VARIABLE STARS OF LONG PERIOD 



BY 



EDWARD C. PICKEMNG. 



The problem présents itself to every owner of a télescope, how to 

 use it in order to obtain results of real scientific value. The nnmber of 

 télescopes of moderate size, 10 to 30 cm. aperture, is very large, and 

 ont of ail proportion to the results obtained with them. Such investiga- 

 tions as the motion of stars in the line of sight, micrometric measures 

 of position, and détails in the surfaces of the planets are now pursued 

 with such skill, aided by the most powerful télescopes in the world, that 

 work with small instruments is, in most cases, wasted. One kind of 

 work, the observation of variable stars by Aïigelandeu's method, is 

 usually recommended for small télescopes, but here the results will be 

 of but little value uuless made according to a clefinite system. It lias 

 been the aim of the Harvard Collège Observatory to render such obser- 

 vations of real value, as will be seen by the références given below to 

 its publications. Yisual observations, by Argelander/s method, of va- 

 riable stars of small range are of little value, since both the accidentai 

 and the system atic errors may amount to several tenths of a magnitude, 

 or a large portion of the total change. The photo meter in use here with 

 the 15-inch Equatorial gives results whose probable error is less than 

 three one hundredths of a magnitude. A variation of a tenth of a mag- 

 nitude can be determined with certainty. Observations on a single night, 

 with such an instrument, are more valuable in determining small and 

 rapid variations, than those extending over months, made by the eye 

 alone. See Harvard Circulars, Nos. 23, 30 and 41. Again, if the vari- 

 ation is regular, it is much better to observe the light-curve once for 

 ail with the photometer, when subsecpient observations will only be 

 required to détermine the period and its variations, if any. On the other 



