570 



SCIENCE. 



[N. S. Vol. XXIV. No. 618. 



authors take advantage of the completion of 

 the last zone to make a discussion of their 

 results, and a comparison between them and 

 the determinations of other astronomers, a 

 brief synopsis of which follows. 



Systematic Differences in the Potsdam Oh- 

 servations. — Small systematic differences ap- 

 pear between the two observers, and between 

 the photometers which they employed. For a 

 mean interval of 0.43 magnitude the mean 

 difference in brightness of two stars, with 

 photometer CI, is smaller by 0.02 when meas- 

 ured by M than by K. With photometer D 

 a difference of 0.01 was found for a measured 

 interval of 0.28. 



Differences also appear between the observa- 

 tions of the fundamental stars and those of 

 the zones. Taking all four zones together a 

 mean difference of 0.49 magnitude was meas- 

 ured less by 0.030 in the zone observations 

 than in those of the fundamental stars, when 

 photometer CI was used, and less by 0.007 

 for photometer D. On the assumption that 

 this quantity increases according to the meas- 

 ured difference, this corresponds to a differ- 

 ence of 0.061 for a full magnitude for pho- 

 tometer CI, and to 0.018 for D. The latter 

 instrument apparently gives results of slightly 

 greater precision. A comparison of the dif- 

 ferent photometers also shows small differences 

 for both magnitude and color. 



For the determination of the differences 

 between M and K, as affected by magnitude 

 and color, a discussion is given of all except 

 the fundamental stars. They are placed in 

 four groups, white, yellowish white, whitish 

 yellow, yellow and deeper shades. For all, 

 M — K^ — 0.03 magnitude. This result 

 agrees well with that of the third zone, but 

 is somewhat at variance with that of the first 

 zone, where the difference was + 0.02. In the 

 estimation of color intensities no significant 

 differences appear. 



The precision attained is indicated by a 

 comparison of the measurements of 2,485 

 stars, which were observed by both M and K. 

 These give a mean difference between the 

 observers of rt 0.11 magn. The probable 

 error of a single observation is =i= 0.052, and 

 for a catalogue value based on two evenings' 



observations, ztz 0.037. There seem to be no 

 marked differences in these values as influ- 

 enced by magnitude or color. 



Comparison between the Potsdam Measure- 

 ments and the Bonn Estimates. — The compari- 

 son with the estimated magnitudes of the 

 Bonn Durchmusterung (BD) presents some 

 curious features. For naked-eye stars of all 

 colors, a magnitude by the BD corresponds 

 to 0.9035 of the Potsdam scale, while for the 

 fainter stars the corresponding value is 1.0566. 

 The logarithms of these quantities are 0.361 

 and 0.423. The corresponding values for the 

 three preceding zones were 0.329, 0.362, 0.366 

 for bright stars and 0.400, 0.457, 0.484 for 

 faint stars. The variations in these quanti- 

 ties seem pretty large. 



When taken with respect to color, the dif- 

 ference, Potsdam — BD, decreases systematic- 

 ally from + 0.41 for white stars to — 0.01 

 for yellow stars. This difference, Potsdam — 

 BD, shows also a marked but systematic varia- 

 tion according to the declination of the stars. 

 This, however, might perhaps be explained by 

 a large atmospheric absorption at Bonn, for 

 which no correction was applied. 



Comparison between the Determinations at 

 Potsdam, Harvard and Oxford. — The Harvard 

 values used are those of volumes XIV. and 

 XXIV. These are referred to as Pickering I. 

 and Pickering II. Later volumes of the Har- 

 vard photometry will be discussed at another 

 time. Tables are given showing the system- 

 atic mean differences, Potsdam — P I., Pots- 

 dam — PII. and Potsdam — Pr (Oxford), for 

 the zone in question, and a special table giving 

 the individual cases in which the differences 

 amount to half a magnitude, or more. The 

 number of such cases is small, one for P I. 

 and five for P II. and Pr. This number is 

 less than in preceding parts of the work, which 

 is explained by the authors as due to the fact 

 that both at Harvard and at Oxford the pole- 

 star was the standard of comparison, and that 

 errors would naturally be fewer in this zone, 

 60°-90°, which lies near the pole-star. 



These tables are followed by others giving 

 the results for all four zones combined. Since 

 the systematic differences are practically the 

 same for all four zones, an inspection of the 



