Helminths of the mallard Anas platyrhynchos Linnaeus, 1758 from Austria, with emphasis on the morphological variability of Polymorphus minutus Goeze, 1782*

The mallard Anas platyrhynchos is the most abundant water bird species in Austria, but there is no record of its helminth community. Therefore, this work aimed to close that gap by recording and analysing the parasite community of a large number of birds from Austria for the first time. A total of 60 specimens shot by hunters in autumn were examined for intestinal parasites. The following taxa were recovered (prevalence given in parentheses): Cestoda: Diorchis sp. (31.7%) and Fimbriarioides intermedia (1.7%); Acanthocephala: Filicollis anatis (5%), Polymorphus minutus (30%) and one cystacanth unidentified (1.7%); Trematoda: Apatemon gracilis (3.3%), Echinostoma grandis (6.7%), Echinostoma revolutum (6.7%) and Notocotylus attenuatus (23.3%); Nematoda: Porrocaecum crassum (1.7%) and one not identified (1.7%). The frequency distribution of parasites showed a typical pattern in which 39 birds (65%) were either not parasitized or were harbouring up to five worms, whereas more intense infestations occurred in a lesser number of hosts. Compared to other studies from central and eastern Europe, an extremely depauperate helminth community, particularly of the cestodes and nematodes, was found. Polymorphus minutus was observed as having highly variable morphology and, therefore, molecular genetic characterization by DNA barcoding was carried out. Species identification was confirmed by comparing data with the reference cytochrome c oxidase subunit 1 gene sequence from P . minutus available in GenBank. maximum time from delivery to the last dissected animal three days. For parasite inspection, the digestive tract was removed, cut open and inspected under a dissecting microscope. Parasites were removed and stored in 4% formalin or 70% ethanol for identification and quantification. Nematodes and trematodes were identified with the help of the keys by McDonald (1974) and McDonald (1981), respectively. Cestodes were identified using the key by Khalil et al . (1994). Acanthocephalans were identified with the help of the key by Smales (2015) and relevant species genetic characterization by DNA barcoding of species of genera belonging phylum Acanthocephala. Forty-two acanthocephalans were used for the molecular genetic analysis. DNA extrac-tion was conducted in a clean using QIAmp DNeasy Blood and Tissue Kit (QIAGEN, Hilden, Germany) following the protocol of the manufacturer. A partial fragment of approxi-mately 650 base pairs (bp) of the mitochondrial cytochrome c oxidase subunit 1 gene (COI) was used for DNA barcoding, amplified with the following primer pairs: H14AcanCOIFw1 (5-TTCTACAAATCATAARGATATYGG) as forward primer and H14AcanCOIRv2 (5-AAAATATAMACTTCAGGATGACC AAA) as reverse primer (5-GTTWATATATGTWTTGGTTAGATTATG)


Introduction
The common mallard Anas platyrhynchos Linnaeus, 1758 is by far the most abundant waterfowl in central Europe. The mallard is native, widespread over Eurasia and well adapted to cultivated landscapes (Bauer, 2005). Therefore, it seems rather surprising that recent investigations on the parasite community hosted by this bird species are scarce. During a conference on the current state of helminth parasite databases in Europe in 2003, it was stated that 'information on helminth faunas are scattered in a large number of papers, often published in hardly accessible journals and written in national languages' (Sitko, 1997). Several important books, including checklists of digeneans (Sitko et al., 2006), nematodes (Sitko & Okulewicz, 2010) and acanthocephala (Sitko, 2011) in birds in the Czech and Slovak Republics as well as additions to an existing checklist of tapeworms (Ryšavý & Sitko, 1992) were published. Those publications covered the major groups of helminth parasites occurring in some avian hosts. Although these checklists are highly valued contributions, there are no recent studies including not only systematic findings but also descriptions and analyses of the ecological aspects of the parasite communities of avian hosts. Such analyses have been described as useful tools in many aspects of ecology, including parasite infracommunity ecology (Simberloff & Moore, 1997). Even on a small scale such analyses can be helpful, as shown, for example, by Sheehan et al. (2016) who distinguished between migrating and stationary cormorant populations by analysing their helminth communities. The presence or the absence of a parasite in an ecosystem may indicate the presence or absence of its definitive host, even if it has not been observed. As importantly, perhaps, as they are often less-studied, the presence or absence of a parasite in an ecosystem indicates the presence of its intermediate host/s. On a larger scale, the assessment of environmental conditions and changes in ecosystems using parasites as bioindicators has been one of the emerging topics in the field of environmental parasitology (Lafferty, 1997;Vidal-Martínez et al., 2010;Nachev & Sures, 2016).
Trematodes, in particular digenea, are by far the largest group of internal metazoan parasites, with 18,000 nominal species (Cribb et al., 2001) and an important parasite group of birds. Although this huge variety has been documented, new species and lineages are discovered whenever an ecosystem is under investigation and the knowledge, particularly of this group, has been described as patchy at best (Sures et al., 2017). Due to their complex life cycles, including a vertebrate (e.g. a bird) as definitive host and at least one intermediate host, their survival depends on manifold factors including the presence of the suitable hosts and the availability of appropriate environmental conditions for transmission of the life stages from one host to another. The diversity of trematodes found in an ecosystem is considered a suitable indicator for ecosystem health by, amongst others, Huspeni & Lafferty (2004) and Shea et al. (2012). High diversity, in this instance, refers mostly to the larval stages of digeneans found in snails and the diversity of larval stages in mallards is poorly known. Studies on A. platyrhynchos report varying numbers of trematode species: Chu et al. (1973) reported five species of intestinal trematodes from wild ducks in Korea; Fedynich & Pence (1994) reported six intestinal species from mallards in Texas; Kavetska et al. (2008) identified 17 species in Polish hosts; and 43 species of trematodes from the intestine and other locations have been recorded in different studies of the mallard from the Czech and Slovak Republics by Sitko et al.(2006). Echinostoma revolutum (Fröhlich, 1802) was the only species present in all studies, being described as the most widely distributed of the echinostome species. Echinostoma revolutum is pathogenic to humans when the intermediate hostsraw snails and frogsare consumed (Chai, 2009).
As already stated by Janovy (1997), there is a lack of research on avian cestodes compared to other groups of parasites. This has been highlighted in more recent studies emphasizing that the diversity of cestode species is highly dependent on the number of host species examined. This has been best documented in game waterfowl. Marinova et al. (2013) recorded 26 cestode species in A. platyrhynchos from Bulgaria. A similarly high diversity was described by Kavetska et al. (2008), who identified 23 different species, 22 belonging to the family Hymenolepididae and one from the family Dilepididae, from mallards in Poland. Fewer species (13) were recorded by Fedynich & Pence (1994), although it is worth noting that these authors did not cite how many different species they subsumed as Diorchis spp. Thirteen species were described by authors from former Czechoslovakia (Pav & Zajicek, 1963). It is noteworthy that, even in neighbouring countries such as Poland and former Czechoslovakia that display a high diversity of species, only four are described from bothnamely, Aploparaxis furcigera (Rudolphi, 1819), Dicranotaenia coronula (Djuardin, 1845), Fimbriaria fasciolaris (Pallas, 1781) and Sobolevicanthus gracilis (Zeder, 1803). This emphasizes the need for molecular identification and clarification.
For this work we had the opportunity to get 60 freshly shot mallards directly from the hunters, from hunting grounds in northern and eastern Austria. The first aim of this project was to contribute to the knowledge of the intestinal parasite community of A. platyrhynchos by presenting the first report from Austria. The second was to confirm the identity of any specimens of the genus Polymorphus recovered in this survey. Analyses of prevalence and intensity of infection in this helminth community were carried out to give insight into species dominance, richness and evenness, thus allowing comparisons with helminth community structures from central European and other avian host-parasite systems. Further, some insights might be gained into the ecological status of the aquatic ecosystems of north-eastern Austria.

Study site
Sampling for this study took place in the north-eastern part of Austria ( fig. 1), which includes the capital Vienna (1.9 million inhabitants), with its industrial impact on the environment. The surrounding region is a typical man-made environment characterized by a population density of 84 inhabitants/km 2 (Statistics Austria, 2014) dominated by agricultural activity and managed forests. The region lies within the catchment area of the Danube River and floodplains, smaller rivers and numerous lakes and ponds are present. Water quality is monitored in Austria regularly. The most recent available report from 2019, including data from 2016, states that 99% of the investigated water bodies were in 'very good' or 'good' physico-chemical conditions. Looking at the biological parameters, including macrozoobenthos and phytobenthos, 44% of surface water bodies must be considered 'mediocre' or worse (BMNT, 2019). Hunting is executed within the whole region, except for congested areas. By far the most abundant waterfowl is the mallard A. platyrhynchos, which is native, widespread over Eurasia and well adapted to cultivated landscape. The numbers of wild fowl are estimated to be 10,000-20,000 breeding pairs in Austria (Bauer, 2005).

Animal sampling
In total, 60 specimens of A. platyrhynchos were received from six hunting grounds situated in the eastern parts of Austria ( fig. 1). They were shot by hunters during autumn, using commercially available lead shot. Freshly killed animals were transported to the University of Veterinary Medicine in Vienna, where sex and weight were determined. Specimens were stored at 4°C until dissection, which was conducted as soon as possible, with a 2 F. Jirsa et al.
maximum time from delivery to the last dissected animal being three days. For parasite inspection, the digestive tract was removed, cut open and inspected under a dissecting microscope. Parasites were removed and stored in 4% formalin or 70% ethanol for identification and quantification. Nematodes and trematodes were identified with the help of the keys by McDonald (1974) and McDonald (1981), respectively. Cestodes were identified using the key by Khalil et al. (1994). Acanthocephalans were identified with the help of the key by Smales (2015) and relevant species descriptions. The parasitological terms used are those defined by Bush et al. (1997). Measures of the helminth community structure were the Shannon-Wiener index and evenness, Simpson's index and the Berger-Parker index as proposed by Magurran (1988). Specimens of P. minutus were deposited in the South Australian Museum with registration numbers AHC 47,980 and 47,981. All other specimens were deposited in the Museum of Natural History, Vienna.

Molecular genetic methods
Acanthocephalans were all placed in 70% ethanol prior to examination as temporary wet mounts cleared in beech wood creosote for identification. As the morphological identification of P. minutus was ambiguous, specimens of adult P. minutus from waterfowl from the UK, registered in the Natural History Museum, London (BMNH 1923(BMNH .12.19.92-95, 1927(BMNH .10.14.13-32, 1927(BMNH .10.17.3-4, 1927(BMNH .12.15.121-122, 1935(BMNH .4.16.152-157, 1946(BMNH .5.14.107-120, 1956(BMNH .8.16.11-25,1969(BMNH .347-355, 1973; from the Rhine, Germany, on loan from the Karlsruhe Institute of Technology, as well as cystacanths from Gammarus pulex (SAM AHC47567); from Germany from Gammarus roeselii (SAM AHC47568-71); from the Czech Republic (A54/1), from G. pulex from Ireland, and from Echinogammarus berilloni from France on loan from the Karlsruhe Institute of Technology, were examined for comparative purposes. For further support of the morphological identification, we conducted a molecular genetic characterization by DNA barcoding of species of genera belonging to the phylum Acanthocephala. Forty-two acanthocephalans were used for the molecular genetic analysis. DNA extraction was conducted in a clean room using the QIAmp DNeasy Blood and Tissue Kit (QIAGEN, Hilden, Germany) following the protocol of the manufacturer. A partial fragment of approximately 650 base pairs (bp) of the mitochondrial cytochrome c oxidase subunit 1 gene (COI) was used for DNA barcoding, amplified with the following primer pairs: H14AcanCOIFw1 (5-TTCTACAAATCATAARGATATYGG) as forward primer and H14AcanCOIRv2 (5-AAAATATAMACTTCAGGATGACC AAA) as reverse primer (Reier et al., 2019) for the species F. anatis and Poly-1+ (5-GTTWATATATGTWTTGGTTAGATTATG) as forward and Poly-2-(5-AATAAATGCTGATAYAAWAR AGG) as reverse primers for the genus Polymorphus. Polymerase chain reactions (PCRs) were conducted in a final volume of 25 μl and contained 18.9 μl distilled water, 2.5 μl 10× PCR buffer, 1.5 mM MgCl2, 0.2 mM of each dNTP, 0.5 μM of each primer, 0.5 units TopTaq Polymerase and 1 μl template DNA. The amplification conditions started with an initial denaturation at 94°C for 3 min followed by 35 cycles of denaturation at 94°C for 30 s, primer annealing at 48°C for 60 s and extension at 72°C for 60 s. The temperature was held at 72°C for 7 min to complete elongation. The PCR products were sequenced (both directions) by Microsynth (Balgach, Switzerland) using the PCR primers. Sequences were manually aligned and checked using the program BioEdit (Hall, 1999

Helminth community
The 60 hosts harboured a total of 504 intestinal worms. The majority (39 hosts) carried zero to five worms, only six were heavily parasitized with 21-116 worms; the dispersion pattern is given in fig. 2. All species identified during this survey are listed in table 1, together with the statistical evaluation of the parasite community. A total of nine identified species, two acanthocephalans, four trematodes, two cestodes and a nematode were recovered, as well as a nematode that could not be identified further. Overall, the dominant parasite species was the cestode Diorchis sp. Clerc, 1903 (P = 31.7%, mean intensity (MI) = 8.7) followed by the acanthocephalan P. minutus (P = 30.0%, MI = 6.3) and the trematode Notocotylus attenuatus (Rudolphi, 1809) Kossack, 1911 (P = 23.3, MI = 6.9). Mixed infections occurred as well with two or more species from one or more of the systematic taxa. The diversity characteristics of the helminth community are given in table 2.

Genetic analysis of Acanthocephala
Only seven out of 42 specimens included in the genetic analysis gave positive PCR products. This could have been due to suboptimal fixation methods for molecular genetic analysis (e.g. use of formalin), which could have inhibited PCR (e.g. Kruckenhauser & Haring, 2008;Zimmermann et al., 2008;Jaksch et al., 2016).
The calculated NJ tree revealed high bootstrap support for most of the nodes (fig. 3a). The specimens of this study occurred within two distinct clades: The basal split in the tree separated sequences of the species F. anatis from the remaining clades belonging to the family Polymorphidae ( fig. 3a). The sequences of F. anatis showed a mean sequence divergence of 0.1%.
The sequences of P. minutus investigated in the present study were found in a main clade comprising three subclades (PspT1, PspT2 and PspT3), showing the same topology as in Zittel et al. (2018). The sequences of this study clustered in subclade PspT3, together with sequences processed by Zittel et al. (2018) and one sequence obtained from GenBank.
We calculated a mean genetic distance of 1.6% (range 0.4-1.9%) within the subclade PspT3, and one of 9.7% between PspT3 and the other two subclades PspT1 and PspT2. The mean genetic distance between the subclade PspT3 and the closest related single sequence of P. obtusus was 11.5%, while the distances to all other clades and single sequences were between 21 and 32%. The MJ network of the subclade PspT3 showed a high intraspecific variation with a high Hd of 0.94 and a considerably low π of 0.008. The network consisted of 15 haplotypes, whereof three haplotypes were found among the specimens of this study (fig. 3b). One specimen (E59-6) was separated by seven mutation steps from the main haplo-group, including sequences processed by Zittel et al. (2018), mostly from Germany. Two samples (E59-7 and E11-1) were closely related to each other and separated by six mutation steps from this main haplo-group ( fig. 3b).

Helminth community
The observed aggregated distribution is in accordance with most publications on helminth parasites of field populations of birds (e.g. Goater & Holmes, 1997 and references therein). Comparing the diversity characteristics with the literature is not easy, as publications vary in the evaluation of community structures and infection rates to a high degree. Although data for comparison are scarce, we can say that diversity indices (Shannon-Wiener, Simpson) do not point to a high diversity, but the distribution of parasite species (evenness) has a value of 0.73, indicating a rather evenly distributed species diversity across the sampled specimens (the value of '1' would mean a completely even distribution). Also, the Berger Parker index is rather low, indicating a not too high proportional importance of the most abundant species. Very recently, Rzad et al. (2020) compared the community structure of trematodes in mallards from Poland and the Czech Republic and stated that a higher diversity was observed in Poland due to a smaller anthropogenic influence compared to that in the Czech Republic. That the Austrian region under investigation might be considered even more influenced and altered by humans, may be the reason for the depauperate helminth community observed in this study as compared to others. Of course, these alterations may not be influencing the occurrence of ducks directly, as they are well adapted to human presence (Bauer, 2005). As stated by the Austrian Government, 44% of surface waters are not in good conditions regarding their macro-zoobenthic community (BMNT, 2019). The massive restructuring and diverse use of all waterbodies in the region might have major influences on the occurrence and abundance of suitable intermediate hosts for avian parasites and influencing their community structure in this indirect way. Amongst those uses is, for example, the damming of many streams to produce electricity with over 1300 so-called 'small hydro-electric power stations' in the region (kleinwasserkraft, 2021) or the use of ponds for fish farming, where anti-parasitic drugs might be used to protect crowded fish populations. These fish farms currently consist of a network of 121 farmers using 13 km 2 of ponds in the region (Niederösterreichischer Teichwirteverband, 2021). An additional factor that reduces the biodiversity of intestinal helminths is chronic lead intoxication (Pruter et al., 2018). Lead levels in liver described by these authors are congruent with the levels in liver of mallards from Austria given by Plessl et al. (2017), who reported at least 3.9% of the mallards showing acute lead pollution.

Nematoda
Most striking is the occurrence of only two individual nematodes within the sample. This stands in contrast to most literature on the nematodes of mallards. Pav & Zajicek (1963), Birova & Macko (1984), Fedynich & Pence (1994), Kavetska et al. (2008), Yoshino et al. (2009) and Syrota et al. (2018) identified between four and nine species. Only De Jong (1976) reported negative results from mallards from the Netherlands. We identified Porrocaecum crassum (Deslongchamps, 1824), which has been recorded as occurring in low prevalences and low intensities in other studies (2% and 1-2 in Canada; 5% and 1-4 in eastern Slovakia, respectively) (Birova & Macko, 1984;Fedynich & Pence, 1994). Birova et al. (1992) also described a general seasonal fluctuation in nematode numbers with a rapid reduction by October and lowest numbers in November and December. In particular, for P. crassum they observed low intensities (1-4) over the whole year, with fluctuations in prevalence between the lowest, 6.1% in October and the highest 39.5% in February (Birova et al., 1990). Our samples were obtained in October and November, so the time of sampling might contribute to the paucity of nematode findings. A pronounced seasonality has also been described for Tetrameres fissispina (Diesing, 1861), which has been reported as the most prevalent nematode in mallards

Journal of Helminthology 5
in Eastern Slovakia (Birova et al., 1992) but was absent in our study. Interestingly enough, this species has been considered to be an incidental finding by other authors (Fenton et al., 2018), which might point to a non-regular appearance in general, most possible depending on the occurrence of suitable intermediate hosts, which are crustaceans. Amidostomum acutum (Lundahl, 1848), another well-known nematode of mallards (Birova & Macko, 1984;Fedynich & Pence, 1994), was absent as well. The parasite has a direct life cycle with eggs developing in the environment (Anderson, 1992), its presence does not depend on intermediate hosts and its absence in this study must remain unexplained and calls for further investigation.

Cestoda
The cestode Diorchis sp. was the dominant parasite species in our study, occurring in 31.7% of the mallards with intensities between 1 and 26 worms per host. We only recorded one other species of cestode: Fimbriarioides intermedia (Fuhrmann, 1913) was found in a single bird host. Our findings stand in complete contrast to most studies published, in which cestodes showed much higher diversities and very often higher infection intensities. Even in the aforementioned study from the Netherlands, in which no nematodes were discovered, a total of nine cestode species were identified (De Jong, 1976). Because cestodes have a life cycle which involves a crustacean (copepod, ostracod or amphipod) as intermediate host (Hiepe et al., 1985), it could be argued that since most crustaceans are much more sensitive to pollution than, for example, molluscs, the occurrence of cestodes might be reduced in Austrian waterbodies. Consequently, fewer cestode species might be found infecting Austrian mallards. But this would need further investigation.

Trematoda
The diversity in trematode species found in other studies is highly variable. We identified four species of trematodes, which is comparable to five intestinal species found by Pav & Zajicek (1960) in a relatively small area in the former Czechoslovakia. Fedynich & Pence (1994) reported five species from Texas and Chu et al. (1973) reported six species in wild ducks from Korea. A much higher species diversity of 12 from a larger study area including 13 regions in Bohemia and Moravia was reported by Pav & Zajicek (1963); this number was exceeded by Kavetska et al. (2008), who identified 16 trematode species from the northwestern part of Poland. Most remarkably, only one species, E. revolutum (Froelich, 1802) Looss, 1899 seems to be globally distributed and occurs in all the aforementioned reports as well as from Bangladesh (Yousuf et al., 2009) and Mexico (Farias & Canaris, 1986). Echinostoma revolutum uses lymnaeid snails as the first intermediate host and metacercariae develop in the second intermediate host, another snail, bivalve or amphibian. Definitive hosts include mammals and birds, which become infected by ingesting infected snails or amphibians (Hiepe et al., 1985). Displaying so little host specificity indicates that the parasite may not have highly specific requirements to complete its life cycle and can, therefore, become prevalent in aquatic systems worldwide. On the one hand, however, species differentiation within the family Lymnaeidae is still under investigation (Vinarski, 2013), and on the other hand, the systematic status of all worms identified as E. revolutum is also under discussion. Echinostoma revolutum could well represent a complex of closely related taxa, which are morphologically indistinguishable (Sitko et al., 2006). In more recent publications authors have, therefore, used the term 'E. revolutum group' for these worms (Syrota et al., 2018). Further molecular investigations might identify a higher number of echinostome species, presently subsumed under E. revolutum. The prevalence of E. revolutum in our study was 6.7% and quite low compared to the 15% found by Fedynich & Pence (1994) or the 53% by Pav & Zajicek (1960). Intensity of infection of 1-19 was comparable with the two studies mentioned. The most prevalent trematode in our study was N. attenuatus, which is also distributed worldwide and was found in all the aforementioned studies except in the one from Texas (Fedynich & Pence, 1994). Notocotylus attenuatus has an abbreviated life cycle, with no second intermediate host involved, metacercariae develop from cercariae that emerge when their aquatic lymnaeid intermediate hosts attach to solid surfaces, such as mollusc shells or water plants (Harper, 1929). Host specificity, neither for the intermediate nor the definitive host seems to be well developed and, therefore, the widespread occurrence of the parasites is not surprising. Omnivorous bird species, such as the mallard, can become infected easily by ingesting infested food items such as molluscs, crustaceans or water plants. Prevalence of infection (23.3%) in our studies is much lower compared to that of other studies (40% by Pav & Zajicek (1960) and 60% by Birova & Macko (1984)). Apatemon gracilis uses fish as its second intermediate host and, therefore, our finding of A. gracilis in mallards is worth mentioning. Although the infection was of low prevalence (3.3%) and exhibited low intensities (3-8), this finding proves that mallards are piscivorous. Compared to other studies, the numbers in our study are low. Infection occurred with high prevalence (42%) and MI of 17.5 in the study from Texas (Fedynich & Pence, 1994), and P = 43% and MI = 10.3 in the study by Pav & Zajicek (1960). Birova & Macko (1984) observed that the parasite vanished after the restoration of the wetlands resulted in a richer fauna of snails and crustaceans, which may have led to a shift in diet of mallards towards prey easier to catch than fish.

Acanthocephala
The acanthocephalan species recovered in this study accord well with other studies from this host. Specimens of the genus Polymorphusnamely, P. minutus and its junior synonym P. magnus (the status of P. magnus is discussed in the introduction) and F. anatis are the two taxa recorded from Europe (Pav & Zajicek, 1963;Macko, 1979;Birova & Macko, 1984;Kavetska et al., 2008;Crompton & Harrison, 2009;Syrota et al., 2018). One other acanthocephalan genus, Corynosoma, was recorded in mallards from North America, the species being Corynosoma constrictum Van Cleave, 1918(Van Cleave & Starrett, 1940Fedynich & Pence, 1994). Prevalence in our study for P. minutus (30%) was relatively high compared with earlier studies from central Europe. Macko (1979) and Birova et al. (1990) found only 1.5% and 2.2% prevalence, respectively, of the host population infected in Slovakia. By contrast, Crompton & Harrison (2009) found 50% of the mallards infected in Kent, UK. The prevalence of F. anatis from our work (5%) compares well to other publications. Infection intensities for acanthocephalans are reported to be highly variable and also subject to seasonal fluctuations, similar to those found in nematodes. Highest intensities occur in summer, most probably due to the period of maximal reproduction of their gammarid intermediate hosts (Birova et al., 1990;Crompton & Harrison, 2009). Ecosystem health seems to be an important factor for the occurrence of acanthocephalans: Birova & Macko (1984) observed a significant rise in F. anatis infections Journal of Helminthology 7 (prevalence as well as intensity) in mallards from the East Slovakian lowlands after water system regulations had been implemented. They argued that these measures changed many seasonal water bodies into permanent ones, therefore allowing continuous development of intermediate hosts and making the mallards more exposed to possible infections.

Genetic results
Even though no reference sequences of F. anatis were available, the identity of F. anatis was unequivocal, due to its distinct morphological traits. The identity of P. minutus was more difficult to establish because of the taxonomic confusion discussed above. The specimens of this study clustered in a subclade (PspT3), already defined by Zittel et al. (2018). These authors suggested that cystacanths of P. cf. minutus comprising different lineages or cryptic species used different intermediate hosts.
In this study, we also detected considerable variation in some characters, particularly proboscis armature and neck length, in individuals from several of the Austrian mallards. Grabner et al. (2020) found similar variations in the proboscis armature of the three genetically distinguished subclades (PspT1, PspT2 and PspT3). We also calculated a lower number of hook rows (12-14; table 3) in the genetically investigated specimens of Austria, while previous descriptions of P. minutus referred to 14-18, usually 16, hook rows (e.g. Lühe, 1911;Amin, 1992). Due to the morphological variations measured in this study, it can be expected that more lineages of P. cf. minutus may occur in Austria. Unfortunately, it was not possible to elucidate whether these morphologically examined specimens belonged to one of the other two subclades (PspT1, PspT2), or represented a further genetic lineage, since these specimens did not yield PCR products in the COI amplification, perhaps due to bad preservation. Polymorphus minutus has, however, been previously described as exhibiting a wide range of variability within and between different host species across a broad geographical range (Van Cleave & Starrett, 1940;Zittel et al., 2018;Grabner et al., 2020) as can be seen in table 3. However, the genetic results further implicated the specimens from Austria as belonging to the subclade PspT3, which uses G. pulex and G. roeselii as intermediate hosts (Zittel et al., 2018;Grabner et al., 2020). According to Grabner et al. (2020), it can be expected that the subclade was introduced by the invasive gammarid G. roeselii. Those authors suggested the invasion from south-east Europe to Germany, which would include a passage through Austria, where G. roeselii is a prevalent species in rivers (Poeckl et al., 2003). Further investigations regarding the species status of P. minutus are necessary, since there is a high probability that the three lineages known to date represent separate species.
Acknowledgements. We cordially thank Anja Joachim from the University of Veterinary Medicine, Vienna, for providing lab space for dissection and space in deep-freeze chambers. Michael J. Mühlegger is thanked for the dissection of the birds. Michaela Hejl and Anton Legin are thanked for the translation of Slovakian and Russian literature, respectively. We thank Elisabeth Haring for valuable discussion of genetic data. In addition, the cooperation with the confederations of hunters from Lower Austria and Upper Austria, mediated by Alois Gansterer, is highly acknowledged.
Financial support. Open access funding was provided by the University of Vienna.

Conflicts of interest. None.
Ethical standards. The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional guides on the care and use of laboratory animals.