OF THE SOLAR SPECTRUM ON VEGETABLE COLOURS. 
211 
can easily and completely absorb, or if the brush in applying- it be allowed to rest on, 
or be passed twice over any part. The effect then becomes that of a coarse and ill- 
printed wood-cut, all the strong- shades being run together, and a total absence pre- 
vailing of half lights. 
220. To prevent this confusion, gum-arabic may be added to the prussiated solu- 
tion, by which it is hindered from spreading- unmanageably within the pores of the 
paper, and the precipitated Prussian blue allowed time to agglomerate and fix itself 
on the fibres. By the use of this ingredient also, a much thinner and more equable 
film may be spread over the surface ; and when perfectly dry, if not sufficiently deve- 
loped, the application may be repeated. By operating thus I have occasionally 
(though rarely) succeeded in producing pictures of great beauty and richness of 
effect, which they retain (if not thrown into water) between the leaves of a portfolio, 
and have even a certain degree of fixity — fading in a strong light and recovering 
their tone in the dark. The manipulations of this process are, however, delicate, and 
complete success is comparatively rare. 
221. If sulphocyanate of potash be added to the ammonio-citrate or ammonio- 
tartrate of iron, the peculiar red colour which that test induces on persalts of the 
metal is not produced, but appears at once on adding a drop or two of dilute sulphuric 
or nitric acid. This circumstance, joined to the perfect neutrality of these salts, 
and their power, in such neutral solution, of enduring, undecomposed, a boiling heat, 
contrary to the usual habitudes of the peroxide of iron*, together with their singular 
transformation by the action of light to proto-salts, in apparent opposition to a very 
strong affinity, has, I confess, inclined me to speculate on the possibility of their ferru- 
ginous base existing in them, not in the ordinary form of peroxide, but in one isomeric 
with it. The non-formation of Prussian blue, when their solutions are mixed with 
prussiate of potash (Art. 209.), and the formation in its place of a deep violet-coloured 
liquid of singular instability under the action of light, seems to favour this idea. Nor 
is it altogether impossible that the peculiar “ prepared” state superficially assumed 
by iron under the influence of nitric acid, first noticed by Keir, and since made the 
subject of experiment by M. Schonbein and myself may depend on a change super- 
ficially operated on the iron itself into a new metallic body isomeric with iron, unox- 
idable by nitric acid, and which may be considered as the radical of that peroxide 
which exists in the salts in question, and possibly also of an isomeric protoxide. A 
combination of the common protoxide with the isomeric peroxide, rather than with 
the same metal in a simply higher stage of oxidation, would afford a not unplausible 
notion of the chemical nature of that peculiar intermediate oxide to which the name 
of “ Ferroso-ferric” has been given by Berzelius. If (to render my meaning more 
clear) we for a moment consent to designate such an isomeric form of iron by the 
name siderium, the oxide in question might be regarded as a sideriate of iron. Both 
* See my paper on this subject in Philosophical Transactions, cxi. p. 293. 
f See Annales de Chimie, tom. liv. p. 87. 
2 e 2 
