DEPARTMENT OF MERIDIAN ASTROMETRY. 193 



script for the printer is well advanced. It has been decided, however, 

 to delaj' publication, awaiting a systematic search of the catalogue 

 for large proper-motions. This work, which has already been 

 started, will be pushed to an early termination in order that the cata- 

 logue may be speedily published. 



PHOTOMETRIC OBSERVATIONS AT SAN LUIS. 



The photometric expedition to San Luis terminated in February 

 1913, and the instrumental equipment was returned to Albany. 



The object of the expedition was the determination of the magni- 

 tudes of those stars on our observing list, in the southern hemisphere, 

 which had not been observed by Harvard, the Harvard system being 

 used as a base to which our observations should be differentially tied. 



The photometer employed in this work is designed to bring the 

 magnitudes of the star and an artificial star to equality by reducing 

 the light of the artificial star by means of a wedge. 20,758 observa- 

 tions were taken, including the observation of fundamentals; 12,769 

 by M. L. Zimmer and 7,989 by Heroy Jenkins. The number of 

 Harvard comparison stars observed was 1,328, for the determination 

 of 6,725 stars. It was intended that each star to be determined 

 should receive two observations, and with few exceptions this plan 

 was followed. In some cases three observations were secured. 



The preliminary reductions of these observations have been carried 

 out. They indicate a probable error of ± 0^*08 for a single obser- 

 vation, or ±0^06 for two. 



The mean value for absorption is 0^/25 (sec. z — 1), the value ob- 

 tained for the northern Harvard measures, though somewhat larger 

 than the mean of the values for the Arequipa station. While the 

 absorption from individual nights at times deviated considerably 

 from the mean value, in general there was little difference, and 

 as the zenith distance of the south pole was only 57°, an error of O"^*! 

 in the absorption would introduce an extreme error of 0^08 in the 

 derived magnitudes. The mean value for the absorption has there- 

 fore been used in the reductions. 



Aside from the observations mentioned, about 2,500 observations 

 were taken for the determination of the scale value. 



BUILDINGS AND APPARATUS. 



The standard Riefier clock was not sealed when mounted after its 

 return from San Luis. As this clock evidenced considerable but 

 inconsistent changes with variation in barometric pressure and temp- 

 erature, it was deemed advisable to seal the glass cases inclosing the 

 clock. Previous attempts at sealing had proved failures. There- 

 fore care was taken to repack the valve through which the winding 

 rod passes. All possible air-passages were tested by immersing the 



