e are elements of similarity between (i) the Mafulu
people, (2) the Dutch New Guinea dwarfs, and (3) one or more of the
Andamanese, Semang and Aetas; but in my comparison of the Mafulu
and the dwarfs of Dutch New Guinea with the other previously known
dwarf races I would specially draw attention to their similarity in
shortness of stature and (as regards most of the Mafulu and a few of
the Dutch New Guinea people) colour of hair; and this impels me to
venture to say a few words on the larger question.

I have searched through much existing literature concerning the
various hitherto discovered dwarf races of the world with reference
to the question whether, even assuming that these people have an
original primary ancestry from which the taller negroid races also
are descended, they must be regarded as having become a related type,
separate and distinct from the latter, as now existing, or whether
they must all be treated as merely separate local variations, each of
them having failed to develop, or retrograded, and in other respects
become different in type from taller negroid races among or near
to whom they are found. And I am struck by the fact that, though
the natural tendency to local variation in stature, shape of head,
colour and other matters is brought forward in support of the latter
theory, no one seems, in connection with the general question, to have
noted the fact that, whilst the hair colour of negroes, Papuans and
Melanesians is black, the hair of all these various dwarf people seems
to be predominantly brown, and that this variation explanation, if
regarded as applying to these dwarf races separately and independently
of one another, involves a remarkable coinciding double variation
(in stature and predominant colour of hair) exhibited by all these
dwarf people as compared with the taller negroids.

On the other hand, if there has been an original separation of
descendants of common primary ancestors of all the negroid races,
which, through variation, has resulted in two main types, one
predominantly full-sized and always black-haired, and the other always
short and predominantly brown-haired, and the pygmies (negritoes and
negrilloes) are to be regarded as being all descendants of the latter
type, who have since for some reason become geographically separated,
there would appear to be nothing remarkable in the double variation.

But in that case we are, I take it, justified in regarding the dwarf
rac