l82 REPORTS OI' INVESTIGATIONS AND PROJECTS. 



This shows that there are no systematic errors, affecting the determination 

 of parallax, depending on these things. A very careful search of this sort 

 has failed to detect systematic errors of any sort affecting the parallax by 

 so much as o.oi''. 



(b) The measured positions are more accurate for stars near the middle 

 of the plate than for those near the edge, and for stars of moderate brightness 

 than for either brighter or fainter ones (whose images are either over- or 

 under-exposed). On the average, the accuracy of the photographs is fully 

 equal to that of any other method of determining the relative positions of 

 stars. Most of the outstanding error is due to real displacements of the 

 star-images on the plate, and can not be eliminated by repeating the measure- 

 ment. Plates taken at an interval of two years agree practically as well 

 (when allowance is made for the real motion of the stars in the interval) as 

 those taken on consecutive nights. 



(c) Professor Pickering has very kindly offered to have the spectral 

 types of all the stars observed in this work determined at the Harvard Obser- 

 vatory, and the results for two fields have already been sent to Dr. Russell. 

 It seems probable that the comparison stars are distributed pretty evenly 

 among the different types. The mean parallax of the stars of each type can 

 then be determined with very great accuracy. This will afford the first 

 direct evidence whether red or white stars of the same apparent brightness 

 average nearer to us. 



(3) Determinations of the parallax of many of the "parallax stars" of 

 the present work are found among the long series of such observations pub- 

 lished since this work was begun. The results of various observers, using 

 different methods, agree, on the average, with those of Dr. Russell within 

 the limits set by the probable errors computed from the "internal" agreement 

 of the observations of each series. 



The preparation of a memoir, giving a full account of Dr. Russell's work, 

 is under way, its completion awaiting data which Professor Pickering has 

 kindly offered. 



Schlesinger, Frank, Allegheny, Pennsylvania. Grant No. 422. Completion 

 of parallax computations begun under the direction of Dr. G. B. Hale. 

 (For previous report see Year Book No. 6, p. 175.) $300. 



This work has been somewhat delayed by the pressure of Dr. Schlesinger's 

 other duties, but it is now ready for publication. The parallaxes of 23 stars 

 have been deduced. In some cases it was possible to deduce independent 

 values of the parallax for the same star. A comparison among these, and 

 also with the best previous determinations for some of these stars, indicates 

 that the experiment of determining parallaxes photographically with a long- 

 focus telescope has been very successful. The results of this research are to 

 be published in the Astrophysical Journal. 



