burnt match on the floor, whereas it is one of my orderly habits
never to leave burnt matches around. And at last the burnt match
became a sort of clue, for I suspected that it had been used to
light one of the candles that sat in holders of every sort, on the
top of the library shelves.

I tried getting up at night and peering over the banisters, but
without result. And I was never sure as to articles that they had
been moved. I remained in that doubting and suspicious halfway
ground that is worse than certainty. And there was the matter of
motive. I could not get away from that. What possible purpose
could an intruder have, for instance, in opening my sewing-basket
or moving the dictionary two inches on the center table?

Yet the feeling persisted, and on the second of August I find this
entry in my journal:

Right-hand brass, eight inches; left-hand brass, seven inches;
carved-wood--Italian--five and three quarter inches each; old
glass on mantelpiece--seven inches. And below this, dated the
third: Last night, between midnight and daylight, the candle in
the glass holder on the right side of the mantel was burned down
one and one-half inches.

I should, no doubt, have set a watch on my nightly visitor after
making this discovery--and one that was apparently connected with it
--nothing less than Delia's report that there were candle-droppings
over the border of the library carpet. But I have admitted that this
is a study in fear, and a part of it is my own.