INTRODUCTION, 
Little is known of the avifauna of Madura but it seems not to 
be specialized in any way. The avifauna of Bali which is rich for the 
size of the island and includes some montane forms is closely allied to 
that of East Java and much more remotely to the very distinct Austro- 
Oriental omis on the neighbouring island of Lombok. The two islands 
are, of course, separated by the narrow, but deep strip of water, form- 
ing part of the well-known * f Wallace’s Line", which can be regarded 
as the southern end of the zoo-geographical boundary between the 
Malaysian and Austro- Oriental sub-regions. Most, but not all, so- 
called Balinese forms are also found in East Java. One of the 
exceptions is a sturnme genus Leucopsar, peculiar to Bali, The 
Austro-Oriental influence is shewn by the presence of Trkhoglossus f 
Lichmera and PtUinopus cinctus which reach their western limit in 
Bali. 
A few subspecies have been described from various small islands 
in the Java Sea on which both Bornean and Javan species are found, 
but these islands are most remarkable for providing stepping stones 
for an infiltration to the west of Austro-Oriental forms which even 
reaches the Thousand Islands off the west point of Java and Billiton. 
The birds of the Karimon-Java Islands show a close affinity to 
those of Java. A species represented in vSouth Borneo and Java by 
different subspecies, if it occurs in the Karimon-Java Islands, is either 
represented there by the Java race, or by a race which is, according to 
the present state of our knowledge confined to the islands, or else 
found in the same, or a very slightly altered state, on other islands 
in the Java Sea. 
Bawean has a few indigenous subspecies : its avifauna is mixed 
Javan and Bornean, but the former element is dominant. 
The Kangean Islands with some well-marked endemic subspecies 
are also mostly Javan in their avifauna, but there is also a strong 
Austro-Oriental element. 
The fauna of Christmas Island is highly specialized. The avifauna 
is Austro-Oriental in its present affinities which is surprising when the 
nearness of Java is considered. In more detail we find that of the 
Austro-Oriental subspecies the Christmas Island races are most like 
those found in the chain of islands stretching from Sumbawa to Timor 
Laut, and, furthermore, when, within this limited area, subspecific 
distinctions can be recognized in a species, the Christmas Island form 
is more like that found in Sumba, Savu, Timor &c., than that inhabit- 
ing Sumbawa and Flores. 
The birds of Cocos-Keeling are of no great interest. They include 
oceanic forms, introduced species, and stragglers from Java and 
elsewhere. There is no indigenous, peculiar, land-bird except Rallus 
philippensis andrewsi whose sub-specific status I have been unable 
to check. 
In the future I hope to attempt a more detailed analysis of the 
Malaysian avifauna. 
XIV 
