Dr Wollaston an the Apparent Direction of' Eyes, <SfC. 117 



which we may be led to infer the presence of that alkali in 

 them. This has been observed in sodalite, analcime, chabasie, 

 albite, pitch stone, and several others. Unfortunately, how- 

 ever, a yellowish colour may be produced by other substances 

 besides soda, so that it is a test which cannot altogether be relied 

 on with certainty. Thus, a similar effect is occasioned, though 

 in a less degree, by the fluate of lime, and, perhaps, by lime 

 under other circumstances. However this may be, it is cer- 

 tain that many minerals that contain soda give a very distinct 

 yellow colour to flame ; it is a circumstance, therefore, which 

 may be useful to the chemist and mineralogist, and as such I 

 mention it. 



I beg leave to observe, in conclusion, that experiments on 

 the colour communicated to flame should be performed with a 

 tallow candle, the colour of which is better fitted for the pur- 

 pose than that of a spirit-lamp. 



Akt. XXII. — On the Apparent Direction of Eyes in a Por- 

 trait. By W. H. Wollaston, M. D. F. R. S. and V. P- 

 With a Plate. 



This very curious paper of Dr Wollaston's, of which we pro- 

 pose to give a brief abstract, appeared in the Philosophical 

 Transactions for 1824. As it is one of those scientific papers 

 which may be easily comprehended by general readers, we 

 could have wished to transfer the whole of it to our pages ; 

 but it is partly illustrated with engravings on steel by Mr 

 Perkins, which cannot be copied, and we are, therefore, oblig- 

 ed to confine ourselves to an abstract of the most popular and 

 interesting part of it. 



In examining the eyes of a person opposite to us, who looks 

 horizontally within a range of about 20° on cither side, we 

 shall find that the white parts of his eyes increase and decrease 

 according as they are turned to or from the nose. When the 

 eyes of the person are looking straight at us, the two portions 

 of white are nearly equal, so that, by the relative magnitudes 

 of the white parts of each eye, we can estimate in what degree 

 the eyes deviate in direction from the J'ace to zvhich they be- 

 long. 



