Certain sewage effluents have been found (Allen, Blezard, and Wheatland, 1946) to become highly toxic when treated with doses of chlorine much smaller than are required to give residual chlorine detectable by the ortho-tolidine test. Evidence from laboratory tests suggested that this toxicity was mainly due to a compound formed by interaction of the chlorine with small quantities of thiocyanate derived initially from gas liquor admitted to the sewage and surviving treatment at the sewage works. Key & Etheridge (1934) has shown that oxidation of the thiocyanate in a percolating filter or an activated-sludge plant, depends on the establishment of the necessary bacterial flora, and that a sudden increase in the concentration of thiocyanate in the sewage may result in a proportion of it being discharged in the effluent. Admission of gas liquor to sewage is common practice, and chlorination of such effluents would have serious effects on the living organisms, including fish, in surface waters to which they were discharged. Further work has revealed the nature of the compound responsible for toxicity, and has established the relation between toxicity and concentration of the substance in solution. An experimental study has been made of conditions affecting its formation during chlorination and of means of removing it from solution.