138 REPORT— 1882. 



is replaced by the clilorides of sodium, ammonia, or zinc. The bands 

 are nearest the red with potassic chloride, and nearest the blue with zinc 

 chloride. 



Before leaving this part of the report, we may just refer to some 

 experiments made by Professor Melde ' to decide the question whether two 

 coloui'ed liquids have the positions of their absorption bands altered when 

 mixed together. He found indeed that snch an effect conld be observed, 

 but all his experiments admit of an obvious explanation. To take an 

 ideal case, suppose in Fig. 3 the absorption of a liquid for different waves 

 of light to be graphically represented by the curve AaB, with a maximum 

 at a, and suppose this liquid to be mixed with another whose curve of 

 absorption is represented by CD, the simple addition of the two curves 

 will produce a third curve CA'5 D with a maximum at b. This maximum 

 is obviously not in the same position as before, but takes place at a wave- 

 length at which the two tangents to the two original curves are inclined 

 equally but in opposite directions. All the somewhat complicated rules 

 deduced by Professor Melde are easily explained in this way, and he 

 would have observed exactly the same phenomena if he had put his two 

 liquids in front of each other instead of mixing them together. 



III. Relations of the Spectra of different Elements. 



Various efforts have been made to connect together the spectra of 

 different elements. The attempts in this direction generally assume that 

 certain lines in one spectrum correspond to certain lines in another 

 spectrum, and the question is raised whether the atom with the higher 

 atomic weight has its corresponding lines more or less refrangible. In 

 the opinion of the writer of this report no definite judgment can as yet 

 be given as to the success of these efforts ; some of the relations traced 

 no doubt are interesting and deserve further attention, but most of them 

 are far-fetched, and will very probably be proved to have no foundation 

 in fact. Lecoq de Boisbaudran* has led the way in these speculations, 

 and some of the similarities in different spectra pointed out by him are 

 certainly of value. But whether his conclusion, that 'the spectra of 

 the alkalies and alkaline earths, when classed according to their refran- 

 gibilities, are placed as their chemical properties in the order of their 

 atomic weights,' will stand the test of further research, remains to be 

 seen. Lecoq begins by comparing together the spectra of potassium 

 and rubidium, as well as those of calcium, strontium, and barium. The 

 spectrum both of potassium and rubidium begins, when heated up in the 

 Bunsen burner, with two red lines ; then follow, in the case of potassium, 



» Poffff. cxxiv. p. 91 (1865), and cxxvi. p. 264. ' C. H. Ixix. pp. 445, 606, 657. 



