Study  of  the  diffuse  PAH/VSG  component in external galaxies is an
almost  completely open question:  a) the IR bands at 3.3, 6.2, 7.7, 8.6
and  11.3 microns have indeed been observed in external galaxies, but in
practically  all  cases this is in starburst galaxies where the emission
is  intense  and comes from very dusty regions, irradiated by the strong
UV  fluxes  of  very  young  stars;  b)  on  the other hand, the "Arome"
experiment detected a diffuse emission from the Milky Way in the 3.3 and
6.2  microns  lines, showing that the carriers of the bands were also an
important component of the Diffuse Interstellar Medium of our Galaxy.
  Such  a  study is however strongly limited by the small column density
of  carriers as long as the line of sight is not within the plane of the
studied  galaxy:  this is why edge-on galaxies offer the most favourable
cases (more than 100 times larger optical path than face-on galaxies).
  We  propose here to extend the results from "Arome" in our Galaxy to a
small  sample of external nearby spiral edge-on galaxies by mapping them
with  ISOCAM  in the dedicated PAH filters and with PHOT-P in a few long
wavelength bands and finally by using PHOT-S on the central region.
  Beyond  the  proof  of an universal presence of a PAH/VSG component in
the  Diffuse Interstellar Medium of spiral galaxies, those observations,
combined  with  maps  at  radio  wavelengths  (HI,  HII, CO) and near-IR
(stellar population), will allow us to study :  a) the dependance of the
ionization  state,  size and relative abundance of the IR bands carriers
with  the  conditions  of  excitation  and  density  at the scale of the
galactic  structure (nuclear region, bulge, disk, giant complexes); this
will  help to disentangle the complex problem of the competition between
photo-destruction,    shielding   by   dust   from   exciting   photons,
sticking/ejection  of  PAH on/from grain mantle etc; b) the ISM at large
galactic  longitudes  and  latitudes;  this will be totally new, and may
give significant inputs to more general problems such as the hidden mass
or the metallicity gradient.