DETERMINATION OF ALKALIES IN MINERALS. 2l3 



cess of drying is going on. When the amount of soda is very 

 small it is best to allow the solution on the glass to dry in the 

 slowest possible manner. Should the quantity of soda be still 

 smaller or the nature of the crystals doubtful, resort may be had 

 to polarized light, when the prismatic crystals of chloride of 

 platinum and sodium will be at once rendered visible by their 

 beautiful colors, as they possess polarizing properties, whereas 

 the crystals of chloride of platinum and potassium, besides dif- 

 fering in form, do not polarize light. 



38. This method of detecting a small quantity of soda in the 

 presence of potash I have employed since June, 1850, while 

 engaged in the examination of the collection of urinary calculi 

 belonging to the Dupeytren Museum at Paris; at that time it 

 was employed daily in the laboratory of Messrs. Wurtz and 

 Yerdeil; the special reason for devising it was to examine the 

 nature of the trace of alkali almost invariably found in the uric 

 acid calculi after combustion. 



39. The reason for making special reference to the date of the 

 original employment of this method is to claim priority in its use, 

 as Mr. Andrews announces it in a late number of the Chemical 

 Gazette as a new method. Were not the method so well known 

 and so constantly employed in the laboratory of Wurtz and 

 Yerdeil at the period above mentioned, I should not now set 

 up any reclamation in the matter. 



40. The amount of soda that can thus be detected is ex- 

 ceedingly small, as the liquid can be concentrated to the very 

 smallest bulk. When the amount of potash is proportionally 

 large compared with that of the soda it is better to put the 

 chloride of platinum in a drop of the solution of the alkaline 

 chlorides placed in a watch-glass, allow the potash-salt to settle, 

 take a little of the clear liquid, place it on a slip of glass, evap- 

 orate slowly, and examine in the way already mentioned. 

 We should avoid the use of alcohol as a solvent for the salts 

 employed. 



41. For the full appreciation of this method it requires some 

 experience, and on first trials the extreme results will not be 

 readily obtained; too much care can not be taken with refer- 

 ence to the evaporation ; sometimes, if the evaporation be a 

 little too speedy, no indication of the presence of soda will be 

 evinced; so in all doubtful instances the glass should be la : d aside 



15 



