Sea ice plays a critical role in structuring ecosystem dynamics throughout the Scotia Sea (SS) region, and variations in ice extent are hypothesized to affect predator populations in this area directly. A paradigm guiding recent research in the Western Antarctic Peninsula (WAP) region of the SS, the “sea-ice hypothesis”, suggests that declines in the seasonal extent and duration of sea ice, owing to climate change, have led to declines in “ice-loving” Adélie penguin (Pygoscelis adeliae) populations, while “ice-avoiding” chinstrap penguin (P. antarctica) populations have increased (1-5). However, 30 years of field studies in the WAP, coupled with more regional surveys throughout the Scotia Sea, refute this hypothesis; both Adélie and chinstrap penguin populations in this region are declining dramatically. Here, we present evidence supporting an alternative, more robust, hypothesis that explains both increases and decreases in penguin abundance as a result of changes in abundance of Antarctic krill (Euphausia superba) in the SS region. Linking trends in penguin abundance with trends in krill biomass can explain why populations of both Adélie and chinstrap penguins increased after seals and baleen whales were over-harvested (6) and, more recently, are decreasing in response to climate change. Thus, while the “sea-ice hypothesis” predicts that chinstrap penguins will benefit from climate change, the “krill-biomass hypothesis” leads to the contrasting prediction that chinstrap penguins are highly vulnerable to the current regime of climate warming.