The density wave theory usually serves as a basis for the understanding
of the creation and maintaining of spiral arms in disk galaxies. But
there are now various forms of this theory of which none of them is
capable to serve in a sufficiently general way. Of course, this is due
to the fact that hardly a galaxy is a simple system which might serve
as a template. In the contrary, the galaxies with the most pronounced
density wave phenomena are those who have undergone an encounter quite
recently, such as M51 or M81. Rather undisturbed galaxies show the
effects of the density wave only weakly and for most of them the
sensitivity and resolution of today's telescopes is not sufficient to
resolve, e.g., the emission of all velocity tracers properly. Thus
we started the investigation of the molecular gas around a spiral arm
in the nearest galaxy with well defined spiral arms, M31. The region
chosen lies close to the giant star cloud NGC206 on the major axis.
We are thus able to map the velocity field and the distribution of
the molecular gas with high spatial resolution.
 This yields the basis for the questions that can only be addressed
with ISO: How is the star forming rate influenced across the spiral
arm profile? What are the temperature profiles of the dust in this
arm? Is there a visible processing of dust properties when comparing
the interarm regions upstream and downstream of the spiral arm? And
in particular: What is the influence of a giant star cloud like NGC206
on the surrounding ISM? IRAS has shown, that the general distribution
of the dust temperatures in M31 seems to be only very little influenced
by locally enhanced radiation. But its low resolution cannot give a
detailed answer to this question. Only ISO is capable to provide the
high quality of data needed to answer these questions.