Both these states, though in name republics, had certain magistrates, called commonly, in history, kings. These kings were, however, in fact, only military chieftains, commanders of the armies rather than sovereign rulers of the state. The name by which such a chieftain was actually called by the people themselves, in those days, was _tyrannus_, the name from which our word _tyrant_ is derived. As, however, the word _tyrannus_ had none of that opprobrious import which is associated with its English derivative, the latter is not now a suitable substitute for the former. Historians, therefore, commonly use the word king instead, though that word does not properly express the idea. They were commanders, chieftains, hereditary generals, but not strictly kings. We shall, however, often call them kings, in these narratives, in conformity with the general usage. Demaratus, who had fled from Sparta to seek refuge with Darius, and who was now accompanying Xerxes on his march to Greece, was one of these kings. It was a peculiarity in the constitution of Sparta that, from a very early period of its history, there had been always two kings, who had held the supreme command in conjunction with each other, like the Roman consuls in later times. This custom was sustained partly by the idea that by this division of the executive power of the state, the exercise of the power was less likely to become despotic or tyrannical. It had its origin, however, according to the ancient legends, in the following singular occurrences: At a very early period in the history of Sparta, when the people had always been accustomed, like other states, to have one prince or chieftain, a certain prince died, leaving his wife, whose name was Argia, and two infant children, as his survivors. The children were twins, and the father had died almost immediately after they were born. Now the office of king was in a certain sense hereditary, and yet not absolutely so; for the people were accustomed to assemble on the death of the king, and determine who should be his successor, choosing always, however, the oldest son of the former monarch, unless there was some very extraordinary and imperious reason for not doing so. In this case they decided, as usual, that the oldest son should be king. But here a very serious difficulty arose, which was, to determine which of the twins was the oldest son. They resembled each other so closely that no stranger could distinguish one from the other at all. The mother said that she could not distinguish them, and that she did not know which was the first-born. This was not strictly true; for she did, in fact, know, and only denied her power to decide the question because she wished to have both of her children kings.