a6s 
Mr Herschel on Achromatic Object-Glasses, 
artists do actually, previous to working their glasses, make some 
estimate of the ratio of the dispersive powers, on which to ground 
a calculation of the radii, sufficient for their own satisfaction, I 
may assume, for the present, that a knowledge of the dispersive, 
as well as refractive powers of the media, may be obtained, re- 
marking, only, that when an optician has the good fortune to 
meet with a parcel of glass from one melting-pot, ' sufficiently 
pure for his purposes, it is well worth his while to bestow the 
utmost pains on the accurate determination of this most import- 
ant point. This will require the sacrifice of no portion of his 
glass capable of being used for large lenses, as neither the re- 
fractive nor dispersive powers of specimens made at one casting, 
can be supposed liable to such variations as materially to affect 
his results. A fragment cut from the corner of one of his plates 
will suffice, if properly used, for all his wants. 
The imperfections to which refracting telescopes are chiefly 
liable, are well known to originate in two sources ; — -the want of 
proportionality in the dispersive actions of glasses of different 
kinds on the differently coloured rays, and the spherical figure 
of the lenses. The former of these imperfections is demonstrably 
insuperable in the ordinary case of a double object-glass, where 
only flint and crown glass are used. The best we can do is to 
work the lenses so as to produce the same compound focus, not 
for all the rays, for that is impossible, but for the two brightest 
and strongest colours in decided contrast with each other, 
* From some experiments on the colours developed by crystals 
in polarized light % I am induced to conclude, that the colours 
we ought to take pains to unite, in order to produce the whitest 
possible pencil, are the brighter red, bordering on orange, and 
that part of the spectrum where the blue is most vivid, and be- 
gins to pass into green. Supposing these rays perfectly united, 
all the rest will be nearly so, and the two extremities of the spec- 
trum will both deviate one way from the exact focus, while the 
intermediate portion will deviate the other, thus producing the 
phenomenon always observed in well adjusted achromatic tele- 
scopes when thrown out of focus, viz. a purple or lilac fringe 
surrounding the image of a white object, on one side of the fo- 
cus, and a green on the other. This is the criterion of a good 
* Phil. Trans. 18S0, i. p. 08, 
