Analytical methods which are accurate to about 1 per cent have been developed for the determination of small amounts ( ca . 500 γ) of bis(ß-chloroethyl)-sulfide (H), ethyl-bis(ß-chloroethyl)amine (EBA), tris(ß-chloroethyl)amine (TBA), ß-chloroethyl-benzylsulfide (benzyl-H), and ß-chloroethyl-ethylsulfide (ethyl-H). The determinations are made by micro titration of the HCl liberated upon complete hydrolysis of the vesicants. A description is given of an apparatus suitable for applying vapors of vesicants to unit areas of skin. A very precise and reproducible micropipetting technique is described for the introduction of the vesicants into the penetration apparatus. By means of this penetration apparatus studies have been made of several factors which may influence the rate at which vesicant vapors penetrate into skin. Model experiments have been performed in which H was allowed to vaporize and the vapor was absorbed on a surface such as that of diethylene glycol or vaseline. It has been found that if the surface of liquid H is increased by spreading the agent on filter paper, the rate of evaporation is markedly increased. Furthermore, if the vapor is agitated by means of a magnetically driven fan, the rate of absorption by diethylene glycol is greatly accelerated. With vaseline as the absorbing surface it has been found that the area of the absorbing surface has an effect on the rate of absorption of H vapor. More H is absorbed by vaseline spread on filter paper to give a rough surface than is absorbed by a smooth film of vaseline. Measurements of the rate of penetration into human skin of H, EBA, TBA, benzyl-H, and ethyl-H vapors have been performed at 21–23°C. and 30–31°C. by means of the penetration apparatus described in this paper. The measurements were carred out on human volunteers under conditions of controlled temperature and humidity. When human skin is exposed to air saturated with H vapor, the H penetrates the skin of the forearm at a rate of about 1.4 γ per cm. 2 per minute (temperature 21–23°C.; relative humidity 46 per cent). This value was found to hold in experiments in which H vapor was applied for 3 to 30 minute intervals, thus indicating that the permeability of the skin to H vapor is not altered during a 30 minute exposure. Agitation of the H vapor by fanning did not result in any measurable increase in the rate of penetration. Two of the volunteers were Negroes; the permeability of their skin to H vapor did not differ appreciably from that found for the other subjects. When human skin is exposed to air saturated with EBA vapor, the vesicant penetrates at the rate of 2.8 γ per cm. 2 per minute (temperature 22°C., relative humidity 50 to 52 per cent). The amount of EBA penetrated is linear with exposure time for exposure periods of 5 to 20 minutes. Under similar conditions, it was found that TBA penetrates at a rate of about 0.18 γ per cm. 2 per minute (temperature 22–23°C.; relative humidity 45 to 48 per cent). This value was found to hold in experiments in which TBA vapor was applied for 30 to 60 minute intervals. The amount of TBA penetrated is linear with exposure time. In the case of benzyl-H, a linear relationship between the amount lost from the penetration cup and exposure time was also observed but the plot did not pass through the origin. It is suggested that this anomaly is due to retention on the skin surface of an appreciable quantity of benzyl-H as a result of rapid physical adsorption or chemical combination with a constituent of the skin. The rate of penetration of benzyl-H may be calculated from the slope of the plot and is found to be 0.35 γ per cm. 2 per minute (temperature 22°C., relative humidity 55 to 60 per cent). The results with ethyl-H showed great variation among individual subjects and no satisfactory value for the rate of penetration can be given as yet. Measurements were also made of the rate of penetration of H, EBA, and TBA vapors at 30–31°C. (relative humidity 47 to 49 per cent). At this temperature, a linear relationship was observed between the amount penetrated and the time of exposure. H vapor penetrated at a rate of 2.7 γ per cm. 2 per minute, EBA vapor at 5.1 γ per cm. 2 per minute, and TBA vapor at 0.29 γ per cm. 2 per minute. Three of the subjects in the EBA measurements were Negroes. The permeability of their skin to EBA vapor did not differ from that found for the white subjects. Despite this fact, their skin failed to vesicate after an exposure period twice that which caused 50 per cent vesication in the white subjects. Calculation of the precision of the data showed that the average deviation, standard deviation, and standard error were not appreciably different for the data obtained with human subjects as compared with data for control experiments in which human skin was not involved. Consequently, no significant differences in the rate of penetration into the skin of individual subjects can be discerned from the data presented in this communication. The increase in the rate of penetration of H, EBA, and TBA vapors from 21–23°C. to 30–31°C. is approximately proportional to the increase in volatility of each agent. These results indicate that at the same gas concentration in milligrams per liter, H, EBA, and TBA vapors would all penetrate at about the same rate. The data presented above permit a determination of the approximate amount of each vesicant which must penetrate to cause vesication in about 50 per cent of the exposed sites. This amount has been designated by the symbol V 50 . The V 50 for H and TBA at 21–23°C. is the same, being about 6γ; at 30–31°C., the V 50 is 4 to 5 γ. On the other hand, the V 50 for EBA at 22°C. and 30–31°C. is about 26 to 28 γ. Thus, per gamma penetrated, H and TBA vapors are about equally effective in producing vesication while EBA vapor is only ⅕ to ⅛ as effective.