“Graphite!” exclaimed Thorndyke. “I have found no graphite, but I have found traces of cocoa—spiral vessels and starch grains—and of hops—one fragment of leaf and several lupulin glands. May I see the graphite?”

I passed him the slide and he examined it with keen interest. “Yes,” he said, “this is undoubtedly graphite, and no less than six particles of it. We had better go over the coat systematically. You see the importance of this?”

“I see that this is evidently factory dust and that it may fix a locality, but I don’t see that it will carry us any farther.”

“Don’t forget that we have a touchstone,” said he; and, as I raised my eyebrows inquiringly, he added, “The Yale latch-key. If we can narrow the locality down sufficiently, Miller can make a tour of the front doors.”

“But can we?” I asked incredulously. “I doubt it.”

“We can try,” answered Thorndyke. “Evidently some of the substances are distributed over the entire coat, inside and out, while others, such as the graphite, are present only on certain parts. We must locate those parts exactly and then consider what this special distribution means.” He rapidly sketched out on a sheet of paper a rough diagram of the coat, marking each part with a distinctive letter, and then, taking a number of labelled slides, he wrote a single letter on each. The samples of dust taken on the slides could thus be easily referred to the exact spots whence they had been obtained.

Once more we set to work with the microscope, making, now and again, an addition to our lists of discoveries, and, at the end of nearly an hour’s strenuous search, every slide had been examined and the lists compared.

“The net result of the examination,” said Thorndyke, “is this. The entire coat, inside and out, is evenly powdered with the following substances: Rice-starch in abundance, wheat-starch in less abundance, and smaller quantities of the starches of ginger, pimento and cinnamon; bast fibre of cinnamon, various seed cortices, stone-cells of pimento, cinnamon, cassia and black pepper, with other fragments of similar origin, such as resin-cells and ginger pigment—not turmeric. In addition there are, on the right shoulder and sleeve, traces of cocoa and hops, and on the back below the shoulders a few fragments of graphite. Those are the data; and now, what are the inferences? Remember this is not mere surface dust, but the accumulation of months, beaten into the cloth by repeated brushing—dust that nothing but a vacuum apparatus could extract.”

“Evidently,” I said, “the particles that are all over the coat represent dust that is floating in the air of the place where the coat habitually hangs. The graphite has obviously been picked up from a seat and the cocoa and hops from some factories that the man passes frequently, though I don’t see why they are on the right side only.”

“That is a question of time,” said Thorndyke, “and incidentally throws some light on our friend’s habits. Going from home, he passes the factories on his right; returning home, he passes them on his left, but they have then stopped work. However, the first group of substances is the more important as they indicate the locality of his dwelling—for he is clearly not a workman or factory employee. Now rice-starch, wheat-starch and a group of substances collectively designated ‘spices’ suggest a rice-mill, a flour-mill and a spice factory. Polton, may I trouble you for the Post Office Directory?”