l, if
properly made, show every scale and every line, be free from
"blow-holes" or blemishes of casting, besides having a fairly even and
square surface surrounding the cavity from whence the fish has been
extracted.

This first mould can be cast into again by plaster of Paris, and will,
in the case of most fishes, turn out a satisfactory reproduction of
the original. Some fishes there are, however, so curiously shaped as
to make the first or "female" mould so "undercut" as to render it
impossible to get a return cast. In this case, nothing avails but the
destruction of the first to release the copy. There are several ways
of doing this; one of the most simple is sawing with many cuts the
edges of the first, or, as it now becomes, the "waste" mould as near
to the cavity as you dare, before casting into it; having done which,
and allowed several hours, or a day even, to elapse, you proceed to
break it away, piece by piece, by gentle blows with a hammer, leaving
the enclosed fish to make its appearance little by little. When this
plan is adopted, the last cast or copy must be run double the
thickness to that you wish to destroy, otherwise you may break the
copy instead of the "waste" mould. Another way is to make the first
mould very thin, or to put thin successive coats of plaster over the
fish, with brown paper between each coat, and subsequently breaking
them away, layer by layer, after the fish is extracted and the mould
is filled in by plaster.

In casting into the first mould for the "copy" or perfect cast, it
will, of course, be necessary to lay it concavity uppermost, and to
surround it with a wall of board like the last, brushing over the
concavity, and indeed the whole of the tablet surrounding it, with
soft soap and water, or oil, or thin pipe-clay and water; or, if the
mould has been baked dry, soaking it in water alone will be sufficient
to prevent the copy sticking. Recollect that the flatter the
tablet--surrounding the cavity left by the fish--is made, the better
will be that of the model.

Supposing that your cast, or model fish, has been turned out in good
condition, you will see that there are still certain inequalities of
the tablet, and certain roughnesses around the fins, mouth, etc.; these
latter must be "relieved" and undercut by the aid of the
"undercutting" and "relieving" tools (see Figs. 29 and 30), the tablet
must be pared flat by a long broad flat chisel called a "firmer," and
the edges a