390 Dr. W. Spottiswoode on a Mode of 



The illusion succeeds in the open as well as in-doors with 

 the sound of a loud-ticking clock, and with the human voice ; 

 but with shrill sounds it succeeds best, notably with the sharp 

 click of a metronome, and even with a metronome-bell. 



These results point to the explanation foreshadowed by 

 Lord Rayleigh, namely that the diffraction of sounds of medium 

 and great wave-length around the head, thus bringing the 

 lower and upper partial tones of the compound sound in un- 

 equal intensity to the two ears, plays a great part in our per- 

 ception of the direction of sounds. When the effects of dif- 

 fraction are such as to be relatively negligible, as for shrill 

 sounds (whose wave-length is small) , then Steinhauser's theory 

 of the relative intensities appears to hold good. Any one may 

 at once convince himself of the fact that diffraction may thus 

 produce a difference in the relative intensities with which the 

 partial tones of a complex sound reach the ears, by the simple 

 experiment of comparing the note of a musically-ticking clock 

 placed in front of the head with its note when placed behind. 

 They appear somewhat different, the difference being one of 

 timbre rather than of total loudness. 



Another experiment with the pseudophone which gives rise 

 to acoustical illusions, consists in setting one flap to catch sounds 

 from the front, while the other catches sounds from behind 

 or above the observer. Under these circumstances the sounds 

 seem, as the observer moves his head, to come sometimes from 

 the right, sometimes from the left, or sometimes from the 

 ground. 



Lastly, most of these experiments with the pseudophone can 

 be repeated simply by holding the hands in front of the ears 

 as flaps ; but here the illusion does not always succeed, as the 

 observer is conscious that his hands are reflecting to the ear 

 sounds from a certain direction, and so the judgment is sophis- 

 ticated. 



XLIV. A Mode of Exciting an Induction-coil. By William 

 Spottiswoode, M.A., LL.D., President of the Royal Society. 



To the Editors of the Philosophical Magazine and Journal. 

 Gentlemen, 



I HAVE lately tried a mode of exciting an induction-coil, 

 which I have not seen elsewhere described, and which 

 appears to promise valuable results. It consists in connecting 

 the primary circuit directly with a dynamo- or magneto-machine 

 giving alternate currents. In my own case I have used one 

 of M. de Meritens's excellent machines driven by a three-and- 

 a-half horse-power Otto silent gas-engine, The speed of the 



