OF THE SOLAR SPECTRUM. 541 
by projection. If the distance of the white surface with respect 
to the lens is a little modified, we see the lines depicted with more 
or less distinctness. In order to see. the fine lines of the less 
refrangible parts of the luminous spectrum, it is necessary to 
have an aperture of only one-fourth of a millimetre in diameter 
at the utmost; but in order to study the great lines, and espe- 
cially the lines H and those beyond, the aperture must have a 
diameter of one to two millimetres. In this case the two lines 
H of the extreme violet are very distinct, and we see some lines 
beyond in the rays whose colour is of a weak gray-lavender 
tint. 
_ Moreover, and this remark is important, as the position of the 
lines depends solely on the position of the apertire and on that 
of the prism, the apparent movement of the sun does not make 
them change place, so that it is useless to have a heliostat for 
the purpose of keeping them in the same position; a simple 
light-director which is turned with the hand serves the same 
aeeDoses only the spectrum is more or less illuminated. 
Thus, from what we have just related, an idea of the forma- 
Reon of the solar spectrum and of the dark lines which separate 
the luminous bands may be conceived. We will not enlarge 
any more on this head, nor on the primitive colours of light, 
as this question does not belong to the subject of which we have 
- to treat. 
§ II. Of the Chemical Spectrum. 
If we introduce a pencil of solar rays into a dark room through 
an aperture made in a shutter, if we refract it through a prism 
of flint-glass, and project the spectrum produced upon different 
substances which undergo chemical modifications resulting from 
the action of the solar rays, the chemical rays which act on these 
substances are not situated, as we know, in the same parts of 
| the spectrum, so that the chemical spectrum has not the same 
| extent for each impressible substance. We shall see examples 
of this further on. There exist besides a great number of sub- 
_ stances which undergo chemical modifications under the influ- 
ence of the agent which accompanies light; but amongst the 
most impressible we must without doubt reckon most of the 
salts of silver. In a former memoir, I showed that, with regard 
to some of these salts, and probably with all, a very curious 
phzenomenon took place: in the solar spectrum there are rays 
