Since 1970, WHO has conducted a collaborative study on the use of "excess mortality" from respiratory diseases (that is, the number of deaths actually recorded in excess of the number expected on the basis of past seasonal experience) in order to assess in 13 different countries the severity of influenza epidemics. The use of computer-produced seasonal expected and actual curves permits quick visual assessment of influenza activity in any one country, as well as comparisons between different countries. The study demonstrates that an excess in the observed over the expected number of total deaths does not necessarily indicate an excess in deaths from influenza, and it is therefore limited, at least for the present time, to deaths from respiratory disease. It provides a week-to-week record of deaths from acute respiratory disease in countries where weekly returns are available and a retrospective analysis of the disease pattern in the collaborating countries. This study will be continued for a number of years and, apart from its value to the individual countries, should provide useful comparisons between countries with different climates.