136 
ASTRONOMY: ADAMS AND SHAPLEY 
Although the mean period of RR Lyrae is remarkably constant, con- 
spicuous transitory irregularities are present both in the period and in 
the form of the light curve, much resembling those of the cluster-type 
variable XX Cygni previously discussed in these Proceedings. In 
particular, there is a slow but definite variation in the length of the 
period which seems to complete its cycle in about sixteen years; and an 
oscillation in the time of the rise to maximum light which at times appears 
to have a uniform period of about forty days, and at other times to be 
erratic in both period and ampHtude. The formula: 
Max. - 2414856.451 -F0?566831E - 0?024 sin (0?0340E - 104?5) 
+ [0^004 + 0^013 cos ^ (E - 45)] 
in which E represents the number of periods, predicts the individual 
maxima with high accuracy for the first seven of eight hundred periods 
after the initial epoch, and by dropping the short period harmonic it 
gives the mean maxima accurately for at least fifteen years from the 
initial date. The complete discussion of these features, based on all 
the available series of light measures and on the recent observations of 
the spectrum, will be published as a Contribution from the Mount Wilson 
Solar Observatory. 
THE SPECTRUM OF b CEPHEl 
By Walter S. Adams and Harlow Shapley 
MOUNT WILSON SOLAR OBSERVATORY, CARNEGIE INSTITUTION OF WASHINGTON 
Received by the Academy, February 8, 1916 
About 130 years ago Goodricke discovered the periodic fluctuations 
in the light of the naked-eye star 5 Cephei — the eighth variable known 
to astronomers. None of those previously recorded exhibited the type 
of light variation peculiar to this fourth magnitude star, but of the four 
or five thousand variables found since that time more than half show 
the same character of light fluctuations. The class has, accordingly, 
received the appropriate designation of Cepheid variables. A century 
after the discovery of the light variations, 6 Cephei was also found to 
vary in radial velocity — the sixth spectroscopic variable to be placed on 
record. Again the variation differed from that of other stars. The 
period of the oscillation of the spectral Hnes is identical with that of the 
light variation, but the orbital elements, derived on the hypothesis that 
the star is a binary, permit no satisfactory explanation of the light 
variations. When other velocity- variables of this character were found 
