Magnier et al. (1993) recently identified 174 OB associations in M 31
using the PLC method (Battinelli 1991) with the Magnier et al. (1992)
catalogue. The objectivity of the PLC method combined with the depth a
nd accuracy of the Magnier et al. BVRI CCD data make these identificat
ions suitable for photometric studies. We have studied the following p
roperties of the 11 richest associations, 9 groups of associations, th
e eastern and western arms of M 31, and the entire sample of associati
ons: the extinction E(B-V), the slope of the differential luminosity f
unction (dLF) in B and V, and the slope of the initial mass function (
IMF). The data are not of sufficient quality to determine the ages of
associations more accurately than to show that they are younger than a
few 10(7) years old. The extinction ranges between 0.20 and 0.41 magn
itudes (+/-0.05 mag); the western arm is significantly more reddened t
han the eastern arm. We have also found evidence of a difference betwe
en the young stellar populations of the eastern and western arms. The
difference is most pronounced in the slopes of the B band dLF of the t
wo arms (east: 0.46 +/- 0.02, west: 0.79 +/- 0.04, Delta: 0.33 +/- 0.0
45). There is also a marginal difference seen in both the slope of the
V band dLF (east: 0.48 +/- 0.04, west: 0.64 +/- 0.05, Delta: 0.16 +/-
0.064) and the slope of the IMF (east: -2.09 +/- 0.30, west: -3.02 +/
- 0.33, Delta: 0.93 +/- 0.45). The combined weight of these three piec
es of evidence suggests a real difference between the stellar populati
ons of the two arms. It is possible that the difference is not physica
l, but an observational effect due to, e.g., the different reddening,
varying photometric scatter between the two arms or different resoluti
on of the images.