Introduction

Murdoch, John A study of the Eskimo bows in the U.S. National Museum E E1H3 : SMITHSONIAN INSTITUTION. UNITED STATES NATIONAL MUSEUM. A STUDY OF THE ESKIMO BOWS IN THE U. S. NATIONAL MUSEUM. JOHN MURDOCH. From the Report of the Smithsonian Institution, i883-'84, Part II, pages 307-316, and plates I-XII. WASHINGTON GOVERNMENT PRINTING OFFICE. 1890. NOV 3 - 1966 1138704 III.-A STUDY OF THE ESKIMO BOWS IN THE U. S. NATI0:NAL MUSEUM. By John Murdoch. While endeavoring to work out the method of construction of the bows collected by our party* among the Eskimos of Point Barrow, Arctic Alaska, I was led to make a comparative study of all the Es- kimo bows in the National Museum with the view of determining the types of construction to be found among them, and their geographical distribution. It is the purpose of this paper to present the general conclusions arrived at from this study, which I propose to treat in detail in a mo- nograph of the ethnological collection of the expedition, which I am engaged in preparing. I am indebted to Professor Otis T. Mason, of the National Museum, for much cordial assistance and co-operation in the prosecution of this study and in the preparation of the illustrations. 1 have confined myself to the discussion of the forms of bow in use among the Western Eskimos, namely, those inhabiting the shores of the Arctic Ocean from the Mackenzie River westward to Bering Strait, of Bering Sea and the Gulf of Alaska, with the outlying Asiatic branches on the mainland of Siberia and Saint Lawrence Island. These regions are very fully represented in the Museum by the collections of Boss and MacFarlane from the Mackenzie River region, Dall, Turner, Nelson, and others, from the Alaska coast. Nelson, from Saint Lawrence Island, and the North Pacific Exploring Expedition, from the mainland of Siberia, while the material from the eastern tribes is very scanty and unsatisfactory. Starting from the island of Kadiak in the south, there is abundant material from the whole coast as far as the northern shore of Norton Sound, from the Diomede Islands, Point Hope, Wainwright's Inlet, Point Barrow, and the Mackenzie region, as well as from Saint Lawrence Island and the Siberian shore. Unfortunately, the region about Kotzebue Sound, including the great peninsula between this and Norton Sound, is not represented in the collection. The field of investigation is practically untrodden. Although it has long been known that the Eskimos used cords of elastic sinew to coun- -- teract the brittleness and lack of elasti^jity in the spruce and fir the -- only wood at their disposal for making bows authors have confined U. S. International Polar Expedition to Point Barrow, 1881-83. 307 308 REPORT OF NATIONAL MUSEUM, 1884. themselves tx) a general statement of_ the fact, without going intojbe details of nonstriiction. I have found that the bows of the Western E skimos are constructed upon three well-defined types, each quite distinctly limited m_it8 geo- graphical distribution. No one of these types can be considered asTIe^ rived from either of the others, but all are plainly developed from a single original type still to be found only slightly modified in the region around Cumberland Gulf, where the mechanical arts seem to have re- mained in many respects more primitive than either in Greenland or Alaska. (Fig. 1, back and side view of a bow of reindeer antler from Cumberland Gulf, No. 34053, collected by L. Kumlien.)t The main part of the reinforcement or backing always consists of a continuous piece of stout twine made of sinew, generally a three-8tran3 braid, but sometimes a twisted cord, and often very long (sometimes forty or fifty yards in length). One end of this is spliced or knotted into an eye, which is slipped round one "nock" of the bow, usually the upper one. The strands then pass up and down the back and round A the nocks. comparatively short bow, having along its back some dozen or twenty such plain strands, and finished off by knotting the end about the " handle," appears to have been the original pattern. The , bow from Cumberland Gulf (Fig. 1) is such a one, in which the strands have been given two or three turns of twist from the middle. They are kept from untwisting by a "stop" round the handle, which passes be- tween and around the strands. ii The three Western Eskimo types may be described as follows: I. The Southern Type. Of this there are two slightly different patterns, found often side by side. *For example: "They ingeniouely remedy the defect [i. �., the want of elasticity m the material] by securing to the back of the bow and to the knobs at each end a quantity of small lines, each composed of a plat or 'sinnet' of three sinews. The number of lines thus reaching from end to end is generally about thirty; but besides these several others are fastened with hitches round the bow, in pairs, commencing eight inches from one end and again united at the same distance from the other, making the whole number of strings in the middle of the bow sometimes amount to sixty. These being put on when the bow is somewhat bent in the contrary way, produce a spring so strong as to require considerable force as well as knack in stringing it and giving the requisite velocity to the arrow." (Parry's Second Voyage, p . 511.) "These bows [in the Yukon delta] are made of spruce, which has little elasticity when dry and is very liable to break. To remedy this defect the bow is bound with cords twisted from deer sinew [as shown in a figure, which gives the general appearance very well]. This gives it great strength and overcomes the brittleness of the V ood." tPalVs Alaska^and its Resources, p. 228.) "Only some old bows had a finer form. They were larger and made with care; for instance, they were covered with birch bar^ and strengthed by an artistic plaiting of sinew on the outer side." (Nordenekiold's Voyage of the Vega, ii, p. 108.) tWhen a scale accompanies a figure each division represents one inch. Figures without a scale are natural size, unless otherwise specified. ESKIMO BOWS, 309 A 1. broad and flat bow tapering to the nocks, which are formed by simple rounded knobs, and narrowed and thickened at the handle so as to be half as wide and twice as thick as the broadest part of the bow. The back is flat and the belly often keeled from end to end, and this keel is sometimes deeply furrowed for its whole length ; the edges are generally square and sometimes grooved longitudinally. (Figs. 2, 3^ and 4 show the general pattern of this type.) The bow when unstrung is either straight, slightly sprung toward the back, or, rarely, arched, and is sometimes stiffened along the back with an extra rib of wood or ivory. The backing is occasionally tightened with wedges. Its length is from 50 inches to 5 feet, averaging about 55 inches, with its greatest breath about 2 inches (rarely 2^ to 2^ inches). A 2. bow of essentially the same size and outline as the first form, but with about one foot of each end bent up toward the back so as to lie parallel to the string when the bow is strung, as in the Tatar bow, with the backing generally stretched over bridges at the bends. (Fig. 6, No. 36028, from the mouth of the Kuskoquim Eiver, collected by E. W. Nelson.) The backing starts in the ordinary way and consists wholly of straight parallel strands passing round the nocks, or secured by pairs of halfhitches at various points on the bow. The last strand is wrapped spirally round the others to keep them from spreading apart, though occasionally one end of the cable is wrapped with a separate piece, and A very rarely the whole wrapping is separate from the rest. separate piece of twine, thong, or withe serves to stop the backing down to the handle, and there are sometimes other separate stops on the broad part of the bow (as in Fig. 2). The whole of the broad part of the bow is occasionally seized down with spaced spiral turns of twine (Fig. 4, No. 7972, from Bristol Bay, collected by Dr. Minor), which, in one case at least, are made by the end of the last strand. The strands of the back- ing vary in number from 11 to 37 (usually about 25). They are some- times all of the same length, in which case the outer strands are hitched round the bow a short distance from the nocks instead of passing round the latter (Fig. 2, back and side view, and Fig. 2 a, one end of No. 36032, from near Cape Eomanzoff, collected by E. W. Nelson). More commonly 4-22, usually 6 or 7 strands are shorter than the rest and only extend from the broadest part of one end to the corresponding point at the other (Fig. 3, No. 72408, from Bristol Bay, collected by the late C. L. McKay. Fig. 3 a, the broadest part of the same bow, to show the attach- ment of the short strands), thus giving special strength and elasticity to the middle of the bow. These shorter strands are sometimes the outer ones of the backing, but more commonly about the middle of it. Earely, as in the case of one bow from