16 
HAGEN. 
the correction. A large correction is often considered more 
advantageous than a small one. In this case the PD. gives 
elaborate tables (VI and VIII) for reducing its system to 
other catalogues. 
A last word should be said about the designation of colors. 
The PD. designates them by letters, starting from the funda¬ 
mental colors, white, yellow, red, with the respective letters 
W, Y, R, and combining them thus : YW, WY, RY, YR. If 
variable stars were included in the work, an additional fun¬ 
damental color for very intense red would have to be added, 
suppose with the letter P. These four colors, with their 
combinations by two, would give exactly ten designations. 
Why not use ten figures instead ? Letters are not equally 
understood in all languages; their combinations are limited ; 
they do not yield to computation, like taking the arithmeti¬ 
cal mean. A scale of ten was used by Schmidt, and was 
followed by Safarik, Krueger, and Chandler. Fortunately, 
the PD. scale of letters agrees perfectly with Chandler’s nu¬ 
merical scale (in his catalogues of variable stars) and can be 
translated into it by the following table : 
0 1 234 56789 
W, YW, WY, Y, RY, YR, R, etc. 
The additional signs -f and —, by which the PD. intervals 
are subdivided, can also be given in these numbers— e. g., 
YW+ = 1*2, WY — = 2T. 
It may be added that the numerical scales of Schmidt, 
Krueger, and Safarik agree among themselves, but differ 
from the above. 1 
In conclusion, attention may be called to the forthcoming 
“ Revised Harvard Photometry,” which will combine the 
results of seven previous Harvard catalogues and compre¬ 
hend the brighter stars from the north pole to the south 
pole, with their spectra and the magnitudes of other cata¬ 
logues, beginning from Herschel. 
1 Vierteljahrs-Schrift, vol. XXXIV, page 297. 
