Jotyv—On Variable Stars. 185 
The plate, when developed, would thus reveal a number of circles, 
or part-circles, of stars down to the magnitude at which they 
failed to impress their paths on the film. 
The examination of these traces would reveal, on comparing 
successive plates, in the case of any appreciable variation in bright- 
ness of a particular star, either a strengthening or weakening of 
the linear image, or possibly a complete fading out of the trace. 
Irregularities in the driving gear or atmospheric influences 
would, in affecting all stars upon the plate in a similar fashion, be 
probably in this way differentiated from real variations in bright- 
ness. 
A full exposure of six minutes has been recommended by the 
Paris congress as suitable for securing good measurable images of 
eleventh magnitude stars. A linear velocity of one millimetre in 
from ten to thirty minutes would probably secure sufficient linear 
definition of stars down to the eleventh magnitude. But of course 
the most suitable rate to meet any particular requirement would be 
matter of trial. The radius of the circular motion is also of impor- 
tance, as the number of complete linear images obtained at any 
particular exposure would be the more reduced the larger the 
radius employed, the images moving off or entering upon the 
plate in its extreme positions. In connexion with this latter 
consideration, a circular movement commends itself in preference 
to any other. It would also probably cause a minimum of confu- 
sion in the overlapping or crossing of images. 
02 
