(a) Rigid quarantine for all persons suffering from the disease and all contacts. During the epidemic quarantine was advocated by many people. It was pointed out that the disease spread most rapidly in camps, in ships, and in quarters generally where large numbers of persons were closely associated; that it was quite as contagious and more rapidly fatal than most diseases which are regularly quarantined; that while it was admitted that there is no laboratory method to make certain the diagnosis, and no method of telling how long convalescents are capable of transmitting the disease, as there is, for instance, in diphtheria, still there is no question of the value of the arbitrary quarantine used in measles, scarlet fever and smallpox, all of which are diseases in which the parasitic causes are not known. Further, the opinion was expressed that complete isolation and quarantine would not only protect the community from influenza, but that it would also in a measure protect the patient from contact with numerous outside strains of pneumococci and streptococci, and so lessen secondary infection and reduce the general mortality. There are many reasons why quarantine is not applicable in epidemic influenza. Most important of all is probably the inability to make certain the diagnosis, especially during the early stages in light cases. This would work detrimentally in several ways. Really ill patients would delay calling a physician until late, for fear of unnecessary quarantine. Many needless and unjust quarantines would result when the diagnosis was uncertain and the physician anxious to carry out quarantine measures efficiently. Many patients would have contacts running about and infecting their neighborhoods while a delayed diagnosis was being made. Influenza was so contagious during the epidemic that it would have necessitated general quarantine not only of all infected persons but also of all contacts to have obtained any favorable results, and since nearly everyone was either a patient or a contact, all lines of business would literally have been paralyzed by the procedure. If it is true that the infected person is most dangerous to others before he has developed symptoms himself, he is a carrier impossible of detection and control. Points in favor of the hypothesis that infected persons spread the disease before they develop symptoms are found in the following facts. As the disease passed from community to community officials became alert for the appearance of the first case. In army barracks and in large institutions it was often possible to determine the first case at its development. The case was, in many instances, removed at once and isolated, but I have seen no instance in which such a measure was successful in curbing the disease. As subsequent cases appeared they were likewise immediately removed, but the cases continued to spread just the same. Bloomfield (38) cited the