24 
RESEARCHES IN STELEAR PHOTOMETRY. 
The whole time covered by the observations, starting from the first, was divided 
into parts corresponding in length with the star’s assumed period. The quantity 
t in the twelfth column is the time elapsed in days since the beginning of each 
of these parts. The last column, “ J Mag,” gives the residual between the ob¬ 
served magnitude and the reading from the mean light-curve corresponding to 
the time t. 
Table 12 gives the data for finding the mean light-curve, following the form 
used by Turner in his reductions of the Rousdon variable star observations' 
with some additions. Dividing the assumed period, 282 days, into twelve parts 
we have in this case 23.5 day groups. The heading of Table 12 gives in two 
lines the number of the group and the day number corresponding to the last 
MAGNITUDE 
Fig. 4.—Magnitude-Curve for T Andromeda. 
day of the group. The first column gives the Julian day of the beginning of 
each period, starting from an arbitrary date, the first observation. Then follow 
for each group of each period, the mean t and magnitude in the Harvard system 
(found by averaging the quantities in the last two columns of Table 11 for the 
interval covered by the group) J M, the difference between the mean magnitude 
and the reading from the mean light-curve for the time t, and the number of 
nights’observations in the group. Finally, at the foot of the table will be found 
the general means of the tabulated quantities t, M and J M , followed by the 
total number of observations in each group. These general means of t and M are 
platted to form the mean light-curve (fig. 6, page 27). 
1 Memoirs of the Royal Astronomical Society, vol. 55 , lix et seq. 
