Аннотация:http://arxiv.org/abs/2008.13682I consider the kinetics of water freezing and show that, at small sub-zero temperatures, (i) the time of ice nucleation within the bulk water environment is enormous and therefore cannot take place either in lakes of in living cells; (ii) that the ice nucleation needs some ice-binding surfaces to occur, but (iii) even this kind of ice nucleation can take place, as a rule, only at the temperatures that are a few degrees below 0oC. Further, I discuss factors that can drastically reduce the ice nucleation time at nearly-zero temperatures both in open reservoirs, where water contacts with air, and in cells, where there is no such contact.