The more he thought over these facts, the more dissatisfied he grew. It certainly did not look as if his effort to connect the Berlyn-Pyke tragedy with the crate affair was going to be successful. And if it failed it left him where he had started. He had no alternative theory on which to work. He recalled the four points by which he had hoped to test the matter. On each of these he had now obtained information, but in each case the information tended against the theory he wished to establish. First there was the breakdown of the car. Was that an accident or had it been prearranged? Obviously, if it had been an accident it could not have been part of the criminal’s plan. Therefore, neither could the resulting disappearance of Berlyn and Pyke. Therefore, the murderer must have been out after some other victim whose disappearance he had masked so cleverly that it had not yet been discovered. Now, Makepeace had stated definitely that the breakdown could not have been faked. Of course it would be necessary to have this opinion confirmed by the makers of the magneto. But Makepeace had seemed so sure that French did not doubt his statement. The second point concerned the movements of the car on the fatal night. French began by asking himself the question: Assuming the murdered man was Pyke, how had his body been taken to the works? He could only see one way—in the car. Suppose the murder was committed on the way from Tavistock. What then? The murderer would drive to the works with the dead man in the car. This, French believed, would be possible without discovery, owing to the distance the works lay from the town. He would then in some way square the night watchman, unpack the duplicator, put the body in its place, load the duplicator into the back of the car, drive off, somehow get rid of the duplicator, return to the road near Colonel Domlio’s house, make the two lines of footprints and decamp. At first sight this obvious explanation seemed encouraging to French. Then he wondered would there be time for all these operations? Taking the results of the tests he had made and estimating times where he had no actual data, he set himself to produce a hypothetical time-table of the whole affair. It was a form of reconstruction which he had found valuable on many previous occasions. It read: