The complete coding region of hemagglutinin genes from 26 influenza A viruses of H9N2 subtype isolated from chicken flocks in China during 1996–2001 was amplified and sequenced. Sequence analysis and phylogenetic studies of H9N2 subtype viruses on the basis of data of 26 viruses in this study and 71 selected strains available in the GenBank were conducted. The results revealed that all the mainland China isolates showed high homology (94.1%–100%) and were assigned to a special sublineage in the major Eurasian lineage, in contrast to the high heterogeneity of Hong Kong SAR isolates. All the 29 mainland China isolates and six Hong Kong SAR strains also had the following common characteristics: sharing the same sequence of proteolytic cleavage site with one additional basic amino acid, RSSR, with only two exceptions; having the same amino acid motif of the receptor-binding site, YWTNV/ALY; 23 of 28 isolates bearing seven potential glycosylation sites and the remaining five having six; and sharing characteristic deduced amino acid residues Asn-183 at the receptor-binding site and Ser-130 at the potential glycosylation site. We concluded that the H9N2 subtype influenza viruses circulating in chicken flocks in China since the 1990s and Ck/HK/G9/97-like viruses isolated in Hong Kong SAR should have a common origin, whereas Qu/HK/G1/97-like viruses including human strains isolated in Hong Kong SAR might originate from other places. The available evidence also suggests that the H9N2 viruses of special lineage themselves and factors prone to secondary infections may contribute to the widespread and dominant distribution of viruses of this subtype in chicken flocks in China and other Asian countries.