40 PROCEEDINGS OF THE CANADIAN INSTITUTE. 



A paper read by R. W. Moore, A.B., M.S., before the American 

 Chemical Society in September of last year, and published in the 

 " Analyst " for December, shews the very great superiority of 

 Reichert's process for the identification of cocoa-nut oil used as an 

 adulterant of butter-fat, and was the immediate cause of my under- 

 taking the work described in the following pages. My friend, the 

 late lamented H. Sugden Evans, F.C.S., Chief Analyst for the 

 Dominion, requested me to examine the process with a view to 

 ascertaining the limits of accuracy obtainable by it under fixed 

 conditions. 



I prepared from a genuine butter, pure, dry fat, by allowing the 

 salt and water to separate by subsidence in a tall beaker on the 

 water bath at about 80° to 90*^ C, pouring off the svipernatant fat, 

 and fi'eeing from suspended caseine, etc., by filtration (using a steam 

 jacket). The fat so obtained had a specific gravity of 0.914 at 100° 

 Fah. (water at 100° Fah. = 1). 



The liquid fat was weighed by diffei'ence from a light glass tube, 

 having a lip, directly into the flasks used in saponification. 



Tlie difficulty of weighing exactly 2.5 grams of fat suggested 

 the use of an approximation to that quantity, and the subsequent* 

 calculation into volatile acidity per gramme of fat used. 



1 employed an approximately normal solution of alcoholic potash 

 for saponification, the alcohol having a specific gravity of 0.823 

 (equivalent to 90 per cent, anhydrous alcohol). The saponification 

 was effected in stoppered flasks of strong glass (the stoppers being 

 tied down), on the water bath at 100° C. ; and 10 c.c. alcoholic pot- 

 ash was used for each gram of fat taken. One hour was allowed 

 for the completion of this process. 



The melted soap was decanted into a flask of about 250 c.c. 

 capacity, about 2-3 c.c. of dilute alcohol used to rinse, and 20 c.c. of 

 approximately normal sulj)liuric acid (accurately 0.98 N. ) added for 

 each gram of fat used. During the subsequent distillation the flask 

 was placed on a sand bath, and a pretty strong heat supplied by a 

 Berzelius's alcohol lamp. 



A tube leading to the bottom of the flask permitted a current of 

 air to be aspirated through the whole apparatus. A 24-inch LieVjig 

 was used and kept well cooled. The distillate was collected in a 

 graduate which was connected with a Richard's filter pump, by 



