MOUNT WILSON OBSERVATORY. 237 



PARALLAX OF NEBULA. 



Four methods have thus far been employed by Mr. Hubble for de- 

 termining the distances of nebulous stars and diffuse nebulosity: 



(1) Spectroscopic method for certain stars. This has been applied 

 to 4 out of the 25 available spectra. 



(2) From apparent magnitudes, on assumption of small dispersion in 

 absolute magnitude? ; 40 magnitudes have been obtained so far by polar 

 comparisons with the 10-inch telescope. 



(3) Proper motion. The available pubhshed data have been com- 

 piled, and a program is being prepared to determine the proper motions 

 of the remaining stars. 



(4) Apparent magnitudes of stars superposed on, and blotted out by, 

 diffuse nebulosity, both luminous and dark. A half-dozen plates have 

 been secured for this purpose, but no measures have been made as yet. 



TRIGONOMETRIC STELLAR PARALLAXES AND PROPER MOTIONS. 



During the year Mr. van Maanen made 534 photographs, including 

 817 exposures, at the 80-foot focus of the 60-inch reflector; 347 of these 

 with 638 exposures were taken for parallax measures; 138 with 167 

 exposures were for proper motions; and 8 with 12 exposure? for mis- 

 cellaneous purposes. 



For 22 fields the necessaty plates were secured for the determination 

 of parallaxes, thus giving a total of 122 fields finished to date. Among 

 these were the central stars of 4 additional planetary nebulae, N. G. C. 

 40, 2022, 6210, and 6543; one spiral nebula, N. G. C. 4051 ; and Camp- 

 bell's hydrogen-envelope star, B. D. +30°3639. Including B. D. 

 +30°3639 among the plane taries, the mean absolute magnitude of 

 the central stars of the 1 1 objects for which the parallaxes have now been 

 completed is +8.2, thus confirming the low absolute magnitude of the 

 central stars of planetaries mentioned in the last report. 



When the first 100 parallaxes were finished, an investigation was 

 made of the systematic errors affecting these results. The following 

 conclusions were derived by Mr. van Maanen from this study: 



(1) A comparison of 16 fields observed in common with other ob- 

 servers indicates a systematic error of — 0''001. 



(2) A comparison of the proper motions in right ascension with those 

 given by Boss indicates an error of — OTOOl. 



(3) A comparison with Kapteyn's tables for mean parallaxes indi- 

 cates for the Mount Wilson parallaxes an error of +0':f001. This com- 

 parison also shows that the parallaxes are apparently free from sys- 

 tematic errors which are functions of either right ascension or proper 

 motion. 



(4) A comparison of the mean absolute magnitudes and the mean 

 r-component of the proper motions, expressed in kilometers per second, 

 for different groups of stars, 'with accepted results for intrinsic bright- 

 ness and stellar velocities, indicates an error of +0''001 or +0''002. 



