OF THE SOLAR SPECTRUM ON VEGETABLE COLOURS. 
203 
thermic spectrum so characterized in Art. 136. It also acquires a whitish narrow 
border, indicated by the dotted line, and very conspicuous on the green ground of 
the paper. The action continuing, the spot y is marked out by the extension of the 
border in that direction, soon after which the spot appears, in brown. Lastly ap- 
pears S with feeble traces of further irregular and interrupted action. Measure- 
ments of these spots as they appear, leave no doubt, of their identity in situation with 
the thermic spots a, (3, y, l of Art. 136, and that they are referable to the drying of 
the paper is shown by the fact, that a film of the liquid dried in a porcelain saucer 
changes from green to dark brown at a definite point of dryness. Moreover, on 
wetting the paper, the brown spots disappear, and in their place we find a train of 
Prussian blue, of varying intensity, but of uniform breadth (not swelling and con- 
tracting, as is the case with the heat-spots formed by simple drying, and therefore ob- 
viously due to direct radiation), and terminating in two insulated and tolerably well- 
defined circular spots or solar images, holding precisely the places of y and l (viz. at 
- 357 and — 45T). 
206. If in lieu of the perchloride of iron, we substitute a solution of that curious 
salt the ammonio- citrate of iron, the photographic effects are among the most various 
and remarkable that have yet offered themselves to our notice in this novel and fertile 
field of inquiry. The two solutions mix without causing any precipitate, and pro- 
duce a liquid of a brown colour, which washed over paper is green (being strongly 
dichromatic). If this be done under the prism, the action of the spectrum is almost 
instantaneous, and most intense. A copious and richly coloured deposit of Prussian 
blue is formed over the whole of the blue, violet, and extra-spectral rays in that direc- 
tion, extending downwards (with rapid graduation) almost to the yellow. If arrested 
when the blue is most intense and thrown into water, the impression is fixed, as in 
the accompanying specimen (see fig. 10.). But if the action of the light be continued, 
strange to say, the blue and violet rays begin to destroy their own work. A white 
oval makes its appearance in the most intense part of the blue (fig. 11.), which extends 
rapidly upwards and downwards. At a certain point of the action, the upper or more 
refrangible extremity of the white impression exhibits a semicircular termination, 
beyond which is a distinct and tolerably well-defined conjugate image, or insulated 
circular white spot, whose centre is situated far beyond the extreme visible violet. 
207. If paper washed over with the mixed solution in question is exposed wet to sun- 
shine, it darkens to a livid purple and rapidly whitens again. If the exposure be con- 
tinued, the white again darkens gradually to a brownish violet hue. But in the shade 
it slowly resumes its original tint, after which it is again and again susceptible of the 
same round of action. The most singular and apparently capricious varieties of co- 
loration and discoloration however arise (as is so frequently the case in photographic 
experiments) from different dosage of ingredients, order of washes, &c., so as to make 
the study of the phenomena in a high degree complicated*. A certain adjustment 
* The whitening is very obviously due to the deoxidation of the precipitated Prussian blue and the formation 
2 D 2 
