er, and put the goldfish in that. Then they can swim around,
and Lizzie can't get at them because they'll be protected by the
glass."

She made a sectional view of the tank with its double sides and held it
up triumphantly to Joe.

"Will that solve the puzzle?" she asked.

"Why, I really believe it will!" exclaimed Joe, after a moment's
thought. "Yes, I could have metal clips, made water-tight with rubber,
fitted inside the tank. Taking five inches off each side wouldn't mean
much loss. Then I could slip four sheets of glass down in the metal
clips, and, as you say, fill the intervening space with water for the
goldfish."

"Exactly," cried Helen. "The audience can't tell whether the fish are
in the tank with you or not, for the water and glass, being
transparent, will make it look exactly as if you and the fish and the
seal were in the same bit of water."

"Good!" cried Joe. "I'll do it! The very thing, Helen!"

The fact that persons looking directly at a glass tank of water can not
tell how far back the water and tank extends is taken advantage of by
moving picture operators in producing novel effects. Only the other day
was shown a scene where a man apparently went down to the bottom of a
river. Fish swam all about him, there was a portion of a wreck half
buried in the sand and mud, and waving aquatic plants seemed to reach
out and twine themselves around the man, while fish swam above and
below him.

This effect was produced by having the man go through certain actions
behind a square glass tank in which the sand, aquatic plants, wreck and
fish had been placed. The fish could swim about, but the man was not in
the water at all but behind the tank, the water and glass offering no
obstruction to the camera.

It was that principle which Helen suggested to Joe. He and the seal
would be doing their act behind a glass and water screen, in which fish
were swimming and aquatic plants gently waving.

Joe lost no time in putting the plan into operation. He ordered the
glass and metal holding clips, with the water-tight rubber gaskets,
from the same firm in New York that had originally made Benny's tank.
They still had the patterns, and knew just the proper size and kind of
glass to send, and Joe had no difficulty in malting his tank a double
one.

The experiment was tried in private one afternoon just before the
performance. The plants and goldfish were put in the thin outer tank
which extended all around the inn