3 
INTRODUCTION 
Avitli Capo York in the northeast as number 1: thence we carry the numbers through the islands 
of Torres Straits^ down the eastern coast through Queensland, N. S. Wales, Victoria and Tasmania, 
along the southern sliores tlirougli South Australia and West Australia, and around the continent 
through the Northern Territoi’v to Cape York again. The small inset maps give fuller detail of 
those areas within which some attempt at systematic collecting has I)een made: some of these extra 
localities are unnumbered, as their index reforeuee to the inset map is sutricien*.:. 
Tlie limits of Australia are clearly delined except at one point, wliere the islands of Torres Straits 
form a scries of natural steppingstones to New Guinea. There is no deep-water ehaunel here to form 
a natural boundary, and the geological formation of llio islands (see Transactions Irish Academy, 
XXX, Part Xr, 1S94) is Australian rather than Papuan. Therefore we liaA’C adopted the political 
boundary of Queensland, which includes Darnley Is., Panks Is., and i^Iurray Is. No other could have 
been used here unless the coastline of the continent itself were taken, and that would exclude such 
essentially Australian islands as Thursday Is., and Prince of Wales Is. 
The great majority of Australian butterflies are to be found within fifty miles of the coast, and 
only a very small percentage of the sj^ecies occur in the far interior: consequently the great hulk of 
our mainland records are from points not far from tlie coastline. ^lost of the localities named 
may he considered as including tJie suiTOunding country within a lifteen-mile radius: when two points 
are less tliaii that distance apart it may he assumed that either altitude or geological formation is 
different. 
Geographical variation is more dependent upon geological and climatic conditions than upon 
distance. An altitude of a thousand feet, or a sudden transition from scrub to forest country, may 
result in a gi-eater change in the appearance of the local butterflies than a distance of hve hundred 
miles. Yet the varying altitudes of a single belt of scrub country do not a]>]iear to affect the 
local butterflies to any great extent: this is well illustrated in the Cairns-Kuranda district: most of 
the scrub siiecies taken at Cairns close to sealevel also occur at Kuranda (1000ft.) and many of 
them at Herherton (3000ft.). The forest country approaches Ivuranda from some points very closely 
indeed, consequently thei’e is a sprinkling of forest sjiecies in Kuranda that are not taken at 
Cairns: and sj^ecies attached to the mangrove swamps which occur near Cairns are absent from 
Kuranda. The number of species to be taken during the winter months is much greater in Cairns 
than in Kuranda, on accoiint no doubt of the smaller variation of temperatures near the coast. Tlie 
character of the ruling vegetation evidently affects the butterflies much more than does the latitude. 
Many of the butterflies of the east coast occur also in West Australia: in most cases they are modified, 
hut can easily l)e recognised as geographical races, hut in some feAV species such as X. Idugi and 
C. cgprotus we can detect no points of difference. 
Now that most of the butterflies from the lands and the larger islands of the Indo-Australian Kegion 
are knoAvn and desciibed, the complete adoption of the trinomial system of. nomenclature for the 
Australian butterflies need no longer be delayed. The advantages of this system are self-evident: 
it indicates at a glance the relationship witli other races from beyond AustriAian limits, and it 
provides easily available data for the study of geographical distribution. | 
I* 
V 
In the past our Australian lists have been built upon a hybrid system of b^iomial nomenclature 
in Avhich collective and racial names have been indiscnminately used. In thfe genus FapiUo for 
instance, four of our huttcrflies have been known as iilyssesj sarpedotij parmu^iiSj and egipiits : the 
first two of these are collective names and the last two racial names. If we use? idysses and sarpedon 
and Avish to be consistent, Ave must change parmalns to aristeiis and egipius ti amhrax'. or if avc 
prefer the racial names then idysses must give place to joesa and sarpedon to choredon'. either the 
one or the other method should be adopted, the present mixture is altogether i iadmissiblc. The use 
of the trinomial system clears up such a difficulty at once, and sIioaa's choredon to be the Australian 
race of sarpedon^ joesa of idysses^ parmatus of aristeus and egipius of amhrax, 
i 
