One of gerontology's most intense and ongoing debates centers around the future shape of the human survival curve. Is average life expectancy destined to peak between the ages of 85 and 90, as some would argue, or will new ways be found to sustain the dramatic increase in life expectancy that unfolded during the 20th century? This article considers issues inherent in projecting the number of tomorrow's older people, with a focus on the future course of mortality. Although overall mortality declined in most countries during the past century, the decline has been irregular and has reversed in some countries because of unforeseen epidemics, cause-specific changes, and sociopolitical upheavals. On the positive side, scientists have identified several promising findings; mortality trajectories at very old ages may decline, old-age mortality can be altered by social interventions, the number of centenarians is exploding, new chemical entities have a demonstrable effect on life expectancy, and there has been a linear increase in worldwide record-setting life expectancy during the past 160 years. New projections made by the United Nations peer, for the first time, 300 years into the future. These projections foresee a rise of about 25 years in life expectancy at birth in North America over the next 3 centuries, which some experts today believe to be quite conservative. It is important to recognize the inherent uncertainty in life expectancy projections and how assumptions of future mortality change must-from a policy perspective-be considered in conjunction with potential concomitant changes in fertility and migration.