The nucleotide sequence of the mRNA encoding the fusion (F0) protein of a virulent strain of Newcastle disease virus was determined. A single open reading frame in the sequence encodes a protein of 553 amino acids with a calculated molecular weight of 59058. The amino acid sequence predicted several structural features involving the fusion-inducing hydrophobic stretch (residues 117-142) and the cleavage-activation site (residues 112-116) to generate the disulfide-linked F1 and F2 subunits. The cleavage-activation site as well as a part of the fusion-inducing sequence were compared among a series of virulent and avirulent strains by the chain-termination method using a synthetic oligonucleotide primer. It was found that without exception, the cleavage-activation site of virulent strains consisted of two dibasic residues with an intervening glutamine, Arg-Arg-Gln-Arg-Arg, whereas the corresponding region of avirulent strains was made of a sequence with single basic residues scattered among uncharged residues, Gly-LysArg-Gln-GlySer-Arg. On the basis of these observations and the previous results showing a strict correlation between the pathogenicity and the cleavability of the fusion protein of NDV (Y. Nagai, H-D. Klenk, and R. Rott, Virology, 72, 494-508, 1976), we propose the importance of the dibasic residues for efficient proteolytic activation of the fusion protein and for the pantropic property of NDV. Some strains were found to have Leu-Ile-Gly as the N-terminus of F1, whereas others contained Phe-Ile-Gly, indicating that Phe-X-Gly is not always conserved at F1 N-terminus of paramyxovirus.