228 Messrs. Dorée and Gardner. Origin and  [Feb. 10, 
thanks to Professor Hobday for the care and attention bestowed on the animal. 
while there. 
The feeces collected during each diet period were dried in the water oven, 
roughly powdered, and extracted thoroughly in a Soxhlet’s apparatus with 
ether. The extract in dilute ethereal solution was then saponified by sodium 
ethylate, and the precipitated soaps filtered off and well washed with ether. 
The ethereal filtrate so obtained was repeatedly shaken with water containing 
soda, to get rid of alcohol, some colouring matter, and excess of soap. It 
was then dried and the ether distilled off. The unsaponifiable matter differed 
widely with different diets in quantity and in character, being in some cases 
a viscid oil, and in others solid. . 
We hoped to be able to separate the cholesterol from this unsaponifiable 
matter by conversion into the dibromide, and taking advantage of the sparing 
solubility of the latter substance in petrol, but preliminary experiments 
showed that with the small quantities of cholesterol obtained the method was 
of no value from a quantitative point of view. We found it better to 
crystallise fractionally from absolute or from 85 to 90-per-cent. alcohol. The 
cholesterol was identified by microscopical examination, by the melting point 
after recrystallisation, and by conversion into the benzoate and the acetate. 
It was found better to convert the smaller crops of crystalline matter at 
once into benzoate, as this substance is very sparingly soluble in absolute 
alcohol. 
The acetate of cholesterol is readily prepared by boiling cholesterol with 
acetic anhydride and sodium acetate for a few minutes, pouring into water, 
and recrystallising from alcohol. The benzoate can be prepared by heating 
to 140° for half an hour or more with benzoyl chloride, treating the product 
with hot alcohol, and recrystallising from alcohol. We find, however, that it 
is a much better method in the case cf cholesterol to dissolve this substance 
in pyridine in the proportion of about 1 gramme to 10 to 20 ac. add a 
solution of a little more than the theoretical quantity of benzoyl chloride in 
pyridine solution, and allow to stand for several hours. The liquid is then 
poured into water, when the benzoate of cholesterol is precipitated, and may 
be recrystallised from absolute alcohol. The yield by this method is quanti- 
tative. The acetate of cholesterol melts at 114° C., and on cooling shows 
colour changes if pure. The benzoate melts at 145°5 C. to a turbid liquid, 
which suddenly becomes clear at 178° to 180°, and, in cooling, a brilliant display 
of opalescent colours is exhibited, among which a brilliant blue, appearing 
at about the temperature of the higher melting point, followed by a violet 
blue just before complete solidification, are most prominent. These colour 
changes are well marked and very characteristic of cholesterol. 
