14 Mr. W. Sutherland on the 



p. 135). Cuthbertson found first in the He family that 

 10 6 (N'-1)=141 (J, 1, 4, 6, 10), the corresponding result 

 for the halogens being 192 (1, 4, 6, 10). For the first^two 

 members of the oxygen and nitrogen families the values are 

 275 (1, 4) and 299 (1, 4), but for the higher members of 

 these families the values are not expressed accurately enough 

 by the integers 6 and 10. The elements of the carbon family 

 have not yet been examined in the gaseous state for N' — 1, 

 so for these I shall use the relation (N'-1)V=(N-1)B, 

 which will be assumed to be the same as the refraction 

 equivalent or atomic refraction of these elements in their 

 compounds. On account of the attention which has been 

 given to atomic refraction in the past it will be convenient 

 to multiply N' — 1 by 11215 to get the values tabulated as 



(N'-i)v: 



Next as to the values of B to be used in testing (8). If 

 the experimental data were complete enough, the best way 

 to get values of B, the limiting volume of a gramme-atom of 

 each gas, would be to derive it from (2a) 3 , calculating a by 

 the principles of the kinetic theory. But on account of many 

 gaps in the data we are compelled to make provisional ap- 

 proximate estimates of some values of B in the following- 

 way. Molecular diameters for a certain number of gases 

 according to the kinetic theory with the best data available 

 are given in a note appended. For the halogen atoms, both 

 in their organic and inorganic compounds, the values of B 

 are F 9, 01 19, Br 26, and I 36, which are nearly 9 

 (1, 2, 3, 4) (Phil. Mag. [5] xxxix. p. 1). Now for "the 

 gramme-atom of 01 as an element gas the kinetic theorv 

 gives B = 2'77 x 10 19 x 11215 (2a) 3 = 16*5. Again the density 

 of solid iodine, namely, 4*94, gives B = 127/4*94 = 25*6, which 

 is different from 36 in compounds and not double 16*5. 

 Yet for the halogens in the state of elements I shall assume 

 B = 6*6 (1, 2, 3, 4), giving greatest weight to the result for 

 solid iodine and to the law 1, 2, 3, 4. Ooncerning B we 

 must give to the inert gases of the helium family a more 

 detailed study. In the first place we must investigate their 

 virial parameters, which have not been calculated in any of 

 my papers on molecular attraction. Let T e and p c denote 

 the critical temperature and pressure of a substance, then 

 for both compounds and elements I, the parameter of the 

 virial of the attractive forces for a gramme, is proportional 

 to R 2 T c 2 /jo c , where RM is the usual gas constant. The con- 

 stant of proportionality is different for compounds from that 

 for elements. In the absence of knowledge to the contrary 

 I shall assume that the formula for such diatomic gases as 



