SCIENTIFIC ABSTRACT Comets are believed to retain most of the original content of the pristine Solar Nebula. Therefore, their study provides a unique way to understand the history of our Solar System and its relation to the interstellar medium. We propose to observe one or two periodic comets within the ISO CP. The primary goals are to investigate the dust and gas composition of these objects, to probe the molecular excitation mechanisms and the coma physical conditions, to monitor the cometary activity and outgassing processes as a function of heliocentric distance, to compare the behaviour of different objects, to investigate the relation between gas and dust and the anisotropy of cometary activity. The present program concentrates on gas and dust mapping with CAM. The dust distribution can be readily observed by broad band filters in the thermal infrared region. The strongest molecular signal is expected in the water band at 2.7 microns using 11 positions of the CVF, at full resolution. Spectral imaging of the full 2.5-5 micron region with the CVF will allow to observe the distribution of other species in addition to H2O: CO2 in its 4.25 micron band, methanol and other still unidentified species emissions in the 3.2-3.5 micron region. For a launch in fall 1995, the most suitable periodic comet during ISO's planned life seems to be P/Kopff, which we plan to observe at 2 different periods to study its evolution. For a launch in spring 1996, we plan to observe P/Kopff and P/Wild 2. This proposal is complementary to other CP cometary proposals to be done with PHT, LWS and SWS. OBSERVATION SUMMARY Cometary observations are highly time-dependent. The following targets and observing periods were selected in function of the expected comet activity and of the instrument Sun and Earth constraints. Target list and special requirements: For a launch in autumn 1995: - P/Kopff in Apr. 1996 (best period before perihelion: delta = 1.1 AU, rh = 1.8 AU; phase angle = 31 deg.; expected Q[gas] = 4.e28 s-1). (#1, 2, 3). - P/Kopff in Oct. 1996 (best period after perihelion: delta = 1.2 AU, rh = 1.9 AU; phase angle = 30 deg.; expected Q[gas] = 4.e28 s-1). (#4, 5, 6). - P/Kopff in Nov. or Dec. 1996 (repeated observation at a larger heliocentric distance to see evolution). (#7, 8, 9). For a launch in spring 1996: - P/Kopff in Oct. 1996 (best period after perihelion: delta = 1.2 AU, rh = 1.9 AU; phase angle = 30 deg.; expected Q[gas] = 4.e28 s-1). (#1, 2, 3). - P/Wild 2 in Apr. 1997 (best period around perihelion: delta = 1.1 AU, rh = 1.6 AU; phase angle = 39 deg.; expected Q[gas] = 3.e28s-1). (#4, 5, 6). - P/Kopff in Nov. or Dec. 1996 (repeated observation at a larger heliocentric distance to see evolution). (#7, 8, 9). Spectral images to be observed: - Broad band filter images with LW6 (7-8.5 microns), LW7 (8.5-10.7 microns), LW9 (14-16 microns) and LW3 (12-17 microns). AOT CAM01 with 6" pixels. Beam switching is more efficient than single pointing to get low flux levels away from the central peak. We use a pseudo beam switching mode with a sky reference, since AOT CAM03 is not available for a comet. These observations will yield the dust continuum distribution. They need only short integration times and will be performed before each observation with the CVF. With 2000 sec, we get S/N better than 70 at the peak, and better than 10 down to a surface brightness of 60 mJy/arcsec2. - CVF scan, with full spectral resolution of the water band (2.55-2.75 microns). AOT CAM04 with 11 CVF positions and 3" pixels. This will yield the water distribution. With 5200 sec, we expect S/N better than 8 at the peak of the H2O band. - CVF scan of the 2.5-5 microns region, with half the full spectral resolution. AOT CAM04 with 40 CVF positions and 3" pixels. This will yield the distribution of water and other species. With 10600 sec, we expect S/N of the order of 5 to 10 at the peaks of the bands.