The object of this investigation was to ascertain the value of the bicoloured guaiac reaction on the cerebro-spinal fluid in mental hospital practice. The reaction, which was first described by de Thurzo (i), is similar in principle to the Lange gold sol test, in that under certain conditions precipitation occurs from a colloidal system. Its originality depends on the fact that two dyes, naphthol green and brilliant fuchsin are contained in the system, one of which attaches itself to the precipitating colloid (brilliant fuchsin), whilst the other (naphthol green) remains in the supernatant fluid. It is claimed that it is not so susceptible to possible fallacies as the gold sol test (such as chemical uncleanliness), that it is possibly more selective in its action, and that it has the same practical value in the laboratory diagnosis of neuro-syphilis. Results have been reported on hospital cases (2) showing fair agreement with the Lange and Wassermann tests, but so far no records are available showing the value of this test in a series of neuro-syphilitic cases from mental hospitals, with the exception of a brief report on 10 cases (3). The present report deals with 325 fluids obtained from cases in L.C.C. mental hospitals, and these included 125 cases of general paralysis treated by malaria and other pyrexial therapies.