AMERICAN LANGUAGES. 61 



in the fifth volume of the new series of the transactions of the American Philo- 

 sophical Society. 1 The fact was not made known to Mr. Duponceau until his 

 memoir was in the possession of the French Institute, and is first mentioned in the 

 preface attached to it when printed, accompanied with the reflection that " in sci- 

 ence, especially in metaphysical science, it is not well to be in haste to generalize." 2 



When Mr. Gallatin prepared his " Notes on the semi-civilized nations of Mexico, 

 Yucatan, and Central America," in 1845, he examined Najera's treatise with a good 

 deal of care. Some exceptions or objections were found to his theory, and some 

 affinities to other dialects were detected, yet the theory appeared to him to be 

 substantially correct ; and he drew from thence a moral similar to that of Mr. Du- 

 ponceau. " The distinct character of the Otomi, teaches us we must be cautious 

 in drawing too general conclusions. It appears certain that almost all the Indian 

 languages are similar in their grammatical forms, structure, and general character. 

 But it does not follow that there are no exceptions." 



Mr. Gallatin commenced his labors in this field of inquiry, in 1823, with an 

 attempt to classify the Indian tribes of North America in families according to their 

 respective languages, at the request of his friend Baron Alexander Von Humboldt. 

 That essay having been communicated by Humboldt to M. Balbi, was noticed by 

 the latter in the Introduction to his " Atlas Ethnographique," and there attracted 

 the attention of the American Antiquarian Society, whose officers applied to Mr. 

 Gallatin for a copy. As he had kept none, but had collected many additional 

 materials, the essay was re-written and much enlarged, and occupies the greater 

 part of the second volume of " Archteologia Americana," published in 1836. 



The synopsis was originally intended to embrace all the tribes north of the semi- 

 civilized nations of Mexico; but the want of materials confined the inquiry towards 

 the south to the territory of the United States. Within that territory, east of the 

 Rocky Mountains, nearly all the dialects were ascertained, as well as most of the 

 families of languages north of the United States. 



Mr. Gallatin's general conclusions were not dissimilar to those of Mr. Duponceau. 

 Being less imaginative, and less under the influence of excitement produced by the 

 development of unexpected and extraordinary characteristics in the American 

 languages, he employed a cooler and stricter logic in their examination. He had 

 the advantage also of a wider scope, and a field of observation prepared, to a con- 

 siderable extent, by previous labors. Hence, his view is more comprehensive, and 

 his exposition of details more complete. 



His own account of the result is that " it appears to confirm the opinions 

 already entertained on that subject by Mr. Duponceau, Mr. Pickering, and others ; 

 and to prove that all the languages, not only of our own Indians, but of the natives 

 of America from the Arctic Ocean to Cape Horn, have, as far as they have been 

 investigated, a distinct character, common to all, and apparently differing from any 

 of those of the other continent with which we are most familiar." 



1 In the title to that treatise, the name is written Naxera ; but the author usually employed the 

 other form of Na/era, the sound in Spanish being the same in either case. 



3 Dans les sciences, et surtout dans les sciences metaphysiques, il ne faut pas se hater de generalise! - . 



