93 
for the displacement in the secondary wave differing from 
that proposed by Professor Stokes. The form proposed by 
Professor Eowland I shall show has a resemblance to that 
which I now propose, but is open to the objection which I 
make against Stokes’ solution, that the coefficients in the 
solution do not tend to a zero limit in the case of a boundary 
at an infinite distance. As Professor Rowland’s result 
depends upon a solution of the same equations as my own 
result, I am able to compare the methods and point out the 
origin of our differences. 
In the paper on “ The Solution of the Equations of Vibra- 
tion of Light,” previously quoted, I have shown that if 
^ = Acos^(at - r) + Bsinp(ai( ~ r) 
be a component of a vibration in a spherical light wave, A 
and B consist of a series of terms of the orders — 1, — 2, etc., 
and I have given the law by which we may derive all terms 
of higher order from those of the order— 1 (say u_i); and 
that the terms U-i are quite arbitrary, subject to the one 
condition 
XM +y u + Zy., = 0 
i— 1 1 
which is the form which the equation of continuity takes. 
Now if we make the assumption that can be written 
in the form where Y stands for a spherical har- 
monic function, we may obtain Professor Rowland’s solution 
of the equations (page 417). 
Now in order to satisfy the equation of continuity. Pro- 
fessor Rowland gives to r\, Z (Fi, Gi, according to his 
notation, page 418) such a form that not merely is 
I -1 -1 I -1 
but that xi[, + y-q + z^ = 0. 
His next assumption (page 419) is that Y_(,^+l) shall be a 
zonal harmonic, limiting the case to that of symmetry 
about an axis. 
By these hypotheses, the solution is far removed from a 
