obtained from a Monochord Sound-box and String. 153 



Tests as to Avoidance of Spurious Effects. 



To test whether the floor was sufficiently rigid and to find 

 the amplitude and period of its vibration, i£ any, a plate was 

 exposed, the operator meanwhile jumping upon the floor. 

 This gave a simple harmonic curve of very small amplitude 

 and of about double the period of the slowest vibrations of 

 the string. Hence any disturbance due to motion of the 

 experimenters would, if present, cause a slight displacement 

 of successive waves. But as this never appears in the photo- 

 graphs, it may be assumed absent throughout. It need 

 scarcely be added that the experimenters, the one bowing the 

 string and the other operating the plate, each assumed a set 

 position which was carefully preserved till the plate had been 

 exposed. Further, on this head, it made no essential 

 difference to the curves obtained, whether the stand carrying 

 the two legs of the optical lever was on the monochord table 

 or on the lantern table. 



A second possible source of disturbance was a general 

 tremor of the tables bodily, which might be expected to give 

 a motion of a spot of light reflected from a mirror fixed on 

 the monochord. To test for this, the optical lever was 

 inverted and fixed on the monochord belly so that it now 

 ceased to be a lever. On bowing the string and shooting a 

 plate, the spot reflected from the mirror gave a line practically 

 straight. 



Thirdly, the suspicion arose as to whether the leg of the 

 optical lever might quit contact with the belly at the upper 

 limit of the motion or might press in to the metal pin to various 

 extents. To test this, the contact between the leg and the 

 metal pin was included in an electric circuit with a cell and 

 telephone. The telephone was taken to a distance of about 

 70 feet in an adjoining room, where all sound produced by 

 vigorously bowing the string was inaudible. With a very 

 weak constraint on the optical lever, the telephone gave 

 scraping sounds corresponding to the making and breaking 

 of contact. But with the constraint employed in the experi- 

 ments no sound was heard from the telephone. 



Apart, however, from these direct tests, we have indirect 

 and confirmatory evidence of an equally convincing nature 

 thnt the curves obtained correspond to actual motions of the 

 belly, and are not illusory. Thus the immediate repetition of 

 an experiment under like conditions gave like results. 

 (Compare figures 4 and 5, also 6 and 7.) Again, the variation 

 of pitch of the string always gave a change in the character 

 of the curve obtained for the motion of the belly (see 

 column 2). And this change could be obtained in its main 



