MARSUPIALS. 291 



If we summarize the facta relating to tlie distribution of IMarsiipial.s which we have just re- 

 corded, we shall find that the following inferences may be drawn from them. First, that the 

 north is thoroughly isolated, and distinct from the rest of the continent. With the solitary exception 

 of Ph.vlangista vi'LPiXA, which may have reached it by overland migration, not a single species 

 inhabits it which is known elsewhere. At the same time, unlike the south-west, it has no un- 

 known types of life. It has mcrelj^ a oollectiou of new sj^ecies of genera already well known as 

 inhabiting New South Wales, although most of them are so distinct as to indicate a long period 

 of separation, only one (Petaukis Ariel), besides the Phalangista, being near any of the New 

 South Wales species. 



Next, it appears plain that the south-west corner, viz., that part of the continent stretching 

 from Shark's Bay southwards, and round past Perth and Swan River, is a j)eculiar district. 

 This, although possessing fewer endemic species than the north, has more of special organization. 

 Here have been found those species of which no other similar creature has yet been foimd in any 

 part of the world. As to actually distinct species, although not distinct forms, we find twenty- 

 eight siDecios out of thirty-nine (that is, about three-fourths) peculiar to itself. Four are common 

 to it and the country to the north of it (North-west Australia), but are not found in South 

 Australia, with which, however, it .shares six. The difference between it and North xVustralia 

 seems to be this, that although both equally isolated at some former period, the one (the south- 

 west) has been so at a more ancient date, and for a longer period; and when the space between 

 was raised out of the sea, and the intermediate regions turned into dry land, the character of 

 the district surrounding North Australia has been of a more desolate, inhospitable character than 

 that between South-west Australia and New South Wales, the communication between which, 

 although difficidt, has not been impracticable, as seems to have been the case with the north ; 

 from which it maj' perhaps be inferred in the absence of any special psculiarities in its condition, 

 that the north has been last raised, and that the upheaved bottom of the sea to the south is more 

 desolate from having had less time to become clothed with a jjroductive soil. 



New South Wales stands nearly in the same position as South-west Australia as an isolated 

 region. It has forty species, of which twenty-nine are peculiar (or forty-one out of sixtj% if Van 

 Dieman's Land be taken along witli it). Eleven of these are found in South Aiistralia, and ii\o. 

 in West Australia. 



Van Dieman's Land, altliough belonging to the same division of the continent as New South 

 Wales, has an unusually large number of endemic species. Out of twenty species, twelve, that is, 

 rather more than the half, arc pecidiar to itself, and amongst them are more of the Marsupial 

 Carnivora than are found in any other part of the continent, one of them being the Thylaoini s, 

 which is a peculiar type found nowhere else. The general strain, however, of the species found in 

 Van Dieman's Land is that of New South Wales. They are, with tlie excejjtion of the Thy- 

 LAciNus, all species of genera also existing in that region. Notwithstanding its large jiroportion 

 of endemic si^ecies, therefore, it is not entitled to "be regarded as a distinct and independent 

 zoological province in the same sense, or at least in the same degree, that the north and south- 

 west are ; nor does this disproportion tell us anything about the length of time for which ^'an 

 Dieman's Laud has been separated from the mainland. I have already explained that I do not 

 think that time has anything to do with the increase of species, except as giving greater oi^portunity 

 for the occurrence of physical changes in the conditions of life, through which change in species 

 mav be induced. 



