Heating Effect of Had "in m and its Emanation. 325 



the centre o£ one of the mercury columns and the steady 

 deflexion of the galvanometer determined. The thickness of 

 mercury traversed by the rays was 4*4 millimetres. On 

 account of the large quantity of mercury and the distance 

 between the platinum coils and the outside cylinder, there 

 was a marked lag between the deflexion and tlie heating 

 effect. For example, the deflexion reached half its maximum 

 value in ten minutes. 



x Iff. O. 



It is difficult to determine the heating effect with the same 

 accuracy as for smaller coils ; but a number of fairly con- 

 cordant experiments gave a heating effect of 1*034. Similar 

 experiments were made with a thickness of mercury of 

 1*46 cm. The deflexion in this case reached half its maximum 

 value in 23 minutes. On account of the large lag, the small 

 variations in the balance point during the time of observation 

 became more important. Suitable corrections were made 

 for the lag of the apparatus, and also for the effect of the 

 decay of the emanation. As a mean of several observations 

 the heating effect was found to be about 1*05. It is, how- 

 ever, difficult to fix the value of the heating effect in this 

 case closer than half of one per cent. 



The heating effect due to the y rays from radium could 

 be determined with greater accuracy with a preparation of 

 radium in equilibrium, for under such conditions there is no 



