218 RETORT OF NATIONAL MUSEUM, 1900. 



tied around the head at one end and at the top of the shaft at the other 

 end. When this weajion is s(^t ready for action, the barbed head is 

 placed on the end of the foreshaft. The line having been wrapped 

 neatly around the top of the shaft, almost to its end, a loop or slip-knot 

 is formed at the hist turn, and drawn tight. When the game is struek, 

 the head is withdrawn, the slip-knot untied, the line unwound, and the 

 heavy portion of the shaft drops into the water, the feather projects 

 into the air, and the apparatus acts both as a drag and as a signal. 

 Excepting the iron point, which might easily be replaced b}^ one of 

 bone, the whole apparatus is aboriginal, and the wide prevalence of 

 this particular combination of parts leads to the belief that we have 

 here an early and unchanged American harpoon arrow. It is interest- 

 ing also from the point of view before mentioned, that it is a step in 

 the progress of the toggle head. If a Columbia River Indian were to 

 fasten a spur on the end of the cup-shaped socket-, the combined barbed 

 and toggle heads, to be more fully illustrated and described, would 

 be realized. This form of harpoon head, in which the socket is on the 

 movable part instead of being in the end of the shaft, is quite well 

 diffused in the Amazon drainage and on the Pacific coast. It is not 

 found in the shell heaps or mounds of eastern United States, but is 

 connnon in western Canada and universal among the Eskimo. 



The harpoon arrows of the tribes in British Guiana are used for 

 shooting fish, pacu {Pacu myletes), which abound at all seasons of the 

 year, according to Im Thurn, in most of the large rivers of Guiana. 

 When the river is high and the water is turbid with rain the pacu 

 are distributed equally in all parts of the stream and are almost invis- 

 ible. When, however, in the dry season, the river is low and the 

 water clear, when the rocks which form the rapids are partially 

 uncovered, and the "pacu grass," a small water plant {Lacis)^ which 

 clothes these rocks, comes into flower, then the pacu collect at the 

 falls to feed on the leaves. Large numbers of Indians then camp at 

 the sides of the falls to shoot these fish. Such a scene is highly pic- 

 turesque. The place is generally a wide extent of river bed, apparently 

 inclosed by the foi'ested banks, and entirely occupied by a curious 

 confusion of rocks and white, rushing water. On a rock in the midst 

 of, and almost covered by the tumbling water, stands an Indian, his 

 feet crushing the delicate, star shaped, pink flowers of the lacis, and 

 every muscle in his naked, cinnamon-colored body bearing witness to 

 the intensitj^ of his watch. His bow is half drawn, the arrow is in 

 position, but its point rests idly on the rocks. The water is rushing 

 and tumbling so wildly that an unpracticed eye can see nothing below 

 its surface. But the Indian sees. Quickly the bow is raised, the aim 

 is taken, the arrow flies, and its shaft is there, dancing and tumbling 

 in the water, carried here and there ))y the terrified rusnes of an 

 unseen pacu. in the body of which the arrowhead is embedded. But 



1 



