[6] Mr. Lizars (On _Tobacco_, p. 54) has the impudence to cite Pereira (vol. ii. p. 1426) as an opponent of smoking, because he calls nicotine a deadly poison! And on p. 58 he similarly misrepresents Johnston. This is the way in which popular writers contrive to marshal an array of scientific authorities on their side. In the case of tobacco, however, it is difficult to find physiologists who will justify the popular clamour. They have a way of taking the opposite view; and when Mr. Lizars cannot get rid of them in any other way, he insinuates that all writings in favour of tobacco "have been _got up_ from more than questionable motives." (p. 137.) This is in the richest vein of what, for want of a better word, we have called radicalism; and may be compared with Mr. Parton's belief that physicians recommend alcoholic drinks because they like to fatten on human suffering! (_Smoking and Drinking_, p. 56.)
It is desirable that this point should be thoroughly understood before we advance a step farther. Here is the _pons asinorum_ in the study of narcotics, but it must be crossed if we would get at the truth concerning alcohol and tobacco. Alcohol is a poison, says the teetotaler, who means well, but has not studied the human organism; alcohol is a poison, and once a poison always a poison. Nothing can seem more logical or reasonable, so long as one knows nothing about the subject. A quart of brandy is admitted to be poison; is not, therefore, a spoonful of brandy also poison? We reply, by no means. Physiological questions are not to be settled by formal logic. Here the quantity is the all-essential element to be taken into the account. Common salt, in large doses, is a virulent poison; in lesser doses it is a powerful emetic; in small doses it is a gentle stimulant, and an article of food absolutely essential to the maintenance of life. In the spirit of the teetotaler's logic, then, it may be asked, If a pound of salt is a poison, is not a grain of salt also a poison? We reply, call it what you please, you cannot support life without it. So from the poisonous character of the quart of brandy, the poisonous character of the spoonful is by no means a legitimate inference. The evil effects of the small dose are to be ascertained by experiment, not to be taken for granted. Logic is useful in the hands of those who understand the subject they reason about; but in other hands it sometimes leads to queer results. It was logic that used up the one-hoss shay.