Six chronic smokers of mid- to high-carbon monoxide (CO) yield cigarettes smoked ultralow- (1.6 mg CO), low- (5.9 mg CO) and high- (14.3 mg CO) yield commercial cigarettes under controlled smoking conditions in which either puff number or puff spacing was manipulated. CO exposure (pre- to postsmoking increments) was directly related to the number of puffs taken for all cigarette yields. CO exposure from the high- and low-yield cigarettes was equivalent when the number of puffs taken from the low-yield cigarettes was increased by 50% (from 8 to 12 puffs). In contrast, CO exposure from ultralow-yield cigarettes was still marginally lower than exposure from high-yield cigarettes after a 4-fold increase in puff number (8 to 32 puffs). Puff spacing did not affect biological exposure to CO. The study showed that the number of puffs taken during smoking can clearly affect biological exposure to CO, but that compensation for lowered yield using increased puffs is much more difficult when ultralow- as compared with low or "light"- yield cigarettes are smoked.