A review of the cavernicolous Trichopolydesmidae (Diplopoda, Polydesmida) from the Carpathian-Balkan arch and the Rhodope Mountains, with descriptions of two new genera and three new species Author Antic, Dragan https://orcid.org/0000-0002-1231-4213 University of Belgrade-Faculty of Biology, Institute of Zoology, Studentski trg 16, 11000 Belgrade, Serbia & Serbian Biospeleological Society, Trg Dositeja Obradovica 2, 21000 Novi Sad, Serbia dragan.antic@bio.bg.ac.rs Author Vagalinski, Boyan Institute of Biodiversity and Ecosystem Research at the Bulgarian Academy of Sciences, 2 Yurii Gagarin Street, 1113, Sofia, Bulgaria Author Stoev, Pavel https://orcid.org/0000-0002-5702-5677 National Museum of Natural History, Bulgarian Academy of Sciences, Tsar Osvoboditel Blvd. 1, Sofia 1000, Bulgaria & Pensoft Publishers, Sofia, Bulgaria Author Akkari, Nesrine https://orcid.org/0000-0001-5019-4833 3 text ZooKeys 2022 2022-04-19 1097 1 46 http://dx.doi.org/10.3897/zookeys.1097.83916 journal article http://dx.doi.org/10.3897/zookeys.1097.83916 1313-2970-1097-1 66F17436DF364FFDB2D3021F14D40D62 6593D2730A295F46B63A738B12E43A22 Bacillidesmus filiformis (Latzel, 1884) Figs 1 , 17A-C Brachydesmus filiformis Latzel, 1884: 128, 129. Bacillidesmus filiformis - Attems 1898 : 481, figs 97, 98; Attems 1940 : 170, fig. 244; Strasser 1962 : 443, 444; Strasser 1966a : 341-343; Kime and Enghoff 2011 : 72. Diagnosis. As for the monospecific genus. Material examined. Lectotype ♂ (NHWM MY3754), designated herewith, "SO Ungarn ", leg. Latzel , don. Latzel 1919. One microslide with only one gonopod. Body in two pieces in ethanol: head with rings 1-6 and rings 8-19; second gonopod, antennae and ring 7 missing. Paralectotype . 1 ♀ (NHWM MY10266), whole body in ethanol, same data as for lectotype . Distribution. Unknown. Remarks. In the original description, Latzel (1884) stated that he analyzed one pair (1 ♂, 1 ♀) that he had collected personally in "suedoestlichen Ungarn" (= southeastern part of the Kingdom of Hungary). Later, Strasser (1962) assumed that the species came from "present-day Yugoslavia north of the Danube". This refers to today's Vojvodina, northern Serbia. However, the southeastern part of the Kingdom of Hungary included both Banat Mountains and Southern Carpathians (= Transylvanian Alps) in present-day Romania. Bearing in mind that this area is already inhabited by three trichopolydesmid genera, it seems more plausible that Bacillidesmus filiformis could have originated from present-day Romania, rather than northern Serbia which is characterized mainly by agricultural fields. It also remains unknown if this species is cavernicolous or epigean. This taxon was originally described as Brachydesmus filiformis Latzel, 1884. Attems (1898) analyzed both Latzel's specimens of Brachydesmus filiformis , and based on numerous differences with the genus Brachydesmus Heller, 1858, he correctly established a new genus, Bacillidesmus . At the same time, Attems (1898) gave the first gonopod drawing of this taxon (Fig. 17A ). Later, in his famous " Polydesmoidea III", Attems (1940) provided a new drawing of the Brachydesmus filiformis gonopod (Fig. 17B ), which is slightly different from his 1898 drawing. After studying Attems' microslide with only one gonopod in poor condition (Figs 1G , 17C ) we can confirm that it coincides a bit more with his schematic drawing from 1940. Unfortunately, the second gonopod, as well as ring 7 and both antennae of the lectotype, are most likely lost. It remains unclear whether Attems could have used the now-lost gonopod for the first drawing, or in both cases he used this one, which is still present today, but over time there have been partial changes in its position on the microslide or a partial deformation. Given that Attems (1898) also made a drawing of the antenna, which is missing today, it is very possible that there was another microslide with the second gonopod and antenna/antennae, which we failed to find. However, based on Attems' (1898 , 1940 ) drawings and the newly examined type material of the gonopod, some conclusions could be drawn here. The genus Bacillidesmus had remained monospecific until Strasser (1962) provisionally included therein a new taxon from Bulgaria, based on a single female. Just a few years later, and this time with males in the hands, Strasser (1966a) confirmed that two more taxa belonged to the genus Bacillidesmus , viz., B. bulgaricus bulgaricus Strasser, 1962 and B. bulgaricus dentatus Strasser, 1966a. However, after a detailed examination of the type material of Bacillidesmus filiformis , as well as material of B. bulgaricus bulgaricus and B. bulgaricus dentatus , and two related new species from Serbia and Bulgaria, we believe that Bacillidesmus should include only Brachydesmus filiformis , while the remaining aforementioned taxa should be assigned to a new genus, Balkanodesminus gen. nov., which we describe below. The new genus differs significantly from Bacillidesmus both in somatic and gonopodal characters. The most striking difference in the gonopod structure is that in Bacillidesmus filiformis the solenomeral branch is simple, without a distal solenomeral process, while in Bacillidesmus bulgaricus Bacillidesmus bulgaricus , Bacillidesmus bulgaricus Bacillidesmus dentatus and the two new species it is transversely bifid. In addition, these two genera differ significantly in several somatic traits: Bacillidesmus has regular rows of metatergal setae, mainly four, whereas Balkanodesminus gen. nov. shows 4-8 irregular rows; sensilla basiconica on antennomere 6: completely enclosed inside the pit in Bacillidesmus , vs. partially exposed outside the pit in Balkanodesminus gen. nov.; setae on paraprocts: 2+2 long setae in Bacillidesmus , vs. 2+2 long and ca. 10+10 shorter ones in Balkanodesminus gen. nov.; setae on hypoproct: 1+1 long distal setae in Bacillidesmus , vs. densely setose, including two long distal setae in Balkanodesminus gen. nov.; femora of all male legs swollen in Bacillidesmus , vs. only femora of legs 1-3 swollen in Balkanodesminus gen. nov.; anterior male legs in Bacillidesmus with ventral denticles, vs. denticles absent in Balkanodesminus gen. nov. These diferences are sound enough to propose a new genus for the taxa described by Strasser ( B. bulgaricus bulgaricus , B. bulgaricus dentatus ) and the two newly described species. Moreover, Bacillidesmus filiformis seems to show more affinity to some of the Carpathian genera (which is another proof that this genus could be from the Carpathians, see under Banatodesmus and Trichopolydesmus ), while Balkanodesminus gen. nov., from the Balkan Mountains, shares many similarities with Rhodopodesmus gen. nov. (see below).