79 



would be, consequently, the lowest death point. One does 

 not see for what reason death would correspond to a sec- 

 ond cooling to the subcooling point ; and, besides, the con- 

 clusions of Bachmetjew have against them that his experi- 

 mental results disagree with the conclusions in 27 per cent 

 of the cases (cf. Payne, 1926). 



Revival after hard freezing was reported by Duval and 

 Portier (1922) for the caterpillars of Cossus cossiis. 

 These authors froze the caterpillars in test tubes im- 

 mersed in cooling mixtures. The temperature was 

 measured with a thermometer placed in the tubes in 

 between the animals. They found that death resulted 

 when the temperature dropped below -21°, while, at 

 higher temperatures, -15 to -17° and -20°, the caterpil- 

 lars could be frozen so hard as to be breakable, and be 

 revived on warming, even if the thawing was rapid (in 

 water at 40° ) . The portions broken otf in the frozen state 

 showed activity after warming. Caterpillars congealed 

 at -15° to -17°, then immersed in liquid air or exposed 

 to the temperature of boiling chloroform (-63°) for 50 

 minutes, or cooled only to - 22°, could not be revived. The 

 authors think that the rigidity of the frozen caterpillars 

 above -21° is due to the congelation of the hodif fluids, 

 while the cellular fluids would still be unfrozen. The lat- 

 ter would congeal at about - 21°, thus inducing death. As 

 a matter of fact, a sudden rise in temperature, indicative 

 of a congelation, was sometimes observed in the neighbor- 

 hood of -20°. The resistance of the caterpillars was 

 found to be less in the Spring than in the Winter. 



A work which perhaps explains some conflicting results 

 of the older authors is that of Payne (1926a, 1926b and 

 1927). This author determined with thermocouples the 

 freezing and the subcooling points of various insects and 

 found these points to be always low in Winter and to cor- 

 respond to a low water-content. The freezing point 

 varied in one case from - 2.4° in midsummer to - 8° in 

 midwinter, and the subcooling point from about -6° to 

 -14°. The changes in freezing temperatures, and there- 



