« Particles with Light Atoms. 567 



for simplicity that the number of H atoms are counted as 

 they emerge from the plates A and B. Let H be the steady 

 magnetic field and X the electric field applied between A 

 and B. Then it is clearly seen that, if u be the velocity of 

 the H atom, 



for assisting fields, H<? it -f X<? = H 2 eu, 



for opposing fields, He u — Xe = Hx^w. 



Subtracting 2eX= (H 2 — R^eu, 



i 2X 



and n= — =y-, 



±i 2 — i±i 



so that the velocity of the H atoms can be determined 

 directly. In practice, it was found that the curve PQR 

 over the experimental range was nearly a straight line. The 

 initial field H was varied in different experiments, but was 

 usually about 4000 gauss. The steady voltage employed 

 was about +1500 volts. The ratio of the number of scintil- 

 lations observed on reversal of the electric field varied in 

 different experiments according to the magnetic field H and 

 the voltage, but lay between 1*8 and 3 in the various expe- 

 riments. Each experiment was complete in itself, for not 

 only were the scintillations counted on reversal of the electric 

 field, but also the number for two magnetic fields on either 

 side of the fixed field which give nearly the same ratio of 

 number of scintillations as that obtained by reversal of the 

 electric field. 



The paraffin film had a stopping power of 3*2 cm. of air, 

 and the stopping power of the silver plate together with the 

 aluminium screens was 11*4 cm. The range of H atom, 

 which passed between the parallel plates and produced scin- 

 tillations on the screen, thus lav between 28 — 3 - 2 = 24'8 cm. 

 and 28-14-6 = 13-4 cm. 



The corresponding velocities are *96i/ and '78u where w 

 is the maximum velocity of an H atom due to an a. particle 

 from radium C. As in the experiments on the magnetic 

 deflexion the average velocity was found to be *87w , In 

 calculating the relative effect due to a magnetic and electric 

 field, a small correction is necessary to allow for the fact that 

 the electric field was only effective the length of the plates, 

 while the magnetic field acted on the H rays from the slit A 

 to the zinc sulphide screen. Making this correction, estimated 

 to be about 12 per cent., the deflexion due to 1000 volts 

 between the plates was found to correspond in five different 

 experiments to 235, 227, 260, 235, 221 gauss respectively, 

 w 7 ith an average value of 238 gauss. 



