Cepheid variables are well known for their use as a reliable distance indicator. M 31 was among the first galaxies to have the distance measured, in part using Cepheid variables identified by Hubble (1929). A later, more extensive search for Cepheid variables was performed by Baade & Swope (1962, 1964) and Gaposhkin (1963). Since then, substantial work has gone into improving the determinations of the periods and apparent magnitudes of these Cepheids, and in making corrections for the effects of extinction and metallicity variations (e.g., Welch et al. 1986; Freedman & Madore 1990). The result of this work has been a reliable determination of the distance to M 31 with a remaining uncertainty of only 10%, and possibly as little as 5%.
Another important use of Cepheids is due to the fact that their ages
can be determined from the pulsation period. The Cepheid variability
occurs when the star passes through the instability strip, after the
star has evolved away from the main sequence. The mass of the star
determines the period of the variability, and the mass is also
directly related to the main-sequence lifetime of the star.
Therefore, the age of the star can be determined from the period of
the oscillation. This relationship was first discussed in the 1960s
(Young 1961; Efremov & Kopylov 1967), and has
since been confirmed observationally (see Tsvetkov 1988 for an
overview). The typical range of Cepheid periods observed in the Local Group
galaxies, between 
1 and 50 days, corresponds to ages of 250
Myrs and 25 Myrs, respectively. Thus, Cepheid variables can be used to
trace stellar populations with intermediate ages.
Substantial work has already been done to identify Cepheids in M 31 (Baade & Swope 1962, 1964; Gaposhkin 1963). These searches were limited to a small portion of M 31 (see Fig. 1 (click here)) and ignore the spiral arm regions, where star formation is most active. As part of an unrelated variability study in M 31, we have surveyed a large portion of the eastern and southern spiral arm regions. We have used this dataset to identify Cepheids. In this article, we report on the analysis of the data and identification of the variables. In a companion paper, we will discuss the distribution of the Cepheids in terms of the star formation history.
Figure 1:
This figure shows the locations of the Cepheids identified by previous
researchers (Baade & Swope 1962, 1964; Gaposhkin 1963 --
's) and our new Cepheid candidates (squares) overlayed on an HI contour
map (Unwin 1980). The inset shows the locations of our 9
fields (squares) and the Baade's Fields I -- III. Baade's field IV is
beyond the southern edge of this figure