The single-breath carbon monoxide diffusing capacity (DLCOsb) was measured together with ventilatory lung function tests as part of a survey of a general population sample living in Northern Italy (n = 2,481). Based on answers to an interviewer-administered questionnaire, subjects free of respiratory symptoms or diseases were identified. Data from subjects who had never regularly smoked cigarettes were used to derive reference equations for the test indexes, and data from the remaining subjects who had smoked were used to derive regression equations incorporating a term expressing cigarette consumption (cube root of pack-years) and a term indicating current smoking decrement, in order to obtain expected DLCOsb percent predicted. Neither number of cigarettes smoked daily or duration of smoking, in smokers, nor duration of smoking or years since quitting smoking, in ex-smokers, entered significantly the multiple-regression model. The mean values of DLCOsb were only slightly affected by the increasing degree of airway obstruction. When subjects with confirmed asthma were analyzed, after stratifying for different levels of FEV1/FVC ratio, increased mean value of DLCOsb (over 100%) was found in those with an FEV1/FVC ratio between 75 and 65%. This cross-sectional analysis suggests that there is a decrease in DLCOsb with cumulative cigarette consumption even in healthy subjects. Further, it confirms the clinical observations of high DLCOsb values in asthmatic patients, at least in those with an initial degree of chronic airflow obstruction.