Abstract
Temperature and time affect the nucleation and freezing of insects extrinsically, functioning together as rate of cooling. Intrinsic factors include the nucleating agents, both indigenous and acquired, and the water in contact with them. The nucleation process is subject to many influences, including chance, and is therefore to some extent unpredictable.No substance has yet been identified as the primary nucleating agent of an insect. It is presumed that a variety of tissue substances and structures possess nucleating ability, and those in the haemolymph are considered most likely to initiate freezing. In feeding insects, foreign nucleating agents (motes) contained in the digestive tract are usually more efficient than are indigenous agents. A rough estimate of the density of the most efficient class of nucleating agents in a population of similar insects may be obtained from the variability of supercooling points and of the site of nucleation.A volume effect, reflecting the probability that a larger volume of water contains more and better nucleators than does a smaller volume, may be detected in insect supercooling ability, provided that proper account is taken of all other influences. Appendages give a fairly accurate measure of the potential supercooling ability of an insect's tissues when this ability is obscured by prior nucleation in gut contents.Contamination, by airborne or other motes, that affects the supercooling of liquid samples, dissected tissues, and wounded or regurgitating insects may sometimes be avoided or circumvented by various means and potential supercooling ability is then realized.Repeated freezing of water drops, haemolymph drops, and intact insects produced consistent supercooling in some and erratic amounts of supercooling in others. Consistent supercooling can readily be explained as repetitive action of the same nucleating agent. Erratic supercooling appears to involve inhibition, destruction, substitution, and reactivation of nucleating agents, but the nature of such actions remains unknown.

This publication has 3 references indexed in Scilit: