no SUNKLANDS OF PORT PHILLIP BAY AND BASS STRAIT 
on their own activities to migrate to new regions, and the migra- 
tion is consequently slow : the transference of cocoons by 
adventitious means is improbable. They cannot pass over moun- 
tain ranges that are bare of vegetation or covered by ice and snow, 
cannot live in desert tracts where there is an absence of moisture 
and organic matter, or migrate over arms of the sea. Some genera 
are restricted to warmer climates, while ice and snow, if they cover 
a region for a long period, cause their extermination. The 
earthworm population disappeared in those parts of Europe 
covered during the Ice Age, and those on the fringe are now 
spreading nortWards : Spencer mentions his difficulty in obtaining 
earthworms near the glacial tarn of St. Clair in Tasmania, pre- 
sumably due to the former ice cover during the Margaret 
glaciation. It is evident that a land bridge was essential for the 
migration of earthworms from Victoria to Tasmania, and it was 
once covered with a mesophytic flora, implying the existence of a 
moist or wet climate and a continuous belt of ordinary humus. 
Perionyx is found only in regions of heavier rainfall — in Aus- 
tralia, in the cooler wet portions of Victoria and Tasmania, or 
the tropical rain belt of the north. That the migration was 
southwards is suggested by the fact that the root genus of the 
Megascolecinae is Diplotrema found in Queensland and New 
Caledonia, and Queensland is thought to be the centre from which 
its distribution took place. 
Spencer (1896) found in Central Australia a single species of 
earthworm, one of the Acanthodrilinae-Acanthodrihis eremius^ 
Spencer in patches of damp black earth in the George Gill, 
McDonnell, and James Ranges. These patches were surrounded 
by a wide tract of desert country quite impassable to earthworms, 
and he considered A. eremius to be the only surviving species of 
the earthworm population of Central Australia before its climate 
changed and its desiccation ensued ; the Acanthodrilids migrated 
to the tropical rain belt, or to the highlands of the south-east. 
They are found in north, north-east, and north-west Australia — 
for the most part in the tropical rain belt — also in the south-west, 
the cool rain belt; but on Mt. Kosciusko, the highest part of 
Australia, quite a different population is found belonging to the 
Tubificidae and Phereodrilidae which almost certainly existed 
when the Acanthodrilids prevailed in Australia. 
The probable absence of ordinary humus on the eustatic land 
bridges was a barrier to migration; the Megascolecinae passed 
over to Tasmania on one of the more permanent land surfaces, 
presumably that during the Yolande or Riss glacial stage in the 
Middle Pleistocene. 
