come to that yet."

"When it has come to it, there will be very little time left to talk
it over."

"Did you not say, brother, that the _Nelson_ might hear our signals
without our hearing hers? If so, there is a chance for Willis yet."

"Certainly, Jack, because she has the wind in her favor to act as a
speaking-trumpet, whilst we had it against us acting as a deafener."

"Is there any other influence that affects sound besides the wind?"

"Yes, I have already mentioned that temperature has something to do
with it. Sound varies in intensity according to the state of the
atmosphere. If, for example, we ring a small bell in a closed vessel
filled with air, it has been observed that, as the air is withdrawn by
the pump, the sound gradually grows less and less distinct."

"And if a vacuum be formed?"

"Then the sound is totally extinguished."

"So, then," objected Willis, "if two persons were to talk in what you
call a vacuum, they would not hear each other?"

"Two persons could not talk in a vacuum," replied Ernest.

"Why not?"