ABSTRACT OF PROCEEDINGS. XXX111. 



into normals with the zenith distance as argument. An 

 examination of these results clearly indicates that the so 

 called constant of refraction needs, not only a correction, 

 but one for every zenith distance. It may be remarked 

 that similar conclusions have been obtained by recent 

 investigators in this connection. From the results of this 

 work the author arrives at a value of the constant 



a = 60"-283 

 for B = 760 mm. at (C) and t = (C) 



It would appear from this investigation, however, that the 

 formula from which the refractions are computed requires 

 modification. The formula may be retained unaltered and 

 the desired result obtained by correcting the value of Log a 

 of the tables in a manner shewn in the paper. Thus we 

 have Log a a = + 0*000122 [52° 30' 33" -s] 



in which s equals the zenith distance in arc. 



The conclusions arrived at by the author are as follows: 

 That if observations of zenitli distance of celestial objects 

 are taken between limits of time separated by Some hours, 

 then greater accuracy in the reductions, to obtain correct 

 positions, can be obtained by taking fully into consideration 

 the fluctuations of the height of the barometer and especi- 

 ally the variation of the temperature, indicated by the 

 readings of the thermometer, when computing the refrac- 

 tions for a series of observations extending over a period 

 of several hours duration. Adopting a state of the 

 atmosphere for a mean of the times of observation does 

 not seem sufficient. Further, the refraction table, (Bessel), 

 in use at the Sydney observatory would represent the 

 observed refractions much better, if a correction be applied 

 for the difference in the force of gravity at Greenwich and 

 Sydney. This correction is represented by a very simple 

 equation which is a function of the latitudes of the two 

 places. The author also considers that the refractions 



c _Nov. 1, IP05. 



