MOLYBDENUM. 139 



graphs we get somewhat noticeable results. From the car- 

 bon dioxide series, Mo = 91.711, ± .113, a figure having no 

 unusual interest. From the other series, if S = 31.987 and 

 O = 15.9633, we get Mo = 92.979, ± .354 ; but if we take 

 S = 32 and = 16, then Mo becomes = 92.133. In this 

 case the higher values for oxygen and sulphur lead to a 

 lower number for molybdenum. In the carbonate series 

 the assumption of 12 and 16 for C and 0, respectively, makes 

 Mo = 92.033. In other words, if we assume the ordinary 

 even numbers for C, 0, and S, Svanberg and Struve's two 

 methods yield more nearly concordant results than when 

 the revised values for these elements are taken. 



Berlin,''' a little later than Svanberg and Struve, deter- 

 mined the atomic weight of molybdenum by igniting a 

 molybdate of ammonium and weighing the residual M0O3. 

 Here, again, a loss of the latter by volatilization may (and 

 probably does) lead to too low a result. The salt used was 

 (NIl4)4Mo50i -. 3 HoO, and in it these percentages of M0O3 

 were found : 



81.598 

 81.612 

 81.558 

 81.555 



Mean, 81.581, =h .0095 



Hence Mo = 91.9817, ±: .0776 ; a result agreeing quite 

 well with those of Svanberg and Struve. 



Until 1859 the value 92 was generally accepted on the 

 basis of the foregoing researches, but in this year Dumas f 

 published some figures tending to sustain a higher number. 

 He prepared molybdenum trioxide by roasting the disulph- 

 ide, and then reduced it to metal by ignition in hydrogen. 

 At the beginning the hydrogen was allowed to act at a 

 comparatively low temperature, in order to avoid volatiliza- 

 tion of trioxide ; but at the end of the operation the heat 



* Journ. fur Prakt. Chem.,49, 444. 1850. 

 f Ann. Chem. Phann., 105, 84, and 113, 2\ 



