[plaskett] THE SPECTROSCOPIC BINARY THETA 2 TAURI 165 



choice of different lines with different wave lengths for measurements. 



Another reason is to be found in the difference between the velo- 

 city of the system obtained here + 42-6 km per second and that obtained 

 from its stream motion + 39-2. ti 2 Tauri is of special interest as being 

 one of the moving stream in Taurus described by Prof. Boss. 1 His com- 

 puted radial velocity for Û 2 Tauri is 40- 5 km per second, 2 • 1 km lessthan 

 the Ottawa value. His velocity is based on Kustner's determination of 

 the radial motion of three other stars of the group-. In a later discussion 

 of the Taurus stream by Wilson 2 in which the computed values are based 

 on the radial velocities of 8 stars of the stream determined by Camp- 

 bell, and hence of much greater weight, the velocity of # 2 Tauri comes 

 out at 39-2 or 3 -4 km smaller than the Ottawa value. 



It seems to me probable therefore that the Ottawa y is over 3 

 kms too high, and, though it is possible to explain this systematic 

 difference by incorrect identifications or wave lengths, it is more likely 

 due to some cause which may be also operative in producing the^ 

 curious humps in the curve and causing the early observations to have 

 such unreasonably large residuals. 



It is of interest to interpolate here that if Boss's value of the 

 proper motion and of the distance of the convergent be accepted the 

 value of the parallax of ft 2 Tauri is 0-023, equivalent to a light journey 

 of about 140 years. 



The similarity between the velocity curve of ft 2 Tauri and that of 

 the Cepheid variable W. Sagitarii 3 is quite marked, the deviations 

 from simple elliptic motion occurring in exactly the same relative 

 positions in the orbits and being of approximately equal relative 

 magnitudes. 



Moreover except in the longer period and higher eccentricity the 

 elements are quite similar and it may be that the abnormal effects 

 are produced by the same causes. Although the range must be small 

 it is possible that accurate photometric observations might show 

 d 2 Tauri to be variable and it would be of interest to have this tested. 



Although it is possible that a better orbit would be obtained if 

 a considerable number of additional spectra were made and measured, 

 the character of the spectrum lines is such as to render this ad- 

 ditional work of doubtful value. 



I have pleasure in acknowledging the interest and helpful sug- 

 gestions of Dr. King in this work. 

 Dominion Observatory, 



Ottawa, Feb. 1915. ____ 



•Astronomical Journal XXVI p. 31. 

 2 Popular Astronomy XX p. 359. 

 3 Astrophysical Journal XX p. 1/2. 



