330 



SCIENCE. 



LN. S. Vol. XVII. No. 426. 



I have elsewhere given reasons for suspect- 

 ing the genuineness of result (1). There 

 are also reasons for the small weight as- 

 signed to result (4) aside from the some- 

 what larger value given. 



It is proposed to continue this series for 

 another year. It wiU. then be terminated 

 unless means can be had for giving the 

 investigation a wider scope. For a number 

 of years I have been hoping that I might 

 be able to set up an instrument of different 

 construction and have a second series of 

 observations carried on simultaneously with 

 my own for a period of at least two or three 

 years. At present the necessary means 

 are not available, but I have not entirely 

 abandoned this project. 



Micrometrical Measures of Individual Stars 



in the Great Globular Clusters: E. E. 



Barnard. 



The great power of the forty-inch re- 

 fractor of the Yerkes Observatory has been 

 utilized in a systematic micrometrical sur- 

 vey of between 600 and 700 small stars in 

 the globular clusters M S, M 5, M 13 and 

 M 92. 



The paper deals mainly with the meas- 

 ures of stars in M 13 Herculis and a com- 

 parison of these with measures made by 

 Dr. Schriner, of Potsdam, in 1891, of pho- 

 tographs of the cluster. This comparison 

 shows a generally close agreement between 

 the photographic and visual measures. 

 There are a few discordances amounting 

 to one or more seconds of are. There does 

 not seem to be any proof that these are 

 due to motion in these stars, but rather 

 due to the difficulty of making the photo- 

 graphic measures. In the ten years ' inter- 

 val there does not appear to be any certain 

 proof of motion in any of the stars luider 

 observation. In the work with the large 

 telescope the stars are referred to a stand- 

 ard star in each cluster. This star is ac- 

 curately measured with reference to known 



stars, and its absolute position given, 

 from which the exact place of any one of 

 the small stars observed can be easily de- 

 duced. The measures were made by the 

 method of position angle and double dis- 

 tances, though the relative position of the 

 stars to the standard stars are given in 

 the order of ^a and ^'5. 



On Some of the Variable Stars in the 

 Cluster M 5, Lihrce: E. E. Barnard. 

 These are observations of some of the 



variable stars discovered in this cluster by 



Professor S. I. Bailey. 



The smaller stars all have periods of 



nearly half a day, but there are three 



bright stars — the brightest in the cluster— 



which have relatively long periods. These 



periods are : 



star No. Period, Days. Light Range. 



42 25.7739 1 J m. 



84 26.5760 IJm. 



50 106.17 Im. 



The first two rise rapidly to maximum 

 and decline slowly to minimum. No. 50, 

 which seems to have the longest period in 

 the cluster, differs markedly from the other 

 two in that its rise and decline are both 

 slow and uniform. All three are slightly 

 yellowish at maximum. 



Several of the small, quick-period vari- 

 ables were under observation. The best 

 observed of these was No. 33, whose period 

 is 0^ 121 2m 7s.6^ and whose light range is 

 about one magnitude (1™.1). The light 

 curve for this star is rather remarkable. 



The normal condition is faint at about 

 14J mag. At about one hour before 

 maximum it begins to rise. Its light in- 

 creases rapidly, and the duration of maxi- 

 mum is very short. The star then declines 

 about as rapidly as it rose, for about forty 

 minutes. It then seems to halt in the de- 

 cline, and from this on sinks very slowly 

 to minimum, not reaching its faintest or 

 normal condition until seven or eight 

 hours after maximum. 



