192 CALIFORNIA ACADEMY OF SCIENCES. [Proc. 3D Ser. 



The remainder of the difference between the values of <p 

 as found from the northern stars and from the southern stars 

 (o".36) is probably due to slight errors in the declinations 

 of the stars used, and to bisection error. 



Conclusion. 



In conclusion it is desired to state that limitations of time 

 have prevented the complete reduction of these observations 

 and of the series taken during the fall months (1899 

 Oct.-Dec). It is hoped that, in the near future, time will 

 be available in which to carry out these reductions by 

 correcting the declinations used and then repeating such 

 portions of these computations as will be necessary. It is 

 also desired to make reductions which will include the 

 relative humidity and a term depending upon the zenith 

 distance. 



It will be noticed from the table (p. 189) that there is a 

 large range in the values of A, viz., from — 0.00584 to 

 -J- 0.00662. This discordance is due partly to the values of 

 the declinations adopted, but is also very clearly a function 

 of the zenith distance. By introducing a term depending 

 upon the zenith distance, and re-solving by Least Squares, 

 this discordance can be greatly diminished. 



From this investigation the following conclusions can be 

 drawn: — 



1. That this preliminary reduction gives for the Constant 



of Refraction 



a = 60". 1 59 



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



2. That for the epoch 1899 June 22, the latitude of the 

 Lick Observatory Meridian Circle was 



9 = + 37° 20' 25". 37. 



3. That the final reduction will show that the Constant 

 of Refraction of the Pulkowa Tables is too large. 



4. That the observing room of the Lick Observatory 

 Meridian Circle is of a very good design, and that there is 

 no need of mounting Meridian Circles in the open air. 



