134 



TUCKER— EFFECT OF DIURNAL VARIATION 



If we take an exchange of signals between Greenwich and a sta- 

 tion 90° west, and the group of stars is observed at midnight, the 

 derived difference of longitude would be o^03 in error, but no differ- 

 ence in the exchanges at both epochs of observation would appear. 

 If the group of stars were observed at sunset at each station, the 

 errors of projected clock rates would be of contrary signs, and there 

 should be a difference of 0^.06 in the exchanges. 



A Longitude 90°. 



A Longitude 180°. 



A Longitude 120°. 

 6 P.M. + o'.03 2 A.M. — o'.oi 



Midnight 00 8 A.M. — 0.02 



6 A.M. —0.03 2 P.M. +0.01 



Noon GO 8 P.M. +0.02 



6 P.M. + o .03 2 A.M. — .01 +0 .04 



