of the combinations of phosphorus. 327 
of the horn silver formed, with the greatest care, 18.4 of the 
liquid chloride afforded only 54.5 of chloride of silver, which 
agrees as nearly as could be expected with the former expe- 
riment. In two other experiments, made with equal care, and 
in which the liquid was poured into a solution of nitrate of 
silver, 6 grains gave 17.1 of horn silver, and 29.4 gave 89.9 
of fused horn silver. 
In examining minutely the circumstances of the action of 
the liquid chloride, or solutions containing phosphorous and 
muriatic acids, or nitrate of silver, I found no difficulty in 
explaining the cause of the error in the former experiments. 
Phosphorous acid acts upon nitrate of silver, and more 
rapidly in proportion to its concentration, and gradually 
produces a copious precipitate from it ; so that if there be an 
excess of nitrate of silver, and the precipitate be not imme- 
diately separated from the solution, there is always a consi- 
derable increase of weight. M. Dulong, and M. Berzelius, 
whose experiments agree with my former ones, may have 
been misled by a precipitation from the nitrate of silver by 
phosphorous acid, as I am sure I was. M. Berzelius does 
not state how he prepared his liquid chloride of phosphorus; 
but M. Dulong, who objects to my process by corrosive 
sublimate, and employs, instead of it, the action of chlorine 
on phosphorus in forming his fluid, must have been exposed 
to other sources of error. He speaks of acting on dry phos- 
phorus by dry chlorine ; but it must be always extremely 
difficult to free a gas that cannot be kept over mercury, of 
all its vapour ; and as perchloride always forms during the 
action of phosphorus on chlorine, a part of which produces a 
fluid, and easily volatile hydrate with water, and soluble in 
