We have spoken of persons who are incapable of deriving stimulus from the use of tobacco, but are always narcotized by it. We doubt if perfectly healthy persons are ever affected in this way. In a considerable number of cases we have observed that this incapacity occurs in people who are troubled with some chronic abnormal action or inaction of the liver; but we have as yet been unable to make any generalization which might serve to connect the two phænomena. In the great majority of cases, however, the incapacity has been probably induced by chronic narcosis resulting from the long-continued abuse of tobacco. Recent researches have shown that confirmed drunkards have after a while modified the molecular structure of their nervous systems to such an extent that they can never for the rest of their lives touch an alcoholic drink with safety. For such poor creatures, teetotalism is the only hygienic rule. It is fair to suppose that under the continuous influence of tobacco-narcosis the nervous system becomes metamorphosed in some analogous manner, so that after a while tobacco ceases to be of any use and becomes simply noxious. This is likely to be the case with those who begin to chew or smoke when they are half-grown boys, and keep on taking enormous doses of the narcotic until they have arrived at middle age. As Mr. Parton seems to find a difficulty in realizing that any one who smokes at all can smoke less than from ten to twenty large cigars daily, (for he always uses these figures when he has occasion to allude to the subject), we presume this to be about the ration which he used to allow himself. If so, no wonder that he found it did not pay to smoke. He probably did the wisest thing he could do when he gave up the habit; and his mistake has been in endeavouring to erect the limitations of his own experience into objective laws of the universe. To sum up the physiological argument: we have endeavoured, as precisely as possible in the present state of knowledge, to answer the question, Does it pay to smoke? From the outset we have found it necessary to a clear understanding of the problem to keep steadily in mind the generic difference between the effects of tobacco when taken in narcotic quantities and its effects when taken in stimulant quantities. The first class of effects we have seen to be always and necessarily bad; though not so extremely and variously bad as hygienic reformers appear to believe.[30] With regard to the second class of effects, we have seen reason to believe that they are almost always good. We have seen reason to believe that, in the first place, the stimulant dose of tobacco retards waste; and, in the second place, that it facilitates repair:-- I. By its action on the sympathetic ganglia, aiding digestion,-- II. By its action on the medulla oblongata, aiding the circulation,-- III. By its action on the interstitial nerve-fibres, aiding the general assimilation of prepared material.