HOW COLD CAUSES DEATH 245 



allow for any expansion during gelation l . Even when the intracellular 

 formation of ice produces fissures in the tissues or ruptures in individual 

 cells, this only has the same effect as a localized injury. 



The injury or death is therefore due to changes or disturbances 

 produced in the protoplasm by the low temperature, and owing to the 

 dissimilar specific characters of different plants death need not always be 

 produced in exactly the same way. This is shown by the fact that some 

 plants are killed at temperatures above zero, while others are killed by the 

 formation of ice at temperatures which they can withstand if the formation 

 of ice is prevented. Other plants, again, are not injured if the formation of 

 ice does not progress too far, while some are resistant to all degrees of cold. 



Owing to the progressive withdrawal of water with decreasing tempera- 

 ture all plants which cannot withstand desiccation must ultimately be killed 

 by cold. In the case of most plants more than half the water is frozen 

 at 3 to -8 C., and hence it remains to be explained why certain plants 

 incapable of withstanding desiccation can resist temperatures of from 2oC. 

 10 30 or even 5oC. If at these temperatures the whole of the free water 

 is frozen, it is evident that the local withdrawal of water by freezing acts 

 differently to its total removal by desiccation. It has, further, still to be 

 determined whether bacteria killed by drying are resistant to extremely 

 low temperatures. 



Without doubt all plants kept permanently frozen would ultimately 

 die, so that death can hardly be due in all cases to the withdrawal of water, 

 as Muller and Molisch suppose 2 , although frequently this may be the 

 primary cause of death. Many seeds capable of withstanding desiccation 

 are in fact killed when frozen after soaking in water, and the high resistance 

 to cold shown by desiccated objects is simply a special case of a general 

 phenomenon. Since we are here dealing with biological adaptations, it is 

 possible that tropical plants subject to desiccation but not to low tempera- 

 tures may be found to afford instances of plants killed by slight cold, but 

 nevertheless resistant to desiccation. 



Muller and Molisch consider the behaviour of the above-mentioned 

 seeds to be due to the sudden withdrawal of water when freezing follows 

 sub-cooling. Under certain circumstances rapid changes are actually 

 injurious, but it appears that rapid thawing and freezing exercise little or 

 no special injurious effect. Furthermore, seeds seem to be unaffected by 

 the most rapid drying, or absorption of water. Muller has indeed himself 

 shown that the potato and beet-root are also killed when sub-cooling is 

 avoided and the formation of ice takes place gradually. It is uncertain 



1 A rise of pressure in the cell will drive out water through the permeable cell-wall until 

 equilibrium is restored. 



2 Muller-Thurgau, Landw. Jahrb., 1886, Bd. xv, p. 534 ; Molisch, Das Erfrieren der Pflanzen, 

 1897, p. 534. This conclusion is as one-sided as that ascribing all death by heat to proteid-coagulation. 



