man said. "Each one has twelve
openings and is built with a low dome to keep in the heat. The flues or
chimneys are in the sides of the furnace. Within, and just beneath the
openings or working-holes, stand the great clay pots of molten batch.
These pots are made for us from New Jersey clay; formerly we used to
make them ourselves, but it was a great deal of trouble, and we now
find it simpler to buy them. They vary in cost from thirty to
seventy-five dollars, according to their size."

"And they are liable to break the first time they are used," whispered
Giusippe in a jesting undertone.

Mr. Wyman caught his words.

"Ah, you know something of glass-making then, my young man?"

"A little."

"The pots are, as you say, a great lottery. Sometimes one will be in
constant use three months or longer, and do good service; on the other
hand a pot may break the first time using and let all the melt into the
furnace. Then we have a lively time, I can tell you, ladling it out,
and taking care in the meantime that none of the other pots are upset."

Giusippe nodded appreciatively.

Many a day just such a catastrophe had occurred when he had been
working; vividly he recalled how all the men had been forced to come to
the rescue.

"Are the pots filled to the top with batch?" asked Mr. Cabot.

"Yes, we charge them pretty solid; but the raw material loses bulk in
melting, so they have to be filled in as the melt settles. At the end
of ten or twelve hours we have a refilling or _topping out_, as we
call it; usually this is enough. The first fill must become fluid and
its gases must escape before any more material is added; we also have
to be sure when we put the pots in the furnace that the temperature is
high enough to melt the batch immediately, or the glass will go bad."

"What do you use for fuel?"

"Crude oil. In the West they can get natural gas, and there they often
melt the batch in tanks instead of pots. But we find crude oil quite
satisfactory. You can readily understand that we cannot burn any fuel
that gives off a waste product such as coal dust or cinders, because if
we did such matter would get into the melt and speck the glass, causing
it to be imperfect. Much of the work done by the earliest glass-makers
was specked in this way, and in fact the genuineness of old glass is
sometimes determined from these very imperfections."

"I see," Mr. Cabot nodded.

"After the melt is in a fluid state it throws to the