M.-P. VOL. I.] CRAWFORD CONSTANT OF REFRACTION. 



2. Details of Observations. A night's program consisted 

 in observing the above list, together with three nadirs, one 

 before, one during, and one after the observing of the stars. 

 As has been pointed out, the nadirs are not necessary for 

 the refraction determinations, but, were taken for the 

 reduction of the latitude, which is a problem practically 

 inseparable from the main one undertaken here. 



No transits were observed during these observations, the 

 whole attention being devoted to the observations for zenith 

 distance. The telescope was set to the nearest 2' and not 

 disturbed until the observation had been completed. The 

 bisection was made (with but a very few exceptions) at the 

 central transit wire, by means of the declination micrometer. 

 For the sake of uniformity every star was bisected but once 

 during its transit. Because of unavoidable circumstances a 

 few of the stars had passed the meridian before the bisection 



