52 



even in the very beginning of the operation ; as it proceeded, there formed a 

 white precip. at the bottom of the vessel which contained the f. alk. : the pro- 

 cess was continued for upwards of an hour after the fumes had almost ceased 

 to appear, during all which time the heat was kept up much stronger than at 

 the beginning. 



The retort was not red hot, though I should imagine it could not want much 

 of it. The cake of neutral salt remaining in the retort after the vessels were 

 taken down weighed about 30-0-0 ; 3-5-0 of loose flowers of arsen. were sub- 

 limed into the neck of the retort ; little or none was found in that part of the 

 retort contained within the furnace, & upwards of 4.4 of moist flowers of 

 arsen. were found in the tube. The arsen. sublimed into the neck of the 

 retort was found no ways to differ from common arsen. 



3- 18 -16 of a solution of dry pearl ashes in equal weight of water were 

 saturated by 2-8-11 of aquafortis, whose spe. grav. was 1-398 : the loss in 

 mixing was 10*9. The operation was performed in a flor. flask, & the aqua 

 fortis was previously diluted with water to avoid too sudden an efferv. : being 

 evaporated it yielded 2*1 1-19 of dry crystals ; therefore I part of saltpetre con- 

 tains -936 of aqua fortis, 759 of dry pearl ashes, & -559 of ditto freed from 

 air. 



3'9-4 of arsen. was sublimed into the neck of the retort & tube in the 

 foregoing distillation; therefore there remains 16-I0-20 of arsenic in the cake 

 of neutral salt, supposing that no arsen. was carried over in the red fumes, as 

 in all probability there was not : therefore 1 part of neut. arsen. salt contains 

 •506 of dry f. alk. saturated with air or -373 of the same alcali deprived of air, 

 & -551 of arsen. Therefore the proportion of f. alk. and arsen. in the neut. 

 arsen. salt is 1 part of dry f. alk. saturated with air to 1,082 of arsen. 



If the cake of salt remaining after distillation be dissolved in a proper quan- 

 tity of hot water it readily shoots on cooling into crystals, which do not at all 

 grow moist in the air, & require about 3^ times their weight of water to 

 dissolve them. 



A solution of these crystals scarcely alters the colour of syrup of violets ; if 

 anything, they give it a reddish cast ; they turn tournsol paper a brownish-red. 

 A solution of fixed alcaU being dropt into a solut. of these crystals makes 

 an effervescence. The quantity of alcaline solution which must be added be- 

 fore it ceases to effervesce is such, that the dry alcaline salt shall be about f 

 of the weight of the neutral arsen. salt, or about f of the alcali already con- 

 tained in the neut. salt. 



Chalk or whiting also make a slight efferv. with this solution. Macquer 

 seems not to have taken notice of this phenom., since he says, it seems in all 

 respects a perfectly neutral salt ; whereas these experiments plainly show an 

 excess of acid, unless you suppose that the salt he made differed in this 

 respect from mine. There is another point in which we differ, relating to the 

 precipitation made by metallic solutions : he says, that blue vitriol and a solut. 

 of iron in aqua fortis make a precip. immediately on dropping into a solut. of the 

 neut. arsen. salt ; but that green vitriol and a solution of copper in aqua fortis 

 do not make a precip. till after having stood some time : whereas I have always 

 found that all 4 substances make a precip. the instant they are dropt in. 



