_Diagnostic standards for the 1918 epidemic._—All cases of illness recorded on the reports, which have occurred during either the 1918 or the 1920 epidemics, or in the interval between them, have been put into four groups as regards diagnosis of influenza. Cases are designated as “Yes,” “Probable,” “Doubtful,” and “No.” Cases of illness occurring during the months of 1918 and 1919 in which influenza was epidemic and in which the patient remembers that he had the more definite symptoms, (fever, headache, backache, pain in the extremities, pneumonia) and in which he was sick at least three days and in bed at least one day, have been designated as “Yes.” The symptoms chosen are those most likely to be remembered. The individual frequently does not remember all. Statements of the absence of fever are often unreliable. Usually the headache, backache or pain in the extremities, or even all of these are well recollected. Cases occurring particularly during the epidemic period in which the more definite symptoms are unknown, but who were sick three days or longer and who were in bed at least one day, were probably influenza. This is particularly true if there were no other symptoms suggestive of some other definite disease. Such cases were designated “Probable.” Cases have been designated as doubtful when the evidence of illness falls short of the above desiderata. Cases of true influenza may fall into this group, either because of the extreme mildness of the symptoms and course or because of the inaccurate memory of the individual concerning the events of his illness sixteen months previously. Our results show that the group of doubtful cases is relatively very small and the number of true cases lost in this group will be negligible. One important reason for adhering to the above classification is that it corresponds closely with that used by Frost and Sydenstricker, so that our results may easily be compared with theirs. _Standards for 1920._—For 1920 the illnesses were so recent in the minds of the patients that we have required rather full information for making the diagnosis of “Yes.” For this designation certain symptoms are arbitrarily required. Certain additional symptoms, if present, serve to strengthen the diagnosis of influenza. The required symptoms are fever, confinement to bed for one day or more and at least two out of the following three, headache, backache and pain in the extremities. The additional symptoms which influence the classification are sudden onset, prostration, lachrymation, epistaxis, and cough. Cases designated as probable are those in which the symptoms as enumerated above are incomplete in one or more details, but yet in which the diagnosis of influenza would be justified. “Probable,” therefore, means that the case is to be accepted among the list of true influenza cases. This is particularly so when the case occurs during the epidemic period.