and of Simple Substances, 183 



mide and chloride of barium they had not the precision and dis- 

 tinctness necessary for making calculations from them. 



In the spectra of those haloid salts of barium whose light con- 

 sists of more strongly refracted rays, the distances of the indi- 

 vidual lines from each other are less than in the spectrum of each 

 haloid salt whose corresponding lines are produced by light of 

 greater wave-length. For this relation I have also succeeded in 

 finding an expression. 



The distance of the more strongly refracted bright line of the 

 chloride of barium from a definite point of the scale, is to the 

 distance of the other bright line from this same point as the 

 distances from the same starting-point of the corresponding lines 

 of the spectra of iodide and bromide of barium. In order to find 

 the starting-point, whose distance from that of the most strongly 

 refracted line of the chloride-of-barium spectrum is expressed by 

 y, I formed from the spectra of chloride of barium and bromide 

 of barium the following equation, in which the unit is a degree 

 of the scale : — 



Distance of the first chloride-of-barium line y 



Distance of the second chloride-of-barium line y ■+■ 3*9 



Distance of the first bromide-of-barium line y + 3 



Distance of the second bromide-of-barium line y + 8* 2* 



from which y = 9'0. 



From the spectra of chloride of barium and iodide of barium 

 the equation is obtained, 



— rrrR = ^n n from which y = 9*5. 

 y + 3'9 y + 16-5' u 



From those of bromide of barium and iodide of barium, 



y-1-9 2/ + 3 , 



, 1g g = , 5 n J from which 2/ = 10'5. 

 y + 16'5 y + 8*2 9 



From the above equations we obtain for the position of the com- 

 mon starting-point 96*1—?/, that is, 96-1 — 9*7 = 86*4 degrees*. 



From these and the preceding equations, it follows that if two 

 spectra of the chloride, bromide, or iodide of barium are known 

 by the first equations, the distances from each other of the prin- 

 cipal lines in the spectrum of the third compound may be cal- 

 culated from them. From the latter equations, which have 

 given the starting-point, the position of these lines may be cal- 

 culated. 



* I have designated the starting lines in the spectra of the metals by the 

 letter a . 



