TABLK LXI. 



There was no change in No. 7, but the changes in No. 5 and No. 21 require further 

 investigation. For this, the differences from the mean pendulum are put down for 

 each station in Table LXII. 



TABLE LXII. 



Owing to the difference in period of the various pendulums, the values of M S 

 in the last three columns should not be the same at each of the different stations. To 

 make them strictly comparable, a correction should be added, which is equal to the 



/M 

 difference (M S) at Potsdam multiplied by the value of ( s -ll.J 



* Corrected for temperature lag. 

 f Last four observations only. 



{ If S., and S y , represent the time of swing of any pendulum at the Ik-Id station and at Potsdam respec- 

 tively, and G., and G ; , the values of " y " at these stations 



S/Sp = A/ := O, 



M 



= a 



and (S,-S P )-(M.-M I> ) = (o 1) (8,-M,,), 



or correction = (a 1) X dilTcrenrc from mean pendulum at Potsdam. 



M< and M,, represent here the times of swing of the mean pendulum at the field station and at Potsdam 

 respectively. 



77 



