oks,
and with a writing-table facing an ordinary window, which looked out
upon the garden. Our first attention was given to the body of the
unfortunate squire, whose huge frame lay stretched across the room. His
disordered dress showed that he had been hastily aroused from sleep.
The bullet had been fired at him from the front, and had remained in
his body after penetrating the heart. His death had certainly been
instantaneous and painless. There was no powder-marking either upon his
dressing-gown or on his hands. According to the country surgeon the lady
had stains upon her face, but none upon her hand.

"The absence of the latter means nothing, though its presence may
mean everything," said Holmes. "Unless the powder from a badly-fitting
cartridge happens to spurt backwards, one may fire many shots without
leaving a sign. I would suggest that Mr. Cubitt's body may now be
removed. I suppose, doctor, you have not recovered the bullet which
wounded the lady?"

"A serious operation will be necessary before that can be done. But
there are still four cartridges in the revolver. Two have been fired and
two wounds inflicted, so that each bullet can be accounted for."

"So it would seem," said Holmes. "Perhaps you can account also for the
bullet which has so obviously struck the edge of the window?"

He had turned suddenly, and his long, thin finger was pointing to a hole
which had been drilled right through the lower window-sash about an inch
above the bottom.

"By George!" cried the inspector. "How ever did you see that?"

"Because I looked for it."

"Wonderful!" said the country doctor. "You are certainly right, sir.
Then a third shot has been fired, and therefore a third person must have
been present. But who could that have been and how could he have got
away?"

"That is the problem which we are now about to solve," said Sherlock
Holmes. "You remember, Inspector Martin, when the servants said that on
leaving their room they were at once conscious of a smell of powder I
remarked that the point was an extremely important one?"

"Yes, sir; but I confess I did not quite follow you."

"It suggested that at the time of the firing the window as well as the
door of the room had been open. Otherwise the fumes of powder could not
have been blown so rapidly through the house. A draught in the room was
necessary for that. Both door and window were only open for a very short
time, however."

"How do you prove that?"

"