GASEOUS REFRACTIVE INDICES 287 



by collecting a large number of observations, and by carefully 

 comparing the observed changes in refractivity, are we likely 

 to make any progress in unravelling these difficult problems. 

 But though the importance of this branch of the subject has 

 been insisted on by every investigator who has approached it, 

 several circumstances have conspired to prevent results of 

 interest being obtained from its study. Not only is the number 

 of gaseous compounds comparatively small, but the number of 

 elements whose indices were known was till recently so re- 

 stricted as to limit still further the choice of compounds on 

 which the comparison could be made; and, to crown all, the 

 first compounds whose indices were measured now prove to 

 have been the most anomalous in the whole range of physics. 

 It is not surprising, therefore, that research in this direction 

 was not pursued very eagerly. Biot and Arago could only 

 make use of water and ammonia ; and the errors in their work, 

 due to the primitive condition of both chemistry and physics at 

 the time, render the results of little interest except historically. 

 They concluded that the refractivities of the compounds are 

 the sum of the refractivities of their constituents, at least in 

 compounds such as ammonia, " where the condensation is not 

 very great." They are thus responsible for formulating the 

 additive law for gaseous indices. 



Dulong (1826) published the results of a systematic research 

 on twenty-two gases and vapours. By assuming a certain 

 probable value for the refractive power of carbon, he was able 

 to utilise, in a fashion, the figures for compounds which con- 

 tained this element, and he concluded that the refractive powers 

 of gases have no relation to their densities, and have nothing 

 to do with the heat of combination, and that, as a rule, the 

 refractive power of a compound is greater than that of its 

 components, when the compound is neutral or alkaline, and 

 less when acid. He added that the difference of refractivity 

 appears to depend on the electrical state proper to the molecules 

 of each sort of matter, and that on the undulatory theory they 

 appear to retard light in proportion to their positivity. To 

 these conclusions he was probably led by the values he 

 found for N 2 0, NO, and NH 3 which, as will be shown later, are 

 anomalous. But it must be admitted that it was a very shrewd 

 guess which, in 1826, connected refractivity with electrical 

 state. 



