122 Sir J. F. W. Herschel on the Action of the Rays 
therefore increasing the measures” (Phil. Mag., Dec. 1842, 
p- 456.), I cannot help observing, that not only is no real ad- 
vantage gained by such elongation, but the very reverse. For, 
in a spectrum formed on a surface so inclined, each of the co- 
loured solar images, of whose succession it consists, is not only 
defective in its individual definition everywhere but at two 
points in its circumference, but also, instead of being circular 
as it would be were the surface perpendicularly exposed, is 
dilated into an ellipse, having its longer axis in the direction 
of the length of the spectrum, and the overlapping of the con- 
tiguous ellipses being necessarily dilated in the same propor- 
tion, every abrupt change in the intensity of photographic 
action becomes softened and smoothed down as it were by 
being spread over more space, and rendered, by consequence, 
less salient to the eye than it otherwise would be. In the case 
of ioduret of silver, indeed, this is not quite of so much moment, 
because the change of intensity in the action of the spectrum 
is, as I have shown in Article 129 of the paper already cited, 
and in Articles 214, 215 of its continuation (Phil. Trans. 
1842), so excessively abrupt at the point of union of the blue 
and violet rays, that it is not in the power of such dilatation 
materially to mask it. But in innumerable other cases where 
consecutive maxima and minima occur, these features (which 
are always characteristic of the ingredients used, and on that 
account especially interesting and important) cannot fail to be 
grievously marred by thus as it were flattening them down. 
It is this consideration which decided me, from the very be- 
ginning of my inquiries on this subject, always to use an 
achromatic lens for forming my spectra (unless in cases where 
the use of two kinds of glass is objectionable); and when it 
has been required to increase the proportion of their length 
to their breadth for peculiar purposes (as in Article 69 of the 
above-cited paper), to do so, not by increasing the length, but 
by diminishing the breadth of the spectrum in the manner 
there described, so as, at the same time, to preserve the circu- 
‘larity of the sun’s image, and also to diminish the overlapping 
of successive images, and thereby increase the homogeneity of 
the light at each point of the length. And here I must take 
occasion also once more to insist (whatever may have been 
said to the contrary) on the indispensable necessity of using 
achromatic lenses for photographic practice with the camera 
obscura by those who desire perfection; being myself fully con- 
vinced that we have hitherto seen nothing comparable to what 
photography is capable of performing when the camera shall 
come to be studied and improved with a view to this especial 
purpose, as the telescope has been for its own. And I ear- 
