414 Dr. Thomson on fJuNE, 



the reason of the gradual deposition of sulphuret of silver, and 

 hence the reason of the length of time necessary to throw down 

 the acids in the liquid by means of nitrate' of silver. The 

 muriatic acid was iiinnediately thrown down in the state of 

 chloride of silver ; but the hyposulpliurous acid was converted at 

 first into hyposulphite of silver, which slowly deposited sulphuret 

 of silver. Thus the ninriatic acid was thrown down in the state 

 of chloride of silver, while the sulphur of the hyposulphurous 

 acid was precipitated in the state of sulphuret of silver. 



Thus the action of water upon the subbichloride of sulphur 

 cannot be completely understood till Xve know the properties of 

 hyposulphurous acid and of some of its salts; but when these 

 are once known, nothing is simpler or more beautiful than the 

 series of decompositions which take place during the analytical 

 experiments which I have described in this paper. It is not 

 surprising then that I did not succeed in my attempts to analyze 

 the chloride of sulphur in the year 1803. I at that time had no 

 notion of the existence of hyposulphurous acid, and was not 

 aware of any other acid compounds of sulphur, except sulphuric 

 and sulphurous acids. It will be seen by consulting my paper 

 published in 1803, that there was no sulphurous nor sulphuric 

 acid in the chloride, but that one or other of them made their 

 appearance when the chloride was agitated with water. It will 

 be seen too that when the liquid was precipitated by nitrate of 

 silver, the precipitate had a brown colour, and of course must 

 have been contaminated with sulphuret of silver. I was not able 

 at the time to form any notion of the source of this sulphuret, 

 and did not even attemj)t to determine its quantity. Indeed I 

 thought the quantity of it was so small as to be insignificant, as 

 far as the analytical results were concerned ; but the preceding 

 <letails are sufficient to show us that this notion was very ill 

 founded. Had I subjected 100 grs. of the subbichloride of sul- 

 phur to analysis, as 1 did in 1803, it is obvious that the weight 

 of the sulphiu'et of silver formed during the analysis would have 

 amounted to 3o*27 grs. ; while the chloride of silver would have 

 amounted to 195*1 grs. Now if the 35'27 grs. of sulphuret had 

 been reckoned chloride (as I did in the analytical experiments 

 published in 1803), they would have indicated 8-7 grs. of chlo- 

 rine ; whereas in reality they indicate 4"47 grs. of sulphur. 

 Thus the quantity of chlorine would have been overrated to the 

 araomit of 8*7 grs. ; while the quantity of sulphur would have 

 been made 4^ grs. beloA\- the truth. 



It was equally out of the power of Berthollet and Bucholz at 

 the periods when they made their respective analyses to take 

 a correct view of the phenomena. They were just as ignorant 

 of the nature and properties of hyposulphurous acid as 1 myself 

 had been. Of course their experiments do not lead us to any 

 correct views respecting the constituents of the chloride of sul- 

 phur which they subjected lo analysis ; neither are they suscep- 



