196 i CALIFORNIA ACADEMY OF SCIENCES. [PROC. 3D SER. 



0.00024690, from v^hjch the following probable errors have 

 been deduced: 



^ <S V. v/. 



r x =,^f 0.0000130 and r = 0^031 = oi'52". 



*?.V 



We, ther^fbrej have from this solution 



Z=5638'.8i'.9 and' x+o.oooioi 0.000013, 

 giving 



loga=loga -\- o.oooioi [5638'.8 z] . 



We are, ; therefore, led to the conclusion that the so-called 

 Constant of Refraction needs not only a correction, but a 

 Correction for 'every zenith distance. In other words, the 

 formula from which refractions are computed needs to be 

 modified. Or, the formula may be retained unaltered, and 

 the desire^ result pbtained by correcting the log p table of 

 the refraction tables used (Pulkowa) by the amount 



A log/*=p ; oopioi [56 38'.8 z]. 



R. T. C. 



