Myotis gerhardstorchi, Horáček & Trávníčková, 2019

Horáček, Ivan & Trávníčková, Eva, 2019, Myotis Gerhardstorchi Sp. N. And Comments On The European Fossil Record Of Myotis Frater Group (Mammalia, Chiroptera), Fossil Imprint 75 (3 - 4), pp. 315-342 : 317-322

Myotis gerhardstorchi sp. n.

Pl. 1, Figs 1, 2, Pl. 2, Figs 1, 2, Pl. 3, Fig. 1

H o l o t y p e. A nearly complete rostrum with preserved both left and right toothrows (I1, 2) C–M3, compressed on dorsal side, partly cemented with flowstone incrustation. (Ber26/C1; Pl. 1, Fig. 1a–d)

I1M3 7.89, CM3 7.11, P4M3 5.05, M1M3 4.26, M2M3 2.79, M1M2 3.42, P4M1 2.74, CM1 4.84, P2P3 1.17, CH 2.42, CL 1.31, CW 1.15, P2L 0.81, P2 W 0.76, P3L 0.57, P3 W 0.52, P4L 1.33, P4 W1 1.34, P4 W2 1.27, P4 W3 1.48, M1L 1.81, M1 W1 1.82, M1 W2 1.91, M1 W3 2.01, M2L 1.78, M2 W1 2.01, M2 W2 2.12, M2 W3 2.16, M3L 0.95, M3 W1 1.87, M3 W2 1.65, M3 W3 1.06 (measurements in mm) .

All the material will be deposited in the collections of the Palaeontological Department, National Museum, Prague, the Czech Republic.

P a r a t y p e s. Ber26/C2: an incomplete rostrum (I1,2) CP2(P3)P4M1 on right side, (I1–C)P2(P3)P4–M3 on left side (Pl. 1, Fig. 2); Ber26/C4: fragment of left maxilla M1– M3 (Pl. 2, Fig. 2).

O t h e r m a t e r i a l. Ber26/C3: fragment of right mandible (i1–p3)p 4m 1(m2–m3) (Pl. 2, Fig. 6); Javoříčko VII/226: fragment of left maxilla (I2–P3)P4(M1–M2) (Pl. 2, Fig. 5); Javoříčko XI/547: fragment of right maxilla (C–P3) P4(M1) (Pl. 2, Fig. 4).

D e r i v a t i o n n o m i n i s. In memory of Gerhard Storch, one of the top personalities in the study of fossil bats, with a brilliant capacity to integrate both the neontological and palaeontological aspects of the topic.

Ty p e l o c a l i t y a n d s t r a t u m t y p i c u m. Beremend 26 (Villány Mts., Hungary), Early Pliocene, MN 15b.

D i a g n o s i s. A larger, medium-sized representative of the genus, resembling extant M. dasycneme in skull size and proportions, from which it differs by robust unicuspids, higher degree of M3 reduction and less distinct M2 ectoflexus. In these characters it resembles M. bechsteinii (and related fossil forms such as M. gundersheimensis , M. kormosi , etc.), from which it differs by distinctly shortened unicuspid row with large but compressed premolars, P3 being displaced from toothrow, and by large, sharply pointed canine, round on section. In fine dental characters and shape of rostrum, it shows clear similarities to Asiatic forms M. sicarius (from which differs in smaller size, larger premolars and less reduced M3), and M. frater , which is of course much smaller.

D e s c r i p t i o n. A larger, medium-sized form of Myotis , resembling M. dasycneme in size and compression of unicuspid row. Rostrum is conspicuously broad and massive, with markedly shortened premaxillae, and short and narrow anterior palatal vacuity in between them. Its distal margin is situated at a level of middle of canine alveolus (⅔ in the paratype, Ber26/C2). Canine is high (almost twice higher than P4), sharply pointed, round at its cingular base, with a sharp distal cutting ridge passing from the distal crown base to the crown tip, and a straight broad ridge at the mesiopalatal crown corner. Labial wall is rounded without a trace of undulation. Cingulum continuous, moderately thick without distinct caspules or talonal extensions.

P2 is particularly massive, with sharp distal ridge and distinct cingulum, almost round on section. In lateral view its height amounts about ⅔ of P4 height. Also P3 is relatively large (about ⅔ of P2 size) and sharply pointed, but displaced palatally from the toothrow, and not visible in the lateral view.

P4 with a high labial crest, massive crown of subcircular outline with robust palatal heel, not extending into separate talonal basin. Distal crown margin almost straight, without distinct undulation, mesial cingulum broadly rounded, buccal cingulum narrow and straight, without marked undulation. Molars are relatively high and massive, with sharp paralophs extended from the base of paracone and terminating in pronounced protoconules, clearly exceeding a level of preprotocrista. M1 and M2 are characterised by narrow postprotocrista continuing as a sharp crest to talonal base of metacone (note: this is a plesiomorphic state of postprotocrista design – comp. Horáček and Špoutil 2012). This arrangement is associated with a broad metaloph ridge passing from hypoconal extension of protoconal complex (“metaconule”) to base of metacone, attaining the occlusal level of postprotocrista. The hypoconal extension forms an undulation of the palatal wall of the protocone complex, at occlusal level marked by a short rudiment of transcrista, despite almost no undulation appearing at the palatal cingulum of the crown. Distal margin of the molar crown is almost straight, without marked separation of the protoconal complex. Both parastyle and metastyle wings of ectoflexus are markedly enlarged, and buccally they considerably exceed a level of mesostyle that makes the outline of the ectoflexus almost straight.

M3 is moderately reduced, its mesio-distal length is obviously smaller and the metacone is less distinct than in other compared species (Pl. 3). Paraconule and paraloph are well developed. In contrast to other species, M3 bears distinct hypoconal thickening with robust transcrista, by which the distal margin of the protocone complex is distinctly separated from the metacone.

All these characters are with minute variations pronounced in all three rostral fragments from the type locality. The maxillary fragments from Javoříčko VII and XI, which bear P4 only (Pl. 2, Figs 4, 5), correspond to them in alveolar conditions of unicuspid row, and in shape and size of P4 .

The mandibular fragment from Beremend 26 (Pl. 2, Fig. 6), tentatively co-identified with the new species for roughly corresponding size and resemblance of M. dasycneme (Pl. 5, Fig. 5) in proportions of p4 and m1, the only teeth preserved, differs from the extant species by narrower p4 and shape of m1, with elongated mesial wall of trigonid, broad trigonidal fovea with a low position of the fovea base, resembling the arrangements characteristic for the M. frater group (see below). By combination of a small and narrow p4 (0.81 × 0.55) and large m1 with elongated trigonid, it differs from the taxa of the M. bechsteinii group ( wuesti KORMOS, 1934 , kormosi HELLER, 1936 , gundersheimensis HELLER, 1936 , aemulus HELLER, 1936 ) and other fossil species of corresponding size ( schaubi KORMOS, 1930 , steiningeri KORMOS, 1934, baranensis KORMOS, 1934 , podlesicensis KOWALSKI, 1956 ).

C o m p a r i s o n s. Of the extant European species, M. dasycneme is the first to be taken in account. It resembles the above-described form in size, shape of rostrum and considerable compression of P2–P3. It differs, of course, in shape of canine (anterior position of mesiodistal ridge and flat palatal wall), smaller size and form of P2 and P3, slender heel of P4 with pronounced undulation of distal margin or lesser degree of M3 reduction. Its molars lack buccal extensions of para- and metastyle wings of ectoloflexus and metaloph thickening of distal postprotocrista. *

In most species of Myotis (and most other vespertilionid genera), hypoconal extension of protoconal wall is relatively low compared to the protocone, and the metaloph is not developed. Consequently, the postprotocrista directly continues to presumptive hypocone and via the transcrista to the distal crown base, while the fossa gradually opens to the distal crown margin (comp. Pl. 3, Figs 4, 7, 8).

This is the case also in M. bechsteinii (and numerous fossil items co-identified with it and/or the fossil species supposedly related to it – see above). At the same time, of course, the forms of that group correspond to the new species in size of individual teeth, shape of the upper canine and premolars, though they markedly differ in size of premolars and arrangement of unicuspid row, position of P3 and in narrower rostrum prolonged in unicuspid section. M2 and M 3 in M. bechsteinii as well as in M. gundersheimensis show a distinct extension of the metastyle wing of the ectoflexus, but the parastyle wing remains quite narrow (Pl. 3). Yet all these forms, as well as other European fossil species of corresponding size for which the maxillary dentitions are available (incl. M. schaubi and M. podlesicensis or numerous Miocene or Late Oligocene forms – comp. Horáček and Hanák 1984, Ziegler 2000, 2003, Rosina and Kruskop 2011, Rosina and Semenov 2012) differ significantly from the new species in arrangement of unicuspid row.

There is an extensive similarity between the new species and M. sicarius in the above-mentioned diagnostic characters, but the latter form, now endemic to Sikkim and Nepal, is

* At this point we feel obliged to elucidate the usage of some terms applied in this paper, namely the term metaloph (or “premetaconule crista” by Aguiar Fracasso et al. 2011), commonly used to denote the crest between postprotocrista and base of metacone (comp. e.g. Menu 1985, Godawa Stormark 1998). Yet, following Cope-Osborn proposal, the term “loph” is to be applied to the thickened enamel ridges which perpendicularly interconnect the main cusps of tribosphenic design, i.e. paracone-protocone (paraloph) and metaconehypocone (metaloph), respectively. Since in all early bats which lack hypoconal extension of protoconal complex ( Ageina , Icaronycteris , Archaeonycteris , etc.) as well as in presumptive ancestors of this order ( Nyctitheriidae , Adapisoriculidae ) postprotocrista connects protocone with base of metacone, also in the modern bat the crest extending postprotocrista to base of metacone, though commonly denoted as metaloph ( Menu 1985; or premetaconule crista by Aguiar Fracasso et al. 2011) is to be considered as part of postprotocrista and a distinct plesiomorphy of molar crown design (comp. Horáček and Špoutil 2012 for details). Of course, where hypocone and/or hypoconal extension of the protocone distal wall is developed (termed “metaconule” as a rule where it is shaped in form of a distinct cusp-like structure), the respective crest can be accompanied with a distally thickened ridge interconnecting that structure with the base of the metacone. This ridge, metaloph s. str., is obviously not identical with the above-mentioned crest (it is figured but not identified by Aguiar Fracasso et al. 2011: fig. 1b); in some clades it accompanies the crest, in others, one or both are absent. In Myotis -like bats, where taxonomically significant odontological criteria are rather scarce, a careful attention to fine morphology of upper molars incl. consequent distinguishing these structures can essentially improve the scope of comparative studies, particularly when the fossil record is taken in analysis.

considerably larger, and shows a higher degree of P2–P3 and M3 reduction (App. III, Tabs A1, A 2). In M. sicarius , P2 is still relatively large, but infilling a narrow space between C and P4, it is mesiodistally compressed; P3 is dot-like, partly under the mesiopalatal cingulum of P4. The metastyle wings of M1 and M2 ectoflexus are much less developed in sicarius , while mesostyles are prominent buccally to form a trilobatelike outline of the ectoflexus in the occlusal view.

A broad measure of agreement in the above-mentioned characters in shape of rostrum and relative sizes of individual teeth is also with members of the extant M. frater group (comp. Pls 4–7, and also Pl. 3, Figs 1, 2), which are, of course, much smaller (App. III, Tabs A1, A 2).

Regarding all this, the new species, which reveals a mosaic of resemblances to all taxa composing Clade III, i.e. M. sicarius , M. bechsteinii group, M. frater group and M. dasycneme - Leuconoe -like traits characterising M. daubentonii , can be considered as the form closely related to the stem line of Clade III.

Myotis frater group in mid-European fossil record

Bats of M. frater group bear an uncommon combination of characters for which their taxonomic position was considered uncertain ( Tate 1941, Kuzyakin 1950, Wallin 1969, Findley 1972, Tsytsulina and Strelkov 2001), but which at the same time allows distinguishing this group from other members of the genus: enlarged braincase, shortened but conspicuously broad rostrum, compressed unicuspids with displacement of P3 from the tooth row, greatly pronounced angulus mandibulae, mesially tapered processus coronoideus, etc.

The group, formerly denoted as a single species, was surveyed in detail by Tsytsulina and Strelkov (2001), who demonstrate that it is composed of several taxonomically distinct forms occupying isolated vicariant ranges: Myotis f. frater ALLEN, 1923 ( China, Fujian Prov., Rong To Valley, Tibet and Taiwan), Myotis f. kaguyae IMAIZUMI, 1956 ( Japan), Myotis f. longicaudatus OGNEV, 1927 (Russian Far East), Myotis f. eniseensis TSYTSULINA et STRELKOV, 2001 (Central Siberia). Similarly to Horáček et al. (2000), Tsytsulina and Strelkov (2001) demonstrated that Myotis f. bucharensis KUZYAKIN, 1950 (three localities in border region of Tajikistan and Uzbekistan) differs in more characters from the other forms of the group, and so represents a separate species. A detailed analysis by Ruedi et al. (2015) combining both molecular and morphological approach supplemented the M. frater group with a new species from Taiwan, M. soror (denoted as Myotis sp. 3 in Ruedi et al. 2013).

A tentative odontological diagnosis of the Myotis frater group (based on comparison of extant taxa) could be as follows:

(i) A complete Myotis dentition with considerably reduced small premolars, shortened but broad rostrum, (ii) shortened premaxilla with a broad but short anterior palatal vacuity, with its distal margin at a middle level of canine alveolus, (iii) strong upper canine with a broad circular cingulum particularly distinct along labial edge of the crown, (iv) P2 is relatively large while P3 is displaced palatally from toothrow, so P2 appears nearly in contact with mesial cingulum of P4, (v) P4 relatively robust (compared to M. daubentoni , M. dasycneme or M. mystacinus ), yet with only indistinct talonal extension, (vi) M1 and M2 with paralophs and protoconules (absent in M1 of bucharensis , indistinct in eniseensis; comp. Tsytsulina and Strelkov 2001) and (vii) enlarged metastyle wings of ectoflexus, (viii) the crest between hypoconal extension of the protocone wall and metacone distinct (particularly high in bucharensis but low in eniseensis), accompanied by metalophus ridge (which is absent in bucharensis ), (ix) M3 moderately reduced (more than in M. daubentonii or M. dasycneme ) with retained protoconule.

Mandible is characterised by (x) a pronounced angulus mandibulae, and (xi) anteriorly tapered processus coronoideus, clearly extending the mesial margin of ramus mandibulae, (xii) i3 is significantly larger that i1 and i2, with two large cusps at the labial side of the crown and two isolated cusps extending its crown lingually, (xiii) in contrast to most other species of the genus, i3 bears a distinct cingulum along its labial base, (xiv) the lower canine exceeds p 4 in height only indistinctly, its lingual cingulum is conspicuously thick, and terminates in a robust cingular cuspid at the mesio-lingual corner of the crown, which attains almost half of the tooth height, (xv) p2 is about a half or ⅓ of p2 size and tends to be displaced lingually from a toothrow, (xvi) compared to m2 and m3 trigonid of m1 is enlarged with anteriorly extended mesial wall, (xvii) trigonid fovea of m1 is thus conspicuously broad and deep with a marked cingulids at its lingual base, (xviii) m3 is moderately reduced, its talonid is relatively narrow, but long, with high entoconid crest and robust postcristid.

The odontological differences of M. bucharensis lie particularly in shape of the upper canines, with an inflated distal cingulum laterally exceeding a level of P2 tip, and the distal margin of the protoconal complex of M1 and M2, with absence of a broad metalophus (vii, viii). In mandibular dentition, a robust elongated p4 is to be mentioned.

Studying extensive materials of fossil bats from a number of the Pliocene and Pleistocene localities of Central Europe (comp. e.g. Horáček and Ložek 1988), we found a set of items differing from other local forms, both extant and fossil, which show a broad measure of agreement with the characteristics of the M. frater group, as proposed above (including their size corresponding to respective extant

1.5

1.4

1.3

1.2

1.1

3

P

2 1.0

P

0.9

0.8

0.7

0.6

0.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 CM3

Tex-fig. 3. Maxillary variables C–M3 vs. P2–P 3 in the reported fossil forms and extant Asiatic forms with clouds of variation of the extant W Palearctic representatives of the genus.

1.8

1.7

1.6

1.5

L

1 m 1.4

1.3

1.2

1.1

1.0 1.8 2.0 2.2 2.4 2.6 2.8 3.0 3.2 3.4 3.6 3.8 cp4

Tex-fig. 4. Mandibular variables cp4 vs. m1L in the reported fossil forms and extant Asiatic forms with clouds of variation of the extant W Palearctic representatives of the genus. 1 – Myotis cf. gerhardstorchii sp. n., Beremend 26; 2–6 – M. frater group: 2 – Urwista, 3 – Javoříčko VII, 4 – Javoříčko III, 5 – Sovinec 4; 6 – Včeláre 4D/12; 7 – M. mystacinus group, Včeláre 4D/12; 8 – M. danutae, Podlesice ; 9 – M. delicatus, Gundersheim ; 10 – M. dasycneme subtilis, Podlesice. At – M. aemulus holotype, Gundersheim; Et – M. exilis holotype, Gundersheim; Gt – M. gundersheimensis holotype, Gundersheim; Ht – M. helleri ( M. insignis holotype), Gundersheim; Pt – M. praevius holotype, Gundersheim; Rt – M. rapax holotype, Gundersheim; aem – M. aemulus, Gundersheim ; exi – M. exilis, Gundersheim ; gun – M. gundersheimensis, Gundersheim ; hel – M. helleri, Gundersheim ; j3d – M. delicatus / dasycneme, Javoříčko III ; kor – M. kormosi, Gundersheim ; s – M. dasycneme, Sovinec 4.

species) – the complete list of them is in Appendix II, the best-preserved items are figured in Pls 4–7, their metrical characters are summarized in Appendix III ( Tabs A1, A 2). Yet, as is frequent in bat fossil assemblages, only few maxillary fragments are available, while the vast majority of the material is composed of mandibular fragments. On the other hand, this opens a possibility of comparison with other fossil species of the genus, described exclusively based on mandibular characters.

The largest collection comes from Q 1 site Sovinec 4 (northern Moravia, the Czech republic; three rostral fragments, three maxillary fragments and 38 mandibular fragments), a perfectly preserved specimen is available also from another Q 1 site – Včeláre 4D (SE Slovakia; one almost complete and three fragmentary mandibles), while the items found in MN 16 Urwista ( Poland), MN 15 Javoříčko VII and MN 17 Javoříčko III (both north Moravia, the Czech Republic) are both less preserved and less frequent (comp. App. I). Both in dental phenotype and in measurements, these specimens fit impressively well to the conditions of extant form (comp. Pls 5–7 and App. III, Tab. A1). Yet all specimens from Urwista, several items in Sovinec 4 (Sov 4/9, 19, 22), and other sites (Javoříčko VII: JavVII/29, Javoříčko III: JavIII/223) exhibit dimensions slightly larger, exceeding the upper limit of variation of extant forms ( Tsytsulina and Strelkov 2001) by 6–10 %. Some of them (e.g. Sov 4/22, JavIII/233) also show certain differences in shape of the canine, which is conspicuously large, but mesio-distally compressed at its base, though in indexing characters (xv– xviii) they exhibit good agreement with other forms of the M. frater group .

In morphometric space of uni-, bi- and multivariate comparisons, the fossil items attributed here to the M. frater group form a relatively compact cluster, close to extant forms of the clade (comp. Text-figs 3–6). It is situated quite apart from variation domains of most other W Palearctic members of the genus, not only the larger forms, both extant ( bechsteinii , emarginatus , nattereri , dasycneme ) and fossil ( kormosi HELLER, 1936 , aemulus HELLER, 1936 , gundersheimensis HELLER, 1936 , rapax HELLER, 1936 , praevius HELLER, 1936 , podlesicensis KOWALSKI, 1956 or gerhardstorchi sp. n.), but also from those of comparable or smaller size.

By some metrical characters, several items (particularly those from Urwista – comp. Pl. 7, Figs 1–2) fall into variation span of extant M. capaccinii . Yet M. capaccinii differs from the M. frater group (including the fossil items in question) by smaller lower incisors (incl. i3), low but elongated canine with enlarged and medially tapered talon and without mesiolingual cingular cusp, by large and high p2 and p3 amounting to ⅔ of p4 height, p4 square on section, narrow trigonid of m1 with a high base of trigonidal fossid, etc., notwithstanding the prolonged rostrum or unreduced premolars in the upper jaw. Similar differences, particularly the lesser degree of premolar reduction, narrower m1 trigonid and different shape of c1 and p4 also discriminate the M. frater group from M. brandtii , M. mystacinus , M. daubentonii , fossil taxa M. paradaubentoni TOPÁL, 1983 , M. janossyi TOPÁL, 1983 , M. estramosensis TOPÁL, 1983 , M. exilis HELLER, 1936 and M. helleri KOWALSKI, 1962 (= M. insignis HELLER, 1936 ), all of which are also significantly smaller on average (comp. Godawa 1993 and App. III, Tab. A2).

M. delicatus HELLER, 1936 , M. dasycneme subtilis KOWALSKI, 1956 and Myotis danutae KOWALSKI, 1956 are the other species which come in account, yet they all are somewhat larger than the discussed items, except for the above mentioned large specimens of our fossil material.

8

6

4

2

%

14.25 0

:

2

Factor -2

-4

-6

-8

-10 -25 -20 -15 -10 -5 0 5 10 15 Factor 1: 60.06%

Tex-fig. 5. Plot of PCA Factor 1 and 2 scores (based on maxillary variables: CM3, P4M3, M1M3, M2M3, M1M2, P4C1, CM1, P2P3, length and width of individual teeth P4–M3, and proportions P2P3/P4 and W/L of individual teeth C–M3) of the reported fossil forms and extant Asiatic forms, with clouds of variation of the extant W Palearctic representatives of the genus.

4

2

%

10.01

:

2

0

Factor

-2

-4

-10 -8 -6 -4 -2 0 2 4 6

Factor 1: 79.31%

Tex-fig. 6. Plot of PCA Factor 1 and 2 scores (based on mandibular variables cm3, p 4m 3, m 1m 3, cp4, p 4m 1, m 1m 2, m 2m 3, p4L, m1L) of the reported fossil forms and extant Asiatic forms, with clouds of variation of the extant W Palearctic representatives of the genus. For further explanations see Text-fig. 4.

The former two taxa were synonymized by Topál (1985) and interpreted as ancestor clade of M. dasycneme by Horáček and Hanák (1989), who suggested (with regards to remains from Javoříčko, Sovinec (both the Czech Republic), Urwista and Żabia (both Poland)) a gradual increase in size and degree of premolar reduction as the morphocline characterising the Pliocene history of that clade (comp. also Text-fig. 4). Of course, while the specimens of delicatus from MN 15 Gundersheim (incl. its holotype) conform well to characters of dasycneme -clade (small i3, elongated base of c1, compressed p3, narrow trigonid fossid of m1, broad molars), M. dasycneme subtilis KOWALSKI, 1956 from MN 14 Podlesice (emended to M. delicatus subtilis by Horáček and Hanák 1989) shows significantly smaller dimensions (comp. Godawa 1993), falling even into the variation span of large items of the M. frater group. The single paratype specimen of subtilis that was at our disposal resembles them even in shape of the lower canine. Also, the remaining species described from MN 14 Podlesice, Myotis danutae , could come in account as a taxon related to stem line of the Myotis Clade III, eventually. Kowalski (1956) and Topál (1983, 1985) mentioned suspected relations of this form to M. emarginatus , mostly based on elongated p4 and relatively short canine, yet at the same time it differs just in these characters from the extant species quite distinctly. When the senior author of this paper examined a paratype series of danutae many years ago, he concluded that the idea on its relationship to emarginatus seemed to be insufficiently substantiated, and danutae is to be looked upon as the form without apparent relations to any of the extant W Palearctic clades. In some characters it even resembles the M. frater group (e.g. in shortened premaxilla, relatively short and broad rostrum), yet its premolars are only slightly reduced, with no displacement and/or marked disproportions in degree of P2–p2/P3–p3 reduction characteristic, e.g. for the M. mystacinus group, or M. dasycneme .

In short, the status and phylogenetic relations of subtilis and danutae are worth being thoroughly re-examined, also in regards to the possibility of their relations to stem line of Clade III. However, this topic is obviously beyond scope of the present paper.

Nevertheless, the possibility that the European Pliocene communities were enriched by more than one relic form related to the stem line of Clade III, not necessarily related to the extant species of that group, cannot be excluded. Also the fact that the phenotype variation observed among the fossil specimens surveyed above (e.g. in shape of lower canine) amounts to dental differences separating extant M. longicaudatus and M. bucharensis , put the assumed species homogeneity of the European M. frater group into question.

In any case, it seems clear that the forms attributed here to the M. frater group represented a quite rare component of fossil bat assemblages. In Sovinec, they are represented by 41 items of about 1,500 jaw fragments, in Javoříčko III 12 items of 1,360 jaws, in Javoříčko XI one of>2,000, in Javoříčko VII two items of ca. 300, in Urwista 9 items of 1,700 and in Včeláre 4D three items of ca. 300 jaws, i.e. 0.1– 2 % of local bat community on average, notwithstanding absence of these forms in the vast majority of other fossil bat assemblages.

Of course, it should be remembered that the extant forms of that clade also rank in all regions of their appearance among very rare bats (comp. Wallin 1969, Tsytsulina and Strelkov 2000, Ruedi et al. 2015). For instance, M. bucharensis was found only in three localities in 1915, 1959 and 1960 ( Kuzyakin 1950, Bogdanov 1960, Tsytsulina and Strelkov 2001, Benda et al. 2011), while no further record is available since 1963, when the colony discovered in 1960 disappeared, due to which the species is considered extinct ( Horáček et al. 2000).