For all the reasons you cited Andrew, my question was not a very good one and I
should have known that.  As fossil collectors we are well aware that exterior
morphology, which is all that a fossil is cannot really be taxonomically
slotted according to the best criteria which would involve charactericstics of
the living organism.  People who id fossils to species level, especially the
older ones in fully lithified sediments are probably deluding themselves.
What I should have asked is what mollusc "seems" to have the longest fossil
record based on shell shape alone, and this really stops at the genus in any
practical  sense as you note, or better yet what shell morphology seems to have
the longest fossil record.  I suppose one could stretch a bit farther than
shell shape if you had the fossils in-situ in their sediment and could make a
guess at habitat, but that's stretching it.  In other words, what are the
successful long-standing shell shapes, success here measured in terms of
durability over the ages. Interestingly the so-called primitive shell shapes
such as Pleurotomaria (and that little limpet-shaped thing, can't remember the
name right now) are still relatively unknown to us because we can't get at
them.  The pace of change is probably so slow in the abyssal depths that body
design can remain unchanged over aeons.  I wonder how adaptable the creatures
would be if they had to change, or have they lost their propensity for change
over the aeons and are now theoretically at risk.
Many of you have provided lots of genus level candidiates.  I like
Pleurotomaria.  We have fossils of pleurotomarid types in our collection back
to the Ordovician I think (Joe knows for sure).

"Andrew K. Rindsberg" wrote:

> What living species of mollusk has the longest fossil record? That is an
> interesting question that can be examined from several points of view.
>
> Some shells are so simple that several populations that do not reproduce
> together have much the same shell. Since they do not reproduce together,
> they are considered separate species and this can be tested by DNA
> analysis. More detailed study of the shells often shows subtle differences
> between these hidden species, but not always.
>
> Also, some species have such variable shells that forms overlap between
> species, as in the oysters.
>
> The result is that living populations can be separated into species (true
> lineages of breeding populations, more or less) in a way that fossil
> populations cannot, even in principle. So we may reasonably suspect that a
> long fossil range of a simple or highly variable shell represents several
> real species, but cannot test or prove this speculation without new kinds
> of evidence that are not currently available.
>
> Thus, we have extremely long-ranging genera like Nucula, Nuculana, Solemya,
> all of which are hundreds of millions of years old. I suspect that the
> genetic code of these organisms has changed with time, if only to resist
> new diseases. Still, the shell forms are remarkably stable, and the laws of
> statistics do support the idea that some taxa should be much longer-ranging
> than others. Also complicating this picture is the fact that simple animals
> really do survive major extinctions better than complex ones, so you can
> expect Nucula to be a survivor.
>
> When we look at the older literature on mollusks, we find many more
> long-ranging genera. But as time goes on, people examine the details and
> realize that the genera should be split up. So genera that formerly had
> long ranges, like Nautilus and Ostrea, now are split into tens or even
> hundreds of genera that represent whole families and orders of mollusks.
> You will still find paleontologists referring to Ostrea in Cretaceous
> rocks, for instance, but this represents a situation where the researcher
> does not know what genus to place the shells in, and is merely using
> "Ostrea" as a temporary holding tank.
>
> What we see in the fossil record is that some species disappear and others
> appear. There are always some new appearances and extinctions, even in the
> quietest of times, though of course there are also mass extinctions. So if
> you went backward through time, you would see fewer and fewer familiar
> species. In the 19th century, Charles Lyell developed the idea that you
> could determine the age of a fossil deposit by the proportion of modern
> molluscan species it contains. This is the original concept of the Cenozoic
> epochs, which were then modified a bit and can now be listed from oldest to
> youngest as Paleocene, Eocene, Oligocene, Miocene, Pliocene, Pleistocene,
> and Holocene. It turned out later to be a better idea to base the epochs on
> particular sequences of rock layers at particular "type sections", but this
> was the original concept and it was based on Mollusca!
>
> In Pleistocene rocks, almost every shell is of a familiar species, though
> not always of a current subspecies. In Pliocene and Miocene rocks, many of
> the shells are indistinguishable from those of modern species, and most
> genera are familiar. In Oligocene and older rocks, there might be one or
> two species out of hundreds that cannot be distinguished from those of
> modern kinds, and they typically belong of one of those simple or variable
> forms mentioned above, like Calyptraea aperta. I don't know of any
> Cretaceous forms that are still extant, but would not be very surprised if
> someone has an example.
>
> So, how long can a species having a complex shell last without substantial
> change? Let's say a few tens of millions of years.
>
> Incidentally, abalones and Nautilus were not very good guesses. The
> abalones have a very imperfect fossil record, because they live on rocky
> coasts where shells tend to be pounded into bits and then eroded away. And
> the living species of Nautilus, so far as I know, have no fossil record at
> all. Of course, modern Nautilus can still be called a "living fossil"
> because it is very similar to fossil forms, but it is not identical to any
> of them.
>
> Andrew K. Rindsberg
> Geological Survey of Alabama