Objective: The aim of the study was to evaluate two different self-help methods as intervention among pregnant smokers. (In Norway, 32% still smoke in the 18th week of pregnancy, averaging 10 cigarettes per day). Design: A randomized controlled experimental trial. Methods: Three hundred and nine pregnant smokers were randomized into three groups with similar back-ground characteristics. At the ultrasound examination around the 18th week of pregnancy Group 1 received ''A Pregnant Woman's Self-Help Guide to Quit Smoking,'' a 40-page manual giving day-to-day instruction for 10 days, with detailed programs for quitting on the 7th day and exercising maintenance on the remaining days, including a reminder for every pregnancy visit. At the same time, Group 2 received a 20 half-page brochure containing information and illustrations about smoking hazards in pregnancy with a 10-day quitting program. Group 3 was the control group. Data was obtained by questionnaire answered at the ultrasound examination around the 18th week of pregnancy and at delivery. Results: Of the two manuals, only the more comprehensive one used in Group 1 produced significant results in that more women quit (12%) and reduced their smoking (55%) compared with the other two groups combined, by 4% and 48%, respectively (P < 0.01). Conclusions: It is concluded that the more comprehensive self-help manual, comprising personal instruction and repeated support, proved more effective than the smaller brochure distributed without face-to-face instruction. The impact on smoking cessation by self-help manuals only affected the light smokers (less than 10 cigarettes per day). To obtain better results in the high-risk group, intensified multicomponent intervention seems necessary.