The influence of headspace flushing on the specific methane (CH4) production of blank samples with just inoculum in Biochemical Methane Potential (BMP) tests was studied. The three most common ways were applied: flushing with nitrogen (N2) gas, flushing with a mixture of N2 and CO2 (80/20 v/v), and no flushing. The results revealed that removing the oxygen is crucial to avoid aerobic respiration, which caused both hindered activity of methanogens and loss of methane potential. Furthermore it was demonstrated that 20% of CO2 in the flush gas increased significantly the methane production by over 20% compared to the flushing with pure N2. In order to mimic the same headspace conditions as in full-scale treatment plants, using a flush gas with a similar CO2 concentration as the expected biogas is suggested.