BY THEIR OPTICAL PROPERTIES. 4S5 
fluorescence taking, as it were, a fresh start, with a different colour, we may be pretty 
sure that we have to deal with a mixture of two fluorescent substances. 
It might be inferred a priori, that fluorescence at any particular part of the spectrum 
would necessarily be accompanied by absorption, since otherwise there would be a crea¬ 
tion of vis viva; and experience shows that rapid absorption (such as corresponds to a 
well-marked minimum of transparency indicated by a determinate band of absorption in 
the transmitted light) is accompanied by copious fluorescence. But experience has 
hitherto also shown, what could not have been predicted, and may not be universally 
true, # that conversely, absorption is accompanied, in the case of a fluorescent substance, 
by fluorescence. 
From what precedes it follows that the colour of the fluorescent light of a solution, 
even when the incident light is white, or merely sifted by absorption, may be a useful 
character. To illustrate this, the electric light, after transmission through a deep-blue 
glass, was thrown on solutions in weak ammonia of two crystallized substances, sesculin 
and fraxin, obtained from the bark of the horse-chestnut, and of which the latter occurs 
also in the bark of the ash, in which, indeed, it was first discovered. Both solutions ex¬ 
hibited a lively fluorescence; but the colour was different, being blue in the case of 
aesculin, and bluish-green in the case of fraxin. A purified solution obtained from the 
bark exhibits a fluorescence of an intermediate colour, which would suffice to show that 
scsculin would not alone account for the fluorescence of the solution of the bark. 
When a substance possesses well-marked optical properties, it is in general nearly as 
easy to follow it in a mixture as in a pure solution. But if the problem which the 
observer proposes to himself be:—Given a solution of unknown substances which presents 
well-marked characters with reference to different parts of the spectrum, to determine 
what portion of these characters belongs to one substance, and what portion to another, 
—it presents much greater difficulties. It was with reference to this subject that the second 
of the objects mentioned at the beginning of the discourse had been spoken of as that 
the attainment of which was by far the more difficult. The problem can, in general, be 
solved only by combining processes of chemical separation, especially fractional separa¬ 
tion, with optical observation. When a solution has thus been sufficiently tested, those 
characters which are found always to accompany one another, in, as nearly as can be 
judged, a constant proportion, may, with the highest probability, be regarded as belong¬ 
ing to one and the same substance. But while a combination of chemistry and optics is 
in general required, important information may sometimes be obtained from optics alone. 
This is especially the case when one at least of the substances present is at the same time 
fluorescent and peculiar in its mode of absorption. 
To illustrate this the case of chlorophyll was referred to. An eminent French chemist, 
M. Fremy, proposed to himself to examine whether the green colour were due to a single 
substance, or to a mixture of a yellow and a blue substance. By the use of merely 
neutral bodies, he succeeded in separating chlorophyll into a yellow substance, and 
another which was green, but inclining a little to blue; but he could not in this way get 
further in the direction of blue. He conceived, however, that he had attained his object 
by dissolving chlorophyll in a mechanical mixture of ether and hydrochloric acid, the 
acid on separation showing a fine blue colour, while the ether was yellow. Now solu¬ 
tions of chlorophyll in neutral solvents, such as alcohol, ether, etc., show a lively fluor¬ 
escence of a blood-red colour; and when the solution is examined in a pure spectrum, 
the red fluorescence, very copious in parts of the red, comparatively feeble in most of the 
green, is found to be very lively again in the blue and violet. Now a substance of a pure 
* Fluorescent substances, like others, doubtless absorb the invisible heat-rays lying beyond 
the extreme red, in a manner varying from one substance to another. Hence, if we include 
such rays in the incident spectrum, we have an example of absorption not accompanied by 
fluorescence. But the invisible heat-rays differ from those of the visible spectrum (as there 
is every reason to believe) only in the way that the visible rays of one part of the spectrum 
differ from those of another, that is, by wave length, and consequently by refrangibility, 
which depends on wave length. Hence it is not improbable that substances may be dis¬ 
covered which absorb the visible rays in some parts of the spectrum less refrangible than 
that at which the fluorescence commences; and mixtures possessing this property may be 
made at pleasure. Nevertheless, the speaker has not yet met with a pure fluorescent sub¬ 
stance which exhibits this phenomenon. 
