Alexandromys evoronensis, Kovalskaya et Sokolov, 1980

Lissovsky, Andrey A., Petrova, Tatyana V., Yatsentyuk, Svetlana P., Golenishchev, Fedor N., Putincev, Nikolay I., Kartavtseva, Irina V., Sheremetyeva, Irina N. & Abramson, Natalia I., 2017, Multilocus phylogeny and taxonomy of East Asian voles Alexandromys (Rodentia, Arvicolinae), Zoologica Scripta 47 (1), pp. 1-12 : 7-8

4.3 | Taxonomic status of A. evoronensis and A. mujanensis

Trees constructed on the basis of both mitochondrial and nuclear data sets are similar in very close phylogenetic proximity of A. maximowiczii , A. evoronensis and A. mujanensis . Our results of the JML test found that mtDNA-based genetic distances are significantly shorter in the pair A. maximowiczii – A. mujanensis , than the distances calculated on the basis of analysis of posterior distributions of nuclear species-trees from *BEAST with known ratio of mitochondrial to nuclear mutation rates. Such result suggests hybridisation with mitochondrial introgression scenario ( Joly et al., 2009). The same result was obtained for A. maximowiczii – A. evoronensis pair but at lower confidence level. It should be noted that in the analysis of the larger taxa set with cytb (Figure 2), A. maximowiczii is not monophyletic relative to A. evoronensis and A. mujanensis . Monophyly is broken by the specimen of A. maximowiczii from Hentiyn Nuruu, Mongolia. Similar results showing paraphyly of A. maximowiczii related to A. evoronensis and A. mujanensis were obtained with the control region of mtDNA ( Haring et al., 2011).

Genetic distances between these three taxa calculated using both mitochondrial and nuclear data sets (Table 1) are notably shorter than distances between recognised species of Alexandromys . In cytb, these distances are even shorter than intraspecific distances within A. oeconomus ; in nuclear BRCA1, these sets of distances are comparable.

Two of the species from the group under discussion, A. evoronensis and A. mujanensis , were described as separate species on the basis of different chromosome structure ( Kovalskaya & Sokolov, 1980; Orlov & Kovalskaya, 1978). Laboratory experiments showed that offspring of interspecies hybridisation in these three species are sterile ( Meyer et al., 1996). The results of those experiments seem to contradict our results of JML test implying gene flow between A. maximowiczii and two other species, especially A. mujanensis . There are two possible explanations of such contradiction. In the first case, establishment of postzygotic reproductive isolation could be very recent and took place after hybridisation events suggested by JML test. In the second case, laboratory hybridisation experiments may not reflect real natural situation. Indeed, supposed postzygotic reproductive isolation is a result of chromosomal rearrangements. Laboratory experiments cited above used limited number of specimens from one population per species. Meanwhile, all the three species share notable chromosomal variation ( Kartavtseva et al., 2008; Lemskaya et al., 2015; Sheremetyeva, Kartavtseva, & Vasil’eva, 2017); thus, gene flow could occur through some populations that were not involved in the hybridisation study.

The shallow difference between Maximowicz’s vole and two other species was earlier found in numerous morphological and allozyme studies (Frisman, Korobitsyna, Kartavtseva, Sheremetyeva, & Voyta, 2009; Lissovsky & Obolenskaya, 2011; Meyer et al., 1996; Pozdnyakov, 1996; Voyta et al., 2013). Thus, it is clear that A. evoronensis and A. mujanensis represent taxa at a very low level of speciation. The only reason to increase their taxonomic status to independent species lies in the results of experimental hybridisation that is in conflict with other data sets. Thus, species status for the taxa in question would be justified if one considers postzygotic reproductive isolation of the same taxonomic weight as speciation time and previous gene flow. Taking into consideration the discussion above, we suggest recognising A. evoronensis and A. mujanensis as subspecies of A. maximowiczii : A. m. evoronensis and A. m. mujanensis .