The evidence that cigarette smoking and atmospheric pcllution are causes of lung cancer is largely statistical. The first evidence was indirect; that is, i1. was noticed that in many countries the incidence of lung cancer had increased and that the increase could be correlated with changes in the prevalence of cigarette smoking and of certain types of atmospheric pollution.Since then much direct evidence has been obtained. The relationship between cigarette smoking and lung cancer has been demonstrated retrospectively by comparing the smoking habits of patients with and without lung cancer and prospectively by observing the mortality from lung cancer in groups of persons of known smoking habits. Conclusions can be drawn from these studies only after careful examination of the results. In particular it is important in retrospective studies to test a) the reproducibility of the data, b) the representativeness of the data, and c) the comparability of the special series and their controls. The resul1.s of retrospective studies are all similar and all show a close relationship between cigarette smoking and the disease.The results have been confirmed by pro~pective studies which are lesF. open to bias. The results can be explained if cigarette smoking causes lung cancer or if both are related to some third common factor. Ancillary data (pathological changes in the bronchial mucosa, animal experiments, etc.) support the causal hypothesis.The evidence relating to atmospheric pollution is less definite and it is difficult to get direct evidence of a relationship in the individual. It is clear that pollution has little effect in the absence of smoking, but the mortality associated with a given amount of smoking is generally greater in large towns than in the countryside and among men who have emigrated from Britain than among men who have lived all their lives in less polluted countries.
PREDICTION OF CARBON MONOXIDE (CO) ATMOSPHERIC POLLUTION CONCENTRATIONS USING METEROLOGICAL VARIABLES
