ve me a leading hint too? I
looked at them and wondered what it was that they had told Thorndyke.
The little fragment of the red paper label had a dark-brown or thin
black border ornamented with a fret-pattern, and on it I detected a
couple of tiny points of gold like the dust from leaf-gilding. But I
learned nothing from that. Then the shorter piece of reed was
artificially hollowed to fit on the longer piece. Apparently it formed a
protective sheath or cap. But what did it protect? Presumably a point or
edge of some kind. Could this be a pocket-knife of any sort, such as a
small stencil-knife? No; the material was too fragile for a
knife-handle. It could not be an etching-needle for the same reason; and
it was not a surgical appliance--at least it was not like any surgical
instrument that was known to me.

I turned it over and over and cudgelled my brains; and then I had a
brilliant idea. Was it a reed pen of which the point had been broken
off? I knew that reed pens were still in use by draughtsmen of
decorative leanings with an affection for the "fat line." Could any of
our friends be draughtsmen? This seemed the most probable solution of
the difficulty, and the more I thought about it the more likely it
seemed. Draughtsmen usually sign their work intelligibly, and even when
they use a device instead of a signature their identity is easily
traceable. Could it be that Mr. Graves, for instance, was an
illustrator, and that Thorndyke had established his identity by looking
through the works of all the well-known thick-line draughtsmen?

This problem occupied me for the rest of the day. My explanation did not
seem quite to fit Thorndyke's description of his methods; but I could
think of no other. I turned it over during my solitary lunch; I
meditated on it with the aid of several pipes in the afternoon; and
having refreshed my brain with a cup of tea, I went forth to walk in the
Temple gardens--which I was permitted to do without breaking my
parole--to think it out afresh.

The result was disappointing. I was basing my reasoning on the
assumption that the pieces of reed were parts of a particular appliance,
appertaining to a particular craft; whereas they might be the remains of
something quite different, appertaining to a totally different craft or
to no craft at all. And in no case did they point to any known
individual or indicate any but the vaguest kind of search. After pacing
the pleasant walks for upwards of t