210 M. Dumas on the Equivalents. 



great accuracy, D^imas judged it not unnecessary to make a de- 

 termination of it by a new method. This consisted in heating 

 pure granulated silver in a tube in a stream of chlorine, and de- 

 termining the equivalent from the difference between the original 

 v,-eight of the silver and that of the chloride formed. The tube 

 was provided with a stopcock by which it could be exhausted, and 

 a similar tube was used as a counterpoise. Three weighings 

 were necessary: — 1st, that of the exhausted tube; 2nd, of the 

 exhausted tube containing the silver; 3rd, of the exhausted tube 

 containing the chloride of silver formed. Various experiments 

 were made with different weights of silver. Assuming the equi- 

 valent of silver at 108, as found by Marignac's very accurate de- 

 terminations, the result of Dumas's experiments gave 35-5 as 

 the equivalent of chlorine, which thus confirms the number 

 at present adopted. 



The determination of the equivalent of copper was made partly 

 by the reduction of oxide of copper, and partly by converting 

 copper into sulphide of copper, but the results were not so con- 

 cordant as to give any certain datum. The number lies, how- 

 ever, between 31 and 3.2. 



These numbers do not agree with Front's law; and Dumas 

 therefore concludes that that law must be accepted in the less 

 general form, that the equivalents of simple bodies are almost all 

 multiples by whole numbers of the equivalent of hydrogen, but 

 that, with reference to chlorine, the unit with which it is com- 

 parable is half as great as that of hydrogen. 



Dumas next discusses the question as to whether there are 

 elements whose equivalents have the relation 1:1 or 1:2. 

 Molybdenum and tungsten have been considered to stand in the 

 latter relation ; and the only question has been whether their 

 equivalents were as 46 : 92 or as 47 : 94. Dumas determined 

 the equivalent of molybdenum by heating molybdic acid in hy- 

 drogen. The molybdic acid was prepared in a pure form, and 

 in beautiful large crystals, by heating sulphide of molybdenum 

 in a stream of oxygen. The molybdic acid placed in an unglazed 

 porcelain tray was heated in hydrogen, at first in a glass tube at 

 a moderate temperature, and then the tray was transferred to an 

 unglazed porcelain tube, and the reduction concluded at a very 

 high temperature. The results gave 48 as the equivalent of 

 molybdenum. 



Great difficulties were experienced in determining the equiva- 

 lent of tungsten. The readiness with which tungstic acid is 

 reduced and forms lower degrees of oxidation, renders the pre- 

 paration of pure tungstic acid very difficult. It was obtained 

 by heating pure tungstate of ammonia in a flat porcelain dish, 

 at first at a low temperature and then at a red heat, until all the 



