184 Proceedings of the Royal Society of Edinburgh. [Sess. 
and race of Tasmanians and Australians with the latter subsequently 
resulting from a racial cross with the primitive stock. 
To Sergi’s theory, based, be it remembered, on facts which he has firmly 
established, let us now apply our own results. 
Commonality of origin of Australians and Tasmanians is shown in the 
present work by the results of Mollison’s variation index for Australian and 
Tasmanian. On no other grounds can the remarkably uniform results 
attained by us with this index be explained. Read in conjunction with 
Sergi’s study of lophocephaly we regard Homo tasmanianus as proved. 
That the Tasmanian aboriginal was as nearly as possible the pure 
descendant of Homo tasmanianus we regard as certain on account of (1) 
the results attained by Sergi himself; (2) the remarkably small range of 
variation found in the Tasmanian for the observations of the present work • 
(3) the equally small amount of variation recorded by Buchner (27) in his 
recent works on Tasmanian prognathism, craniotrigonometry, and curvature 
indices ; and (4) the close approach to unity for the coefficients of correlation 
already recorded by us in our biometrical study of the Tasmanian (2). It 
is thus clear that the morphological studies of Sergi, the craniological 
researches of Buchner and ourselves, and the biometrical work of Dr Cross 
and ourselves all alike testify to the purity of type of the Tasmanian and 
the truth of Sergi’s hypothesis. 
That the Australian aboriginal is a hybrid is, we believe, proved by (1) 
the results recently recorded by Sergi ; (2) the study in the range of 
variation adduced in the present work — a study which clearly proves that 
the Australian is more variable than an admittedly crossed race like the 
modern Italian ; (3) our own previously recorded study in the Australian 
coefficients of correlation ; and (4) Broca’s statement, quoted by Topinard 
(28), that it is only when the variations reach 15 or 18 per cent, that we 
can say with certainty that they are due to mixture of race. In the 
present work the Australian range of variation has been proved to be 
40 per cent, more than in the Tasmanian. 
If it be argued that we deduce too much from this range of variation — 
a study as yet largely in its infancy as regards the precise meaning to be 
attached to it — we reply in the words of Cossar Ewart, than whom there 
is no greater living authority on the subject of variation by crossing of 
types (29) : — 
“ Domestic animals reproduce themselves with great uniformity if kept 
apart ; but the moment one mixed up two different races, strains, or breeds, 
one did something that was difficult to put in words, but the result was 
what has been best described as an epidemic of variations.” 
