Abstract We are in accord with Smith in the belief that operation within forty-eight hours of onset carries too high a mortality to warrant much consideration, especially in the wards of a public charitable hospital whose patients are not infrequently poor surgical risks. Furthermore, these patients often do not come to the hospital within forty-eight hours of the acute onset. Our opportunities have been limited as regards statistics for this group. Our entire Group 1 series has little positive value since it is too small in numbers. We are inclined to the view of Cutler and Whipple that any early operation, preferably cholecystectomy, should be done within six to twelve hours of onset. So far as we have been able to determine, apparently the peak of the disease is reached in from one to four days. In Group 2, after forty-eight hours, seven acute, ten subacute, five ulcerative, four gangrenous, two empyemas and one hydrops were present; in other words, considerable acute pathology remained. The chronic cases, almost a third, showed subsidence of the pathology. Nevertheless the mortality for this group had dropped by 60 per cent, from the 19 per cent of Group 1 to 7.8 in Group 2. In Group 3, in which operation occurred after six days (most of them about the fourteenth day), we still find evidence of acute gall-bladder pathology. There was evidence of an active pathology in 130 cases out of 427 supposedly subsided cases. In all of these we waited, and in the majority of instances they showed minimal clinical manifestations when operated upon. Yet acute lesions were present in 31 per cent of cases. The mortality, however, dropped from 19 per cent in Group 1 and 7.8 per cent in Group 2 to 5.4 per cent in Group 3. There were nine deaths in Group 3 due to accidental injuries of ducts and blood vessels. These deaths represented almost 40 per cent of the total mortality of this group; if these nine deaths were eliminated the mortality would be 3.5 per cent.