Class COLLEMBOLA Lubbock, 1870

Collembola

Museums

Museum list

Regional Maps

Class COLLEMBOLA Lubbock, 1870

Springtails

Compiler and date details

April 2018 - ABRS

August 2011 - Penelope Greenslade, Environmental Management, School of Science and Engineering, the University of Ballarat, Ballarat, Victoria 3350, Australia

2007 - updated, Penelope Greenslade with assistance from J. Trueman, Australian National University, Canberra

1994 - Penelope Greenslade, School of Botany and Zoology, Australian National University, Canberra, Australian Capital Territory, Australia

Janssens, F. & Christiansen, K.A. 2011. Class Collembola Lubbock, 1870. pp. 192-194 in Zhang, Z.-Q. (Ed.). Animal biodiversity: An outline of higher-level classification and survey of taxonomic richness. Zootaxa 3148

Introduction

The Collembola are minute arthropods. Together with the Diplura and Protura, they are sometimes grouped as the entognathous Hexapoda. The Collembola and Protura are now considered to be classes within the Arthropoda with a close relationship with firstly, the Insecta and secondly, the Crustacea. Although Collembola are not now considered to be insects (Kristensen 1991), for convenience they are often grouped with them. Molecular studies are now beginning to shed light on the relationships of classes within the Arthropoda but so far have not produced definitive results. However, the monophyly of hexapoda on mophological grounds is accepted by several authors (Cameron et al. 2004; Gullan & Cranston 2000; Luan et al. 2005).

For many years Collembola were considered to be primitive insects on the basis of their having, like insects, a pair of antennae inserted anteriorly on the head, a thorax with three pairs of legs, and a segmented abdomen with appendages. However, they differ from insects in several features: they are soft-bodied; the mouthparts are internal and enclosed in extensions of the labrum, labium and plicae orales; wings, malpighian tubules and metamorphosis are absent; the abdomen is always primitively six segmented; tracheae are rarely present; and there is, primitively, a springing organ or furca ventrally, arising from abdominal segment IV. The furca at rest is flexed forward against the body and held in place by a hooked organ, the tenaculum, inserted ventrally on abdominal segment III. Another unique feature of Collembola is the ventral tube on abdominal segment I, which carries a pair of extrusible sacs that are involved in water and salt regulation and, in some species, functions as an adhesive organ. In common with Protura, 'salivary' glands are present in the head. These glands have ducts that connect to the buccal cavity. There are also a pair of nephridia in the head that empty via a duct to a ventral groove which leads to the ventral tube.

Although directional locomotion is by walking, when disturbed, Collembola can leap considerable distances by rapidly releasing the furca,. After its release, a combination of muscular and hydrostatic forces extends the furca, causing it to hit the ground with force and project the animal a considerable distance into the air. There is some evidence that leaping is partly, at least, directional.

The general anatomy of the Collembola is dealt with in detail by Denis (1949), Schaller (1970), Manton (1972) and Hopkins (1997). Greenslade (1991) briefly describes the morphology of the group in relation to the taxonomy and ecology of the Australian fauna. A series of comprehensive volumes revising and updating Gisin's (1960) Collembolenfauna Europas is being published by the Görlitz Museum in Germany, edited by W. Dunger. Those on the Tullbergiinae, the Symphypleona, the Isotomidae and the Hypogastruridae are already published and the remainder of the families will appear shortly.

BIOLOGY
Collembola are essentially inhabitants of the soil and soil surface and are almost ubiquitous. They are often extremely abundant, even in habitats considered to be inhospitable for most other terrestrial invertebrates. Examples are the marine littoral zone, the surface of still, fresh water, on or under snow in montane and polar environments, and in deserts and caves. Maximum densities in soil and litter have been recorded of over half a million individuals per square metre (Massoud 1971a; P. Greenslade, unpublished results) but, in Australia, densities usually range from 2,000-30,000 per square metre depending on habitat, season and climate. Highest numbers are found in organic rich, humid environments, such as soil and leaf litter of forests, arable land and improved pastures. The lowest densities, if only active individuals are counted, occur in arid environments.

Life histories are simple since there is no metamorphosis. Eggs are generally laid in soil or leaf litter and hatch in a few days. The first instar closely resembles the adult in general appearance, but has fewer setae and a relatively undifferentiated cuticle. The second and third instars become progressively more similar to the adult in chaetotaxy until, after five to seven instars, sexually mature individuals appear. Development time from egg through to adult can take as little as one week in a few species, but is normally three to five weeks but in colder climates, such as polar regions, can take much longer. Adults usually continue to grow and moult throughout life. Fertilisation in most families takes place indirectly: the male deposits a stalked sperm sac for the female to take up. Synchronised aggregation at the time of moulting suffices to ensure that this occurs. In other families, an elaborate dance, instigated by the male, stimulates receptivity of the female. In some taxa, males are highly sexually dimorphic possessing structures that enable them to hold the female, often by entwining antennae.

Collembola feed primarily on micro-organisms such as fungi, bacteria, algae and yeasts. They do not normally feed directly on sterile, dead or live organic matter, and decomposer micro-organisms are necessary to stimulate feeding. The mouthparts in all families, except the Neanuridae, Brachystomellidae and Odontellidae, are similar and consist of a pair of mandibles with apical teeth and a posterior, ridged, molar plate, and maxillae with lamellae and teeth. In the Neanuridae a trend towards external digestion may be associated with simplification of mouthparts. Some neanurid taxa are carnivorous on Protozoa, rotifers and/or nematodes, others are saprophagous. A few species, including the clover springtail (or lucerne flea), Sminthurus viridis Linnaeus (Sminthuridae), feed on live plants, and some Onychiuridae can feed on plant roots. Collembola are able to feed selectively on micro-organisms and some members of the Brachystomellidae are specialised feeders on fungal fruiting bodies, at least during some parts of their life cycle.

There is some evidence to show that Collembola affect soil processes by grazing on micro-organisms. They are believed to act as catalysts in distributing micro-organisms, so contributing to the breakdown of organic matter, increasing the rate of cycling of plant nutrients and enhancing soil fertility and structure. They are also known to feed on fungal root pathogens, such as Rhyzoctonia, and may exert a measure of disease control. Alternatively, a few fungi are known to possess protective mechanisms against predation and are lethal to Collembola. Passive dispersal to new habitats can be by means of wind or water currents. Phoresy is unknown in the group.

The physiology of Collembola is becoming better known (Joosse & Verhoef 1987; Hopkins 1997). For instance, studies on the ability of cold climate species to supercool, and of cryptobiosis in species in xeric habitats have been made. Certain species are able to detoxify insecticides, are fairly immune to radiation damage and can tolerate heavy metal pollution by eliminating chelated metals during ecdysis. Others can regulate water and salt losses over a fairly wide range of conditions. A relatively recent development is that of the use of some Collembola species in ecotoxicology and a considerable literature is developing on the topic (Fountain & Hopkins 2005).

Predators of Collembola include mesostigmatid Acari, Formicidae, Coleoptera, Hemiptera and Araneae. Some beetles have developed elaborate devices for trapping jumping Collembola and certain ants lure Collembola by secreting an attractant chemical. Collembola can avoid capture by leaping, by cryptic colouration or morphology associated with immobility, or by producing a noxious secretion.

Collembola are a convenient and sensitive group for use in environmental assessment because of their ubiquity, short life cycles, abundance and close association to the soil. They are especially valuable in monitoring subtle habitat changes.

BIOGEOGRAPHY
Many Collembola genera have a worldwide distribution and this applies to about half of the 133 genera and subgenera currently known in Australia. Only 22 are endemic, although this is probably an underestimate since the generic classification in some families, particularly the Neanuridae, requires revision. Genera with extra-Australian distributions include a northern group in Australia with oriental affinities and a southern Gondwanan group allied to the New Zealand and South American, even southern African faunas. There are also genera which show relationships with faunas of the southwest Pacific or Madagasca. The biogeography of the Australian arid zone and Tasmanian faunas at the generic level is discussed by Greenslade (1982, 1987). Recent phylogeographic studies in Australia using molecular data are showing that single species may be variable, differences being related to past climate and topographic changes (Garrick et al. 2004). This work has the potential to make a significant contribution to understanding evolutionary processes and paleoclimatic events.

SIZE OF THE ORDER
When Salmon (1964, 1965) published his index of world Collembola 3,424 species and 396 genera were described. Four years later Rapoport (1971) estimated that this number had increased to 3,874 species and 416 genera. Janssens (2007) lists 7,833 species but no doubt many synonyms are included. Hopkins (1997) listed 348 genera and estimated the world fauna at 50,000 species including all undescribed species. In the present Catalogue of the Australian fauna, 133 genera and subgenera and 367 species and subspecies (about 78 introduced) are recorded, an increase of nearly 10% since 1993. There are, in addition, a small number of genera mentioned in the introductions to families that are represented only by undescribed species. Australia has probably just under 5% of the described world species and about a third of the described genera. I estimate, however, that the total number of species of Collembola in Australia is around 2,000. The figure of 1,630 species given by Greenslade (1991) is a conservative estimate.

Recent significant books published on Collembola are, for biology, Hopkins (1997) and, for taxonomy, the Synopses of Palaearctic Faunas (ed. Dunger, 1999–2004).

HISTORICAL
Lubbock (1899) recorded the first two Collembola from Australia, Anoura dendyi and Anoura tasmaniae, both from Tasmania. The next taxonomic studies of Australian Collembola published were the pioneer works of Schött (1917) and Womersley (1932-1942). Schött worked in Sweden on a collection of Collembola made mainly in northern Queensland during the Mjöberg expedition of 1911 to 1913. The greater part of this material is deposited in the Naturhistoriske Riksmuseum, Stockholm. A smaller representative collection has been deposited in the collection of the South Australian Museum, Adelaide. Part of Schött's Australian material is also in the National Museum of New Zealand, Wellington.

Womersley was brought to Australia from England by R.J. Tillyard of CSIRO to work on the clover springtail in Western Australia. Two years later he was appointed entomologist in the South Australian Museum, where he worked and published on the taxonomy of Australian Collembola and mites until his death in 1962. Womersley's collection is deposited in institutions in Perth, Adelaide and Canberra, with most of his material in the South Australian Museum.

After Womersley, little work was carried out on the taxonomy of the native collembolan fauna until the 1970s. R. Yosii (who in all papers published from 1980 to the present day, changed the spelling of his name to Yoshii) published on the neanurid Collembola in the South Australian Museum (Yosii 1966) but his other work has concentrated on other parts of the western Pacific region and southeast Asia. M.M.H. Wallace contributed to the chapter on entognathous hexapods in The Insects of Australia (Wallace & Mackerras 1970) and to its Supplement (Wallace 1974) and published on the ecology of the pest species, Sminthurus viridis.

The major contributor in recent years has been Penelope Greenslade whose interest in taxonomy of Australian Collembola developed from ecological studies in the late 1960s on soil invertebrates. She has published extensively on the taxonomy and ecology of Collembola in Australia, the Australian subantarctic islands and Antarctica, often with other authors.

Classification
The classification used here is based on Janssens & Christiansen (2011). Other previous references include: The Insects of Australia (Greenslade 1991) and Dunger (1994-2004). The basis is that of Gisin (1960) and brought up to date by additions from Betsch (1980), Deharveng (2004), Massoud (1967, 1971b), Richards (1968) and Szeptycki (1979). From a total of 30 families, 20 are recognised in Australia. No fossil families have been found in Australia nor have four extant monobasic families. Subfamilies have not been used as they are not well supported.

Types
For information on many of the locations of depositions of types, I am indebted to the valuable catalogues of Weidner (1962), Mari Mutt (1978) and Vilkamaa (1988) from the Hamburg Museum, the Illinois Natural History Survey and the Zoological Museum, University of Helsinki, respectively and more recent catalogues as listed on the World List of Collembola (www.collembola.org). From 1973 to the present day, I have visited a number of other institutions known to have holdings of Australian Collembola, both within Australia and overseas, and listed material. Depositions of types in institutions other than those visited have been verified by correspondence except where indicated. In many cases the depositions given by Salmon (1964, 1965) in his index are not correct and so all have been checked for accuracy and his references to 'hypotype' have been omitted as they have no status in formal taxonomy. When necessary, Womersley's alcohol collection has been examined for the presence of type material.

Type Designation
Early workers often did not designate holotypes or type localities and I have designated lectotypes and defined type localities here wherever it seemed advisable. Familiarity with Womersley's and Schött's collections has enabled me to solve a number of problems concerning the real identity of species. I have visited many of Womersley's type localities and other collecting sites and made new collections.

Schött Material
Schött neither labelled types of any kind nor published holotype designations and the labels on his slides give the locality simply as 'Australia'. Much of Schött's material was preserved in alcohol and in the 1950's Salmon mounted some of this material and selected lectotypes and paralectotypes. Salmon's selections have not all been accepted, because, in his catalogue (Salmon 1964, 1965), he did not publish his designations except by mentioning their deposition which was not always correct. This is considered here to be an inadequate lectotype designation. Where a species was collected from one locality only (and hence there is no problem with the identity of the type locality) and if the slide is in good condition, then a lectotype has been selected from Schött's mounted specimens in preference to that selected by Salmon. In many cases, Salmon's mounts are not satisfactory. He used polyvinyl lactophenol as a mounting medium, which tends to produce shrivelled specimens, and rarely cleared his specimens, and they are very difficult to demount. Consequently, the lectotype has been selected in preference from Schött's rather than from Salmon's mounted specimens.

Womersley Material
Womersley's type series occasionally included several species, or even species belonging to different tribes and families as was noted by Lawrence (1968). Clearly, lectotype designation is indicated wherever this might be a possibility. Lectotypes have not been designated in several cases, however, when there is no doubt that a syntype series is conspecific, i.e. when it is from the same locality collection, date and habitat, is mounted on the same slide, and specimens are morphologically identical.

The species Womersley described fall into six groups:

1. Type labelled and designation published. This is usually for species based on a single specimen.
2. Cotypes labelled and designation published.
3. Cotypes labelled but designation not published.
4. Syntypes labelled and designation published.
5. No types of any kind labelled nor designations published where a species is described from a single locality.
6. No types of any kind labelled nor designations published, where a species is described, the material is listed from several localities and dates.

For the sixth group, it was necessary to select a type locality as well as type specimens and here the collection of the earliest date has been selected if the specimens are in good enough condition. For the fourth and fifth categories, the lectotype was selected from the syntype series listed by Womersley in the published description as it is assumed to be the material he examined. Lectotypes have been selected from Womersley material when the syntype series has been split between institutions. As noted earlier, Womersley deposited specimens in a number of Australian and a few overseas institutions. There is evidence that one of Womersley's collections deposited in the University of Western Australia was later lost (Murphy 1971). The species concerned, however, on which evidence for this was based, was probably listed under the wrong date in Womersley's original description. This is apparent from the Western Australian Museum's register where dates of accession and collection are recorded (T. Houston, in litt.). I have no evidence of other lost material except for a few small collections usually consisting of single specimens.

Distribution
An attempt has been made to include all published distributional records in the Catalogue, as well as some records of reliably identified species which have not been published before, but the individual references to these records are not always given. The distribution records are based on identified material in the SAMA collection. Where possible, earlier records have been verified, or listed here as errors if found to be incorrect. As a result, some records of Womerley's are not included if the specimens were originally misidentified. In some cases, verification of Womersley's determinations has not been possible due to the poor state of the specimens or the fact that they could not be found. The distributions of very few species are well known in Australia.

Ecology
The detailed ecology of the vast majority of species is unknown and the descriptors mainly used here have been the terms terrestrial, littoral or aquatic. Even the feeding habits of some of the best known species are uncertain and so have only been mentioned if known with certainty. These tend to be species of cosmopolitan distribution and apparent economic importance, such as Sminthurus viridis and Onychiuridae.