Host relations

The records of Cimex lectularius and C. pipistrelli in bat roosts in Central Europe, Serbia, Bulgaria and some other European countries are summarized in Table 1 and Supplementary material Table 2. Generally, the number of colonies monitored annually varied among bat species. It was high in non-dwelling bats with high fidelity to shelters, such as Rhinolophus hipposideros, Myotis myotis and M. emarginatus. In species with low philopatry the long-term monitoring is difficult. Even in common species (e.g. Pipistrellus pipistrellus, Nyctalus noctula) it is difficult to locate summer colonies. If found, the roosts cannot be checked in the following years as they are often destroyed or the bats relocate during large-scale renovations of buildings (especially prefabricated houses). Furthermore, such roosts are often difficult to check for both bats and cimicids because of their crevice character, in contrast to the spacious roosts of philopatric bat species. Therefore, the numbers of records for each bat species are not representative with regard to the incidence of cimicids in their roosts. However, the incidence can be considered high at least in some species like Nyctalus spp. and Pipistrellus spp., given the number of records on mist-netted individuals (Balvín et al. 2012b) or roosts inhabited by these species (Supplementary material Table 2).

Cimex pipistrelli was confirmed as a parasite of the attic-dwelling bat species M. emarginatus, Myotis myotis and, possibly, M. blythii, as well as the crevice-dwelling bat species Myotis brandtii, M. daubentonii, Nyctalus noctula, N. leisleri, Vespertilio murinus, Pipistrellus pygmaeus and, possibly, P. pipistrellus. It was newly recorded for Myotis nattereri (Lužnice, Czech Republic, Supplementary material Table 2). The new record for Nyctalus lasiopterus was already mentioned by Balvin et al. (2012b).

Cimex lectularius was confirmed to parasitize on the attic-dwelling bat species Myotis myotis, M. emarginatus and, possibly, M. oxygnathus, as well as the crevice-dwelling bat species Eptesicus serotinus and Pipistrellus sp. (see below).

Among attic-dwelling bat species, both cimicid species were shown to be very common in roosts of Myotis myotis in central Europe. Less than 25 % of these synanthropic roosts inspected by the authors were negative. In the Czech Republic, out of about 140 monitored roosts about 80 were inspected for the presence of cimicids, which were found in 64 (∼80 %) roosts. In total, Cimex lectularius was recorded in 36 roosts of M. myotis and C. pipistrelli in 46. No cimicids were found in the cave roosts in Serbia and Bulgaria.

Myotis myotis was the only bat species hosting both cimicid species in one roost, though only in few isolated cases. Mixed infestation was found only at the colony in Dubá (Czech Republic) in 2013 and 2014, whereas in 2009 only C. lectularius was discovered (Supplementary material Table 2). In 2009, this bat colony was complemented by several hundred bats from a nearby roost in Doksy. In this roost, only C. lectularius was found in the year the bats moved to Dubá, so other, unknown bats had to have brought C. pipistrelli. A mixed infestation was also revealed found in Zemianske Kostol'any (Slovakia), but among remains of bodies of unknown age. A change of infestation from C. pipistrelli to C. lectularius over years has likely been recorded in Úštěk and Držovice (Czech Republic). However, samples from these roosts were not numerous and may not reflect the composition of the species. Part of samples from other roosts also contained only a few individuals but comparing the numbers of roosts with consistent and inconsistent record of Cimex species it is likely that populations of cimicids in single roosts mostly consisted only of one species. Based on this limited evidence, the mechanisms of coexistence or competition between C. pipistrelli and C. lectularius can only be speculated on. However, the co-occurrence of the two species is likely caused by the co-occurrence of different bat species in a single roost. Attic-dwelling bats like M. myotis, frequent hosts of C. lectularius, often share attics with crevice-dwelling bats like Pipistrellus spp. (e.g. Host'ovce) or Eptesicus spp. (e.g. Oleksovice; see Supplementary material Table 2 or Czech bat Conservation Society database, unpublished), frequent hosts of C. pipistrelli. Myotis oxygnathus was confirmed as a host of C. lectularius only in a colony mixed with M. myotis (Martonyi, Hungary, Supplementary material Table 2). In the same area of the Aggtelek Karst in Hungary, two more roosts inhabited only by M. oxygnathus were visited and no trace of cimicids was found. As M. oxygnathus is physically and ecologically almost identical to M. myotis, the absence of cimicids in the roosts of this species may not be a coincidence, considering the at least 80 % incidence of cimicids in synanthropic M. myotis roosts in our record. It is possible that M. oxygnathus does not constitute a suitable host for cimicids for an unknown reason.

In roosts of M. emarginatus, only C. lectularius was confirmed. It was found in 11 synanthropic roosts, often shared with Rhinolophus spp. A similar number of roosts with similar microclimates was negative. There is a single published record of C. pipistrelli from M. emarginatus (Usinger 1966), while the records of C. lectularius are at least three (Usinger & Beaucournu 1967, Protić & Paunović 2006). C. pipistrelli was found only in two roosts of M. emarginatus shared with M. myotis. It is possible that M. emarginatus is the only host that C. lectularius does not share with C. pipistrelli, at least in Central Europe and Serbia.

Further south, however, M. emarginatus has been shown to host C. emarginatus (Simov et al. 2006). The colony from which the type material originated flew away in 2005. Since then, only a single record (female) of C. emarginatus is available from mistnetted Myotis cf. alcathoe (Bulgaria, Monastery Sveti Archangel, Malashevska planina Mts., Blagoevgrad district, 41°51′23.04″ N, 22°59′31.92″ E, 10 September 2011, B. Petrov, I. Alexandrova lgt.). If the determination of the bat species is correct, it is also the first record of a cimicid for this newly described species. In 2006, C. emarginatus was unsuccessfully attempted to be confirmed by investigating a large unfinished building of a hotel near the type locality in Primorsko, which was inhabited by multiple colonies of M. emarginatus and many other bat species (Myotis blythii, Myotis sp., Miniopterus schreibersii, Rhinolophus spp., see Benda et al. 2003). Only two specimens of C. pipistrelli were collected in this study (Supplementary material Table 2).

The presence of C. lectularius in roosts of Pipistrellus sp. was shown indirectly. The bugs attacked people in a gamekeeper's house near Hnanice, South Moravia, and a hunting hide nearby. In both buildings, colonies of Pipistrellus sp. were recorded (P. pipistrellus or P. pygmaeus; not distinguished at that time). Although these bugs were unusually small, similar to C. pipistrelli from Pipistrelhis spp., in other morphological characters they clearly corresponded to C. lectularius from other bat species (Balvín et al. 2012a). The third record of C. lectularius from Pipistrellus sp. was made from a roost shared with M. myotis (Host'ovce, Slovakia).

Roosts inhabited only by Rhinolophus spp. were always free of cimicids, which is consistent with the literature. Unlike vespertilionid bats, Rhinolophus spp., at least in the synanthropic roosts in Central Europe, rarely form tight clusters (Gaisler 1966) and readily move through the roosting space. This likely makes them an unsuitable host for cimicids. Until recently, Plecotus spp. was similarly considered an unsuitable host for cimicids, regarding the entire lack of published records for such common bats. Unlike Nyctalus or Pipistrelhis spp., most of the maternity colonies of Plecotus spp. are comprised of only a few individuals (Anděra & Horáček 2005), which were believed to be incapable of hosting a population of cimicids, similar to the diffusive colonies of Rhinolophus spp.

Out of the 140 summer roosts that are annually monitored in the Czech Republic, at least 23 are inhabited by species that commonly host cimicids together with Plecotus or Rhinolophus spp. (Bartonička & Gaisler 2010). Rhinolophus spp. are also listed as hosts of cimicids in records from colonies mixed with, for example, Myotis emarginatus (Usinger 1966, Usinger & Beaucournu 1967, Protić & Paunovic 2006). A recent record of C. lectularius has been made from Plecotus auritus (Balvín et al. 2012b). Furthermore, the first record of Cimex (C. lectularius) from bats from Serbia was made from Rhinolophus ferrumequinum caught by harp trap. More recently, C. lectularius was found on three more individuals of R. ferrumequinum caught using the same technique, as well as on five specimens of M. emarginatus. These bats came from a mixed colony of the two species. One of the visits of the colony was made in August 2013, when only a few M. emarginatus individuals remained among about a thousand R. ferrumequinum bats. However, the bugs were numerous and recently fed. It is therefore clear that Plecotus and Rhinolophus spp. are able to serve at least as occasional or temporary, though likely less suitable, hosts of cimicids.

Moreover, the preference for a specific host was not detected in host-specificity experiments. C. pipistrelli repeatedly sucked on the bat species in whose roosts they have never been observed (Zedníková 2010). In conclusion, the bugs of the genus Cimex appear to be common ectoparasites of 20 bat species in Europe (Table 1). The difference between the ranges of host species of C. pipistrelli and C. lectularius may suggest different host preferences. These preferences can result from different ecologies of the respective bat species (tree or building dwellings versus large attics), as C. lectularius is found mostly on attic-dwelling bat species. However, the historical distribution of some of the host bat species, especially Myotis myotis, may have played a more important part.

Geographic distribution of cimicids and their hosts While Cimex lectularius as a parasite on man is cosmopolitan, the distribution of the bat-related lineage has never been reviewed. To our knowledge, the records come from the following countries: Afghanistan (Usinger 1966), the Czech Republic (Povolný 1957), Finland (Poppius 1912), France (Usinger & Beaucournu 1967), Germany (e.g. Eichler 1937), Serbia (Protić & Paunović 2006, misidentified as C. pipistrelli) and Slovakia (Usinger 1966). Our records extend the known distribution to Hungary, Switzerland (Table 1) and Ukraine (Table 2 in Balvín et al. 2012b, record from a mist-netted bat).

The distribution of Cimex pipistrelli has been recently reviewed by Péricart (1996). More recent records of C. pipistrelli are by Krištofík & Kaňuch (2006, Slovakia) and Simov et al. (2006, Bulgaria, Greece). The species was newly recorded from Lebanon, Ukraine and Spain; however, all these findings come from mist-netted bats and were listed already by Balvín et al. (2012b).

As cimicids are parasites of bats in their summer roosts, the geographic distribution of cimicids follows the breeding areas of their host species. Records from overwintering bats are singular (Simov et al. 2006). The host range recorded for C. pipistrelli and C. lectularius (Table 1), comprising many bat species with diverse ecologies, may suggest that their distribution evenly covers all Europe. However, this is not true based on comprehensive data available from five countries covering a Northwest-Southeast transect across Europe: the Czech and Slovak Republics, Hungary, Serbia and Bulgaria. Although the incidence of cimicids in roosts of crevice-dwelling bat species in central Europe cannot be exactly determined, considering the numerous records in our material or the literature, cimicids can be regarded as more or less frequent in their roosts. The incidence of cimicids in roosts of attic-dwelling bats in Central Europe appears to be high.

This situation is in contrast with the frequency of records from Bulgaria. Records from 1997–2006, coming from the comprehensive survey of about 500 roosts in caves, buildings, bat-boxes and tree holes, as well as from examination of about 20000 captured bats were summarized by Simov et al. (2006). They comprise the finding of the newly described C. emarginatus in a roost of M. emarginatus and only two records of C. pipistrelli from Nyctalus noctula. Since 2011, an even more extensive survey as part of Natura 2000 has covered about 1600 bat roosts. Furthermore, an additional ca. 10000 mist-netted bats were examined for ectoparasites during the period 2007-2013 in the Tabachka Bat Research Station. During these surveys, no cimicids were found. In addition to these surveys, 1) the second record of C. emarginatus was made; 2) C. pipistrelli was found in Primorsko, as mentioned above; 3) C. pipistrelli was found in 2004 (though only at a photograph) and in 2013 in a bat-box inhabited by Nyctalus noctula at Sedemte Prestola Monastery (Western Balkan Mts., see Supplementary material Table 2); and 4) unidentifiable Cimex specimens were found in 2006 in guano under a roost likely inhabited by Nyctalus noctula in the entrance of Devetashka cave (Lovech Province, see Supplementary material Table 2).

Numerous roosts of Myotis emarginatus were examined in southern Vojvodina, a northern province of Serbia, documenting the likely reason for the absence of C. lectularius in bat roosts in the southern Balkans. If the non-dwelling bat species M. myotis and M. emarginatus can be regarded as the principal hosts for C. lectidarius, as suggested by the limited records from other bats, the absence of C. lectularius on bats in the Balkans can be explained by the characteristics of roosts, as suggested by Simov et al. (2006). The temperature in deep caves, where Myotis spp. typically roost, is usually below 15 °C, while the humidity often reaches 80–90 % (e.g. Paksuz et al. 2007). The combination of low temperature and high humidity has been shown to be unfavourable or even lethal for C. lectularius, and its development is arrested below 13–15 °C (Kemper 1936, Omori 1941). In central Europe, these bat species are synanthropic, inhabiting warm and dry attics. In more southern areas, they mostly occupy their natural habitats of large caves, which are often too humid and cold for cimicids. In Vojvodina province, northern Serbia, C. lectularius seems to be common in roosts of M. emarginatus in buildings, whereas in the roosts in caves in central and south Serbia, cimicids have not been recorded.

However, the reason for such low frequency of records of C. pipistrelli in Bulgaria and Serbia is less clear. Only about one third of the bat roosts examined during the Natura 2000 mapping in Bulgaria were in humid and cold caves, which are often inhabited by M. myotis and M. emarginatus. The rest inhabited shallow, dry caves or their entrances (e.g. the entrance of Devetashka cave, Bulgaria, where remains of bugs were found), tree holes, bat-boxes and buildings which should be suitable for cimicids.

As an explanation for the low frequency of records of both Cimex species in southern Europe, it is possible that stable colonies of attic-dwelling bats like M. myotis provide long-term local reservoirs for infestations in roosts of crevice-dwelling bats. Colonies of such bat species as Pipistrellus spp. or Nyctalus spp. often split and change roosts during the breeding season, a phenomenon that has been described as the fission-fusion behavioural model (Kummer 1971). This behaviour efficiently reduces the numbers of bugs in the roosts and may be occurring, at least partially, for this purpose (Bartonička & Růžičková 2012). Infestations of a local population of a bat species can eventually be eliminated through this behaviour. This is supported by genetic data on both C. lectularius (Balvín et al. 2012a, Booth et al. 2015) and C. pipistrelli (Balvín et al. 2013, Wawrocka in litt.), which show no host-associated structure and suggest frequent switching between bat species within regions. In southern Europe, stable infestations in roosts of attic-dwelling bats are absent and therefore cannot be the source of re-infestation of crevice-dwelling bats.

However, this is only partly true for Nyctalus species, at least N. noctula. Almost all records of C. pipistrelli from southern Europe [Bulgaria, Greece (Simov et al. 2006) and Italy (Lanza 1999)] and Lebanon (Balvín et al. 2012b) are from N. noctula. Only the record from Spain (Balvín et al. 2012b) was from N. lasiopterus. The southern limit of the breeding area of N. noctula is 48° N (Kaňuch & Celuch 2004). In summer, only males and non-reproductive females are found south of this limit. As suggested by Simov et al. (2006), it is possible that all the records of C. pipistrelli from southern Europe are only temporary transmissions by N. noctula. However, at least in the bat-boxes in Cherven and Sedemte Prestola Monastery (Bulgaria), C. pipistrelli was found in two subsequent years (Simov et al. 2006, Supplementary material Table 2). It is therefore possible that stable populations may occur here. N. noctula transmits cimicids much more often (Balvín et al. 2012b) and is much more migratory than other bat species. Therefore, if the hypothesis of local reservoirs of cimicids in M. myotis roosts is valid, in situation when the reservoirs are missing, the local populations in N. noctula roosts are not likely to die off like in other crevice-dwelling bat species.

In conclusion, the distribution of Cimex lectularius and C. pipistrelli is not consistent across the areas of distribution of their host bat species (i.e. different host relations are found in different areas). The distribution of cimicids is presumably shaped by the ecology of bat species. However, it is possible that some bat species provide reservoirs of cimicids, and changes in their ecology across their area of distribution may affect populations of cimicids on other bat species.