In answer to Art's question, there is no set amount or number of
differences required to determine that something is a different species,
genus or family. What is more important is the pattern of relationships. In
particular, the rules of phylogenetic classification demand that all taxa
must be monophyletic. This means that all the descendants of a lineage must
be included within the taxon. In everyday terms, this means that the only
things that you should name (above the species level) are the ones that you
remove from the tree with a single cut. Kantor's (1991) tree of
relationships of oliviform gastropods, which I reproduce below, does not
fulfill this requirement.

       |---- Benthobia
|------|
|      |---- Pseudoliva (two species)
|
|------ Melapium
|
|              |-----Amalda dimidiata
|         |----|
|----*----|    |-----Amalda montrouzieri
          |
          |    |-----Oliva bulbosa
          |----|
               |     |----Olivella borealis
               |-----|
                     |----Olivella verreauxii


Kantor places Olivella in Olivellidae and Amalda and Oliva in Olividae, but
Olividae is not monophyletic--to remove Olividae takes two cuts--one at the
asterisk, and one to remove the Olivellas. To be consistent, the
classification should recognize either three subfamilies or three families,
Ancillidae/Ancillinae, Olividae/Olivinae and Olivellidae/Olivellinae. Each
of these can be removed with a single cut.

I don't care whether they are ranked as subfamilies or families--that's an
arbitrary decision, and hence not a scientific one. However, I'd advocate
retaining the traditional classification using subfamilies since it is
consistent with the structure of the tree, whereas the classification that
Kantor proposed is not.

>As far as I understand the figure by Kantor, Olividae are derived from
>Olivellidae. Before they "became" Olivinae (subfamily), the subfamily
>Ancillinae arose from the Olividae stock.

Given that Kantor placed Amalda and Oliva in Olividae, his figure shows
Olivellidae being derived from Olividae, not vice versa.

Gary

---------------------------------------------------------------------
Gary Rosenberg, Ph.D.                     [log in to unmask]
Malacology & Invertebrate Paleontology    gopher://erato.acnatsci.org
Academy of Natural Sciences               http://www.acnatsci.org
1900 Benjamin Franklin Parkway            Phone 215-299-1033
Philadelphia, PA 19103-1195 USA           Fax   215-299-1170