The observed temperatures of IIF (flashing temperatures) fell essentially into two groups with respect to developmental stage and into two groups with respect to the type of nucleation. With respect to developmental stage, IIF in oocytes and one-cell embryos and simultaneous IIF in the blastomeres of two- to eight-cell embryos occurred between —35 and —43°C. That range overlaps the calculated Th, a fact that leads us to conclude that their flashing is a consequence of homogeneous nucleation. In theory, one might expect a slight lowering of Th with an increase in the number of blastomeres per stage. Because embryo volume remains nearly constant with development, the volume of individual blastomeres has to decrease. However, as indicated above, the dependence of Th on water volume is weak, and our failure to see a volume effect may simply be a consequence of the noise in the system. further
Morulae, in contrast, undergo simultaneous IIF of blasto-meres at a much higher mean temperature, namely —23°C. Therefore, the nucleation of their blastomeres must be by a heterogeneous route. Internal heterogeneous nucleators are absent in the earlier stages, as evidenced by the fact that they only nucleate when temperatures have dropped to near Th. Consequently, it seems very unlikely that intracellular heterogeneous nucleators would abruptly appear at the morula stage. If they do not, then the likely explanation is that nucleation in morulae is the result of contact between supercooled water in blastomeres and external ice that has penetrated their plasma membrane.
With respect to embryos that contain > two blastomeres (i.e., two-cell to morula), the second division into two groups is with respect to the type of nucleation. In the great majority of cases (79% to 91%, depending on the developmental stage), the flashing of individual blastomeres occurred simultaneously; but in a minority of cases (9% to 21%), it occurred sequentially with a variable time interval between blastomere flashing. Morulae had the highest percentages of sequential freezing (21%). They differed even more sharply with respect to the time interval between the flashing of the first blastomere and that of the succeeding one (Fig. 5). In the morulae, that time interval was 0.6 sec. In 2-8-cell embryos, it was 10-18 sec. This difference suggests that the plasma membranes of morulae differ in significant ways from earlier stages. That in fact is the case.