houtman abrolhos isLANi)s. 169 



a result of natural phenomena related to the growth of reef-building corals. 

 Neither elevation nor subsidence are required to account for it. The only 

 other possible explanation is that this ring-shaped reef is all that remains of 

 a former mass perhaps at one time part of the outer rim islets some distance 

 away from it. In that case erosion would play a more prominent part in the 

 explanation of the present form, but -would be governed by phenomena of 

 normal coral growth. One can find other cases in the Pelsart Group where 



Text-figure 12. 



small coral growths are taking the form of miniature atolls, and neither 

 subsidence nor the heaping of debris by waves and currents are responsible 

 for the general architecture of these structures. Just as the one mentioned 

 above occurs where coral growths are most luxuriant in the Easter Group 

 lagoon, i.e. nenr its northern end, so in the Pelsart Group the northern open 

 side of the atoll is the site of the greatest lagoon coral growth. In the latter 

 case it is obvious that the general trend of water in the lagoon is from south 

 to north, and the water is more free from sand and debris to the north where 

 the lagoon is open to the sea. 



Faunistic Notes on the Easter Group. 



No large mammals occur on Rat Island except those which have been 

 introduced, and mention has already been made of the cat which was intro- 

 duced by the guano workers to keep down the rats. 



The island has been noted by many ornithologists for the vast number of 

 Noddy Terns which are found regularly every year during the months of 

 October and November nesting upon the bushes. The nests are practically 

 nothing but "stations" on the tops of the shrubs where the birds lay 

 their eggs (PI. 12). Very often one meets with shrubs almost covered 

 with the nests of these birds, the owners refusing to leave them when dis- 

 turbed, unless actually pushed off. Large numbers of Sooty Terns also nest 



