4 Prof. A. Schuster on a Simple 
plate is not that which gives no light at the minimum, con- 
sider a certain retardation A,B, introduced into half the 
beam. The effect is the same as if the impulses, instead of 
starting from A,, A,,...As, were sent off simultaneously 
from 5,, B.,.. B;, where A,B,=A,B,=,,. .A, Beane 
If B,A, and B,A, are lines drawn 
parallel to the principal plane of the 
focussing-lens so that the optical dis- 
tances of B, and A, and of B, and A, to 
the focus are the same, the impulses 
starting from points between By and 
B, may interfere with those starting 
from points between A, and Ag; but 
the parts of the grating A,A, and 
A,A, are not effective as regards interference, the impulses 
from A,A, arriving too, soon and those from A,A, too late. 
If the useless portions are cut off by means of screens, the 
bands become black again, and we return to the best retarda- 
tion, but with a reduced aperture, and consequently a reduced 
resolving power. ‘The introduction of these screens does not 
alter the width of the bands; but these will be further 
apart than for the best retardation on account of the reduced 
cross-section of the effective portion of the beam. If the 
retarding plate is thicker than that which gives the most 
distinct bands, it may similarly be shown that the blackness 
of the darkest portions may be restored by screening off the 
central portions of the beam; but the bands will be closer 
together in this case. The point at which the bands disappear 
altogether is that at which each homogeneous component of 
the impulse gives a central diffraction-image equal in width 
to the distance between the bands. In each case the width 
of the bands is easily seen to be the same as that given by 
the interference of two points of light separated laterally in 
the beam by a distance equal to that of any two of the pulses 
which reach the focus simultaneously. 
Powell’s experiment in which bands are obtained by intro- 
ducing plates into part of the beam traversing a hollow prism 
filled with a refracting substance is only a modification of 
Talbot’s, and is explained in a similar manner. 
2. An objection might be raised to the above explanation 
in so far as it is not perhaps at once obvious how darkness 
may result by the interference of a succession. of impulses 
which are all in the same direction. The answer to the 
objection is, that if the impulses belonging to the first half 
of the beam are retarded in such a way as to fit in exactly 
half-way between those belonging to the second half, the 


