short time, the body of the sibaro begins to tremble. He throws off the cloth and rises, and begins, with outstretched arm and a fixed look at the distance, slowly to turn to the rhythm of the music. At the same time a time-keeping convulsion, beginning in his fingers, extends from limb to limb, finally engaging the whole body, till at last the man dances in spasmodic leaps, which continue till he collapses in exhaustion. The music now ceases, and the time has come for the head of the family to question the begu which has taken possession of the medium, first asking its name. The begu, having given its name, then asks why it has been called; and in response to this overture the whole occasion of the trouble is related, and the spirit's good advice is requested. The most important question is, whether there is any hope of the recovery of the patient, and what must be done to secure that desirable result. If the family are not satisfied, as they are not likely to be with the unfavorable answer that is generally given, the music and the dancing are repeated, or the process is applied to the second sibaro. It sometimes happens that the two mediums do not agree in their revelations, and then the drumming and the dancing and the questioning are kept up till they are of accord. If the final answer is that there is no hope for the sick man, he is left to his fate, which has most probably been made more certain by his having had to endure the prolonged torture of witnessing these ceremonies; if a more favorable answer is given, all that the spirit requires as a condition of recovery is performed in good faith.
If the ceremonies are interrupted by the death of the patient during their performance, the music ceases and lamentations take its place; the company go away, leaving only the nearest relatives of the deceased at the house; a few shots are fired, either to drive away evil spirits, or to give notice of the death, and preparations are begun for the funeral.
The existence of cannibalism among the Battas and some peculiarities connected with it suggest some questions respecting its origin. The principal question is whether it is a survival from the original barbarism of the people, or is an offense of later beginning. All the evidence I have met in my investigations points to the latter conclusion as more probable. Among the evidences is the fact that the practice occurs, not among the more degraded tribes, but among those which are most distinguished from their neighbors by intelligence and culture. Other facts, favoring the same view, are: 1. The Battas have traditions of a primitive time when man-eating was unknown among them. They say that it originated during a long civil war, in the course of which the hostility of the opposite factions became so embittered that they went to the extremity of eating captured enemies. 2. Cannibalism is unknown among other people evidently related to the Battas—the inhabitants of the Island of Nias, for instance, whose language is nearly the same, and who are of a lower degree of civili-
