Studies of modern famines have found disproportionately high mortality amongst adult men. The most commonly suggested root of this ‘female mortality advantage’ is biological, and it seems to be strongest when starvation is the main cause of death. The present study is the first to investigate the phenomenon in an early-modern society. Looking at the famines of 1597 and 1623 in northwest England, it finds some evidence for a female mortality advantage in 1623, but that this was concentrated in the first 12 months of the crisis (after the 1622 harvest). The female advantage was also much greater in north Lancashire and Westmorland than it was in the wealthier western Lancashire plain. Together this supports the idea that there was actual starvation during the 1623 crisis, at least in these areas at these times. There are, however, some reasons to suppose that the most mortal phase of the crisis, around the winter of 1623–1624, took place at a time when food was becoming more widely available, and hence should be attributed to diseases that followed the famine.