HOtJTMAN ABROLHOS ISLANDS. 149 



the sea and overhangs very considerably (PI. 14. fig. 6). This limestone 

 conglomerate forms the margin of: the island for a little distance along the 

 western shore, but gives place to a sandy beach again as the northern shore 

 is reached. The sand dunes are not so high here as on the eastern side. The 

 coral limestone, which is exposed at certain places along the shore, can be 

 traced inland and evidently forms the foundation of the island. Its surface 

 is about 6 or 8 feet above sea-level, and it agrees in structure with that to be 

 referred to later in connection with the Wallaby Isles. The central part of 

 the island is quite low and flat, and only a few feet above sea-level (a salt 

 lake, dried up at the time of our visit, occupies part of it). No rock 

 exposures are visible once the shore is left, the whole being covered with 

 sand which, as we have remarked above, is responsible for the height of the 

 island above sea-level. The sand-hills are thickly covered with low halophytic 

 bushes of a hard and brittle nature, and the terrestrial fauna appears to be 

 poorer than that of the southern islands. The lack of fresh water is probably 

 responsible for this. 



Shore-collecting was not very successful at North Island, and as the other 

 islands appeared likely to offer better opportunities for all-round work we 

 spent very little time there. A number of specimens of Heterodontus phillipi 

 were captured whilst we lay at anchor close to the shore, together with the 

 Parrot fish (Coris auricularis, one of the Labridas) ; Groper (Aclierodus 

 gouldii) ; Skipjack (Caranx platessa) and Buffalo Bream (Carangidse). The 

 latter looks very much finer than it tastes ! A number of birds wore noted 

 and are enumerated in Mr. Alexander's report. Further reference will be 

 made to the structure of North Island in the section dealing with the 

 formation of the four islands. We may remark here that the wide expanse 

 of reef-flat represents the result of erosion and that North Island will 

 probably be cut down still further in the future. 



THE WALLABY ISLES. 



This group of coral islets is, in some ways, the most interesting of all, for 

 it is on certain of these that the land fauna requires, as an explanation of its 

 character, the assumption of land connection with the mainland of Australia. 

 The group consists of two large islands (the East and West Wallaby Islands) 

 together with a considerable number of small islets, many of which are not 

 depicted at all on the small-scale charts. The East Wallaby Island lies 

 really north-east of West Wallaby Island, from which it is separated by a 

 very shallow channel about a mile across. 



At first sight it appeared somewhat difficult to piece out the former 

 history and the structure of the irregular collection of islands, reefs and 

 lagoons of the Wallaby Group. It became much more clear after our 

 second expedition, and I now consider that the group represents a stage in 

 the development of an atoll form. 



