1820.] of different inorganic Bodies. 91 



my new researches, a difficulty presented itself which I was not 

 aware of in my former ones, and which laid me under the neces- 

 sity of making in the first place a great number of experiments 

 in order to find a method of operating which should always, give 

 the same results. The difficulty consists in this, that just when 

 the oxymuriate begins to be decomposed, the oxygen carries 

 along with it a portion of oxymuriate in the form of a white 

 smoke, and deposits it in the tubes. This smoke is deposited 

 so slowly that when I passed the oxygen through a glass tube 

 two feet in length, curved in different directions, and drawn out 

 at the end so as to form a tube almost capillary, the last half of 

 the tube was not covered with dust ; but a piece of glass against 

 which I allowed the jet of gas to strike became entirely covered 

 with it at the place where the gas struck it. As the formation of 

 this saline powder seemed to be merely the mechanical effect of 

 the boihng (for it increased with this last), 1 endeavoured to 

 avoid boiUng by mixing it with from six to ten times its weight 

 of fused muriate of potash. By this method 1 gained my object, 

 and I then proceeded in the following manner : Oxymuriate of 

 potash, the solution of which was not at a'l rendered muddy by 

 nitrate of silver, was reduced to a fine powder, and dried in a 

 temperature higher than 212°, but not sufficiently so to soften 

 the salt. It was then put while hot into a retort which contained 

 muriate of potash in powder. It had been heated to disengage 

 all moisture, and its weight was known. When the mixture had 

 recovered the temperature of the atmosphere, the retort was 

 weio-hed again. It was then shaken to mix the two salts. By 

 means of a caoutchouc tube, a glass tube 16 inches long, and a 

 fine in diameter, was attached to the retort. Into this tube 

 some pieces of dry muriate of lime had been put. It was curved 

 into a spiral, and the opening surrounded by filtering paper to 

 retain all the saline smoke in powder. Care was taken to weigh 

 the tube both alone and in connexion with the retort. The quan- 

 tity of oxymuriate decomposed varied from 15 to 10 grammes. 

 The retort was heated in a sand-bath till it became red within, 

 and till the muriate began to soften. When the experiment was 

 finished, the oxygen gas within the retort was changed for atmo- 

 spheric air. The retort was then weighed along with the tube, 

 and finally the tube alone. 



In all these experiments, a trace of sublimate was found in the 

 beak of the retort. It was oxymuriate undecomposed. It 

 weighed in all 0-003 gr. The tube with the pieces of muriate 

 of lime and paper had increased in weight from 0*02 to 0-023 gr. 

 By subtracting this quantity of weight from what the retort had 

 lost, we obtained the quantity of oxygen disengaged, and at the 

 same time that of the muriate of potash remaining in the retort. 

 Four experiments were made in this manner, and taking into the 

 account the small quantity of sublimate, 100 parts of the oxymu- 

 riate gave 



