304 



PRITCHAKD S WEDGE PHOTOMETER. 



the mean of the 3, which mean was used as the standard. The observations were all 

 made within 1| hours of the meridian. November 4th was bright moonlight. There 

 was a slight moonlight during part of the observations of November 1st, but not 

 enough to be troublesome ; the background on which the stars disappeared became 

 perfectly black before they vanished, which was not the case on November 4th 

 with stars 4 and 6. These were lost to the eye, while the sky background was still 

 visible in the diaphragm hole. 



For some not very obvious reason, the extinction mean of the three stars varied 

 very much on the four nights. It was respectively (in wedge-readings) 2.051, 

 2.113, 2.326, and 2.494 inches. It looks very much as if the eye, after practice, 

 became able to follow the star farther before extinction. 



For the first three stars the average deviation of the determination for each night 

 from the mean of the four nights is ± 0.062 in. of the wedge, indicating a probable 

 error for one night's work of ±0.054 in., and a probable error of half that amount 

 for the determination by the four nights' observations. Assuming from page 323 

 that O.Tin, corresponds to one magnitude, the probable error of the finally deter- 

 mined difference of magnitude between the assumed standard (mean of Nos. 1, 2, 

 and 3) and either of the first three stars is ±0.04 of a magnitude. For star No. 5 

 (read only five times each night instead of ten), it is about ±0.05 of a magnitude; 

 the single-niffht error is about the same for No. 4, but having been observed only 

 on three nights the final error is correspondingly larger, — about ± 0.06 m. The 

 probable error for No. 6 is nearly twice as great; it was obviously too faint to admit 

 of a good determination by the apparatus used. 



