1. Feeding tests on monkeys (Macacus rhesus), dogs and cats are unsatisfactory for detecting the presence of enterotoxin, owing to the variable susceptibility of these animals by the oral route.
2. Using Dolman's method, in which the material is injected intra-peritoneally into kittens, the production of enterotoxin has been demonstrated by: (a) sixteen out of thirty-eight strains of Staph. aureus, isolated from cases of acute or chronic mastitis or from normal udder milk; (b) four out of five strains of Bact. coli, mostly from calves with “white scours”. No enterotoxin was obtained from fifteen strains of Str. agalactiae from mastitis in cows.
3. The formation of enterotoxin under natural conditions has been observed: (a) In udder milk seeded with Staph. aureus or naturally contaminated with that organism and stored at atmospheric temperatures (18 and 22°C.). The substance remains active in cheese prepared from such milk. (b) In layer cake made with cream naturally contaminated with Staph. aureus.
4. A small outbreak of poisoning due to potted meat paste was shown to be caused by a non-haemolytic Staphylococcus.
5. A few feeding experiments on man with milk or cream, in which food-poisoning staphylococci had grown, were negative, but on one occasion a Staphylococcus from a case of mastitis yielded a culture filtrate which caused symptoms of food poisoning.
6. Enterotoxin has the following properties. It is resistant to heat (95°C., 30 min.), to low concentrations of formalin sufficient to destroy the haemolytic toxin, to acid (pH 5·0), and to rennet, but is destroyed by trypsin.
It diffuses freely into the culture medium but only slightly through collodion. It is antigenic. Its properties are such that enterotoxin can be classed as a bacterial exotoxin.