An experimental study of hydrate formation has been performed, using a simulated natural gas-a. mixture of methane, ethane, and propane in a 90:7:3 molar ratio-and a liquid large-molecule guest substance (LMGS), which provides guest molecules to fit into the 51268 cages of a structure-H hydrate. Except for the use of the gas mixture, the hydrate-forming procedure used in this study was the same as that tested in our previous studies [Ohmura et al., Energy Fuels 2002, 16, 1741-1147; Tsuji et al., Energy Fuels 2004, 18, 418-424], i.e., spraying liquid water downward through a gas phase onto a liquid-LMGS layer lying on a pool of water under a prescribed temperature-pressure condition (275 K, 2.9 MPa) in a chamber into which the gas mixture was being supplied to compensate for its loss due to hydrate formation. We selected three LMGSs for comparison: tent-butyl methyl ether (TBME), which was found to give the highest rate of hydrate formation when pure methane is used as the guest gas [Tsuji et al., Energy Fuels 2004, 18, 418-42,1], 2,2-dimethylbutane (neohexane), and methylcyclohexane (MCH). The rate of hydrate formation from the gas mixture was determined to be increased by the presence of any of these LMGSs, compared to the rate observed in the absence of any LMGS. An unexpected fact was found in the rate of hydrate formation from the gas mixture plus an LMGS, compared with the rate observed with pure methane plus the same LMGS under the same temperature-pressure condition. That is, the former rate may be much higher or, on the contrary, appreciably lower than the latter rate, depending on the species of the LMGS used. Accordingly, it turns out that the rate of hydrate formation from the gas mixture plus neohexane or MCH is higher than that from the same mixture plus TBME, in strong contrast to the nature of hydrate formation from pure methane plus an LMGS that we previously revealed.