The arrow was called sawa.
Of all the specimens of arrows in the University Museum, scarcely any show such perfect workmanship as those of Ishi. His proportions and finish are of very high order.
At the time of the rediscovery of the remnant of the tribe, a number of arrows were secured from the huts, which doubtless represent his average work. Later, while with us, he made scores of arrows of various shapes and sizes. Apparently some arrows, those of great length, measuring a yard, and having large heads, were purely for ornamental purposes, or intended to be given as presents, or possibly to be used in time of war. His hunting shafts were of two kinds—obsidian pointed, and blunt. For shooting small game, such as birds and rabbits, the latter were used. For killing deer, bear, and predatory animals, sharp arrows wore used. Here, if the objcet shot at were missed, a broken arrow-point resulted. The arrow shafts were made of several kinds of wood. Those obtained from his hut in Tehama County seem to be of hazel, humoha, and this was a favorite wood with him. A native bamboo-like reed was also a great favorite. Dogwood and mountain mahogany he also used. Other shaft woods pointed out by him were bakanynu'an (Philadelphus Lewisii), sawa'i ("arrow bush," Paeonia Brownii), and loko and habaigili'i, unidentified. Later, as the result of a modification of ideas he underwent in our company, he adopted the commercial 5/16-inch birch dowel as the ideal material, probably because of its accessibility.
In the case of cane arrows, a wooden "foreshaft," six to eight inches long, was invariably added, and such foreshafts were sometimes added to wooden arrows. They were of hazel, buckeye (bahsi), wild currant (wahsu'i), and perhaps other woods. The foreshaft was normally heavier material than the main shaft.
In general it may be said that his typical hunting arrow was a hazel stick, with a foreshaft, the entire length being 29 inches. The diameter at the middle was 11/32 inch ; and the total weight was 330 grains. The feathering of the arrow consisted of three plumes from a buzzard's wing, 4 3/4 inches long, % inch wide. They were trimmed straight to the forward end. where their width was about 1/8 inch, and terminated 3/4 inch from the nock of the arrow. At each end the feathers were bound down with sinew.
In gathering wood for arrows he generally selected the tall, straight shoots of hazel where it grew in competition with other shrubs or trees, cutting them about a yard long, their greatest diameter being little more than three-eighths of an inch. These he stripped of bark with his thumb nail. He always made arrows in groups of five. Thus he would select the best of his sticks, and collecting them in groups, bind them together securely with a cord. In this bundle they were permitted to season, lying in a horizontal position. After any period from a week to a year these sticks might be used.
The first process in manufacture was that of straightening his shafts. To do this he either made a small heap of glowing embers from a fire or utilized a hot stone. He applied pressure with his thumbs on the convex side of any irregularity or bend in a shaft, and holding this near the heat, passed the wood back and forth before the stone or coals. When the wood was warm, it gave very readily to pressure. In less than a minute any curve or crook could be straightened out. The wood after cooling always retained its new position. Glancing down the axis of his shaft from time to time, Ishi gauged its straightness. To burn or discolor the wood was evidence of bad technique. Smoothing was accomplished by scraping and rubbing the arrow shaft between two pieces of sandstone. He sometimes finished the shaft by rolling it back and forth on the thigh with his right palm, while he worked it with a piece of sandstone held in his left hand. By this means he could "turn" a shaft almost as accurately as if a lathe were used.
Where a foreshaft was to be added, the length of the main shaft was 21 inches. At the smaller end he cut a notch for the bow string with a bit of obsidian, making this nock 5/32 of an inch wide and 3/16 inch deep. In larger arrows he deepened this to 1/2 inch. The other end of the shaft was next drilled out to accommodate the foreshaft.
His method of drilling was as follows: Placing a sharp piece of bone, point up, in the ground, and stead3nng it with his toes, he rotated the shaft perpendicularly upon this point. The motion here was identical with that employed in making fire by means of a drill and base stick, the stick being rolled between the palms with downward pressure. The excavation averaged an inch in depth and a quarter of an inch in diameter, and ran to a point. During this drilling process the lower end of the shaft was tightly bound with sinew or cedar cord to keep it from splitting. One end of the foreshaft was formed into a spindle and made to fit this socket, leaving a slight shoulder where the two segments met. Salmon glue or resin was used to secure union, and the joint was bound with macerated tendon for the distance of an inch or more.