When 0.5 ml of a suspension of E. coli (5 times 107 per ml) in the sterile filtrate of “late” strangulation fluid was injected intraperitoneally on mice, all the mice died within 24 hours. None of the mice injected with the same numbers of microbes suspended in “early” strangulation fluid filtrate or in a casein medium died during this period. E. coli cultured in media containing “early” or “late” strangulation fluid or the casein medium did not vary significantly in rates of growth or virulence. A direct effect on the bacteria by substances in the “late” strangulation fluid filtrate did not appear to represent a factor of importance for the virulence-enhancing property of the “late” filtrate.