Trematodes of Genera Gyrabascus and Parabascus from Bats in European Russia: Morphology and Molecular Phylogeny

Simple Summary The ecology of bats determines their unique parasitic fauna. Most species of worms from bats are highly specialized parasites. We studied parasitic worms of bats that died of natural causes, using classical morphological and molecular phylogenetic approaches. Original drawings, descriptions, and results of molecular phylogenetic analysis for five species of trematodes were provided. We established a broad morphological variability in the studied trematode species, which means that the identification of closely related species may be problematic for researchers. We proposed a taxonomic key for the reliable identification of the studied trematode species. The results of our study contribute to the knowledge of bat helminths and host-parasite relationships in general. Abstract Morphological variability of trematodes from bats (Chiroptera) is poorly studied. Since the variability of adult digenean specimens may be rather high, morphological features are often insufficient for the identification of closely related species, and confirmation with the use of molecular data is required. The aim of our study was to combine the morphological and molecular phylogenetic analyses of several bat trematodes from the genera Gyrabascus and Parabascus (Pleurogenidae): Gyrabascus amphoraeformis, Gyrabascus oppositus, Parabascus lepidotus, Parabascus duboisi, and Parabascus semisquamosus, of which G. amphoraeformis and G. oppositus are little known in European Russia. We made detailed morphological descriptions of these trematodes from several definitive hosts, analyzed morphometric features, and generated new partial sequences of the 28S rRNA gene. A broad variability of trematodes of the genera Gyrabascus and Parabascus was revealed both from various host species and from specimens of the same host species. We propose a new taxonomic key for the identification of the studied species. Certain host specificity of these trematodes was revealed.


Introduction
Bats (Chiroptera) are the only mammals capable of active flight which strongly affects their food chains. Chiropterans harbor a unique helminth fauna, and the host specificity of their parasites is generally high [1,2].
Presently, 35 species of helminths are known for bats in the Middle Volga region, including 23 trematodes [15][16][17][18][19][20][21][22][23]. The study of trematodes, as well as of other parasitic worms, is necessary due to their epidemiological and epizootic importance. Some species of trematodes are known as causative agents of dangerous helminthiasis. The parasitological Smolny village (54 • 43 23 N, 45 • 17 03 E). No animals were killed intentionally for our research. Some dead specimens of bats were kindly provided by the staff of the Mordovia Nature Reserve and National Park "Smolny". Several dead bats were provided by rural residents. Animals died of natural causes or were killed by domestic cats. Necropsy was performed on bats within approximately 1-9 h of their death. Only alive motile adult trematodes were collected for further investigations.
For the morphological examination, the trematodes were recovered from the intestine and killed by careful heating in water under identical conditions. The trematodes were stained with aceto-carmine, dehydrated in an ethanol series (70-96%), cleared in clove oil, and mounted in Canada balsam [45,46]. Trematode specimens for molecular phylogenetic analysis were fixed in 96% ethanol and stored at +4 • C.
In total, 109 specimens of trematodes were studied: 25 Gyrabascus oppositus, 1 Gyrabascus amphoraeformis, 28 Parabascus duboisi, 25 Parabascus lepidotus, and 30 Parabascus semisquamosus. Drawings were made using an MBI-9 light microscope with the Levenhuk M500 BASE Digital Camera and drawing tube RA-7. All the measurements are given in micrometers. For a comparative analysis of the morphology and measurements of trematodes, we used only those works that contained morphological drawings and morphometric data on the trematodes species of interest.
The taxonomic identification and morphological examination of the helminths were carried out in the Laboratory of Population Ecology of the Institute of Ecology of the Volga Basin of the Russian Academy of Sciences (Togliatti, Russia). The trematodes were identified according to Zdzietowiecki [28], Khotenovsky [29,47,48], Skvortsov [49], Sharpilo, Iskova [33], Odening [27], Kirillov et al. [17], and Sokolov et al. [30]. The voucher specimens of trematodes are stored in the parasitological collection of the Institute of Ecology of Volga Basin of RAS (IEVB RAS), a branch of the Samara Federal Research Center of the Russian Academy of Sciences.

DNA Extraction, Amplification, Sequencing, and Phylogenetic Analysis
In order to obtain 28S rDNA and sequences, the specimens were dried on ethanol in a dry block heater for 1.5 h at 35 • C, digested with a mixture of 49 µL 0.1% Chelex-100 and 1 µL Proteinase K (concentration 10 mg/mL), and incubated for 1 h at 55 • C and 25 min at 95 • C. After that, the water solution of the total DNA was placed into a sterile 500 µL tube and frozen. All DNA was extracted from single worms.
The D1-D3 domain of LSU rDNA (approximately 1000-1300 bp long) was amplified using several primers. The thermal cycle parameters are shown in Supplementary Table S1. The newly obtained sequences from both forward and reverse primers were assembled using Chromas Pro 1.7.4. (Technelysium Pty Ltd., Brisbane, Australia). After assembling and trimming low-quality parts of contigs, the sequences were mounted in general alignment. All specimens were used for PCR after the preliminary morphological examination.
Sequences for general alignment were downloaded from the GenBank database with a custom script based on the "ape" package in the "R studio" [51,52]. Newly obtained sequences were aligned together with others using the "Muscle" algorithm as implemented in the "R studio" "msa" package [53,54]. Information on sequences is given in Supplementary Table S2. The alignment was then trimmed manually in SeaView software to a length of approximately 90% of sequences [55]. The final length of alignment was 1260 bp.
The evolutionary model for Maximum likelihood and Bayesian inference analysis was chosen with MrModeltest v. 2.4 [56]. The best-fitted model was GTR + G + I. Maximum likelihood analysis was performed through the Cipres portal [57] with non-parametric bootstrap with 1000 pseudoreplicates. Bayesian analysis was performed using MrBayes 3.2.7 with computational resources provided by Resource Center "Computer Center of SpbU" [58]. Trees were run as two separate chains (default heating parameters) for 15 million generations, by which time they ceased converging. The quality of the chains was estimated using built-in MrBayes tools and additionally with the Tracer 1.6 package [59]. Based on the estimates by Tracer, the first 25,000 generations were discarded for burn-in.

Molecular Phylogenetic Analysis
We generated 14 new sequences of partial 28S rRNA genes for five trematode species and distinguished their relationships with closely related digeneans (Figure 1). We did not discuss the relationships among species of the families Microphallidae, Lecithodendriidae, Phaneropsolidae, and Prosthogonimidae in detail. In our analysis, Microphallidae is the sister taxa to Lecithodendriidae + Phaneropsolidae + Stomylotrematidae. Stomylotrema vicarium (Stomylotrematidae) forms a sister clade to Phaneropsolidae with a relatively low nodal support in Bayesian Inference analysis. In Maximum likelihood analysis, Stomylotrematidae is a separate clade sister to Lecithodendriidae. Pleurogenidae is a sister to the previously described taxa with a relatively low posterior probability. Prosthogonimidae are a basal clade to Microphallidae + Lecithodendriidae + Phaneropsolidae + Stomylotrematidae, and Pleurogenidae. Pachypsolus irroratus (Pachypsolidae) is basal to the other trematodes under consideration.
Among pleurogenids, Gyrabascus spp. form a sister clade to Parabascus spp. with a relatively high Bayesian probability. Three newly generated sequences of G. oppositus (ex N. leisleri, N. noctula, and P. nathusii) cluster together with the previously obtained sequence of G. oppositus (GenBank No MK575195, ex P. kuhlii). Gyrabascus amphoraephormis ex M. brandtii clustered with other previously obtained specimens of these species. Two specimens of P. duboisi (both ex M. daubentonii and M. brandtii) were found to be closely related to the previously known P. duboisi sequence (GenBank No AY220618). Four newly obtained sequences of P. semisquamosus (ex N. noctula and P. nathusii) formed a compact clade without any clear correlation with the host species. Parabascus joannae formed a sister clade to P. semisquamosus. Four newly obtained specimens of P. lepidotus (ex N. noctula and V. murinus) clustered together without any clear correlation with the host species and form a sister clade to P. joannae + P. semisquamosus. Collyriclum and Loxogenes were found to be closely related. Both genera are sister groups to the other pleurogenids.  General description (based on 25 adult specimens): Body pear-shaped, elongated, with maximum body width at testes level. Body narrows toward conical anterior end of body and widens toward rounded posterior end. Body densely covered with spines except at posterior end. Oral sucker subterminal, drop-shaped, with elongated anterodorsal side. Prepharynx short. Esophagus long. Intestinal bifurcation approximately at border of anterior and middle thirds of body. Intestinal branches long, extending beyond testes level but not reaching posterior body end. Ventral sucker equatorial, transversely elongated, always larger than oral sucker. Testes oval, lying behind ventral sucker in hind body, approximately at the same level or one slightly behind the other. Cirrus sac absent. Seminal vesicle convoluted, lying freely in parenchyma approximately at level of ventral sucker, more or less overlapping with it. Genital pore submedial, opens at ventral sucker level on opposite side from ovary. Ovary round or oval, submedially located at lateral edge of ventral sucker, may be partially overlapped by it. Vitellarium consists of numerous oval or irregularly shaped follicles. Follicles located between level of intestinal bifurcation and ventral sucker, or slightly in front of it. Vitelline fields do not extend below ventral sucker level. Uterus forms numerous transverse loops, occupies all space in hind body, behind level of ventral sucker, and completely overlaps testes. Excretory pore terminal.

Systematics and Morphological Characteristics
Remarks. Trematodes G. oppositus from N. noctula and from P. nathusii are morphologically similar and differ in body size only (Table 1 and Figure 2). The body size varied both in specimens from different hosts and in specimens from the same host. Trematodes from P. nathusii had the same width as those from N. noctula but a smaller body length. So, trematodes ex P. nathusii had a body length from 652 to 948 (in N. noctula 941-1230) and a width of 348 to 460 (in N. noctula 326-450). Correspondingly, the body length to width ratio in trematodes ex P. nathusii is 1.8-2.2:1 (average 2.0:1), while for trematodes ex N. noctula it is 2.6-3.1:1 (average 2.8:1) ( Table 1). Specimens from P. nathusii are wider in the posterior half of the body and less elongated, while those from N. noctula are more elongated. The size of oral and ventral suckers varied both in specimens from different host species and in specimens from the same host species. Thus, the oral sucker width of trematodes from N. noctula varied from 54 to 75, while that of trematodes from P. nathusii varied from 49 to 69. The width of the ventral sucker of specimens from N. noctula varied from 91 to 114, while that in specimens from P. nathusii, from 102 to 126. Accordingly, the oral sucker width of trematodes G. oppositus from different hosts ranged from 49 to 75, while the ventral sucker width ranged from 91 up to 126. The width ratio of the oral sucker to ventral sucker remained relatively constant. Larger suckers were noted in specimens from N. noctula. The pharynx size varied in G. oppositus from the two hosts, being 39-51 in specimens from N. noctula and 30-37 in specimens from P. nathusii (Table 1). This feature changed only slightly in specimens from one host species. Considerable differences were noted in the esophagus length of trematodes G. oppositus both from the two host species and from the same host species. A longer esophagus was noted in G. oppositus from N. noctula, 148-217 (from P. nathusii 126-177) ( Table 1). Variability was noted in the size of testes and ovaries in trematodes both from the two host species and from the same host species. The reproductive organs were larger in G. oppositus from N. noctula. The size of eggs in trematodes varied slightly (Table 1).   General description (based on 28 adult specimens): Body fusiform, ovoid, lanceolate, or oval. Body densely covered with spines except at posterior end. Oral sucker round, subterminal. Prepharynx not visible. Ventral sucker pre-equatorial; equal to or somewhat smaller than oral one. Intestinal bifurcation located at anterior edge of ventral sucker. Intestinal branches long, extending beyond testes level, not reaching hind body. Testes round or oval, located below ovary level or at some distance from it. Testes approximately at same level or one somewhat behind the other. Cirrus sac elongated, club-shaped, located in ventral sucker region or directly behind it, lies across the body, or at an angle to longitudinal axis of body. Proximal part of cirrus sac may touch the ovary, more or less overlap it. Distal part of cirrus sac in most cases reaches the anterior edge of testes or, less often (in three cases), proximal end of cirrus sac touches the testes. Both testes are at some distance from cirrus sac. Convoluted seminal vesicle occupies all proximal parts of cirrus sac. Genital pore submedial at level of ventral sucker or somewhat behind it. Ovary round or oval, located at ventral sucker level or somewhat behind it. Vitellarium consists of numerous irregularly shaped follicles, begins at level of intestinal bifurcation or slightly above it, reaches level of posterior edge of ventral sucker or goes somewhat beyond it. Uterus forms numerous loops, fills all the space below level of ventral sucker, partially overlaps testes. Eggs oval. Excretory pore terminal.
Remarks. Specimens of Parabascus duboisi from M. daubentonii and M. brandtii differ in morphology and size of individual organs (Table 1, Figures 3 and 4). The largest specimens of this trematode as well as the smallest ones were noted in M. daubentonii. Trematodes from M. brandtii were intermediate in size (Table 1). The body length to width ratio varied both in specimens from different hosts and in specimens from one host species. This value varied in trematodes from the two host species in the range from 1.5 to 3.0 ( Table 1) Table 1). The width of the oral sucker to ventral sucker ratio remained relatively constant in specimens from both host species, varying within 1.0-1.4: 1. The pharynx size and the width of the oral sucker to pharynx ratio in P. duboisi differed insignificantly. A greater variability was observed in trematodes from one host and even from one host specimen. This was also noted in the case of the length of the esophagus and ovary size. The variability in the size of testes in trematodes from different host species was more marked. The largest testes were noted in trematodes from M. brandtii. This also applies to the size of the cirrus sac ( Table 1). The body shape changed significantly, both in specimens from different host species and in specimens from the same host species. Variability in the position of the cirrus sac and the interposition of the ovary and testes was noted. Testes could be located directly behind the ovary. In this case, one of the testes could touch the ovary, or the testes were located at some distance from it. The variability of the length of vitelline fields was small (Figures 3 and 4). The egg size in P. duboisi was constant and did not depend on the host species. Intestinal branches long, extending well beyond testes level, but do not reach hind body. Ventral sucker pre-equatorial, larger than oral sucker, located in posterior part of anterior third of body. Testes rounded or oval, some distance behind ventral sucker, approximately at same level or one slightly behind the other. Cirrus sac elongate and club-shaped, situated immediately behind ventral sucker at an angle to longitudinal axis of body. Convoluted seminal vesicle located in proximal part of cirrus sac. Proximal part of cirrus sac located at lateral or posterolateral edge of ovary and may overlap it to a greater or lesser extent. Convoluted seminal vesicle occupies whole proximal part of cirrus sac. Genital pore is submedial at level of ventral sucker on opposite side from ovary. Ovary rounded or oval, located submedially behind ventral sucker. Vitellarium consists of numerous oval, rounded, and pear-shaped follicles located between intestinal bifurcation and ventral sucker. Vitelline fields do not extend beyond level of posterior edge of ventral sucker. Uterus forms numerous loops, occupies all space behind ovary. Terminal part of uterus with well-defined metraterm. Excretory pore terminal.
Remarks. The specimens of P. semisquamosus from N. noctula and P. nathusii are morphologically similar (Figure 4) but differ in the size of the body and individual organs. The body size of specimens from different host species varied widely. The largest trematodes were noted in N. noctula: body length 1354-1892 with a width of 277-460; the smallest ones were from P. nathusii: body length 1031-1523, with a width of 246-415. Parasites from P. nathusii had approximately the same width as those from N. noctula, with a smaller body length (Table 1). Accordingly, specimens from P. nathusii were wider in the mid body, and those from N. noctula were narrower. The variability of the body size was less marked in specimens from the same host species. The size of oral and ventral suckers of P. semisquamosus from various hosts was approximately the same, as was their ratio. Considerable differences were noted in the size of oral and ventral suckers from the same host. Thus, the oral sucker width in trematodes from N. noctula was 46-67 (from P. nathusii , and the ventral sucker width was 75-106 (from P. nathusii 73-110). The same applies to the size of the pharynx (Table 1). The esophagus length of P. semisquamosus varied in specimens from the two host species. A longer esophagus was noted in specimens from N. noctula: 187-335, as compared with 150-268 in specimens from P. nathusii. High variability was observed in the size of reproductive organs of trematodes, regardless of the host species. Larger testes, cirrus sac, and ovary were noted in trematodes from N. noctula. Egg size was a constant characteristic in trematodes from both host species.  Remarks. Significant variability in size and morphology was noted in P. lepidotus from V. murinus and N. noctula ( Figure 5). The largest specimens were noted in N. noctula: body length 844-1160 with a width of 385-619; the smallest one, in V. murinus: body length 441-770 with a width of 272-519. The variability in body size was especially pronounced in P. lepidotus from V. murinus. In this host, trematode specimens strikingly different in width were noted. The body length to width ratio in specimens ex V. murinus was 1.1-1.8:1 (average 1.6:1), and that in specimens ex N. noctula was 1.5-2.5:1 (average 2.1:1). (Table 1). The size of the oral and ventral suckers from the two host species varied. The oral sucker width of specimens from N. noctula varied from 51 to 83, while that of specimens from V. murinus varied from 43 to 67. The width of the ventral sucker of specimens from N. noctula varied from 63 to 98, and that in specimens from V. murinus, from 45 to 75. In P. lepidotus from V. murinus, the oral sucker could be smaller than the ventral sucker, but more often the suckers were equal in size. In trematodes from N. noctula, the oral sucker was always smaller than the ventral one. The oral sucker to ventral sucker ratio in specimens from different hosts was constant. This ratio was more variable in trematodes from the same host species. The pharynx of P. lepidotus specimens from N. noctula was slightly larger than that of the specimens from V. murinus. The width of the oral sucker to pharynx ratio was approximately the same in trematodes from various hosts. Considerable variability in the esophagus length was noted in P. lepidotus specimens from both host species. The esophagus of specimens from N. noctula (217-293) was much longer than that of specimens from V. murinus . The size of the testes, ovary, and cirrus sac varied greatly in P. lepidotus specimens from the two host species. These features were less variable in trematodes from the same host species. The largest organs were noted in specimens from N. noctula. Differences in the body shape of parasites from different host species were noted. In P. lepidotus specimens ex V. murinus, the body is pear-shaped with a tapering anterior end and a rounded, widened posterior one. In specimens from N. noctula, the body is more spindle-shaped with tapering anterior and posterior ends. Differences in the location of the ventral sucker were also noted. The position of the cirrus sac was different in specimens from various host species. In P. lepidotus specimens from V. murinus, the cirrus sac was always located submedially at the ventral sucker level, partly overlapped by it. In specimens from N. noctula, the cirrus sac was located medially and directly behind ventral sucker. Variability was noted in the length of vitelline fields in specimens from various hosts. In P. lepidotus specimens from V. murinus, the posterior edge of the vitelline follicles did not extend beyond the ventral sucker level. In P. lepidotus specimens from N. noctula, the vitelline fields sometimes extended beyond the level of the lower edge of the ventral sucker. The anterior edge of the vitelline fields in P. lepidotus specimens from both host species was at the level of intestinal bifurcation or slightly above it. Egg size was constant regardless of the host species.

Discussion
In this study, we presented morphological descriptions of five species of trematodes from the genera Gyrabascus and Parabascus from various species of bats from Mordovia (Russia) and novel molecular phylogenetic data on these parasites. The combined use of molecular and morphological methods made it possible to perform a reliable identification of Gyrabascus amphoraeformis, Gyrabascus oppositus, Parabascus duboisi, Parabascus lepidotus, and Parabascus semisquamosus.
The general topology of the tree obtained in this study is in good agreement with previous publications [30,60,61]. The two exceptions are the families Stomylotrematidae and Phaneropsolidae. In Shchenkov et al. [60] and Sokolov et al. [30], Stomylotrematidae was the closest to Lecithodendriidae, while Phaneropsolidae formed a sister clade to Lecithodendriidae + Stomylotrematidae branch. In Dellagnola et al. [62], Stomylotrematidae and Phaneropsolidae were sister clades to each other and close to Microphallidae. In Dellagnola et al. [62], Lecithodendriidae appeared to be the basal clade to other microphallids. In Fernandes et al. [61], Phaneropsolidae is a sister clade to Lecithodendriidae, while Stomylotrematidae was left out of the analysis. In our analysis, the phylogenetic position of Stomylotrematidae was unstable between ML and BI analyses. All digenean specimens incorporated into our molecular phylogenetic analysis belonged to five distinct species, which were difficult to distinguish based on morphological features only. No impact of host species on the clusterization of the species under consideration was revealed.
Morphometric features such as the body length and width, and the size of the reproductive organs varied broadly, apparently depending to a large extent on the age of the parasites. The dependence of morphometric changes on the trematode age has been noted by several authors [69][70][71][72][73][74] and is especially evident in trematodes from the same host species [73].
Morphological variability of the body shape and the position of the cirrus sac noted in our study also depended on the degree of development of the trematode specimen. The variability of the length of the vitelline fields, the body length to width ratio, and the oral sucker to pharynx ratio were less pronounced. Features such as the oral sucker to the ventral sucker width ratio and the size of eggs were relatively stable ( Table 1).
The observation that the variability of parasites depends on the host species has been made in a number of studies [70,71,[73][74][75][76][77][78][79]. We also found that the morphological variability of trematodes depended on the species of the host. A host-induced variability was recorded in P. oppositus, P. duboisi, P. semisquamosus, and P. lepidotus in respect of the body size, body shape, and the size of inner organs (Table 1 and Figures 2-5).
We gave the first complete description of G. amphoraeformis from M. brandtii, including morphometric data. The comparison of our specimen of G. amphoraeformis with the descriptions of other authors [27,28,33,36,38,48,[63][64][65]67] demonstrated a good agreement in the main morphological and morphometric characteristics, except for the egg size. In our specimen, the eggs were somewhat smaller than in descriptions by other authors (Section 3.2 and Table 2). The specimens of G. oppositus described by us ex N. noctula differed from the previously described specimens from other host species in being longer (Table 3). It should be noted that this species has been previously recorded in P. kuhlii, M. schreibersi, and Eptesicus serotinus Schreber, 1774 [28,30,39]. Pipistrellus nathusii and N. noctula are new hosts for G. oppositus. Specimens of P. semisquamosus examined in our study fully corresponded to the descriptions available in the literature [17,24,[27][28][29]33,34,64,80]. Specimens of P. duboisi examined here differed in body length, size of the oral sucker, and cirrus sac (Tables 1 and 4) from those described by other authors [17,24,[27][28][29]63,66,67]. The difference was mostly due to the fact that our specimens from M. daubentonii were relatively larger than the parasites described earlier. Our specimens of P. duboisi from M. brandtii fully corresponded to the literature data in respect of the measurements (Table 4). Note: 1 -diameter.
The specimens of P. lepidotus from V. murinus obtained in our study were smaller in respect of body size than those described in earlier studies [17,24,28,33,35,37,38,47,63]. Accordingly, their organs were also smaller ( Table 1). Specimens of P. lepidotus from the N. noctula mainly corresponded to the morphological and morphometric characteristics given earlier (Table 5). Related species of trematodes can be identified erroneously, therefore, it is necessary to take into account the morphological and morphometric variability in their diagnosis. Zdzitowiecki [28] noted that P. duboisi and P. lepidotus are morphologically very similar. The results of our study confirm this. It was often difficult to distinguish these species. For example, such a diagnostic morphometric feature as the size of the oral and the ventral sucker was not always applicable. In our material, there were trematode specimens (both P. duboisi and P. lepidotus) in which the size of the oral sucker was equal to that of the ventral sucker. Another diagnostic feature that could not be used in these two species was the esophagus length. There were specimens in which the values of this feature were the same. We conclude that the following morphological and morphometric features should be taken into account when identifying these species: the body length to width ratio, the oral sucker to pharynx width ratio, the oral sucker to the ventral sucker width ratio, and the location of the ventral sucker relative to the mid body.
Zdzitowiecki [28] has also noted that G. oppositus is similar in many morphological features to P. lepidotus and that many early descriptions of the latter [36][37][38]63] could also apply to G. oppositus. This is probably why Khotenovsky [29] reduced G. oppositus to synonyms of P. lepidotus. In addition, these species of trematodes have a common host, N. noctula. Therefore, when identifying these trematode species, one should take into account the following morphological features: the body shape, the oral sucker shape, the size and position of the ventral sucker relative to the mid body, and the presence/absence of the cirrus sac.
In turn, G. oppositus is similar to G. amphoraeformis. Skvortsov [49] considered the former as a synonym of the latter. Distinctive morphological features for these two species are the body shape, the oral sucker shape, the position of the ventral sucker relative to the mid body and the location of the genital pore.
It is much easier to distinguish P. semisquamosus from P. lepidotus and P. duboisi. Here, the distinctive morphological and morphometric features are the body shape, the body length to width ratio, the oral sucker to the ventral sucker width ratio, the position of the ventral sucker relative to the mid body, and the length of the intestinal branches.
Based on our results, we propose a key for identifying Gyrabascus spp. and Parabascus spp. involved in our study. It should be noted, however, that the applicability of this key is somewhat limited due to a broad morphological variability of these trematodes, which is especially pronounced in immature specimens. Special care should also be taken during the identification of adults that could have been deformed during fixation and whole mount preparation.  (Looss, 1907).
An analysis of the literature data and the results of our own studies showed that trematodes of the genera Gyrabascus and Parabascus exhibit a certain host specificity. In our study, P. lepidotus was found only in N. noctula and V. murinus. P. semisquamosus and G. oppositus were noted only in P. nathusii, N. noctule, and N. leisleri. Parabascus duboisi was recorded only in Myotis spp. We found two specimens of G. amphoraeformis in M. brandtii.
Gyrabascus amphoraeformis mainly occurs in Myotis bats. Though several authors reported this species from N. noctula, P. pipistrellus, E. serotinus, and Barbastella barbastellus Schreber, 1774 [24,28,33], all of them provided drawings of specimens from Myotis spp., and so these reports should be treated with caution. However, the presence of G. amphoraeformis in P. kuhlii has been confirmed by molecular analysis [40], indicating that G. amphoraeformis may indeed parasitize bats other than Myotis spp.
According to the literature data, the hosts of P. semisquamosus are N. noctula and P. pipistrellus (Table 6). Khotenovsky [29] is the only source where M. daubentonii is indicated as its host. The description and drawing of the parasite in this work match those of P. semisquamosus, but we believe that an error could have occurred in identifying the host. Sharpilo and Iskova [33] report P. semisquamosus from Myotis mystacinus Kuhl, 1817 (the drawing and description of the specimen from N. noctula). Skvortsov [24] doubts the findings of P. semisquamosus in Myotis bats since the trematode is a specific parasite of Nyctalus and Pipistrellus bats. We have never found P. semisquamosus in Myotis spp. in our long-term studies of bat helminths [15][16][17][18][19][20][21][22][23]. Parabascus lepidotus has previously been recorded mainly in E. serotinus. Other bats mentioned as its hosts are E. nilssoni, N. noctula, P. kuhlii, P. pipistrellus, P. auritus, V. murinus, M. blythii, and M. nattereri [17,24,28,29,33,35,37,38,47,63]. Findings of P. lepidotus in Myotis bats require confirmation, as the authors may have dealt with a closely related species, P. duboisi, which is a common parasite of Myotis spp. No drawings of P. lepidotus from Myotis bats are given in the studies cited above, except for Zdzietowiecki [28]. His work contains a drawing of a parasite from M. nattereri, which, though we cannot be certain, seems very similar to P. duboisi. We have critically reviewed whole mounts of trematodes from bats of the Samarskaya Luka (Russia) [16,17] and found that of all Parabascus spp. only P. duboisi parasitizes Myotis bats.

Conclusions
The combined use of molecular and morphological methods in our study made it possible to reliably identify closely related trematode species Gyrabascus amphoraeformis, Gyrabascus oppositus, Parabascus duboisi, Parabascus lepidotus, and Parabascus semisquamosus. A broad morphological variability of Gyrabascus spp. and Parabascus spp. was revealed, both from various host species and from various specimens of the same host species. We reevaluated morphological characters for a reliable identification of the closely related species of the genera Gyrabascus and Parabascus involved in our study and proposed a key for their identification.
Our data complement and expand the knowledge of bat parasites. We provided the first record of Gyrabascus amphoraeformis from bats in the Volga basin and the first record of Gyrabascus oppositus from bats in the Middle Volga region. We also established three new hosts of Gyrabascus oppositus: N. leisleri, N. noctula, and P. nathusii.