Linton — The Marquesas Islands 305 



while two other holes, one on either side, were pierced through the angle formed by the top 

 and sides about half-way between the breakwater and rear end of the deck. The upper lash- 

 ing holes, which served for the attachment of the ends of the side planks, did not pierce the 

 cavity of the bow piece. 



The neck, or portion lying between the breakwater and the figurehead, was two to three 

 times the length of the deck. Its upper surface was flat, and was the same width as the deck 

 immediately in front of the breakwater, tapering gradually toward the forward end. Its lower 

 surface was hollowed for about one-half its length, the outer end of the excavation being cut 

 to leave a semicircular projection like that already mentioned in the description of the ends 

 of canoe underbodies. Beyond the end of the excavation a heavy ilange projected downward, 

 its lower edge being flush with the bottom of the end of the underbody when the bow piece was 

 in position. On the outer side this flange curved up smoothly to the base of the figurehead. 

 The lower edges of the sides of the excavated portion fitted accurately to those of the under- 

 body, and were pierced with lashing poles. Above the edge, however, the sides sloped inward, 

 the flat upper surface of the neck projecting beyond them on either side in a longitudinal 

 flange. 



The angle which the neck formed with the body of the canoe seems to have varied con- 

 siderably. In the specimen photographed the neck is practically parallel to the upper edge ot 

 the underbody, and the early accounts and pictures would seem to indicate that this was usual. 

 According to informants, however, it curved upward sharply beyond the end of the underbody, 

 and all the recent canoe models show this feature in more or less exaggerated form. In the 

 double canoe shown by Christian (10, p. 180), the neck of the bow piece of one canoe is of 

 the form already described and curves upward at only a slight angle, while the neck of the 

 other canoe curves up much more abruptly. In the second specimen the upper surface of the 

 neck does not project over the sides, but curves down into them smoothly, as in the normal 

 stern pieces. 



The form of the figurehead and its relation to the neck can best be described by com- 

 paring it to a long handled shovel laid back up, the neck corresponding to the handle and 

 the figurehead to the blade. The base of the figurehead projected out sharply from the neck 

 on either side, while the forward end was semicircular or was bluntly pointed. The upper sur- 

 face was convex, the point of greatest thickness being at the middle of the rear end. It was 

 carved into a single, large, flat face, chin forward. The convention was the usual one except 

 that the broad mouth was crescent shaped, and the ears were represented by thin semicircular 

 flanges which projected from either side of the head just beyond its juncture with the neck. 



Ths Stern Piece 



Two forms of stern piece were employed in the Marquesan built up canoe. 

 The one shown in the accompanying picture (PI. XLii^ H) is apparently the com- 

 moner type. For the sake of convenience in description this form may be divided 

 into three parts — the body, deck and tail. 



The body of the stern piece was that part which rested upon the underbody of the canoe. 

 It formed a third to a half of the total length of the stern piece, supporting the cleck and taper- 

 ing off into the tail at the rear. The lower surface of the body was deeply hollowed, while 

 the outer sides curved inward gradually meeting in a distinct ridge along the median line of 

 the upper surface. The lower edges were cut to conform accurately to the upper edges of 

 the stern of the underbody, and were pierced with holes for lashings. The rear end of the body 

 terminated in a heavy, downward projecting flange which rested against the stern of the under- 

 body and was identical in form and arrangement with the corresponding flange on the bow 

 piece. 



The deck, although hewn in a single piece with the body, had the appearance of simply 

 resting upon it, its sides and rear edge projecting beyond the sides of the body. Its upper 



[45] 



