212 Profs. E. Rutherford and H. T. Barnes on the 
These results, as far as they go, are in agreement with the 
view that the heating effect of radium is, at any time, pro- 
portional to its activity measured by the a rays; or, in other 
words, that the heat emission of radium is an accompaniment 
of the expulsion of @ particles. The heat emission is not pro- 
portional to the @ or y rays; for in the experiment the amount 
of 8 and y rays remaining in the radium, after partial de- 
emanation, was only 6 per cent. of the total, while the heating 
effect was about 30 per cent. of the total. 
These experiments are not, however, sufficient to justify us 
in concluding that in all cases the heat emission accompanies 
the expulsion of a particles and is proportional to the number 
expelled. It will be necessary, in addition, to show that the 
emanation and emanation X each supplies an amount of heat 
proportional to its activity, measured by the @ rays, and also 
that the heating effect of each of the succession of changes 
of emanation X is always proportional to its activity, 
measured by the @ rays. 
Decrease of Heating Hffect of Radium immediately after 
De-emanation. 
The air-calorimeter, already described, is not suitable for 
rapid observations of the variation of the heat emission of the 
de-emanated radium or the emanation immediately after the 
latter has been removed. For this purpose the differential 
platinum thermometers, previously described, were used. 
The emanation was condensed in a small glass tube 3 cms. 
long, 8 mms. internal diameter. As the object of these expe- 
riments was to determine the initial drop of the heating effect 
of the radium, onlv a few minutes were occupied in heating 
the radium and condensing the emanation. In consequence of 
this, a small amount of the emanation was carried over into 
the pump and was not condensed in the tube. The radium 
tube was rapidly removed and placed inside the air-bath, and 
observations made on its heating effect, as soon as its tempe- 
rature was steady. The results are shown in fig. 5, curve B. 
The heating effect of the radium about 10 minutes after 
removal of the emanation had fallen to about 45 per cent. of 
the original value. Itthen decayed more slowly toa minimum 
corresponding to about 25 per cent. of the original value. 
The gradual decay of the heating effect to a minimum is 
connected with the removal of the emanation and consequent 
decay of activity of the emanation X left behind. The curve 
of decrease of the heating effect to a constant minimum should 
thus be the same as the curve of decrease to zero obtained 

