sulphur by my estimation and 42 by his own: the sp. gravity of his artificial realgar is not given. Westrumb estimates realgar at 100 arsenic and 25 sulphur, and Thenard at 100 arsenic and 33 sulphur. But from the above it must be concluded to contain 100 arsenic and 45 to 50 of sulphur. One hundred parts of the same realgar heated in caustic potash were resolved into 78 orpiment taken up by the liquid and 22 arsenic precipitated. It appears to me most probable that a true subsulphuret would be most convenient for the printers’ use, or one containing 100 arsenic and 32 sulphur, that is, 2 atoms arsenic and 1 sulphur. The object being to deoxidize indigo and obtain it in solution in a green state, we may suppose that 1 atom arsenic takes the oxygen from the indigo and then forms arseniate of lime which precipitates, whilst the other atom in union with the sulphur, takes the green indigo and unites it to the potash, making a quadruple compound of arsenic, sulphur, green indigo and potash in solution. If this view be right the heaviest and darkest coloured realgar of commerce must be the most advantageous for this purpose. Some printers however prefer the protosulphuret. _Deutosulphuret._ Proust, by heating 100 arsenic with 300 sulphur in one instance got 222 parts, and in another 234 parts of a transparent deep greenish yellow sulphuret, (Jour. de Phys. 59--p. 406. 1804). Now it is very remarkable that if we take the atom of sulphur at 13 and that of arsenic 21, one of this and two of the former will be found as 100 to 124, together 224; but if sulphur be 14, then the proportion will be 100 to 133, together 233. It seems more than probable that Proust had accidentally used that degree of heat in the combination which is requisite for forming the deutosulphuret. It is probable too that Laugier always used a higher heat, as he uniformly obtained the same (lower) sulphuret whatever were the proportions, the excess of either being sublimed or separated by the heat. _Trisulphuret_, _quadrisulphuret_, &c. When a solution of the oxide of arsenic is treated with quadrisulphuret of lime, little precipitate appears; but if muriatic acid be dropped in, a fine yellow precipitate is formed. This I have reason to think is sometimes a trisulphuret, and at other times a quadrisulphuret or higher; but it is difficult to investigate these compounds, and on that account I speak with some uncertainty. 33. _Sulphuret of cobalt._ Sulphuretted hydrogen does not precipitate cobalt from solutions containing that metal; but hydrosulphurets precipitate it.