112 
Mr. Baxendell also read the following communication 
“On the Determination of the Mean Form of the Light-curve 
of a Variable Star.” 
Since the publication in the last number of the Society’s 
Proceedings of my “ Note on T Aquilae,” I have been 
requested to give an explanation of the method I employ to 
obtain the mean light-curve of a variable star. Many years 
ago I adopted a method which I afterwards found was very 
similar to, if not identical with, the one used by Professor 
Argelander for the same purpose ; but I have recently em- 
ployed another plan which appears to be decidedly preferable, 
as it saves time, renders unnecessary a great deal of tedious 
computation, and gives results of greater weight and value 
than those obtained by the former method. It is simply this : 
Having determined the elements of a variable star, calculate 
the times of all the maxima — or minima, as the case may be— 
that have occurred during the period over which the observa- 
tions extend ; then, taking the original light-curve, read off 
the values of the ordinates for each day, and arrange them 
in a table having as many columns as there are days in the 
star’s mean period, or, in many cases, it will be sufficient to 
take every second, or even third day, always reckoning, of 
course, from the last preceding maximum or minimum, and 
then adding together the values in each column, and taking 
the means, we have the numbers from which to lay down the 
star’s mean light-curve. 
