The statement that each naukrary was obliged to furnish one ship can hardly be true of the time before Solon: as Pollux states it, we should be led to conceive that he only infers it from the name ναύκραρος (Pollux, viii, 108), though the real etymology seems rather to be from ναίω (Wachsmuth, Hellen. Alt. sect. 44, p. 240). There may be some ground for believing that the old meaning, also, of the word ναύτης connected it with ναίω; such a supposition would smooth the difficulty in regard to the functions of the ναυτόδικαι as judges in cases of illicit admission into the phratores. See Hesychius and Harpokration, v. Ναυτόδικαι; and Baumstark, De Curatoribus Emporii, Friburg, 1828, p. 67, _seq._: compare, also, the fragment of the Solonian law, ἢ ἱερῶν ὀργίων ἢ ναῦται, which Niebuhr conjecturally corrects. Rom. Gesch. v. i, p. 323, 2d ed.; Hesychius, Ναυστῆρες—οἱ οἰκέται. See Pollux, Ναῦλον, and Lobeck, Ῥηματικὸν, sect. 3, p. 7; Ἀειναῦται παρὰ Μιλησίοις? Plutarch, Quæst. Græc. c. 32, p. 298. But the phratries and gentes are a distribution completely different from this. They seem aggregations of small primitive unities into larger; they are independent of, and do not presuppose, the tribe; they arise separately and spontaneously, without preconcerted uniformity, and without reference to a common political purpose; the legislator finds them preëxisting, and adapts or modifies them to answer some national scheme. We must distinguish the general fact of the classification, and the successive subordination in the scale, of the families to the gens, of the gentes to the phratry, and of the phratries to the tribe,—from the precise numerical symmetry with which this subordination is invested, as we read it,—thirty families to a gens, thirty gentes to a phratry, three phratries to each tribe. If such nice equality of numbers could ever have been procured, by legislative constraint[82] operating upon preëxistent natural elements, the proportions could not have been permanently maintained. But we may reasonably doubt whether it did ever so exist: it appears more like the fancy of an author who pleased himself by supposing an original systematic creation in times anterior to records, by multiplying together the number of days in the month and of months in the year. That every phratry contained an equal number of gentes, and every gens an equal number of families, is a supposition hardly admissible without better evidence than we possess. But apart from this questionable precision of numerical scale, the phratries and gentes themselves were real, ancient, and durable associations among the Athenian people, highly important to be understood.[83] The basis of the whole was the house, hearth, or family,—a number of which, greater or less, composed the gens, or genos. This gens was therefore a clan, sept, or enlarged, and partly factitious, brotherhood, bound together by,—1. Common religious