1. Moisture in the reagents affects the yield seriously because of the ease with which the intermediate imido ether is hydrolyzed. The hydrogen chloride must be thoroughly dried by passing it through two wash bottles containing sulfuric acid. If the wash bottles are not effective, a tube containing glass beads mixed with phosphorus pentoxide should be used after the wash bottles. It is well to protect the side arm of the flask by means of a calcium chloride tube. The absolute alcohol used should be at least 99.5 per cent (p. 249).

2. If commercial acetonitrile is used directly, the yields are poor. It should be dried over calcium chloride for at least a week, filtered, and distilled, the fraction boiling at 76–81° being used.

3. If a thinner flask is used, there is danger of its being cracked during the experiment.

4. If a proper safety bottle is placed between the drying train and the flask of acetonitrile, the treatment with hydrogen chloride does not require constant attention during this time.

5. When weaker solutions of ammonia were used, the yields were lower.

6. The grinding of the imido ether hydrochloride is very important, as lumps of this material become coated with ammonium chloride, which prevents the complete action of the ammonia.

7. By a similar procedure benzamidine hydrochloride can be prepared. Dry hydrogen chloride is passed into a cooled solution of 51.5 g. (51 cc., 0.5 mole) of benzonitrile in 25 g. of absolute ethyl alcohol until 21.3 g. of the gas is absorbed. The reaction mixture is allowed to stand for forty-eight hours, and the solid cake of imido ether hydrochloride is quickly crushed in a dry mortar. The solid is transferred to a liter flask, and an 8 per cent solution of dry ammonia in absolute ethyl alcohol containing 12 g. of ammonia is added slowly in small portions. The reaction mixture is shaken for twenty-four hours, then allowed to stand for forty-eight hours and filtered to remove the ammonium chloride which has precipitated. (During the shaking the solution may develop a faint pink color. This is probably due to the presence of traces of glyoxal in the alcohol for this material in basic solution furnishes colored condensation products with benzamidine.) The filtrate is allowed to evaporate to dryness in the open air, and the benzamidine hydrochloride thus obtained is dissolved in water. The solution is acidified with concentrated hydrochloric acid, decolorized with charcoal, filtered, and the filtrate is evaporated almost to dryness at room temperature. The crystals of benzamidine hydrochloride dihydrate are filtered and air-dried. The yield is 60–70 g. of material which melts at 70–73°. The filtrate on evaporation to dryness furnishes 15–30 g. of less pure product, melting at 74–82°, which contains a small amount of ammonium chloride but which can be used satisfactorily in most reactions. The total yield is 80–95 per cent. (A. R. Ronzio and J. B. Ekeley, private communication.)