In a recent report (Topley, Wilson and Lewis, 1925) we described some observations on the rôle of the Twort-d'Herelle phenomenon in the epidemic spread of mouse-typhoid. In the experiments there recorded the lytic filtrate was administered by the mouth. The results obtained did not suggest that such administration exerted any appreciable influence on the spread of disease. It was shown, in a separate series of experiments, that intraperitoneal inoculation of such a filtrate afforded a marked degree of protection against subsequent inoculation of living cultures of B. aertrycke, provided that sufficient time had elapsed to allow of an immunising response, that no immediate protection was afforded by the inoculation of the lytic filtrate, and that neither immediate protection, nor subsequent immunisation, was afforded by the administration of the lytic filtrate by the mouth.