In biological systems, nucleation of ice from a supercooled aqueous solution is a stochastic process and always heterogeneous. The average time any solution may remain supercooled is determined only by the degree of supercooling and heterogeneous nucleation sites it encounters. Here we summarize the many and varied definitions of the so-called ‘‘supercooling point,’’ also called the ‘‘temperature of crystallization’’ and the ‘‘nucleation temperature,’’ and exhibit the natural, inherent width associated with this quantity. We describe a new method for accurate determination of the supercooling point, which takes into account the inherent statistical fluctuations of the value. We show further that many measurements on a single unchanging sample are required to make a statistically valid measure of the supercooling point. This raises an interesting difference in circumstances where such repeat measurements are inconvenient, or impossible, for example for live organism experiments. We also discuss the effect of solutes on this temperature of nucleation. Existing data appear to show that various solute species decrease the nucleation temperature somewhat more than the equivalent melting point depression. For non-ionic solutes the species appears not to be a significant factor whereas for ions the species does affect the level of decrease of the nucleation temperature.

The temperature at which a solution spontaneously freezes when cooled below its equilibrium freezing temperature, Tf, is denoted variously as the ‘‘kinetic freezing point’’ [26], the ‘‘temperature of crystallization’’ [36], and the ‘‘nucleation tem- perature’’ [21]. For biological solutions, or even for whole organisms, this temperature of spontaneous freezing is also often called the ‘‘supercooling point’’ (SCP) [19,40], the notation we shall adopt in this paper.