Microdyromys misonnei ( Vianey-Liaud, 1994 )

Lu, Xiaoyu, Costeur, Loïc, Hugueney, Marguerite & Maridet, Olivier, 2021, New data on early Oligocene dormice (Rodentia, Gliridae) from southern Europe: phylogeny and diversification of the family, Journal of Systematic Palaeontology 19 (3), pp. 169-189 : 174-176

Microdyromys misonnei ( Vianey-Liaud, 1994)

( Fig. 3I–T)

1971 Pseudodryomys aff. fugax Hugueney et al. : 2432. 1994 Branssatoglis misonnei Vianey-Liaud : 147–149.

Type locality. Hoogbutsel (MP21).

Holotype. IRSNB M1786 , right M2.

Specimens and measurements. UCBL-FSL_218056 to 218067 and 218073 to 218074 for St-Martin-de-Castillon C; UCBL-FSL_218034 to 218043 and 218072 for Montalb́an 1D; for measurements see Table 1 and Supplemental material, Appendix 2 ( Figs 3I–T, 4j–s).

Description of the specimens from St-Martin-de-Castillon C. DP 4s are more-or-less triangular. The occlusal surface is moderately concave. The protocone is the highest cusp. The paracone and metacone are about the same height but lower than the protocone. The anteroloph is relatively long and lower than other transverse crests, about half of the width of the tooth. It is curved and free, following the anterior side of the tooth. In the central valley, the precentroloph is well developed and connects to the paracone, whereas the postcentroloph is absent. The posteroloph also connects to the protocone but remains free labially.

P4. All teeth display an oval shape. The occlusal surface is strongly concave, with the protocone, paracone and metacone all the same height. The main pattern of the tooth is similar to that of DP 4. The anteroloph is still curved and connects to the protocone and metacone, but longer and higher than in DP 4; in some specimens, it is as high as other crests. The precentroloph is more developed than in DP 4 and the postcentroloph is absent as in DP 4. The posteroloph can be either free or connected to metacone on the labial side.

M1/M2. Teeth display a sub-rectangular shape. The anterior part of M2 is generally wider than M1 but the rest of the morphology is similar. The occlusal surface is concave at all wear stages. The protocone is included in the endoloph and forms a short lingual wall connecting to the posteroloph. The anteroloph is free on both sides, occasionally connecting with the protocone on some M2; it is curved posteriorly on its lingual side. The protoloph is straight and transverse, but the metaloph bends anteriorly near the protocone. The anterotrope and posterotrope are both absent. The precentroloph is well developed, almost reaching the protocone. The postcentroloph is less developed than the precentroloph, about half the length of the central valley, and it connects lingually to the precentroloph in some specimens. A very weak prototrope can be observed in most of the teeth whereas the metatrope is absent. The posteroloph is free labially.

M3. All teeth display a sub-triangular shape, strongly reduced posteriorly. The occlusal surface is slightly concave and the pattern is more complicated than on other teeth due to a stronger morphological variation. The anteroloph is straight and connects with the protocone. There is either the prototrope or the metatrope in the central valley, never both together. The precentroloph and the postcentroloph can be weak or well developed and in some specimens they are discontinuous. The posteroloph extends posterolabially and connects with protocone and metacone.

dp4. It is slightly narrower than p4. The metaconid is strong and elongated posteriorly, but disconnected from the entoconid. The anterotropid and centrolophid are absent. The metalophid is robust and connects to the metaconid and protoconid. The anterolophid is very weak. The mesolophid is free and also weak. The posterotropid is absent. The posterolophid is more robust than the metalophid and forms a rounded wall on the posterior border of the tooth.

p4. The morphology is similar to that of dp4. However, the anterolophid is present and also a short anterotropid; it is oblique, weak and short, shorter than in m1. The metalophid is curved posteriorly near the labial side. The centrolophid is present but very low and difficult to observe. The posterolophid is transverse in the middle.

m1/m2. All teeth are rectangular and strongly concave. m1 differs from m 2 in having the posterior part slightly wider than the anterior part. The main cuspids are distinct from the crests. The metaconid is elongated and forms a robust lingual crest. The entoconid is strong and connects to the posterolophid in a low level. In most specimens, the entoconid connects with the mesolophid; in some rare cases this connection is absent. All transverse crests are about the same height. The anterolophid disconnects from the protoconid labially. There is only one weak anterotropid. The metalophid bends backward but does not reach the metaconid lingually. Both the precentrotropid and postcentrotropid are absent. The centrolophid can be weak or developed, even lingually free in some cases. The mesolophid is straight and connects with the mesoconid. The posterotropid is well developed.

m3. it displays a triangular shape, reduced posteriorly. The main pattern of the teeth is similar to that of m1 and m2, but the cuspids are weaker except the metaconid. The anterotropid is weak and the metalophid is completely straight. The centrolophid is well developed in all specimens. Because of the reduction of the posterior part of teeth, the posterotropid and posterolophid are strongly curved backward.

Remarks. Microdyromys misonnei was initially described as Branssatoglis misonnei . It was transferred to the genus Microdyromys by Freudenthal & Mart́ın-Súarez (2007a). Most of the new specimens from Montalb́an 1D (MP23) are referred to this species. Freudenthal & Mart́ın-Súarez (2007a) made a complete description of the population from Montalb́an. Our specimens from Montalb́an 1D are characterized as upper molars, by the prototrope which can be either absent, short or long; the anterotropid in lower molars always present but weak; and a well-developed posterotropid in m1 and m2. These characteristics fit the description and variability described by Freudenthal & Mart́ın-Súarez (2007a). Apart from the material from Montalb́an 1D, some specimens from St-Martin-de-Castillon C (MP24) are also referred to the same species (see Description, above). The new specimens from Montalb́an 1D seem smaller than the former collection for P4 and M3 ( Figs 5, 6). In lower cheek teeth, the length of teeth is similar, but they are noticeably narrower than in the old collection. Such differences for the same locality are difficult to explain as the method used to measure the teeth is the same as the one used by Freudenthal & Mart́ın-Súarez (2007a). Discrepancies between the two studies in identifying premolars and third molars at specific level could explain such differences. Additionally, for m3 only, the material from Hoogbutsel is slightly larger than that from Montalb́an 1D. These differences do not seem to be the result of an evolution of a lineage with time but might partly be due to a previously underestimated size variability of the species. However, our new specimens from St-Martin-de-Castillon C are much larger than in the two other localities. Vianey-Liaud (1994) noticed a noticeable size evolution for the genera Glamys , Gliravus and Branssatoglis along the course of the Palaeogene. She hypothesized that geographic isolation of taxa might lead to significant size differentiations whereas morphological changes often remain weak. Likewise, we hypothesize here that the different sizes might also be due to a biogeographic factor considering the distance between the different localities. Waiting for further data or additional comparisons with other localities, we tentatively refer the specimens of St-Martin-de-Castillon C to M. misonnei due to the morphological similarities and despite the size differences with Hoogbutsel and Montalb́an 1D.