BRIEF DESCRIPTIVE ECOLOGY: WHAT DO FLEAS DO?

Transcription

1 PART I BRIEF DESCRIPTIVE ECOLOGY: WHAT DO FLEAS DO?

2 1 Composition of the order Siphonaptera is a relatively small order of secondarily wingless holometabolous insects. According to a recent taxonomic scrutiny by Medvedev et al. (2005), the order includes 2005 species and 828 subspecies belonging to 242 genera and 97 subgenera. In addition, 409 specific, 147 subspecific, 65 generic, and seven subgeneric names are considered to be synonyms. However, some recently discovered species are still absent from Medvedev et al. s (2005) database (e.g. Barriere et al., 2002; Durden & Beaucournu, 2002; Pampiglione et al., 2003; Beaucournu & Wells, 2004; Hastriter & Haas, 2005). Nonetheless, at present, this database is the most complete one available. In this chapter, I briefly outline the composition of the order and provide basic information on the higher-level flea taxonomy. 1.1 Infraorders and families Cladistic analysis of fleas using 50 morphological features of the head, thorax and abdomen (Medvedev, 1994) resulted in the above-generic taxonomic scheme including four infraorders and 18 families as follows. Order Siphonaptera Infraorder Pulicomorpha Family Pulicidae Family Tungidae Family Malacopsyllidae Family Rhopalopsyllidae Family Vermipsyllidae Family Coptopsyllidae Family Ancistropsyllidae 3

3 4 Composition of the order Infraorder Pygiopsyllomorpha Family Lycopsyllidae Family Pygiopsyllidae Family Stivaliidae Infraorder Hystrichopsyllomorpha Family Hystrichopsyllidae Family Chimaeropsyllidae Family Macropsyllidae Family Stephanocircidae Infraorder Ceratophyllomorpha Family Ceratophyllidae Family Leptopsyllidae Family Ischnopsyllidae Family Xiphiopsyllidae Species composition of different flea families ranges from 594 in Hystrichopsyllidae to two in Malacopsyllidae (Medvedev et al., 2005) (Fig. 1.1). Large siphonapteran families (Pulicidae, Tungidae, Rhopalopsyllidae, Pygiopsyllidae, Stivaliidae, Hystrichopsyllidae, Stephanocircidae, Ceratophyllidae and Leptopsyllidae) contain species that exploit different host orders and even host classes (e.g. pulicids, ceratophyllids and leptopsyllids parasitize both mammals and birds). Other families are more host-specific. For instance, fleas of the family Ischnopsyllidae are associated exclusively with bats (Chiroptera), Chimaeropsyllidae with elephant shrews (Macroscelidea) and Malacopsyllidae with armadillos (Cingulata). Soricomorph hosts are parasitized mainly by fleas from the hystrichopsyllid tribe Doratopsyllini, whereas lagomorphs by fleas from the pulicid tribe Spilopsyllini. 1.2 Temporal pattern of discovery of flea species The earliest description of a flea species dates back to 1758, when Linnaeus described the house flea Pulex irritans and the sand (chigoe) flea Tunga penetrans. The next descriptions of flea species were made 40 years later by Bosc (Nosopsyllus fasciatus) and Schrank (Ceratophyllus gallinae and Monopsyllus sciurorum). Discoveries and descriptions of new flea species are continuing at present (e.g. Hastriter, 2000a, b, 2001a, b, 2004; Lewis & Haas, 2001; Lewis & Stone, 2001; Barriere et al., 2002; Durden & Beaucournu, 2002; Hastriter & Whiting, 2002; Hastriter et al., 2002; Hastriter & Eckerlin, 2003; Beaucournu & Wells, 2004; Lewis & Eckerlin, 2004; Acosta & Morrone, 2005; Hastriter & Haas, 2005; Beaucournu et al., 2006).

4 Temporal pattern of discovery of flea species 5 Figure 1.1 Number of valid (a) genera and (b) species in siphonapteran families. Data from Medvedev et al. (2005).

5 6 Composition of the order Figure 1.2 Frequency distribution of dates of description for 2005 flea species. Data from Medvedev et al. (2005). The frequency distribution of dates of flea descriptions demonstrates that the majority of flea species was discovered in the first half of the last century (Fig. 1.2). However, periods when many flea species were described alternated with periods with few descriptions (even reaching zero per year). The three highest peaks in the twentieth century (Fig. 1.2) coincide with the descriptions of many new flea species by highly productive taxonomists (e.g. Jordan & Rothschild, 1915a, b; Rothschild, 1915a, b; Wagner, 1929; Ioff et al., 1946). For example, K. Jordan and N. N. Rothschild, separately and together, described more than 600 flea species, F. Smit and J. Wagner each described about 140 species, I. Ioff and R. Traub each described about 100 species and K.-C. Li, R. Lewis, J.-C. Beaucournu, G. Holland and C. Baker each described about 50 species. Interestingly, the number of researchers involved in description of flea species grew steadily until the beginning of the 1990s and then dropped sharply (Fig. 1.3a). However, description effort per researcher does not follow this pattern. Number of flea species described per researcher per year peaked between the 1920s and 1960s (Fig. 1.3b). The rate of discovery of new flea species is not caused only by the abovementioned subjective reason. Biological parameters of flea species as well as

6 Temporal pattern of discovery of flea species 7 Figure 1.3 Distribution of (a) number of researchers who provided taxonomic description of new flea species (including synonyms) and (b) description effort (number of descriptions per researcher) from 1758 to Data from Medvedev et al. (2005).

7 8 Composition of the order those of their hosts are also involved (see Chapter 13). In addition, the latest discoveries of the new species of Siphonaptera were made mainly in remote geographic regions such as Tasmania, southeastern Asia, central Africa and South America. However, a new species, Jordanopsylla becki, was discovered in North America (Nevada) as late as in 2000 (Hastriter, 2000a), and a new genus (Psittopsylla) was described recently in Mexico (Lewis & Stone, 2001). on flea biology and their host associations is still needed. About 600 flea species (about 30% of the total number of known species) are known from only a single host and a single record (Medvedev, 2002). The highest numbers of such species were found in the Madagascar, Eastern African, Papuan, Malayan and Caribbean biogeographical subregions. These poorly known species were collected mainly from rodents (murines and cricetines) and shrews, although some species were recorded from bats, lagomorphs, carnivores and marsupials.

8 2 Hosts of Siphonaptera Fleas are characteristic parasites of birds and mammals. Occurrences of fleas on reptiles are accidental (e.g. Tillyard, 1926; Dunnet & Mardon, 1974), although they are able to digest blood of these hosts (Belokopytova et al., 1983; Vashchenok, 1988). Accidentally, fleas are able to feed even on haemolymph of ticks (Bilyalov et al., 1989). In this chapter, I address general patterns of distribution of fleas within the two classes of higher vertebrates. 2.1 Avian and mammalian hosts The majority of fleas parasitize mammals (more than 94% of species: Vashchenok, 1988; Beaucournu et al., 2005), whereas their association with birds is much less frequent. Analysing host associations of 1951 flea species, Medvedev (1997a, b) found that fleas were recorded on 16 mammalian and 21 avian orders. Furthermore, 1835 of the 1951 species analysed were harboured by 1606 mammalian hosts, whereas only 214 species were recorded on 543 bird hosts. Among the latter, only 60 species (about 3% of the total number of flea species) can be considered as specific bird parasites (Medvedev, 1997a, b). The number of flea mammal associations compared with flea bird associations suggests that fleas are mainly parasites of mammals. Parasitism on birds is, thus, secondary, and fleas parasitic on birds are commonly thought to have originated from fleas parasitic on mammals, with this switch occurring at least 16 times during the siphonapteran evolution (Hopkins, 1957; Holland, 1964). In particular, species characteristically parasitic on birds are found in six of 18 siphonapteran families (Holland, 1964), suggesting that the switch from mammalian to avian hosts was related to ecological rather than to phyletic reasons (Beaucournu et al., 2005). Some bird fleas have likely originated from ancestors 9

9 10 Hosts of Siphonaptera parasitic on arboreal mammals (Traub et al., 1983). For example, 18 specimens of five species of rodent fleas were collected from nests of woodpeckers (Haas & Wilson, 1985). Interestingly, the occurrence of mammalian flea species in bird nests increases in winter (Roman & Pichot, 1975). This supports the idea that the switch from mammals to birds is associated with ecological factors. Bird fleas could also switch from burrow-dwelling mammals to burrow-dwelling birds (see Holland (1964) for examples with Actenopsylla and Ornithopsylla) or could exploit birds of prey or owls switching from the mammalian victims of these predators (Scharf, 1998). Traub et al. (1983) suggested that fleas of the family Ceratophyllidae probably evolved from ancestors infesting squirrels (Sciuridae) in the early Eocene (40 45 million years ago). Furthermore, within the family, the adaptation to exploit avian instead of mammalian hosts probably occurred independently in different genera. However, the opposite was also suggested. For example, it is thought that formerly a bird flea Ceratophyllus lunatus reverted to mammals (it is the only species of the genus that does not exploit birds) (Hopkins, 1957). An ability to feed successfully on mammals has been reported for some other bird fleas (e.g. Frontopsylla frontalis: Shevchenko et al., 1976). 2.2 Realized and available hosts Distribution of fleas among mammalian and avian orders in six different regions located on different continents is summarized in Table 2.1. Evenfroma superficial glance at this table, it becomes clear that (a) most flea species exploit mammals (in particular, rodents and, in Australia, marsupials), and (b) there are no data on fleas for many bird and mammal species. This is still true for well-studied regions such as central Europe or North America. The reason for the lack of reports of fleas from some host species or orders can be that either these species do not harbour fleas and/or they merely have not been examined for parasites. In some cases, the former appears to be true (for example, for cetaceans and proboscids), whereas it is impossible to distinguish between the two reasons in other cases. Nevertheless, if the former reason is true, this means that fleas under-use the available pool of host species. Some host species may not be used by fleas due to obvious reasons, such as aquatic life (e.g. pinniped carnivores, although some flea species are associated with penguins and many sea birds: Holland, 1964), absence of pelage (e.g. elephants) and lack of shelters necessary for successful development of pre-imaginal stages of many flea species. Even if a host species possesses a shelter, but its physical conditions (physical structure, bedding material, temperature and/or relative humidity) are unfavourable for pre-imaginal fleas, this host would probably be unsuitable for fleas, as is the case

10 Table 2.1 Number of flea species recorded on birds and mammals belonging to different orders in six geographical regions Number of host species infested Region Class Order with fleas Number of flea species recorded Number of flea species per host species Proportion of host species infested with fleas Number of orders with no flea records Poland Aves Podicipediformes Anseriformes Falconiformes Galliformes Charadriiformes Columbiformes Cuculiformes Piciformes Passeriformes Mammalia Erinaceomorpha Soricomorpha Chiroptera Carnivora Lagomorpha Rodentia Artiodactyla China Aves Anseriformes Falconiformes Charadriiformes Columbiformes Apodiformes Piciformes Passeriformes (cont.)