328 Recently published Ornithological Works. [Ibis, 
forms in other groups of islands ; but this is by no means 
always the case, and there are many instances of com¬ 
paratively closely allied species being found together even 
in the same patch of forest. Finally, a revised list of the 
faunal regions into which the islands can be most con¬ 
veniently arranged, is given. 
A map and a number of photographs of characteristic 
scenery in the islands agreeably assist one to follow the 
facts and arguments put forward in the paper, which is one 
of very considerable importance, and should be studied by 
all interested in the problems of the distribution of birds 
in tropical lands. 
Matheivs on Australian Birds. 
[The Birds of Australia. By Gregory M. Mathews. Vol. viii. pt. 5, 
pp. 241-316, pis. 395-399. London (Witherby), December 15th, 1920.] 
This part, which concludes the volume, gives us an 
unusually simple task, as it contains only various Pachy- 
cephaline forms now separated in the genera given below. 
First, we have the conclusion of Lewinornis rufiventris , 
to which many a generic name was given of old, not without 
reason, as it shows a certain affinity to Golturicincla. Only 
one species is allowed, the others being reduced to nine 
subspecies, of which dulcior of North Queensland, gawler- 
ensis of South Australia, and ivaddelli from the same country 
are new. Gilbertornis has two species— rufogularis , which is 
said to have a peculiar note, and inornatus , which is proved 
to be the proper appellation of gilbertii , with its three sub¬ 
species. The robust Alisterornis lanioides was considered 
worthy of specific rank, even by Gould ; its range lies to 
the northward and it has a curious habit of feeding on small 
crabs. There are four subspecies. Timixos olivaceus, with 
one less, is followed by Mattingleya griseiceps, which is the 
Eopsaltria inornata of Bamsay and almost certainly Pachy- 
cephala peninsula; of Hartert. Muscitrea simplex , which is 
said to be synonymous with Tephrodornis grisola of Blyth, 
has two subspecies, which are certainly different from the 
extralimital form. 
