failing this, the nobles appoint, and, "unless positive disqualification exists, the eldest son is usually chosen." Africa furnishes evidence of varied kinds. Though the Congo people, the coast negroes, and the inland negroes, have formed societies of some size and complexity, notwithstanding that kinship through females obtains in the succession to the throne, yet we read of the first that allegiance is "vague and uncertain"; of the second, that, save where free in form, the government is "an insecure and short-lived monarchic despotism"; and of the third, that, where the government is not of mixed type, it is "a rigid but insecure despotism." Meanwhile, in the two most advanced and powerful states, stability of political headship goes along with departure, partial or complete, from succession through females. In Ashantee the order of succession is "the brother, the sister's son, the son"; and in Dahomey there is male primogeniture. Further instances of this transition are yielded by extinct American civilizations. Though the Aztec conquerors of Mexico brought with them the system of kinship through females, and consequent law of succession, yet this law of succession was partially, or completely, changed to succession through males. In Tezcuco and Tlacopan (divisions of Mexico) the eldest son inherited the kingship; and in Mexico the choice of a king was limited to the sons and brothers of the preceding king. Then, of ancient Peru, Gomara says, "Nephews inherit, and not sons, except in the case of the Incas": this exception in the case of the Incas having the strange peculiarity that "the first-born of this brother and sister [i. e., the Inca and his principal wife] was the legitimate heir to the kingdom"—an arrangement which made the line of descent unusually narrow and definite. And here we are brought back to Africa by the parallelism between the case of Peru and that of Egypt. "In Egypt it was maternal descent that gave the right to property and to the throne. The same prevailed in Ethiopia. If the monarch married out of the royal family, the children did not enjoy a legitimate right to the crown." When we add the statement that the monarch was "supposed to be descended from the gods, in the male and female line," and when we join with this the further statement that there were royal marriages between brother and sister, we see that like causes worked like effects in Egypt and in Peru. For in Peru the Inca was of supposed divine descent; inherited his divinity on both sides; and married his sister to keep the divine blood unmixed. And in Peru as in Egypt there resulted royal succession in the male line, where, otherwise, succession through females prevailed.
With this process of transition from the one law of descent to the other, implied by these last facts, may be joined some processes which preceding facts imply. In New Caledonia a "chief nominates his successor, if possible, in a son or brother": the one choice implying descent in the male line and the other being consistent with descent in either male or female line. And in Madagascar, where the system of