452 
ASTRONOMY: SHAPLEY AND SHAPLEY 
By consulting the table of verbal pronouns on the folder at the end of 
my Preliminary report, the reader will at once see the importance of the 
pronouns mentioned above, in determining the linguistic classification 
of Potawatomi. Complete tables for Potawatomi will be pubhshed in 
the American Anthropologist. 
^Amer. Anthrop., New York, N. S., 6, 369-^11. 
^Smithsonian Inst., Rep. Bur. Amer. Eth., 28, 221-290b. 
THE LIGHT CURVE OF XX CYGNI AS A CONTRIBUTION TO 
THE STUDY OF CEPHEID VARIATION 
By Harlow Shapley and Martha Betz Shapley 
MOUNT WILSON SOLAR OBSERVATORY, CARNEGIE INSTITUTION OF WASHINGTON 
Presented to the Academy, June 29, 1915 
Probably as many as 90% of the 4500 stars whose light intensities 
are known to be variable may be placed in three well-defined classes. 
(1) The eclipsing variables are of various colors (or spectral types); 
their periods range from a few hours to a year or more in length, and 
their variation may be from less than one-tenth to several stellar magni- 
tudes. (2) The long period variables are generally reddish stars; the 
cycle of their light-changes occupies from one hundred to six or seven 
hundred days, and the amount of their Ught variation is usually several 
magnitudes. (3) The largest class, however, since it includes the great 
number of variables in globular clusters and in other special regions 
of the sky, is that known as short period variables or Cepheid variables. 
Practically all types of spectra are represented, though types A and F 
predominate so far as now known. The periods in general are much 
less than 50 days, and for a very large subdivision (the cluster type 
variables) average about 12 hours. The variation, which is practically 
continuous, is nearly always of the order of one magnitude in range, and 
is characterized generally by a more rapid increase than decrease of 
brightness. 
The cause and characteristics of eclipse variation are definitely known. 
But this can not be said of the other two main types of variable stars. 
Though many suggestions have been made, the true cause of long period 
variation remains more or less obscure. The interpretation of Cepheid 
variables has been much debated and much evidence in favor of various 
theories has been collected, but no explanation has as yet received 
general acceptance. The behavior of the spectral lines of Cepheids 
has led to the widely adopted assumption that they are spectroscopic 
binaries and that the light variations are in some way related to orbital 
