191 5.] McDIARMID— VARIABLE STARS. 67 



33.6 minutes and the two eclipses have a depth of 0.62 mg. and 

 0.57 mg. respectively. From a discussion of the light curve follow- 

 ing the theory as outlined by Professor Russell in A. J ., 36, 5 ; 36, i 

 results were obtained giving the dimensions of the system in terms 

 of the radius of the orbit. It was found that the two stars were of 

 nearly the same size and had the same surface brightness. 



In the case of T Leonis Minoris as in TW Cass, we have two 

 minima, they are however of very different depth, the primary 

 having a loss of 2.46 magnitudes while the secondary has only 0.05 

 magnitudes. The period is 3 days, o hours, 28 min., and 38.0 sec, 

 and is accurately known. Combined with the visual observations 

 I have used the Harvard photographic measures as far back as 

 1889, and have been able to establish a definitive period. The ob- 

 servations are not so complete as in the other systems, the length 

 of the period being so nearly three days ; also weather conditions at 

 special times have entered largely into this. 



From a study of the light curve along the lines of the eclipsing 

 theory it has been found that the stars are of nearly the same size 

 but are very different in surface brightness, the ratio being i : 18. 



The third system TV Cass., whose period, i"^ 19'' 30™ 11.7% has 

 long been known, having been observed by Ashbury and Yendell, 

 was placed under observation in October, 191 3, at Dr. Shapley's 

 suggestion. At that time nothing was known about the secondary 

 eclipse. From my observations it was found that a secondary 

 eclipse of 0.09 magnitude did really exist, coming 21 minutes before 

 the time of mid period. The orbit of the system like that of TW 

 Cass, is eccentric, but in this case the components of the eccentricity 

 can not be separated. 



In the two previous stars it was found from the Hght curve that 

 the stars were of constant brightness between eclipses. The light 

 curve of TV Cass, is somewhat different as there seems to be a 

 gradual rise in the curve between primary and secondary eclipse 

 which corresponds to an increase in brightness of the system. The 

 explanation is, that the radiation of the bright star on the side of the 

 fainter one as they approach the time of secondary eclipse tends to 

 brighten its surface and thus give rise to the phenomenon observed 

 in the light curve. From a study of the light curve it was found 



