can lay no claim to his power except that of personal merit and the consent of his former equals; and his death is instantly followed by the disruption of his dominions." Even when there has arisen a headship of the compound group which lasts beyond the life of its founder, it remains for a long time not equal in stability to the headships of the component groups. Pallas, while describing the Mongol and Calmuck chiefs as having unlimited power over their dependents, says that the khan had in general only an uncertain and weak authority over the subordinate chiefs. Of the Caffres we read: "They are all vassals of the king, chiefs, as well as those under them; but the subjects are generally so blindly attached to their chiefs that they will follow them against the king." Europe has furnished kindred examples. Of the. Homeric Greeks Mr. Gladstone writes: "It is probable that the subordination of the sub-chief to his local sovereign was a closer tie than that of the local sovereign to the head of Greece." And, during the early feudal period in Europe, allegiance to the local ruler was stronger than that to the general ruler.
In the compound group, as in the simple group, the progress toward stable headship is furthered by the transition from succession by choice to succession by inheritance. During early stages of the simple tribe, chieftainship, when not acquired by individual superiority tacitly yielded to, is acquired by election. In North America it is so with the Aleuts, the Comanches, and many more; in Polynesia it is so with the Land Dyaks; and, before the Mohammedan conquest, it was so in Java. Among the hill-races of India it is so with the Nagas and others. In some regions the transition to hereditary succession is shown by different tribes of the same race. Of the Karens we read that "in many districts the chieftainship is considered hereditary, but in more it is elective." Some Chinook villages have chiefs who inherit their powers, though mostly they are chosen.
Similarly, the compound group is at first ruled by an elected head. Sundry examples come to us from Africa. Bastian says that "in many parts of the Congo region the king is chosen by the petty princes." The crown of Yariba is not hereditary—"the chiefs invariably electing, from the wisest and most sagacious of their own body." And the King of Ibu, says Allen, seems to be "elected by a council of sixty elders, or chiefs of large villages." In Asia it is thus with the Kukis: "One, among all the rajahs of each class, is chosen to be the Prudham or chief rajah of that clan. The dignity is not hereditary, as is the case with the minor rajahships, but is enjoyed by each rajah of the clan in rotation." So has it been in Europe. Though by the early Greeks hereditary right was in a considerable measure recognized, yet the case of Telemachus implies "that a practice, either approaching to election, or in some way involving a voluntary action on the part of the subjects, or of a portion of them, had to be gone through." The like is true of ancient Rome. That the monarchy was