Forty-two crossbred lambs (33·4 kg initial body weight; twenty-four wethers and eighteen ewes) were used in a 42 d experiment with a 2 × 3 factorial treatment arrangement to determine effects of forage level and source on splanchnic tissue mass. Diets were 250 and 750 g/kg of chopped lucerne (Medicago sativa) (A), ryegrass (Lolium multiflorum)–wheat (Triticum aestivum) (RW) or bermudagrass (Cynodon dactylon) (B) hay, with the remainder being maize-based concentrate. Five lambs per treatment were slaughtered at the end of the experiment and measurements made of internal organs and contents of the gastrointestinal tract (GIT). Digestible organic matter intake (DOMI) on the 7 d preceding slaughter was 0·89, 0·83, 0·90, 0·83, 0·77 and 0·61 (SE 0·05) kg/d, and live-weight gain was 0·20, 0·17, 0·18, 0·10, 0·10 and 0·07 (SE 0·02) kg/d for diets A-25, RW-25, B-25, A-75, RW-75 and B-75 respectively. Total GIT mass (fresh) was higher (P < 0·05) for 750 than 250 g forage/kg and for B than RW (4·80, 4·57, 5·55, 5·84, 5·99 and 6·91 kg for diets A-25, RW-25, B-25, A-75, RW-75 and B-75 respectively). Nonfat organic matter was 259, 295, 292, 303, 277 and 264 g for the total GIT; 93, 102, 103, 106, 95 and 97 g for the reticulo-rumen (forage level × type (diet A v. diets RW and B) interaction; P < 0·05); and 204, 196, 202, 177, 156 and 127 g for the liver (SE 10) with diets A-25, RW-25, B-25, A-75, RW-75 and B-75 respectively. In summary, differences in properties of forage A and the grasses at 250 g/kg diet may have influenced GIT mass independent of energy intake and digesta mass. Conversely, with 750 g dietary forage/kg, higher digesta mass for diet B than diet RW appeared responsible for high reticulo-rumen mass relative to DOMI. Greater digesta mass for 750 than 250 g forage/kg may have elevated intestinal tissue mass/DOMI with diets A and B but not with diet RW, for which NDF digestibility was highest.