276 NATURAL SCIENCE. 



JUNE, 



more curved. They are also two or three times as large as anything 

 found in overlying deposits. The largest measures lof inches 

 (26*3 cm.) in length by if (4'5 cm.) in width, and is remarkable 

 both for its thinness (not reaching f in.) and its strong curvature. 

 The others measure respectively 23 by 5 by 1-5 cm. and 67 by 5*5 by 

 1*3 cm., and are also thin, though the curvature is less marked. 

 All are roughly chipped, and show no trace of polishing. Though 

 implements of this character occur with relics of Neolithic date, yet 

 they are by no means distinctively Neolithic. Similar ones occur in 

 France in the Magdalenian period, and in the British Museum there 

 are undoubted Palaeolithic fragments of equally large flakes. 



A necklace of perforated teeth was discovered round the head of 

 the large skeleton. All the teeth are ground down and polished, and 

 many are ornamented by transverse striae, so that all distinctive 

 characters in the crowns are obliterated. M. Riviere's opinion that 

 they are canines of deer is probably true of the greater number. The 

 perforations are central between crown and fang ; they are cleanly 

 cut and polished, considerably wider at the surface than the centre, 

 as is usually the case with holes bored by flint flakes. 



The fish-vertebrae are of rather small size, scarcely exceeding 

 a centimetre in diameter, and are perforated through the axis by 

 holes a millimetre in width. They were found round the necks of 

 all three skeletons and seem to belong to the genus Salmo. 



Two Cowrie shells perforated just at the anterior end of the 

 mouth were also found, and seem from their position to have been 

 used as anklets. The species is not, I believe, a Mediterranean 

 one, but seems to be identical with an Atlantic form in the Mentone 

 Museum, labelled Cypraa millepunctata, though colour and marking" 

 have disappeared. M. Riviere refers also to a necklace of small shells 

 of Nassa similar to those found with the skeleton of 1873, but it has 

 now disappeared. 



The bone ornaments discovered were three in number, lying on 

 the forehead of the skulls, and had been, doubtless, attached to the 

 thong on which the teeth were threaded. All have the form of a 

 spindle, deeply constricted in the middle, with a smoothed surface 

 ornamented by rows of short transverse lines. One measured 

 2 by f in., the others, one larger and one smaller. Internally the 

 texture is somewhat cancellous. M. Riviere is of opinion that they 

 are made from stag's horn. 



Coming to the question of the age of the new skeletons, one is 

 confronted with the same difficulties that have attended other 

 discoveries in this region. What importance is to be attached to 

 great thicknesses of cave-deposit when there are no stalagmitic 

 layers to mark particular levels ? To what depth are we to suppose 

 Neolithic man could dig graves ? How much value is to be attached 

 to the manufacture of ornaments, and how much, on the other hand^ 

 to cranial characteristics ? 



