564 



SCIENCE 



[N. S. Vol. XXVII. No. 693 



of the curves were approximately ex- 

 ponential and had a period from 36 to 40 

 minutes. There thus seems to be a period 

 of decay of about 38 minutes. 



4. At least eight or ten of the curves 

 showed a slow rate of leak remaining for 

 many hours. In two or three cases this 

 rate of decay was identified with that of 

 thorium excited, having a period of decay 

 of about 11 hours. 



One may, I think, conclude from these 

 results that there is a simple exponential 

 decay with a period of 28 minutes, another 

 one of 38 minutes, and some evidence of a 

 much slower rate of decay with a period of 

 over 50 minutes. If the period is over 60 

 minutes one can be certain that thorium 

 excited is present. The period of 28 min- 

 utes agrees with that of radium. The re- 

 sults of this paper then show that in 

 general the decay curves of the radio- 

 activity of the atmosphere will be irregular 

 and made up of several rates of leak due 

 to radium-excited activity and in many 

 cases also to thorium excited. The effect 

 of smoke particles is undoubtedly the same 

 as that of snow and rain, as observed by 

 other experiments, though the rate of 

 decay is in general diiferent. 



071 the Charge carried by the Negative Ion 

 of an Ionized Gas: R. A. MniLiKAN and 

 L. Begeman, University of Chicago. 

 The paper is a discussion of the sources 

 of error in preceding determinations of e, 

 and a description of attempts to eliminate 

 some of these sources of error. In its 

 essentials the method employed was that 

 used by H. A. Wilson. The source of ioni- 

 zation was radium. The potential differ- 

 ence established between plates 5 mm. apart 

 in the cloud chamber ranged from 1,600 

 to 3,000 volts. The expansion was of such 

 value as to cause the fall in pressure to be 

 between 22 and 24 cm. of mercury, its 

 initial value being about 75 cm. The cloud 



was timed as it fell between the cross 

 hairs of a short focus micrometer telescope, 

 so set that the distance between these cross 

 hairs corresponded to an actual fall of 2 

 mm. The degree of accuracy with which 

 results could be duplicated was shown from 

 a typical set of observations. 



The results of ten different sets of ob- 

 servations made with fields of the indicated 

 strengths were as follows: 



Distriiution of Ionizing Energy aiout 



X-Bay Tubes: Wm. E. Ham, University 



of Chicago. 



The distribution of ionizing energy about 

 X-ray tubes was examined by an electro- 

 scope method in order to determine whether 

 the uniformity in distribution obtained by 

 previous observers by photographic and 

 fluoroscopic methods would be found to 

 hold. 



"Working at one and one half meters dis- 

 tance from the center of the tube there was 

 found to be a wide variation in ionizing 

 energy in certain planes. Thus: 



1. In the plane detemiined by the axis 

 of the cathode and the normal to the target 

 there are large variations, the energy along 

 the normal being about one half as great 

 at a point 60° from the normal, where it 

 is a maximum. 



2. In the plane at right angles to the 

 preceding plane and containing the normal 

 to the target there are also variations as 

 shown by a large number of curves. 



