894 Mr. A, L. McAulay on the Recoil of 



into a clip in the front of the holder, about 2 mm. from the 

 source. Rays from the source traverse 4 cm. of a magnetic 

 field of several thousand gauss before reaching the ionization 

 chambers. j3 rays are thus almost completely removed. The 

 chambers are the same in shape and size, and roughly symme- 

 trically placed with regard to the source. Tbe case of the 

 electroscope and one of the ionization chambers are electrically 

 connected to earth, the electrode is charged to 60 volts, and 

 the other chamber, with a dummy electroscope attached to it, 

 is maintained at 120 volts. In this way the charge conveyed 

 to the electrode by the y-ray ionization is approximately zero, 

 but the somewhat thinner face of the earthed chamber causes 

 a slight excess of ionization on this side. This is compensated 

 by weak radiation from a tube containing a little radium D, 

 placed near an aluminium window in the side of the other 

 chamber. An exact balance can be obtained by adjusting 

 the position of the radium D tube. 



The electroscope is the result of a preliminary research, 

 and is of a new type. The principal objects aimed at were 

 high sensitivity and a very small volume. The design finally 

 adopted depends for its sensitivity on the fact that the leaf 

 when charged points upwards instead of downwards. An 

 increase in potential causes it to move towards the vertical, 

 and as it moves, the gravitational couple tending to restore it 

 to its original position decreases, instead of increasing as it 

 does in the case of the ordinary electroscope. It was found 

 that with the right dimensions it was easy to use the electro- 

 scope in this manner without the leaf collapsing or going to 

 the side. At the same time, if the inside of the case is not 

 suitably designed, these and other troubles do arise. The 

 sensitivity of such an electroscope may be made as high as 

 100 divisions on a microscope scale per volt, and its behaviour 

 appears satisfactory in this condition. In the present research 

 the sensitivity was about 30 divisions per volt. The leaf did 

 not creep and the sensitivity varied very little from one end 

 of the scale to the other. Greater sensitivity was not found 

 to be an advantage without greater refinement in the rest of 

 the apparatus. The construction of the electroscope is very 

 simple, and it will work satisfactorily when made quite 

 roughly. The case is 1*5 cm. in diameter and 0*6 cm. thick. 

 The electrode enters along a diameter from below, and the 

 leaf is attached to its top, which ends in a knife edge slightly 

 above the centre of the case. The leaf is 4 or 5 mm. long, 

 and when uncharged rests along the wedge-shaped side of 

 the knife edge. T\ T hen charged, it is attracted towards the 

 edge of a plate of thin sheet brass, cut in the arc of a circle, 



