. 142, non pl. 1, figs. 2–4; 2000, p. 32; Mlíkovský, 2002, p. 118.
Holotype. PIN, no. 2614/95, cranial half of right coracoid; Mongolia, Gobi-Altai aimag, Shargain-Gobi, Sharga locality; uppermost middle Miocene, Oshin Formation
Description. In the humerus, the tuberculum dorsale is subtriangular and slightly elevated above the
plane of the bone; the caudal shaft ridge is not strongly pronounced and is oriented towards the tuberculum dorsale; the proportions of the humerus are shortened; bone shaft moderately gracile; the fossa brachialis is close to the tuberculum supracondylare ven- trale; the epicondylus dorsalis is well defined, but not large. The coracoid has a groove passing from the medial edge of the shaft to the ventral surface of the sternal expansion; the impressio m. supracoracoidei is moderately expressed; general proportions shortened; see also the description of the coracoid in Zelenkov and Kurochkin (2012).
Comparison. It differs from M. minutissima in the slightly larger size (corresponding to modern Nettapus coromandelianus or the smallest specimens of Anas crecca), craniocaudally somewhat more elongated (advanced character), and medially significantly more strongly protruded (plesiomorphic character) processus acrocoracoideus of the coracoid.
Remarks. Anas soporata is assigned here to the genus Mioquerquedula on the basis of the similar structure of the processus acrocoracoideus, cotylascapularis, processus procoracoideus and generally
similar coracoid proportions to M. minutissima. Previously, only fragmentary coracoids were reported for Anas soporata (Zelenkov and Kurochkin, 2012, 2015); however, the revision made it possible to attribute several fairly complete specimens from the type locality of Sharga to this species. Based on these coracoids (as well as the almost complete specimen MNHN, SA 10283 from Sansan), it can be seen that the coracoid of Anas soporata was shortened compared to that of modern Anatini, somewhat like that of Nettapus. While the length of the glenoid articular part (facies articularis humeralis + cotyla scapularis) is similar, the length of the bone (from the apex of the processus procoracoideus to the angulus medialis) is noticeably shorter in Mioquerquedula than in Anas s.l. In extant Nettapus, the coracoid is shortened somewhat more than in Mioquerquedula; an even shorter coracoid occurs in Malacorhynchus, which M. soporata resembles somewhat in the morphology of the processus acrocoracoideus. Previously, one of the distinguishing features of M. soporata was considered to be the presence of a pneumatic opening under the facies articularis clavicularis (Zelenkov and Kurochkin, 2012). We attribute the presence of this opening in the M. soporata holotype to variability (probably intrageneric), since individual variability in the development of pneumatization was found in extant Nettapus (our data). It should be noted that significant variability of the acrocoracoid part also characterizes the coracoids of M. minutissima (see above). It is possible that M. soporata may actually represent a separate anatid genus (this may be indicated by a slightly longer processus acrocoracoideus, an advanced character for Anatinae), but the general morphological similarity of
M. soporata and M. minutissima and their identical age indicates at least their equal evolutionary grade, and likely close affinity. A judgement about the generic independence of M. soporata would still be premature, but can be later confirmed by additional evidence for oth species. First and foremost, it seems important to note the incorrectness of assigning M. soporata to the Anas s.l. group, as was done earlier (Kurochkin, 1985; Zelenkov and Kurochkin, 2012, 2015).
The proximal humerus from Sharga (PIN, no. 4869/107) is here assigned to this species. Although poorly preserved, it reliably differs from Anas in the subtriangular tuberculum dorsale, as in primitive ducks (in Anas, this tubercle is always elongated). The short and raised tuberculum dorsale also characterizes Mioquerquedula sp. from the Rudabánya locality (Zelenkov, 2017). In these humeri, a slightly pronounced caudal shaft ridge is visible, oriented towards the tuberculum dorsale, while in Anas s.l. this
ridge is not expressed. An incomplete humerus from Sharga (PIN, no. 4869/54), also assigned to M. soporata here, corresponds in size to N. coromandelianus, as does the holotype. PIN, no. 4869/54 differs from Anas s.l. in the noticeably less robust shaft (a plesiomorphic feature for Anatinae) and its shortening: while absolute dimensions of the distal end are similar to those of small specimens of extant A. crecca, this humerus is shorter (the distance from the distal margin to the distal point of the crista deltopectoralis is 10% shorter in M. soporata).
Further, a number of bones from the Sansan locality are here attributed to M. soporata, including the
coracoid (MNHN, no. SA 10283) that is almost identical to the holotype of A. soporata both in morphology and in size. This specimen is characterized, among other things, by the presence of a groove passing onto the ventral surface of the bone, as in M. minutissima. The only difference between MNHN, no. SA 1232 and the holotype of M. soporata is the shape of the base of the processus acrocoracoideus in ventral view: in the holotype of M. soporata, the angle between the impressio bicipitalis and the adjacent part of the shaft at the level of the facies articularis humeralis is smoothed out, while in MNHN, no. SA 10283 it is cut, resulting in an angular notch. This difference most likely is a feature of intraspecific variability, as variations in the shape of the caudal edge of the processus acrocoracoideus are characteristic of the coracoids of M. soporata from Sharga and extant anatids.
The scapula (PIN, no. 4869/80) corresponds in size to that of extant N. coromandelianus and A. crecca.
This specimen is characterized by a relatively short acromion, as in modern Nettapus, and, apparently, unpronounced curvature (however, the latter feature is difficult to assess due to the incomplete preservation of the bone shaft). Almost complete (without os metacarpale minus) carpometacarpi from Sharga (PIN, no. 4869/242) and Sansan (MNHN, no. SA 1250) are tentatively attributed here to M. soporata: they are very similar in size and proportions, in the orientation of the processus extensorius, the length of the distal symphysis, and other features. The weathered nature of PIN, no. 4869/242, apparently explains the somewhat smaller proximal width of this specimen compared to MNHN, no. SA 1250. In the size of the proximal end, PIN, no. 4869/242 corresponds to extant N. coromandelianus, but is somewhat elongated and has a slightly thinner os metacarpale major. The general proportions of carpometacarpus are similar to those of Anatini. This specimen is similar to Nettapus and Anas and
differs significantly from Oxyurinae in the high processus extensorius and the elongated symphysis distalis. Advanced features indicating its attribution to Anatinae include the presence of a well-defined notch in the dorsal margin of the trochlea carpalis in caudal view and a proximodistally shortened base of the os metacarpale alulare (in Oxyurinae, in particular, the base of the os metacarpale alulare is elongated, the processus extensorius is low, and the notch in the dorsal margin of the trochlea carpalis is unpronounced). Importantly, the dorsal margin of the trochlea carpalis protrudes proximally only slightly more than the ventral one (this is clearly seen when viewed caudally and ventrally), as in Anas s.l. and unlike the Mergini, but more than in Mionetta. This specimen also differs from Mionetta in a number of features: the ventral surface of the processus extensorius and the proximal end is not concave (concave in Mionetta), on the caudal surface of the os metacarpale minus near its base there is a fossa (unpronounced in Mionetta), the fossa infratrochlearis caudalis is barely expressed and small (deep in Mionetta, Oxyurinae and Mergini). All these characters make this specimen similar to Anatinae.
Several bones were previously incorrectly identified as Anas/Mioquerquedula (Zelenkov and Kurochkin, 2012). The femur (PIN, no. 4869/5) belongs to the Strigidae; another femur (PIN, no. 4869/99) also does not belong to Anatidae and, apparently, represents Phasianidae. PIN, no. 4869/19, fragment of an ulna, also does not belong to Anatidae.
Range. Middle Miocene of France (Sansan; MN 6) and Mongolia (Sharga; MN 7+8).
Material. From the type locality, in addition to the holotype: PIN, nos. 4869/71, 102, 144, 189, 192, 240, 243, 244, incomplete left coracoids; 4869/156, 159, 246, incomplete right coracoids; 4869/80, left incomplete scapula; 4869/107, proximal ends of right humeri; 4869/54, incomplete left humerus; 4869/25, distal epiphysis of the right humerus; 2614/119, proximal half of right ulna; 4869/242, right carpometacarpus; 4869/241, distal fragment of the right carpometacarpus. From Sansan, France: MNHN, no. SA 10283, right coracoid; SA 1250, right carpometacarpus; SA 1213, distal half of the left tibiotarsus.
Fred