In this novel study, we have for the first time identified evolutionarily conserved capsid residues in an individual chronically infected with norovirus (GGII.3). From 2000 to 2003, a total of 147 P1-1 and P2 capsid sequences were sequenced and investigated for evolutionarily conserved and functionally important residues by the evolutionary trace (ET) algorithm. The ET algorithm revealed more absolutely conserved residues (ACR) in the P1-1 domain (47/53, 88 %) as compared with the P2 domain (86/133, 64 %). The capsid P1-1 and P2 domains evolved in time-dependent manner, with a distinct break point observed between autumn/winter of year 2000 (isolates P1, P3 and P5) and spring to autumn of year 2001 (isolates P11, P13 and P15), which presumably coincided with a change of clinical symptoms. Furthermore, the ET analysis revealed a similar receptor-binding pattern as reported for Norwalk and VA387 strains, with the CS-4 and CS-5 patch (Norwalk strain) including residues 329 and 377 and residues 306 and 310, respectively, all being ACR in all partitions. Most interesting was that residues 343, 344, 345, 374, 390 and 391 of the proposed receptor A and B trisaccharide binding site (VA387 strain) within the P2 domain remained ACR in all partitions, presumably because there was no selective advantage to alter the histo blood group antigens (HBGA) receptor binding specificity. In conclusion, this study provides novel insights to the evolutionary process of norovirus during chronic infection.