18 
SIR J. F. W. HERSCHEL ON THE CHEMICAL ACTION OF THE 
of testing the perfection actually attained in this respect. Neither the refractive nor 
dispersive indices best adapted for an ordinary double achromatic lens are exactly 
those most fitted to collect the chemical spectrum. To determine those which are 
so, the method pointed out in my Treatise on Light* seems, on the whole, best 
adapted, viz. by over correcting the crown lens, by a flint one somewhat too concave, 
and then separating the lenses, till it is found on trial that the adjustment is complete. 
The dispersive ratio best adapted to the purpose will then be found by the formula 
there stated. For the refractive indices, those corresponding to Fraunhofer’s ray F 
will probably suffice. 
53. Another and highly important practical conclusion which seems to be pointed 
to by this experiment, is the possible future production of naturally coloured photo- 
graphic images. The feeble traces of colorific power above noticed might become 
exalted by the use of other combinations, which it therefore became extremely inter- 
esting to vary in every possible way. Mr. Talbot has already observed, that in copy- 
ing, photographically, designs on coloured glass, the red portions would occasionally 
impart a corresponding tint to the impression of that particular portion of the design 
so coloured. The rationale of this fact, as will afterwards appear, is somewhat more 
complex than it would seem to be on perusing the experiment just cited. The bright 
red varieties of coloured glass by no means transmit the red rays in a state of purity; 
and those varieties which do so or nearly so, are incapable of producing the effect in 
question, if the paper be fresh prepared and completely free from all traces of pre- 
vious photographic action. At the moment, however, the direct explanation seemed 
satisfactory, and proved an incentive to further inquiry. 
54. July 9, 1839. — A highly concentrated spectrum was therefore formed, by re- 
ceiving the rays after emergence from the prism used in the last-described experiment 
1 . Colours of the luminous spectrum. 
2. Colours impressed on the paper. 
Extreme redf 
None. 
Mean red 
None. 
Orange 
Faint brick-red. 
Orange-yellow .... 
Brick-red, pretty strong. 
Undecided ; red passing into green. 
Dull bottle-green. 
Dull bottle-green, passing into blueish. 
Very sombre blue, almost black. 
Black, but when the exposure was long continued, metallic 
yellow, like an imperfect gilding. 
Black, passing into the same metallic yellow by long exposure 
in the less refrangible portions of the violet ray. 
Violet-black, or purplish-black. 
Yellow 
Yellow-green 
Green 
Blue-green 
Blue 
Violet | 
Beyond the violet 
* Encyclopaedia Metropolitana, art. Light. 
f As in this table I use for the first time in this paper the expression “ extreme red,” it will he well to de- 
fine what is meant here and in the sequel by that term. The extreme red rays intended, then, are those which 
are brought into view by defending the eye while looking at the spectrum with one or more thicknesses of that 
sort of deep blue glass which is common enough in commerce, and which I am told owes its colour to smalt. 
