It has been suggested that stabilization of liposomes and proteins during freeze-drying requires only that they be maintained in a vitrified (glassy) state. In the present paper we show that vitrification is indeed necessary. However, dextran, which exists as a glass at a higher temperature than does trehalose and thus might be expected to stabilize liposomes more effectively, preserves DPPC liposomes only when extremely large quantities of the dextran are added. Dextran does not stabilize egg PC liposomes and, in fact, inhibits the stabilizing effects of trehalose. Dextran also does not depress Tm in the dry phospholipids and shows no interaction with the polar headgroup, as assessed by infrared spectroscopy. Trehalose, by contrast, depresses Tm in dry egg PC by about 60 degrees C and depresses vibrational frequency of the phosphate in the polar headgroup to the frequency seen in the hydrated lipid, an effect we ascribe to hydrogen bonding between the sugar and the polar headgroup. We conclude that while vitrification may be required it is not in itself sufficient to preserve freeze-dried liposomes.