Royal Society. 219 



The researches which form the subject of this paper were sug- 

 gested by an inquiry into the most effectual method of ascertaining 

 the quantity of nitrate of potassa existing in crude saltpetre. The 

 author found that by the action of hydrochloric acid the nitrate of 

 potassa was converted into the cldoride of potassium; and converse- 

 ly, that the chloride of potassium might, by the proper regulation of 

 the temperature, be reconverted into the nitrate of potassa Ijy the 

 action of nitric acid. These mutual conversions afforded excellent 

 means of determining, with great exactness, the relative equivalent 

 numbers, in the theory of definite proportions, belonging to these 

 salts, and to their respective constituent elements. The author, ac- 

 cordingly, pursued the investigation of these numbers by several suc- 

 cessive steps, of which the details occupy the greater part of the 

 present paper. He first determines the equivalent of chloride of 

 potassium by decomposing chlorate of potassa into oxygen and chlo- 

 ride of potassium ; the proportion between which gives the ratio 

 which the respective equivalent numbers of each bear to one another, 

 and also to that of chlorate of potassa. The equivalent of nitrate of 

 potassa is next obtained by converting the chlorate and the chloride 

 of potassium into that salt ; and from these data the equivalents of 

 chlorine and of nitrogen are deduced. A similar train of inquiry is 

 next instituted with the corresponding salts having sodium for their 

 base : chlorate of soda being decomposed into the chloride, and 

 into the nitrate ; nitrate of soda into chloride ; and chloride of so- 

 dium into nitrate of soda. The results of these different series of 

 experiments coincide so closely with one another as mutually to con- 

 firm their general accuracy in the most satisfactory manner. For 

 the purpose of determining the equivalent numbers of the element- 

 ary bodies themselves, (namely, chlorine, nitrogen, potassium, and 

 sodium,) the author employed the intermedium of silver, the several 

 saline combinations of which with chlorine and with nitric acid were 

 found to afford peculiar advantages for the accurate determination of 

 the relative weights of the constituents of these salts, when subjected 

 to various combinations and decompositions. The conclusions to 

 which the author arrives with regard to the equivalent numbers for 

 the six elementary bodies in question, tend to corroborate the views 

 of the late Dr. Turner, and to overturn the favourite hypothesis that 

 all equivalent numbers are simple multiples of that for hydrogen. 

 He finds these numbers to be as follow : 



Oxygen 8- 



Chlorine 35-45 



Nitrogen 14-02 



Potassium 39-08 



Sodium 2305 



Silver 107-97 



The author intends to pursue these inquiries, by applying similar 

 methods to the investigation of other classes of salts. 



January 31. — A i)aj)er was read, entitled, " Some account of the 

 Art of Photogenic Drawing, or the Process by which Natural Ob- 



