L. H. Dudley Buxton 
99 
While the high standard deviations of Lycia are noticeable, the extremely low 
standard deviation of the Lycian gypsies calls for special coinment. We have 
already drawn attention to the fact that these people formed a strictly endogamous 
community and did not mix with their neighboui's. The evidence of the low 
standard deviation, especially in contrast with the great variation shewn by people 
living in the same place but belonging to a different social milieu, supports the 
claim of these gypsies to be a pure blooded race. 
The crania (see Table III) belong to very small series and it is difficult to argue 
with certainty. It appears that as a general rule the modern Greeks are slightl)' 
TABLE III. 
GephnJic and Craitinl Indicef^. 
Provenance 
Lexikas 
Corfu 
Ci'etc 
Crete, Khauia 
Lycia (Greeks) 
Samos 
Cyprus 
Athens 
Adult Living 
Males 
Modern Crania 
No. 
42 
IfiOO 
179 
586 
60(?) 
Geplialic 
Index 
84-55 
79-20 
52 
17 
L3 
80-27 I 93 
82-54 
79-6 
Sex 
Cranial 
Index 
2 units 
80 
82-9 
77-8 
79-9 
Ancient Crania 
Date 
Neolithic 
Bronze Ago ... 
Late Roman ... 
Ancient 
" Beginning of Bronze Age "... 
Early Middle Minoan 
Middle Minoaii 
Early Late Minoan ... 
Late Minoan III 
"Ancient" 
Bronze Age ... 
Hellenic and Roman 
Early Hellenic to Hellenistic 
No. 
1 
4 
6 
46 
13 
5 
7 
G 
14 
7 
13 
Sex 
Cranial 
In lex 
+ 2 units 
83 
75 
77 
79 
77 
75 
77 
78 
81 
79 
80 
79 
79 
98 
1 
10 
5 
5 
1 
40 
35 
20 
38 
more brachycephalic than the ancient skulls from the small place, even after making 
allowance for the difference between the measurements of the living and the dead. 
This difference appears still more striking when the distribution into dolicho- 
cephalic, etc. is made. Even granting however that this difference exists, con- 
sidering the small number of crania, it is perhaps more remarkable that the ancient 
skulls should be so extremely close considering the very large standard deviations 
which we have found. The comparison has been made in Table IV. 
There are however certain points of resemblance and difference which need 
special attention. First there is in general a closer approximation between the 
earliest cranial indices and modern ones than among those of later date. This holds 
true in every case. Secondly there is only one case of a considerable difference 
between the ancient crania and the modern. This occurs in Leukas. The one 
early (Neolithic) skull has a cranial index of 81, and this index is not very different 
from that of the modern men of Leukas. The evidence of the five later skulls with 
