Few, if any, yeast cells survived rapid cooling to -196 degrees C and subsequent slow warming. After rapid freezing, the suspensions absorbed latent heat of fusion between -15 degrees and 0 degrees C during warming, and the relation between the amount of heat absorbed and the concentration of cells was the same as that in equivalent KCl solutions, indicating that frozen suspensions behave thermally like frozen solutions. The amount of heat absorbed was such that more than 80 per cent of the intracellular solution had to be frozen. The conductometric behavior of frozen suspensions showed that cell solutes were still inside the cells and surrounded by an intact cell membrane at the time heat was being absorbed. Two models are consistent with these findings. The first assumes that intracellular freezing has taken place; the second that all freezable water has left the cells and frozen externally. The latter model is ruled out because rapidly cooled cells do not shrink by an amount equal to the volume of water that would have to be withdrawn to prevent internal freezing.