Methane emissions from geological reservoirs may have played a major role in the sudden events terminating glaciation, both at the start of the Bølling/Allerød and also at the end of the Younger Dryas. These reservoirs include Arctic methane hydrates and also methane hydrate stored in offshore marine sediments in tropical and temperate latitudes. Emissions from hydrate stores may have resonated with tropical wetland emissions, each reinforcing the other. Because methane is such a powerful greenhouse gas, much smaller emissions of methane, compared with carbon dioxide, are required in order to have the same short-term impact by climate forcing. The methane-linked hypothesis has much geological support from sea-floor evidence of emission. However, Greenland ice-core records have been interpreted as showing methane as a consequential factor, rather than the leader, of change. This interpretation can be challenged on the grounds that temperature gradients in Greenland ice record local changes and local timing of a step-like shift in weather fronts, while methane concentrations record changes on a hemispheric and global scale. There are large remaining hydrate reservoirs in the Arctic and in shelf sediments globally, and there is substantial risk of further emissions.