180 Mr. R. H. M. Bosanquet on the 



A few words may be said about the improvements in the 

 arrangement of the bottle-notes * with which my investigations 

 on beats have been conducted. It was found that the clips 

 used to hold the india-rubber tubes, through which water was 

 admitted to the bottles, were not sufficiently tight, and the 

 arrangement of a small reservoir to each note was not conve- 

 nient. The corks were also frequently defective ; and it was 

 found difficult to procure them sound when of the proper size. 

 A number of india-rubber corks were obtained, two inches 

 in diameter ; and the bottles were fitted with these. A number 

 of glass stopcocks were also obtained, such as are used in ordi- 

 nary chemical apparatus. In this way the difficulty as to the 

 tightness of the water-tubes was completely overcome. 



An india-rubber tube was carried from the water-supply of 

 the laboratory, and slipped on to any of the glass-stopcock 

 tubes into which water was to be introduced. An aspirator 

 was similarly employed at first to remove water when required. 

 But it was ultimately found more convenient simply to detach 

 the end of the india-rubber tube from the water-supply, and let 

 the end hang down over the waste. When the india-rubber tube 

 is full of water, this constitutes a simple and efficient aspirator. 

 By these means any of the notes can be tuned with considerable 

 accuracy ; also the notes can be made to change slowly and 

 continuously at any desired speed, by running the water slowly 

 in or out. This is an important point in the demonstration 

 of the properties of difference-tones, as some, for instance 

 De Morgan, have maintained that the combination-tones only 

 exist at definite points. By this arrangement the change of 

 the difference-tones can be followed continuously throughout 

 their whole course. The ordinary difference-tone was long 

 ago demonstrated by Tyndall continuously, so that the point 

 is not new; but the present method of demonstration is very 

 convenient. 



The bottles do not give good notes much above the middle 

 of the treble staff (about af). For notes above this, so far as 

 I have examined them, I have used ordinary organ-pipes, both 

 stopped and open, as in this region little depends on the quality 

 of the notes employed. But the particular phenomena I have 

 hitherto worked at are most clear in the lower part of the 

 scale, in which the peculiar purity of the bottle-notes is avail- 

 able for the demonstrations. 



During the last severe winter the laboratory was almost 

 uninhabitable, and little was done. An arrangement for 

 heating by gas with ventilation proved a failure ; the moisture 



* Phil. Mag. 5th ser. vol. viii. p. 292. 



