248 Dr, Thomson* s Experiments to determine the Weight [April, 



80. It was the knowledge of these facts that led Dr. WoUaston 

 to pitch upon 3*45, and Dr. Prout upon 3*5, as the atomic 

 weight of iron. 



I find by a very careful analysis, that the crystals of protosul- 

 phate of iron are composed as follows ; 



1 atom sulphuric acid = 5*0 



1 atom protoxide of iron = 4*5 



7 atoms water = 7*875 



17-375 



For 17*375 grs. of this salt, when dissolved in water and pre- 

 cipitated by muriate of barytes, give 14*75 grs. of sulphate of 

 barytes, indicating five grains of sulphuric acid. When precipi- 

 tated by ammonia, and the precipitate well dried, it weighs five 

 grs. of peroxide of iron, which is equivalent to 4*5 grs. of protox- 

 ide of iron. The remaining 7*875 grs. must be the water of 

 crystaUization. This view of the subject is confirmed by the 

 following experiment : 



I dissolved 17*375 grs. of the crystals of protosulphate of iron 

 in distilled water : 13*25 grs. of chloride of barium were dissolved 

 in another portion of water, and the two liquids were mixed. 

 After the precipitate had subsided, the residual liquid was neither 

 affected by muriate of barytes, nor sulphate of soda. It is 

 obvious, therefore, that it contained neither sulphuric acid nor 

 barytes; consequently 17*375 grs. of crystallized protosulphate 

 of iron contain exactly five grs. of sulphuric acid. The remain- 

 ing 12*375 grs. consist of protoxide of iron and water. It may, 

 perhaps, be objected to the determination of the weight of an 

 -atom of protoxide of iron at 4*5 from this experiment, that we 

 are not absolutely certain that the portion of water in 17*375 

 grs, of these crystals is exactly 7*875 grs,, or 7 atoms ; but the 

 facts ascertained are perfectly decisive on this point. By the 

 analysis of Berzelius, the water in 17*375 grs. of protosulphate of 

 iron weighs 7*888 grs. By my analysis, it weighs 7*819 grs. 

 Now the mean of these two numbers is 7*853 grs. which is only 

 .^ ggp ths of a grain less than the theoretical quantity ; but if the 

 water in this salt were 8 atoms instead of 7, its weight would 

 amount to 9 grs. instead of 7*875 ; and only 3*375 grs. would 

 remain for the oxide of iron ; but we are perfectly certain that 

 its weight greatly exceeds that quantity. It is evident that the 

 quantity of water in the salt must be a whole number of atoms : 

 '7*875 being the weight of seven atoms of water, and the hquid 

 "'Amounting sensibly to that quantity as nearly as can be esti- 

 mated by experiment, we are entitled to consider it as the exact 

 quantity ; consequently an atom of protoxide of iron must 

 weigh 4*5. This gives 3*5 for the weight of an atom of iron, 

 almost the number pitched upon by Wollaston, and the very 



