Trachysphaera costata ( Waga, 1857 a)

Antić, Dragan, Šević, Mirko, Macek, Oliver & Akkari, Nesrine, 2021, Review of Trachysphaera Heller, 1858 (Diplopoda: Glomerida: Glomeridae) in Serbia, with taxonomic notes on the genus, Zootaxa 5047 (3), pp. 273-299 : 279-282

2. Trachysphaera costata ( Waga, 1857 a)

Figs 4 , 12A .

Glomeris costata Waga, 1987a: 175–177 .

Glomeris costata —Stronczyński (1855: 170), nomen nudum.

Gervaisia costata — Waga (1857b: 829–932, figs 2–4);? Daday (1889: 73, plate 2, fig. 22); Verhoeff (1906: 810, 816, 817, figs 8–11, 19, 20; 1908: 529, fig. 4); Schubart (1934: 55, 56, figs 71, 75, 76); Jermy (1942: 54–57, figs 192–197); Jawłowski (1927: fig. 1); Lang (1954: 68–70, figs 48, 49); Stojałowska (1961: 82, 83, figs 54, 55).

Gervaisia costata genuina Latzel, 1884: 89 .

Trachysphaera costata — Tabacaru (1988: 52); Golovatch (2008: 100–103, figs 11–17; 2010: 1–5, figs 3–7); Kania et al. (2005: 81–84, figs 1–3); Kime & Enghoff (2011: 35, 125); Kocourek (2013: 31, fig. 77); Kocourek et al. (2017: 88, 89, figs 2–5).

? Trachysphaera hyrtlii Wankel, 1861: 253 , 254, figs 1–3 [synonymized by Latzel (1884: 86); it is very possible that this is synonym of T. schmidtii ].

Gervaisia rotundata Lignau, 1911: 39–41 , plate 1, fig. 1, figs 1–3 [synonymized by Golovatch (2008: 100)].

Trachysphaera rotundata — Golovatch (1976: 41, 42, plate 5, figs 1–4; 1981: 423–425, figs 5–7; 1990: 344–352, figs 9–12, 15–19, 25–32, 46–48, 50–55, 59–62); Kime & Enghoff (2011: 37).

Gervaisia asiaeminoris Verhoeff, 1940: 44 , figs 40–42 [synonymized under rotundata by Golovatch (1990: 344)].

Gervaisia pigmentifer Verhoeff, 1941: 296 , 25–26 [synonymized under rotundata by Golovatch (1990: 344)].

Trachysphaera pigmentifera (sic!)— Strasser (1975: 589, fig. 8).

Material examined. Serbia ( IZB): 31 ♀♀, 3 juvenile ♀♀, Lazareva Cave, village Zlot , near Boljevac, 15.06.1996 ; 1 ♀, same locality, but 10.2013, D. Antić leg.

Remarks. Undoubtedly, together with T. schmidtii Heller, 1858 [= T. acutula ( Latzel, 1884) etc., syn. n., see below], T. costata is the most widespread and most intricate taxon in the genus and caused a lot of confusion in the past. Solving the taxonomic problems referring to this species seems to be hardly possible at the moment. Both costata and schmidtii were described almost at the same time, under two different generic names. Fortunately, the true identity of T. schmidtii has since been resolved by Sillaber (1984, 1987) (see under T. schmidtii ).

The problem concerning the identity of T. costata was emphasized by Golovatch (1990: 350) who stated that it had to be confirmed. Until the treatment of this species by Schubart (1934), there had been much confusion concerning T. costata and its morphological characteristics. Thus, Latzel (1884) synonymized T. schmidtii with T. costata , and the former name remained in oblivion until its revival ( Sillaber 1984, 1987). Similarly, numerous species now considered as synonyms of T. schmidtii have since been described (see below).

What called our attention is that the name costata appeared twice before Waga’s (1857b) paper, hitherto considered as the original description of this taxon. In both occasions, in “Sprawozdanie z podróży naturalistów odbytćj w r. 1854 do Ojcowa. Biblioteka Warszawska” from 1855 and 1857, this taxon is mentioned in the genus Glomeris Latreille, 1802 . In 1855 [see Stronczyński (1855: 170)], costata is mentioned under the genus Glomeris , which we consider here as nomen nudum. The same combination “ Glomeris costata ” is mentioned by Waga (1857 a: 175– 177)], where he provided a full description of the species in Polish. An identical description appears somewhat later in French ( Waga 1857b). This means that Glomeris costata Waga, 1857 a is the original combination to consider for the species. The problem with the authorship of these reports from 1855 and 1857 is known for decades, and only recently did Pawłowski (2005) suggest how they should be cited, and we accept his suggestions here with minor modifications. The reason for putting the authors of these reports in square brackets in the reference list, is that they are not explicitly shown in these works. The species T. costata was originally described from Ojców, southern Poland and in both papers, Waga (1857a, 1857b) stated that he counted 16 pairs of legs on the largest specimen he checked and presumed as a male. We are not able to verify this presumption as on one hand 16 pairs of legs characterize subadult females according to Tabacaru (1963). On the other hand, both subadult and adult males would present 16 pairs of walking legs plus three ultimate pairs of modified legs (leg-pairs 17 and 18 plus telopods). Leg-pairs 17 and 18 are sometimes in close proximity of telopods, giving the impression of all three pairs as one structure. Thus, the probability that Waga (1857 a, b) did actually examined a male is not excluded though it could only be verified once the type material is examined. Furthermore, Waga was, understandably, probably very impressed by the transverse ridges on the tergites after which he named the species, stating that these ridges are very pronounced. However, it also remains unclear if Waga’s focus on these ridges was because they were unknown within Glomerida at the time, or whether they were actually strongly developed like in e.g. T. schmidtii , T. cristangula , T. lobotarsus , T. acutula sensu Verhoeff (1898, 1899 , 1906) (see below about acutula in sense of Verhoeff) and not as in what we consider today as T. costata . Another question that remains open is whether Waga was actually looking at the same taxon that was later redescribed by Schubart (1934) as costata , which paradoxically has low transverse ridges, or alternatively T. acutula sensu Verhoeff , which would better match his observation of strong ridges, given that both taxa can be found in the region where the species was originally described. For the presence of T. costata and T. acutula sensu Verhoeff in Ojców see Dziadosz (1966: 7).

What gives hope that the true identity of this taxon will be resolved in the future is the paper of Kania et al. (2005), where the authors stated that males of T. costata were found in the Ojców National Park (where the species was described). Unfortunately, apart from the original photographs of the two individuals on the basis of which it is not possible to say much, the drawings from Waga (1857b), Stojałowska (1961) and Schubart (1934) are reproduced in this paper, thus no final conclusions can be drawn. Until the type material of this taxon is found and examined, or a correct redescription of the males from the type locality is presented, we can only speculate about the true identity of this animal, also bearing in mind that sympatry or even syntopy is known in the genus Trachysphaera ( Gruber 2009; this study).

The identifications of T. costata are still largely based on the drawings of the telopods and habitus presented by Schubart (1934: 55, 56, figs 71, 75, 76), who collected material in Silesia, a historical region on the borders of Poland, Germany, Slovakia and the Czech Republic, today mostly in Poland. Since Silesia also included the part of the territory of Poland where the species was originally described, it is most likely that Schubart (1934) indeed depicted the telopods of the true T. costata . Telopod illustrations of what was claimed to represent T. costata also appeared prior to Schubart’s (1934) drawings, but in Latzel (1884, fig. 42) and Berlese (1885, fig. 5) (see below) they certainly showed the telopods of T. schmidtii , and likely in Daday (1889, fig. 22) as well.

The general confusion concerning the identity of T. costata was probably exacerbated by the assignment of numerous parthenogenetic populations to this taxon throughout its distribution area [see Enghoff (1978)]. Multiple female and/or juvenile samples recorded in the past seem to have been too easily and too carelessly identified as T. costata . Moreover, several “good” species have been either misidentified as or synonymized with T. costata , leading to the wrong general impression that T. costata is a very widespread species with great variations in the structure of the telopods. Although we present no detailed revision of this species, mainly focusing on its presence in Serbia instead, such an idea is rejected here.

In the latest surveys of the Serbian millipede fauna, Makarov et al. (2004) and Antić et al. (2013) list T. costata for all regions in Serbia, including its northern, eastern, western and southern parts. The first record of T. costata from Serbia belonged to Attems (1929: 312) who listed this species for Kačanik and Ljuboten, both at Mt. Šar Plani- na, southern Serbia, but incorrectly listed those localities as lying in Albania. A little later, however, Attems (1943: 83) again listed these two localities (and again incorrectly in Albania), but this time he referred to T. noduligera ( Verhoeff, 1906) (= T. schmidtii ), not mentioning T. costata whatsoever. It is thus clear that Attems (1943) simply corrected his own earlier (1929) misidentification. Yet, because neither Makarov et al. (2004) nor Antić et al. (2013) noted that correction, the combination T. schmidtii has never appeared in the fauna of Serbia (see below).

In addition to the literature data, Makarov et al. (2004) also listed some new localities for T. costata in Serbia. However, after rechecking this material, we have found that in all but one case it actually represents T. schmidtii (see below). Unfortunately, we have been unable to find the sample from the Kovačevića Cave ( Fig. 12A , blue circle with question mark), but it seems to have represented T. schmidtii as well.

Only one previous record of T. costata from Serbia, that from the Lazareva Cave ( Makarov et al. 2004) ( Fig. 4 ), seems to be reliable. We have had the opportunity to study the material which was collected on two occasions, the first time more than 30 juvenile and adult females, and the second time only one female. Given the large number of females collected and the lack of male specimens, we may assume we deal with a parthenogenetic population of T. costata , but we cannot exclude the possibility that we are wrong. Based on this, it is clear that T. costata is no longer the most widespread species of Trachysphaera in Serbia and that its formerly wider distribution was based on misidentifications with T. schmidtii . Another locality worth mentioning here is the village of Rsovci, Mt. Stara Planina, where 12 females (juvenile and adult) have been found. This sample could also represent a parthenogenetic population of T. costata , but as we do not dare claim that, we have provisionally identified those specimens only as a Trachysphaera sp. (see below).

In conclusion, bearing in mind that some Trachysphaera species can live in sympatry, or even syntopy, we are not excluding the possibility, as stated above, that Waga’s costata could be actually the same as Verhoeff’s acutula (see below under cristangula ). Similarly, the animal described by Schubart (1934) as costata could be a different species. However, we opt here to consider T. schmidtii and T. costata as recognized by Sillaber (1984, 1987) and Schubart (1934), respectively.

Type locality. Ojców , Poland .

General occurrence. Widespread (?), ranging from Central and Eastern Europe, across the Balkans, Crimea, Turkey, Israel and the Caucasus to northern Iran.

Occurrence in Serbia. Only a population from the Lazareva Cave is assigned to this species for the time being ( Fig. 12A ).

Figures

FIGURE 4. Trachysphaera costata (Waga, 1857a), ♀♀ from Lazareva Cave (IZB), habitus. A, B. Dorsal views. C, D. Lateral views. Scale bars: 1 mm.

FIGURE 12. Distribution of the genus Trachysphaera Heller, 1858 in Serbia. A. Identified specimens. B. Unidentified specimens.