﻿388 Mr. J. A. Crowther on the Coefficient of 



If we assume that combination does not influence the 

 absorption of the combining atoms, we can easily calculate 

 the value of the ratio X/p for a compound from the knowledge 

 of its value for the constituent elements. A comparison 

 between the values so calculated and those obtained by 

 direct experiment will then enable us to test the truth of our 

 assumption. 



Let X and Y be two elements of atomic weight m 1 and m 2 

 respectively. Since, the number of corpuscles in the atom is 

 proportional to the atomic weight, the number of corpuscles 

 in the atoms of X and Y respectively will be hm x and hn 2 . 

 Since the absorption per corpuscle is, as previously explained, 

 equal to the ratio \/p, the absorption per atom of the element 



X will be — km h and that of Y, — km 2 . 

 Pi 92 



If, now, the compound is represented by the formula 

 X a Y 6 , the absorption per molecule will on our assumption be 

 equal to the sum of the absorptions of the individual atoms, 

 and must therefore equal 



a - 1 km-i + o — hn 2 . 

 Pi P2 



But the absorption per molecule of a compound will also 

 equal the product of the average absorption per corpuscle 

 multiplied by the number of corpuscles in the molecule ; that is, 



it will equal —(ahn 1 -\-bkm 2 ), where \ and p are the values 



for the compound. Thus equating the two expressions we 

 have 



A/i 7 Ao 



a . — m x -(- b . — m Q 



* ..... Pl P2 " 



p ami + bm 2 



(compound) 



which is known when the values of the absorption per corpuscle 

 for the component elements are known. 



The following table shows the nature of the agreement 

 between the values calculated from the above formula and the 

 direct experimental values, for fourteen different compounds. 



The value of the absorption of oxygen could not be directly 

 determined. The value used in calculating the results in the 

 above table was calculated from the observed absorption 

 of the different oxides, by means of the formula obtained 

 above. 



On the whole, the agreement between the observed and the 



