1887.] 
Mr W. Durham on Chemical Affinity. 
53 
Now, it is apparent at once that the quantity dissolved is almost 
exactly inversely as the difference of heats, which is in complete 
accordance with the laws laid down in my former paper. With the 
nitrates the same relationship cannot be made so clear, as the 
metals are in combination with O, and the data obtainable are 
insufficient to trace the various affinities. The results, however, are 
quite in accordance with the above, although the range is much 
greater. Thus we have— 
Quantity of M(N0 3 ) 2 dissolved. 
Ca = 1 1 1 grains. 
Sr = 50 „ 
Ba = 7 „ 
The differences are in the same proportion as in the chlorides. The 
sulphates I have already noticed. 
The salts of those metals which form insoluble oxides or hydrates 
I leave for future treatment, as the data obtainable are defective for 
my purpose. So far as they go, however, they are in complete 
accordance with those laws I have stated. 
These facts seem to me to prove, without doubt, that solution is 
entirely due to chemical affinity, and that chemical affinity does not 
act, as has hitherto been supposed, in units, but in all proportions 
according to the circumstances, and that in chemical combinations 
of all degrees every atom acts upon every other atom according to 
its affinity and the position in which it is placed. This way of 
regarding chemical affinity reduces a perfect chaos of empirical 
results into an orderly and systematic arrangement. 
3. Thermometer Screens. Part IV. By John Aitken, Esq. 
(Plates II., Ill, IV.) 
The object of this paper is to describe a new thermometer screen, 
and to give the results of some trials made this autumn and winter 
with a Stevenson screen as generally used, and one modified in the 
way described in a previous part of this investigation.* Also to 
give comparative readings taken with those screens and with the 
new one. 
*“ Thermometer Screens,” Proc. Roy . Soc. Edin . , Part 117, p. 661. 
