Order SIPHONAPTERA

Siphonaptera

Museums

Museum list

Regional Maps

Order SIPHONAPTERA

Fleas

Compiler and date details

17 November 1998 - Andrew A. Calder, CSIRO Entomology, Canberra, Australian Capital Territory, Australia

Introduction

The Siphonaptera, or fleas as they are commonly known, are a relatively small order of parasitic insects. The order comprises some 2525 described species and subspecies worldwide (Lewis 1998) in 16 recognised families. The Australian flea fauna exhibits a high degree of endemism, a feature shared with the mammals with which most species are associated. Ten of the species recognised in Australia are thought to have been introduced by humans; most of these are now cosmopolitan due to their association with humans or commensal animals. Of the 80 indigenous species or subspecies, 72 are endemic (Dunnet & Mardon 1991).

The familial classification adopted here is essentially that proposed by Hopkins & Rothschild (1962, 1966) and Mardon (1981). An alternative, significantly different classification of the Australian species was proposed by Smit (1982). Medvedev (1994) has proposed a classification of Siphonaptera families and a scheme of their phylogenetic relationships. As the current arrangement of the suprageneric categories of Siphonaptera is in a state of flux, the families and genera of fleas are ordered alphabetically. Nine families, 34 genera and 91 species and subspecies are listed for the Australian fauna, including one species described from Christmas Island (Indian Ocean), five species from Macquarie Island and a monotypic genus from mainland Antarctica. Endemic species are recorded in seven of the nine families present in Australia.

Fleas are highly specialised, small (1–10 mm in length), wingless, laterally compressed insects. The adults are bloodsucking ectoparasites of mammals and birds; their mouthparts are modified for piercing and sucking. The body is strongly sclerotised and covered with posteriorly directed setae and spines, sometimes arranged in combs, that facilitate progress through the hair or feathers of the host. They have long legs which enable them to jump in a highly characteristic manner. The short antennae lie in deep grooves on each side of the head; in males only of nearly all species they are erectile and are used to grasp the venter of the female during copulation (Smit 1982). The abdomen consists of 11 segments; segment X in both sexes bears dorsally a sharply defined pincushion-like area (sensilium) containing a number of sensory structures (trichobothria). The larvae are apodous and vermiform, usually living in refuse in the lairs or nests of their hosts. These maggot-like larvae are not parasitic. They have chewing mouthparts and feed on a variety of organic matter. One species, Uropsylla tasmanica Rothschild, however, whose larvae live in the skin of dasyurids in Tasmania and Victoria, appears to be truly parasitic (Dunnet & Mardon 1991). Most life stages show remarkable biological adaptations in that all stages, except perhaps the egg, can withstand unfavourable environmental conditions for remarkably long periods (Dunnet & Mardon 1991).

Many flea species are significant as transmitters of disease from mammalian hosts to humans, especially of plague from rodents. The Oriental Rat Flea, Xenopsylla cheopis (Rothschild), is most frequently implicated in transmission of plague as well as transmission of murine typhus (Rickettsia) and tularemia (Francisella). Australia has been substantially free of bubonic plague with only occasional outbreaks (none recently) at seaports. Both sylvatic plague and tularemia are unknown in Australia; murine typhus is endemic in country towns (Dunnet & Mardon 1991).

In addition to the Human Flea, Pulex irritans Linnaeus, several other species may bite humans as well as their primary hosts, which are usually domestic mammals, rodents or birds e.g. Ctenocephalides felis (Bouché), Ceratophyllus gallinae (Schrank), Nosopsyllus fasciatus (Bosc), Xenopsylla cheopis. Such species can become serious domestic pests and many people become sensitised to their bites (Freeman 1980). As fleas are so structurally modified for their parasitic way of life (Snodgrass 1946), their relationships with other insects are somewhat obscure (Dunnet & Mardon 1991). Kukolová-Peck (1991) considers that the group probably emerged from the mecopteroid stem group in the early Carboniferous and includes them in the Mecopterida. Similarly, Kristensen (1991) assigns the Siphonaptera to the Section Mecopterida (=Panorpoid orders) which also includes the Mecoptera and Diptera as well as the Trichoptera and Lepidoptera. Two fleas from the Cretaceous Koonwarra Fossil Bed were recorded by Riek (1970) although Willmann (1981) expressed considerable doubt as to their identity. Jell & Duncan (1986) accepted Riek's identifications and proposed two new genera Tarwinia and Niwratia for the species. Niwratia elongata Jell & Duncan (family uncertain) has a typical siphonapteran body and pronotal comb, but has long slender legs. Tarwinia australis Jell & Duncan and two undescribed Pulicidae also from the same deposits, and also with long slender legs, were considered fleas.

The early published work on Australian fleas is scattered. It was not until the work of Dunnet and co-workers that all previous literature on Australian fleas was brought together comprehensively in a monograph on Australian fleas (Dunnet & Mardon 1974).

Very little work on the Australian fauna was undertaken during the last century. The first described Australian flea was Pulex echidnae (Denny 1843) which now constitutes the monotypic genus Bradiopsylla; it is a specific parasite of the echidna. The next references to Australian species were by Olliff (1886) and Skuse (1893). Olliff described the new genus Echidnophaga with the species ambulans, also collected from the monotreme Tachyglossus aculeatus; Skuse erected the genus Stephanocircus for the species dasyuri, a very common and well known species. At the beginning of the twentieth century, Rainbow (1905) described a single species Ceratophyllus rothschildi now known as Acanthopsylla rothschildi rothschildi.

Only when Charles Rothschild became interested in fleas and established the Rothschild Collection of fleas of the world at Tring, was the Australian fauna worked on in a systematic way. From 1900 to 1923, Rothschild published 14 papers as sole author, and another eight in collaboration with Karl Jordan, which describe or refer to Australian fleas. After Rothschild's death, Jordan published a futher 10 papers from 1925 to 1946, which also referred to the Australian fauna. Ferguson (1923) reviewed the state of siphonapteran knowledge; his collection, previously in the School of Tropical Health and Medicine, University of Sydney, is now housed in the Australian National Insect Collection, Canberra. Miriam Rothschild (1936) published on a collection of fleas from Western Australia describing six new species in three families. Smit (1957, 1958, 1962, 1973, 1975) described four species of Australian fleas, including the nest flea Idilla caelebs.

The most recent work on Australian fleas is by four workers, particularly Dunnet and co-workers. Holland (1971a, 1971b, 1971c) described several pygiopsyllid fleas and erected the genus Austropsylla for a flea from Tasmania. Mardon & Dunnet (1971a, 1971b, 1972a, 1972b) described numerous pulicid and pygiopsyllid fleas. Mardon (1973) described a new species of flea from WA, and Dunnet & Mardon (1973) described a new monotypic genus of bat flea, Coorilla, from NSW. Traub & Dunnet (1973) revised the Australian Stephanocircus and erected a new genus Coronapsylla for a flea known only from Antechinus species. Holland (1964), Rothschild (1975) and Traub & Starke (1980) have reviewed the phylogeny, classification, host relations, physiology, and medical and veterinary importance of fleas and give valuable leads into the biological literature.

Acknowledgements

Final preparation and editing of the flea section of the Catalogue was conducted in the Division of Entomology, CSIRO, Canberra, with support from the Australian Biological Resources Study (ABRS). Use of the Division's resources and facilities are gratefully acknowledged. I am very grateful to Dr D.W. Walton who entered a large proportion of the entries in initial draft files which I then checked and edited.

I would particularly like to thank the staff at the CSIRO Black Mountain Library for their patience and help above and beyond their call to duty in finding many obscure references and processing innumerable requests for inter-library loans. The BMNH Library kindly provided a photo-copy of Enderlein's 1903 work. Dr Brian Pitkin of the Department of Entomology, BMNH, kindly provided a copy of Smit & Wright (1978) and Dr Brian Cooke, formerly of the Animal and Plant Control Commission, South Australia, is thanked for providing a map of the release localities for Xenopsylla cunicularis Smit from the Commission's database.

I am also indebted to Dr A. Wells of ABRS for editorial advice.

Database Notes

The information on the Australian Faunal Directory site for the Siphonaptera is derived from the Zoological Catalogue of Australia database compiled on the Platypus software program. It incorporates changes made to the work published on 15 August as (Calder, A.A., 1996)

In preparation of this work, all publications containing original descriptions have been seen. In only one case was any difficulty encountered in verifying the date of publication of a name. Pulex kerguelensis Taschenberg, 1880 was described twice in the same year by the same author (1880a, 1880b). It has proved impossible to establish the actual date of publication of each description. Thus, I have relied on the entry in the Zoological Record for 1880 (Insecta p. 194) and have accepted the description in Taschenberg's monograph (1880a) as the first published and that in the Notes from the Leyden Museum (1880b) as the second.

Hopkins & Rothschild (1956) in the second volume of the Catalogue of the Rothschild Collection of Fleas stated that 'Although the holotype concept was accepted by Jordan and Rothschild from the beginning of their work on fleas, they did not in their earliest papers publish any statement or clear implication that one particular specimen was regarded as the type to the exclusion of all others, and in consequence some of the specimens we have listed as holotypes are actually lectotypes. In such instances our listing of specimens as holotypes must be regarded as a selection of lectotypes'. However, Smit & Wright (1978) in their publication which lists the primary type specimens of Siphonaptera in the British Museum (Natural History) (BMNH—now the Natural History Museum) deemed it necessary to designate lectotypes of these same specimens. Since their work was governed by the second edition of the International Code of Zoological Nomenclature in which Article 74(c) states: 'Lectotypes must not be designated collectively by a general statement; each designation must be made specifically for an individual nominal species, and must have as its object the definition of that species'. Thus Smit & Wright (1978) considered that the authors of the Rothschild Flea Catalogue had not designated lectotypes validly by listing specimens as holotypes, and further the practice did not indicate whether those specimens listed were in fact holotypes or lectotypes. I concur with this view, and these type specimens that are listed in the Catalogue are considered to be lectotypes designated by Smit and Wright rather than lectotypification by inference of holotype as now allowed in Article 74(b) of the latest and third edition of the International Code of Zoological Nomenclature.

Common names of fleas have been taken from Naumann (1993). The scientific and common names of mammalian and bird hosts in the Ecology section follow the Census of Australian Vertebrate Animals (CAVS). A list of host species and fleas recorded from them has been compiled from the host data for each valid species and is given in Appendix IV.

The location and status of type specimens has been checked only in the Australian National Insect Collection (ANIC). The list prepared by Spratt (1983) was useful for locating other siphonapteran types held in Australian Institutions; these have not been checked recently.

Limital Area

Distribution data in the Directory is by political and geographic region descriptors and serves as a guide to the distribution of a taxon. For details of a taxon's distribution, the reader should consult the cited references (if any) at genus and species levels.

Australia is defined as including Lord Howe Is., Norfolk Is., Cocos (Keeling) Ils, Christmas Is., Ashmore and Cartier Ils, Macquarie Is., Australian Antarctic Territory, Heard and McDonald Ils, and the waters associated with these land areas of Australian political responsibility. Political areas include the adjacent waters.

Terrestrial geographical terms are based on the drainage systems of continental Australia, while marine terms are self explanatory except as follows: the boundary between the coastal and oceanic zones is the 200 m contour; the Arafura Sea extends from Cape York to 124 DEG E; and the boundary between the Tasman and Coral Seas is considered to be the latitude of Fraser Island, also regarded as the southern terminus of the Great Barrier Reef.

Distribution records, if any, outside of these areas are listed as extralimital. The distribution descriptors for each species are collated to genus level. Users are advised that extralimital distribution for some taxa may not be complete.