According to Dr. Rogers, however, Smith, during his residence in Kirkcaldy, went out one Sunday morning in his dressing-gown to walk in the garden, but once in the garden he went on to the path leading to the turnpike road, and then to the road itself, along which he continued in a condition of reverie till he reached Dunfermline, fifteen miles distant, just as the bells were sounding and the people were proceeding to church.  The strange sound of the bells was the first thing that roused the philosopher from the meditation in which he was immersed.[227] The story is very open to criticism, but if correct it points to sleepless nights and an incapacity to get a subject out of the head, due to over-application.

The persistency of his occupation with his book, according to Robert Chambers in his Picture of Scotland, left a mark on the wall of his study which remained there till the room was repainted shortly before that author wrote of it in 1827.  Chambers says that it was Smith’s habit to compose standing, and to dictate to an amanuensis.  He usually stood with his back to the fire, and unconsciously in the process of thought used to make his head vibrate, or rather, rub sidewise against the wall above the chimney-piece.  His head being dressed, in the ordinary style of that period, with pomatum, could not fail to make a mark on the wall.

M’Culloch says Smith dictated the Wealth of Nations but did not dictate the Theory of Moral Sentiments.  Whether he had any external ground for making this assertion I cannot tell, and, apart from such, the probability would seem to be that if he dictated his lectures in Edinburgh to an amanuensis, as seems probable, as well as his Wealth of Nations, he would have done the same with his Theory.  But M’Culloch professes to see internal evidences of this difference of manual method in the different style of the respective works.  Moore met M’Culloch one evening at Longman’s, and they were discussing writers who were in the habit of dictating as they composed.  One of the party said the habit of dictating always bred a diffuse style, and M’Culloch supported this view by the example of Adam Smith, whose Wealth of Nations, he said, was very diffuse because it had been dictated, while his Theory, which was not dictated, was admirable in style.  But in reality there is probably more diffuse writing in the Theory than in the Wealth of Nations, which is for the most part packed tightly enough.  Another Scotch critic, Archibald Alison the elder, the author of the Essay on Taste, even surpasses M’Culloch in his keenness in detecting the effects of this dictating habit.  He says that Smith used to walk up and down the room while he dictated, and that the consequence is that his sentences are nearly all the same length, each containing as much as the amanuensis could write down while the author took a single turn.[228] This is excessive acuteness.  Smith’s sentences are not by any means all of one length, or all of the same construction.  It need only be added that the habit of dictating would in his case arise naturally from his slow and laboured penmanship.