We now come to the great Homeric appellatives, Danaan, Argive, and Achæan. As Thucydides has said (i. 3), Δαναοὺς δὲ ἐν τοῖς ἔπεσι, καὶ Ἀργείους, καὶ Ἀχαιοὺς ἀνακαλεῖ. Why has the great historian arranged the three names in this order? It cannot be with reference to the comparative frequency of their use: for the first is employed the smallest number of times, and the third is by far the most frequent. For the present let us postpone seeking after the cause; and simply note it as probable, even if no more than probable, that there _is_ a cause. _Modes of formation for Names of Peoples._ Let me, by way of preface to the examination of these names, consider the various ways in which, so far as we have the means of tracing them (which is but to a limited extent), the names attached by Homer to the inhabitants of particular countries are derived. They appear to come either 1. From an eponymist directly, who is also an original founder, as Δαρδανοὶ, Τρῶες, from Dardanus, and Tros, in relation to Dardania and Troja respectively. 2. From the land they live in: and thus from an eponymist, if there has originally been one for the territory. For example, we find Ἰθακήσιοι from an island Ἰθακὴ, which again was derived from Ἰθακός. In a case like this, when the appellation of the people comes not directly, but mediately from the name-giver, a territorial designation intervening, we can draw no inference as to the oneness of race between them and him. Thus in the case before us, Ἰθακήσιοι, though connected with Ἰθακὴ, has not as of necessity, any connection whatever with Ἰθακὸς personally. 3. From the land they live in, as described by its most prominent physical characteristic. For example, the Thracians (Θρῇκες), must evidently be so called from the roughness of the country, as a cognate word to τρῆχυς, which is thus applied to Ithaca, τρηχεῖ’, ἀλλ’ ἀγαθὴ κουρότροφος. Odyss. ix. 27. Again, from Αἰγίαλος, the district afterwards called Achæa, we have, in later Greek[613], the name Αἰγιαλεῖς for the inhabitants. This does not occur in Homer, but we have what is equivalent to it in the name of Αἰγιάλεια, who was wife of Diomed, and daughter of Adrastus, the former king of Sicyon in Ægialus. This is an instance of the application of the principle, not to the inhabitants at large, but to an individual inhabitant. 4. The name of a population may be derived secondarily from that of another population. Thus while we must derive Ἕλληνες from Ἕλλας, this in its turn can only be drawn from the Ἕλλοι. 5. In the single case of the Athenians, we find the name of a population derived from that of a deity. 6. It is presumable, though not certain, that entire populations took their name from ruling individuals or races. It seems hardly possible to explain, for example, the name Καδμεῖοι, which nowhere connects itself with any of the foregoing sources of eponymism, otherwise than by reference to an individual Cadmus, whom Homer mentions in Od. v. 333.