ALLEN ON GEOGRAPHICAL DISTRIBUTION OF MAMMALS. 367 



I consider, so closely are they related, rather as provinces than regions, 

 and may be termed respectively the Papuan Province and the Australian 

 Province. The former is situated almost wholly between the equator and 

 the twentieth degree of south latitude. The latter embraces that portion 

 of Australia south of this line, together with Tasmania. The boundary 

 between the two regions can of course be drawn only approximately, 

 but may be provisionally assumed as the vicinity of the isotherm of 

 700 p.* The reason for uniting the northern portion of Australia 

 with New Gujnea as a part of the Papuan Province lies in the fact that 

 not only so many of the mammalian genera are common to the two, but 

 that these genera are absent from the more southern portions of Aus- 

 tralia, where they are replaced by others wholly restricted to South 

 Australia and Tasmania. Three-fourths of all the genera of Marsupials 

 (excluding, of course, the American family Didelphidce) are, so far as at 

 present known, restricted to the Australian Province, as are several gen- 

 era of Muridce and the Ornithorhynchus. Of the remaining Marsupial 

 genera, six only are limited to the Papuan Province. 



The Papuan Province. The Papuan Province embraces not only New 

 Guinea, but the Molucca and Aru' Islands on the west and the Solomon 



here following sufficiently shows. The principle I still hold as applying to Australia 

 with the same force as elsewhere, only I make the division more to the northward, as 

 a little more care would have led me to do originally. The York Peninsula, and most 

 probably the whole northern coast region north of 20 S. lat. (except the high arid 

 interior), has certainly closer affinities, as regards mammals, with New Guinea than it 

 has with any portion of South Australia. Of the strictly Papuan genera, only two out 

 of nine are restricted to New Guinea, the rest being common to both North Australia 

 and Papua. Of the other North Australian genera, about one-half occur generally 

 throughout the continent, but the remainder are essentially South Australian, rep- 

 resented by only stragglers in Northern Australia. On the other hand, more than twenty 

 genera occurring in Southern Australia and Tasmania, are wholly unrepresented in the 

 portion of Australia I here assign to the Papuan Region. In other words, we get the 

 same wide faunal differences between the tropical and temperate portions of the 

 Australian Realm that we get elsewhere under similar climatic conditions. 



In the same connection, Mr. Wallace cites my separation of Temperate South Africa 

 as a primary region as another instance of the misleading nature of the principle of 

 the distribution of life in zones. This I have also seen fit to abandon (see anted, p. 351 ) 

 on a detailed re-examination of the subject, not because the principle is erroneous, but 

 in consequence of certain peculiar geographical conditions, namely, the comparatively 

 small area subject to a temperate climate and to its limited extension into the temperate 

 region. It is, in fact, wholly within the warm-temperate belt, and widens rapidly north- 

 ward to abut very broadly against the tropical zone. Only a very small portion really 

 comes under the influence of temperate conditions. Here we get, as usual, a temperate 

 aspect in the fauna, and I still maintain my separation of South Africa as a faunal divi- 

 sion, simply lowering its grade from a primary region to a u province " of the great Iiido- 

 African Realm, simply from the fact that thesmallnessof its area and warm-temperate, 

 rather than temperate, conditions have prevented, as would be naturally expected, any 

 great amount of differentiation . 



* Mr. E. Blyth, in a paper (Nature, vol. iii, p. 428, issue of March 30, 1871) published 

 almost simultaneously with my own cited in the last foot-note, included a portion of 

 Northern Australia in his " Papuan Sub-region ", namely, " York Peninsula and eastern 

 half of Queensland (as far as the dividing range), on the main laud of Australia". 



