224 



SCIENCE. 



THE POLARIZATION OF SOUND. 



AN EXAMINATION INTO THE NATURE OF VIBRA- 

 TIONS IN EXTENDED MEDIA.* 



By S. W. Robinson, 

 Professor of Physics and Mechanics, Ohio State University. 



The phenomena of polarization of light have hereto- 

 fore presupposed transversal vibrations of particles of 

 the luminiferous ether. Such vibrations have not only 

 been supposed transversal when polarized, but primarily, 

 or when in their primitive condition. It is proposed now 

 to show that no necessity exists for considering- the vi- 

 brations in light as transversal in front of a polarizer; 

 and furthermore to show that in all probability luminous 

 vibrations are primarily longitudinal. 



It is well known that light can be radiated, reflected, 

 refracted, diffracted, diffused, can be made to interfere 

 and can be polarized. All these effects are known to be 

 common to sound, except the last, and it is for the sole 

 purpose of explaining this in light that the convenient 

 theory of transversal vibration has been set up by physi- 

 cists, for the single case of luminous vibration. It is to 

 be noticed that transversal vibrations are not to be 

 assumed impossible when sufficient cause exists. It is 

 simply assumed that the cause is insufficient when a ma- 

 terial particle is made to vibrate from the action of a dis- 

 turbance at a remote single centre transmitted to the 

 particle considered ; the centre, the transmission, and the 

 particle considered, being supposed as belonging to a 

 homogeneous medium of indefinite extent. As regards 

 the nature of the vibratory mpvements of particles of 

 luminiferous elher may we not justly ask that, if we can 

 go through such a range of density as from platinum to 

 hydrogen without a change in the nature of the vibra- 

 tions where, as we rise in the scale of etheral tenuity, 

 shall longitudinal end and transversal begin ? Why 

 should the lunvniferous ether, now considered as a sub- 

 stance, have a peculiar form of vibration ? If ether un- 

 dula'ionscan be polarized, why not undulations generally? 

 These questions are not answered by the highest author- 

 ities. The short of it all seems to be that if polarized 

 light had never been discovered probably the dev ce of 

 transversal vibrations never would have been set up. 

 Indeed, the eminent author M. J. Jamin, says in his 

 three volume work on Physics at the outset, in his lesson 

 on polarization, and subsequent to the treatment of in- 

 terference, diffraction and other phenomena : " What has 

 been said previously of the movement of luminous waves 

 is absolutely independent of the directions of the vibra- 

 tion." This is good authority for limiting the transver- 

 sal theory to polarization, authority with which doubtless 

 all physicists will agree on this point. It is, therefore, 

 only necessary to polarize sound to place all the known 

 effects of luminous waves in common with sound waves, 

 or to make the theory of longitudinal vibrations univer- 

 sal. 



Assent to the above notions will be the more readily 

 given after noticing the consideration that, in polarized 

 light it is not necessary to suppose the vibrations trans- 

 versal till after passing the polarizer, and that the lat- 

 ter probably imparts an effect equivalent to a lateral 

 impulse, as due to its one-sided action upon the ray trans- 

 mitted, thus giving cause for vibration which are more or 

 less transversal ; the same being true for undulations in 

 air, water, iron, platinum, hydrogen, luminiferous ether, 

 etc., without exception. 



But independent of all questions of polarization we find 

 powerful evidence of the unity of system, for vibrations 

 in all possible media ; evidence which, in one case at 

 least, is employed as the basis of a rigorous mathemati- 

 cal demonstration of the impossibility of existence of per- 

 petual motion. This latter named evidence is the famous 



♦Copyright 1881, by S. W. Robinson. 



principle of Helmholtz regarding the action of natural 

 forces among mutually interacting material points, viz.: 

 that the forces must be central forces and functions of the 

 distance, and hence motions of remote particles can only 

 be longitudinal with reference to the centre of force. 

 This principle, considered aside from luminiferous ether, 

 will be universally accepted as truth. But what is the 

 criterion for making an exception of any homogenous 

 substantial medium ; even luminiferous ether? Indeed, 

 if any criterion exists for such an exception, it consists 

 simply in a desired convenient means for explaining 

 polarization : a theory of transverse vibrations, which, 

 though convenient and probable beyond the polarizer, has 

 thoughtlessly and without need, been extended to the 

 front of the polarizer, and to undulations in the primitive 

 conditions, where, as shown above, no necessity exists for 

 transversal vibrations. That the necessity for transversal 

 vibrations in primitive rays of light is entirely wanting, 

 let it be granted for the sake of an argument that the 

 source of light, such as the sun or a gas burner, is capa- 

 ble of exciting vibrations in the adjacent medium, which 

 are in all possible directions immediately at the radiant. 

 At a considerable distance from the radiant, the effect 

 upon a single particle will be the resultant of action of all 

 the particles immediately surrounding the radiant, and 

 transmitted by and through the intervening medium; 

 such a resultant impulse can hardly be admitted to be 

 otherwise than longitudinal. Or again, a particle near 

 the radiant imparts an impulse to the adjacent particle. 

 It the passive particle is in direct line with the active one, 

 the impulse received will be in direct line also, and the 

 direction of motion of the two particles coincides. The 

 similar direct action of a second particle upon a third 

 will also be in the same line, and so on indefinitely. At 

 a distance of several hundred or thousand times the 

 diameter of the radiant the line of vibration indicated will 

 be almost perfectly longitudinal irrespective of where the 

 first particle considered is situated about the radiant. 

 That is to say, a particle at the surface of a radiant 

 vibrating in a line of direction tangential to it or transver- 

 sal, may be considered as transmitting its vibration from 

 particle to particle in a direct line, and hence to the best 

 advantage ; and still, at a distance this direct line be- 

 comes nearly a line of longitudinal vibration for a remote 

 particle. 



Thus all lines of direction of vibration will pass through 

 the radiant or be tangent to it, so that in sunlight trie 

 rays, and the directions of vibration, will all lie within 

 the visual angle of the sun, or within about half a degree 

 of arc, and hence almost perfectly longitudinal. 



These considerations all confirm the principle of Helm- 

 holtz. And that remarkable principle, together with 

 all considerations presented above, and all which can 

 possibly result from a careful study of the subject of 

 transmitted impulses, go to confirm a universal law of 

 longitudinal vibrations for primitive rays in all possible 

 substantial media, and to antagonize the notion of trans- 

 versal vibration. 



Hence, if light can be polarized, why not undulations 

 generally. The writer, after much study of the subject, 

 became convinced of the possibility of this about eight 

 years ago, and six years ago apparatus was made for put- 

 ting the matter to an experimental test. Want of time pre- 

 vented, and further study determined a modification of 

 the apparatus which was made over two years ago. This 

 apparatus was successful in verifying all my preconceived 

 notions in the matter, but owing to extended study and 

 matured views of the principles involved, the experiments 

 simply confirmed, without developing new theories or 

 unanticipated facts. I propose now to describe the ap- 

 paratus and give the results. 



The means adopted for polarizit.g the undulations is 

 the same as that for polarizing light by reflection. It is 

 well known that when sound passes lrom one medium 

 into another whose velocity of sound differs, the sound is 



