22 Mr. W. Sutherland on the Fundamental 



multiples of 3*8; so that the values of f (R){^(Cl)--^(Br)} 

 are represented by the form N x 1*06, in which N is an 

 integral number characteristic of each metal: as ^(Cl) — ^r(Br) 

 remains constant, the values of ^ (R) for the metals are pro- 

 portional to a series of integral numbers. 



It will be well to arrange these integral numbers according 

 to the natural classification of the elements ; the values for 

 •JSr and JBa can be derived from the values for CI— Br in 

 Table V., they are 8 and 8. 



Table VII. — 

 Integers proportional to ^(R) for the Metals. 



Li. 

 9 



Na. 

 9 



8 



Cu. 

 6 



Ag* 

 5 



^Mg. 

 11 



*Oa. 

 10 



8 



Pa. 



8 



9 



JOd. 



6 



3 



iPb. 

 6 



Ml. 

 8 



Ms. 

 6 





This table shows that in each main family the integer 

 decreases with increasing atomic mass : if the value for -JSr 

 were 9 the series of values in the Mg family would run 11, 

 10, 9, 8 ; in the zinc family the series is regular, namely, 

 9, 6, and 3 ; but the data are not extensive enough or 

 accurate enough to make sure whether in each natural family 

 the integers form an arithmetical progression ; however, the 

 data of the last table are very suggestive of the probable truth 

 of the following addition to our third result, namely, that the 

 values of ^r(R) in each family form an arithmetical progression. 



An inspection of the heats of formation of the iodides in 

 Table V. leads to our fourth conlusion, for it is noticeable 

 that there is a large range in the values from 74'2 for KI to 

 — '6, or say for HI. This makes it probable that in 

 (R) +/(RI) + (I) the value of the constant part (I) is small 

 compared to the variable part (R) +/(RI) : we are therefore 

 encouraged to assume (I) = 0. It must be remembered that 

 (I) denotes that part of the heat of combination of a gramme- 

 atom of iodine which is independent of the nature of the atoms 

 with which it is uniting, the iodine being supposed gaseous 

 at 18° C, and therefore having a diatomic molecule : thus (I) 

 has nothing to do with the heat of dissociating the iodine 

 molecule into atoms, which on thermodynamical grounds 

 Boltzmann has calculated as 14*2 kcal. per gramme-atom 

 (Wied. Ann. xxii.). It appears, then, that we cannot be 

 far wrong in putting (I)=0, and then from the equation (1) 

 above we must put (C1)=0 and (Br)=0. In explanation of 

 this result to which our assumption has led, it may be urged 



