( 529 ) 



In this form the adnlt femalf is a pale silvery grey all over almve, while below 

 the belly is only a little paler, and there is no defined white middle part to the belly 

 as there is in New Guinea examples of subsp. ti/pieus. The dorsal line is strongly 

 marked, and there is a yellowish snffnsion on the fore-back on each side of the line, 

 and also on the rnmp just at the root of the tail. Young specimens are similar, and 

 so are the young males, but in the latter sex as the animal gets older the general 

 colour gets paler,* first anteriorly and then throughout, until it is almost white, 

 with just a thin sprinkling of darker hairs among the white. Tlie old mf/lt' in the 

 Tring Museum rejaresents this last stage. The dorsal line still remains conspicuous, 

 which is not the case in the well-known albinistic examples of this species. Also 

 in adult and old mnles the underside of the neck gradually gets sutfused with rufous 

 brown, probabl}' glandular in its origin. 



The series brought home by Mr. Meek is very interesting as showing the way in 

 which, while the femah's always retain their grey colour, the males, which are at 

 first like the females, gradually become in age quite different. 



The single specimen of subsp. intercasiellanns available is a very old male, older 

 than any of the examples of kiriwinae, but its general colour is still greyish, and it 

 has no trace of the glandular darkening on the throat. 



The skull oi kiriwinae is longer and rather slenderer than that o'lintercastellaniis, 

 and has a much narrower interorbital and intertemi^oral region, which is especially 

 noticeable as this part becomes narrower with age, and the type of intercastellanus 

 is extremely old. The jiostorbital processes do not overhang the orbits so far 

 laterally, but seem to be situated farther forwards as compared to the brain-case. 

 Another curious peculiarity present in all the specimens is that the zygomata are 

 bowed inwards anteriorly opposite the ascending portion of the malar, so as to form 

 a distinct concavity in their outline at this point. In the two Tring specimens, 

 which still contain their skulls, this concavity can be distinctly felt beneath the skin. 

 It is also noticeable that in old females the supraorbital crests are almost as well 

 developed, so that the animal is consequently as " cavifrons " as in the males. 



Altogether the Kiriwina Cuscus, although undoubtedly very closely allied both 

 to the typical Ph. orientalis and to subsp. iniercastellanus, seems sufficiently distinct 

 as a local race to deserve a subspecific name. 



The following are the skull-measurements of the adult male selected as the 

 type (Brit. Mus. 96.11..5.15):— Extreme length from back of occipital crest 92-3; 

 basal length {<■) 84: greatest breadth .")S ; nasals, length 35-5, greatest breadth 13; 

 interorbital breadth 11-6; tip to tip of postorbital processes 12-2; intertemporal 

 breadth 4; palate length 40 ; palatal foramen 6-2. Horizontal length of p' 4-8; 

 length of ms'~^ 1.5. 



Of this form there are a fine pair in the Tring Museum, and three males of 

 different ages and an aA\\\i female in the British Museum. 



15. Perameles sp. (probably P. doryana Quoy ^t Gaim). 



One young specimen. Woodlark Island. 



This Bandicoot is unfortunately too young for certain determination. 



* This appears to be the general coarse of the colour change, but one fact r.ither militates against the 

 correctness of the above .account, namely, that on the whitisli anterior back of the olde-st British Museum 

 specimen (the type) some patches of new hairs are ccmiing up (ireij, exactly the reverse of what should be 

 the case. I can at present suggest no explanation of this curious anomaly. 



