OK ARCfi^OLOGlCAL AND ETHNOLOGICAL RESEARCHES IN CRETE. 407 



that the bones had been transferred from some other cemetery or 

 graves, and thus the term ' ossuary ' is advisedly applied to the locality. 



This circumstance was attended with advantages and disadvantages, 

 for while the bones were so closely packed that a comparatively small 

 amount of labour brought many examples to light, and thus saved much 

 time that AV(ni!d have been consumed in exposing each skeleton if 

 separately interred, yet the disadvantage remained that it was rarely 

 possible to assign a number of bones to the same skeleton, and there is 

 thus a certain lack of information as regards the proportions of the limbs. 

 The condition of the bones was unsatisfactory as regards transport, their 

 substance being extremely friable : this necessitated great care in 

 excavating, and measurements could often only be made while the bone 

 or skull was still half imbedded in the soil. About thirty-five skulls and 

 two or three hundred bones were preserved in various ways, but the 

 results of these attempts are not yet available. 



Including the skulls discovered in the rock-shelter at Agios Nikolaos, 

 there were altogether about a hundred skulls available for examination. 

 In the case of male skulls sixty examples, and of female skulls twenty-two 

 specimens were actually measured. Skulls of later date were found in 

 the museum at Candia, including skulls from Zakro, collected by 

 Mr. Hogarth ; a child's skull was found at Agios Nikolaos ; and .several 

 modern Cretan skulls were also measured. 



We are now chiefly concerned with the sixty-two male and twenty-two 

 female skulls from Pahvokastro. To these may be added the data refer- 

 ring to two male skulls and one female skull obtained at the same place in 

 1902 by Mr. Bosanquet, and measured by Dr. Myers at Cambridge. This 

 brings the total to sixty-four male and twenty-three female crania. 



The important feature to notice here is the breadth or cephalic index 

 of these crania, which is on the average 73"4 for males and 73 for female 

 skulls ; both are therefore dolichocephalic. The specimens are thus 

 concordant in this character with the majority of other early Cretan 

 skulls and, it may be added, with most early Greek crania from the 

 mainland. 



But an important point (upon which information was desired by 

 many who are interested in the prehistoric ethnology of Greece and the 

 Levant, including Crete) is the inquiry as to how far this character of 

 long-headedness is general among this early population, and is not merely 

 the expression of an average. 



In the scanty material previously available, short (brachycephalic) 

 skulls are admittedly infrequent, and the accession of comparatively 

 abundant new material provides a fresh contribution to our stock of 

 information. The result of the investigations is, then, to show that in this 

 early Cretan population longheadedness is quite predominant : of forty-six 

 male crania available for examination thirty (6o-3 per cent.) are dolicho- 

 cephalic, twelve (26"15 per cent.) are of mean proportions, and only four (or 

 8'55 per cent.) are .short. The corresponding percentage figures for female 

 skulls are 70'G dolichocephalic, 23'53 of mean proportions, and 5'S7 per 

 cent, of short skulls. 



Such a proportion of short skulls had previously been anticipated by 

 some (certainly by Mr. Myres), and it is evident that a proportion of 

 8 per cent., or even .5 per cent., in a population constitutes a factor that 

 cannot be ignored in a full discussion of race afhnities. 



It remains to be remarked that the crania here described are ahuo&t 



