346 
ASTRONOMY: H. SHAPLEY 
observations at Harvard on the spectra in a few bright clusters® and 
from the work of Adams and van Maanen on the double cluster in 
Perseus.^^ 
The accompanying table contains the parallaxes and space coordi- 
nates of the 70 open groups which are sufficiently rich, condensed, sym- 
metrical, and distinct from the background to make practicable the use 
of diameter as a criterion of distance. Melotte's well-known cata- 
logue^^ of clusters contains more than twice as many open groups as are 
listed here, but a large number of his clusters are so ill-defined and 
poor in numbers that they appear to be little else than chance groupings 
of Milky Way stars. The omission of these irregular and scattered 
aggregations does not operate selectively in matters of distance or 
distribution. 
The first column of the table contains the number of the cluster in 
Dreyer's New General Catalogue or its indices. A few that are not 
listed by Dreyer retain their numbers in Melotte's catalogue. Paral- 
laxes of the fourteen clusters marked with an asterisk have been de- 
rived directly from measured apparent magnitudes combined with 
absolute magnitudes estimated on the basis of observed colors, or 
spectra, or both. The adopted mean absolute magnitudes of early 
type stars depend on the studies of Kapteyn,^^ Plummer,^^ and Char- 
lier,^^ and on my own results' for the luminosities of blue stars in globular 
clusters. Apparently without exception the brightest stars in open 
clusters (as in globular clusters) are giants in luminosity. 
By using the measured diameters and adopted distances of these 
specially studied clusters to determine a parallax-diameter curve, the 
parallaxes of the other open clusters have been estimated on the basis 
of diameters alone; in most cases, however, the relative distances have 
been roughly checked with the aid of the magnitudes of the brightest 
stars as estimated by Bailey or as derived from Mount Wilson photo- 
graphs. The diameters are the means of measures by Bailey, Melotte, 
Shapley, and Miss Davis. In keeping with the probable accuracy, 
most of the parallaxes have been rounded off to a single significant 
figure after computing the linear galactic coordinates, which are given 
in the fourth and fifth columns. The very largest parallaxes are un- 
certain because of the lack of precision in the corresponding end of the 
parallax-diameter curve; the very smallest are uncertain because of the 
large effect of small errors in measurement of diameter; most of the 
diameters, however, fall within limits for which the curve is well-de- 
fined and for which the measures of diameter by the four observers are 
