Rates of smoking relapse remain high, despite the wide availability of cessation aids. Presumably factors such as craving, withdrawal symptoms, and smoking urges are key contributors to relapse, but empirical support for this presumption is not conclusive and is complicated by the high variability in symptoms across individuals and time, as well as by the lack of an absolute symptom threshold for response. Data were analyzed from 137 female smokers, aged 18-40 years, who completed 30 days of a protocol for a longitudinal smoking cessation trial. Subjects were assigned a quit date and followed regardless of subsequent smoking status. At baseline, subjects completed written measures of nicotine craving, withdrawal symptoms, and smoking urges. They also completed these measures daily for 30 days, beginning on their quit date, Scores were standardized within subjects and graphed to identify temporal symptom patterns. A total of 26 women quit smoking and 111 relapsed (at least one cigarette puff). The intensity of subjects' craving, withdrawal, and smoking urges Factors 1 and 2 peaked on the day of relapse by an average of 1.4, 1.1, 1.2, and 1.1 standard deviations, respectively, with symptoms rising during the previous 2-5 days and dropping precipitously over the 2 days subsequent to relapse. Additionally, women who relapsed had higher absolute (unstandardized) symptom scores on their quit day than those who were abstinent for 30 days. These findings imply that escalation of withdrawal symptoms, craving, and smoking urges during a quit attempt may contribute to smoking relapse. Frequent symptom monitoring might be clinically important for relapse prevention.