Importantly, adolescents whose parents smoke are more likely to smoke themselves (e.g., Jackson, et al., 1998; Patton, et al., 1998b; Wang, et al., 1999). They are also less likely to quit smoking and more likely to relapse if they try to quit (Hansen, Collins, Johnson, & Graham, 1985; Patton, et al., 1998a). This finding, of a relationship between the smoking behaviors of adolescents and those of their parents, has been replicated in several longitudinal studies and in a variety of settings. There are a number of possible reasons for this association. As noted above, a heritable genetic susceptibility to tobacco use likely plays a role. Additionally, one longitudinal study, based on a nationally representative sample, provides compelling evidence that there may be a prenatal effect, such that a mother's smoking habits during pregnancy may affect the likelihood that her child will smoke during adolescence (Kandel, Ping, & Davies, 1994). This relationship was found among daughters but not among sons. Further possible explanations include parental modeling of smoking behavior and increased access to cigarettes at home. Interestingly, one longitudinal study, based on a predominately non-Hispanic white sample of three generations of mothers and daughters, suggests there may also be a "do as I say, not as I do" effect. This study found that daughters whose mothers smoke tend to view their mothers as less likely to punish smoking behavior than do daughters of non-smoking mothers, although mothers who smoke are not less likely than other mothers to report that they would discipline this behavior (Chassin, et al., 1998). Thus, parents who smoke but discourage their children from smoking may be sending mixed messages.