also exhibiting great variations. The early missionaries were slow to discover these facts, and it was a source of discouragement for them to learn that, after long study to acquire a language, it was spoken by only a single group of the natives, while adjacent to them dwelt others who spoke a totally different one.[17] [Illustration: “SINGING-GIRL” SCULPTURED IN TRACHYTE From débris of Temple 22, Copan. Slightly larger than life ] Even where a group of Amerinds speak related languages, or dialects, there are, and were, such wide variations that the one is not understood by those speaking the other. Therefore we have in North America not only a large number of distinct languages, but within these separate languages an immense number of dialects or sub-languages, sometimes as many as twenty in one stock varying from each other as much as, say, English and German. At least sixty-five of the separate stock languages are distinguished in North America which appear so radically separated from each other that it is believed impossible that they ever should have sprung from the same parent, unless it may have been at a time so remote as to be beyond the scope of present investigation. In the classification according to these languages it has been necessary to have a general designation for each stock, and in selecting the names to be thus used, Powell and others have observed the law of priority of mention, as far as possible, and have derived the stock name from the author first mentioning it in print since 1836, the date of Gallatin’s great work, which is taken as the foundation. The termination “an,” or “ian,” is added to distinguish the family or stock name from a tribal name, for often a tribe bears the name given to the whole stock. As examples, Algonquian may be mentioned as a stock name taken from the tribal name of Algonquin, and Mayan from the tribal name, Maya. This is not always strictly adhered to outside of the Bureau of Ethnology because of its frequent inconvenience, and in the case of Mayan the term Maya is preferably used by some investigators and writers as being simpler, and Brinton gives it as the stock name.[18] Following the distribution of tribes as closely as possible at the time of the first contact with white men, Powell and his able associates of the U. S. Bureau of Ethnology in Washington have produced a map, based on Gallatin’s.[19] The separate stocks north of Mexico are each represented by a different colour, every colour standing for a variation in language as great as that between Hebrew and English, not related as English and Spanish. Fifty-eight are thus shown, but south of the Mexican border are perhaps a dozen more. Continuous study may succeed in bringing some of the stocks into relationship or in dividing them still further. In their beginning, languages probably changed rapidly; memory was deficient, intercourse slight, and comparatively short separations of tribes speaking originally the same