132 Professor Julius Wilhelm Brilhl [May 26^ 



Later researches have confirmed this. The univalent halogens 

 give, like hydrogen, constant atomic refractions, both in the element- 

 ary state and in their compounds. The multivalent elements, on the 

 other hand, such as oxygen and nitrogen, display different optical 

 values, according to the kind of bond. 



In the course of such researches the behaviour of oxygen as a 

 quadrivalent element, which had been previously conjectured, was 

 established with certainty, and afterwards confirmed synthetically by 

 Collie, Tickle, and others. 



§ 20. The theory which accounted for the optical abnormalities 

 of certain classes of bodies, making them in fact abnormalities no 

 longer, has proved extraordinarily fruitful. It formed the starting- 

 point of aU subsequent discoveries in the subject, and indeed we may 

 describe the progress of this branch of science during the last 25 

 years as based essentially on this conception. 



For not until we had fathomed the mystery of the benzene re- 

 fractive increment 6, was it possible to know for certain that the- 

 variable valency of the multivalent elements is always of determining, 

 influence on the optical behaviour of bodies. Thus for the first time 

 a spectrochemical method was called into being for the study of 

 chemical structure, and the foundations were laid of what we now 

 call " Spectro-chemistry." 



YII. 



§ 21. We must now return once more to the formula for 



— ~= — j P had proved not constant 

 for the temperature in the case of fluid bodies, and was, therefore, re- 

 placed by Gladstone and Dale's more satisfactory ratio ( ~ j P. 



For 20 years and more this did admirable service. As, however, the 

 number of observations kept on increasing, even this formula betrayed 

 imperfections which flnally led to its abandonment. It is impossible 

 here to follow the argument in detail, and we must be content with 

 the remark that comparisons of bodies in different states of aggregation 



failed to yield satisfactory constants. The values of ( — -^ — j P f or a 



fluid or solid substance always came out considerably greater than for 

 the same substance in the state of gas or vapour. 



% 22. Then by a happy chance two physicists, L. Lorenz, of 

 Copenhagen, and H. A. Lorentz, of Leyden, came forward simulta- 

 neously in 1880 with a new expression for refraction. One of them 

 started from the ordinary theory of light, the other from Maxwell's 

 electromagnetic theory of light based on Faraday's views, and they 



