Prior attempts to make and test vaccines against the parainfluenza viruses (PIV), epidemiologic studies, and new information about the viruses’ surface glycoproteins provide the background for new vaccine development. The parainfluenza viruses are somewhat different from one another, so some aspects of their biology must be considered separately. This is particularly true of their epidemiology. PIV-1 and PIV-2 have similar epidemiologic patterns: they tend to be epidemic in the fall and early winter, usually appearing every other year and causing croup in children between one and four years of age. Infections usually do not occur in the first six months of life. PIV-3, on the other hand, behaves epidemiologically more like respiratory syncytial virus (RSV). Infections are common in the first six months of life, and large winter epidemics do not occur; this virus is largely endemic, with periodic epidemicity. Not much is known about PIV-4, but it is thought to be considerably less important than the other three types, and thus will not be considered here.

The parainfluenza viruses are species in the Paramyxoviridae family. Considerable information exists about the two surface glycoproteins of paramyxoviruses. One, the HN protein, contains both hemagglutinating and neuraminidase activity (Scheid et al., 1972). These two activities may occupy separate sites on the HN molecule (Portner, 1981). In vitro studies of the other surface glycoprotein have shown that it has the capacity to fuse membranes and is responsible both for the formation of syncytia and for entry of the virus into the cell (Scheid and Choppin, 1974). Antibody to either glyco-