Accidental Poisoning of Children

amount of dilution of the figures due to erroneous diagnoses occurring during the non-epidemic period. Information about the presence of influenza virus A infections during the trial period was largely provided by the laboratories of the Public Health Laboratory Service, whose staff, as can be seen from the reported printed at page 1178, worked closely with clinicians and medical officers of health. Particular credit must be given to the general practitioner " spotters," who must have found it difficult to observe the exacting precautions necessary in obtaining specimens for virus tests. In previous years detailed epidemiological conclusions on the origin and course of epidemics in this country have necessarily been somewhat speculative, because of the lack of laboratory information. This lack of information was particularly unfortunate in 1951, when viruses of two different groups were present and probably caused illnesses with different degrees of severity. The figures of the Ministry of National Insurance and Pensions provide an accurate measure of the spread and extent of influenzal infections, but without support from the laboratory these figures lose much of their value from the epidemiological point of view. Many of the problems of influenza epidemiology will probably remain unsolved until much more detailed information of the type collected by the Public Health Laboratory Service has been accumulated and correlated with information from other sources. An alternative approach to the epidemio.logy of influenza was described by Professor C. H. Stuart-Harris at the meeting of the Royal Society of Medicine already mentioned. He tested sera which had been collected for other purposes for their content of antibody to influenza virus soluble antigens, and showed the value of this method in tracing the appearance and progress of the virus in a community. This method might avoid some of the inevitable defects of an investigation such as that conducted by the Public Health Laboratory Service, but to provide useful information many specimens of sera would need to be tested. None of the methods used at present for the study of influenza epidemiology can give the whole picture, which is likely to emerge only when every source of information is used to the best advantage.