Several studies have previously examined the effects of smoking and abstaining from smoking on working memory (WM) but have yielded inconclusive findings, which may be due to the tasks used (e.g., n-back) as well as the possibility that smoking has different effects on the different WM domains (i.e., verbal and visuospatial). Thus, the present study used a repeated measures design to assess the effects of abstaining from smoking on both visuospatial (symmetry span) and verbal (operation span) WM capacity (WMC) using highly reliable, well-validated, and theoretically-driven WM span tasks. Verbal n-back was also administered to examine its relationship to these complex WM span tasks and to compare the results of the present study to those of previous findings. Smokers (n = 23) and nonsmokers (n = 21) participated in two sessions separated by approximately one week. During one session, smokers completed the WM tasks after abstaining from smoking for at least 12 hours whereas in the other session smokers did not abstain from smoking and were tested after smoking. Nonsmokers served as a comparison group. Results indicated that smokers had lower verbal WM span when they were nonabstinent (smoking) compared to when they were abstinent and to nonsmokers. Although there was no difference in verbal n-back performance between the smoking sessions, smokers did manifest lower verbal n-back performance than nonsmokers. In contrast, there was no difference in visuospatial WM span between the smoking sessions or between smokers and nonsmokers. Taken together, these findings: (a) replicate previous n-back results showing differences in WM performance between smokers and nonsmokers; (b) indicate that smoking impairs nonabstinent smokers' verbal WMC and that abstaining from smoking may restore verbal WMC to levels found in nonsmokers; and (c) demonstrate that smoking does, in fact, exhibit differential effects on the different WM domains.