4 PREFATORY LETTER. the nineteen others, ten are west of the Stony Mountains ; and seven of these inhabit, south of the sixtieth degree of north latitude, the islands and the narrow tract of land contained between the Pacific Ocean and the continuation of the California!) chain of mountains, as far south as the forty-seventh degree of north latitude. Six of the remaining nine families, the probable remnants of ancient nations, are found amongst the southern tribes, either annexed to the Creek confederacy, or in the swamps of West Louisiana. The three others are the Catawbas, the Pawnees, and the Fall or Rapid Indians. Some new families, or totally distinct languages, will hereafter be found in the quarters already indi- cated: West Louisiana, the wandering tribes on the upper waters of the Arkansas and of the Missouri, and west of the Stony Mountains, in the territory drained by the Columbia river. Many distinct languages or dialects of the Eskimaux, of the Athapascas, and of some of the other great families, will be added to the present enumeration. But I believe that the classification now submitted will, as far as it goes, be found correct. I feel some confidence, that I have not been deceived by false etymologies ; and that the errors, which may be discovered by further researches, will be found to consist in having considered as distinct families some which belong to the same stock, and not in having arranged as belonging to the same family any radically distinct lan- guages forming separate families. The only exceptions, in that respect, refer to the Minetare group and the Shyennes, both stated as being Sioux, and to the Sussees, annexed to the Athapascas, in regard to whom the evidence is not conclusive. It must, however, be understood, that the expression " family," applied to the Indian languages, has been taken in its most extensive sense, and as embracing all those which contained a number of similar primi- tive words, sufficient to show that they must, at some remote epoch, have had a common origin. It is not used in that limited sense in which we designate the Italian, Spanish, and French as languages of the Latin stock, or the German, Scandinavian, Netherlandish, and English as branches of the Teutonic ; but in the same way as we consider the Slavonic, the Teutonic, the Latin and Greek, the Sanscrit, and, as I am informed, the ancient Persian, as retaining in their vocabularies conclu- sive proofs of their having originally sprung from the same stock. Another important observation relates to the great difference in the orthography of those who have collected vocabularies. Those which proceed from the native language of the writer, may be reconciled with- out much difficulty ; and it is almost sufficient, in that respect, to note whether he was an Englishman, a German, a Frenchman, &c. But the guttural sounds which abound in all the Indian languages, and even
