The study here reported was undertaken to determine whether it is possible by the use of the available methods of skin testing to anticipate the skin eruptions and other signs of hypersensitiveness that not infrequently occur with the therapeutic use of dinitrophenol. Without consideration of the controversial subjects of allergy, atopy or anaphylaxis and without referring to the none too clear conceptions of antigen antibody reactions as applied to nonprotein substances, I wish to submit the results of my experiment. METHODS The three possible methods of skin testing were employed for the direct test, and the intradermal method alone was used for the indirect test. The Patch Test.—Ten milligrams of sodium dinitrophenol powder 1 was placed on the flexor surface of the forearm or of the back and covered with waxed paper, held in position with a wide strip of adhesive plaster. Readings of the reactions were recorded at