300 
crystals, and in this paper I shall confine myself to the 
former. 
We observe that a plane drawn through the direction 
of the ray or pencil and the axis of the crystal is called a 
principal plane. It is found that one of the two rays into 
which the incident ray is divided is refracted according to 
the ordinary law, and is therefore called the ordinary 
ray, the other according to a more complicated law, 
and is called the extraordinary ray, But we especially 
remark that the vibrations constituting the ordinary ray 
lie wholly perpendicular to the principal plane of the crystal, 
those of the extraordinary ray parallel to the princi- 
pal plane. Consequently both rays will consist of 
polarised light, according to the definition we have 
just given of polarisation, and this polarised light will arise 
from the vibrations of the incident ray being resolved 
perpendicular and parallel to the principal plane. 
Of all uniaxal crystals the best known is Iceland spar, 
which possesses the power of double refraction in an emi- 
nent degree. It occurs in the form of a rhombohedron, 
and the axis of the crystal corresponds with the shortest 
diagonal. 
Now suppose a plate of this crystal, cut perpendicular 
to its axis, is interposed between the polarising and 
analysing plates, when the first and second planes of 
incidence are perpendicular to each other. 
The colour of light depends on the length of the wave. 
The length of a wave of violet light is ‘o000167, and that 
of a wave of red light ‘oo00266 of an inch. The wave- 
lengths of the other colours lie between these. White 
light is a compound of all colours, and therefore its vibra- 
tions are of all lengths lying between these extremes. It 
will therefore manifestly simplify the explanation we are 
going to submit to the reader, if we commence by assuming 
the light to be monochromatic, and its vibrations of the 
same length. Let a small pencil of monochromatic light, 
P (Fig. 1), be reflected from a plate of glass, A, at an angle 
of about 56°, pass through the crystal C, and then be 
reflected a second time from the plate B at an angle of about 
56° in the direction BE. The plane PAB is supposed to 
be perpendicular to the plane of the paper, the plane A B E 
to coincide with it. The crystal is supposed to be bounded 
by parallel surfaces perpendicular to its axis, and in a 
position perpendicular to the line AB. Let ad, cd, be 
two straight lines drawn in the crystal, one in the plane of 
the paper and the other perpendicular to it. The eye is 
supposed to be placed at E. 
The vibrations of light proceeding from A will all be 
parallel to the plane of the paper. In general, each ray as 
it enters the crystal will be resolved into two, and thus 
will arise two sets of waves which will interfere with each 
other. Hence a succession of dark and bright curves will 
be visible to an eye placed at E. Moreover, as the crystal 
is symmetrical round A B, it may be expected that these 
curves will be circular. It may also be expected that the 
diameters of the bright circles will depend on the length 
of the wave, and therefore on the colour of the light. 
We have already defined the principal plane of the 
crystal for a given ray, to be the plane passing through the 
ray and through the axis of the crystal, which in this case 
coincides with AB. As therefore the vibrations of the rays 
incident on the crystal are all parallel to the plane of the 
paper, those which enter the crystal along ¢ d@ are all 
perpendicular to a principal plane of the crystal, and those 
which enter it along @ @ are all parallel to a principal plane. 
Therefore every ray which enters the crystal along ¢é and 
c d, will be transmitted to the second plate unresolved, and 
unchanged, and will therefore be incapable of being 
reflected to an eye situate at E. Consequently a dark 
cross will appear to the eye at E, corresponding to the 
lines a4, cd, intersecting the system of rings we have just 
described. Now suppose white light to be substituted for 
monochromatic light. White light is composed of light of 
all colours, and we have seen that the diameters of the 
| NATURE 
[ Aug. 11, 1870 
bright rings are different for each colour. Consequently, 
instead of dark and bright rings, we shall have a series of 
coloured rings intersected by a dark cross. This pheno- 
menon is shown in Fig, 2. 
If the second plate be turned round an axis in the 
direction A B, till the first and second planes of reflection 
coincide, the reader who has followed this investigation 
will perceive that the dark cross intersecting the rings will 
be changed into awhite cross. This bright cross is shown 
in Fig. 3. 
If the second plate is turned round the same axis into 
any position intermediate to those we have just described, 
the rings will be intersected by two crosses, inclined to 
each other at an angle equal to that between the planes 
of first and second reflection. There will be coloured ares 
between the crosses, but those distances which give a 
maximum light for the arcs outside the crosses will give a 
minimum light for the ares inside the crosses. This 
phenomenon is shown in Fig. 4. These are the appear- 
ances observed when the surfaces of the crystal plate are 
perpendicular to the axis of the crystal, and to the axis of 
the incident pencil. Dr. Ohm has investigated the phe- 
nomena which occur when the axis of the crystal is inclined 
at any angle to its surface, and the crystal itself is placed 
in any position between the polarising and analysing 
plates. His mathematical investigations will be found in 
the seventh volume of the “ Munich Transactions,” and 
the following are the principal results which he has 
obtained. 
We shall suppose the light monochromatic. There will 
be a succession of dark and bright curves which are either 
parabolas, ellipses, or hyperbolas. For a given position 
of the crystal, the ellipses and hyperbolas are concentric ; 
but the centres are never in the centre of vision, unless the 
axis of the crystal is either perpendicular or parallel to its 
surfaces. When the bright curves are parabolas, their 
vertices are equidistant from one another. When the 
bright curves are ellipses or hyperbolas, the difference 
of the squares of the semiaxes of two consecutive 
curves is a constant quantity. We have seen that when 
the axis of the crystal is perpendicular to its faces, the 
bright curves are circles. As we cut the plates so that the 
axis may be inclined at angles more and more acute to 
the surfaces, the bright circles become ellipses, which 
elongate continually till they become parabolas. 
Suppose two plates of crystal, with parallel surfaces of 
equal thickness, and with their axes inclined at the same 
angle to the parallel surfaces, be placed in contact, so that 
their axes may lie in the same plane, but not in the same 
straight line, and then introduced between the polarising 
and analysing plates, a succession of dark and bright 
ellipses and hyperbolas, with their centres in the centre 
of vision, will be seen. It was this experiment which led 
Dr. Ohm to the results we have endeavoured to lay before 
the reader. W. H. L. RUSSELL 
NOTES 
TWELVE months ago the Erdington Orphanage, founded, 
built, and endowed solely by Mr. Josiah Mason, at a cost of 
nearly 250,000/., was opened at Birmingham. Mr. Mason, 
exactly following the example of Mr. Peabody, has now in con- 
templation another public work of even greater ultimate impor- 
tance, namely, a college and schools for scientific and technical 
instruction, open to all classes, and if the hopes of the founder 
should be realised, capable of expansion into one of the noblest 
institutions in the kingdom. As yet the plan is only broadly 
formed, and some time must elapse before it can be carried into 
effect ; but the Birmingham Daily Post states that a beginning 
has been made, and that for the purpose above mentioned, Mr. 
Mason has agreed to buy a large block of land in Edmund 
Street, exactly facing Ratcliff Place, between the Town Hall and 
