354 REPORTS ON THE STATE OF SCIENCE, ETC. 
5. As an example of the method, a small disc of a transparent insulator called 
bakelite was placed in a straining-frame and compressed along a diameter. The 
disc carried a network of reference lines and the appearances were projected on a 
screen, upon which the isoclinics and isochromatics were traced with a pencil. 
Bakelite was selected because its stress-optical coefficient (recently determined 
as 50 brewsters by Mr. I. Arakawa in a M.Sc. dissertation of the University of London) 
is at least five times that of celluloid, and therefore, for a given thickness, a far larger 
number of isochromatic lines is brought into the field. The difficulty with this 
material is usually that it shows considerable optical effect in the unstrained state, 
but it was found possible, by a method due to Mr. H. T. Jessop, to obtain almost 
perfect annealing of this particular specimen. 
The disc compressed along a diameter was selected as a type of stress because its 
mathematical solution is simple and well known (See Love, Theory of Elasticity, 
3rd edition, Art. 155) and would afford a basis for comparison. It was found, however, 
in practice, that the compressing pieces flattened the disc and the pressure was spread 
out over a quite considerable arc. 
The straining-frame used did not allow of the total pressure being found directly. — 
The ratio of this to the stresses has been deduced from the observations themselves. 
—— Isochromatic Lines. 
- Isoclinic Lines. 
65° ~~~ Lunes of Principal Stress, 
order of Isochro- 
matics us shown By 
. Roman Numer 
5° 
70° 
5° 
t FoR 
(7994.2) IV IL I 
Fic. 4 
In like manner it will appear that neither the stress-optical coefficients of the bakelite, 
nor its elastic constants, needed to be determined; and these points are of some 
importance, as they show how, in many cases, the method is largely independent of 
subsidiary determinations. 
6. The isoclinics (drawn for every 5° of #) and the isochromatics, of orders 1 to 8, 
are shown in fig. 4, for one quadrant of the disc only, the others being deducible 
from symmetry. The isoclinics are shown in fig 4 by thin lines and the isochromatics 
by thick ones, the isochromatic of order zero being the boundary itself. The lines of 
principal stress, drawn from the data supplied by the isoclinics, are shown by the 
dotted lines, which should be coaxial circles according to theory. Those surrounding 
Pl 
the point Y show a certain amount of ellipticity which agrees with what one would 
expect if the pressure is spread out on either side of Y. 
The first step was to use the method of §4 to obtain P and Q along OX, P being 
here horizontal. The radius r of the circular image was 3}in., and measurements 
