98 THE AUSTRALIAN NATURALIST. 
‘have P. gambrisius, which in the male bears a strong resem- 
blance on the underside to some Solomon Island forms. This 
also has only one female form. In the Northern Moluccas 
P. tydeus is found, whose single female is allied to the light 
amanga form of New Guinea. New Britain and New Ireland 
have each a single male and two female forms, but in each case 
the light form is wanting. In the Solomon Islands each large 
island as yet visited has been found to possess forms belonging 
to two groups according as the white area of the hindwing 
extends along the costa or not (P. hecateus and P. ptolycus) ; 
then further east we have the fine form, P. oberon, from Santa 
Cruz, which is closely allied to our own P. xgeus. 
5. New Guinea as a centre of dispersion, as exemplified by 
Papilio zyeus and Troides priamus.—In considering the origin 
of a species it is not enough to take the central point of its 
present distribution. 
The butterfly fauna of Hastern Australia may be divided 
into two groups of genera: First, those that decrease in the 
number of species as we range North, and these forms have no 
doubt entered our continent from the South ; secondly, those 
genera that are almost entirely absent from the South, but 
increase in number of species as we proceed North, and further 
develop in New Guinea and the Malayan Islands. These 
species have reached us v¢a Torres Straits. 
The longer a species is in existence in a country the more 
likely it is to be found in a variable condition. This is amply 
borne out by the immense variation of P. egeus in New Guinea, 
while this variation decreases as we proceed West, or South or 
Hast. Further, when we take species to the east and west of 
New Guinea we find a remarkable resemblance. For instance, 
the specimens of P. gambrisius, from Ambonia, and the form 
from Choiseul, in the Solomon Islands, show a very close 
resemblance on the underside, so that we are constrained to 
the view that these species represent somewhat corresponding 
points in the western and eastern spreading of this group of 
species. The Santa Cruz form at the extreme east is very 
close to the Australian form, and the Ugi species, P. erskinez, 
also on the east, shows a relation to P. wgeus. P. inopinatus, 
on the west, would also be instantly recognised as related to 
our species. Similar facts us these are shown in the cases of 
P. agamemnon, P. ambrax and PB. ulysses. 
As a further illustration of this, I have brought for exhi- 
bition this evening a selection of my sub-species of Trotdes 
priamus. The green-winged J'roides obtain their maximum 
development in New Guinea, where beside the common 
T. priamus they are represented by T. tithonus, T. goliath, 
T. chimera, and the various forms of 7’. paradisea, none of 
which occur outside the limits of New Guinea. This immense 
