Food volume could influence both the portions that people take and the amount that they eat, but these effects have had little investigation. The influence of food volume was tested by systematically reducing the flake size of a breakfast cereal so that the cereal was more compact and the same weight filled a smaller volume. In a crossover design, 41 adults ate cereal for breakfast once a week for 4 weeks during 2011 and 2012. The cereal was either standard wheat flakes or the same cereal crushed to reduce the volume to 80%, 60%, or 40% of the standard. A constant weight of cereal was provided in an opaque container and participants poured the amount they wanted into a bowl, added fat-free milk and noncalorie sweetener as desired, and consumed as much as they wanted. Results from a mixed linear model showed that as flake size was reduced, subjects poured a smaller volume of cereal, but still took a greater amount by weight and energy content (both P values <0.0001). Despite these differences, subjects estimated that they had taken a similar number of calories of all versions of the cereal. They ate most of the cereal they took, so as flake size was reduced, breakfast energy intake increased from a mean±standard error of the mean of 286±18 kcal to 358±19 kcal, an increase of a mean±standard error of the mean 34%±7% (P<0.0001). These findings demonstrate that variations in food volume associated with the size of a food's individual pieces affect the portion served, which in turn affects energy intake.