Butyrate is recognised as efficient in healing colonic inflammation, but cannot be used as a long-term treatment. Dietary fibre that produces a high-butyrate level when fermented represents a promising alternative. We hypothesised that different types of dietary fibre do not have the same efficiency of healing and that this could be correlated to their fermentation characteristics. We compared short-chain fructo-oligosaccharides (FOS) and type 3 resistant starch (RS) in a previously described dextran sulfate sodium (DSS)-induced colitis model. Seventy-two Sprague–Dawley rats received water (control rats) or DSS (50g DSS/l for 7d then 30g DSS/l for 7 (day 7) or 14 (day 14) d). The rats were fed a basal diet (BD), or a FOS or RS diet creating six groups: BD-control, BD-DSS, FOS-control, FOS-DSS, RS-control and RS-DSS. Caeco-colonic inflammatory injuries were assessed macroscopically and histologically. Short-chain fatty acids (SCFA) were quantified in caeco-colon, portal vein and abdominal aorta. At days 7 and 14, caecal and distal macroscopic and histological observations were improved in RS-DSS compared with BD-DSS and also with FOS-DSS rats. Caeco-colonic SCFA were reduced in FOS-DSS and RS-DSS groups compared with healthy controls. The amount of butyrate was higher in the caecum of the RS-DSS rats than in the BD-DSS and FOS-DSS rats, whereas distal butyrate was higher in FOS-DSS rats. Partially explained by higher luminal levels of SCFA, especially butyrate, the healing effect of RS confirms the involvement of some types of dietary fibre in inflammatory bowel disease. Moreover, the ineffectiveness of FOS underlines the importance of the type of dietary substrate.