Memoirs presented to the Cambridge philosophical society on the occasion of the jubilee of Sir George Gabriel Stokes, bart., Hon. LL. D., Hon. SC. D., Lucasian professor.

312 PROF. LIVEING, EFFEOTS OF DILUTION, TEMPERATURE, ETC. ON THE
towards the less refrangible side, and the diffuse absorption at the most refrangible end
extends lower down the spectrum than with an aqueous solution of equal concentration.
The general relation between the spectra of the two solutions will be seen on comparing photographs (1) and (2) of plate 19, of which the former is given by the aqueous,
the latter by the alcoholic solution. The shift of the bands towards the red is visible
in the photographs, but as the plate had to be shifted between the exposures, no reliance
can be placed on the appearance of a shift in such photographs, when the amount of
displacement of the bands is small. This defect is, however, met by direct eye-observations,
with the two spectra in the field of view at the same time. In this way it is seen
that all the bands that are visible are shifted towards the red, but are by no means
all equally shifted. At the same time the strong groups of bands in the yellow and
green have, by the action of the alcohol, undergone a modification of their general
appearance which simulates the addition of some new bands; but by examining solutions
of different concentrations I have satisfied myself that no new bands make their
appearance, but the simulation of them is due to the widening and unequal shift of
the bands, whereby their overlapping, and the consequent relative positions of the maxima
of absorption, are modified. The modifications are such as we may reasonably ascribe
to the influence of the bulky colloid molecules of the alcohol, amongst which the vibrating
absorbent molecules move and from which they can hardly ever get free, loading them
but loading them unequally, and on the whole degrading the rates of their vibratory
motions.
A very remarkable, and by far the most excessive, modification of the bands that I
have observed, is produced by passing dry hydrochloric acid into the alcoholic solution.
The third photograph of plate 19 shews the effect. The colour of the solution is changed
by the acid from pink to bluish green, and the reason of this is obvious from the
photograph. The molecules seem so loaded as to be nearly incapable of taking up the
more rapid vibrations corresponding to the bands in the indigo and blue, while they
seem to absorb more strongly those of slower rate in the yellow and citron. At the
same time these are more degraded than by alcohol alone, and the group in the yellow
so spread out that some of the components are distinctly separated. Of course the acid
makes the solvent a complicated mixture, including ethyl-chloride and water as well as
the unaltered components.
The modifications of the spectrum by glycerol are of the same character as those
produced by alcohol. The bands are generally shifted towards the red, and are more
diffuse, but otherwise not much modified. Plate 20 shews the spectrum of the glycerol
solution above and below that of an aqueous solution of didymium nitrate of nearly,
but not exactly, equal concentration. Observed directly by eye it is seen that the
band in the red at X 679 is not sensibly affected, the group in the yellow and the
less refrangible of the two groups in the green, are distinctly shifted towards the red,
but otherwise not affected in character; while the more refrangible group in the green
is not sensibly shifted, but appears weakened by diffusion. The still more refrangible
bands are all rendered more diffuse by glycerol, and are also degraded with the exception