32 Mr. kir wan’s Experiments and Obfervations on 
Hill confefledly contain fome undetermined proportion of wa- 
ter, and by the quantity of water he commonly underffands 
that which is retained by cryftallization : thus, in his firft: vol. p. 
i 37. he fays, that 100 grs. of vitriol of iron contain 23 of iron, 
39 of vitriolic acid, and 38 of water. But in his treatife De 
Produciis Vulcanicis , § 12. he fays, that 100 grs. of vitriol of 
iron contain 24 of iron, 24 of dephlegmated vitriolic acid, and 
52 of water; and this laft calculation fcarcely differs from 
mine, as I aflign to 100 grs. of vitriol 25 of iron, 20 of real 
vitriolic acid, and 55 of water. The difference manifeftly 
arifes from the quantity of water Hill contained in his de- 
phlegmated acid. The mold material difference between us 
regards the quantity of the mineral acids taken up by alkalies ; 
for, according to his and Mr. scheffer’s experiments, they 
take up more of the vitriolic than of the nitrous, and 
more of the nitrous than of the marine ; whereas, according 
to Mr. iiomberg’s, Dr. plummer’s, Mr. wenzel’s, and my 
experiments, this does not happen. This difference arifes in 
all probability from the different degrees of evaporation by 
which the cryftals of thefe falts are obtained ; for which reafon 
I did not examine the quantity of the cryftals, which muff be 
variable, but that of dry fait, left after thorough evaporation. 
With regard to the quantity of earth and metallic bafis in dif- 
ferent falts, Mr. bergman’s experiments and mine agree almoft: 
intirely. 
The advantages refulting from thefe inquiries are very confi- 
derable, not only in promoting' chymical fcience, which, being 
a phylical analyfis of bodies, eflentially requires an exadl deter- 
mination, as well of the quantity and proportion, as of the 
quality of the conftituent parts of bodies, but alfo in the 
pra&ical way. Thus, in the firft: place, it is well known, that 
feveral 
