392 



Prof. F. J. Cole. 



[Jan. 17, 



an essential factor in the success of the test, we should still leave unexplained 

 the differential response. It is, however, easy to show that the box itself, 

 owing to its small size and heavy construction, is ill adapted for the 

 transmission of vibrations, and is certainly the last means one would adopt to 

 transmit aerial vibrations freely to the animal. 



Tested with the trombone under the conditions of many of the following 

 experiments, both by means of the optical lever and by polarised light, two 

 of the live boxes showed no traces of strain in the latter case, and in the 

 former only an extremely slight vibration to the G, F sharp, and F below 

 middle C. 



The animal having been placed in the live box, and the current of water 

 started, the whole was then transferred to the stage of the microscope, and 

 the following three tests applied : — 



(1) On a specially prepared physical table, constructed of alternate layers 

 of sawdust and three massive cast iron slabs,* which served to filter off 

 vibrations from the ground, and to ensure that any response observed was 

 due to a stimulus conveyed through the atmosphere, i.e. was a true auditory 

 and not a tactual reaction. Judged by the mercury test this table was quite 

 efficient. 



(2) On an elongated rectangular resonator provided with a sliding piston. 

 Here the rationale of the piston is to make it possible to eliminate the factor 

 ■of the specific note of the resonator itself (a point of obvious importance), 

 and I may as well state at once that the position of the piston never 

 appeared to affect the results in the least. For example, a female tested on 

 the resonator responded equally well to middle B fiat whilst the length of the 

 resonator was being reduced from 135 to 30 cm. I should also observe here 

 that the bell of the trombone was usually placed near the open end of the 

 resonator, but not, of course, touching it. 



(3) On an organ reed, made to my instructions, which gave at least two 

 good octaves, one above and the other below middle C. This was effected by 

 means of a sliding piston and an adjustable lip, and was much more 

 ^satisfactory than tlio makers anticipated that it would be. 



In the cases of 1 and 2, a B fiat or tenor trombone was employed to 

 provide the stimulus, and it was generally found necessary to produce a ff 

 jiote. I have observed, for example, that' a crescendo note is only responded 

 to when the louder portion is reached. 



Tlie intervals mentioned in this paper arc those of the harmonic chromatic 

 scale of C major. 



The responses which the animal makes to sound are of two kinds, and 

 * My colleague, Prof. O. J. Burch, F.R.S., kindly suggested this to me. 



