144 GOLD, PLATINUM, PALLADIUM, c. 



with silver, and the chloride of silver collected, 

 washed, dried, fused, and weighed ; in the third, 

 I endeavoured to approximate as nearly as pos- 

 sible to the quantity of water of crystallization, 

 by distilling it wer into a small receiver. Per- 

 fect accuracy in this last experiment was not at- 

 tempted, but merely such an approximation as 

 enabled me to judge how far the quantity of 

 water deduced from the other two experiments 

 would be borne out by the analysis. I shall 

 state at once the general result which was ob- 

 tained. 



The loss of weight which the salt sustained 

 was 1 8 grains ; one-half of this was chlorine, 

 determined from the chloride of silver formed. 

 This chloride, in two different experiments, 

 weighed 36 grains, indicating 8*87 grains of 

 chlorine. The small portion of chlorine gas re- 

 maining in the retort was estimated to weigh 

 O13 grain ; these two quantities together mak- 

 ing 9 grains the other 9 grains of loss were wa- 

 ter. Of this I actually obtained 7 grains by the 

 process above described ; and by passing a cur- 

 rent of air through the retort, and this through 

 a solution of nitrate of silver, I satisfied myself 

 that the weight of the residual chlorine gas did 

 not exceed O13 grain. There can be no doubt, 

 therefore, that the whole 9 grains were pure 

 water. There remained in the retort 25 grains 

 of gold. We see from paragraph 2d, that the 



