182 
Proceedings of the Royal Society 
° r -g, 2~, &c. ; but the most instructive method is to produce the 
phenomenon of diffraction, which is usually accompanied by that 
of interference. 
Diffraction is the name given to the lateral deviation of light in 
passing the edge of an obstacle, i.e., of an opaque body. Having 
adopted the undulatory theory, we are ready to admit that such a 
deviation ought to take place, both from our experience of similar 
effects in air and water, and from our general ideas of the structure 
and equilibrium of fluids, from which we conclude that no single 
particle of a fluid can be disturbed without disturbing those sur- 
rounding it on all sides, that is, propagating a disturbance in all 
directions. When light, proceeding from a luminous source of 
very small apparent diameter, passes the edge of a dark body and 
is received upon a screen, instead of a sudden transition from light 
to darkness at the line where the geometrical shadow commences, 
we observe a gradually diminishing illumination for some distance 
inside of that line, and outside of it we observe maxima and 
minima of illumination arranged in bands parallel to it, if it is a 
straight line. In order to effect the measurement of the length 
A, and understand the principle of the process, it is not necessary 
to follow the mathematical investigation of the position and in- 
tensity of these maxima and minima. That investigation is based 
upon the axiom that each point of a wave of light is a centre of 
force, the molecule there situated tending to propagate the energy 
with which it is animated in all directions around it, so that, if it 
were at any instant the only molecule agitated, it would imme- 
diately become the actual centre of a spherical wave. In the case 
of the uninterrupted propagation of a spherical wave, it is the 
envelope of all these elementary undulations to which is trans- 
mitted the vibratory movement of each molecule, and which, by 
reason of symmetry, is a spherical surface concentric with that 
which it succeeds. Diffraction takes place when part of the wave 
is intercepted by an obstacle, and the symmetry is destroyed 
which kept the surface of the wave concentric with its first posi- 
tion. The propagation of a spherical wave does not require that 
contiguous molecules be allowed free play. If we look at a 
luminous source through a fine grating, we see it in the same 
