FLINT IMPLEMENTS AND PALiEOLITHS 91 



a satisfactory striking-platform, and so they had to learn to 

 provide themselves by flaking with a flat surface upon which 

 blows could be struck with precision. There is no doubt that 

 the sub-Crag rostro-carinate implement, though generally 

 much larger and of a more imposing appearance than the 

 primitive implement represented in fig. 2, is nevertheless made 

 on almost the same lines. The ventral surface of the rostro- 

 carinate formed by the removal of a large flake from the 

 original flint nodule represents the natural flat surface of 

 tabular flint, and in both cases blows were delivered on each 

 side of this flat surface. But whereas it seems that in the case 

 of the primitive implements represented in fig. 2 the result 

 aimed at was the production of two scraping hollows, in the 

 rostro-carinate the keel or gable seems to have been the desired 

 object. 



It will be noticed that there is a slight difference in the 

 method of production of the rostro-carinate from that of the 

 implement represented in fig. 2. In the latter all the blows 

 were delivered on the flat ventral surface, and while this is the 

 case with most of the blows which went to form the keel of the 

 rostro-carinate, yet one or two were delivered on the dorsal 

 surface of the specimen. The author has found by many 

 experiments in flint-making that in forming the keel it is some- 

 times necessary to deliver some blows on the dorsal surface. 

 The section of the rostro-carinate implement is triangular as in 

 the case of fig. 2. 



Fig. 4. — It has been noticed, from an examination of a 

 number of rostro-carinate specimens found in deposits of less 

 antiquity than the detritus-bed below the Pliocene Red Crag, 

 but more ancient than the river valley gravels, etc., that the 

 ventral surface, or " striking-platform " as it might be termed, 

 was gradually extended further backwards towards the posterior 

 region of the implement. 



This extension of the ventral surface made possible the cor- 

 responding extension of the keel, until an implement was pro- 

 duced having a cutting edge extending the whole length of its 

 dorsal surface. Such a specimen is illustrated in fig. 4. The 

 section of this implement is still triangular, and the majority of 

 the blows forming the keel were delivered upon the flat ventral 

 surface. Several specimens of this type have been found from 

 time to time in river-gravels associated with early palaeolithic 



