by many as suprrcoolcd li([iii(ls. This dctiiiit ion is uscrul, 

 in a sense, sinee it reminds one of tlie faet tliat i»lassi's re- 

 semble liquids in that they ])iesent a random arrauiivment 

 of theii" moleenles. But sueh a eoneeption ii>nores the faet 

 that, while a snpereooled litiuid is in so nnstabli' a state of 

 etinilibrium that the eontaet with a erystal invariably i)ro- 

 duees its erystallization, a ^lass ean be put in eontaet with 

 a erystal without losing* its stability. A snpereooled liquid 

 has been bronght from the zone L (Fig. l2J)) into the zone 

 (' of erystallization tem])eratures, where it is highly un- 

 stable; a glass has been brought, from the zone L, below 

 the zone of erystallization temperatui-es, into the zone V 

 where it is praetieally stable. The ehange in temperature 

 whieh leads to the formation of a glass is represented dia- 

 grammatieally by the upper arrow in Fig. 29, that whieh 

 leads to the formation of a snpereooled liquitl is re])rt'- 

 sented by the lower arrow. 



From what has been said it follows that a body ean erys- 

 tallize (/.('., freeze, to use the common term) only within 

 a limited range of temperatures (C and D, Fig, 29). In 

 the zone of the vitreous state (V, P^ig. 29) freezing is im- 

 l)ossible, the temj)erature being too low. Since the crystal- 

 lization range can be reached from above and from below, 

 there are two ways of freezing a body: one is to cool it 

 down from the liquid state, the other to warm it up from 

 the vitreous state. We are so accustomed to the idea that 

 freezing is more intense at lowei- temperatures, that the 

 statement which jirecedes appears paradoxical. 



As we shall see later, for aqueous colloids, the zone of 

 erystallization temperatures extends over some tens of 

 degrees below zero ; consequently these colloids cannot be 

 frozen (crystallized) if the tem])erature within them is 

 below this range. 



The reason why plants or animals freeze in nature, when 

 the atmos]iheric temperature drops to - 30*^ or -40", is 

 that the dro]) is too slow. The objects which are being 

 cooled remain for a few minutes at the dangerous (freez- 

 ing) temperatures. To avoid freezing, the temperature 



