PAUMOTU GROUP. 333 



about on the coral shelf. This, where I measured it, was five hundred 

 ieet wide, but it is not of equal width in all parts. Among the coral 

 bjocks was some sand, and in many of them were found large speci- 

 mens of the chama and other shells. I was informed at Raraka, that 

 there were a few inhabitants on Vincennes Island, but none were seen 

 by us. They were said to live on the southern end of it. 



After finishing our observations, we returned on board, and made 

 sail for Aratica, or CarlshofF Island. We arrived off it in time to 

 secure its connexion with Vincennes Island : the distance was found, 

 by patent log, and astronomical observations, to be twenty miles to 

 the westward. We then stood on and off its eastern point for the 

 night. The next morning at daylight we began its survey. The 

 tender was despatched round its northern shore, whilst the Peacock 

 and Vincennes took its southern side, running close along the reef, 

 which continued submerged until near its southwestern end, which is 

 twelve feet high and thickly wooded. On rounding the point, we saw 

 a white flag waved by several natives on the beach. I immediately 

 despatched a boat, with an officer, who brought off two of the prin- 

 cipal natives, one of whom spoke a little English, and proved quite 

 intelligent. One of these natives was tattooed only on one side, from 

 the pubis to the sternum, bounded by broad blue bands, which divided 

 and terminated under each ear. 



NATIVE OF PAUMOTU GROUP. 



He reported that there were about twenty natives on the island, 

 and that they had frequent intercourse with vessels that had visited 

 them. They informed me that water was to be had on the island. 

 Finding ourselves short of this necessary article, I despatched several 

 boats to procure it. Aratica is eight miles in length by five in 

 breadth. 



