“The most obvious method is that of keeping the prisoner in such a state of debility as would confine him to his bed. But such debility could be produced by only starvation, unsuitable food, or chronic poisoning. Of these alternatives, poisoning is much more exact, more calculable in its effect and more under control. The probabilities, then, were in favour of chronic poisoning.

“Having reached this stage, I recalled a singular case which Jervis had mentioned to me and which seemed to illustrate this method. On our return home I asked him for further particulars, and he then gave me a very detailed description of the patient and the circumstances. The upshot was rather startling. I had looked on his case as merely illustrative, and wished to study it for the sake of the suggestions that it might offer. But when I had heard his account, I began to suspect that there was something more than mere parallelism of method. It began to look as if his patient, Mr. Graves, might actually be Jeffrey Blackmore.