Sir.—The article entitled "Effect of Maternal Cigarette Smoking on Apgar Scores" by Garn et al (Journal 1981;135:503-506) is a valuable contribution to the medical literature because data are presented to show a statistical relationship between maternal cigarette smoking and lower infant Apgar scores. The correlation of an increased incidence of low Apgar scores with maternal smoking is quite convincing.
However, this correlation is the only fact demonstrated by the data presented and it is scientifically unsound to assume that a cause-and-effect relationship is demonstrated as well. It seems that such a relationship is implied by the comments of the authors and by the marginal comment by T. A. Merritt in the same issue (1981;135:595-596). If this is so, where are the data to support it?
If one adopts the very reasonable hypothesis that a cause-and-effect relationship does exist between maternal smoking and lower Apgar scores, then one is bound