BRILLIANT finish to a most remarkable case,” I commented as our visitors’ footsteps died away upon the stairs, “and a most magnificent piece of bluff on the part of my revered senior.”

Thorndyke smiled and Polton looked shocked.

“I shall not contest your description, Jervis,” said the former, “but, in fact, the conclusion was practically a certainty.”

“Probability,” I corrected.

“In practice,” said he, “we have to treat the highest degrees of probability as certainties; and if you consider the evidence in this case as a whole, I think you will agree that only one possible conclusion emerged. The element of bluff was almost negligible.”

“Probably you are right,” I admitted. “You usually are, and you certainly were in this case. But the evidence was so complex and conflicting that I find it difficult to reconstitute it as a whole. It would interest me very much to hear you sort it out into a tidily arranged argument.”

“It would interest me, too,” said he, “to retrace our investigation and observe the curious way in which the different items of evidence came to light. Let us do so, taking the events in the order of their occurrence and noting the tendency of the evidence to close in on the final conclusion.

“This was a very singular case. The evidence did not transpire gradually but emerged in a number of successive and perfectly distinct stages, each stage being marked by the appearance of a new fact which reacted immediately on our previous conclusions. There were seven stages, each of which we will examine separately, noting how the argument stood at the end of it.

“The first is the inquest, including the post mortem. Perhaps we had better deal with the body first. There were only two points of interest, the neck and the teeth. The dislocation of the neck appeared to me to have occurred before death and I took it to be, most probably, the immediate cause of death. As to the teeth, there was nothing very striking in their appearance; just a little pitting of the enamel. But from the arrangement of the little pits in irregular transverse lines, corresponding roughly to the lines of growth, I did not believe them to have been due to the heat but to have existed during life. I thought it possible that deceased might have had mottled teeth which had been bleached out in the fire; but, as I had never seen a case of mottled teeth, I could not form a definite opinion. I just noted the facts and satisfied myself that the pitting showed clearly in Polton’s photograph of the dead man’s face.

“And now let us consider the body of evidence which was before us when the inquest was finished and the inferences that it suggested. To me — and also to Blandy — the appearances as a whole conveyed the idea of deliberate arson; of a fire which had been arranged and started for a definite purpose. And since the death of Cecil Moxdale seemed to be part of the plan — if there was a plan — it was reasonable to suspect that this was the purpose for which the fire was raised.

“What especially led me to suspect arson was the appearance of preparation. The room, itself crammed with highly inflammable material, seemed to have been expressly prepared for a fire. But most suspicious to me was the information given by Haire to Green. It seemed designed to create in Green’s mind (as it actually did) the fear that a fire might occur. But more than this; it prepared him, if a fire should occur, to decide at once upon the way in which it had been caused. Nor was that all. Haire’s statement even suggested to Green the possibility of a fatal accident; and in the event of such a fatality occurring, it provided Green in advance with the data for identifying any body that should be found.

“Then there were the objects found in the ruins which confirmed Green’s identification. They were marked objects composed of highly refractory material.”

“They would have to be,” I objected, “if they were found. All the combustible objects would have been destroyed.”

“True,” he admitted. “But still it was a striking coincidence that these imperishable objects should happen to bear the initials of a man whose corpse was unrecognizable. The clay pipe was especially significant, seeing that people do not usually incise their initials on their pipe-bowls. But a clay pipe is, as nearly as possible, indestructible by heat. No more perfect means of identification, in the case of a fire, could be devised than a marked clay pipe. To me, these most opportune relics offered a distinct suggestion of having been planted for the very purpose which they served.

“But there is one observation to make before finishing with the positive aspects of the case. It was assumed that the man who was in the house when the fire broke out was a live man; and it was agreed that that live man was Cecil Moxdale. Now, I did not accept, unreservedly, either of these assumptions. To me, the appearances suggested that the man was already dead when the fire started. As to the identity, the probability seemed to be that the man was Moxdale; but I did not regard the fact as having been established conclusively. I kept in my mind the possibility of either a mistake or deliberate deception.

“And now, what conclusions emerged from these considerations? To me — and to Blandy — they suggested a crime. My provisional hypothesis was that Haire had made away with Moxdale and raised the fire to cover the murder; that the crime had been carefully planned and prepared; and that, for some reason, Haire was especially anxious that the body should be identified as that of Cecil Moxdale. That, as I said, was the positive aspect of the case. Now let us look at the negative.

“There were two facts that conflicted with my hypothesis. The first was that when the fire broke out, Haire was in Dublin and had been there for five days. That seemed to be an unanswerable alibi. There was no trace of any sort of fire-raising apparatus known to the experts or the police; indeed, no apparatus was known which would have been capable of raising a fire after an interval of five days. The large and complicated appliances used for the automatic lighting of street lamps do not come into the problem; they would not have been available to Haire, and, in fact, no trace of anything of the kind was found. Apparently, it was a physical impossibility that the fire could have been started by Haire.

“The second objection to my hypothesis was in the nature of the injury. A dislocation of the neck is, in my experience, invariably an accidental injury. I have never heard of a homicidal case. Have you?”

“No,” I answered; “and, in fact, if you wanted to dislocate a man’s neck, I don’t quite know how you would go about it.”

“Exactly,” he agreed. “It is too difficult and uncertain a method for a murderer to use. So that, in this case, if the broken neck was the cause of death, the man would appear to have died from the effects of an accident.

“Thus, the position at the end of the first stage was that, although the case as a whole looked profoundly suspicious, there was not a particle of positive evidence of either arson or murder.

“The second stage was introduced by the disappearance of Haire. This was most mysterious. Why did Haire not return at the expected time? There was no reason why he should not, even if my hypothesis were true. For if he had raised a fire to cover a murder, his plan had succeeded to perfection. The fire had been accepted as an accident, the body had been identified, and the man’s death had been attributed to misadventure. And not only was there no reason why Haire should not have come home; there was a very good reason why he should. For his absence tended to start inquiries, and inquiries were precisely what he would have wished to avoid. I could think of no explanation of his disappearance. There was a suggestion that something had gone wrong; but there was no suggestion whatever as to what it was. Nevertheless, the fact of the disappearance tended to make the already suspicious group of events look even more suspicious.

“The third stage was reached when we learned that Moxdale senior was dead and heard of the provisions of his will. Then it appeared that Haire stood to benefit to the extent of four thousand pounds by the death of Cecil Moxdale. This, of course, did not, by itself, establish a probability that Haire had murdered Moxdale; but if that probability had already been suggested by other facts, this new fact increased it by supplying a reasonable and adequate motive. At this stage, then, I definitely suspected Haire of having murdered Moxdale, though still not without some misgivings. For the apparently insuperable difficulty remained. It seemed to be a physical impossibility that Haire could have started the fire.

“Then came Polton’s astonishing discovery; and immediately the position was radically altered. Now, it was shown, not only that it was possible for Haire to have started the fire, but that it was nearly certain that he had done so. But this new fact reacted on all the others, giving them an immensely increased evidential value. I had now very little doubt that Haire had murdered Moxdale.

“But the mystery of Haire’s disappearance remained. For he was all unaware of Polton’s discovery. To him, it should have seemed that all had gone according to plan and that it was perfectly safe for him to come back. Then why was he keeping out of sight? Why did he not return, now that his uncle was dead and the stake for which he had played was within his grasp? I............