OF ARTS AND SCIENCES. 329 



« = 4. « = 14 - - . 12. 



The following observatious, taken with a break corresponding to the 

 removal of the first 10 inches of the bar, bear also the same date: — 



V= 26. V — 13. 



showing that the interruption of a large portion of the pulse of 

 propagation is accompanied by a much less retardation of the phase. 

 There was reason to believe that only about half of this pulse was 

 interrupted in this way. All attempts to measure the amount of the 

 direct pulse have given widely different results ; and the rough exper- 

 iments, upon which the above opinion was founded, do not deserve 

 special notice. 



The phenomena are essentially the same for a short bar of about 

 20 inches in length ; that is, of course, as far as they go. In the 

 notation used above, the results are reduced as follows : — 



ti rr 4.5 1) := 14.5 



d = 2 

 3 

 4 

 6 

 6 

 7 

 9 

 12 

 17 



A small rod of one fourth inch diameter was now substituted, and 

 much smaller differences of phase were obtained, corresponding to the 



