Influenza viruses are recognized for their ability to cause serious infection and disease, in all age-groups of people. These annual influenza epidemics vary in severity, and are attributable to the way in which these viruses undergo rapid genetic evolution. Up until late 1997, this century~s historical record suggested that only H1, H2 and H3 subtypes could affect humans. However this assumption came unraveled in the face of the H5N1 outbreak in Hong Kong in which 18 people became infected with an H5N1 avian (chicken) influenza virus. Six of these cases were fatal. While it is recognized that avian and swine (pig) viruses can infect man without reassortment of the viral genes, the nature of the human cases in Hong Kong aroused great concern that we were possibly facing the beginning of a pandemic. The transmission of H5N1 influenza viruses from domestic poultry to humans emphasizes the need to have information available on all influenza subtypes circulating in the animal reservoirs of the world. It is well established that the aquatic birds of the world perpetuate each of the known subtypes of influenza viruses. However, it is not known which subtypes and variants are dominant. It is essential for the control of pandemic influenza that we implement systematic and continuous surveillance for influenza viruses in birds and animals of the world, particularly in Asia, and production and storage of viruses, reagents and technologies to facilitate rapid characterization and exchange of information. It is also essential to understand the fundamental differences between the Eurasian and American lineages of influenza viruses, the molecular changes that occur when influenza viruses transmit to other hosts and the dominance of particular subtypes in their natural reservoirs. If we can gain knowledge on these issues, we might be better prepared to respond rapidly to an emerging influenza pandemic.