AbstractFour‐day food intake reports, collected with electronic diaries, and saliva probes for cotinine analysis were obtained from 39 smokers and 43 nonsmokers. The nonsmokers had a higher caloric intake than the smokers, but this was related to more frequent sport activities. The distribution of energy intake across carbohydrates, fat, and protein was identical for both groups, but the smokers consumed bread, sweets and desserts, cheese, fruit juices, muesli (a cereal and yoghurt mixture), and water less frequently, but also more soft drinks, coffee, and alcohol. Only coffee and alcohol drinking were correlated with saliva continine. For two additional days, the recording of food intake was continued together with the continuous recording of activity and heart rate, intermittent measurements of blood pressure, assessments of subjective state, and performing a mental task. On one of these 2 days, the smokers had to abstain from smoking. Abstinence lowered heart rate and increased caloric intake by about 500 kcal, whereas actometer readings, blood pressure, subjective state, and mental performance were unaffected. Also unaffected were the percentage intake of fat, protein, and carbohydrates and the frequency of consuming particular food items. Post hoc correlations between subjective state, cotinine measures, excess intakes of total calories, macronutrients, and specific groups of food items were obtained only occasionally and in an inconsistent fashion. It is suggested that effects of smoking on taste and eating as a hedonic substitute when abstaining might represent important factors for food intake beyond the effects of nicotine. The new methods developed for this study appear to offer a way to obtain detailed information under particularly naturalistic conditions in studies on smoking and related problems. © 1994 Wiley‐Liss,