SANDY CRYSTALS IN ICE CREAM: THEIR SEPARATION 
AND IDENTIFICATION 
By Harper F. ZoLLER, Chemist, and Owen E. Williams, Dairy Manufacturing 
Specialist, Dairy Division, Bureau of Animal Industry, United States Department of 
Agriculture 1 
Ever since Sato ( 6) 2 isolated crystals of lactose, sucrose, calcium 
phosphate, calcium citrate, etc., from the crystalline sediment occurring 
in sweetened condensed milk, there has been considerable speculation 
as to which of these substances are responsible for most of the sandiness 
found therein. 3 
The ice-cream industry utilizes large quantities of condensed milk, 
particularly evaporated skim milk and sweetened condensed whole milk. 
The tendency during the last few years toward building up the total 
solids of the ice-cream mix with milk solids other than milk fat has 
created a demand for the products mentioned. With their use there 
arose simultaneously the problem of “sandy” ice cream. Not infre¬ 
quently wholesale quantities of ice cream develop this sandy or crystal¬ 
line texture—sometimes large, hard crystals, sometimes fine and gritty, 
like starch. This wide variation in the size of the crystals caused much 
perplexity as to their composition in view of the work cited above on 
condensed milk. Since ordinary ice cream contains from 12 to 16 per 
cent of sucrose it was but natural that many should believe the sandi¬ 
ness, especially the large crystals, to be due to this substance; the fmer 
might be due to any of the suspected substances. 
In order to settle this question of “what constitutes sandiness” we 
took up the problem of isolating the “sandiness” after it had formed 
and of identifying it. 
DEVELOPMENT OF SANDY CRYSTALS 
A large quantity of a high solids-not-fat ice cream was prepared in a 
regular commercial power freezer. The composition of the mix was as 
follows: Fat, 10 per cent; sucrose, 14 per cent; milk solids not fat, 12 
per cent. The raw products were ripened 40 per cent cream, evaporated 
skim milk, and pure cane sugar. The mix was frozen in the regular 
time with brine at about 10° F. (- 12.22° C.). It was immediately 
packed, after freezing, in 2-quart tinned cans and placed in the hardening 
< The authors express their indebtedness toG.L. Keenan and E. T. Wherry, of the Bureau of Chemistry’ 
to the former for his eourtesy in preparing the photomicrographs and to the latter for checking the crystal' 
lograpluc fca tures of lactose. ybUiI 
2 Reference is made by number (italic) to “Literature cited,” p 795 
/ U 7 Uld , be “77" <^ate the impression that Sato actually isolated sucrose crystals from all samples 
of condensed milk. In fact, the sucrose crystals were isolated only rarely from condensed milk, and in 
those cases he specifically states that they were condensed to an unusually low water content. 
Journal of Agricultural Research, 
Washington, D. C. 
zi 
Vol. XXI, No. 10 
Aug. 15, 1921 
Key No. A-Lx 
50937 °—21 - 7 
(791) 
