II. The Arctic type, to which the bows just described belong, in use from the Kaviak peninsula to the Mackenzie and Anderson rivers; and III. The Western type, confined to St. Lawrence Island and the mainland of Siberia. I have shown how these three types differ from each other and from the original type, and have expressed the opinion that these differences result from the different resources at the command of the people of different regions. I have also endeavored to account for the fact that we find sporadic examples of the Arctic type, for instance as far south as the Yukon, by the well known habits of the Eskimo in regard to trading expeditions. Outside of the region treated in my paper above referred to, there is very little material for a comparative study of Eskimo bows, either in the Museum or in the writings of travelers. Most writers have contented themselves with a casual reference to some of the more salient peculiarities of the weapon without giving any detailed information. Beginning at the extreme north of Greenland, we find that the so-called “Arctic Highlanders” have hardly any knowledge of the bow. Dr. Kane saw none during his intercourse with them, but Dr. Bessels[N295] mentions seeing one bow, made of pieces of antler spliced together, in the possession of a man at Ita. In Danish Greenland, the use of the bow has been abandoned for many years. When Crantz[N296] wrote it had already gone out of use, though in Egede’s[N297] time it was still employed. It appears to have been longer than the other Eskimo bows. Nordenskiöld[N298] reproduces a picture of a group of Greenlanders from an old painting of the date of 1654 in the Ethnographical Museum of Copenhagen. The man holds in his left hand a straight bow, which appears to have the backing reaching only part way to the ends like a western bow without the end cables, and yet twisted into two cables. If this representation be a correct one, this arrangement of the backing, taken in connection with what Crantz and Egede say of the great length of the bow, would be an argument in favor of my theory that the St. Lawrence Island bow was developed from the primitive form by lengthening the ends of the bow without lengthening the backing. The addition of the end cables would then be an after invention, peculiar to the western bow. In Baffin Land the bow is very rudely made, and approaches very closely to my supposed primitive form. Owing to the scarcity of wood in this region the bow was frequently made of reindeer antler, a substance still more unsuitable for the purpose than the soft coniferous woods used elsewhere. There are in the Museum three specimens of such antler bows, brought from Cumberland Gulf by Mr. Kumlien. [Footnote N295: Naturalist, vol 8, No. 9, p. 869.]