rictly intermediate in its character. It has accordingly been
designated by the term Sub-Semitic.[127] But it shares this character
with all or nearly all the other African languages, which bear strong
marks of affinity to the Egyptian and Semitic tongues. On this
subject Dr. Latham says, "That the uniformity of languages throughout
Africa is greater than it is either in Asia or in Europe, is a
statement to which I have not the least hesitation in committing
myself."[128] To the north the Indo-European area is bounded by a
great group of semi-barbarous populations, mostly with Mongolian
features, and speaking languages which have been grouped as Turanian.
These Turanian languages, on the one hand, graduate without any break
into those of the Esquimaux and American Indians; on the other,
according to Mueller and Latham, they are united, though less
distinctly, with the Semitic and Japhetic tongues. They not improbably
represent in more or less altered forms the most primitive stock of
language from which both the Semitic and Japhetic groups have
branched. Another great area on the coasts and in the islands of the
Pacific is overspread by the Malay, which, through the populations of
Transgangetic India, connects itself with the great Indo-European
line. Mr. Edkins, in his remarkable book on "China's Place in
Philology," has collected a large amount of fact tending to show that
the early Chinese in its monosyllabic radicals presents root-forms
traceable into all the stocks of human speech in the Old World; and
the American languages would have furnished him with similar lines of
affinity. If we regard physical characters, manners, and customs, and
mythologies, as well as mere language, it is much easier thus to link
together nearly all the populations of the globe. In investigations of
this kind, it is true, the links of connection are often delicate and
evanescent; yet they have conveyed to the ablest investigators the
strong impression that the phenomena are rather those of division of a
radical language than of union of several radically distinct.

This impression is farther strengthened when we regard several results
incidental to these researches. Latham has shown that the languages of
men may be regarded as arranged in lines of divergence, the extreme
points of which are Fuego, Tasmania, Easter Island; and that from all
these points they converge to a common centre in Western Asia, where
we find a cluster of the most