COPYRIGHT　©　2015　by　the　Editorial　Office　of　Chinese　Journal　of　Gastroenterology.　Background:　As　the　empirical　studies　on　human　body　are　restricted　extremely,　the　establishment　and　selection　of　suitable　animal　models　are　important　for　researches　on　ulcerative　colitis　(UC).　Aims:　To　compare　the　symptoms　and　colonic　pathology　of　rat　models　with　experimental　colitis　induced　by　dextran　sulfate　sodium　(DSS)　and　trinitrobenzene　sulfonic　acid　(TNBS),　so　as　to　provide　a　reference　for　selecting　animal　models　in　UC-related　studies.　Methods:　Drinking　4%　DSS　freely　for　7　days　or　intrarectal　administration　of　single　dose　100　mg/kg　TNBS-50%　ethanol　were　used　to　establish　experimental　colitis　model　in　Sprague-Dawley　rats.　The　disease　activity　index　(DAI)　was　assessed　dynamically　during　the　course　of　experiment.　The　whole　colon　was　removed　in　batches　for　measurements　of　colonic　damage　score　and　activity　of　myeloperoxidase　(MPO)　at　different　time　points.　Results:　The　DAI　score　reached　the　peak　at　the　7th　day　and　the　2nd　day　in　DSS　group　and　TNBS　group,　respectively,　and　decreased　gradually　afterwards.　Six　and　one　deaths　occurred　during　the　experimental　course　in　DSS　and　TNBS　groups,　respectively.　In　DSS　group,　the　duration　of　inflammation　was　short,　the　colonic　injury　was　moderate　and　recovered　after　drug　withdrawal.　At　the　18th　day,　the　colonic　damage　score　and　MPO　activity　was　0.25±0.50　and　(0.80±0.33)　U/g,　respectively,　and　no　significant　differences　were　seen　between　DSS　group　and　normal　control　group.　In　TNBS　group,　the　duration　of　inflammation　was　longer　and　the　colonic　injury　was　more　severe.　At　the　21st　day,　the　colonic　damage　score　and　MPO　activity　was　3.60±0.55　and　(1.60±0.39)　U/g,　respectively,　and　chronic　inflammation　was　observed　histologically.　Conclusions:　Both　DSS　and　TNBS　can　induce　experimental　colitis　model　in　rats.　The　course　of　TNBS-induced　colitis　model　presents　a　transformation　of　acute　to　chronic　inflammation,　and　may　be　more　suitable　for　treatment-related　studies　of　UC.