Page:Austin Freeman - The Mystery of 31 New Inn.djvu/13



upon one of my earlier novels, in respect of which I had claimed to have been careful to adhere to common probabilities and to have made use only of really practicable methods of investigation, a critic remarked that this was of no consequence whatever so long as the story was amusing.

Few people, I imagine, will agree with him. To most readers and certainly to the kind of reader for whom an author is willing to take trouble, complete realism in respect of incidents and methods is an essential factor m maintaining the interest of a detective story. Hence it may be worth while to mention that Thomdyke's method of producing the track chart, described in Chapters II and III, has been actually used in practice. It is a modification of one devised by me many years ago when I was crossing Ashanti to the city of Bontuku, the whereabouts of which in the far interior was then only vaguely known. My instructions were to fix the positions of all towns, villages, rivers and mountains as accurately as possible; but finding ordinary methods of surveying impracticable in the dense forest which covers the whole region, I adopted this simple and apparently rude method, checking the