COMSTOCK STUDIES IN ASTRONOMY 63 



The coefficients in equations A and B have been so de- 

 termined as to reproduce with all possible fidelity the re- 

 fractions of the Pulkowa Tables, but they may be made to 

 represent the actual refractions with greater precision by 

 the application to the constant coefficients of the formulae 

 of certain corrections depending upon the latitude of the 

 place at which the refraction is required, the amount of 

 moisture in the air and the wave length of the light whose 

 refraction is to be computed. These corrections are de- 

 veloped in Vol. IX, Publications of the Washburn Observa- 

 tory. The most important of them, and the only one which 

 need be considered here, is that depending upon the lati- 

 tude. Its effect will be sufficiently taken into account by 

 adding to the bracketed coefficient in the equations A and 

 B, the quantity 



C = 225 sin (<p - 60) sin (<p + 60) 



where 9 denotes the latitude and C is given in units of the 

 fifth decimal place. 



