In this connection the recent work of Allison and Katz (_J. Ind. Eng. Chem._ =11=, 336, [1919]) is of interest. They have designed an instrument, “the odorometer,” for measuring the intensity of odors in varying concentrations in air. It is based on the principle given above. A measured volume of air is passed through the liquid and then diluted to a given concentration. The mixture is then passed through a rubber tube with a glass funnel at the open end. Only one inhalation of the mixture is used to determine the intensity of the odor. The position of any strength of odor on the scale depends upon the sensitiveness and judgment of the operator, but with one person conducting the entire test, the results have been found quite satisfactory. (See tables on pages 360 and 361.) SKIN IRRITANTS Substances which seem useful for producing skin burns are studied both on animals and on man. Dichloroethyl sulfide (mustard gas) is used as a basis of comparison. Several methods are available. =Direct Application.= This method consists of the direct application of the compound itself to the skin, using a definite quantity (0.005 cc. or 0.005 mg.) over a definite area (5 square centimeters) of the skin. With such a quantity of mustard gas a rather severe burn on animals is produced. No precautions are taken to prevent evaporation from the skin since it is believed that in this way the test will approximate fairly closely the field conditions. =Vapor Tests.= Preliminary tests with vapors of volatile compounds are best made by placing a small amount of the material on a plug of cotton in the bottom of a test tube enclosed in a larger test tube which acts as an air jacket. After about an hour at room temperature the mouth of the test tube is applied to the skin. The concentration is not known, but one is dealing practically with saturated vapor. If an exposure of from 30 to 60 minutes produces no effect, one is safe to assume that the compound is not sufficiently active to be of value as a skin irritant. If quantitative results are desired, the apparatus shown in Fig. 119 is used. Dry air is blown through the bubbler, which is connected with a series of glass skin applicators. The concentration is determined in the usual way. The skin applicator consists of a small cylinder about 1.5 to 2 cm. in diameter and about 4 cm. long with a small glass handle attached on top. The opening is 1 cm. in diameter. When the concentration of the gas is constant, the exposure to the skin is made directly for any desired length of time. The skin irritant efficiency is judged by comparing the per cent of positive responses to approximately equal concentrations of the vapors, using mustard gas as a standard.