112 Prof. Thomson, On Secondary Rant gen Radiation. 
instead of using graphite. A correction had to be applied for the 
secondary radiation due to the gold-leaf alone. 
The saturation current was measured in two cases. 
1. When the plates were separated by 1 mm. of air, at 
atmospheric pressure. There is but little diminution in the 
intensity of the rays in passing through a layer of this thickness 
so that the ionization will be proportional to the product of the 
intensity of the rays and their absorption by the air. 
2. When the plates were separated by 15 mms. of air at 
atmospheric pressure. The secondary rays are practically com- 
pletely absorbed by this thickness of air so that the ionization 
in this case measures the energy in the stream of secondary 
radiation coming from the metal. As the absorption of the 
secondary radiation by the metal is much greater than that of 
the primary the stream of secondary radiation is equal to I/A 
where I is the energy in the secondary radiation emitted by a 
layer of unit thickness, and A the fraction of energy absorbed 
when the secondary radiation passes through this layer. 
The connection between the secondary radiation and the 
atomic weight would probably be brought out more clearly by 
determinations of I, than by those of I/A. The advantage would 
however be neutralized to a considerable extent by the fact that 
the values of I/A can be determined with much greater accuracy 
than those of I. 
An inspection of the curves will show that there is a very 
close connection between the values of I/A and the atomic 
weight. In every case except nickel an increase in atomic weight 
is accompanied by an increase in the stream of radiant energ} r . 
From the curves we should conclude that the atomic weight of 
nickel is greater than that of cobalt. 
The slope of the curve is by no means uniform : at places 
it is very steep, indicating that a small increase in atomic weight 
is accompanied by a large increase in the secondary radiation. 
It will be noticed that the places where the jumps occur 
depend on the hardness of the rays, and that with hard rays the 
jumps occur at higher atomic weights than with soft rays. 
Thus with hard rays the first big jump occurs between titanium 
and chromium, while with soft rays it occurs between aluminium 
and calcium. 
The steepest parts of the curves show a tendencj^ to occur at 
the middle of the groups of the periodic series. 
The unit of secondary radiation is quite arbitrary and is 
different in the different curves. 
The measurements from which the curves were drawn were all 
made by Mr G. W. C. Kaye of Trinity College. I wish to thank 
him for the assistance he has given me. 
