32 Director's Animal Report. 



Mendaiia visited the group (1595) they had much better canoes 

 and sails than two hundred years later. 



As regards the Paumotu Archipelago — these islands were little 

 visited in former times on account of the difficulty of the naviga- 

 tion. The gentlemen of the Wilkes exploring expedition were the 

 first to describe these parts, and there was no information given 

 concerning" the sails. The Paumotuans had large double canoes 

 which traded between the innumerable small islands, and from a 

 model of a canoe in the Bishop Museum from Manihi Island it was 

 ascertained that the sails were similar to those of the Caroline Isl- 

 ands. The canoe model was made within the last twenty years, 

 and there is little to show that the style of sail was not of modern 

 introducftion. There are two sails on the model, suspended from 

 two vertical masts. 



RAY-SKIN RASPS. 



By Allen M. Walcott, Assistant in the Museum. 



Among the specimens in the Gilbert Islands sedlion of Poly- 

 nesian Hall are the four rasps shown in Fig. 10. These implements 

 are more or less common throughout the islands of the Pacific. 

 The outside or rasping portion is the skin from the back of a spe- 

 cies of Trygon or sting-ray not uncommon in the waters about the 

 islands. An}- convenient piece of wood makes the handle and core, 

 No. 3 having for a center a portion of the leg of a foreign chair. 

 As this skin when dry cannot be bent readih^ it is, while wet and 

 pliable, sewed firmly around the wood with coconut fibre thread. 

 It will be seen from the illustration that the tubercles on the differ- 

 ent rasps vary much in size, and this is due to the age of the fish 

 from which the skin is taken. 



The various grades were adapted to the work to be done. 

 Almost their sole use was to enable the Gilbert Islanders to so 

 shape the edges of the boards of their canoes that when sewed 

 together they were water-tight. As the Gilbert Islands are low 

 coral atolls the trees are generally not large enough to be used for 

 dug-out canoes ; hence the necessity for vising planks. These were 

 obtained from the breadfruit tree {Artocarpus iucisa). For the 

 making of the great proas, with a length of seventy feet and a 



