Eulimnadia palustera, Timms, 2015

Timms, Brian V., 2019, A redescription of Eulimnadia rivolensis (Brady, 1886) (Branchiopoda: Spinicaudata: Limnadiiidae), and its transfer to Paralimnadia, Memoirs of Museum Victoria 78, pp. 57-64 : 63

Synonymy of E. palustera

This species was originally assigned to Eulimnadia on the sole criterion of an apparent spine beneath the cercopod base ( Timms, 2015). However, this spine is not a typical subcercopod spine of most Eulimnadia but a rather sharp triangular ventroposterior corner of the telson. Hence, an assignment to Paralimnadia is necessary. Furthermore, three other features suggest placement in Paralimnadia : a cercopod with a spine approximately midlength and not at about 80% of its length, 13 antennomeres rather than about 8, and a sex ratio approximating 1:1, all generally (but not absolutely) indicating Paralimnadia ( Timms, 2016a, 2016b).

Given the placement of P. palustera within Paralimnadia , its eggs are identical with those of P. rivolensis being astroform with 14–20 projections subtended by 3–8 sharp-edged grooves (fig. 4). Egg morphology has proved to be the most reliable character separating species within Eulimnadia (Belk, 1998; Rabet, 2010; Rogers et al., 2012; Timms, 2016a) and Paralimnadia ( Timms, 2016b) . The next most reliable species indicator in both genera is the nature of the cercopod setae. Both P. palustera and P. rivolensis have about 8 medium length (i.e. 1–2× cercopod diameter) setae (cf. fig. 6 in Timms, 2015 and figs 1–3). Again, both species have about 21 telsonic spines, although spacing is different in the two species. In P. rivolensis , all are evenly sized and spaced, except for the first three, which are larger and more spaced. In P. palustera , the telsonic spines are mixed in size (cf. fig 6 in Timms 2015 and figs 1–3). Two characters generally of poor differentiating ability are the first antennae and rostrum, although in these two species, there are only minor differences (cf. fig 6 in Timms and figs 1–3).

The claspers are somewhat different between the two species. P. palustera has a distinct hamulus medially on the hand (endite IV), while P. rivolensis has just a slight swelling there. The palps are variable, with 3 palpomeres in the paralectotype of P. rivolensis , but only 2 indistinct ones in most other specimens examined. P. palustera generally has 3 palpomeres but may have the second division indistinct or incomplete. Sometimes there are spines at palpomere junction 1–2 in P. rivolensis . Similar variability has sometimes been observed in a few other Paralimnadia species ( Timms 2016b).

Distribution. South-western Western Australia, south-eastern South Australia, southern Victoria and Tasmania. There is a single record from central Australia, which is difficult to accept considering the prominent maritime distribution across southern Australia. It has not been collected in Victoria since 1910, its habitat in the swamps of eastern Port Philip Bay being drained and urbanised in the early 1900s. Widespread drainage in the south-east of South Australia seems to have denied it habitat there. The most recent collection from near the type locality is dated 1975, and my expeditions there in the spring of 2010 and winter of 2016 were unsuccessful. Sites in central Tasmania seem (as of March 2018) also to be drained, so that perhaps it now only occurs in refuges of Flinders Island, Kangaroo Island and south-western Western Australia.