Polar Stratospheric Cloud (PSC) measurements by ground-based LIDAR were carried out at Dumont d'Urville, Antarctica, during the years 1989 to 1993. From such measurements it can be seen that there are cases of PSCs that are not consistent with the simplest nitric acid trihydrate (NAT) theories. Several cases of long-lasting, non-depolarizing PSCs were detected at temperatures below or close to the NAT freezing threshold, at about 195 ¿K, suggesting the presence of durable supercooled droplets. PSC cases showing depolarizing (frozen) particles well above the NAT expected threshold are also shown. These results seems to be more consistent with recent laboratory and in situ findings, suggesting a close link between sulfate and PSC particles through the HNO3-H2O-H2SO4 ternary system. In this framework, non-depolarizing clouds observed below and close to 195 ¿K would arise from the uptake of HNO3 and water by the background particles before the freezing of the ternary system and the successive growth in nitric-acid hydrates. Depolarizing ''warm'' PSCs are also consistent with the laboratory-observed high melting point of the frozen sulfate core, that remains after NAT evaporation. A problematic PSC case, that cannot be easily explained by this theory is also shown. ¿ American Geophysical Union 1995