Influenza A virus was discovered in 1933, and since then four major variants have caused all the epidemies of human influenza A. Each had an era of solo world prevalence until 1977 as follows: H0N1 (old style) strains until 1946. H1N1 (old style) strains until 1957, H2N2 strains until 1968. then H3N2 strains, which were joined in 1977 by a renewed prevalence of H1N1 (old style) strains. Serological studies show that H2N2 strains probably had had a previous era of world prevalence during the last quarter of the nineteenth century, and had then been replaced by H3N2 strains from about 1900 to 1918. From about 1907 the H3N2 strains had been joined, as now. by H1N1 (old style) strains until both had been replaced in 1918 by a fifth major variant closely related to swine influenza virus A/Hswine1N1 (old style), which had then had an era of solo world prevalence in mankind until about 1929. when it had been replaced by the H0N1 strains that were first isolated in 1933. Eras of prevalence of a major variant have usually been initiated by a severe pandemic followed at intervals of a year or two by successive epidemics in each of which the nature of the virus is usually a little changed (antigenic drift), but not enough to permit frequent recurrent infections during the same era. Changes of major variant (antigenic shift) are large enough to permit reinfection. At both major and minor changes the strains of the previous variant tend to disappear and to be replaced within a single season, worldwide in the case of a major variant, or in the area of prevalence of a previous minor variant. Pandemics, epidemics and antigenic variations all occur seasonally, and influenza and its viruses virtually disappear from the population of any locality between epidemics, an interval of many consecutive months. A global view, however, shows influenza continually present in the world population, progressing each year south and then north, thus crossing the equator twice yearly around the equinoxes, the tropical monsoon periods. Influenza arrives in the temperate latitudes in the colder months, about 6 months separating its arrival in the two hemispheres. None of this behaviour is explained by the current concept that the virus is surviving like measles virus by direct spread from the sick providing endless chains of human influenza A. A number of other aspects of the human host influenza A virus relationship encountered in household outbreaks are among the list of 20 difficulties that are inexplicable by the current concept of direct spread. Alternative concepts have usually been designed to counter particular difficulties and are incompatible with other features of influenzal behaviour in mankind. The new concept detailed in the appendix provides simple explanations for most if not all of the difficulties. It proposes that influenza A virus cannot normally be transmitted during the illness because it too rapidly becomes non-infectious in a mode of persistence or latency in the human host. Many months or a year or two later it is reactivated by a seasonally mediated stimulus which, like all seasonal phenomena, is ultimately dependent on variations in solar radiation caused by the tilt of the plane of earth's rotation in relation to that of its circumsolar orbit. The carriers, who are always widely seeded throughout the world population, become briefly infectious and their non-immune companions, if infected, comprise the whole of the next epidemic. The reactivated virus particles must encounter the immunity they have engendered in the carrier, thus allowing minor mutants an advantage over virions identical with the parent virus, and so favouring antigenic drift and automatic disappearance of predecessor and prompt seasonal replacement. Antigenic shift and recycling of major variants may also be explained by virus latency in the human host.