TRANSACTIONS OF THE SECTIONS. 135 



as to induce the makers of these flint tools to adopt a plan by which the maximum 

 number were obtained with a minimum amount of waste. 



All the flint flakes and cores from the ten different stations, ranging along the 

 coast from Croyde to Bude, show a singular uniformity in their design, and the 

 method by which they were formed appears to have been as follows. A nodule 

 having been selected, a flat surface or base was tirst formed by striking oflT one end, 

 as near the point as possible. If the flint was cherty, or showed an uneven and 

 hackly fracture, it seems to have been rejected in this first stage of its manufac- 

 ture ; but if, on the other hand, it split with a smooth conchoidal fracture, a series 

 of blows were administered from the flat surface, at intervals round the margin, 

 so as to peel oft" the rough coating of the nodule on three sides. The second 

 series of blows produced the largest flakes, and a third or even a fourth set of flakes 

 could be obtained in this manner before the core was used up. 



This peculiarity was noticed by the author about two years ago, in the coiu'se of 

 a communication to the Society of Antiquaries, and a subsequent examination of 

 many hundred flakes and cores has served to prove that the same process was in 

 use throughout the whole of this district. 



The largest flakes hitherto foimd in North Devon are about three inches in length, 

 but between these and the smallest, which measm'e not more than three-quarters 

 of an inch, there are innumerable gradations in size. The results of the principal 

 excavations which have been made at Croyde and Northam, show that the average 

 proportion of cores to flakes is about 14 per cent. 



On the Esquimaux considered in their relationship to Man's Antiquity. 



ByW.S- Hall. 



On the Circassians or White Khazars*. By H. H. Howobth. 



The author detailed the history of the Circassians and Kabardiens from the 12th 

 century downwards. Before the arrival of the Turks, Comans, &c. the chief race in 

 the plains of the Kuban and the Crimea were the Khazars, a race well known to 

 the Byzantines and Arabians. When the name Khazar disappears, that of Cir- 

 cassian appears, and from a comparison of the language, customs, and traditions 

 of both races, the author proved them to be one race under two names, and that 

 the Circassians are the descendants of the White Khazars. 



On a Frontier of Ethnology and Geology*. By H. H. Howoeth. 

 The Isothermals in Europe are twisted far to the north of the normal course 

 they follow in Asia and America. This twisting has not always existed, as is 

 proved by the fauna and flora of prehistoric times, and by the remarks of the 

 Greek and Roman geographers on the climate of Central Europe. Their gradual 

 movement, synchronous with the gradual pushing back of the limit of the fauna 

 and flora of prehistoric times towards Siberia, the author correlated with the 

 gi-adual intrusion of the Iranian race of men, with its associated animals and 

 plants, into Europe, and the gradual displacement of the Ugrians ; the Ugrian 

 being as much a paL-eontologieal witness to ?uch a condition of things as we find 

 in Siberia, or as existed in prehistoric times, as the musk sheep or the reindeer. 

 The first intrusion of the Iranians into Europe, which the author coidd not date 

 before the 12th century B.C., marked the beginning of a new geologic period. 

 The climactic changes incident to it seem to be caused solely by the presence of 

 the Gulf-stream, so that if we can date the first giving way of the Ugrians, we 

 may possibly also date the first advent of the Gulf-stream. 



On the so-called " Petrified Human Eyes," from the Graves of the Dead, Arica, 

 Peru. By the Rev. A. Hume, D.C.L., F.S.A., iSfc. 

 A very large portion of the Republic of Peru, on the west coast of South Ame- 

 rica, lies in a district where rain is unknown ; and in several important respects, 



* This paper will be printed at length in the ' Journal of the Ethnological Society of 

 London.' 



