138 THE ALUMNI JOURNAL 



concentrated, and specially free from nitric and nitrous acids. Pumice, 

 in small pieces. 



Procedure. — 0.5 to i.o c.c. of the sample of alcohol or of an alco- 

 holic distillate is placed in a 6 inch test tube, i c.c. of the chromic 

 acid solution added, and the liquid diluted with water to 4 to 5 c.c. 

 Two or three small pieces of the pumice are dropped in, the test tube 

 connected with the condenser (see figure) and the liquid distilled into 

 another test tube by boiling briskly over a small flame. When 3 to 4 

 c.c. of the liquid have passed over or when only about 0.5 c.c. remains 

 behind in the first test tube, the condenser is detached and rinsed, with 

 about 2 c.c. of distilled water, into the receiving test tube. To the 

 distillate are added: i drop of ferric chloride solution, 2 drops of 

 albumin solution, and after mixing, 4 to 5 c.c. of the pure sulphuric 

 acid are poured in, slowly and carefully as a layer, generation of 

 much heat being avoided. The zone of contact is then observed, with- 

 out disturbing the liquids, against a white background. A sharply 

 defined, violet zone appears almost at once if the proportion of methyl 

 alcohol is above 5 per cent. With smaller amounts but more than i 

 per cent., the color shows within one minute ; several minutes will 

 be required for the color to appear with less than i per cent, of methyl 

 alcohol. Pure ethyl alcohol, treated in this manner, gives no color. 

 When organic impurities (other than methyl alcohol) are present 

 which cannot be removed, a yellow to reddish color is often obtained, 

 but not violet. The violet color is intensified on standing or warming, so 

 that the entire liquid becomes colored violet to lavender if the test 

 tube be placed in boiling water for some time. Dilution with water 

 will diminish the intensity of the color, which is not affected by glycerin 

 or 50 per cent, sulphuric acid. 



Albumin in strong solutions gives with concentrated sulphuric acid 

 a variety of color reactions. It must not be used in excess, and only 

 traces of ferric chloride should be added to obtain perfect blank tests. 

 Various ferric salts have been recommended, but trials have shown 

 the chloride to be the best suited. This is in agreement with the con- 

 clusions of Shrewsbury C^^). 4 to 5 drops will be found a sufficient 

 quantity of milk for the test. Commercial sulphuric acid, as used 

 by Hehner (^-) was found unsuitable; the acid which happened to be 

 tried gave no reaction with a weak formaldehyde solution, with which 

 a pure acid reacted. An investigation showed that traces of nitric or 

 nitrous acid will affect the reaction to a marked degree and that in 



