THE FORMATION OF ICE IN PLANTS 243 



thermometer, or inserted .the latter into holes bored in potatoes, beet-roots, and the 

 like. The objects were then placed in freezing chambers, with the thermometer 

 scale projecting outside. In an experiment performed with a potato the air in the 

 chamber was at 4-0 to 4-5C., and the temperature of the tuber sank from i5C. 

 to 0-1 C. in one hour, and to 3-2C. in the next fifty minutes. Then ice-formation 

 began and in twenty minutes the temperature rose to 0-8 C., the true freezing-point, 

 at which it remained for an hour, and then gradually fell again. In an ivy leaf, the 

 formation of ice is not active enough to cause a rise to the true freezing-point, which 

 lies at i-5C., whereas sub-cooling may progress as far as 3-4 C. The method 

 of melting is not a well-adapted one to give the true freezing-point of a plant 1 . 



The temperature-curve indicates the progress of the formation of ice, and 

 it rises to the freezing-point after the "maximal degree of sub-cooling has been 

 reached. After remaining at the first freezing-point for some time it slowly 

 sinks as the sap is more and more concentrated and the freezing-point lowered. 

 At the same time the formation of ice takes place more and more slowly. The 

 following data obtained with a kohl-rabi leaf may serve as an illustration : 

 Critical point (of sub-cooling) .... 4-4 C. 

 Temperature four and a half minutes later . . 1-2 C. 



Amount of ice formed 6-69 per cent. 



three minutes later . . 2-25 per cent. 



Temperature then fell, reaching in fifty-six minutes 4-3 C. 

 Total amount of ice formed .... 41-3 per cent. 



The formation of ice does not take place so suddenly in a sub-cooled plant 

 as in a sub-cooled solution, and in many leaves the formation of ice begins at 

 particular points, and thence spreads slowly over the leaf. Whether the ice 

 formed outside the cells excites the solidification of the sub-cooled cell-sap is as 

 uncertain, as is the analogous question whether a crystal can excite crystallization 

 in a supersaturated solution through a membrane permeable by water but not 

 by the crystalline substance. 



The lowering of the freezing-point is directly proportional to the osmotic 

 concentration, and can therefore be calculated for any solution from the table 

 of osmotic values, having given that a decinormal solution of potassium nitrate 

 (i-oi per cent.) freezes at 0-308 C. 2 Hence the freezing-point of concentrated 

 solutions of colloids is only slightly lowered, so that crystals of ice form in a 10 per 

 cent, solution of gelatine just below o C. s 



Hence when the cell-sap is isosmotic with i or with 3 per cent, solutions of 

 KNO 3 , its freezing-point will be 0-3 or 0-9 C. respectively. The fact that 

 the actual freezing-point of the cell is lower than this is due to surface-tension 



relationships between freezing, sub-cooling, and cold-rigor, and his observations on insects yield 

 nothing new of importance. 



1 Miiller-Thurgau, Landw. Jahrb., 1880, p. 177. 



3 Ostwald, Lehrb. d. allgem. Chem., 1891, 2. Atifl., Bd. I, p. 752. A'ioi per cent, solution of 

 KNO 3 develops a pressure of i atmosphere, so that an osmotic pressure of I atmosphere represents 

 a lowering of the freezing-point by 0-088 C. 



3 Guthrie, Philosoph. Magazine, 1876, 5th ser., Vol. II, p. an; Sabanejew and Alexandrow, 

 Zeitschr. f. physikal. Chem., 1892, Bd. ix, p. 88. 



R 2 



