794 
Journal of Agricultural Research 
Vol. XXI, No. IO 
CHEMICAL, ANALYSIS 
Some of the isolated sandy crystals mentioned in the section on separa¬ 
tion, and similar to those photographed in Plate 138, A, were analyzed 
for lactose, without inversion, according to the Munson and Walker 
method, 1 described by Leach (4, p. 598). The average of three deter¬ 
minations gave 99.86 per cent calculated as lactose hydrate from Munson 
and Walker’s tables. It is evident that the quantity of sucrose appearing 
as sandy crystals is insignificant. 
DISCUSSION 
The results of our experiments upon a number of sandy ice creams 
lead us to conclude definitely that the sandiness is due to the crystalli¬ 
zation of lactose from a supersaturated solution. 
The suspicion, fostered by a number of persons in the dairy industry, 
that sucrose is responsible for this sandiness is not founded on scientific 
principles. The solubility, recorded by many investigations, of lactose 
in water at io° C. is about 17 gm. per 100 gm. of water. The solubility 
of sucrose at 12 0 is given as 198.6 gm. per 100 gm. of water. When these 
solutions are cooled they become supersaturated, provided they contain 
the above-mentioned quantities of the respective sugars. In sucrose 
solutions, however, we meet with the experience of the sugar technician, 
Prinsen Geerligs (5, jor), who tells us that sucrose solutions frequently 
act like glue at low temperatures and can be supercooled to a state 
in which they can not be poured. In the highly supersaturated, 
supercooled condition it is very difficult to induce them to crystallize. 
When accomplished it requires weeks and even months for the crystals to 
develop in size, particularly if much colloidal material, such as gum, 
pectin, or protein, is present in the solution. It required several months 
for the perfect crystal of sucrose shown in Plate 138, C, to develop even 
to this size. 
In the case of lactose, because of its low solubility in water, compared 
to sucrose, it is more difficult to form supersaturated solutions of any 
great magnitude. When its solutions are concentrated it crystallizes 
without much excitation. Indeed, it will persist in crystallizing even in 
the presence of complex protective substances which check crystallization 
of many other compounds. 
An ice-cream mix of 10 per cent fat, 12 per cent milk solids not fat, 
and 14 per cent sucrose contains, if we assume that the two sugars com¬ 
pete for only the water in the mix, about 22 per cent sucrose solution 
and about 10.5 per cent lactose solution. When we recall the fact that 
sucrose is about 10 times as soluble as lactose in water, we readily see 
that the physical conditions existing in frozen ice cream are unsuited to 
the crystallization of sucrose. 
1 Munson, L. S., and Walker, Percy H. unification of reducing sugar methods. In Jour. Amer. 
Chem. Soc., v. 28, no. 6, p. 663-686, 1906; v. 29, no 4, p. 541-554, 1907; Final tables in U. S. Dept. Agr. 
Bur. Chem. Bui. 107 (rev.), p. 241-251,1908; A correction, U. S. Dept. Agr. Bur. Chem. Circ. 82, 6 p., 1911. 
