July 27, 1900-] 



SCIENCE. 



123 



variables were made by Argelander's 

 method, using a sequence of comparison 

 stars whose maguitudes have not yet been 

 determined. The results are, therefore, 

 given in grades. The value of one of 

 these grades is somewhat uncertain, but is 

 not far from a tenth of a magnitude, since 

 in a previous work the value of my grade 

 has been 0.085 of a magnitude. The ob- 

 servations were then plotted, using vertical 

 distances to represent magnitudes and hori- 

 zontal distances to represent time, and a 

 smooth curve was drawn through them. 

 The time scale employed in this drawing 

 was very open, in order to read with greater 

 accuracy the ordinates of the curve corre- 

 sponding to intervals of five minutes. The 

 results of the measures are very accordant. 

 Of all the measures on the Lick plates of 

 ten minutes' exposure the average devia- 

 tion from the curve is less than half a 

 grade. 



From these curves it appears that the 

 total increase of light, amounting to 17.5 

 grades in the case of variable No. 11, takes 

 place within 70 minutes; in the case of 

 No. 96, an increase of 16.7 grades occurs 

 within 60 minutes ; and No. 119, 17.0 

 grades, within 75 minutes. The maximum 

 increase during any interval of 5 minutes, 

 is, in the case of No. 11, 1.9 grades ; No. 96, 

 2.5 grades; No. 119, 1.5 grades. During 

 30 minutes No. 11 increases in magnitude 

 10.9 grades, or at the rate of 21.8 grades per 

 hour ; No. 96, 12.8 grades, or at the rate of 

 25.6 grades per hour; No. 119, 8.6 grades, 

 or at the rate of 17.2 grades per hour. The 

 greatest rapidity is met in the case of No. 

 96, where for 5 minutes the increase is at 

 the rate of 30 grades, or about 2.5 magni- 

 tudes per hour, and during 30 minutes has 

 a rate of 25.6 grades, or more than two 

 magnitudes per hour. This rate of change 

 appears to be more rapid than that of any 

 other star known. 



The Algol variable U Cephei, which per- 



haps undergoes the most rapid change of 

 any star not found in clusters, changes at 

 the rate of about one and a half magnitudes 

 per hour, during the half hour when its in- 

 crease and decrease are most rapid. The 

 total times of increase of the three stars, 70, 

 60 and 75 minutes, are 9, 8 and 10 per cent, 

 respectively, of their whole periods. Near 

 the beginning and end of increase, however, 

 the rate of change seems to be relatively 

 much slower. If we allow one and a-half 

 grades for each of these slow changes, mak- 

 ing three grades in all, we find that the 

 remaining increase, amounting to more than 

 four-fifths of the whole change in light, 

 takes place for the three stars in 42, 34 and 

 54 minutes, respectively. That is, in about 

 6, 5 and 8 per cent, of their respective full 

 periods. 



In the case of No. 96 this increase is 

 about ten times as rapid as the correspond- 

 ing decrease. In general it may be stated 

 that the length of periods and form of light- 

 curves are similar to many of those in the 

 clusters Messier 5 and (o Centauri. (See 

 AstrophysicalJournal, Yol. X., 255.) It will 

 be noted that the periods of these three 

 stars are about one-half a day. Several 

 other variables in this cluster appear to 

 have approximately the same period. 



Tlie Series of Parallaxes of Large Proper Mo- 

 tion Stars made luith the Yale Heliometer : 

 By F. L. Chase. 



A large proper motion is, as is well 

 known, the strongest indication of a star's 

 nearness. Some years ago it seemed to us 

 at the Yale Observatory that it would be a 

 promising task to make a rather sweeping 

 survey of all the fainter northern stars 

 having a large proper motion to single out 

 those which show a measurable parallax. 

 Our list was based upon Porter's Catalogue 

 of Proper Motion Stars, and it was our aim 

 to take up all the stars therein contained, 

 which showed an annual ] motion as great 



