Chlorinating agents (chlorine, sulfur dichloride, etc.) rapidly transform mustard gas into an inactive (non-blistering) substance. Sulfur dichloride was a valuable reagent in both laboratory and works in “cleaning up” mustard gas. This reaction also explains why the early attempts to prepare mustard gas by the interaction of ethylene and sulfur dichloride were unsuccessful. Mustard gas is probably formed, but is almost immediately chlorinated by the excess of sulfur dichloride. Sulfur chloride on the other hand has no effect on mustard gas. Chloramine-T and Dichloramine-T (the valuable therapeutic agents introduced by Dakin and Carrel for treatment of wounds) also react with mustard gas. For this reason they were advocated as treatment for mustard gas burns. But as we will see later, they were not altogether successful. DETECTION At first the only method of detecting mustard gas was through the sense of smell. It was then believed that concentrations which could not be detected in this way were harmless. Later this proved not to be the case, and more delicate methods had to be devised. In the laboratory and in the field these tests were not very satisfactory, because most of them depended upon the presence of chlorine, and the majority of the war gases contained chlorine or one of the other halogens. The Lantern Test depended upon the accumulation of the halogen upon a copper gauze and the subsequent heating of the gauze in a Bunsen flame. This test could be made to detect one part of mustard gas in ten million parts of air. Another field detector devised by the Chemical Warfare Service consisted in the use of selenious acid. Here again the lack of specificity is apparent, for while certain halogen compounds did not give the test, arsine and organic arsenicals gave a positive reaction and often in a shorter time than mustard gas. [Illustration: FIG. 36.—Field Detector for Mustard Gas.] The Germans are said to have had plates covered with a yellow composition which had the property of turning black in the presence of mustard gas. These plates were lowered into the bottom of recently captured trenches and if, after a few minutes, they turned black, the presence of mustard gas was suspected. It is also stated that the characteristic yellow paint on the olive of the mustard gas shell had the same composition, and was useful in detecting leaky shell. According to a deserter’s statement, however, reliance upon this test resulted in casualties in several instances. A white paint has also been reported which turned red in the presence of mustard gas. This color change was not characteristic, for tests made by our Army showed that other oils (aniline, turpentine, linseed) were found to produce the same effect.