First records of a leptestherid clam shrimp in Australia ( Crustacea , Spinicaudata , Leptestheriidae , Eoleptestheria )

Eoleptestheria ticinensis, a highly variable Eurasian species, was collected from three widely separated sites in northern Australia. Each population is described and compared with the eight described species of Eoleptestheria, now all synonyms of E. ticinensis. It is postulated that the Australian occurrences of these clam shrimps are initiated or maintained by dispersal due to migrating birds from China.


Introduction
Of the approximately 150 species of clam shrimps in the world (Brendonck et al. 2008), 31 are reported from Australia (Richter and Timms 2005; Timms in press; Timms and Richter in press).Th ese are divided among the families Lynceidae (two species) Limnadiidae (17), Cyzicidae (11) and the Cyclestheriidae (1), but none in the Leptestheriidae.Of the later, Garcia and Pereira (2003) list 34 species worldwide, Brtek (1997) lists 35 valid species and Brendonck et al. (2008) count about 37 species.It is diffi cult to know how many species (and genera) there are because of wide variability within and between populations (e.g.Straškraba 1965) and because some authors synonymise species without argument (e.g.Brtek 1997;Naganawa 1999) and still others do not accept some genera (e.g.Brtek 1997;Dumont and Negrea 2002).
Herein, I report the fi rst Australian records of leptestheriid clam shrimp, which on current understanding belongs to the genus Eoleptestheria.
Leptestheriid clam shrimps (Fig. 1) are characterised by having an elongated delicate carapace with numerous growth lines, head lacking a pyriforme frontal organ, a body of 22-32 segments, a rostral spine in both sexes, females with dorsal extensions to hold the eggs on thoracopods 10 and 11 or maybe up to number 15, a telson with numerous (>40) subequal fi ne dorsal spines, and a caudal furca also with numerous (>30) subequal fi ne spines (adapted from Dumont and Negrea 2002).
It is the purpose of this paper to describe three populations of Eoleptestheria recently found in Australian and to note their relationships.

Methods
Measurements were made using a stereomicroscope and a template placed under the specimens and marked in half millimetres (accurate to ±0.25 mm).Drawings were made with the aid of an ocular drawing tube.Th oracopod terminology is after McLaughlin (1980) and Ferrari and Grygier (2003).In the drawings of the fi fth thoracopod not all setae are shown.Classifi cation follows Martin and Davis (2001), and synonymy Straškraba (1965) and Naganawa (1999).Specimens were sourced from the Australian Museum, Sydney (AM), National Museum of Victoria, Melbourne (NMV), and Th e Department of Environment and Conservation Research Laboratories, Woodvale, Western Australia (DEC).
Description of female.Carapace (Fig 2A) 6-7 mm by 3.3-4.6mm, oval, but dorsally centrally humped, a dorsoposterior angle and no dorsoanterior angle and broadly rounded both ventroanteriorly and ventroposteriorly.Umbo only slightly developed and associated with a small protuberance anteriodorsally.Growth lines 15-22, unevenly spaced, with tighter spacing marginally and interstices between lines granular.Carapace thin, semitransparent and usually brown in colour, especially in the older areas.
Head (Fig 2B ) with a rounded occipital condyle and well separated from the trunk.Conspicuous ocular tubercle and large winged fornices of triangular rostrum terminating in an anteriorly directed rostral spine, about one-third length of the rostrum.Ocellus oval and within rostrum, usually in a ventrobasal position.
First antennae about 1.5 times the length of the base of the second antenna and with 10-13 lobes, each with 2-4 dorsal setae.Second antenna base (Fig. 2D) with about 12 rows of dorsal spines and bearing two rami with 13-14 antennomeres each.Each antennomere with 3-8 dorsal spines and 2-6 ventral setae, all evenly spaced except terminal on basal antennomeres.Flagellum middle antennomeres with most spines and setae, while terminal and basal antennomeres with least spines and setae.
Trunk segments 24 (Fig. 1).Posteriormost 14 segments (Fig. 2E,F), sometimes fewer, armed dorsally with numerous spines inserted on a common broad base, triangular in central segments of the array, pedunculate in the 3-4 most anterior segments.Segments around 17 th (i.e.seventh last segment) with strongest and most (typically 13) spines, and those anterior to and posterior to this segment with fewer spines, e.g. 5 spines on third last segment.Fifth thoracopod (Fig. 3) with fi ve endites on the medial surface, each about the same size.Also a comb-like discoid lobe (Ferriera and Grygier 2003) with many closely packed setae basally at right angles to the fi rst endite.First endite with about 20 anterior and posterior setae, while remainder with about 12 anterior and posterior setae.All setae two segmented, but only the posterior setae plumose.Anterior setae 2-3 times longer on fi rst endite than on endites 2-5.Distal posterior setae tend to be longer than proximal setae on each endite.Fifth endite with a long unsegmented palp with few setae and many setules apically.Sixth endite (= endopod of some authors) also elongated but longer and wider than the palp and with more setae than palp, more medially than externally.Bipolar exopod with distal part (the fl abellum) long and fi nger-like and a similarly shaped but smaller proximal extension.Exopod clothed with a few setae similar in structure to the posterior setae of the endites.Th ese setae limited to apex region of fl abellum and middle external edge.Epipodite fi nger-like, about half the length of the proximal exopod.A triangular lamellar (cf Marinček 1978), edged with setae, protruding from base of fl abellum.Gross examination of other thoracopods reveal slightly diff erent proportions of some components, especially the exopod.Eleventh and twelfth pairs with fl abellum sheathed and carrying eggs.Palp of fi fth endopod of third thoracopod one segmented.
Telson (Fig. 2C) with a concave dorsal surface with about 40 (39-43) small spines of similar size throughout.Paired telsonic setae (fi laments) inserted on slight mound between the fi rst and second denticles.Caudal furca even curved, a little shorter than the dorsal surface of the telson, and with about 40 small, subequal spines arranged on a curved line commencing on medial surface basally but on dorsal surface apically.Th ese spines slightly smaller than the telsonic spines.

Kuranda population
Fig. 4 Material examined.5 females, Queensland, via Cairns, Kuranda, Mrs Armitage, 27 February 2006, NMV J93994.Description of female.Carapace (Fig. 4A) larger 9.0-9.8 by 5.8-6.4mm with more growth lines (26-32), but same shape as in the Toomaroo material and with same number of body segments (24).Head (Fig. 4C) as in Toomaroo material.Dorsal armature similar to that in the Toomaroo material, but with slightly more segments (15) involved.Similar arrangement and number of dorsal spines, ie those on central segments on a quasiequilateral triangular base and number up to 13 per segment, those most posterior segments number fewer (3-7) and on a slightly protruding triangular base, and the most anterior on a column.
First and second antenna similar to those of the Toomaroo material, but with slightly diff erent numbers (9-12 lobes on antenna 1 and 12-13 rami of antenna II).
Telson (Fig. 4D) with more spines (ca 50-60) and more caudal furca spines (ca 50) than in the Toomaroo material, but their arrangement similar, i.e. equal sizes and in a curved line on the claw, basially mesodorsal and apically dorsal.Description of female.Carapace (Fig. 4B) 5.9 by 3.9 mm, slightly humped middorsally, with rounded dorsoanterior and dorsoposterior corners, and 34 closely spaced growth lines.Areas between growth lines with small rounded protuberances tending to lie, between outer growth rings, in meridian lines.Umbo most protruding of the three populations.

Benmore Well clay pan population
Head (Fig. 4E) as in Toomaroo population, but with winged fornices unevenly developed, widest centrally.
First and second antenna similar to those of the Toomaroo material, but with slightly diff erent numbers of lobes and rami (9 lobes on fi rst antenna and length only just longer than peduncle of second antenna and 12-13 rami on second antenna).
Body segments 23.Posteriormost 12 segments with dorsal spines; anterior most and posteriormost with spines on a peduncle, but central segments with spines on a fl at triangular base.Up to 13 spines on segments around the seventh posterior most, descreasing anteriorly and posteriorly.
Limbs unstudied, but segments 1-9 with long exopods (fl abella).Segments 10 and 11, on one side only of the only specimen, with sheathed tubular extensions carrying eggs.
Telson (Fig. 4F) as in Toomaroo material; about 42 dorsal spines and about 30 spines on the caudal claw.Spines subequal, those on caudal furca in a weakly row, basally mesodorsal and apically dorsal.segments sheathed for carrying eggs, but in the Benmore Well clay pan population it is the 10 th and 11 th segments that are so modifi ed.(f ) Th e shape of the telson and caudal claw is similar in all three populations, but the Kuranda population has far more spines than the other two (50-60 on telson cf ca 40; 50 on caudal furca cf ca 30).Th oracopods are not thoroughly studied in Eoleptestheria, and the present study based on few specimens does little to redress the situation.It is clear however that while thoracopods conform to the generalised spinicaudatan structure (McLaughlin 1990;Ferrari and Grygier 2003), they have a character apparently unique to Eoleptestheria, a triangular lamella at the base of the distal exopodite (Fig. 2).Marinček (1978) thought that the basal discoid lobe was also unique to Eoleptestheria, but at least some cyzicids have it too (Ferreir and Grygier 2003).Th e triangular lamella is illustrated for the recently described Chinese species (Hu 1986;Shu et al. 1990;Zheng and Hu 1992), but only in E. dongpingensis is the discoid lobe shown (Hu 1986, Fig 13a).Th e component parts of the thoracopods apparently vary in relative size between individuals and certainly between thoracopods (Marinček 1978).Of most interest is the segmentation and relative size of the palp of the fi fth endite (termed the palpus endopoditalis by many authors).It is one to three segmented but insuffi cient data are available on its variability in segment number and relative size, so that its use in species or population discrimination is presently limited.
Variations between the three populations are not systematic.Smaller size in the Toomaroo population (associated perhaps with their youth-see later) may explain the lower number of growth lines and telsonic and caudal claw spines, but other diff erences seem to be random.Similarities between the three groups far exceed their minor diff erences, so it is concluded they are all belong to the one species of Eoleptestheria.But is this species new or can it be accommodated within a described species?Th e number of valid species of Eoleptestheria is disputed: Straškraba (1965) synonymised three European species and the only then known Chinese species into E. ticinensis (Balsamo-Crivelli 1859), Brtek (1997) accepted 4 of 8 species he listed, but Naganawa (1999) thinks, without giving any analyses, there is only one (but curiously omits E. sangziensis from his list).I am also of the opinion that there is only one widespread and variable species of Eoleptestheria.Evidence for this is provided fi rstly by Straškraba (1965) in his study of the variability of E. ticinensis in Czechoslovakia and on the overlap in characteristics of this species with those of E. inopinata, E. variabilis and E. chinensis ( Table 1.Eoleptestheria spinosa, described after Straškraba study, also lies within the range of variability of the European material, thus confi rming Naganawa's synonymy of it with E. ticinensis.Shu et al. (1990) give a comparative table supposedly separating E. chinensis, E. dongpinensis and E. yanchowensis and similarly Zhang and Hu (1992) give a table separating their E. sanziensis from E. dongpingensis, but the supposed diff erences are minor in all the Chinese forms and could be due to variability of characters in separate populations.Th is argument is strengthened by Petrov and Marinček's (1995) study of age induced variability in the closely related Leptestheria saetosa Marinček and Petrov.Th ese authors show that many of the characters used in the separating of the various species of Eoleptestheria change with age, including proportions of the carapace, presence or absence of marginal hairs, shape of rostrum and occipital condyle, number of trunk segments equipped with dorsal spines, number of telsonic spines, and segmentation in the palp of the fi fth endite.Also it is well known that carapace size and number of growth lines are variable and there is even some variation in number of trunk segments (Straškraba 1965;Richter and Timms 2005).Because of this, not one of the six diff erences between E. sangziensis and E. dongpingensis given by Zhang and Hu (1992) is signifi cant, thus invaliding E. sangziensis as a separate species.
If there is but one variable species of Eoleptestheria, are the populations in Australia suffi ciently diff erent to be given species rank?Most of their characteristics (Table 1) are accommodated within the range of E. ticinensis s.l., except for the lower number of armed trunk segments, and lower number of lobes and rami on fi rst and second antennae respectively.Also in two of the three populations there are more spines of the dorsum of the trunk segments than in overseas populations.All four of these features could be an expression of even wider variation (cf.Straškraba 1965) or of change with age (Petrov and Marinček 1995), or be due to founder eff ects associated with a small number of dispersing eggs (Provine 2004).Th e most parsimonious conclusion is to consider the Australian populations as further variations within the E. ticinensis complex, rather than a separate species.
Eoleptestheria is rare in Australia, though admittedly all three collections are from remote regions, and therefore not likely to be commonly encountered.However, the Lake Toomaroo population has occurred only once in a long term study (so far 15 years) of the lake (Mark Handley, pers. comm.).Two of the three occurrences are sites in far north of Australia and as such are likely to be visited by returning migrating birds from the northern hemisphere on arrival or soon afterwards.Th ese data suggest   the possibility this clam shrimp is not an integral part of the Australian fauna and is occasionally being introduced by migrating birds from overseas, possibly China where Eoleptestheria is known to occur (op.cit.).Th ere is at least one known occurrence of migration of lake fauna the other direction: the widespread Australian copepod Boeckella triarticulata in Mongolia (Bayly 1979).It is postulated that arriving birds deposit egg-laden faeces (Procter et al. 1967;Sánchez et al. 2007;Green et al. 2008) and so introduce eggs of Chinese Eoleptostheria ticinensis s.l.. Similarly, Th iery and Pont (1987) note that three southern European populations of E. ticinensis could have been introduced by migrating birds from central Europe.In Australia, there could be just one founding population and then subsequent dispersal, or two or all three populations could be founders.Th e most likely population to result from secondary dispersal in that in Toomaroo, given its more southerly and inland location.Th is theory of dispersal by birds from China is enhanced by the apparent absence of Eoleptestheria ticinensis in southeast Asia, but this could be due to lack of collecting there or lack of suitable habitat.Finally, it is signifi cant that almost all other large branchiopods in Australia are endemic due to their isolation in remote Australia; the only known exceptions so far are the circumtropical Cyclestheria hislopi (Timms 1986) and now Eoleptestheria ticinensis.
The Toomaroo population was young (<2 weeks old) when collected and did not survive because of fish predation (Timms & Handley 2008).It was however old enough to have reproduced, in keeping with the known short life cycle of E. ticinensis (Popović & Gottstein-Matočec 2006).It will be interesting in years to come to see if this species reappears in Lake Toomaroo as a self-maintaining population.
Much has yet to be learnt on the diversity of Australian clam shrimps, but an outline is available in Richter & Timms (2005).Th eir key to genera needs modifi cation to include Eoleptestheria and updated to include other recent discoveries (see below).Circumtropical Cyclestheria hislopi occurs in northern Australia (Timms, 1986).

Key to Genera of Clam Shrimps in
5 Th e wisespread Eurasian E. ticinensis s.l.occurs in northern Australia as reported here. 6 Two species of Eocyzicus separable by a key in Timms and Richter (in press).In other regions Cyzicus would key out here, but this genus is not in Australia, despite many references to it in books (e.g.Williams, 1980).

Figure 1 .
Figure 1.Lateral view of a whole female of Eoleptestheria ticinensis from Lake Toomaroo, Queensland.Drawing by Jane McRae.

Figure 2 .Figure 3 .
Figure 2. A female of Eoleptestheria ticinensis from Lake Toomaroo.A carapace B head C telson D second antenna E dorsal spination on segments, 11, 12, 17 and 20 F frontal view of spines and their common triangular base on segment 18. Scale bars 1 mm.

Figure 4 .
Figure 4. Females of Eoleptestheria ticinensis from Kuranda, Queensland (A,C,D) and from Benmore Well clay pan, Western Australia (B,E,F).A, B carapaces, growth lines not shown as they are too numerous C, E heads D, F telsons.Scale bars 1 mm.