Introduction
In this article, we provide an update on the fascinating problem of the hymenolepidid cestode species often known as Hymenolepis murissylvatici. This occurs quite frequently in rodents in Europe but has also been associated with the surprising additional ability to infect birds. As the nomenclature for this and some of the relevant related species arises from descriptions in the 18th and 19th centuries, and moreover, these forms are morphologically very similar, a considerable amount of uncertainty about their taxonomy and host-specificity has been propagated through the literature. Here, bringing modern molecular database records together with information from previous studies pertaining to relevant forms in both rodents and birds, we are able to throw further useful light on the nomenclature, host-specificity and general biology of these tapeworms.
Background
The 18th–early 20th centuries were a time when many different parasitic organisms were being described for the first time. Cox (Reference Cox2017) considered the years spanning 1875–1925 to be highly productive, referring to the period as the ‘golden age of parasitology’, during which not only were many new species described, but complex life cycles were also worked out for the first time (Foster Reference Foster1965; Marti Reference Marti2019). In those early years, the discipline required to avoid confusion in reporting new species had not yet been totally refined. Workers did not have access to the tools of the current age for rapidly reporting and disseminating their findings nor for learning easily about new species already described by others and published in learned society journals (Foster Reference Foster1965). Much of the systematic literature was based on proceedings of learned societies, university memoires, and museums, and published in their journals, some of which are not easily accessible even today. The practice of naming newly discovered parasites after their hosts is still common today (Poulin et al. Reference Poulin, McDougall and Presswell2022), but historically the concept that parasitic organisms are mostly host specific or have narrow host specificity resulted in organisms that looked morphologically similar under microscopical examination being given different names according to the hosts from which they had been isolated. This confusion is no better illustrated than in cestode nomenclature, which at the time was clogged with names of poorly described species, as emphasised by Meggitt and Subramanian (Reference Meggitt and Subramanian1927) in the introduction to the section of their paper listing what they considered to be invalid species (page 223). Similarly, Wardle and McLeod (Reference Wardle and McLeod1952) and Czaplinski and Vaucher (Reference Czaplinski, Vaucher, Khalil, Jones and Bray1994) provided, in some cases, long lists of synonymised genera and species, nomina nuda, nomina dubia, species (or genera) incertae sedis and species (or genera) inquirendae. Re-descriptions of species, based on new isolates and benefiting from contemporary improved laboratory methodologies such as electron microscopy, molecular phylogenetics and statistical methodologies (Brooks and Hoberg Reference Brooks and Hoberg2001; Poulin and Presswell Reference Poulin and Presswell2016), as well as reviews updating the taxonomy of helminth orders and families have often continued to include long lists of synonyms (e.g. Dimitrova et al. Reference Dimitrova, Vasileva, Mariaux and Georgiev2023; Erhardová Reference Erhardová1958; Tenora Reference Tenora1967; Verster Reference Verster1969; Wardle and McLeod Reference Wardle and McLeod1952).
Even against this background of widespread and on-going taxonomic refinement, some species have been subject to more taxonomic instability than others. Our focus here, the rodent-infecting tapeworm Hymenolepis murissylvatici, which has been variously known as Variolepis crenata and Microsomacanthus crenata (sometimes improperly spelled as Hymenolepis muris-sylvatici or H. muris sylvatici; International Commission on Zoological Nomenclature, 1999) certainly falls into this category. It has undergone a long, chequered taxonomic history that, as noted above, includes uncertainty about whether it is a rodent specialist, or instead a remarkable generalist, able to infect both rodents and birds. Here, we first review the history of nomenclature changes of this cestode and then, based on available host records, morphometric details and recent molecular information, propose a new taxonomic combination for the species. For the time being, the proposed taxon encompasses only rodent-infecting forms, whilst recognizing that very closely related, perhaps even conspecific, tapeworms occur in birds, either as generalist parasites infecting both rodents and birds or as specialists that relatively recently evolved differential specificity to these host groups. With regard to the general biology of the rodent-infecting species, we refer to infection experiments in laboratory mice, consider studies reporting aspects of its epidemiology in wild rodents, identify key gaps in our knowledge, and suggest avenues for further exploration.
History of nomenclature changes
From the perspective of parasitologists interested in the helminth communities of wild rodents in Europe, until recently this cestode was known as Hymenolepis murissylvatici (or H. muris-sylvatici, or H. muris sylvatici). In fact, historically it was first described by Rudolphi (Reference Rudolphi1819; on page 170) as Taenia muris sylvatici from Muris sylvatici (=Apodemus sylvaticus). Given the brief description, however, this was not accepted by all in the field, as for example Meggitt and Subramanian (Reference Meggitt and Subramanian1927), who included T. muris-sylvatici in their section on invalid species (page 224). These authors also referred to Janicki (Reference Janicki1906), who likewise considered this species to be dubious (page 581 in Janicki (Reference Janicki1906), referred to as T. muris sylvatici).
Nevertheless, T. muris-sylvatici was subsequently recognized as a hymenolepidid by Baer (Reference Baer1930), who examined worms from wood mice trapped near Geneva. He reported the 10 characteristic scolex hooks of this species and renamed it Hymenolepis muris-sylvatici. Joyeaux and Baer (Reference Joyeux and Baer1936) considered Hymenolepis muris-sylvatici to be a valid name for the parasite from A. sylvaticus. This version of the name, as well as its variants (i.e. H. muris sylvatici, H. murissylvatici) was subsequently cited in studies on helminth communities of rodents from Europe, as well as in species lists (Baylis Reference Baylis1939; Lewis and Twigg Reference Lewis and Twigg1972; Murúa Reference Murúa1978). Perhaps surprisingly, Skrjabin and Mathevosjan (1948) referred to Dicranotaenia muris-sylvatici and, over a decade later, Lewis (Reference Lewis1987) referred to Staphylocystis (Hymenolepis) murissylvatici, presumably following Spasskii (Reference Spasskii1950), who earlier had moved the species to the genus Staphylocystis Villot, 1877. Nevertheless, this latter change in generic status was not adopted widely, although Yamaguti (Reference Yamaguti1959) listed S. murissylvatici (Rud., 1819) Spassky, 1950 in his magnum opus.
Even earlier than Rudolphi (Reference Rudolphi1819), Goeze (Reference Goeze1782) had named an avian cestode, Taenia crenata, which he had recovered from a greater-spotted woodpecker (Dendrocopos major), and several authors, including Kostylev (Reference Kostylev1915), thereafter referred to the cestode by this name. However, Spasskii and Spasskaya (Reference Spasskii and Spasskaya1954) erected the genus Passerilepis and moved the species to this genus as P. crenata, and subsequently, this name was adopted by some authors (e.g. Galkin Reference Galkin1979a; Kornakova Reference Kornakova1999; Sultanov and Spasskaya Reference Sultanov and Spasskaya1959). The genus Variolepis was also erected by Spasskii and Spasskaya (Reference Spasskii and Spasskaya1954). A decade later, both genera were reviewed and reorganised, and P. crenata was moved to Variolepis (Spasskii and Spasskaya, Reference Spasskii and Spasskaya1964), but still referred to as P. crenata by Spasskii (Reference Spasskii1965). Variolepis crenata is considered to be an avian generalist, with a large host range, especially among passerine species and has been known by a variety of other names (see Prokopič (Reference Prokopič1967) for a long list of synonyms of V. crenata).
In Reference Spasskii and Spasskaya1960, Spasskii and Spasskaya drew attention to the similarity of Baer’s (1930) description of T. muris-sylvatici with V. crenata, and this initiated a debate that is still ongoing about whether these two really are the same species, capable of parasitizing both birds and rodents. The first substantial comparison was by Prokopič (Reference Prokopič1967), who conducted a meticulous study comparing the morphometrics of worms isolated from wood mice in Czechoslovakia with those from blackbirds and starlings and came to the conclusion that isolates from both mice and birds were indistinguishable microscopically and therefore highly likely to be the same species. Importantly, Prokopič (Reference Prokopič1967) did not include isolates from woodpeckers in his study.
The situation was further complicated by yet more revisions of the nomenclature among the Hymenolepididae. In their major review of the family Hymenolepididae, Czaplinski and Vaucher (Reference Czaplinski, Vaucher, Khalil, Jones and Bray1994) considered Variolepis Spasskii and Spasskaya, Reference Spasskii and Spasskaya1954 to be a junior synonym of Wardium Mayhew, 1954, and Passerilepis Spasskii and Spasskaya, Reference Spasskii and Spasskaya1954 to be a junior synonym of Microsomacanthus López-Neyra, Reference López-Neyra1942. Hanzelová and Ryšavý (Reference Hanzelová and Ryšavý1996) acknowledged this reorganisation of the Hymenolepididae. Czaplinski and Vaucher (Reference Czaplinski, Vaucher, Khalil, Jones and Bray1994) proposed that H. murissylvatici be best placed in Microsomacanthus. Although Tenora (Reference Tenora2004) first listed the Apodemus spp. infective species as V. crenata (Goeze Reference Goeze1782) on the basis of Prokopič (1970a, b), he then recognised the original name, H. muris-sylvatici (Rudolphi Reference Rudolphi1819), and citing Czaplinski and Vaucher (Reference Czaplinski, Vaucher, Khalil, Jones and Bray1994), concurred with these latter authors that it probably belongs to the genus Microsomacanthus Lopez-Neyra, 1962. Based on Czaplinski and Vaucher (Reference Czaplinski, Vaucher, Khalil, Jones and Bray1994), M. crenata was adopted by Behnke et al. (Reference Behnke, Lewis, Mohd Zain and Gilbert1999), Dwużnik et al. (Reference Dwużnik, Gortat, Behnke, Gryczyńska, Bednarska, Mikoszewski, Kozakiewicz and Bajer2017), and was also referred to in Lewis et al. (Reference Lewis, Morley and Behnke2023) in their studies of the helminth fauna of wood mice in Southern England. Klimpel et al (Reference Klimpel, Förster and Schmahl2007a), on the other hand, utilised yet another combination, listing Microsomacanthus murissylvatici as infecting A. flavicollis in Germany.
However, Czaplinski and Vaucher’s proposals that Passerilepis is a junior synonym of Microsomacanthus, that Variolepis is a junior synonym of Wardium, and that H. murissylvatici be best placed in Microsomacanthus were controversial. Indeed, although these authors stated that, according to their key, H. murissylvatici should be classified in Microsomacanthus, seemingly not totally convinced, they also proposed further scrutiny of the species ‘to solve this problem ’. In their key to the Hymenolepididae, Czaplinski and Vaucher (Reference Czaplinski, Vaucher, Khalil, Jones and Bray1994) specified rostellar hook shape as a defining character separating Microsomacanthus and Wardium among the Hymenolepididae. The length and shape of rostellar hooks have been frequently used to distinguish between hymenolepidid species, notably in the case of Passerilepis spp. (Dimitrova et al. Reference Dimitrova, Mariaux, Georgiev and Vasileva2019; Spasskii and Spasskaya Reference Spasskii and Spasskaya1954, Reference Spasskii and Spasskaya1964). Inclusion in Microsomacanthus, is based on the presence of ten diorchoid hooks (Galkin et al Reference Galkin, Regel and Mariaux2006, Reference Galkin, Mariaux, Regel and Skirnisson2008; hook handle longer than well-developed blade and guard shorter than blade (Khalil et al Reference Khalil, Jones and Bray1994)), whereas those of H. murissylvatici are clearly fraternoid (Figure 1c; hook handle longer than guard and blade, guard as long or longer than blade and thicker than blade (Khalil et al Reference Khalil, Jones and Bray1994); but see below for more on this). Neither would it be appropriate to move H. murissylvatici to Wardium, for which inclusion is dependent on aploparaksoid and echinorhynchoid hooks. In the avian cestode literature, Czaplinski and Vaucher’s (Reference Czaplinski, Vaucher, Khalil, Jones and Bray1994) proposals were not accepted by all (Dimitrova et al. Reference Dimitrova, Mariaux, Georgiev and Vasileva2019; Mariaux et al. Reference Mariaux, Tkach, Vasileva, Waeschenbach, Beveridge, Dimitrova, Haukisalmi, Greiman, DTJ, Makarikov, Philips, Razafiarisolo, Widmer, Georgiev, Caira and Jensen2017, Reference Mariaux and Georgiev2020), and even Spasskii and Spasskaya’s (Reference Spasskii and Spasskaya1964) earlier proposal for moving the cestode from Passerilepis to Variolepis was ignored by some. Referring to P. crenata, Haukisalmi (Reference Haukisalmi2015) gave carrion crow (Corvus corone) and thrushes (Turdus spp.) as hosts in Finland, while Hanzelová and Ryšavý (Reference Hanzelová and Ryšavý1996), also referring to P. crenata, included additionally starlings, house sparrows and jays as hosts of this species in Slovakia. Both authors also listed many Microsomacanthus spp., all parasites of avian hosts. Neither study listed rodents as potential hosts, nor recognised H. murissylvatici as a synonym of this species.
Variolepis murissylvatici. Scolex with rostellum inverted (A) and with rostellum partially everted (B), both from worms isolated from wood mice trapped in S. Nottinghamshire, England. C, The 10 fraternoid hooks from the scolex in a compressed preparation of a worm isolated from a wood mouse trapped in N. Norfolk, England. D, The cysticercoid stage removed from Tribolium confusum (Coleoptera) that had been infected with eggs obtained from a worm isolated in S. Nottinghamshire.
Figure 1. Long description
A four-panel micrograph series labeled A through D.
Panel A: A light micrograph of the scolex with an inverted rostellum. A black arrow points to the small, dark, recessed rostellar hooks at the distal tip. Two large circular suckers are visible on the sides. A scale bar at the bottom right indicates 100 micrometers.
Panel B: A light micrograph of the scolex with a partially everted rostellum, which appears as a protruding cone-like structure. A black arrow points to the ring of hooks at the apex of this protrusion. Two suckers are clearly defined below the rostellum. A scale bar at the bottom right indicates 100 micrometers.
Panel C: A high-magnification micrograph showing 10 fraternoid hooks arranged in a circular pattern. Each hook has a distinct curved blade and a bifurcated base. A scale bar at the bottom right indicates 10 micrometers.
Panel D: A light micrograph of a cysticercoid stage. The organism is oval-shaped with a granular internal texture. A black arrow points to the invaginated scolex and hooks located within the central cavity of the cyst. A scale bar at the bottom right indicates 50 micrometers.
The status of H. murissylvatici was yet further complicated when Vaucher et al (Reference Vaucher, Santalla, Casanova, Feliu, Combes and Jourdane2003), following up Czaplinski and Vaucher’s (Reference Czaplinski, Vaucher, Khalil, Jones and Bray1994) recommendation that H. murissylvatici and V. crenata should be compared by isoenzyme electrophoresis, found that none of the eleven enzyme systems studied shared the same electrophoretic profile. Based on the outcome of this study, H. murissylvatici from wood mice and V. crenata from greater-spotted woodpeckers cannot be the same species. Vaucher et al’s (Reference Vaucher, Santalla, Casanova, Feliu, Combes and Jourdane2003) isolates of V. crenata were from the same host, greater-spotted woodpecker, as those obtained originally by Goeze (Reference Goeze1782) in his original description of this species. It may also be relevant that Prokopič (Reference Prokopič1967) had obtained worms from blackbirds and starlings, avian hosts that are known to support another morphologically very similar cestode M. serpentulus (Schrank Reference Schrank1788), which is also known as H. serpentulus. Indeed, Vaucher et al (Reference Vaucher, Santalla, Casanova, Feliu, Combes and Jourdane2003) found that H. murissylvatici and M. serpentulus were very similar with respect to hook shape and size, internal anatomy, and egg structure. For this reason, these authors rejected the idea that V. crenata and M. serpentulus were synonymous species, as had been concluded by Prokopič (Reference Prokopič1967) (which is designated as H. serpentulus in this paper). However, they made no commitment to H. murissylvatici and M. serpentulus being synonymous, because they lacked sufficient material to enable a comparison of their isoenzymes.
From the chronology of publications, it is clearly evident that not all workers were aware of all relevant publications, and/or accepted that the successive changes in nomenclature were appropriate. Even in the current century, and three decades after Spasskii and Spasskaya (Reference Spasskii and Spasskaya1964) had moved the cestode to Variolepis, Rosso et al. (Reference Rosso, Scalet, Ferrari, Manfredi, Rizzolia and Cerbo2005) referred to P. crenata in their study of the helminth fauna of A. flavicollis and C. glareolus from NE Italy, Marinova (Reference Marinova, Georgiev, Bechev and Yancheva2020) listed P. crenata (Goeze, Reference Goeze1782) Sultanov and Spasskaya, Reference Sultanov and Spasskaya1959 as a parasite of birds in Bulgaria and Dimitrova et al. (Reference Dimitrova, Mariaux, Georgiev and Vasileva2019) also referring to P. crenata (Goeze, Reference Goeze1782) Sultanov and Spasskaya, Reference Sultanov and Spasskaya1959 emphasized that the species was one of the most frequently recorded Passerilepis spp. worldwide and included the species in their list of cestodes of African birds (Dimitrova et al Reference Dimitrova, Vasileva, Mariaux and Georgiev2023). Mariaux (Reference Mariaux2024) listed H. murissylvatici as a parasite of Apodemus spp. in Switzerland. The Global Cestode Database (https://tapewormdb.uconn.edu/index.php/parasites/species_details/7764; accessed September 2025) still lists the species as Passerilepis crenata (Goeze, Reference Goeze1782) Sultanov and Spasskaya, Reference Sultanov and Spasskaya1959 and records hosts only as birds.
Morphology of H. murissylvatici
A distinguishing feature of H. murissylvatici from European rodents is the crown of 10 long (22–25 μm) rostellar hooks of fraternoid type (Table 1, Figure 1). Most other cestodes frequently encountered in Apodemus spp. and Microtus spp. (voles) in Western Europe possess no rostellar hooks (e.g. Hymenolepis diminuta, H. hibernia) or have many more and smaller hooks (Rodentolepis nana fraterna (20–30 and 14–18 μm), R. microstoma [23–30 and 11–18 μm], R. straminea (19–36 and 12–14 μm)) (Casanova et al. Reference Casanova, Santalla, Durant, Vaucher, Feliu and Renaud2001; Haukisalmi et al. Reference Haukisalmi, Hardman, Foronda, Feliu, Laakkonen, Niemimaa, Lehtonen and Henttonen2010; Rushworth et al. Reference Rushworth, Boufana, Hall, Brannan, Mastin, Birtles, Craig and Rogan2016; Smyth Reference Smyth1976; Wardle and McLeod Reference Wardle and McLeod1952). An exception to this is Kontrimavichusia asymmetrica (Janicki, 1904), previously R. asymmetrica, which has hooks in the range 18–23 μm (Makarikov and Binkienė 2022), although earlier authors recorded shorter rostellar hooks (12–18 μm in Santalla et al. Reference Santalla, Casanova, Durand, Vaucher, Renaud and Feliu2002) and has over twice as many hooks as H. murissylvatici (n=21–22 in Santalla et al. Reference Santalla, Casanova, Durand, Vaucher, Renaud and Feliu2002 and 18–23 in Makarikov and Binkienė, 2022).
Records of the morphometric values for key characters of V. murissylvatici recovered from rodents. All the authors listed below record the presence of 10 rostellar hooks
Table 1. Long description
The table consists of eight columns: Authors, Host, Mean hook length, Range hook length, Mean egg size, Range egg size, Cirrus pouch size, and Oncospheral hook length.
* Baer, 1931: Host A. sylvaticus; Mean hook length 23; Cirrus pouch size 100–130. Other values not given.
* Rybicka, 1959: Host A. flavicollis; Range hook length 23–25; Cirrus pouch size 160–175. Other values not given.
* Schmidt, 1961: Hosts A. sylvaticus, A. flavicollis and C. glareolus; Range hook length 22.6–23.2; Range egg size 94–101 times 72–75; Oncospheral hook length 18–19.5.
* Prokopič, 1967: Host A. sylvaticus; Range hook length 22–24; Range egg size 82 times 105; Cirrus pouch size 200 times 60.
* Vaucher and Hunkeler, 1967: Hosts A. sylvaticus and A. flavicollis; Mean hook length 22.5; Cirrus pouch size 150–170.
* Tenora and Murai, 1972: Host A. flavicollis; Range hook length 23–25; Range egg size 85–90; Cirrus pouch size 140–168 times 45–50; Oncospheral hook length 18–22.
* Vaucher et al., 2003: Host A. flavicollis; Range hook length 22–25.5; Mean egg size 122 times 111; Range egg size 104–150 times 91–140; Cirrus pouch size 117–228 times 49–78; Oncospheral hook length 20–22.
* Sasaki et al., 2021: Host A. speciosus; Range hook length 22–25. Other values not given.
Note: n g indicates data not given in the cited paper.
a ng, not given in cited paper.
Like many hymenolepidid cestodes, the craspedote proglottides of H. murissylvatici have three testes. Two testes lie anteriorly and posteriorly on the aporal side and one testis is on the poral side (Baer 1930; Figure 2A), although there is some variability in this arrangement as described by Rybicka (Reference Rybicka1959). The ovary is located centrally in the proglottides and may be lobate (Rybicka Reference Rybicka1959; Vaucher and Hunkeler Reference Vaucher and Hunkeler1967; Vaucher et al Reference Vaucher, Santalla, Casanova, Feliu, Combes and Jourdane2003). The vitellarium lies posteriorly to the ovary (Baer 1930; Rybicka Reference Rybicka1959; Vaucher and Hunkeler Reference Vaucher and Hunkeler1967; Vaucher et al Reference Vaucher, Santalla, Casanova, Feliu, Combes and Jourdane2003). A full account of the morphometrics of H. murissylvatici is given in Prokopič (Reference Prokopič1967), in which he also compares measurements obtained by himself from both wood mice and birds with those obtained earlier by Kostylev (Reference Kostylev1915), Baer (1930), Yamaguti (Reference Yamaguti1935), Joyeaux and Baer (Reference Joyeux and Baer1936), Skrjabin and Matevosjan (Reference Skrjabin and Matevosjan1948), and Ryšavý (Reference Ryšavý1957). Useful morphometric data have also been supplied by Rybicka (Reference Rybicka1959) and an excellent detailed description is given in Vaucher et al (Reference Vaucher, Santalla, Casanova, Feliu, Combes and Jourdane2003).
Variolepis murissylvatici. Developing craspedote proglottides (A) and egg (B), both from a worm isolated from wood mice trapped in Nottinghamshire. Abbreviations are as follows: T1, testis on the poral side; T2, two testes on the aporal side; Ov, ovary; Go, genital openings and ducts leading internally, including the vagina and cirrus sac with the cirrus; Oh, oncospheric hooks.
Figure 2B shows a photograph of an egg of H. murissylvatici from a worm isolated from a wood mouse in Nottinghamshire. The literature records some variability in the length of eggs (Table 1). Kostylev (Reference Kostylev1915), Schmidt (Reference Schmidt1961), Propkopič (Reference Prokopič1967), and Vaucher et al (Reference Vaucher, Santalla, Casanova, Feliu, Combes and Jourdane2003) all give the egg length as exceeding 80 μm (82.5, 94–101, 105, and 104–150 μm, respectively). Figure 2B shows the six oncospheric hooks (see Table 1 for their length), orientated in parallel, with traces of very thin polar filaments just visible, as observed also by Vaucher et al (Reference Vaucher, Santalla, Casanova, Feliu, Combes and Jourdane2003).
Genetic identity of H. murissylvatici and proposal for Variolepis murissylvatici comb. nov.
Currently, two genetic sequences in GenBank (accessed June 2026; GenBank accession nos. GU166246 and GU16247; 1365 and 1533 bp, respectively, partial sequences of 28S ribosomal RNA) are both assigned to M. crenatus rather than M. crenata. These are respectively derived from wood mice and yellow-necked mice (Apodemus flavicollis) from Croatia and attributed to Haukisalmi et al. (Reference Haukisalmi, Hardman, Foronda, Feliu, Laakkonen, Niemimaa, Lehtonen and Henttonen2010), who in the paper referred to them as ‘Hymenolepis’ muris-sylvatici. In this study, 28S rRNA sequences from 32 species of hymenolepidid cestodes were used to elucidate systematic relationships between these isolates and to construct a phylogenetic tree for the group. Whilst clearly a hymenolepidid, ‘H’.muris-sylvatici (M. crenatus in GenBank) represented a singular lineage within the family that did not cluster with any of the four diversified clades circumscribed by the Haukisalmi et al. (Reference Haukisalmi, Hardman, Foronda, Feliu, Laakkonen, Niemimaa, Lehtonen and Henttonen2010) analysis (Clade A, Ditestolepis; Clade B, Hymenolepis; Clade C, Arostrilepis; and Clade D, Rodentolepis). The 28S rDNA sequence from ‘H’. muris-sylvatici (M. crenatus in GenBank) was closest to a muskrat parasite, R. evaginata, that also fell outside any of the A-D clades.
A 28S rDNA sequence (LC535242) from a Microsomacanthus sp. isolate from A. speciosus from Japan showed 99.6% similarity to sequence GU166246 (Haukisalmi et al. Reference Haukisalmi, Hardman, Foronda, Feliu, Laakkonen, Niemimaa, Lehtonen and Henttonen2010), indicating that ‘H’. muris-sylvatici (M. crenatus in GenBank and in Sasaki et al., Reference Sasaki, Anders and Nakao2021) also occurs in Japan (Sasaki et al. Reference Sasaki, Anders and Nakao2021). The latter authors concluded that it was highly probable that H. murissylvatici (=M. murissylvatici) and M. crenata were the same species, but then retained the possibility that isolates from rodents may represent a different species from those from birds and urged more molecular work on isolates from birds and rodents to clarify the relationship. These authors were apparently unaware that Passerilepis had been synonymised with Microsomacanthus (see above; Czaplinski and Voucher 1994) and that 28S rDNA sequences from avian Microsomacanthus (=Passerilepis spp. in Wider et al., 2013) were available (JQ950690-JQ950693), including two sequences from Microsomacanthus passeris isolated from Pycnonotus leucotis (white-eared bulbul) and Neolestes torquatus (black-collared bulbul). Makarikov and Binkienė (2022) included these sequences in a phylogenetic tree of the hymenolepidids, showing the close relationship of avian Microsomacanthus to R. evaginata but did not include other isolates from rodents in their work.
Exploiting these 28S rDNA sequences from GenBank and those of related hymenolepidids, we reanalysed their relationships as shown in Figure 3. Percentage similarities between the key species on which this review is focused are shown in Table 2. The phylogenetic tree in Figure 3 clearly shows the same four clades identified by Haukisalmi et al. (Reference Haukisalmi, Hardman, Foronda, Feliu, Laakkonen, Niemimaa, Lehtonen and Henttonen2010): Clade A, Ditestolepis; Clade B, Hymenolepis; Clade C, Arostrilepis; Clade D, Rodentolepis. However, whereas these authors were unable to place ‘H’. muris-sylvatici into a distinct clade, our analysis reveals that the H. murissylvatici isolates from rodents (M. crenatus in Figure 3 and Table 2 and H. murissylvatici in Haukisalmi et al., Reference Haukisalmi, Hardman, Foronda, Feliu, Laakkonen, Niemimaa, Lehtonen and Henttonen2010) show close similarity to the bird-infecting V. farciminosa (96.99–97.35% homology), with which they form a tight clade. Unfortunately, GenBank does not currently have a 28S rDNA sequence from a V. crenata isolated from an avian host. All currently available avian isolates of Microsomacanthus (e.g. M. passeris) form another distinct branch leading to a subclade that does not include V. crenata (M. crenatus in Figure 3) from rodent hosts.
Tree of phylogenetic relationships among the Hymenolepididae infecting rodents in the Palearctic, based on 28S rDNA.
Phylogenetic relationships among the 28S rDNA gene sequences, representing various cestode species, were assessed using Bayesian Inference, implemented in MrBayes v3.2.6 (Ronquist et al Reference Ronquist, Teslenko, van der Mark, Ayres, Darling, Höhna, Larget, Liu, Suchard and Huelsenbeck2012). Due to the large size of the dataset, we fitted a GTR + G model with all six rate parameters free and with variation in the rate of evolution among sites in the alignment. Two independent runs of 10,000,000 generations were sampled every 1,000 generations and 25% of the initial samples were discarded as a burn-in phase. The convergence of two runs and effective sample sizes for the model’s parameters were checked using Tracer v. 1.7 (Rambaut et al Reference Rambaut, Drummond, Xie, Baele and Suchard2018). This tree was constructed entirely from sequences deposited in GenBank (n = 46).
We have highlighted in coloured boxes the four clades identified by Haukisalmi et al (Reference Haukisalmi, Hardman, Foronda, Feliu, Laakkonen, Niemimaa, Lehtonen and Henttonen2010), with which our analysis is largely concordant. The M. crenatus (= H. murissylvatici in Haukisalmi et al., Reference Haukisalmi, Hardman, Foronda, Feliu, Laakkonen, Niemimaa, Lehtonen and Henttonen2010) isolates and Hymenolepididae sp. are highlighted in bold red and the Microsomacanthus spp. from birds are in bold blue.
Note that:
-
• Our analysis shows Haukisalmi et al.’s Clades B (Hymenolepis) and C (Arostrilepis) as subclades of a larger clade that also incorporates the Microsomacanthus spp.
-
• Haukisalmi et al (Reference Haukisalmi, Hardman, Foronda, Feliu, Laakkonen, Niemimaa, Lehtonen and Henttonen2010) included Rodentolepis nana and R. fraterna in Clade D (Rodentolepis), whereas our analysis indicates a deep-rooted division in this clade, and we have excluded these species from Haukisalmi et al.’s Clade D.
-
• Rodentolepis asymmetrica has been redescribed recently and moved to a newly erected genus Kontrimavichusia Makarikov and Binkienė 2022.
-
• Sequences in Clade D, registered as Hymenolepis microstoma in GenBank are more appropriately now known as Rodentolepis microstoma (Czaplinski and Vaucher Reference Czaplinski, Vaucher, Khalil, Jones and Bray1994), although not all have accepted this change in the generic status of the species (Cunningham and Olson Reference Cunningham and Olson2010).
Figure 3. Long description
A Bayesian Inference phylogenetic tree based on 28S rDNA gene sequences. The tree branches from a root on the left toward 46 terminal nodes on the right.
* Clade C (Red Box, Top): Contains Arostrilepis species from various rodents in Mongolia, Russia, U S A, China, Canada, and Finland. A rodent silhouette is positioned to the right.
* Clade B (Blue Box, Middle-Top): Contains Pseudobotrialepis, Hymenolepis, and Rodentolepis asymmetrica species from shrews, rodents, and hedgehogs. Silhouettes of a rodent and a hedgehog are shown.
* Unlabeled Central Group: Below Clade B, several nodes are highlighted in bold red (M. crenatus and Hymenolepididae sp. from Apodemus) and bold blue (Microsomacanthus species from birds). Silhouettes of a rodent and two birds are present.
* Clade A (Green Box, Middle-Bottom): Contains Ditestolepis diaphana and D. tripartita from shrews in Finland. A shrew silhouette is shown.
* Clade D (Yellow Box, Bottom): Contains Hymenolepis microstoma, Rodentolepis straminea, and Staphylocystis furcata from humans, rodents, and shrews. Silhouettes of a human, a rodent, and a shrew are shown.
* Basal Nodes: At the very bottom, outside the colored boxes, are sequences for Hymenolepis nana, Rodentolepis fraterna, and Staphylocystis tiara.
Branch support values (posterior probabilities) are indicated at the nodes, with many showing 100. A scale bar at the bottom represents 0.03 substitutions per site.
Similarity of 28S rDNA nucleotide sequences between rodent (in red) and avian (blue) derived Microsomacanthus isolates and their close relatives. The first figure refers to the number of nucleotides from the isolate listed in each row that are identical to those in the nucleotide sequence from the isolate listed in the corresponding column, given in the second figure. These may vary depending on the degree of similarity between the compared sequences. Percentage of identical nucleotides is given in parenthesis
Table 2. Long description
The table compares 10 isolates across rows and columns. Each cell contains a fraction representing identical nucleotides over total nucleotides, followed by a percentage in parentheses.
Row 1: M. crenatus G U 1 6 6 2 4 7 compared to:
- M. crenatus G U 1 6 6 2 4 6: 100 percent.
- Hymenolepididae L C 5 3 5 2 4 2: 1524/1529 (99.67 percent).
- V. farciminosa J Q 9 5 0 6 8 9: 1067/1100 (97.00 percent).
- C. podicipina P Q 0 8 4 0 8 9: 1414/1518 (93.15 percent).
- M. passeris J Q 9 5 0 6 9 2: 1007/1082 (93.07 percent).
- Microsomacanthus sp. J Q 9 5 0 6 9 0: 1012/1093 (92.59 percent).
- R. evaginata G U 1 6 6 2 4 5: 1245/1337 (93.12 percent).
- R. asymmetrica G U 1 6 6 2 3 3: 1430/1532 (93.34 percent).
- Hymenolepis sp. G U 1 6 6 2 2 7: 1244/1342 (92.70 percent).
Row 2: M. crenatus G U 1 6 6 2 4 6 compared to:
- Hymenolepididae L C 5 3 5 2 4 2: 1360/1365 (99.63 percent).
- V. farciminosa J Q 9 5 0 6 8 9: 1063/1096 (96.99 percent).
- C. podicipina P Q 0 8 4 0 8 9: 1246/1337 (93.19 percent).
- M. passeris J Q 9 5 0 6 9 2: 1003/1078 (93.04 percent).
- Microsomacanthus sp. J Q 9 5 0 6 9 0: 1008/1089 (92.56 percent).
- R. evaginata G U 1 6 6 2 4 5: 1242/1334 (93.10 percent).
- R. asymmetrica G U 1 6 6 2 3 3: 1240/1331 (93.16 percent).
- Hymenolepis sp. G U 1 6 6 2 2 7: 1244/1342 (92.70 percent).
Row 3: Hymenolepididae L C 5 3 5 2 4 2 compared to:
- V. farciminosa J Q 9 5 0 6 8 9: 1067/1096 (97.35 percent).
- C. podicipina P Q 0 8 4 0 8 9: 1410/1513 (93.19 percent).
- M. passeris J Q 9 5 0 6 9 2: 1005/1078 (93.23 percent).
- Microsomacanthus sp. J Q 9 5 0 6 9 0: 1010/1089 (92.75 percent).
- R. evaginata G U 1 6 6 2 4 5: 1244/1334 (93.25 percent).
- R. asymmetrica G U 1 6 6 2 3 3: 1431/1528 (93.65 percent).
- Hymenolepis sp. G U 1 6 6 2 2 7: 1431/1528 (93.65 percent).
Row 4: V. farciminosa J Q 9 5 0 6 8 9 compared to:
- C. podicipina P Q 0 8 4 0 8 9: 1007/1108 (90.88 percent).
- M. passeris J Q 9 5 0 6 9 2: 1020/1104 (92.39 percent).
- Microsomacanthus sp. J Q 9 5 0 6 9 0: 1029/1111 (92.62 percent).
- R. evaginata G U 1 6 6 2 4 5: 992/1099 (90.26 percent).
- R. asymmetrica G U 1 6 6 2 3 3: 1001/1099 (91.08 percent).
- Hymenolepis sp. G U 1 6 6 2 2 7: 977/1102 (88.66 percent).
Row 5: C. podicipina P Q 0 8 4 0 8 9 compared to:
- M. passeris J Q 9 5 0 6 9 2: 1001/1085 (92.26 percent).
- Microsomacanthus sp. J Q 9 5 0 6 9 0: 1011/1096 (92.24 percent).
- R. evaginata G U 1 6 6 2 4 5: 1266/1351 (93.71 percent).
- R. asymmetrica G U 1 6 6 2 3 3: 1419/1499 (94.66 percent).
- Hymenolepis sp. G U 1 6 6 2 2 7: 1262/1356 (93.07 percent).
Row 6: M. passeris J Q 9 5 0 6 9 2 compared to:
- Microsomacanthus sp. J Q 9 5 0 6 9 0: 1064/1089 (97.70 percent).
- R. evaginata G U 1 6 6 2 4 5: 996/1075 (92.65 percent).
- R. asymmetrica G U 1 6 6 2 3 3: 1022/1076 (94.98 percent).
- Hymenolepis sp. G U 1 6 6 2 2 7: 1009/1077 (93.69 percent).
Row 7: Microsomacanthus sp. J Q 9 5 0 6 9 0 compared to:
- R. evaginata G U 1 6 6 2 4 5: 1002/1086 (92.27 percent).
- R. asymmetrica G U 1 6 6 2 3 3: 1025/1087 (94.30 percent).
- Hymenolepis sp. G U 1 6 6 2 2 7: 1010/1088 (92.83 percent).
Row 8: R. evaginata G U 1 6 6 2 4 5 compared to:
- R. asymmetrica G U 1 6 6 2 3 3: 1276/1348 (94.66 percent).
- Hymenolepis sp. G U 1 6 6 2 2 7: 1268/1350 (93.93 percent).
Row 9: R. asymmetrica G U 1 6 6 2 3 3 compared to:
- Hymenolepis sp. G U 1 6 6 2 2 7: 1295/1355 (95.57 percent).
Since V. farciminosa (Goeze, Reference Goeze1782) appears to be a sister species to the Microsomacanthus isolates from Apodemus spp., we propose that the subclade that includes all these sequences is henceforth best referred to as the Variolepis clade. Following Czaplinski and Vaucher’s synonymization of Variolepis with Wardium, the former genus was later reinstated and again recognized as a valid hymenolepidid genus (Dimitrova et al. Reference Dimitrova, Mariaux, Georgiev and Vasileva2019; Mariaux et al. Reference Mariaux, Tkach, Vasileva, Waeschenbach, Beveridge, Dimitrova, Haukisalmi, Greiman, DTJ, Makarikov, Philips, Razafiarisolo, Widmer, Georgiev, Caira and Jensen2017). Given the distinct clade for avian Microsomacanthus spp. evident in Figure 3, this in turn implies that the generic name for the isolates from Apodemus spp. should be Variolepis and not Microsomacanthus. Moreover, given that isolates from rodents and woodpeckers differ markedly in terms of their isozyme systems and differ in key morphologic characters (Vaucher et al Reference Vaucher, Santalla, Casanova, Feliu, Combes and Jourdane2003), the correct name for this cestode from wood mice should be Variolepis murissylvatici comb. nov. Variolepis farciminosa has rostellar hook length and shape very similar to those of V. murissylvatici (10 fraternoid hooks of length 20–24 μm; Mariaux et al., Reference Mariaux and Georgiev2020). This supports the conclusion that V. farciminosa and V. murissylvatici are closely related sister lineages, and therefore most likely congeneric. Furthermore, intriguingly, the close relationship between these species found infecting rodents and birds is indicative of unstable host usage, either in current ‘ecological’ time (whereby there may be generalist species concurrently exploiting both birds and rodents), or in recent evolutionary time (whereby rodent- or bird-specialist species may have arisen following host switches in the recent past).
It is also highly relevant that published descriptions of M. serpentulus (Schrank, Reference Schrank1788) Rudolphi, Reference Rudolphi1810 indicate that this species may be synonymous with V. murissylvatici but, as stated by Vaucher et al (Reference Vaucher, Santalla, Casanova, Feliu, Combes and Jourdane2003), this relationship remains to be assessed and substantiated (or refuted) by biochemical and/or molecular genetic methods. Unfortunately, if they are shown to be synonymous, yet more complications will arise, since the priority for nomenclature will be with M. serpentulus (= H. serpentulus) (International Commission on Zoological Nomenclature, 1999). However, given our molecular phylogeny (Figure 3) and the clear evidence that ‘Microsomacanthus’ isolates from rodents do not fit into the avian Microsomacanthus line, but are closely related to V. farciminosa, the generic name will then need to be amended to Variolepis, hence V. serpentulus comb. nov., rather than V. murissylvatici. Nevertheless, our view is that the most parsimonious solution for the time being is that the cestode infecting Apodemus spp. should be referred to as Variolepis murissylvatici, until such time as molecular genetic analysis indicates otherwise. For these reasons, when considering the cestode from rodents in the following sections of this review, we refer to V. murissylvatici.
The phylogenetic tree (Figure 3) also includes a 28S rDNA sequence from a cysticercoid of Confluaria podicipina (Szymanski, Reference Szymanski1905) Spasskaya, Reference Spasskaya1966 derived from Artemia sp. Adult worms of this species are parasites of grebes (Podicipedidae) across the Holarctic and Neotropical Regions, and also possess ten rostellar hooks, but their shape is aploparaksoid (with a special appendage on the handle and guard) rather than fraternoid (Georgiev et al. Reference Georgiev, Sánchez, Green, Nikolov, Vasileva and Mavrodieva2005; Guagliardo et al. Reference Guagliardo, Graff, Gigola and Tanzola2020; Redón et al. Reference Redón, Berthelemy, Mutafchiev, Amat, Georgiev and Vasileva2015). The single 28S rDNA sequence available in GenBank places this species as a close relative but distinct from the Variolepis clade referred to above (Table 1; 92.15–93.19% homology). As found by Haukisalmi et al. (Reference Haukisalmi, Hardman, Foronda, Feliu, Laakkonen, Niemimaa, Lehtonen and Henttonen2010) and Makarikov et al. (Reference Makarikov and Binkienė2022), the 28S rDNA sequence from R. evaginata did not fit any of the clades or subclades and is represented in Figure 3 by a relatively deep branch on its own, perhaps suggesting that there may be another clade among the Hymenolepididae still to be defined.
Whilst above we have reviewed and amended the nomenclature applying to V. murissylvatici, we recognise that the names given to organisms become more informative and valuable to the extent that they embody known biological traits. Thus, in the sections below we bring together what is currently known about the biology of V. murissylvatici, hoping to firmly link this information to our current species concept.
Life cycle in the wild
Intermediate hosts of V. murissylvatici and V. crenata
As far as we are aware, the only studies of possible intermediate hosts have been carried out on cestodes isolated from birds of the order Passeriformes. Spasskii (Reference Spasskii1965) wrote that Geotrupes sylvaticus (Panzer 1795) (Coleoptera; but now known as Anoplotrupes stercorosus (Hartmann in Scriba, 1791)) was the only known intermediate host of V. crenata from carrion crows (Corvus corone), but pointed out that this beetle was unlikely to be the intermediate host in tropical or steppe zones, where it does not occur. Of twenty beetle species that Prokopič (Reference Prokopič1967) surveyed, he recorded Carabus hortensis (the granulated ground beetle) as the only naturally infected beetle species to carry fully developed infective cysticercoids of V. crenata. Mourad (Reference Mourad1967) successfully infected C. hortensis, using V. crenata worms isolated from the European Jay (Garrulus glandarus). We are not aware of any records employing isolates from the Piciformes (woodpeckers and related species).
However, in 1977 we successfully infected the flour beetle Tribolium confusum with eggs obtained from gravid proglottides of V. murissylvatici isolated from wood mice from Nottinghamshire and then maintained the species for over two decades by routine passage using flour and mealworm beetles (Tenebrio molitor) and laboratory mice (M. musculus) as definitive hosts. Fleas (Ctenophthalmus spp.) have also been suggested as suitable intermediate hosts (Klimpel et al. Reference Klimpel, Förster and Schmahl2007a), but there is no evidence to support their involvement in the life cycle of this cestode species.
Definitive hosts of V. murissylvatici
Mammalian hosts
The species has been recovered mostly from Apodemus spp., i.e. from wood mice (A. sylvaticus), yellow-necked mice (A. flavicollis), Ural mice (A. uralensis (=A. microps)), striped field mice (A. agrarius) and large Japanese field mice (A. speciosus) (Table 3). Non-Apodemus spp. hosts include short-tailed field voles (Microtus agrestis) and bank voles (Clethrionomys glareolus) (Table 3). Chiriac and Hamar (Reference Chiriac and Hamar1966) also mention A. tauricus (Steppe field mouse; = Mus sylvaticus tauricus Barret-Hamilton, 1900; see Kryštufek, Reference Kryštufek2002) as a host but do not provide further details. Laboratory-maintained hamsters and mice can act as hosts but may need to be treated with immunosuppressive drugs to allow the parasite to develop fully to maturity (see below).
Studies recording V. murissylvatici in rodent hosts
Table 3. Long description
The table consists of nine columns: Author, Name used, Location, Host, n (sample size), Prevalence, Intensity, Abundance, and Range.
Key entries include:
* Baer, 1931: Taenia muris-sylvatici in A. sylvaticus, Switzerland.
* Stammer, 1955: H. muris-sylvatici in A. sylvaticus, Germany; n = 127, prevalence 0.79.
* Rybicka, 1959: H. muris-sylvatici in A. flavicollis, Poland; n = 8, prevalence 12.5, intensity 30, abundance 3.75, range 0 to 30.
* Schmidt, 1961: H. muris-sylvatici in Germany; hosts include A. flavicollis (n = 76, prevalence 5.3), A. sylvaticus (n = 163, prevalence 1.8), and C. glareolus (n = 215, prevalence 0.9).
* Lewis and Twigg, 1972: H. muris sylvatici in England; hosts A. sylvaticus (n = 137, prevalence 7.3) and M. agrestis (n = 17, prevalence 17.6).
* Behnke et al., 1999: Microsomacanthus crenata in A. sylvaticus, England; n = 134, prevalence 24.6, intensity 15.4, abundance 3.8, range 0 to 131.
* Lewis et al., 2023: M. crenata in A. sylvaticus, England; n = 440, prevalence 3.4, intensity 8.5, abundance 0.29, range 0 to 26.
Many entries for earlier studies (e.g., Baylis 1939, Bray 1952) list data as n g (not given). The table notes define prevalence as the percentage of animals infected, abundance as the mean worm burden of all sampled animals, and intensity as the mean worm burden of infected animals only.
a 88 C. glareolus were not infected.
b 14 C. glareolus were not infected.
c Prevalence is the percentage of animals infected, abundance is the mean worm burden of all sampled animals including those without the tapeworm, whereas intensity is the mean worm burden of infected animals only.
ng, not given in the cited publication.
na, details not available to us, written in Bulgarian, summarising four original publications in the Bulgarian language.
Avian hosts
Writing about V. crenata, Prokopič (Reference Prokopič1967) listed the following families: Upupidae (hoopoes), Picidae (woodpeckers), Turdidae (thrushes), Laniidae (shrikes), Fringillidae (finches), Motacillidae (pipits and wagtails), Sturnidae (starlings), Oriolidae (orioles), and Corvidae (crows) as hosts. Hanzelová and Ryšavý (Reference Hanzelová and Ryšavý1996) added Passeridae (sparrows) to the list and, referring to P. crenata, recorded the species as abundant across Slovakia (see also the references cited in this paper).
However, given the conclusions of Vaucher et al (Reference Vaucher, Santalla, Casanova, Feliu, Combes and Jourdane2003), it is now unlikely that V. crenata (from passerine avian hosts) and V. murissylvatici (from wood mice) are the same species. As explained earlier, other avian parasitic candidates for synonymy with V. murissylvatici are M. serpentulus (= H. serpentulus) or possibly V. farciminosa. These have been recorded from Passerine species in Europe (Table 4) as well as further afield, throughout the world (Ransom Reference Ransom1909; Jones Reference Jones1945 (American robins, starlings)), including the British Isles, in Wales (Elce Reference Elce1965 (raven, carrion crow, rook, magpie, jay, fieldfare)) and in England (Baylis Reference Baylis1928, Reference Baylis1939 (song thrush, jackdaw, blackbirds, ring-ouzel, carrion crow) and Mettrick Reference Mettrick1956, Reference Mettrick1958 (Corvidae, Turdidae and starlings)). Vaucher et al’s (Reference Vaucher, Santalla, Casanova, Feliu, Combes and Jourdane2003) isolates of M. serpentulus were from blackbirds (Turdus merula).
Records of the occurrence of Variolepis farciminosa and V. serpentulus in birds of Europe
Table 4. Long description
The table is organized into six columns: Country, Locality, Name of cestode used by authors, Host species, Common name of host, and Authors.
Key records for Variolepis farciminosa (often cited as Hymenolepis farciminosa) include:
* England: Multiple records in Hertfordshire, Yorkshire, Cumbria, Essex, Hampshire, Herefordshire, and Worcestershire. Hosts include Starlings (Sturnus vulgaris), Magpies (Pica pica), Jays (Garrulus glandarius), and Jackdaws (Coloeus monedula).
* Wales: Records from Skomer Island and Radnorshire in Starlings and Jays.
* Bulgaria: Records in Sakar Mountains and Northeast region involving Starlings, Golden Orioles, Jays, and Common Redstarts. One record in Vranja identifies an Earth-boring dung beetle (Geotrupes stercorarius) as a host.
* Other countries: Belgium, Lithuania/Russia, Germany, Switzerland, Finland, Poland, France, Czech Republic, and Slovakia report occurrences in hosts such as Blackbirds (Turdus merula), Fieldfares (Turdus pilaris), and Mistle thrushes (Turdus viscivorus).
Key records for V. serpentulus (often cited as Hymenolepis serpentulus) include:
* England: Extensive records in Hampshire, Surrey, Gloucestershire, Cambridgeshire, and Hertfordshire. Hosts include Blackbirds, Ring-ouzels, Carrion Crows, Song-thrushes, Rooks, and Jackdaws.
* Wales: Records from Aberystwyth and Skomer Island in Carrion Crows, Rooks, Starlings, and Redwings.
* France: Records in Gannat for Carrion Crows, Blackbirds, Starlings, Orioles, and Jays.
* Other countries: Scotland, Belgium, Spain, Poland, Germany, and Iceland (Rock ptarmigan).
Taxonomic variations in the cestode names used by authors include Hymenolepis farciminosa, Wardium farciminosa, Hymenolepis serpentulus, Mayhewia serpentulus, Taenia serpentulus, and Passerilepis serpentulus.
Distribution of V. murissylvatici
Based on current records, V. murissylvatici shows a very patchy occurrence in wood mice (Table 3 and Figure 4). Published data record its occurrence in wood mice in the UK from Surrey and Bristol and we have recorded the species in wood mice from N. Norfolk and S. Nottinghamshire (Behnke, unpublished observation). On the European mainland, it has been recorded in mice from Germany (Klimpel et al. Reference Klimpel, Förster and Schmahl2007a; Schmidt Reference Schmidt1961), Italy (Rosso et al Reference Rosso, Scalet, Ferrari, Manfredi, Rizzolia and Cerbo2005), Switzerland (Baer Reference Baer1930; Mariaux Reference Mariaux2024; Vaucher and Hunkeler Reference Vaucher and Hunkeler1967), Czechia (Prokopič Reference Prokopič1967), Croatia (Haukisalmi et al. Reference Haukisalmi, Hardman, Foronda, Feliu, Laakkonen, Niemimaa, Lehtonen and Henttonen2010), Hungary (Murai Reference Murai1972), Romania (Chiriac and Hamar Reference Chiriac and Hamar1966), Poland (Dwużnik et al. Reference Dwużnik, Gortat, Behnke, Gryczyńska, Bednarska, Mikoszewski, Kozakiewicz and Bajer2017; Rybicka Reference Rybicka1959), Bulgaria (Genov Reference Genov1984), and Central Europe (Goüy de Bellocq et al. Reference Goüy de Bellocq, Sarà, Casanova, Feliu and Morand2003). Even further afield, Sasaki et al. (Reference Sasaki, Anders and Nakao2021) reported a Microsomacanthus sp. (most likely V. murissylvatici) isolated from A. speciosus from Japan. Within countries, V. murissylvatici shows a patchy occurrence. Ondriková and Stanko (Reference Ondriková and Stanko2009) found that it was present in an A. agrarius population in Lučenecká Kotlina (n = 57 mice) in Slovakia, but not in Košická Kotlina (n = 63 mice) in the same study, nor in another larger study (n = 96; Ondriková et al. Reference Ondriková, Miklisová, Ribas and Stanko2010).
The location of studies reporting V. murissylvatici in rodents in Europe. See Table 3 for more details. A single report of the species from Japan is not included.
Where V. murissylatici does occur, prevalence and abundance (mean worm burden in all sampled animals, including those not infected, as defined by Margolis et al (Reference Margolis, Esch, Holmes, Kuris and Schad1982) and Bush et al. Reference Bush, Lafferty, Lotz and Shostak1997) in wood mice are usually very low, but occasionally high values for these metrics have been published (Table 3; Murúa, Reference Murúa1978; Lewis and Twigg, Reference Lewis and Twigg1972). Of equal interest are studies in which the species has not been recorded from wild mouse populations. Among more substantial studies in the British Isles that did not detect the species are Elton et al. (Reference Elton, Ford, Baker and Gardiner1931), Sharp (Reference Sharpe1964), Lewis (Reference Lewis1968), Langley and Fairley (Reference Langley and Fairley1982), O’Sullivan et al (Reference O’Sullivan, Smal and Fairley1984), Montgomery and Montgomery (Reference Montgomery and Montgomery1988), Abu-Madi et al. (Reference Abu-Madi, Behnke, Lewis and Gilbert2000), Loxton et al. (Reference Loxton, Lawton, Stafford and Holland2017), and Stuart et al. (Reference Stuart, Paredis, Henttonen, Lawton, Ochoa Torres and Holland2020). Examples of surveys of Apodemus spp. on the Europe mainland that also failed to record the species include those in Czechia (Erhardová Reference Erhardová1958), Slovakia (Tenora Reference Tenora1967), Poland (Kisielewska Reference Kisielewska1970), Finland (Tenora et al. Reference Tenora, Henttonen and Haukisalmi1983), Portugal (Eira et al. Reference Eira, Torres, Vingada and Miquel2006), Spain (Debenedetti et al. Reference Debenedetti, Sainz-Elipe, Sáez-Durán, Galicia, Imaz, Galán-Puchades and Fuentes2015; Fuentes et al. Reference Fuentes, Saez, Trelis, Munoz-Antoli and Esteban2004a, Reference Fuentes, Sáez, Trelis, Galán-Puchades and Estebanb; Portoles et al. Reference Portoles, Granel and Esteban2000; Sáez-Durán et al. Reference Sáez-Durán, Debenedetti, Sainz-Elipe, Sabater-Tena, Galán-Puchades and Fuentes2021; Torres et al. Reference Torres, Trelis, Espert, Ribas, Toledo, Casanova, Roman, Arrizabalaga, Esteban and Feliu2003), Romania (Mészáros and Murai Reference Mészáros and Murai1979), Denmark (Tenora et al. Reference Tenora, Andreassen, Hindsbo and Lodal1991), Serbia (Bjelić-Čabrilo et al. Reference Bjelić-Čabrilo, Čabrilo and Popović2013), and European Russia (Kirillova et al. Reference Kirillova, Kirillov, Ruchin and Alpeev2024). Some studies in Spain (Feliu et al. Reference Feliu, Gracenea, Torregroasa, Sans-Coma, Mas-Coma and Gosálbez1987, Reference Feliu, Renaud, Catzeflis, Hugot, Durand and Morand1997), S. France and NE. Spain (Galán-Puchades et al. Reference Galán-Puchades, Fuentes, Cerezuela, Fons and Mas-Coma1998) and S. Italy (Milazzo et al. Reference Milazzo, Aloise, Cagnin, Di Bella, Geraci, Feliu and Casanova2005), while reporting the presence of the common hymenolepidids of rodents, also recorded infections in wood mice with Hymenolepis sp., perhaps reflecting V. murissylvatici or other unidentified hymenolepidids.
Clearly, the current evidence for V. murissylvatici as a rodent specialist or a generalist capable of infecting both rodents and birds is equivocal. However, the existence of many other Variolepis spp. in avian hosts (Mariaux et al. 2020) and the low and patchy ecological metrics of V. murissylvatici in rodents support the view that it is probably primarily a parasite of birds, with the capacity also to develop to maturity in Apodemus spp. rodents and that the latter become infected inadvertently when feeding on cysticercoid-carrying arthropods in habitats frequented by both taxa, such as woodlands. On the other hand, the geographical distributions of both M. serpentulus (see Jones Reference Jones1968; Slater Reference Slater1967) and V. farciminosa (see Mariaux and Georgiev Reference Mariaux and Georgiev2020) extend into the New World and are more widespread than that of V. murissylvatici. Records for the latter are in Europe, although a further observation in Japan (supported by >99% 28S rRNA identity to European V. murissylvatici) points to a wider Palaearctic range. This lack of records of V. murissylvatici from New World, African and Australasian rodents argues in favour of it being a separate, distinctive species (either a Palaearctic rodent specialist (i.e. Apodemus spp.), or perhaps unrecognised when infecting birds), and this is consistent with only c. 97% 28S rRNA sequence identity with V. farciminosa. It is acknowledged, however, that further faunistic and taxonomic work may still have much to reveal.
If V. murissylvatici turns out to be synonymous with M. serpentulus or V. farciminosa (less likely based on observed genetic identity), given the wide avian host ranges of the latter two species, V. murissylvatici will have to be considered a generalist, rather than a highly host-specific cestode, and this may also be a factor facilitating its survival in rodents. Many of the avian families listed by Elce (Reference Elce1965) include species that live in woodlands, the preferred habitat of wood mice, where the parasite is likely to cycle between arthropod intermediate hosts and insectivorous birds, with rodents perhaps inadvertently picking up infection when feeding on the same invertebrates. Moreover, as noted above, the generally low prevalence and abundance of V. murissylvatici in wood mice (see Table 2) is not consistent with the idea that rodents such as wood mice and other Apodemus spp. are the primary definitive hosts of this cestode.
Life cycle in laboratory rodents and evidence for host resistance
Attempts to keep the life cycle in the laboratory have been successful. We recovered adult V. murissylvatici from wood mice trapped in a woodland located about 10 km south of Nottingham in 1976–1977, fed gravid proglottides to flour beetles (Tribolium confusum, Coleoptera, Tenebrionidae) and obtained viable cysticercoids. Oral administration of mature cysticercoids to laboratory mice resulted in variable establishment of worms (see below), but the strain of mice employed for this work (CFLP) in Nottingham rejected adult worms within two weeks or so (Grencis Reference Grencis1978), and maintenance was only successful if the mice were given subcutaneous cortisone (a wide-spectrum immunosuppressant in rodents). The life cycle was maintained successfully in this way in Nottingham for over two decades. Cysticercoids were sent to Prof. C.A. Hopkins at Glasgow University, who then sent some to collaborators at the State University of Gent in Belgium (Van Haeren and de Rycke 1986). Hopkins had identified the worms provisionally as H. muris-sylvatici, and the Belgian workers continued to refer to the isolate by this name.
Our experience with maintaining this cestode in flour beetles and laboratory mice was that initial establishment after oral administration of cysticercoids was never 100% in mice, and usually more in the region of 50-60% (Grencis Reference Grencis1978), and this was also the experience of Van Haeren and de Rycke (Reference Van Haeren and De Rycke1984). Murine cestodes such as Hymenolepis diminuta and R. microstoma establish much better, with 90–100% of cysticercoids giving rise to worms in the intestines of mice (Hopkins et al. Reference Hopkins, Subramanian and Stallard1972), which is consistent with the idea that V. murissylvatici is not primarily a parasite of rodents, as discussed above. Variable, poor establishment in mice suggests that some of the signals that would normally elicit excystation of cysticercoids from consumed flour beetles may be lacking in the intestines of murine hosts. Alternatively, it could be that despite normal appearance under microscopy, some cysticercoids of this species do not fully develop in flour beetles or perhaps are damaged by immune responses during and/or after completion of development in flour beetles (Van Haeren and de Rycke Reference Van Haeren and De Rycke1984). In addition, differences in intestinal morphology and physiology between mammals and birds (Stevens and Hume Reference Stevens and Hume1995) may also be important factors in determining establishment success of this cestode in laboratory mice.
Experiments in CFLP and NMRI strain mice in Belgium showed that when mice were infected with 6 cysticercoids, fewer than 60% of the worms survived for 8 days after infection, and all had been rejected by day 16 and day 18, respectively (Van Haeren and de Rycke 1984, 1986); this agrees with our own work (Grencis Reference Grencis1978). Worms surviving in male CFLP mice until day 17 had an average length of 92.7 ± 34.2 (SD) mm. Survival was better in cortisone-treated mice, with worms measuring 110.3 ± 9.7 mm on day 17 post infection (PI), and this improvement in worm survival following cortisone treatment was confirmed by Van Haeren and de Rycke (Reference Van Haeren and De Rycke1986). Interestingly, surgical transplantation of 6–8-day-old worms, obtained from donor mice, into recipients resulted in 100% recovery of worms, showing good growth in the week following the operation (Van Haeren and de Rycke 1986), with worms achieving a dry weight exceeding 10 mg by day 14 post-surgery, before eventual expulsion.
Worms grew even better in untreated golden hamsters, achieving a length of 116.2 ± 12.7 mm on day 17 PI, but as found also by Van Haeren and de Rycke (1986), initial establishment was as variable as in mice. Despite the uncertainty of the extent of initial establishment, the latter authors then maintained the species in hamsters. When cysticercoids were administered to female Wistar rats, infection was far less successful, with worm recovery only 17% (Van Haeren and de Rycke Reference Van Haeren and De Rycke1986).
Among the studies conducted in Ghent, Van Haeren and de Rycke (Reference Van Haeren and De Rycke1986) found that a primary infection with V. murissylvatici in NMRI mice induced acquired immunity to a subsequently administered challenge infection, and that a primary infection with R. microstoma also caused an accelerated loss of a challenge infection with V. murissylvatici (Van Haeren et al. Reference Van Haeren, Van der Vorst, De Vliegher and De Rycke1988). Hence, as in earlier studies of cross-immunity between experimental hymenolepidid cestode infections in mice (Alghali and Grencis Reference Alghali and Grencis1986) and as reflected also in the accompanying antibody responses (Van Haeren et al. Reference Van Haeren, Van der Vorst, De Vliegher and De Rycke1988), a degree of cross-immunity exists between V. murissylvatici and R. microstoma.
Epidemiology of Variolepis spp. in avian hosts
Birds are common hosts for cestodes, harbouring the most diverse helminth fauna of any vertebrate group (McLauglin Reference McLaughlin, Atkinson, Thomas and Hunter2008). Hymenolepidids comprise a significant proportion of the species infecting birds (McLaughlin Reference McLaughlin2003) and are consequently commonly reported from surveys of helminths of avian hosts. Although both V. farciminosa and V. serpentulus are regularly described from passerine birds in Europe (Table 4), Asia (e.g. Dutt and Mehra Reference Dutt and Mehra1962; Al-Ankari et al Reference Al-Ankari, Elamin, Al-Atiya and Al-Jabr2003), including Australasia (McKenna, Reference McKenna1998), and America (Hair and Forrester, Reference Hair and Forrester1970; Jones, Reference Jones1968), a greater level of ecological information is available for the former than the latter.
The recorded occurrence of V. farciminosa indicates a predominant Holarctic distribution. Within Europe and North America, passerine birds of the families Corvidae and Sturnidae are most frequently infected. The life cycle is that of a typical hymenolepidid, with natural intermediate hosts being beetles of the species Geotrupes stercorarius (Prokopič and Karapcanski Reference Prokopič and Karapcanski1973) and, experimentally, Pterostichus madidus, Aphodius sp. and Tribolium sp. (Binder Reference Binder1971; Mourad Reference Mourad1967). In the UK, it is the most frequently recorded cestode of starlings (Sturnus vulgaris), where it occurs throughout the year in both juveniles and adults, suggesting a continual exposure to infection, although with an increased frequency in the April to June period (Mettrick Reference Mettrick1958; Owen and Pemberton Reference Owen and Pemberton1962; Pemberton Reference Pemberton1960). Indeed, other studies from localities around St. Petersburg, Russia, have suggested a double summer peak in June and August (Markov Reference Markov1939). Nevertheless, there appears to be little difference in infection levels of V. farciminosa between adults and juvenile birds (Bernard Reference Bernard1987; Owen and Pemberton Reference Owen and Pemberton1962).
In general, the exchange of parasite species, particularly cestodes, between related avian hosts is common (McLaughlin Reference McLaughlin, Atkinson, Thomas and Hunter2008; Rausch Reference Rausch1983), where it has been suggested that ecological rather than physiological factors or phylogenetic relationships are more dominant influences for determining successful infection (Rausch Reference Rausch1983). Indeed, examples exist of conspecific natural infections in phylogenetically different hosts and, more rarely, between avian and mammalian species, although in these cases it is considered that imperfections in immunological or physiological function may have been more influential in susceptibility to infection (Rausch Reference Rausch1983). Nevertheless, such broad host specificity is regarded as atypical (Rausch Reference Rausch1983), and so far no regular exchange of parasites between avian and mammalian hosts has been unequivocally established.
Epidemiology of V. murissylvatici in wood mice
There is very little published information on the epidemiology of V. murissylvatici in its wild rodent hosts. Most surveys of wood mice have found prevalence to be low, generally below 8% (Table 2), with only three published studies reporting higher prevalence (Behnke et al. Reference Behnke, Lewis, Mohd Zain and Gilbert1999; Murúa Reference Murúa1978; Rybicka Reference Rybicka1959). Moreover, some reports were based on very low sample sizes (Dwużnik et al. Reference Dwużnik, Gortat, Behnke, Gryczyńska, Bednarska, Mikoszewski, Kozakiewicz and Bajer2017; Rybicka Reference Rybicka1959), thereby restricting further analysis in relation to possible explanatory intrinsic and extrinsic factors that may have influenced prevalence and abundance of infection. Other reports of the species in wood mice appear simply as species lists (Baylis Reference Baylis1939; Bray Reference Bray1952; Klimpel et al. Reference Klimpel, Förster and Schmahl2007a) or isolates for taxonomic (Prokopič Reference Prokopič1967) and genetic (Haukisalmi et al. Reference Haukisalmi, Hardman, Foronda, Feliu, Laakkonen, Niemimaa, Lehtonen and Henttonen2010) studies.
Murúa (Reference Murúa1978) reported a prevalence of 25% in one of the two sites in Bristol in which he trapped wood mice, but no infections in sympatric bank voles, nor in 61 wood mice and 24 bank voles trapped at Slimbridge, located about 30 miles northeast of Bristol. In fact, the only study to record V. murissylvatici in bank voles was that of Schmidt (Reference Schmidt1961), who found worms in just two of the 215 voles that were examined. Lewis and Twigg (Reference Lewis and Twigg1972) examined 88 bank voles from the same woodlands as those with infected wood mice (Table 2), but none carried V. murissylvatici. Similarly, Klimpel et al. (Reference Klimpel, Förster and Schmahl2007b) found that none of the 29 bank voles they examined was infected with V. murissylvatici, despite the presence of the species locally in yellow-necked mice (Table 3).
Neither seasonal nor between-year variation in occurrence of V. murissylvatici in rodents has been thoroughly investigated. In the Bristol site (Wills Hall), Murúa (Reference Murúa1978) recorded a prominent peak in worm burdens in infected mice during the summer months, but this was based on just one heavily infected animal, and overall occurrence of the cestode throughout the year was sporadic. Behnke et al. (Reference Behnke, Lewis, Mohd Zain and Gilbert1999) did not study seasonal effects, their results being based on surveys conducted in September, towards the end of the summer period in the British Isles and the start of autumn in each of four successive years (1994–1997). Both prevalence and mean abundance were high in 1994 (34.1% and 6.4) and 1996 (32.4% and 5.4), compared with much lower values for both parameters in 1995 (12.9% and 0.7) and 1997 (12.0% and 0.6).
If V. murissylvatici turns out to be also parasitic in passerine birds, then woodlands attracting starlings, thrushes, crows and other birds provide an environment in which transmission to and from wood mice is feasible. Higgins (Reference Higgins1977) compared prevalence among wood mice from three contrasting environments. Prevalence among wood mice from scrubland with recovering deciduous woodland was 53.8% (n = 13), compared with 15.8% among mice from pine-dominated forest (n = 19) and 22.2% among those from a marshy borderland surrounding a pond (n = 9).
Few studies have provided information on intrinsic factors (age and sex) influencing infection with V. murissylvatici in naturally infected wood mice. Behnke et al. (Reference Behnke, Lewis, Mohd Zain and Gilbert1999) allocated mice to three age classes, and there was a significant effect of age on abundance. No infections were detected in the youngest animals (age class 1), and mean worm burdens were higher in age class 3 (12.6, 1.0, 6.6, and 1.6, respectively in 1994, 1995, 1996, and 1997) compared with age class 2 mice (0.3, 0.8, 5.1, and 0.6, respectively) in all four years of the study, but the extent of the difference was not consistent and varied significantly between years. Lewis and Twigg (Reference Lewis and Twigg1972) also found that prevalence was higher in adult mice compared to juveniles (9.0% vs 2.7%) and worm burdens were higher in adult mice (abundance = 0.43 vs 0.03, respectively). Although overall prevalence was lower in Lewis et al. (Reference Lewis, Morley and Behnke2023), prevalence was also significantly higher in adult mice (4.8% vs 0.7%). These data indicate that once established in wood mice worms survive for some time and hence accumulate with host age, in turn suggesting that like many other helminths, including cestodes, V. murissylvatici can manipulate the surrounding environment in its intestinal niche to facilitate and prolong survival (Lynch and Grencis Reference Lynch and Grencis2025; Maizels et al. Reference Maizels, Smits and McSorley2018; McKay 2010; Wang and McKay Reference Wang and McKay2005; Wegener Parfrey et al. Reference Wegener Parfrey, Jirků, Šíma, Jalovecká, Sak, Grigore and Pomajbíková2017).
No studies to-date have found that male and female wood mice differ significantly in terms of prevalence and abundance with V. murissylvatici. Lewis and Twigg (Reference Lewis and Twigg1972) commented that worm burdens tended to be higher in female wood mice (13.7% in females and 3.5% in males) but did not substantiate this difference between the sexes by statistical analysis. Behnke et al. (Reference Behnke, Lewis, Mohd Zain and Gilbert1999) also recorded numerically higher prevalence and abundance in female mice (prevalence = 30.2% vs 22.0% and abundance = 5.5 vs 2.9, respectively), but with other factors taken into account, these were not significant differences between the sexes. In contrast, Lewis et al. (Reference Lewis, Morley and Behnke2023) found bias in the opposite direction, recording numerically higher values in male mice (prevalence = 4.0% vs 2.4% and abundance = 0.30 vs 0.27, respectively), but again, with other factors taken into account, the differences between the sexes failed to achieve significance. In a sample of 41 wood mice trapped in a wood in Nottinghamshire, Higgins (Reference Higgins1977) found a prevalence of 31.0% among adult male mice and 25.0% among females, a difference between the sexes that was not significant (p = 0.7). These contrasting findings suggest that, as in many other helminth infections in mammals, sex biases in prevalence and abundance are context dependent, and swayed in favour of one sex or the other depending on the condition of their host, the demands of pregnancy and lactation in females, and ecological circumstances in the immediate environment of their hosts.
Conclusions and recommendations
The relationship of Apodemus spp. parasitic V. murissylvatici to morphometrically similar avian parasitic cestodes is still unresolved and, to some extent, still confused. As we have documented, based on current evidence, isolates of V. murissylvatici from wood mice and avian parasitic species are very similar morphologically. Although Prokopič (Reference Prokopič1967) concluded that the species was synonymous with V. crenata, Vaucher et al (Reference Vaucher, Santalla, Casanova, Feliu, Combes and Jourdane2003) did not accept this synonymy but suggested a closer affinity with M. serpentulus. Molecular phylogenetics based on 28S rDNA has shown that V. murissylvatici isolates from rodents (referred to variously as M. crenata and H. murissylvatici) are closely related genetically to V. farciminosa. Therefore, we propose that the use of H. murissylvatici (and its variants) and M. crenata is no longer justified.
Until more 28S rDNA sequences from avian isolates are made available, we cannot be unequivocally certain about the taxonomic interrelationships between V. crenata, V. farciminosa and M. serpentulus from birds and V. murissylvatici from rodents, and whether any of these should be synonymised, or whether these are cryptic sister species, morphologically very similar, but sufficiently distinct at the genetic level to be considered different species. Birds and mammals rarely share the same species of helminths, and indeed, on this assumption, Czaplinski and Vaucher (Reference Czaplinski, Vaucher, Khalil, Jones and Bray1994) provided separate keys for the identity of hymenolepidid cestodes for avian and mammalian hosts. It should be possible to establish genetic barcodes for isolates from birds and rodents based on the 28S rDNA genetic region and other genes (e.g. ITS region) or, more ambitiously, on whole genomes. Although further comparison of key isoenzymes from each is another possible approach, as suggested by Czaplinski and Vaucher (Reference Czaplinski, Vaucher, Khalil, Jones and Bray1994), this methodology is rarely implemented in the current era, and is regarded generally as outdated and less informative than genetic methods. At this stage, we cannot eliminate the possibility that a rodent-adapted variant circulates among wild sylvatic rodents such as wood mice while an avian-adapted variant infects only birds, both relying on arthropod intermediate hosts. Having established a close genetic similarity of V. murissylvatici to another Variolepis sp., and until more detailed comparisons have been made between well-defined isolates from birds and Apodemus spp., we conclude that, for the time being, the safest and most parsimonious solution is to refer to V. murissylvatici for the Apodemus spp. infective cestode.
The epidemiology of V. murissylvatici in the wild is very poorly known, and in fact virtually untouched by scientists. While studies on non-protected, wild rodents, based on culled animals from which mature worms can be isolated, are feasible, in the current era this is a highly unlikely route to follow for exploring how the avian infective parasites exist in their avian hosts, because birds are protected in Europe by law (European Council Directive 2009/147/EC on the conservation of wild birds (Birds Directive) and European Commission: Directorate-General for Environment, The Birds Directive – 40 years of conserving our shared natural heritage, Publications Office, 2019, https://data.europa.eu/doi/10.2779/622146). One feasible solution would be to depend on sampling the carcasses of animals dying through accidents, predation, or natural causes. It should also be possible to collect faeces from wild birds (e.g. at bird feeders or if captured for conservation-linked activities such as ringing) to obtain eggs, and these could be a source of material for morphological or molecular analysis. The opportunistic nature of these types of sampling is unlikely, however, to permit methodical epidemiological sampling designs. Furthermore, preliminary identification of the eggs could be problematical, given the variation apparent in the literature (see current variation in published sizes in Table 1) and the presence of many other gastrointestinal helminths in birds.
Since isolates from Apodemus spp. can be maintained through passage in flour and mealworm beetles and laboratory mice, experimental work could be expanded to include laboratory-bred wood mice rather than laboratory house mice. Laboratory-bred wood mouse colonies have been established and maintained in British Universities. The work of Van Haeren and de Rycke (1986) shows clearly that V. murissylvatici stimulates potent acquired immunity in laboratory mice, but it would be interesting to determine whether the same applies to infections in wood mice.
Whatever the outcome of the avenues for future research that we have suggested above, V. murissylvatici is clearly a fascinating, enigmatic organism, meriting further attention by parasitologists working on mammalian and avian host systems.
Data availability statement
The authors confirm that the data supporting the findings of this study are available within the article.
Acknowledgements
We thank the staff at the University of Nottingham library, who helped us to trace papers that were not otherwise easily available for scrutiny. We are grateful to Chris Higgins for permission to use data from his study and to Tom Hartman for assistance with the photographs in Figure 2B. We are grateful also to Anna Bajer and Boyko Georgiev for their comments on earlier versions of the manuscript, for their advice and for the provision of records from Bulgaria by Boyko Georgiev that were not otherwise available to us.
Authors’ contribution
The paper was written jointly by the authors.
Financial support
This research received no specific grant from any funding agency, commercial or not-for-profit sectors.
Competing interests
None.
