306 



REPORT 1851. 



servers changed places and repeated the experiments. The following are 

 the results of eight good experiments. 



The figures of columns A and B mark revolutions of large dial divi- 

 sions, i. e, J^th of the large dial, and divisions, i. e. ^Lth of small dial of 

 Wheatstone's chronograph, its time being in value the same as when used at 

 Dalkey experiments. 



As there seemed to be two sources of error of observation in this mode of 

 experiment, another set of eight good experiments was made, in which the 

 same observer counted the clock, and himself started and stopped the chro- 

 nograph ; and the results below give the readings of the instrument as by 

 each observer for the same epoch of 30 seconds. 



The figures of columns A^ and B* represent revolutions of large dial divi- 

 sions of i^th thereof, and divisions of small dial of ^V'^^ thereof, as before. 



Means of C and D. Means of Cr and D*. 



Differences of the means. . . . 0"-0155 0"-0349 



