Transplantation of pancreases may have clinical utility in the treatment of diabetes, for it has been shown that chemically induced diabetes in rats can be reversed by the transplantation of greater than or equal to four syngeneic fetal pancreases. Allogeneic transplants produce serious immunological problems, but the problems could be ameliorated if tissue-typed organs could be stored in the frozen state. Unfortunately, nearly all attempts to freeze organs have failed. Nevertheless, recent developments in the physical-chemical analysis of freezing injury and its successful application to the freezing of mammalian embryos encouraged us to attempt the freezing of 16 1/2- to 17 1/2-day intact fetal pancreases. The analysis indicated that to achieve success pancreases would have to be cooled less than 1 degree/min and diluted extremely slowly after thawing. Experiments with embryos and red cells indicated that high survivals might require high concentrations of protective solutes and slow warming. These predictions were accurate. After freezing to -78 or -196 degrees and thawing under optimal conditions, the fetal pancreases synthesized 80-100% as much protein as unfrozen controls and they yielded viable allografts. Optimal conditions included suspension in 2 M dimethylsulfoxide, freezing at 0.3 degrees/min, and slow dilution to preclude osmotic shock.