220 Mr. R. H. M. Bosanquet 07i the Theory of Sound. 



ments were all made with the tube employed in the second of 

 the above determinations, to which a fljit movable plug was 

 fitted. I had some difficulty in getting the plug to fit per- 

 fectly ; and, whether'from this or some other reason, these de- 

 terminations do not agree so perfectly as those given above. 

 The method was to shift the plug till the note agreed with one 

 of the tonometer-forks : — 



=-. Description. Correction. 



19 r Mean from determinations ofl .porT? 

 '*" \ four adjacent forks about 6'^^ J 



15 Mean of three forks about a' -554 R 



17 / -SUSR 



25 c'i •503R 



20 c' -43^ 



That there are some inequalities and discrepancies here is 

 evident ; but they are less than those of Wertheim, and dis- 

 tinctly follow the law above announced. 



Less weight attaches to these observations than to the two 

 with both ends open ; it is not practicable to ascertain the 

 effective dimensions of a very short pipe closed by a movable 

 stopper with any thing like the accuracy with which this can 

 be done for a cylinder open at both ends. These results, how- 

 ever, distinctly negative the position which follows from 

 Wertheim's numbers, that the correction for stopped pipes is 

 greater than that for open ones. They do not furnish mate- 

 rial for any distinction between the two. 



The present experiments are those which were made with 

 all care; they confirm the general conclusions T had arrived 

 at previously by a considerable number of experiments of a 

 preliminary character. I have not at present attempted an 

 extensive series of experiments, as I cannot expect final results 

 until I verify a tonometer myself and then use it directly. 



The c^# fork used in the last experiment but one has been 

 verified repeatedly ; and I consider it the best single fork I 

 have. T am confident that the error of that determination 

 does not amount to 1 vibration out of 276, which would cor- 

 respond to about '04 R. 



I must further quote one of my older experiments on a 

 longer tube open at both ends. It is a piece of gas-pipe 33'2 

 in. long and exactly 1 in. in diameter. It gives a very clear 

 note. I copied its octave (at 55° F.) on a fork ; I then took 

 up the fork and determined it at the tonometer at South Ken- 

 sington. It was exactly 396; this gives the correction for one 

 end =*436R. Calculating the vibration-number from cor- 



