402 Professor G. H. Bryan [March 19, 



differences have, however, been noticed by most people to whom the 

 experiments have been shown. While it is easy for an inexperienced 

 person to produce the necessary differences of touch with a yineu- 

 matic player fitted with this controllino: device (the only thing 

 necessary being to shift the sliding weight, or other sub-permanent 

 control), I find it very difficult to obtain the same effects by str'king 

 the keys of an ordinary piano with my fingers. 



5. There appear to be two different ways of accenting particular 

 parts of chords by hand pressure applied to the control lever. With 

 a slight sub-permanent tension, treble notes are usually best accented 

 by depressing the lever before the note has reached the tracker 

 board, and subsequently allowing it to fly up smartly. With a heavy 

 sub-permanent tension, it is necessary to jerk the lever upwards from 

 below just after the note has reached the tracker board. For a bass 

 note with light tension, the lever is firmly pressed down after the 

 note has reached the tracker. AVith heavy tension the lever is 

 previously raised, and then allowed to drop down with the note. In 

 either case the action is supplemented by a corresponding action in 

 pedalhng. 



6. With a heavy sub-permanent tension and the lever supported 

 from below, it is possible to obtain very soft effects in which the 

 treble parts ring out clearly and are not drowned by the bass. With 

 a light sub-permanent tension and the lever pressed down the results 

 are more brilliant. I attribute these differences to the inertia of the 

 controlling weight, which has more influence in retarding the oscilla- 

 tions of the regulating bellows when acting at a long leverage than 

 when acting at a short one. 



7. There is a great satisfaction in being able to slam down a 

 vigorous chord, hand and foot working in unison. 



8. In the earlier experiments the connexion between the lever 

 and the bellows was made firstly with strings and tapes, and sub- 

 sequently with wires passing over pulleys. It was found, however, 

 that the stretching of these connexions greatly interfered with the 

 effects and led to the production of harsh results. That such a 

 cause as the stretching of a wire should have so marked an effect 

 clearly shows how small differences may have very great influence on 

 the playing of a pneumatically operated piano. 



9. Contrary to the usual belief, I obtain the best accentuation 

 effects in parts of chords when the notes reach the openings in the 

 tracker board at exactly the same instant. When they are cut 

 unevenly it is necessary to make allowances for the irregularities, and 

 sometimes this is difiicult or even ineffectual. 



10. With experience it is possible to learn the exact kind of 

 effort required to accentuate a note in any part of the scale, and thus 

 to obtain marked differences between the treble and bass parts of a 

 comparatively short chord. 



11. Where a note or chord is repeated a number of times in rapid 



