354 ON THE HEAD-FORMS 



diameter lies far back, and the tendency of the skull to 

 ellipticity renders it aphtenozygous. The brows are, in the 

 Keltic type, prominent and low, either oblique, or, which is 

 very common in Ireland (see Davis, Cr.Brit., p. 201), " forming, 

 with the projecting superciliary ridges, a horizontal line above 

 the eyes." The forehead above the brows is rather flat, in 

 intelligent men often elevated and square (Edwards^ Kimric 

 type), but in the bulk of the population low, ''gaining," as has 

 been said, "in length what it wants in height/'' The upper pro- 

 file of the skull has generally a gentle and regular curve as far 

 as the upper occipital region ; this is generally protuberant, 

 and whether so or not, is oval in section : this point we owe to 

 Pruner-Bey, but I have confirmed it in a good many instances. 

 " Eeceptaculum cerebelli small," said Eetzius of a particular 

 skull, an Irish one. (Cr. Brit., Description of Ancient Hibernian 

 Skull, plate 55, p. 2). I think the remark applies to the best 

 examples of the type, but my method of measurement does 

 not allow me to test its correctness. 



The facial features in several varieties or crosses of this 

 type have been well described by Dr. Barnard Davis. The most 

 constant are the rather deep-set eye, the sinuous long-nostriled 

 nose, prominent at the tip, and the always angular and often 

 narrow chin. A slight degree of prognathousuess, producing 

 a vertical furrowing of the cheek, is so common, that it may 

 perhaps be a race-character. Length of face varies like length 

 of head, but is generally considerable ; in the Firbolgs of Arran, 

 and in many Walloons, it is conspicuously great. 



Such is the prevailing type in Ireland generally; and I 

 think it is more conspicuous than any other in the greater 

 part of Somerset, and perhaps in South Devon. It is common 

 in other parts of the west also, including certain tracts in the 

 valley of the Bristol Avon, which, according to Dr. Guest, long 

 remained Damnonian. The ovoid head, tending to elHpticity 

 when long, to roundness when short, seems to predominate in 

 all the upper part of Wiltshire and of Gloucestershire, and 

 occurs in more or less force elsewhere, notably about Bideford, 

 and along both coasts of the Bi-istol Channel. But in the 

 Vale of Thornbury and the Forest of Dean, as well as in 



