The most primitive type now existing is that found among the Eastern Esquimaux, consisting of a piece of drift-wood (or two or more pieces of wood, whale-rib, or horn spliced together [2]) "backed" with a cord of plaited or twisted sinews, which is fixed by an eye-loop to one of the nocks of the bow, and is wound up and down between the nocks, passing round them. A bundle of cords is thus formed, stretched tightly between the ends of the bow, and to a great extent covering the "back." Sir Martin Frobisher described this form of sinew backing as "not glued to, but fast girded on" Sometimes, even in the roughest specimens, the longitudinal lacing is gathered up into a compact rope by spiral binding. There are further cross lacings passing round the body of the bow and the backing, so as to keep the latter close against the former.

A specimen from the Barrow collection in the British Museum from Whale Fish Island (?) is backed with a lacing of raw hide, gathered into two bundles twisted up, with a cross lacing of the same material. A second in the British Museum from Parker Bay, Victoria Land, consists of roughly spliced bones reinforced at the back with short whalebone (baleen) strips. Crantz also mentions this material as used for backing Greenland bows. I have figured (Plate V, Fig. 1) a bow of this simple type obtained from the Eastern Esquimaux by Captain Lyon, R.N., circa 1825. In this specimen the body is of a single piece of drift pine, thick and clumsy; the tension of the sinew backing in this, as in most cases, causes the bow to assume the opposite curve to that of the weapon when strung for use. It frequently happens that the bows of the Eastern Esquimaux assume a very unsymmetrical shape, from the rough splicing and the unequal strength of the parts.

When the body is composed of more than one piece of bone, the pieces may be united by being overlapped and fixed with sinew thongs passed through holes, or with rivets of old ships' nails, or by splicing. In the latter case the joints are often strengthened by additional short pieces placed on back and front, with a splicing line bound round the whole.

The Esquimaux bows have been so well described in detail by Mr. John Murdoch [3] that I need not enter into the details of the various modes of " backing " characteristic of the different regions of Arctic North America, my purpose being merely to describe the prominent types which seem to indicate the various epochs in the history of this weapon. Mr. Murdoch refers his three well-defined western types to a single primitive ancestral form, of which the bows of the Eastern Esquimaux with simple backing, such as that described above, are but slightly modified survivals. He cites as an example a bow from Cumberland Gulf of very primitive construction.

In the western regions of the Esquimaux, where the materials are of better quality, and the workmanship far superior, owing no doubt to the ready access to the higher civilization of the West, in the proximity to the Asiatic Continent, the style of backing is more complicated. The cross lacing round the wood, especially, is usually more elaborate; occasionally, as in the specimen figured (Fig. 2), obtained by Capt. Beechey in 1826, to the N.E. of Icy Cape, forming a close transverse binding over the greater length of the bow, the central grip and two extremities alone remaining free. By this means the longitudinal cords are brought into close contact with the wood, and the whole becomes stronger and far more compact. The backing is wound between the nocks as in the eastern forms, but the strands are gathered up closely to form a compact rope-like bundle, kept close against the body by the transverse binding, except at the ends where the strands are more free and less compactly packed. Generally the sinews are twisted together into a single or double rope by means of small ivory levers.