296 Transactions of the Boyal Microscopical Society. 
Example 2. 
Spirit of turpentine : displacement =1-62 
Distance from top 5-00 
1-62 
3-38 
Index = =1-48 nearly (1 • 481). 
It is well known that various readings are obtained by the 
spectroscope for the same substance by different instruments, 
according to the nature of the glass of the prisms, and on that 
account every observation is now referred to the standard obtained 
by Angstrom by means of the diffraction lines of Nobert.* 
Thus supposing the interval between the lines B and G in the 
spectrum is divided into 1000 parts, the readings for E, the line 
for mean rays, are : 
Dispersion. 
Diffraction 
Grating. 
E. 451 
G. 1000 
434 
1000 
400 
lOuO 
624 
1000 
Crown glass. 
Flint glass. 
Disulphide carbon. 
2. Befractometer for Thin Plates and Lenses, &c. 
This instrument depends upon two principles then : 
I. The distance through which an image is displaced by 
refraction through a plate. 
II. The exquisite sensitiveness of contact-films forming the 
various orders of Newton's rings and the central black 
spot. 
I. It is well known that when the index of refraction is 1 -500 
or I, an image is formed by a plate having parallel sides at a dis- 
* The method here proposed is not subject to the well-known inconvenience 
of spectroscopic determinations, where every instrument gives different values, 
and eacho bservation has to be reduced to a common standard of the diffraction 
spectra. I find, according to my own calculation, taking the wave-length of 
line E, 5269 ' 0 tenth-metres, this corresponds to 48205 * 6 waves per inch ; but it 
is not nearly the mean length of wave, for the mean of 9 of Angstrom's principal 
lines is 
54736 • 9 tenth-metres. 
The brightness of the sodium lines, 
D2 43086 • 5 waves per inch, 
Dj 43130 '4 waves per inch, 
naturally accounts for this mean value of E not corresponding to the mean value 
of the rays for white light. 
