as Romance languages, to show that they are derived from
Latin, in contrast with the closely related tongues of
the English, Dutch, and German peoples, which came from
another common stock, the Teutonic. But even the Teutonic
and the Romance languages are not entirely different.
The similarity in both groups of old root words, like
the numbers from one to ten, point again to a common
origin still more remote. In this way we may trace a
whole family of languages, and with it a kinship of
descent, from Hindustan to Ireland. Similarly, another
great group of tongues--Arabic, Hebrew, etc.--shows a
branch of the human family spread out from Palestine and
Egypt to Morocco.

Now when we come to inquire into the languages of the
American Indians for evidence of their relationship to
other peoples we are struck with this fact: we cannot
connect the languages of America with those of any other
part of the world. This is a very notable circumstance.
The languages of Europe and Asia are, as it were, dovetailed
together, and run far and wide into Africa. From Asia
eastward, through the Malay tongues, a connection may be
traced even with the speech of the Maori of New Zealand,
and with that of the remotest islanders of the Pacific.
But similar attempts to connect American languages with
the outside world break down. There are found in North
America, from the Arctic to Mexico, some fifty-five groups
of languages still existing or recently extinct. Throughout
these we may trace the same affinities and relationships
that run through the languages of Europe and Asia. We