Linton — The Marqncsas Islands 303 



which had been worn ont or damaged. The names of the Pua ]\Ia'u canoes were 

 Vaimakamaka, the largest and most important, Vaioto, Taipehe and IMeaupiau. 



The canoe normally consisted of nine parts, the underbody, bow piece, stern 

 piece, two side planks, and four strips used to cover the side seams. Certain other 

 parts were occasionally added. Each of the large sections was usually hewn from 

 a single piece of wood, while the strips were made of bamboo. 



The underbody of one large canoe and part of a second underbody, which 

 had been made in two pieces, were seen in the valley of Hakaui, Nuku Hiva. 

 The history of these objects had apparently been forgotten by the Hakaui 

 natives, but according to Atu Ona informants one of the underbodies belonged to a 

 canoe called Mo'ote which was made in that valley by a tuhuna named Kaneautia. 

 This canoe was sent to the chief of Hakaui by a chief of Atvi Ona in exchange for 

 enanwa, a yellow paint or dye used for anointing the body, prepared in Nuku Hiva 

 by a secret process. The debris of a great temanu tree from which this canoe 

 is said to have been made is still to be seen in Atu Ona. 



Hakaui informants said that the two underbodies had belonged to separate 

 craft, not to a single double canoe. On the other hand Christian (10, p. 180) 

 mentions two old canoes in this valley which, from their location, seem to be the 

 same as those here referred to and gives a photograph of a large double canoe, 

 uncjuestionably Marquesan, w4iich he implies belongs with the reference. In this 

 photograph each of the canoes appears to be complete in itself while the attach- 

 ment consists of three cross poles, easily removable. It seems probable, therefore, 

 that these canoes were used both singly and together. 



A study of the complete underbody — which had been exposed to the 

 weather and was much rotted and warped — brought out the following particulars : 



It was 40 feet, 6 inches in length, with a maximum central width of 4 feet, which 

 tapered to 2 feet, 6 inches at the ends. The original depth could not be accurately determined. 

 The two ends of this underbody were identical. 



The section of an underbody was much better preserved, having been stored under an 

 overhanging clifif which protected it from the weather. If the underbody of which this had 

 formed a part was similar to the complete specimen it must have been over 60 feet in length. 

 Some old canoe models, which appear to be very accurate, clearly show that the two ends of 

 underbodies were not always the same, the forward end often being made much broader and 

 deeper to facilitate the firm attachment of the bow piece. The extreme shallowness and narrow- 

 ness of the outer end of the section proves that it was the stern, and if we assume that the 

 whole underbody was of the broad bowed type it is possible that it was not more than 40 to 

 45 feet in length. The dimensions of the section were as follows: Length 34 feet, 11 inches; 

 maximum width (at a point 4 feet from the inner end of the section ) 3 feet, 4 inches ; width at 

 stern, i foot, 6 inches; maximum depth (outside) i foot, 9 inches; depth at tip of stern 4^ 

 inches. The walls were 1^/2 inches thick at the upper edge, increasing gradually to 2^ inches 

 in the center of the bottom. The cavity of the underbody had, in cross section, the form of 

 a rectangle with rounded corners. The middle of the canoe was flat bottomed, becoming 

 gradually more rounded toward the stern. 



[43] 



