544 



SCIENCE. 



[N. S. Vol. III. No. 67. 



' Inferences from Maxwell's theory con- 

 cerning the motion of pure ether ' (Wis- 

 senschaftliche Abhandl. B. III., p. 526, 

 Wiedem. Am. Vol. LIII., p. 135-143). 



M. I. PUPIN. 



CoLUiMBiA Univbksity, April 2, 1896. 



A METHOD OF DETEB3IINING THE RELATIVE 

 TRANSPARENCY OF SUBSTANCES TO THE 



RONTGEN BAYS. 

 The fact that the Rontgen rays have 

 the power of dissipating the charge of a 

 perfectly insulated electrified body was es- 

 tablished by Professor J. J. Thompson,''^ 

 and furnishes us one of the simplest meth- 

 ods of detecting the i-aj's. This effect is the 

 basis of a very simple method of making 

 quantitative measurements of the intensity 

 of the radiation. If we take a condenser 

 and allow the Eontgen rays to fall upon it, 

 we shall find that there is a very considerable 

 diminution in its insulation resistance, and 

 that the charge of the condenser is gradu- 

 ally dissipated. This is illustrated by the 



curves A and B in the accompanying figure. 

 A was obtained under the ordinary condi- 

 tions. B was obtained when the Crookes 

 tube was in action, and placed about six 

 * London Electrician, February 7, 1896. 



inches from the wooden side of condenser. 

 The curve A was determined before, and 

 again immediately after the determination 

 of B. The two determinations of A were 

 identical, showing that the effect of the 

 Rontgen ray on the insulation disappeared 

 with the cessation of the ray. In making 

 these measurements a Nalder micro-farad 

 condenser was used, the condenser being 

 charged with a standard Clark cell. It is 

 evident, therefore, that it is possible to com- 

 pare the transparency of different sub- 

 stances by allowing the rays to pass through 

 screens made of the substances and placed 

 between a Crookes tube and the condenser 

 and measuring the resulting leakage of the 

 condenser. 



I am now engaged in making a series of 

 measurements, using this method and a 

 condenser especially constructed for the 

 purpose, and hope to give the results in a 

 subsequent number. 



It would seem that the method is capable 

 of giving results much more quantitative 

 in character than any that can be obtained 

 by. photographic methods. 



Wm. Lispenard Robb. 

 Trinity College, March 25, 1896. 



AN APPARATUS FOR THE STUDY OF SOUND 

 INTENSITIES. 

 The study of sound intensities presents 

 many difficulties to the physicist as well as 

 to the psj^chologist; the determination of 

 the equality of loudness of two sounds, as 

 well as of the law of relation between the 

 physical cause and the sensational result is 

 perhaps the most serious one. The facts 

 that sounds must be estimated succes- 

 sively and should be of a constant intensity 

 from beginning to end further complicate 

 the problem. The method of the falling 

 ball has been most frequently used; it con- 

 sists in dropping a ball successively from 

 two diflferent heights and recording tlie 

 minimum difference in height necessary to 



