Ellis’s figure (plate opposite p. 132) shows a bow of the Tatar shape, but gives no details of the backing, except that the latter appears to be twisted. We have no published descriptions of the bows used in other regions. As far as I have been able to ascertain, the practice of backing the bow with cords of sinew is peculiar to the Eskimo, though some American Indians stiffen the bow by gluing flat pieces of sinew upon the back. One tribe of Indians, the “Loucheux” of the Mackenzie district, however, used bows like those of the Eskimos, but Sir Alexander Mackenzie[N303] expressly states that these were obtained from the Eskimo. [Footnote N303: Voyages from Montreal . . . to the Frozen and Pacific Oceans, p. 48.] _Arrows._--With these bows were used arrows of various patterns adapted for different kinds of game. There are in the collection fifty-one arrows, which are all about the same length, 25 to 30 inches. In describing these arrows I shall employ the terms used in modern archery[N304] for the parts of the arrow. The greatest variation is in the shape and size of the pile. The stele is almost always a straight cylindrical rod, almost invariably 0.4 inch in diameter, and ranging in length from 20 to 28 inches. Twenty-five inches is the commonest length, and the short steles, when not intended for a boy’s bow, are generally fitted with an unusually long pile. From the beginning of the feathering the stele is gradually flattened above and below to the nock, which is a simple notch almost always 0.2 inch wide and of the same depth. The stele is sometimes slightly widened just in front of the nock to give a better hold for the fingers. The feathering is 6 or 7 inches long, consisting of two, or less often, three feathers. (The set of sixteen arrows from Sidaru, two from Nuwŭk, and one from Utkiavwĭñ, have three feathers. The rest of the fifty-one have two.) The shaft of the feather is split and the web is cut narrow, and tapered off to a point at each end (Fig. 181). The ends of the feathers are fastened to the stele with whippings of fine sinew, the small end of the feather which, of course, comes at the nock, being often wedged into a slit in the wood (with a special tool to be described below), or else doubled back over a few turns of the whipping and lashed down with the rest. The small end of the feather is almost always twisted about one turn, evidently to make the arrow revolve in flight, like a rifle ball. Generally, if not universally, the feathering was made of the feathers of some bird of prey, falcon, eagle, or raven, probably with some notion of giving to the arrow the death-dealing quality of the bird. Out of the fifty-one arrows in the collection, only nine are feathered with gull’s feathers, and of these all but two are new, or newly feathered for sale to us.[N305] Dr. Simpson[N306] says that in his time “feathers for arrows and head-dresses,” probably the eagles’ feathers previously