SCIENTIFIC ABSTRACT The time variation of mass loss of AGB stars of different luminosity is one of the most longstanding problems in AGB evolution. The uncertainty in the distance to many AGB stars prevents an accurate determination of their bolometric luminosities. The aim of this proposal is to study the mass loss and evolution of AGB stars in the LMC and SMC in order to establish the way in which the mass loss on the AGB varies with time and luminosity. The far-IR energy distribution of AGB stars allow an accurate determination of the dust mass loss rate, and the known distance to these objects yields the bolometric luminosity. This will result in the independent determination of mass loss and luminosity for a statistically significant sample of stars, and will resolve the above mentioned problem in AGB evolution. A comparison with results on galactic AGB stars will give insight in the effect of metallicity on AGB mass loss. OBSERVATION SUMMARY The far-IR energy distribution of a statistically significant sample of AGB stars in the Magellanic Clouds will be studied by means of photometry with PHT and CAM. The shortest wavelength (12 micron) will be observed with CAM (CAM01) for the faint sources and PHT-P (PHT03) for bright sources. For the CAM observations a pixel field of view of 1.5 arcsec will be used. For the PHT observations an aperture of 52 arcsec is used, and these observations will be carried out in chopping mode, with triangular chopping and a chopper throw of 90 arcsec. Both the CAM and PHT-P observations will use the 12 micron IRAS band filter. The 25 micron observations will be done using PHT-P (PHT03), a 52 arcsec aperture, triangular chopping with a 90 arcsec throw and the 25 micron IRAS band filter. The 60 micron observation will be done with PHT-C (PHT22), with triangular chopping and a 150 arcsec chopper throw, using the 60 micron IRAS band filter. In all bands we aim for a S/N ratio of 10 or more. For 15 selected objects, with a range of spectral characteristics, spectra will be taken using PHT-S (PHT40). The S/N ratios for these spectra should be 10 or higher. The observations on each source are concatenated. The scientific rationale for this is that the IR energy distribution and thus the mass loss of AGB stars can only be determined accurately if all bands are observed.