Human adenovirus type 3 (HAdV-3) respiratory infections occurs worldwide in both children and adults, leading to severe morbidity and mortality, particularly in the paediatric age group and especially in neonates. During HAdV infection, neutralizing antibodies are formed against the epitopes located in the hyper variable regions (HVRs) of the hexon protein. These neutralizing antibodies provide protection against reinfection by viruses of the same type. Therefore it is reasonable to speculate that variations of HAdV-3 in the HVRs could impair the immunity acquired by previous infection with a different strain with variation in its HVRs. HAdV-3 has recently become the major agent of acute respiratory infection worldwide, being responsible for 15% to 87% of all adenoviral respiratory infections. However, despite the increased prevalence of HAdV-3 as respiratory pathogen, the diversity of hexon proteins in circulating strains remains unexplored. This study was designed to explore the variation in HVRs of hexon among globally distributed strains of HAdV-3 as well as to discover possible relationship among them, thus possibly shedding light on the cause for the increased prevalence of HAdV-3. In this study, for the first time we analysed the hexon proteins of all 248 available strains of HAdV-3 from the NCBI database and compared them with those of the HAdV-3 prototype (GB stain). We found that the HVRs of HAdV-3 strains circulating worldwide were highly heterogeneous and have been mutating continuously since -their original isolation. Based on their immense heterogeneity, the strains can be categorized into 25 hexon variants (3Hv-1 to 3Hv-25), 4 of which (3Hv-1 to 3Hv-4) comprises 80% of the strains. This heterogeneity may explain why HAdV-3 has become the most prevalent HAdVs type worldwide. The heterogeneity of hexon proteins also shows that the development of a vaccine against HAdV-3 might be challenging. The data on hexon variants provided here may be useful for the future epidemiological study of HAdV-3 infection.