Andrew,
An interesting question, but perhaps more in the realm of philosophy
than genetics??  We already know that most mutations, even minor
ones, are ultimately fatal.  That is to say, given that most species
are extremely well adapted to their environment after millions of
years of cumulative change, any additional change imposed upon a
species, or upon an individual of a species, especially a randomly
selected change, is likelty to "rock the well balanced boat", that
is, have a net detrimental effect.  Consequently an individual so
affected usually dies before reproducing, thereby removing the
mutation from the gene pool.  When an animal's offspring is
"completely changed", that offspring is typically spontaneously
aborted, or stillborn, or dies shortly after birth (obviously,
medical intervention shifts the balance in some human cases, but
animals in the wild don't have that option).  Therefore, the chances
of an individual surviving a major short-term morphologic change are
essentially zero.  I don't know just how "major" a change you had in
mind.  A sinistral Buccinum undatum is the same species as a dextral
on.  An albino or melanistic or rostrate specimen is the same species
as its parents.  Apparently some individuals with major morpholigic
changes do survive though.  We do see quite a few "freak" specimens
of fully mature shells.  It would be interesting to know how many of
these are of genetic origin, and whether any such changes are
transmissible to subsequent generations.
Paul M.