
Secondary Réntgen Radiation. 545 
plate suffered absorption which would be produced by going 
normally through a plate of thickness ¢(1:106), ¢ being the 
actual thickness of the plate. 
The average effective thickness of the plate was-not more 
than 1:°03¢. To compare the absorbability of the secondary 
rays with that of the primary, the ratio of the deflexions of 
the two electroscopes, consequent upon the ionizations pro- 
duced in these electroscopes by the primary and secondary 
radiations, was first obtained. From the observation of the 
defiexion of one of these, the deflexion of the other could then 
be calculated to within about 3 per cent. 
Plates of aluminium of equal thickness were then placed 
before the electroscopes, so that both the primary and secon- 
dary beams suffered absorption by transmission through 
aluminium. ‘The ratio between the two deflexions consequent. 
upon the passage of the primary and secondary beams through 
the electroscopes was again determined. In measuring the 
deflexions, corrections were always made for the result of the 
normal ionization taking place in the electroscopes. Many 
experiments were made with rays differing in penetrative 
power. In some of these the possible error was very small, 
and they showed that within 4 or 5 per cent. the absorption 
coefficients for the primary and secondary radiations were the 
same. Some of the results are given in Table I. (p. 546). 
In the second, third, and fourth sets of experiments given 
in the table, the order of magnitude of the absorption may 
be seen from a comparison of the duration of experiment and 
deflexion of the primary electroscope in each case. 
A sheet of aluminium of ‘04 cm. thickness reduced the 
rate of ionization in the electroscope by about 60 per cent. 
In the first set the intensity of radiation was altered in the 
different experiments, so that the duration of experiments 
was varied little. The ionization in an electroscope was 
diminished by about 40 per cent. by a sheet of aluminium of 
‘02 em. thickness. 
From the results given in the fifth column, it will be seen 
that in each case the rate of ionization in the secondary 
electroscope was diminished by the aluminium plate slightly 
more than that in the primary electroscope. But the effective 
thickness of the plate before the secondary electroscope was 
probably between 2 and 3 per cent. greater than that before 
the primary, so that the actual difference in the absorptions 
must have been exceedingly small if any difference existed. 
Corrections have been made for the normal ionization in 
each electroscope. 
Phil. Mag. 8. 6. Vol. 7. No. 41. May 1904. ie 
