792 Journal of Agricultural Research voi.xxi.No.io 
box. This box was insulated with wood and held a hardening mixture 
of ice and rock salt. The ice cream was repacked each day in this box. 
On the fourth day very fine sandy crystals appeared. The crystals 
were developed by allowing the cans to become “heat shocked” (by 
setting them in contact with the air of the room for about 20 minutes) 
every other day after the sandiness first became perceptible. 
The dimensions of the crystals were observed frequently through the 
low-power objective of a compound microscope. When the sandiness 
first became perceptible to the tongue and between the fingers, the 
crystals were clearly visible in the microscope. The general form of the 
crystals remained the same throughout 20 days’ observation, although 
they grew much in size during this interval. 
When the crystals had grown to a sufficient size they were separated 
from the parent mass. Plate 137 illustrates the appearance of the crystals 
in the frozen ice cream described above, when a drop of the ice cream 
was placed upon a glass slide and then viewed through the microscope 
in polarized light. The crystals showed plainly the natural maize¬ 
shaped form of lactose hydrate. 
SEPARATION OF THE CRYSTALS 
The sandy crystals were separated from several different preparations 
of ice cream by the following method: Frozen cream was allowed to 
melt at the temperature of the laboratory, about 25 0 C. When com¬ 
pletely melted it was poured into large centrifugal tubes of 100 cc. 
capacity and centrifuged at 2,000 revolutions per minute for 10 minutes. 
The fat and liquid layers were poured from the sediment; the liquid 
portion was now entirely free from sandiness, the latter having been 
thrown down in the form of sediment. A drop of the sediment examined 
on the microscopic slide seemed to be made up entirely of crystals of 
one form—some large and some small. 
Before the crystals were proved to consist entirely of pure lactose 
they were isolated from the sediment for the purpose of holding any 
sucrose or lactose crystals in the crystalline form. This was effected 
by saturating water with all the sucrose and lactose that it could hold 
at 2 0 C. It was then warmed to 5 0 , and the sediment in each tube was 
thoroughly shaken with about 75 cc. of it. Then the tubes were again 
centrifuged. This operation was repeated with the addition of 50 per 
cent acetone to the washing fluid. The product was further freed from 
liquid on a Buchner funnel. 
The crystals were obtained in a pure white condition. More than 
yi kilo was obtained in this manner. They were dried to constant weight 
on fdter paper in a boiling-water oven. Plate 138, A, illustrates the 
appearance of these isolated crystals. They show plainly the effect of 
the washing mixtures but retain their characteristic maizelike or wedge¬ 
like form. 
