Neomys, KAUP, 1829

Neomys View in CoL during the Last Glacial Cycle

The scarcity of European fossil material together with the absence of reliable determination criteria represent a major setback, as it has often led to troubles with identification. Therefore, most of the specimens have been ascribed to Neomys sp. or very often Neomys cf. fodiens . However, relevant discrimination of most of the material is lacking, and the aspects of variability, especially the overlap between the two species, are barely considered.

As a result of the issues mentioned above, knowledge of the most recent history of genus Neomys is limited. Regarding the outcomes of our survey of extant populations, we analysed 58 Q 4 specimens from Czech and Slovak fossil sites. Our material showed extreme variability, especially regarding size and shape of the condylar process and both condylar facets (comp. Text-figs 6, 7, 10). We also observed differences in the shape of the coronoid process apex. Considering that most of our material was collected from taphocenoses by accumulation of owl pellets, it is possible that such differences are taphonomic artefacts. Besides that, our Q 4 material shows general similarities with the extant populations regarding basic dental characteristics.

In Europe, genus Neomys has been reported from 83 Holocene and Late Pleistocene (Q 4, Q 3/Q 4) localities of 18 countries (see SM:SF III for details). The earliest records are from the Q 3 biozone in Spain ( Moya-Costa et al. 2023), three localities in France ( Jammot 1977, Reumer 1996), one locality in Italy ( Bartolomei and di Broglio 1964), Netherlands ( van Kolfschoten 1990), Romania ( Rădulescu and Samson 1992) ( N. fodiens ) and Austria ( Rabeder 1972) ( N. milleri ). Generally, Neomys fodiens is considered a resident species in Central Europe since the Middle Pleistocene, while anomalus-milleri clade is expected to be distributed primarily in southern Europe, and to invade Central Europe only during few interglacial stages, including the Holocene ( Rzebik-Kowalska 1998, Kryštufek et al. 2000, Castiglia et al. 2007, Anděra and Hanzal 2022).

A significant number of samples followed by detailed descriptions are presented in Rzebik-Kowalska (2006) from Komarowa Cave in Poland ( Neomys cf. fodiens MNI = 16, Neomys sp. MNI = 2). While the specimens ascribed to Neomys fodiens seem to be more robust in comparison to our Q 4 material (even though their variance ranges overlap in all cases), the specimens identified as Neomys sp. are smaller: one mandible corresponds well to our Q 4 N. fodiens (p4L, m1L, CorH), the other fragment of a lower jaw presents similar metric characteristics to our recent and fossil N. milleri . Material from Ightham Fissures, Dogholes (Late Pleistocene) and Leasowe (Holocene) (both UK; Hinton 1911), all of them described as Neomys fodiens , are smaller than our Q 4 Neomys fodiens regarding height of the coronoid process and length of the molar row. A mandible from Bois Roche ( France, MIS 4–5) described as Neomys cf. fodiens ( Sesé and Villa 2008) possesses a longer premolar than our Q 4 Neomys fodiens . On the contrary, its width is moderate compared to our samples. The first two molars with small deviations correspond to our material in their lengths and widths. Both species were reported (as N. cf. fodiens and N. cf anomalus ) from Grotta Maggiore di San Bernardino in Italy (MIS 3–7) ( López-García et al. 2017), Höhle Fels in Germany (MIS 3–5; N. cf. fodiens , N. cf. milleri , Neomys sp. ) ( Luzi et al. 2022) and Grotta della Ferrovia in Italy (LGM; N. cf. anomalus , N. fodiens ) ( Ceregatti et al. 2023).

As seen from previous paragraphs, knowledge of the most recent history of genus Neomys is limited, mostly due to the absence of reliable determination criteria and fragmentary state of the record. However, by application of combined determination technique based on results of the biometric analysis of extant populations, we were able to identify 41 % of the 58 Q 4 records with high degree of reliability. Even though fossil material of Neomys from this period is generally rare, so that it is not possible to make detailed conclusions, our survey denies the possibility that Neomys milleri represents an apochoric element invading Central Europe during the Late Holocene. Instead, we managed to prove the presence of both species in Central Europe during the Vistulian, in early units probably with higher abundance of N. fodiens , but even before the beginning of Holocene, Neomys milleri is also recorded, and becomes at least locally more abundant by the end of the Preboreal. We believe the two species coexisted in sympatry or parapatry during whole glacial cycle, even though their presence was probably mosaic-like in terms of both time and space. The significantly enlarged span of phenotype variation demonstrated for our Q 4 sample can be tentatively ascribed to the effects of character displacements accompanying the stages of sympatric occurrence.