Molecular and morphological assessment of Delma australis Kluge (Squamata: Pygopodidae), with a description of a new species from the biodiversity ‘ hotspot’ of southwestern Western Australia Author Maryan, Brad Author Brennan, Ian G. Author Adams, Mark Author Aplin, Ken P. text Zootaxa 2015 3946 3 301 330 journal article 10.11646/zootaxa.3946.3.1 21a54f6c-7b8e-4da6-81ac-05703db5bbf1 1175-5326 244607 91C0214A-7A3D-4DEC-9D24-7B10928AC127 Delma hebesa sp. nov. Heath Delma ( Figs. 9 , 10 , 11) Holotype : WAM R144237, male, Bandalup Hill, Ravensthorpe Range ( 33º40'29"S 120º23'54"E ), Western Australia , Australia , collected by R. Teale & G. Harold, 14 October 2000 . Fixed in 10% formalin, stored in 70% ethanol, liver sample stored in ‒80ºC ultrafreezer at WAM . Paratypes : All from Western Australia . WAM R129674, male, 3.8 km W of Kundip, ( 33º41'S 120º09'E ); WAM R131902, female, Hellfire Bay, Cape Le Grand National Park, ( 34º00'15"S 122º10'20"E ); WAM R132154, female, Duke of Orleans Bay, Wharton Beach, ( 33º56'S 122º33'E ); WAM R144238, male, same details as holotype ; WAM R154234, male, Kundip, ( 33º40'26"S 120º11'45"E ); WAM R156978, male, Canal Rocks, ( 33º39'46"S 115º00'45"E ). Other material examined : A full list of material examined is provided at the end of the paper. Diagnosis. A small species of Delma (SVL to 79 mm ) with: ventral scales not markedly larger than adjacent lateral scales; one pair of supranasals; modally 18 midbody scales; modally 10 hindlimb scales in both sexes; 73‒92 ventral scales (males average 76.8, females 85.5); six upper labials with fourth typically below eye; loreal scale row typically interrupted by a ventral extension of supraloreal scale that contacts upper labials; essentially unpatterned head, sometimes with weak dark variegations on sides of head and indistinct narrow bars or smudges on labial scales, nape and forebody. Diagnostic differences between D. hebesa sp. nov. and D. australis are listed under the foregoing species account. Delma hebesa sp. nov. differs from D. torquata from southeastern Queensland in having a larger adult size (SVL to 79 mm versus to 63 mm ), three precloacal scales ( versus two), the fourth upper labial typically below the eye ( versus typically the third below the eye), modally 18 midbody scale rows ( versus 16) and only dark variegations (if present) on head and neck ( versus broad dark bands). Delma hebesa sp. nov. differs from all other Australian species (except D. australis , D. torquata and D. concinna ) in having ventral scales not markedly larger than adjacent lateral scales ( versus markedly larger). FIGURE 9. Preserved holotype WAM R144237 of D. hebesa sp. nov. and head scalation in (A) dorsal and (B) lateral views. Delma hebesa sp. nov. differs from D. fraseri with which it occurs in sympatry (see below) in having a smaller adult size (SVL to 79 mm versus to 140 mm , Bush et al. 2007 ), one pair of supranasals ( versus two pairs), modally 18 midbody scale rows ( versus 16), ventral scales not markedly larger than adjacent lateral scales ( versus markedly larger) and only dark variegations (if present) on head and neck ( versus broad dark bands, often faded in adults). Description of holotype ( Fig. 9 ). Measurements (in mm) and meristic values: SVL 57, HD 3.4, HL 6.9, HW 4.5, HLL 3.3, ML 4.9, RL 0.9, RW 1.3, SL 2.6, EW 1.1, HLS 10, MSR 18, VE 75. Head short and narrowing gradually forward of eyes, of equal width to body posteriorly; obvious tympanic aperture, indicated by round opening posterior to corner of mouth, opening is more narrow on right side; snout moderately long and rounded in dorsal and lateral profiles; body moderately robust of equal width and round in cross-section; hindlimbs visible as well-developed elongate, rounded flaps adpressed to body at lateral extremes of vent; tail relatively short, tapering very gradually distally to a pointed tip. Head scales smooth, non-imbricate and heterogeneous; rostral rounded anteriorly, wider than long, with obtuse apex projecting between supranasals; one pair of supranasals in broad contact, angled backwards posteromedially behind rostral and in short contact with first upper labial; nostril positioned on posterior junction of supranasal with first upper labial and postnasal; one postnasal, wider than high, angled posteroventrally and in short contact with second upper labial; prefrontals symmetrical and in broad contact; supraloreal much higher than wide, and in broad contact with second upper labial; four loreals, the anterior most slightly larger; five supraciliaries, first and fourth the smallest, second and fifth the largest; two supraoculars, first slightly larger than second; two frontals, the anterior most slightly wider and larger; two parietals, one on left side is slightly larger and longer; occipital present; two upper temporals; six upper labials, fourth the widest and positioned below eye, third the smallest, on left side a small scale intersects suture between second and third upper labial; five lower labials, second the largest and widest, fifth the smallest; mental wider than long with suture starting half-way along first upper labial. General form of head and details of scalation illustrated in Fig. 9 A, B. Body scales smooth, non-imbricate, homogeneous, and arranged in parallel longitudinal rows; ventral scales only very marginally wider than the adjacent lateral body scales; three precloacal scales. In preservative, due to the leaching of colours in ethanol after 14 years, the holotype ( Fig. 9 ) has lost the grey colouration on the dorsal surface which is now light brown. All other aspects of dark bars or smudges on lower labials and obscure variegations on body remained. Colouration in life. The following description of colouration in life is based on Figs. 10 A, B, 11A and field observations of D. hebesa sp. nov. from the Esperance Plains. Top of head bluish to light grey and unpatterned or with weak dark variegations. Lower labials whitish with blackish bars or smudges, centred on sutures of mental and anterior two lower labial scales, corner of mouth, grading to obscure variegations or weak bars around ear opening and on lateral scales of forebody. There is only very weak ventral extension of black bars or smudges on to chin and throat. Dorsal body surface bluish to light grey anteriorly, gradually merging to light brown body than to light grey tail. Lower flanks are slightly pinkish to reddish brown. Dorsal surface is uniform, except for some indication of obscure dark spots or variegations, particularly on some sutures of body scales. Ventral surface under head and along body whitish, with blackish variegations, becoming less pigmented under tail. The longest tail measured was 140 mm (208% of SVL) on WAM R131902. FIGURE 10. Adult D. hebesa sp. nov. photographed in life from (A) Scaddan and (B) Mettler Lake, Western Australia (images‒B. Maryan). For comparisons of both D. hebesa sp. nov. and D. australis in life see Thompson & Thompson (2006: 44, ‘grey morph’ and ‘brown morph’) , Bush et al. (2007 : 135) and Henkel (2010: 141) . Variation. Table 5 presents the means, standard deviations and ranges of the characters counted and measured for each sex of D. hebesa sp. nov. Table 8 presents the individual measurements and meristic counts for the type series of D. hebesa sp. nov. As noted in D. australis , the variation in the contact of the supraloreal with the upper labials and interrupting the loreal scale row or not, is similarly recorded in D. hebesa sp. nov. as the following conditions: the supraloreal contacts second upper labial on both sides (as in holotype ) in WAM R129674 and WAM R154234; it contacts second and third upper labial on both sides in WAM R131902 and WAM R144238, and is separated from upper labials by either one large loreal and an elongate postnasal or two smaller loreals in WAM R156978. One specimen (WAM R154234) has five upper labials unilaterally, with the third under eye. All the type material has 18 midbody scales and 2‒4 loreals. TABLE 8. Individual measurements (mm) and meristic counts for the type series of D. hebesa sp. nov. The holotype is marked with an asterisk.

WAM #	Sex	SVL	HD	HL	HW	HLL	ML	RL	RW	SL	EW	HLS MSR VE
129674	♂	61	3.2	6.6	4.3	3.1	5.2	0.9	1.5	2.8	1.2	10 18 76
131902	♀	67	2.6	6.6	3.7	2.2	5.4	0.7	1.3	2.3	1.1	10 18 84
132154	♀	78	3.4	7.6	5	2.1	5.3	1	1.4	3.2	1.3	10 18 85
144237*	♂	57	3.4	6.9	4.5	3.3	4.9	0.9	1.3	2.6	1.1	10 18 75
144238	♂	66	3.2	7.5	4.7	3.6	5.8	1	1.7	2.9	1.2	10 18 78
154234	♂	49	2.5	5.9	3.7	2.4	4.4	0.8	1.3	2.6	1.2	10 18 77
156978	♂	66	3.2	7.6	4.4	3.6	5	1	1.3	3.1	1.1	11 18 78

FIGURE 11. Head patterning of	(A) D. hebesa sp. nov.	adult from	Kundip, Western Australia and (B) D. australis adult from
Cowell, Eyre Peninsula, South	Australia, and arrows	indicating	pigmented or clear suture on lower labials (images‒B.
Maryan).

Variation in colouration and pattern. Most s pecimens are similarly coloured to the holotype in life and in preservative. Occasional preserved individuals (e.g. WAM R131902) have more pronounced dark bars on the lower labials and lateral scales of forebody, while others (e.g. WAM R154234) are a very uniform dark grey on head, body and tail. In life, hatchlings of D. hebesa sp. nov. have a similar colour and head pattern to adults (B. Bush, pers. comm.). Etymology. The specific name hebesa is derived from the Latin adjective hebes , meaning dull, alluding to the matt body texture, without much shine, of this species. Distribution and sympatry. Delma hebesa sp. nov. is widespread on the Esperance Plains bioregion and patchily distributed on the Warren and southern Jarrah Forest bioregions ( Thackway & Cresswell 1995 ) in southwestern Western Australia ( Fig. 8 ). Records extend east to the vicinity of Thomas River and Cape Arid, west to Canal Rocks and near Busselton, and inland to Stirling Range National Park, Ongerup, Ravensthorpe Range, Scaddan and Mount Burdett. The geographic distributions of D. hebesa sp. nov. and D. australis appear to be parapatric ( Fig. 8 ). Currently the two species are known to occur within 80 km of each other in the east: WAM R91740 of D. hebesa sp. nov. from Cape Arid versus WAM R36229 of D. australis from Pine Hill and within 130 km in the west: WAM R42637 of D. hebesa sp. nov. from Ongerup versus WAM R12604 of D. australis from Wagin. Specimens from these parapatric localities do not show any indication of morphological intermediacy. The only recorded instance of sympatry involving D. hebesa sp. nov. is with D. fraseri ( Bush 1981 , 1984 , as D. australis ). Habitat. Delma hebesa sp. nov. occupies the proteaceous scrub and mallee heath on south coast sandplains ( Beard 1990 ; Comer et al. 2001 ; Fig. 12 ). This habitat preference is exemplified of 65 records of D. hebesa sp. nov. from 18 sites, in which 79% were recorded from mallee heath ( Sanders et al. 2012 , as D. australis ). These diverse vegetation communities provide ample cover for D. hebesa sp. nov. , where most specimens, including the type series, have been pit-trapped and raked from leaf litter, spoil-heaps, and mats of dead vegetation and inside abandoned stick-ant ( Iridomyrmex conifer Forel ) nests. It has also been found under logs, mallee roots, rocks and rubbish, especially pieces of corrugated iron in disturbed areas adjacent to uncleared heath ( Bush 1981 ). It also occupies the granitic heath where it is occasionally found under exfoliated granite slabs (B. Maryan, pers. obs.). There are no habitat data associated with specimens from the lower southwestern corner of Western Australia (WAM R129003, WAM R156978, WAM R172507). FIGURE 12. The habitat of D. hebesa sp. nov. , dense scrub heath on sandplain at Stirling Range National Park, Western Australia (image‒B. Maryan). Remarks. Bush (1981: 21) , Wilson & Knowles (1988: 246) and Bush et al. (2007 : 135a) illustrated D. hebesa sp. nov. as D. australis . Additionally, Wilson & Knowles (1988: 96) and Wilson & Swan (2013: 144) refer to D. australis individuals from the far southwest of range as bluish grey, a colouration feature consistent with D. hebesa sp. nov. Historically, D. australis has been documented as not occurring in the humid deep southwest of Western Australia ( Kluge 1974 ; Wilson & Knowles 1988 ; Storr et al. 1990 ). However, there are a few records of D. hebesa sp. nov. in this area where they appear to be scarce when compared to the Esperance Plains. This apparent scarcity is documented by How et al. (1987) , who recorded ‘ D. australis ’ from only Two Peoples Bay in a survey of herpetofauna between Busselton and Albany. Biological surveys conducted on the Esperance Plains to date indicate D. hebesa sp. nov. to be widespread and locally abundant in areas of suitable habitat ( Chapman & Dell 1975 ; Bush 1985 ; Chapman & Newbey 1995 ; Sanders et al. 2012 , as D. australis ). Bush (1983 , 1984 , as D. australis ) provides additional information on reproduction in captivity and field observations of winter aggregations of D. hebesa sp. nov. from the Esperance Plains. The Esperance Plains bioregion is a biogeographically significant area rich in endemic plants, rare ecosystems, and vulnerable and specially protected fauna ( Comer et al. 2001 ). Approximately 87% of the Esperance Plains bioregion has been cleared and developed for intensive agriculture. However, much of the remaining vegetation is afforded statutory protection, including many nature reserves and national parks (Cape Arid, Cape Le Grand, Stokes, Fitzgerald River and Stirling Range National Parks) where D. hebesa sp. nov. is known to occur.