1883.] 



On the Atomic Weight of Manganese. 



45 



This must be the case if, as Forchammer maintains, pure manganous 

 salts are colourless, the pink colour of manganese salts being due to 

 traces of a manganic compound. Forchammer has observed that on 

 fusing manganese sulphate with potassium hydrogen sulphate, a white 

 mass is obtained which gives a colourless solution. "We have been 

 unable to prepare any chloride or bromide without a pink or rose 

 colour giving a correspondingly coloured solution, and this was also 

 the case with specimens fused in hydrogen and hydrochloric acid gas. 

 The effect of a trace of manganic salt in the chloride would be to 

 lower the atomic weight. The chloride and bromide of manganese 

 .are both not only very hygroscopic, but if fused in hydrochloric acid 

 gas (or hydrobromic acid in the case of the bromide) are liable to 

 retain traces of the halogen acids, and this would consequently make 

 the atomic weight too low. 



In order to ascertain the values of the atomic weight of manganese 

 which result from careful analysis of the halogen salts, determina- 

 tions were made of the molecular weights of chloride and bromide on 

 specimens prepared with great care. The number found for the 

 bromide was 214*87, and for the chloride 125 "825, yielding the 

 respective atomic weights of manganese of 54' 9 7 and 54*91. All 

 researches on the oxides of manganese have shown that they are all 

 difficult to obtain in anything like a definite form, with perhaps the 

 exception of the protoxide. 



It occurred to us that the analysis of silver permanganate might be 

 employed with advantage, as this salt is found in a very definite state, 

 und can be easily freed from all the allied metals. The selection of 

 this substance, moreover, involved only the atomic weights of silver 

 and oxygen, and as it seemed feasible to deduce the atomic weight 

 ■of the manganese directly from the percentage loss of oxygen on 

 heating, we expected to get very accurate results. In this we were 

 disappointed, as we have not been able to obtain concordant results 

 by this most direct method. 



Table I. 





Weight of silver 

 permanganate. 



Weight of residue. 

 Ag + MnO. 



Oxygen 

 lost. 



<p 

















In air. 



In vacuo. 



In air. 



In vacuo. 





a- 1 



I.. .. 

 II .. 



III. . 



IV. . 



V. .. 



5 "8688 

 5 -4981 

 7 -6725 

 13 -0997 



12 -5782 



5 -8696 

 5 -4988 

 7-6735 

 13 -10147 



12 -5799 



4 -6320 

 4-3358 

 6-0538 

 10 -3179 

 J 9 -9104 

 1 9 -9141 



4 -63212 

 4-33591 

 6 -05395 

 10 -31815 

 9-91065 

 9 -91435 



1 -23748 

 1 -16293 



1 -61959 



2 -78332 

 2 -66925 

 2 -66555 



227 -673 



226 -965 



227 -422 



225 -943 



226 -22 

 226 53 



