was lost.

"What can I do?" asked Tom hopelessly. "I have nothing to reply to him."

"But I have," quietly returned Kennedy, deliberately folding up the
message and handing it back. "Tell them all to be in the library in
fifteen minutes. This message hurries me a bit, but I am prepared. You
will have something to wire Mr. Clark after that." Then he strode
off toward the house, leaving us to gather the group together in
considerable bewilderment.

A quarter of an hour later we had all assembled in the library, across
the hall from the room in which Lewis Langley had been found. As usual
Kennedy began by leaping straight into the middle of his subject.

"Early in the eighteenth century;" he commenced slowly, "a woman was
found burned to death. There were no clues, and the scientists of that
time suggested spontaneous combustion. This explanation was accepted.
The theory always has been that the process of respiration by which
the tissues of the body are used up and got rid of gives the body a
temperature, and it has seemed that it may be possible, by preventing
the escape of this heat, to set fire to the body."

We were leaning forward expectantly, horrified by the thought that
perhaps, after all, the Record was correct.

"Now," resumed Kennedy, his tone changing, "suppose we try a little
experiment--one that was tried very convincingly by the immortal Liebig.
Here is a sponge. I am going to soak it in gin from this bottle, the
same that Mr. Langley was drinking from on the night of the--er--the
tragedy."

Kennedy took the saturated sponge and placed it in an agate-iron pan
from the kitchen. Then he lighted it. The bluish flame shot upward,
and in tense silence we watched it burn lower and lower, till all the
alcohol was consumed. Then he picked up the sponge and passed it around.
It was dry, but the sponge itself had not been singed.

"We now know," he continued, "that from the nature of combustion it
is impossible for the human body to undergo spontaneous ignition
or combustion in the way the scientific experts of the past century
believed. Swathe the body in the thickest of non-conductors of heat, and
what happens? A profuse perspiration exudes, and before such an ignition
could possibly take place all the moisture of the body would have to be
evaporated. As seventy-five per cent or more of the body is water, it
is evident that enormous heat would be necessary--moisture is the great
safeguard. The experiment w