The present study proposes new bench-scale protocols for evaluating the flammability of bark and its involvement in resistance of trees to fire. Samples of cork from Algerian oak forests (Quercus suber) were selected for flammability testing. A mass loss calorimeter device, arranged in the standard horizontal configuration, was used to determine Heat Release Rate and temperatures. A calibrated epiradiator, arranged in a vertical configuration (to resemble field conditions), was used to determine lethal temperatures in living tissues and to assess the inter-device reproducibility of the data. Both protocols showed good repeatability and reasonable reproducibility. The time to reach lethal temperatures beside living tissues was more than two minutes in all cases and the average time was 230 s. The resistance of cork to fire increased with the thickness of the material, showing that trees in which the cork is less than 3 cm thick are most vulnerable to fire. The importance of corkback tissue in the flammability of cork is also highlighted, indicating important differences in the flammability of industrially processed cork and natural cork. The proposed protocols can be implemented using other devices (i.e. cone calorimeter) to obtain more information about the flammability of different types of tree bark, fire resistance and heat transfer during wildfire.