and on the Ellipticity of the Earth. 



181 



1824. 



Exp. 

 Velocity 

 in feet. 



Contained 

 Annie. 



Nov. 24 

 Dec. 8 



1825. 



Jan. 10 



Feb. 7 



17 



21 



Mar. 2 



22 



June 3 



4 



1040-49 

 1037-34 



1029-04 

 1020-99 

 1039-25 

 1010-28 

 1014*39 

 1026*64 

 1098-32 

 1118-10 



4° 18' 

 85 42 



Calculated 

 Velocity. 



Estimated 



Effect of 



Wind. 



1045-51 

 1043-24 



1011-48 

 1025-66 

 1033-03 

 1010-91 

 1009-78 

 1029-06 

 1092-82 

 1094-60 



Final 

 Velocity. 



+ 

 + 



+ 

 + 



4-0 

 4-0 



4«0 

 2-0 



+ 4-0 

 — 4*0 

 + 4-0 

 + 25-0 



1049-5 

 1047*2 



1015-5 



1027-7 

 1033-0 

 1010-9 

 1013-8 

 1025-1 

 1096-8 

 1119-6 



Differ- 

 ence. 



9-0 

 9-9 



-13-5 



+ 6-7 



- 5-7 

 + 0-6 



- 0-6 



- 1-5 



- 1-5 

 + 1-5 



Mean error of the whole 

 Of a single set .... ... 



4-9 

 0-5 



In most of the above experiments, the experimental and 

 calculated velocities approximate very closely. There is, no 

 doubt, some uncertainty in the estimated effect of the wind, 

 though it is believed it cannot be great. Perhaps it is a little 

 too great in the first two experiments. I cannot reconcile the 

 third very well by any probable supposition. The only one 

 on which the effect of the wind is considerable, is the last, when 

 it was strong and squally, and blowing nearly in the direction 

 of the sound. Upon the whole, the comparison appears satis- 

 factory, though it would have been less objectionable had the 

 velocity of the wind been ascertained by experiment, and its 

 direction more accurately observed. 



I may add, that since my last communication on experi- 

 ments by the pendulum, I have reconsidered the whole; and 

 upon rejecting those evidently affected with some cause not well 

 explained, I have found the following formula : 



P == 39-01326+0*20686 sin 3 (X— 0) (A) 



In which P is the length of the pendulum, A the observed 

 latitude, and the reduction of the latitude. 

 Also g = 0-00330 = ~±j very nearly ; 

 And P at London by computation from formula (A) is 

 39-13937, while I have found it from Captain Kater's experi- 

 ments to be 39-13938, almost exactly the same. P at Paris, 

 by the same formula is 39-12982, or 0*00053 greater than by 

 experiment. And these two instances show the great accuracy 

 of the formula. 



The 



