232 REPORTS ON THE STATE OF SCIENCE. 



broad-heads by nine to one, more than even among the 64 ancient 

 ■skulls examined by Dr. Duckworth. This is a true centre, for we find 

 this dolichocephalic element radiating in all directions into the neigh- 

 bouring eparchies of Pedhiadha, Mirabello, Hierapetra, and Vidnnos. 



The other mountain massifs are not true centres ; for, although the 

 dolichocephals are most numerous, the mountains, towering up 8,000 

 feet, form a barrier to the south, and the long-heads cluster on the 

 northern slopes only. This is true of the White Mountains, where in 

 one village alone of Upper Kydhonia, Lakkous, the home of the late 

 Dr. Jannaris, the 65 men I measured averaged 76'9, against 79'9 in 

 the plains of the same eparchy. The other mountain missif, Mount 

 Ida, plays a similar part, and on the northern side, in the upper portion 

 of Mylopotamo, my records show 83 subjects averaging 765. The 

 mountains to the south of the Messara Plain, though not so lofty, slope 

 steeply to the sea, and being shut off from the main centres, offer a 

 most undesirable region for any invader to occupy in a hostile country. 

 The region is sparsely populated, and the 28 subjects measured in that 

 part which falls within the eparchy of Monophatsion average 76'9, 

 compared with 80"9 in the Messard Plain immediately below. 



These four mountainous regions appear to be the strongholds to 

 which the earlier inhabitants have been driven by successive invaders, 

 and strong confirmation of this hypothesis comes from the method by 

 which I arrived at it. A map of the cephalic index eparchy by eparchy 

 offered no clue. A suspicion of differences between mountain and plain 

 suggested the cleavage line of 1,000 feet altitude as a criterion of classify 

 cation, but this failed in some cases, though successful in others. It 

 was arbitrary and did not always serve as a register of accessibility. It 

 then occurred to me that in a land of such marked physiographical 

 features as Crete, Achseans, Dorians, Venetians, all had probably fol- 

 lowed much the same routes as the Turks in the seventeenth century. I 

 therefore made a map of the Turkish occupation of the island according 

 to the census of 1881, before the latter-day exodus. This showed that 

 from centres on the north coast of the island — Canea, Eethymo, Candia, 

 and Sitia — the lines of immigration radiated southward, stopping short 

 at the foot of the mountains, with but one exception, to which I shall 

 refer later. This general truth is particularly well illustrated in the 

 many lines of occupation stretching south from Candia, the greatest 

 Turkish centre, which all stop short abruptly at the'foot of the Messara 

 Mountains. Three great blank spaces stand out on the map between 

 these lines of immigration — Mount Dicte with the fringes of the neigh- 

 bouring eparchies, the northern slopes of Mount Ida, and both northern 

 and southern sides of the White Mountains. These blank areas are 

 those which we have already found occupied by the predominant dolicho- 

 cephal, with one notable exception, the southern slopes of the White 

 Mountains. This region, where the Turks have never yet held sway, 

 this eparchy of Sphdkia, where the Sphakiots have successfully repulsed 

 Turk and Venetian alike, and, isolated in a sterile, rocky home, proudly 

 claim Dorian descent, is the one outstanding exception to the rule that 

 the mountains are the refuge of the dolichocephals. I believe this 



