OF HIGH REFRAIN GIBILIT Y UPQH GASEOUS MATTER. 
343 
cause it to fluoresce. Both substances quench the effectual rays close to the surface at 
which they enter. 
I endeavoured at the time to apply this fact to the solution of the question whether 
the absorption of chemical energy was the act of the molecule as a whole, or of its 
constituent atoms. I tried to show that on the first of these assumptions it is impossible 
for the self-same rays to be absorbed by a liquid and its vapour. For absorption depends 
upon the rate of molecular vibration, and reaches its maximum when this rate synchro- 
nizes perfectly with the rate of succession of the ethereal waves. Now as the rate of 
molecular vibration depends upon the elastic forces exerted between the molecules, and 
as it could hardly be imagined that these forces would remain undisturbed during the 
passage of a vapour to the liquid condition, the fact of the liquid nitrite of amyl and 
its vapour absorbing the same rays indicated that the absorption was not molecular. 
We were thus driven to conclude that it was atomic*; and this conclusion was fortified 
by the consideration already adverted to, — that were the absorption the act of the mole- 
cule as a whole, no mechanical ground could be assigned for the falling asunder of its 
atoms. Thus actinic action itself pointed out the seat of the absorption. 
A wide, if not entire generality was anticipated for the proposition that the same rays 
are absorbed by a liquid and its vapour. I have now no reason to retract this anticipa- 
tion ; but when it was expressed I believed that liquids in general would be found so 
destructive of the effectual rays as to render transmission through moderate depths of 
them sufficient to rob a beam of all power to act upon their vapours. This idea, enter- 
tained though not expressed, has not been verified, and the deportment of iodide of 
allyl may be taken as representative of a class of facts which contradict it. 
Glass cells were employed varying from one-eighth of an inch to an inch in width. 
Filled with the transparent iodide, these cells were placed between the electric lamp and 
the experimental tube charged with the iodide vapour. The rays after traversing an 
inch of the liquid produced copious decomposition in the tube. A marked distinction 
was thus proved to exist between the liquid iodide of allyl and the liquid nitrite of amyl. 
But the same distinction extends to their vapours. The exceeding absorbent avidity 
of the nitrite-of-amyl vapour, and the rapidity with which it deprives a powerful beam 
of its effective constituents, have been already noticed. It is quite different with the 
iodide of allyl. A tube 5 feet long was charged with the iodide vapour, and after it, in 
the same line, was placed another tube 3 feet long charged with the same vapour. On 
sending abeam through both tubes in succession, the 5-foot tube, through which the light 
first passed, was filled immediately with an actinic cloud ; but a similar cloud was at the 
same time falling in the second tube. A transmission through 5 feet did not seem to 
* When I use the word “ atomic ” in contrast with “ molecular,” I hj no means pledge myself to an absolute 
limit of divisibility. The molecule may resemble a house, the atoms the hard bricks composing that house. But 
while it is both convenient and correct to regard the house as constituted of bricks definitely hounded, it is by 
no means essential to regard the bricks themselves as absolutely indivisible. The divisibility or non-divisibility 
of the atoms does not in the least affect the atomic theory as a, working conception. 
2 z 2 
