that there has been a significant change in policy since 1936, when Levine recommended a period of six to eight weeks in bed, with an appropriate convalescent period. Alternatively, it is possible that there is an increasing occurrence or recognition of milder cases. While most would acknowledge the considerable psychological benefits which are seen in a patient who is able to resume work, there are very few data concerning the physical effects of resumption of work, and it is impossible to answer the oftput questions, " Will I do myself any harm if I go back to work? " or " Should I have another month's rest from work? " The data presented in this series do not provide an answer to these questions except with respect to the mortality (the crudest index) which compares favourably with that recorded in doctors (Morris, Heady, and Barley, 1952 ; McMichael and Parry, 1960)-the mortality at one year is 4.2 %, compared with 4% of the doctors (who survived the first month after infarction), and at three years is 14%, compared with 10% in the doctors' series. Cook, Tuttle, and Kodlin (1962) report from a cardiac workevaluation centre an attempt to establish whether work had any effects on the progress of heart disease. They graded the work capacity of their patients into five categories ranging from normal capacity to confinement to bed. By comparing their assessment at the initial visit with the actual work being carried out when the patients were seen at six-monthly intervals they were able to classify the men into those who complied with the original advice, those who did more than recommended, and those who did less. The mortality of patients suffering from "atherosclerotic heart disease" showed a gradient from 9% in those who did excessive work, to 5 % in those who complied with advice, and to 3% in those who did less work than advised. Further studies are needed to confirm if hard work has any adverse effect on those who have suffered from cardiac infarction. The possible role of psychological stresses has not been considered as it was not possible to measure them.