If to 24 parts of lime dissolved in water we put 98 parts of dry arsenic acid, the compound remains in solution, and is perfectly neutral to the colour test, but so that the addition of a small quantity of either ingredient disturbs the neutrality. If to this solution 24 parts of lime dissolved in water be added, the compound remains a limpid solution, but is very limy to the test. If to this we put in like manner, 24 parts more of lime, the whole compound is thrown down, and yields, when dried, 170 parts of arseniate of lime, the liquid being now free from both elements. Here we see first, two atoms of the deutoxide, neutralized by two atoms of base, namely, 1 of arsenic oxide, and 1 of lime; but (second), when one atom more of lime is added, an union of 2 deutoxide, and 3 of base is effected, which of course is an alkaline salt; when (third) more of lime is added, the 2 deutoxide and the 1 protoxide each attach 1 of lime, and form a still more alkaline salt, which being insoluble, is wholly thrown down, most probably in a compound state of 98 parts arsenic acid, combined with 72 parts lime. In like manner, I find that 42 parts of potash, 28 of soda, and 12 of ammonia, severally neutralize 98 parts of arsenic acid. 1st. 24 lime + 32.7 arsenic acid = insoluble arseniate 2d. -- + 49 ------------ = soluble arseniate 3d. -- + 98 ------------ = neutral arseniate It is a remarkable fact, that when neutral arseniate of potash and nitrate of lead are mixed together to mutual saturation, the precipitate is found to consist chiefly of arsenic acid and oxide of lead, in proportion of 1 of acid to two of oxide, (that is, 98 ∶ 194, or 100 ∶ 198); which does not differ much from the determination of Berzelius. I find however, only one fourth of the nitric acid in the residuary liquid in a free state; which leads me to suspect that the precipitate is a compound of subnitrate and arseniate of lead, in which the arsenic acid and lead are in due proportion, or 98 acid, to 97 oxide. This consideration may be properly resumed hereafter. Hence we conclude, the atom of arsenic weighs 21 (and not 42, as at page 264, VOL. 1), that of the protoxide of common white arsenic, 28; and that of arsenic acid = 98, being a compound of 2 atoms of deutoxide, and 1 of protoxide. Or, 100 Arsenic + 33.3 oxygen = 133.3 protoxide -- + 55.5 ---- = 155.5 arsenic acid 21. _Oxides of cobalt._