The H2 breath test is ideal for orocaecal transit time (OCTT) measurement, as it is non-invasive and inexpensive. Indigestible substrates added to a test meal are metabolised by the colonic bacteria, resulting in the production of H2 which is detected in end-exhalation breath. However, the substrates themselves can alter the transit times in the gastrointestinal tract. The aim of the present study is to compare OCTT and gastric emptying (GE) when lactulose in liquid (L-L), solid lactulose (L-S) and solid inulin (IN-S) are added to a test meal, and subsequently, to examine if inulin alters GE. Firstly, ten male volunteers were tested on three occasions. Volunteers ate a pancake breakfast containing 100 mg of 13C-octanoic acid and either 12 g of L-L, 12 g of L-S or 12 g of IN-S in a randomised order. Secondly, seven male volunteers were tested twice with meals containing either 12 g of IN-S or no substrate (NO-S). L-L induced the shortest OCTT (85·3 (sd 42·8) min) compared with L-S (162·4 (sd 62·6) min) and inulin (292·4 (sd 66·7) min; P = 0·007). GE half-time and lag phase (L-L: 61 (sd 9); L-S: 57 (sd 10); IN-S: 52 (sd 10) min; P = 0·005) were also affected, with L-L being the slowest. Thirdly, inulin reduced GE lag and latency phases (P < 0·05) compared with NO-S. Lactulose accelerates OCTT but delays GE compared with inulin. Inulin accelerates the onset of stomach emptying, but it has no effect on GE half-time. For these reasons, inulin is the preferred substrate for the H2 breath test.