277 



assumption which is ' not in asreemont with what is 

 known on the rate of format ion of ice in terms of tem- 

 perature. 



Some of these fundamental objections were resumed by 

 Mes f 190.5). This author w^onders why dehydration should 

 be harmful when caused by freezing and harmless oth- 

 erwise. Concerning- the formation of ice at tempera- 

 tures far below^ the freezing point, he quotes his experi- 

 ments with ImpaUens stems in which he claims to have 

 shown by the shape of the freezing curves that congel- 

 ation is completed at -6-. Cooling at a low^er tempera- 

 ture, therefore, w^ould not induce any further desicca- 

 tion and one does not see how it w^ould be injurious, 

 except if one admits the theory of the ''specific mini- 

 mum ' '. 



According to Mez, several experiments in which death 

 was attributed by Molisch to freezing and interpreted 

 as identical wdth death by desiccation (water algae, 

 staminal hair of Tradescantia, potato) represent really 

 cases of death by desiccation and not of death by freez- 

 ing. 



Apelt (W07), supporting the argument of Mez, says 

 that in his experiments on potato tubers he ahvays found 

 that the death point was definitely belo^v the freezing 

 point. This fact is interpreted as signifying that death 

 is not due to a desiccation during freezine- 



According to the same author, the dehydration theory 

 has against it that, in one experiment on potato, re- 

 peated freezing (more than 4 or 5 times) at a temperature 

 slightly above the death temperature resulted in death. 

 Since it is assumed that the amount of water congealed 

 in each freezing at the same temperature is the same, 

 one does not see how^ the repetition of the experiment 

 could become injurious. 



Gorke (1907) brought forth a new argument in favor 

 of the dehydration theory by showing that proteins can 

 be precipitated by freezing when the salt concentration 



