54 
SIR J. F. W. HERSC1IEL ON THE CHEMICAL ACTION OF THE 
fluous alcohol will have hardly run off, when the phenomena of the thermic spectrum 
begin to appear in all their characters,, at first faintly, and, as it were, sketched in by 
a dimness and dullness of the otherwise shining and reflective surface of the wetted 
paper ; but this is speedily exchanged for a perfect whiteness, marking, by a clear 
and sharp outline, the lateral extent of the calorific rays, and by due gradations of 
intensity, in a longitudinal direction, their law or scale of distribution, both within 
and without the luminous spectrum. 
133. It may appear at first sight as if the process above described for the prepara- 
tion of the paper were not in exact accordance with the principle of action at first 
explained. But a little consideration will show, that, as a coloured paper dries, it is 
the pile or projecting filaments which the film of moisture retreating into the paper 
lays bare, which first catch and reflect the light. By transferring the colour to the 
under side of thin paper, we in effect only provide for the regular and copious exhi- 
bition of such reflective filaments, unsullied by the application of colouring matter. 
The rationale of this will be obvious on trying a few experiments on papers coloured 
through their whole substance, of which the best tints are blue, and that kind of 
greenish brown which is in very common use among bookbinders ; and the best tex- 
ture a soft and bibulous one, into which liquids readily sink, and form very dark 
spots while wet. 
134. The kind of paper is by no means indifferent in another and very important 
respect. It is with thermographic as with photographic papers ; each has its pecu- 
liar scale of action. Each absorbs in preference some above others of the thermic 
rays, and thereby gives an undue preponderance of effect to those so preferred. It 
is therefore only by trials on a great variety of papers that we can satisfy ourselves 
whether any observed interruptions or deficiencies in the thermic spectrum be really 
due to the absence of calorific rays of corresponding refrangibility in the incident 
beam, or be not simply owing to partial insensibility in the particular paper used. 
The smoked paper I have above recommended seems to be more free than any other 
I have tried from this defect. That prepared with Indian ink is materially less sen- 
sitive, both to the most and least refracted heat-rays. In this respect, the smoked 
compared with the inked paper holds the same sort of relation as the broinuretted 
paper compared with a merely nitrated one, in respect of the chemical action. 
135. Supposing, then, (unless where the contrary is expressed,) that such smoked 
paper is employed, the thermograph of the spectrum in its most complete state, or rather 
in that succession of states in which each part comes most characteristically into view, 
is as represented in Plate II. fig.2, where the dimensions are enlarged in the ratio of three 
to two above those actually measured. On the same enlarged scale, fig. 3. represents the 
luminous spectrum as seen with the naked eye, and fig. 4. as seen through the standard 
cobalt-glass used in the above-recited experiments ; while fig. 5. (still on the same 
scale) represents the photographic spectrum impressed on bromuretted paper*; and 
* See Art. 77. 
