length of the pendulum vibrating seconds. 
75 
Distance from A to a, 329,06 divisions. 
B to b, 366,97. 
Date. 
Readings of the micrometer. 
Divisions. 
A to a. 
B to b. 
Scale. 
+ 39,4 in. 
June 
9*7 
39,0 
630,0 
953,66 
23th. 
7,o 
37,3 
630,0 
955,86 
10,0 
36,5 
630,7 
955,46 
The pieces made to change places 
26th. 
59,o 
87,0 
680,0 
955,oi 
59,o 
84,0 
680,3 
956,81 
5!,o 
75,o 
671,0 
664,5 
955,51 
43,o 
67,7 
957, 16 
4i,5 
68,0 
662,5 
955 76 
Mean of the whole 
955,65 
Hence the distance between the knife edges is 39,4 inches 
+955>^5 divisions of the micrometer. 
Having thus satisfied myself that no injury to the knife 
edges was to be apprehended from moderate use, the pendu- 
lum was again suspended, but now, to my surprise, I found 
the number of vibrations different from what they were 
before the re-measurement. This difference became still 
greater on the following day, and it at length occurred to 
me that the moisture of the atmosphere must have under- 
gone some change, and that an alteration had been thus 
occasioned in the weight of the wooden extremities of the 
pendulum. On referring to the register of the hygrometer 
