152 MR. GEOFFREY SMITH ON THE 



filling in of this sea and the junction of the eastern and western 

 islands, no feasible passage was formed by means of which 

 tlie eastern Astacopsis could invade the territory of the western 

 Chcera'ps or vice versa. The nature of this early barrier is un- 

 certain, but when once the desert condition was established, the 

 central deserts have consituted a barrier against a direct eastern 

 or western migration : on the south-west coast the desert and semi- 

 tropical region of the Australian Bight prevents a northern passage 

 of Astacopsis into the territory of Chceraps, while to the north- 

 east the semi-tropical coastal streams do not appear to have been 

 colonized either by Chceraps moving southwards or Astacopsis 

 migrating northwards. Astacopsis, in fact, remains a typically 

 temperate genus, Chceraps a semi-tropical or tropical one, and the 

 barrier of tempei'ature seems to determine the distiibution of 

 these two forms with complete rigidity. 



An oSshoot from Chceraps, however, viz. Farachceraps hicarinatus, 

 on the filling in of the Cretaceous sea, has gradually extended its 

 range from the west eastwards, invading the centi^al districts of 

 Australia and penetrating thence southwards and eastwards so 

 as to become almost coterminous in distribution with Astacopsis. 

 In the very centre of the Astacopsis distribution, viz, in Victoria, 

 P. hicarinattis is particularly abundant, and here it has given rise 

 to the burrowing land-Crayfishes, Engceus, whose headquarters are 

 in this region. If it is true that P. hicarinatits does not occur in 

 New Guinea or Tasmania, the view that this form is of compara- 

 tively recent origin is amply confirmed, the date of its origin and 

 migration being fixed as subsequent to the separation of New 

 Guinea and Tasmania from the mainland in Tertiary times. We 

 are, however, here confronted with a difiiculty, since, although 

 P. hicarinatus apparently does not occur in Tasmania, one or two 

 species of Engceus, which Ave must hold to be derived from 

 P. hicarinatus, are common in Tasmania. The habits, however, 

 of Engceus are peculiar, and it is quite possible that individuals of 

 this genus have been carried across Bass's Straits from Gippsland 

 in floating tree trunks, at the roots of which the " land-crabs " 

 frequently make their burrows. 



Without unduly pressing this hypothesis, it nevertheless oflTers 

 an explanation of the curious fact of the occurrence of Engceus in 

 Tasmania and the absence of its parent form, P. hicarinatus from 

 that island. 



The occurrence of A. kershavn in Gippsland, related by many 

 of its features, e. g. the absence of spines on the abdomen and the 

 truncated shape of the rostrum, to A. franklinii in Tasmania, 

 cannot be passed over without reference. It is apparently con- 

 fined to Gippsland, but it is surrounded and in close contact 

 with the much more widely distributed A. serratus. Its 

 relationship to the Tasmanian form indicates a close connection 

 at some not very remote time between Gippsland and Tasmania, 

 a connection which may be detected in the similar geological and 



