12 Chant — Experimental Investigation into the 



sented to the dielectric and the effect is greater. In some of 

 his detectors Rutherford used iron wire 0*07 mm in diameter, 

 (i. e., only one-half the diameter of mine), with 160 turns of 

 fine wire on the helix, and much longer wings. Thus he was 

 able to obtain noticeable effects at a distance of over half a 

 mile from the oscillator, which consisted of large plates. My 

 detector, however, was sensitive enough for my purpose. 



In the report* on Hertzian Oscillations presented by B-ighi 

 to the International Congress of Physicists in Paris in 1900, he 

 enumerated 21 kinds of apparatus for indicating the presence of 

 electric oscillations. Of these the magnetic detector is one of 

 the easiest to manage and possibly the simplest for quantitative 

 comparisons, though I know of few investigations in which it 

 has been used.f Its sensitiveness increases as the wings are 

 lengthened, and the greatest difficulty I experienced arose 

 from the fact that it is especially sensitive to the surgings in 

 the connecting wires of the oscillator. An illustration may be 

 interesting. 



A wire three meters long was stretched from one terminal 

 of the secondary of an induction coil across to the side of the 

 room, and sparks were made to pass between the ordinary 

 pointed terminals of the coil. The coil was in the basement 

 of the laboratory. A fully magnetized detector, with wings 

 each about 90 cms long arranged parallel to the wire running out 

 from the coil, was placed in position by the magnetometer, 

 giving the usual deflection of 40 cms . The magnetometer was 

 in a room on the next floor above, and was distant about 35 

 meters, with brick walls, numberless wires and other pieces of 

 metal between ; and yet, as soon as sparks began to pass at the 

 coil the magnetometer deflection was reduced by 2 cms . 



The detector was also used to determine the wave-length of 

 the oscillators used. The wave-length for the cylinders was 

 found to be approximately 80 cms ; that for the 10 cm spheres, 

 approximately 38'5 cms , i. e. 3*85 times the diameter. An 

 account of these experiments will be published later. Using 

 these results the frequencies given in Table I were obtained. 



4. Experiments on the u skin" -effect. 



During the study of the "skin "-effect cylindrical oscillators 

 with the dimensions given above were made as follows : — 

 No. 1. Solid brass. 

 No. 2. Solid Norway iron. 



* Rapports preserves au congres international de Physique reuni a Paris en 

 1900. Tome 2, p. 301 (Paris, 1900). 



f In the June (1901) Phil. Mag. is a paper by C. G. Barkla. in which are 

 described experiments on the relative velocities of electromagnetic disturbances 

 along wires of different diameters, in which the magnetic detector was used as 

 an indicator. 



