H₂S H₂S N=CR / | + + C₆H₄ | \ | NH₂ NHCOR C=N / / | C₆H₄ + (RCO)₂O → C₆H₄ SH \ \ CN CN The hydrogen selenide used in the reaction was prepared from FeSe by the action of hydrochloric acid, or by heating paraffin and selenium, in the proportion of four to one respectively, at 335° to 350°C(54). The selenoquinazoline was prepared from anthranilic nitrile by the following methods--the anthranilic nitrile being prepared from o-nitraniline(57), (a) 20 grams of acetyl-anthranilic nitrile was dissolved in absolute alcohol, and dry hydrogen selenide and dry ammonia passed into the solution for three hours. The quinazoline crystallized out gradually on cooling was filtered out and recrystallized from dilute alcohol. The yield was about ten per cent. (b) 10 grams of acetyl-anthranilic nitrile was heated in a sealed tube at 110° with alcohol saturated at zero degrees with dry hydrogen selenide and dry ammonia. After five hours, the tube was taken out and the quinazoline crystallized out on cooling. Yield was about sixteen per cent. As hydrogen selenide was somewhat unstable and did not dissolve freely in alcohol, freshly prepared sodium selenide was used in the following method and was found to be more satisfactory. It was prepared from sodium hydroxide in absolute alcohol by passing dry hydrogen selenide into the solution for about three hours. In the beginning and end of the reaction, nitrogen was used to exclude the oxygen of the air. The selenide was collected and dried in an atmosphere of nitrogen, and then in a vacuum, in presence of phosphorus pentoxide. When thus prepared, sodium selenide was colorless, but on exposure to air it turns reddish and finally dark colored. The C. P. selenide on the market was black and was found to be entirely useless. (c) 20 grams of anthranilic nitrile and fifty grams of sodium selenide were mixed and heated in a distilling flask in an atmosphere of nitrogen, and forty grams of acetic anhydride dropped into the flask very slowly. The temperature was kept at 115° for half an hour and then raised to distill off the acetic acid formed in the reaction, as the condensation hardly went to completion in the presence of any trace of acetic acid. The whole process took an hour and a half. The flask was removed from the oil bath and, after cooling, dilute alkali was run in, in successive portions, to dissolve out the quinazoline. Into the clear alkaline extracts carbon dioxide was bubbled for an hour, and common salt then added. The precipitate was recrystallized several times from twenty-five per cent. alcohol. The yield was from twenty to twenty-five per cent.