Mustela nivalis Linnaeus, 1766

Marciszak, Adrian, Hilpert, Brigitte & Ambros, Dieta, 2024, The remarkable record of mustelids from Hunas (Bavaria, Germany), Comptes Rendus Palevol 23 (23), pp. 339-383 : 354-358

Mustela nivalis Linnaeus, 1766

( Fig. 7 B-H)

Mustela aff. praenivalis Heller, 1983: 200 ; fig. 44; pl. 6/1-4, 16-18; table 25. – Groiss 1983: 354; table 48. – Koenigswald & Heinrich 1999: 94. – Ambros 2006: 54. – Rosendahl, Ambros, Hilpert, Hambach, Alt, Knipping, Reisch & Kaulich 2011: 19; table 3/2

Mustela praenivalis Baumann, 2011: 8 .

REFERRED MATERIAL. — 5 vertebrae, scapula, baculum, 15 canines, isolated P4 and m1, maxilla fragment, 169 hemimandibles (120♁♁, 49 ♀♀), 102 humeri (63♁♁, 37♀♀), 24 ulnae (15 ♁♁, 9♀♀), 21 radii (16 ♁♁, 3 ♀♀), 100 femora (52 ♁♁, 40 ♀♀), 81 tibiae (62 ♁♁, 19 ♀♀), 33 pelvis bones (551/169 - 120 ♁♁, 49 ♀♀).

DESCRIPTION

The elongated and low mandibular body has a constant lateral thickness for its whole length ( Fig. 8). The lower margin is moderately curved and reaches the curvature peak below the m1 level. The symphysis is long and narrow. Two oval-shaped mental foramina are of similar size but are located on different levels. The mesial one is situated higher, below p2 and the distal one are located below the distal root of p3, and on a lower level than the first. Some specimens have a vestigial, longitudinal, lingual furrow of the mandibular body, but the degree of development varies substantially between individuals from weak to moderate. In most individuals this structure is absent. The strong ramus has a shape of rectangular triangle. Its mesial margin rises gently to the top, while the distal margin is almost vertical. The condylar process is massive and short, elongated laterally and inclined slightly downward on the medial side. It is positioned below the teeth row. The angular process is short and is shaped like a rectangular triangle. The deep masseteric fossa has a rounded mesial margin and reaches the m1/m2 boundary. The teeth row is straight in the occlusal view, with only the p2 crown placed at an angle of 30-40o to the rest of the teeth. The distal halves of p3-p4 are oriented disto-buccally. The teeth are positioned loosely to each other, and between them are visible small diastemas ( Fig. 8).

The elongated and oval p2 is moderately reduced ( Figs 8; 9). The two-rooted p3 is low and moderately massive. The protoconid is pushed strongly mesially and almost exactly to the axis of the tooth. The blunt mesial and rounded distal margins have a weak cingulum. The crown bears an elongated, well-developed, distal, cingular projection. The buccal edge is straight, while the lingual one is weakly convex. The two-rooted p4 is an elongated and robust tooth, with a prominent, medially positioned protoconid, which is moderately separated from the surrounding cingulum. Before the protoconid is a crescent, shallow valley. The stronger cingulum is situated on the distal margin. The mesial and distal margins are blunt, while the distal halves of the buccal and lingual edges are strongly convex. This bulge is more developed on the buccal side. The elongated and narrow m1 is moderately trenchant. The crown has a long and low paraconid and prominent, but relatively low protoconid. The trigonid is moderately long and low in relation to tooth length. The buccal margin is moderately convex. The lingual margin of the paraconid and talonid is straight, while the lingual edge of the protoconid is moderately convex. The short and moderately broad talonid bears a centrally positioned hypoconid. The reduced m2 is a moderately large, oval-shaped and one-rooted tooth with a low crown ( Figs 8; 9).

COMPARISON WITH MUSTELA ERMINEA

The morphometric analysis of the material of the smallest mustelids from Hunas was carried out based on the sex division, determined based on the combination of the length m1 and the height of the mandibular body after m1. By comparing with the sexually divided Silesian population of the extant M. nivalis , it was possible to separate the population into ♁♁ and ♀♀. The taxonomic analysis also included a comparison with the sexually divided Silesian population of the extant M. erminea , M. palerminea and M. praenivalis . Both ancient species have already been subjected to critical analysis ( Marciszak et al. 2021), which showed that many of the previously reported distinguishing features are no longer valid. For this reason, the legitimacy of their designation was not discussed here, and the focus was on comparing the given characteristics of the Hunas population with other species. In particular, the most important similarities and differences were highlighted, which in the end allowed us to reach a credible conclusion.

Compared to M. erminea , cranial and dental material of the smallest mustelids from Hunas show differences in:

1) overall smaller size, well documented by the L m1, where the mean of M. nivalis ♁♁ from Hunas (4.43 mm, 4.02-4.97 mm, n = 103) is distinctly smaller than the mean of both sexes of M. erminea (♁♁ 5.84 mm, 5.51-6.36 mm, n = 151; ♀♀ = 5.19 mm, 4.65-5.57 mm, n = 177), even if slightly overlapping with ♀♀ M. erminea ( Table 6);

2) on average shorter and shallower masseteric fossa, reaching maximally the m1/m2 boundary, with a more triangular, mesial margin, although some individuals from Hunas also have a more rounded, mesial margin. Mustela erminea possesses, longer ones on average, reaching m1/2 up to the trigonid/talonid m1 border, and deeper masseteric fossa, for which the mesial margin is almost always rounded;

3) shape of the coronoid process. In M. nivalis from Hunas, the mesial and distal edges of this process are approximately of the same length and form the same angles with the longitudinal axis of the mandible. In M. erminea , the mesial edge is more inclined than the distal one, which is almost vertical and therefore considerably shorter;

4) presence of weakly to moderately marked, but still stronger than in M. erminea , longitudinal, lingual furrow of the mandibular body;

5) proportionally larger and less reduced p2, where p2 L/ m1 L index of M. nivalis from Hunas (♁♁ 28.3, 22.9-31.9, n = 26; ♀♀ 29.4, 26.9-32.0, n = 5) is higher than that of M. erminea (♁♁ 23.1, 20.4-26.7, n = 151; ♀♀ 24.1, 20.5- 27.0, n = 178). There is apparently some overlap, but the ratios differ notably;

6) narrower p3, which is well documented by ratios, but the ranges of variability strongly overlap: M. nivalis from Hunas (♁♁ 56.0, 43.2-69.2, n = 46; ♀♀ 56.0, 46.8-66.4, n = 18) is higher than this of M. erminea (♁♁ 61.5, 54.2-68.7, n = 197; ♀♀ 60.2, 50.7-69.1, n = 230).The morphotype analysis showed variable results in both species, and the higher proportion of more advanced morphotypes in ♁♁. In ♁♁ M. nivalis from Hunas the most common are morphotypes B (43.5%, wider distally than mesially, without a buccal protrusion) and C (37.0%, distally strongly widened and the occlusal crown outline is therefore strongly asymmetrical). For Hunas ♀♀ the more typical morphotypes are A2 (50.0%), distally also not widened, but with a mesio-buccal protrusion of the crown margin, and B (22.2%). Other morphotypes in both sexes of M. nivalis from Hunas are much less common ( Table 7). In M. erminea one morphotype is clearly dominant, C (89.3%) in ♁♁ and B (74.3%) in ♀♀ ( Table 7);

7) narrower p4, with less marked difference between the mesial and distal halves of the crown, where both the means and the ranges of variability of the Ba/Bp p4 indexes almost do not coincide: M. nivalis from Hunas (♁♁ 88.8, 83.8- 93.7, n = 69; ♀♀ 90.3, 87.2-94.1, n = 29) is higher than that of M. erminea (♁♁ 74.6, 62.2-74.4, n = 248; ♀♀ 81.4, 68.3-85.8, n = 162) ( Table 6). The morphotype analysis documented the distinctly wider (in relation to the mesial) distal half of p 4 in M. erminea . In both sexes (♁♁ 94.3%, ♀♀ 82.1%) of this species the most advanced morphotype C dominated, where the crown is strongly widened buccolingually. In M. nivalis from Hunas (♁♁ 81.2%; ♀♀ 75.9%) the most common is morphotype B, where the buccal margin is still almost straight, while the lingual margin in the distal part holds a moderately developed convexity. In the Hunas population there is also the most primitive morphotype A, with a narrow and elongated crown, with the mesial and distal halves of similar width (♁♁ 14.5%; ♀♀ 20.7%) ( Table 7);

8) m1 with a proportionally lower and shorter trigonid and longer and broader talonid, where the indexes of trigonid and talonid length and breadth of ♁♁ are higher than ♀♀ in both species, and where both the means and the ranges of variability of L ta/L tr m1 and B ta/B tr m1 indexes almost do not overlap. In M. nivalis from Hunas both ratios are as follows: L ta/L tr m1 (♁♁ 40.6, 30.5-51.2, n = 103; ♀♀ 49.1, 39.2-54.4, n = 39), while B ta/B tr m1 (♁♁ 90.6, 87.1-95.8, n = 103; ♀♀ 91.7, 87.2-96.2, n = 39). The values of both indexes in M. erminea are distinctly lower: L ta/L tr m1 (♁♁ 23.9, 18.6-32.4, n = 151; ♀♀ 27.8, 20.8-33.9, n = 177), while B ta/B tr m1 (♁♁ 78.1, 66.5-87.1, n = 151; ♀♀ 82.9, 77.7-86.1, n = 177) ( Table 6);

9) much less developed m1 protoconid broadening, usually, where it is present, occurs in variable size on the lingual margin, while in M. erminea this crown bulging occurs on the buccal and lingual sides. It is well illustrated by the morphotype analysis, where for both sexes of M. erminea (♁♁ 94.7%, ♀♀ 91.5%) the evolutionarily most advanced morphotype D is dominant, broadened buccally and lingually, commonly with a lingual, additional, median root. In M. nivalis from Hunas the situation is totally different and more variable. In ♁♁ morphotype C is dominant (73.8%), only lingually widened, tending to form a lingual root under the metaconid ridge of the protoconid. About one-fourth (24.3%) are of the less advanced morphotype B, which is only buccally widened, and is dominant (48.7%) in ♀♀ M. nivalis from Hunas. There are fewer of morphotype C (18.0%) and the most primitive A (25.6%), with an elongated and narrow crown ( Table 7).

COMPARISON WITH MUSTELA NIVALIS

When comparing M. nivalis from Hunas with the extant Silesian population of this species, we found few distinguishing features; the scale and degree of differentiation between the two groups is quite small, especially in comparison with M. erminea . The size and main indexes used in taxonomic analysis, like p2 L/m1 L, B/L p3, Ba/Bp p4, L ta/L tr m1 and B ta/B tr m1, differ very little. In most cases, indexes and the ranges of variation are almost identical, and the differences found between the sexes in both cases are almost the same. The differentiating features presented below are highly differentiated and tend to be frequent rather than typical for a particular population. The only reliable difference is presence of weakly to moderately marked, but still stronger than in the extant M. nivalis , longitudinal, lingual furrow of the mandibular body.

Mentioned by Heller (1983d, e), considerable bulging of the pars alveolars below the teeth row and accompanying its trough-like depression in M. nivalis from Hunas occurred in such a large degree of diversification of development, it is impossible to find any regularity. Apart from individuals with a fairly strong structure (although not as much as Heller suggested), in most specimens this structure is absent or poorly marked. Moreover, a similar high variability was observed for the extant Silesian M. erminea , which absolutely excludes its usefulness in taxonomic analyses. The same was found for the extant Silesian population of M. nivalis .

COMPARISON WITH MUSTELA PALERMINEA

AND MUSTELA PRAENIVALIS

Finally, the comparison with two ancestor species of small Mustela , M. palerminea and M. praenivalis , do not confirm the statement by Heller (1983d, e). Although the morphometric analysis showed the existence of a few weakly expressed, primitive features, they should rather be treated in the category of ancestral remnants. It is also an expression and proof of evolution within one evolutionary line. It also explains the presence of given features at different stages of advancement from data, different in time and space. Evolution as a process is gradual and heterogeneous, so a given feature may appear earlier in one area, while at a contemporaneous but geographically distant site, it is still primordial. A similar suggestion that the material of the smallest Mustela from Hunas is an intermediate form and evidence of gradual evolution was already proposed by Heller (1983d, e).

Compared to M. palerminea , cranial and dental material of M. nivalis from Hunas shows the following differences:

1)smaller size, well-illustrated by the L m1, where the mean of M. nivalis ♁♁ from Hunas (4.43 mm, 4.02-4.97 mm, n = 103) is smaller than the mean of M. palerminea (EP 5.46 mm, 4.49- 6.12 mm, n = 165; eMP 5.23 mm, 4.46-5.97 mm, n = 36). Overlap in the size class of c. 4.50-5.00 mm is observed, with the large M. nivalis ♁♁ from Hunas comparable with the small (probably ♀♀) M. palerminea ;

2) on average deeper masseteric fossa, with a more triangular mesial margin. Mustela palerminea possesses on average shallower masseteric fossa, with a more rounded mesial edge;

3) shape of the condyloid process, which in M. nivalis from Hunas is more elongated and straighter, while in M. palerminea is usually shorter and hook-shaped;

4) the morphology of the angular process, longer and hooked in M. palerminea , contrary to the short and straight angular process as observed in M. nivalis from Hunas;

5) presence of a longitudinal, lingual furrow of the mandibular body, which is on average distinctly weaker than in M. palerminea . However, the degree of development of this structure in M. palerminea is variable.Together with individuals with a strong lingual furrow, many specimens show a weakly developed structure, similar to that of M. erminea . In larger series from Deutsch Altenburg 2 and Żabia Cave, about 70% of individuals have at most a moderately strong lingual furrow.

The main indexes, like p2 L/m1 L, B/L p3, Ba/Bp p4, L ta/L tr m1 and B ta/B tr m1, differ slightly between M. nivalis from Hunas and M. palerminea . Their variability ranges generally largely overlap, with the means indicating a slightly higher level of evolutionary development in M. nivalis from Hunas. This is not due to evolutionary advancement of M. nivalis from Hunas, but rather to the primitivism of the features characterising M. palerminea . The evolutionary lineage M. palerminea => M. erminea is characterised by a stronger development of progressive features compared to the lineage M. praenivalis => M. nivalis , which is visible in the comparison shown above ( Figs 10; 11).

The comparison of M. nivalis from Hunas with M. praenivalis showed similar morphological differences like those noted in comparison with M. palerminea . In relation to M. praenivalis , M. nivalis from Hunas has a deeper and longer masseteric fossa, which reaches the m1/m2 boundary, weaker lingual furrow, more elongated and straighter angular and condyloid processes, proportionally more reduced p2, broader p3, stronger marked difference between the mesial and distal halves of the p4 crown, and m1 with proportionally higher and longer trigonid and shorter and narrower talonid ( Table 6). There is apparently some overlap, but the ratios differ. The comparison of indexes, starting from the Early Pleistocene through the early Middle Pleistocene, both represented by M. praenivalis , populations of M. nivalis from Hunas (MIS 7), up to the extant M. nivalis , shows a slow but clearly marked increase in the progressiveness of cranial and dental features. However, it is clearly less pronounced than in the evolutionary line of M. erminea => M. palerminea . The above conclusion is additionally confirmed by analyses of the frequency of occurrence of given morphotypes. In the lineage presented above, M. praenivalis => M. nivalis , we observe an increasing share of evolutionarily advanced morphotypes ( Table 7).

COMPARISON OF THE POSTCRANIAL MATERIAL

Numerous and well-preserved postcranial skeletal elements, mainly long bones, but also vertebrae or pelvis, of M. nivalis from Hunas, are morphologically similar to those of M. erminea , but they differ metrically in their smaller size. They are also indistinguishable from bones of extant M. nivalis , which shows great sexual dimorphism, where ranges of variation of sexes almost do not overlap. All long bones of M. nivalis from Hunas are metrically comparable with those of the extant M. nivalis , and the material of this species from Hunas can therefore be sexed. The analysis showed a predominance of ♁♁ and documented presence of only adult, moderately to large-sized individuals.

Considering all the given information, the analysis allowed us to determine the whole cranial and dental material of the smallest Mustela from Huans as M. nivalis . Certain insignificant morphological differences found in relation to extant M. nivalis are a remnant of its ancestor, M. praenivalis . These primary features, expressed to varying degrees in different individuals from Hunas, are already quite weakly expressed. In principle, they should be treated in terms of frequency in terms of the entire population, rather than as the occurrence of a given feature characterizing the population. Thus, Hunas is unique on an Eurasian scale in terms of it allowing the study of a M. nivalis population ( Figs 10; 11) deposited in one spatially limited place, in a relatively short time and in good condition.

In addition, it allows us to capture the already faintly expressed, but still present, primitive features that are no longer present in stratigraphically younger populations. This is direct evidence of the evolution of the phylogenetic lineage of M. praenivalis => M. nivalis . In this lineage, the population of M. nivalis from Hunas documents one of the last accords immediately preceding the appearance of the extant M. nivalis . Ambros (2006) also pointed out the absence of ancient species like M. palerminea and M. praenivalis . She found no distinguishing features M. cf. palerminea from Hunas (in fact, M. nivalis ) and M. erminea . Also, the analysis of postcranial material did not allow her to clarify this problem ( Ambros 2006).