Number average molecular weight (Mn) of saturates (CL), which recovered from Cold Lake oil sand bitumen by column chromatography (JPI-5S-22-83), was estimated from a chromatogram obtained by gaschromatographic distillation (GCD, ASTM D 7169), and compared with Mn measured by cryoscopy and gel permeation chromatography (GPC). When Mn is estimated from a chromatogram obtained by GCD, the following conditions should be satisfied. The first, a chromatogram returns to baseline, and the second, the correlation between boiling point and molecular weight was known prior to the estimation. The second condition is filled with good linear correlations between the boiling points and the logarithm of the formula weights for alkanes and alkylbenzenes, respectively. For CL, the first condition was regarded to be filled because the chromatogram returns very close to the baseline. The Mn was estimated to be 400 using a calibration curve based on alkylbenzenes, because its H/C atomic ratio (1.78) was closer to alkylbenzenes than to alkanes. Cryoscopy is a more basic method for Mn measurement, because no calibrations are required. Here, the Mn was measured by cryoscopy using about 10 mL of p-Xylene solution with the concentrations of 17.2, 8.63, and 1.57 g/kg. As a result, Mn was calculated to be 370. In addition, a conventional gel permeation chromatography (GPC) method was performed, and the Mn was estimated to be 490. Comparing those three methods, GCD method is the most convenient when a sample meets the conditions described above. However, only saturate fraction can be applicable at a moment. Cryoscopy is applicable for any fractions in principle, but it requires relatively large sample amount (c.a. 150 mg), and it is available for a sample with Mn up to about 500. Mn obtained by GPC using polystyrene standards was most inaccurate among the methods described here. It is necessary to adopt new standards other than polystyrene for more reliable molecular weight measurement.