Hacxett—The Secondary Radiation excited by y Rays. 211 
where p* is the fraction of a primary beam of @ rays, falling normally on a 
plate thick enough to absorb it, which is returned as secondary. It has been 
determined experimentally by McClelland, in the case of lead, to be 49 per cent. ; 
for other substances it diminishes with atomic weight, having a value of about 
24 per cent. for aluminium. 
It would be important to obtain measurements of the secondary radiation 
‘produced by a pure pencil of y rays on the side of the plate on which the rays 
are incident, using a magnetic field to remove the 8 rays which always accom- 
pany the y rays when such a precaution is not taken. It would not be necessary 
to make such measurements with great accuracy, for there is sufficient range in 
the value of p for different substances to enable us to decide if the relation 
indicated by formulas (C) and (D) holds good, and thus to discover to what 
extent the y rays have any directive action on the 8 rays produced by them. 
For the discussion of the experimental results in this paper, we use the ex- 
pression (A) or (C) for S,, the secondary radiation emerging from the plate on the 
same side as the y rays. It does not matter which formula is used, since the 
expressions in each case for the radiation emerging from the plate on the 
same side as the primary are very much alike and can be discussed together. 
The only difference is the factor 1 +./1—«, in the denominator of one. From 
Table II. it will be seen that it does not vary much from one element to another. 
The values of \, and x, have as yet only been determined for the 6 rays 
of radium and the secondary radiation produced by them. From some experi- 
ments at the end of this paper, it is seen that the penetrating power of the 
secondary rays produced by y rays is somewhat less than the penetrating 
power of the 8 rays from uranium. The ratios of the two coefficients of absorp- 
tion for tinfoil in the case of the secondary rays from aluminium is 1°83, and 
from lead varies from 2°34 to 1°41. As the uranium 8 rays are not so penetrating 
as the B rays from radium, it is likely that «,°is less for these secondary B rays 
than for the 6 rays of radium; but it probably varies in the same way with the 
atomic weight. As before, since the absorption coefficients of y rays and B rays in 
general are each roughly proportional to the density, their ratio is nearly con- 
stant from one element to another, and contributes nothing to the variation in 
the secondary radiation from one element to another. A glance at the values 
of 1 + VT Sie given in Table II. will show that it decreases slowly with 
increase of atomic weight. In this case also it may be assumed that it varies in 
a similar manner. We therefore conclude from both (A) and (C) that the great 
increase in the secondary radiation for elements of high atomic weight is because 
x, is very much greater for these elements. This coefficient has considerable 
* J. A. McClelland, ‘‘ Energy of Secondary Radiation.””—Trans. Roy. Dub. Soe. vol. ix., part i1., 1906. 
2L2 
