from the collections of the late Dr. Carl Hermann Berendt and the Abbé E. C. Brasseur (de Bourbourg).

When in Spain, in 1888, I found in the Royal Library the MS. of the earlier portion of Sahagun’s “History of New Spain” in Nahuatl. I described it in (42).

The term “Anahuac” has long been applied to the territory of Mexico. Dr. E. Seler, of Berlin, published an article asserting that this was an error, and devoid of native authority. In (43) I pointed out that in this he was wrong, as early Nahuatl records use it in this sense.

The Alaguilac language of Guatemala, long a puzzle to linguistics, is shown in (44) to be an isolated dialect of the Nahuatl.

Nos. (45), (46), (47), (49) and (54), have been already mentioned.

The term Chilan balam, which may be freely rendered “the inspired speaker,” was the title of certain priests of the native Mayas. Many records in the Maya tongue, written after the conquests, go by the name of “the Books of Chilan Balam.” They have never been published, but copies of them, made by Dr. Berendt, are in my possession. Their purpose and contents were described in (50).

There are reasons for believing that previous to the arrival of the Cakchiquels in Guatemala its area was largely peopled by Xincas. Of this little-known stock I present in (58) three extended vocabularies, from unpublished sources, with comments on the “culture-words.”

Some apparent but no decisive affinities between the Otomi of Mexico and the Tinné or Athapascan dialects are shown in (55); and in (59) the ancient Guetares of Costa Rica are proved, on linguistic evidence, to have been members of the Talamancan linguistic stock.

The Matagalpan is an interesting family, first defined in The American Race, and in (60) more fully discussed, as they survive in San Salvador.

In (61) some unpublished vocabularies from the tribe of the Ramas, on the Mosquito coast, place them as members of the Changuina stock, most of whom dwelt on the Isthmus of Panama.

IV. South American and Antillean Languages.