Mr Campbell, The (3 Rays from Potassium. 
215 
the potassium rays. I have no doubt that the extraordinarily 
small mean obtained for the blank experiments is partly due to 
chance, but there is an ample margin. 
§ 9. A few experiments were made with a less intense field 
between the plates of the grid. With a P.D. of 5600 volts the 
following results were obtained. 
9 
Potassium rays : decrease = 3 °/ 0 . 
Uranium rays (1) = ^ %• 
§ 10. At onty one stage in the research were any considerable 
experimental difficulties encountered. Just at the end of the 
observations anomalous results were obtained. The cause of the 
trouble was eventually found in brush discharges which had 
formed at points on the plates of the grid : they had punctured 
the aluminium window, so that the ionised air in the neighbourhood 
of the grid could enter the testing vessel and cause an increase in 
the current. The brushes caused great trouble until the baro- 
metric pressure — which had been extraordinarily low — rose, when 
they disappeared. The hole in the window was mended and the 
same results as before were obtained. 
§ 11. The figures given in § 7 would seem to prove decisively 
the similarity between the potassium and the uranium rays and to 
show that the former, like the latter, consist of charged particles. 
The greater proportional decrease observed with the potassium 
rays is to be expected if the conclusion of the former experiments 
is correct, that the rays from potassium are heterogeneous and 
that their velocity varies from a value approaching that of the 
extremely rapid rays of uranium down to a much smaller value. 
It may be noted that the observations on the deflection in smaller 
electrostatic fields (§ 9) is not quite in harmony with this result, 
for the ratio of the effect observed with potassium to that observed 
with uranium is greater for the larger than for the smaller fields. 
But since the observed decrease in the smaller field does not 
amount to more than twice the error of experiment the discrepancy 
is of no importance. 
There is a possible alternative to the conclusion that the rays of 
potassium are (3 rays : it may be that the charged rays deflected in 
these experiments are not emitted directly by the potassium, but 
are secondary rays emitted from the plates of the grid when struck 
by uncharged rays from potassium. But since this alternative is 
improbable on a prion grounds and explains the experiments no 
better than the simpler conclusion, I think it may be rejected 
safely until some positive evidence is offered in support of it. 
