682 On a new Uork-Wallahij. 



By its gciu'ial cliaiacters llic animal is clearly nlatcd to 

 the West Australian Pctroaale lateral/ii of the Bwan Uiver 

 district and to P. hocketii of ]\[ondrain Island. This lattor 

 MMs described as a snhsjiecies of P. hiteralis, bnt I should 

 now give it specific lank on account of its uniformly larger 

 size and certain ditTerences in the structure of its skull. 



The new form is thus situated far to the east of the pre- 

 viously known members ot" the group, with the large haclcelli 

 between it and the form which it — at least superficially — 

 most resembles. 



It may be culled 



Petrof/ale pearsoni, sp. n. 



Size comparatively small, about as in lateralis, decidedly 

 smaller than in hacketti. General colour, on the whole, very 

 much as in lateralis, paler than in hacketti. Dark lateral 

 lines of the underside, however, more blackish, those of 

 laf(raHi< being dark brown. White patches at base of ears 

 larger and more jirominent. Tail with its u])per and under 

 surfaces, from about 3 inches from the base, contrasted 

 black, the sides dull buffy whitish ; above, the black soon 

 fades off into the brownish terminal tuft, but below it con- 

 tinues to within 2 inches of the tip. This tail-coloration is, 

 on the whole, more as in hackttti than lateralis, but in both 

 there is considerable variation. 



The usual narrow black dorsal line is continued rather 

 more definitely on to the rump than in any of our specimens 

 of lateralis, but the difference may be due to this part being 

 in fresh ))elage, and so showing the line more distinctly. 



Skull in size and general shaj)e quite as in lateralis, 

 smaller and with less heavy supraorbital i-idges than in hacketti. 

 Palatal foramina comj)aratively long, about as long as in 

 liacketti. In the bulhc there is a diU'erence between lateralis 

 and hacketti which had not been ))reviously noticed. In the 

 former they are fairly well swollen, anteriorly as well as 

 posteriorly, so as to jjroduce a transverse convexity (hardly 

 to be called a ridge), in front of which the bone descends 

 nearly vertically towards the level of the glenoid surfaces. 

 In hacketti, however, the whole bulla is larger, but lower 

 and more spread out, its front part evenly and gradually 

 descending towards the glenoid level without marked trans- 

 verse convexity. In pearsoni the bulltxi are most like those 

 of hueketti, altiiough perhaps a little more swollen. In 

 making this comparison six skulls of hackttti and ten of 

 l(it(rulis\\-A'VG been available, so liiat the difference is evidently 

 fairlv constant. 



