155 



One can see, by the preceding analj^sis, that, if the solu- 

 tion is dihite, the curve will stay practically horizontal for 

 a long time since a relatively large quantity of water has 

 to freeze before the freezing point is lowered noticeably. 

 With very concentrated solutions, on the contrary, the 

 liorizontal portion of the curve might be so small that a 

 sharp determination of the freezing point would be dif- 

 ficult. 



At the temperature of congelation of the saturated solu- 

 tion, that is, at the eutectic point (C in the Figure), the two 

 phases, solvent and solute, crystallize. The curve becomes 

 parallel to the time axis (portion CD). For the analysis 

 of this portion which is a freezing curve, we refer to what 

 has been said in the preceding section on the freezing curve 

 of water. A slight complication results, however, from the 

 fact that there are 3 phases present, the liquid solution, the 

 solid solvent and the solid solute, the latter two crystalliz- 

 ing separately, though at the same time. 



It should be mentioned also that, for dilute solutions, the 

 quantity of eutectic mixture is so little as compared to the 

 quantity of ice present that the eutectic plateau almost 

 vanishes while the freezing plateau is large. On the con- 

 trary, for concentrated solutions, in which, as we said 

 above, the freezing plateau disappears, the eutectic plateau 

 becomes considerably broader. 



3. Freezing Curves of Colloids. Little is known on the 

 forces which have to be overcome to freeze water out of 

 colloids. The following investigation is one of the few 

 that we found in the literature, on this subject. 



Fischer and Bobertag (1909) described the curious fact 

 that, when gelatin was dissolved in water by a previous 

 heating, the freezing curve so obtained had a longer hori- 

 zontal plateau than the freezing curve of a suspension of 

 the same quantity of gelatin (about 9%) dispersed in 

 flakes in the water (Fig. 19). A further investigation into 

 the cause of this phenomenon might yield valuable in- 

 formation on the mechanism responsible for the anomalies 

 of similar nature so often met with in the freezing of 

 tissues. 



