The stools of the convalescents were examined and no man was discharged from hospital until his stools were negative for dysentery bacilli upon 3 successive tests in fourteen days. Isolation of the bacilli from convalescents was obtained in 40 patients only for periods under fourteen days while with 27 others such carrying of bacilli lasted from two weeks to one month. As the dysentery bacillus does not invade the blood stream we do not find it in the urine so that to a certain extent the dysentery bacillus carrier is less dangerous than the typhoid one. There have been reports of isolation of Flexner and “Y” type bacilli from monkeys and rabbits but there is nothing to indicate that any other host than man is of importance. Flies are undoubtedly of as much importance in the spread of bacillary dysentery as of typhoid. The possibility of infection through the medium of soiled clothes, sent out for washing, is to be thought of. There have been several instances of transference of the disease by the water supply. In times of war, with large forces of soldiers, bacillary dysentery tends to become the most important disease encountered by military surgeons. During the Civil War there were 285,000 cases of dysentery in the Federal army. It is possible that infectious material may be disseminated as dust and thus contaminate food. PATHOLOGY Injection of dysentery bacilli into the peritoneal cavity of guinea pigs causes a muco-sanguinolent diarrhoea with congestion of and haemorrhage into the caecum. There is also a haemorrhagic peritoneal exudate. In the rabbit lesions similar to those in man are obtained as well as paralysis of the limbs. It is therefore thought that there are two toxins concerned in the pathology of bacillary dysentery, one a neurotoxin which may cause a peripheral neuritis or joint trouble and the other a toxin which acts on the lower bowel, especially the caecum, with the production of congestion and coagulation necrosis of the mucosa. Cases have been reported where the adrenals showed congestion and necroses, as if subjected to the action of a toxin. In man we have an acute inflammation of the mucosa of the large intestines and, in the tropics, we frequently find the lower third of the ileum involved as well. In amoebic dysentery the process rarely extends beyond the ileo-caecal valve. A catarrhal process with hyperaemia and sero-purulent exudate is first noted, to be succeeded by fibrin formation in the mucosa, a process of coagulation necrosis. When the process invades the ileum there is no involvement of Peyer’s patches. Virchow noted the greater intensity of the process in the region of the rectum, sigmoid flexure and ileo-caecal valve.