A study was made of the dietary habits and intake of energy and nutrients of 127 elderly people aged between 65 and 95 (60 men and 67 women). The method employed involved the recording of food consumption over a period of five days. Subjects were divided into two groups according to body mass index: those who were overweight or obese (0) (BMI > or = 25 Kg/m2), who made up 60% of the experimental population, and those with normal bodyweight (NW) (BMI < 25 Kg/m2) (40%). The percentage of subjects who took no breakfast was somewhat higher amongst members of group 0 (5.6% compared to 3.9% of NW subjects). Further, the breakfast of 0 subjects was less varied, smaller and contributed less to daily energy intake and the covering of theoretical energy expenditure than did the breakfasts of NW subjects. 0 subjects also took less time over their breakfasts. More than half (58.8%) of NW subjects took what might be considered a satisfactory breakfast, i.e. this meal provided more than 20% of the total energy intake. Only 29.6% of 0 subjects took breakfasts of similar energetic value. NW subjects consumed greater quantities of sugars and fruits at breakfasts and also more carbohydrates, fibre, vitamin E and iron than did 0 subjects. The shorter length of time spent at breakfast by 0 subjects, their consumption of smaller and less varied breakfasts and the composition of the same, could indicate the existence of less healthy breakfast habits amongst the overweight and obese. This could be the reflection of inadequate food habits all round, or be a conditioning factor prompting 0 subjects to make poorer food choices for the rest of the day. Though some people omit breakfast in an attempt to control their weight, this is probably an inadvisable practice. The control of bodyweight might be more efficiently accomplished by taking an adequate breakfast and better distributing total calorie intake across the day, avoiding its concentration into one or two meals.