260 H. S. H. WARDLAW. 
they were in equilibrium, but with the same milk freed from 
fat in the centrifuge. This was done because the freezing 
point of spun milk is more easily determined than that of 
the same milk still containing fat. It has already been 
shown (loc. cit.) that the freezing points of whole and spun 
milk are practically the same. 
The milk on which all the work described in the present 
paper was done contained 1% of toluol as already stated. 
It was thought that the presence of the toluol might have 
some effect on the freezing point of the milk, but the follow- 
ing determinations of the value of A for spun milk (a) 
without toluol, (b) containing 5% of toluol, show that this. 
is apparently not the case. 
Spun Milk Alone. | Spun Milk + Tolwol. 
0°552° 0°556" 
The effect of the toluol, if any, is thus small. 
All these determinations of freezing point were carried 
out in the manner previously described (loc. cit.). The 
agreement between the freezing points of the correspond-. 
ing liquids is within the limit of accuracy of the method 
there set down. Hach value of / given is the mean of at. 
least three determinations having an extreme difference 
of not more than about 0°005°. 
During the course of a dialysis the volume of liquid put. 
into the celloidin sac does not remain constant, but. 
diminishes, as the following results show. 
Volume of liquid in celloidin sac before and after completion of dialysis. 
Experiment. | Original volume. | Volume of dialysate. 
55 25 cc. 18:6 ce. 
56 25, Lora 
58 25%) 16275 
These figures were obtained for the dialysis of spun milk; 
they show the osmotic effect which occurs before the con- 
