Like the Clallam, their language abounds with the nasal _ng_, both as a prefix and a termination. It has another peculiarity, in great measure its own, in the broad sound of the letter _a_, somewhat as in the words _mass_, _pass_, but even more prolonged. I have distinguished this in the vocabulary by the italic. The letters _f_, _r_, _v_, _z_, are wanting in both languages.
The Clallam and Lummi, as will be observed, contain three classes of cardinals, which I have distinguished as _simple_, _personal_, and of _valuation_, although these terms are not strictly accurate; certain objects, besides men, being counted by the second, and others, as well as money, by the third; I have never fully ascertained the distinctions which govern their use. The words _animate_ and _inanimate_ do not apply; those of _noble_ and _ignoble_, by which the French Missionaries designated the names of objects in the Algonquin languages, probably indicate the idea. Mr. Gallatin, in his "Notes on the Semi-Civilized Nations of Mexico," &c.,[A] mentions the same peculiarity as existing in the Mexican and Maya, in the former of which there are three different terminations suffixed to the simple numbers, according to the objects denoted. A similar distinction is found in the Makak language, and traces of it, at least, are observable in the Pima. I imagine that by inquiry the fact would be found to exist in other Indian tongues. Singularly enough, this feature also characterizes some of the languages of Micronesia.
[A] Trans. Am. Eth. Soc., vol, i., p. 54.
The affinities of the Clallam and Lummi with the Selish are too obvious to require demonstration. Indeed Dr. Latham has already classed the former with the Shewhapmukh, or, as he terms it, Atna, of Frazer River, the northernmost of the Selish dialects. The term Atna, it may be mentioned, is improperly applied as a family name to these languages, as it is a Tákulli (Athabascan or Chepewyan) word, signifying, according to Mr. A. C. Anderson, "strangers."
These two vocabularies were collected, the first at Port Townshend, during a residence of a few months, the other at Simiámoo Bay, while I was connected with the N. W. Boundary Commission. Neither of them underwent more than a partial revision, and inaccuracies have therefore doubtless crept in. They are, however, sufficiently reliable to afford the basis of extending comparisons with the other Selish and the more northern coast families.