144 
POPULAR SCIENCE REVIEW. 
one of mean refrangibility, the angle between and Bi' affords 
a measure of the prism’s power in causing deviation. On the 
other hand, the length of the spectrum VR affords a measure of 
the prism’s dispersive power. If dispersion bore a constant 
proportion to deviation, it is clear that we could not by any 
combination of prisms see a spectrum when looking straight 
towards a star, because the means used to make the deviation 
vanish would also do away with the dispersion. But as this 
proportion has been found not to hold, we are enabled to over- 
come the difficulty. It has been found that flint glass gives a 
spectrum of much greater length for a given deviation of mean 
rays than a prism of crown glass. If, then, we correct the 
deviation caused by a flint-glass prism by means of the deviation 
caused by one or more crown-glass prisms, there will still re- 
main an uncorrected dispersion ; in other words, a spectrum will 
still be visible. This is what is done in the compound prism of 
Amici (shown in fig. 2), in which a flint prism is placed between 
two crown-glass ones. Here the rays of mean refrangibility 
suffer no deviation, so that it is possible to look directly at a 
luminous object through the compound prism; but a spectrum 
is seen because the dispersion produced by the flint prism in 
one direction is greater than the dispersion produced by the two 
crown-glass prisms in the other. 
Fig. 3 represents another form of direct-vision prism, in 
which the light undergoes three internal reflections before 
emerging. This contrivance was devised by Mr. Alexander 
Herschel for the observation of meteors. The construction 
involves practical difficulties, as every face of the prism is 
brought into action, and great care is therefore required in the 
preparation of the prism. These difficulties were successfully 
overcome by Mr. Browning, who takes up eon amove all matters 
connected with spectroscopy. An instrument formed on this 
plan, or by the combination of two such prisms, as shown in 
fig. 5, is called a Herschel-Browning direct-vision spectroscope. 
It is worthy of notice tliat a simple prism, or better, any 
direct-vision prism, 'will serve to exhibit the dark lines in the 
spectra of the stars if only a Huyghenian eye-piece be made use 
of in examining the spectrum. Nothing, perhaps, is more re- 
markable than the circumstance, that whereas Lord Eosse’s 
tele.scope docs not suffice to give any indications whatever of the 
fact tliat the nearest fixed star lias appreciable dimensions, a 
hand-spectroscope, which can be obtained for fewer pence than 
the great reflector cost pounds, will exhibit those dark lines in 
the stellar spectra which enable us to determine the physical 
constitution of these distant orbs. 
In fig. 4 a mode of using a direct-vision prism in combination 
with a small achromatic telescope is shown. The instrument 
