iii the Western Ides of Scotland. 



171 



but it would be difficult to say whether it is really blowu sand or 

 merely the upper portion of the old sandy shore. 



The floor in which the implements lie is firmly cemented by oxide 

 of iron, which pipes down into the subjacent sand in a remarkable 

 way. The flint chips lie scattered about in all directions, and do not 

 appear to be accumulated in one spot more than another. There 

 was certainly no trace of anything in the nature of a mound, and no 

 marked accumulation of shells or bones. "We saw neither hearth- 

 stones, nor heaters, nor any other indication of a fire having been 

 lighted on the spot. In fact, it is exceedingly unlikely that the 

 floor as at present exposed was in any sense a dwelling-place. 



W° ck - 



Sandy P/at 

 adjoining stream . 



Blown-sand. 



Fig. 2. 



Sketch-map of the Sand-hills at Balnahard on the scale of 6 inches 

 to a mile, showing the position of the Neolithic floor. 



The only objects of interest found on the floor in addition to the 

 numerous flint flakes were some elongated water-worn shingle stones 

 with rubbed or broken ends. These are merely fragments of the 

 Torridon mudstones of the island, the cleavage in which frequently 

 causes them to break up into elongated pieces, which become rounded 

 on the shore. The ends of some of these stones are distinctly bevelled 

 by rubbing, and others are broken in such a manner as to suggest that 

 they have been used with considerable force. In the presence of 

 these stones we have a suggestion, but only a suggestion, of con- 

 temporaneity with the shell-mound of Caisteal-nan-Gillean, where 

 exactly similar implements have been found in considerable numbers 1 

 (see p. 167). 



1 Mr. T. C. Cantrill informs me that he has found a number of such elongated 

 pebbles, averaging 6 inches in length, on flint chipping-floors in Pembrokeshire, 

 and believes they are flaking-tools. 



