‘this is effected by making the reflected i image of a 
’ with the unreflected image of another distant object, and re 
110 Measuring refraction under the Compound Microscope. 
before the object-glass, and the telescope and prism. turned and — 
adjusted till the minimum deviation is reached when the angle 
or any particular colored ray is read off, and the refraction onlong 
lated by the formula, 
_sin 3 (a+d) 
singa 
2 
having previously ascertained the value of a by the method al- 
luded to in the first part of this article. The dispersion is meas- 
ured at the same time, more particularly with the aid of the di- 
vided sli 
Such an n arrangement costs. but little, is convenient, and its ac- i 
ioe can be depended on as far as three places of decimals. 
schel in his treatise on light proposes a certain method of 
ssoatabiin the polarizing angle, of opaque substances having at 
least one polished face, of set gems and of substances too s 
to be formed into prisms ; and thence, their index of refraction. 
This method was found in practice to be very uncertain and diffi- 
cult: the following modification gave better results. A little 
piece of apparatus is to be constructed, consist- Py 
ing of a mirror, M, set as seen in fig. 2, and ca- 
pable of being ‘firmly attached to the plate fags. 
in the end of the compound body : it is to 
justed so that the axis of the 
shall lie in its plane. The substance under ex- 
amination, 8, is attached by wax in the position 
seen in the figure, and its polished face brought 
into a position parallel to the face of the mifror : 
NE 8 8 EE NR reas 
qs 
Re ae et NF RS 
distant object as seen by the mirror, coincidé 
bringing the face of the substancé under examination into sucha 
position that a similar coincidence is observed. A candle is then 
placed on a level with the mirror, at about three feet distance, 
ny . . . . . m 
the flame to its center. This is effected by placing the eye be 
hind the flame and turning the mirror on its axis till the retiecth a 
image is seen to coincide with the flame: the error in Be mod 
of adjustment is far within the necessary error of observation. A 
Nicol’s prism is then Aigo between the candle snd cis instru> 
a ; 
ee on of ie Nicol’s prism in 
made pi the ee wi till its polis ed fate ceases to re- 
flect the polarized beam, when the anglei nat es Any deviation & 
in the apparent direction of the flame caused by the Nicol’s prism — 
can be“ascertained beforehand with accu: nd applied as con- 
+ ge 
gs Gs 
