FREE AND PERFECTLY ELASTIC MOLECULES IN A STATE OF MOTION. 
5 
“ There seem to be errors in the formulae of alcohol and ether on p. 49, for they do 
not agree with those in the table. They ought probably to be written 
2 (HCJ + CL2EL and 4 (HC,) + 0,21b. 
« Considering how nearly Waterston approached what is now believed to be the 
true theory, it is disappointing to read his controversy with Odling in 1863 and 1864 
(‘Phil. Mag.,’ vols. 26 and 27), where he seems to oppose the new formula then being 
introduced. He is very dogmatic about the constitution of hydrate of potash : he 
very properly insists that we can only obtain a knowledge of the molecular weight of 
bodies that can be volatilized, and of which the vapour densities can be determined, 
but he does not see the analogy between the hydrate and oxide of potassium with 
alcohol and ether, probably because he regards these latter bodies as combinations of 
water with different quantities of olefiant gas. He writes water HCL = 9, alcohol 
CHJICL = 23, and ether C 3 H 4 .HOj = 37, whilst he considers potassic hydrate 
KO,. HCL = 56, and oxide of potassium KCL = 47, the hydrate having a higher 
molecular weight than the oxide. If we regard these compounds as derived from 
water by the replacement of hydrogen by ethyl and potassium respectively, the 
analogy between the two series is complete (ethyl was discovered in 1849 and is 
mentioned by Waterston). 
H 3 0 = 18 HoO = 18. 
(C 3 H 5 ) HO = 46 KHQ = 56. 
(C 3 H 5 ) 3 0 = 74 KoO =94. 
“ From a remark in the ‘ Phil. Mag.’ (vol. 26, p. 520), I imagined that Waterston 
had arrived at the double atomic weights of many of the metals now adopted, for he 
gives that of iron as 56 and that of aluminium as 27 calculated from their specific 
heats, but there is an error in his arithmetic, for 3'3 divided by the specific heat of 
iron ‘1138 gives 28098, and 3‘3 divided by the specific heat of aluminium ’214 3 
gives 15'399.” 
With the exception of some corrections relating merely to stops and spelling the 
paper is here reproduced exactly as it stands in the author’s manuscript.—Dec. 1891. 
[Authors Introduction .] 
Of the physical theories of heat that have claimed attention since the time of 
Bacon, that which ascribes its cause to the intense vibrations of the elementary parts 
of bodies has received a considerable accession of probability from the recent experi¬ 
ments of Forbes and Melloni. It is admitted that these have been the means of 
demonstrating that the mode of its radiation is identical with that of light in the 
quantities of refraction and polarization. The evidence that has been accumulated in 
favour of the undulatory theory of light has thus been made to support with a great 
