TBANSACTIOiN:S OF THIi aKCTlONS. 45 



of a few particles from tlieir extremities ; but even if the rod and wire were de- 

 stroyed altogether, the building would still be safe. 



On Compass Coi-nctlon in Iron Sh'j;>s. Btj Sir W. Thomson, D.C'.L., F.R.S. 



Effects of Sfre,-^s on the Macpietization of Iron. 

 Bff'fiiv W. Thomson, 'D.C,L.,F.E.S. 



On Contact Ehetridt)j. By Sir "W. Thomson, D.C.L., F.11.S. 



AcorsTics. 



On the Conditions of the Transformation of Pendalum-Vihrations ; with an 

 experimental illustration. By R. H. M. Bosanqtjbt, Fellow of St. John's 

 College, Oxford. 

 Under certain circumstances, a pendulum-vibration of given period can give rise 



to impulses which support vibrations whose periods are h h h ^f the period 



of the original vibration. The conditions under which this tates place are of 



interest. . • j- • 



Wheatstone enunciated the following as an experimental law :— A periodic im- 

 pulse can sustain vibrations whose I'req^uencies are multiples of that of the irnpulse. 



This was supported by an experiment in which the harmonics of a Jew's harp 

 are obtained from it by an adjustable resonator. But a general law cannot be 

 proved by a particular experiment. 



An experiment was adduced in contradiction of the generality of the above law. 

 It can be shown that the stopped pipes of the organ are incapable of supportingthe 

 vibration of resonators tuned to their octave and double octave, while open pipes 

 are capable of doing so. 



As the result of mechanical theory, the law may be enunciated that no pendulum- 

 vibration can be maintained in a vibrating system, unless the acting forces contain 

 impulses of the same period as the vibration maintained. 



The experiments commonly shown, in which a simple pendulum-vibration is 

 made to support its harmonics, generally depend on a transformation of the vibra- 

 tion in the transmission of the impulse. The apparatus exhibited forms a type of 

 the general process of transformation by transmission. 



A metronome vibrating seconds furnishes the fundamental vibration : a number 

 of small pendulums vibrate 2, 3, 4, 5, 6, 7, and 8 times in a second. By making 

 connexions between the metronome and the pendulums with elastic cord in diffe- 

 rent ways, the different kinds of transmission (with and without transformation) 

 can be illustrated. 



When the cord is tight, the impulses are transmitted without transformation ; 

 when the cord goes slack during the vibration, the impulses are transformed into a 

 series of pulls. In the first case the small pendulums are not affected ; in the second 

 they are generally set in vibration. 



The following points are illustrated by the experiment with the partly slack cord, 

 where the impulses constitute a series of pulls : — 



The common exposition of the theory of musical sounds, in which the impulses 

 are compared to the blows of a hammer, really makes a very complex effect the 

 basis of operations. The notes thus constructed differ from sirnple musical tones in 

 having the power of supporting the vibrations of their harmonics. 



T?he cases of the notes of the siren and the harmonium, in which the sound is 

 produced by a series of jets of air, are illustrated by the same experiment. 



An experiment of Prof. Mayer's, for the analysis of the sound of a reed pipe, by 



