846 
been hitherto neglected, presumably be- 
cause the observers deemed it insignificant: 
The largest values, however, of the motion 
of the Earth toward and from the Sun, due 
to this cause, amount to 0.50 kilometer per 
second, in April and October ; and if veloci- 
ties of stars in the line of sight were deter- 
mined directly from absolute wave-lengths, 
instead of relative wave-lengths, uncertain- 
ties of one-half of a kilometer would at once 
arise, in measurements now given to the 
tenth of a kilometer. Expressed in wave- 
lengths that velocity would produce a dis- 
placement of 0.011 tenth-meters at C ; or 0.008 
at F. 
The diurnal correction to velocities in the 
line of sight, due to the Harth’s rotation, 
also seems to have been omitted in reduc- 
ing measures of absolute wave-length. This 
might have an effect up to 0.006 tenth- 
meter at C or 0.005 at F. 
The need of even a higher degree of ac- 
euracy than that yet obtained in relative 
wave-lengths was urged, as an error of 0.01 
tenth-meter in the relative wave-length of 
either a stellar or comparison line (not co- 
incident) whose separation is measured, 
produces an error of 0.7 kilometers per 
second in the velocity of the star as de- 
duced from that line. (Tobe published in 
the Astrophysical Jowrnal. ) 
FRANK SCHLESINGER: Suggestions for the De- 
, termination of Stellar Parallax by 
Photography. 
As the star to be examined for parallax 
will usually exceed each of the comparison 
stars in brightness by six or seven magni- 
tudes, the first problem is to reduce in some 
way most of the light of the brighter stars, 
in order to escape various errors which 
would arise in the measurement of star 
disks very unequal in size and appearance. 
It is suggested that the portion of the pho- 
tographiec film upon which the light of the 
principal star will fall be previously treated 
SCIENCE. 
' [N. S. Von. X. No. 258: 
with some suitable dye, thus greatly reduc- 
ing its photographic action. 
To avoid a second source of error, distor- 
tion of the film after exposure, a modification 
of the method employed by Wilsing is sug- 
gested. let two pictures be taken close to 
each other on the same plate at the first date ; 
let the plate then be stored undeveloped 
ina dark room; after six months make two 
more exposures of the same star and com- 
parison stars on the plate at a little dis- 
tance from the previous impressions; this 
plate can then be developed and measured, 
without fear of any ill effects from distor- 
tion of the film. The first pair of exposures 
on a new plate may then be begun, and this 
stored for six months; and so on until a 
sufficiently long chain of plates is secured 
to give good values both for the parallax 
and for relative proper motions with re- 
spect to comparison stars. 
It is proposed that errors of optical dis- 
tortion caused by peculiarities of the object- 
glass be eliminated from the parallax by ro- 
tating the lens in its own plane through 
180° each time that the telescope is re- 
versed. In this way the objective will pre- 
sent the same position relative to a stellar 
configuration, whether in the east or west, 
and optical distortion, if any, will shift all 
images of the same star alike. 
According to the author’s estimate, a 
single observer working 15 or 18 hours a 
week at the telescope and employing the 
rest of his time in measuring and reducing, 
could give us in three or four years the 
parallaxes of 200 stars with an accuracy 
hitherto attained for only a score. (To be 
published in the Astrophysical Journal.) 
S. I. Battey: Periods of the Variable Stars 
in the Cluster Messier 5. 
This cluster contains about 900 stars on 
the photographs made with the 13-inch 
Boyden refractor, of which about eighty- 
five, or one in eleven, are variable. The 
