Paratypes

[Robert Mill]

There have been a couple of responses to this enquiry which quote the
ICZN. From these, it would appear that the ICZN differs from the ICBN
in what one can and cannot do.

The ICBN is very (?) clear: Art. 9.5 (Tokyo Code 1994) states, as Tim
wrote, "A paratype is a specimen cited in the protologue that is
neither the holotype nor an isotype, nor one of the syntypes if two
or more specimens were simultaneously designated as types."

(For the benefit of the ICZN folk, an isotype is a duplicate of the
holotype; the protologue is "everything associated with a name at
its valid publication, i.e. diagnosis, description, illustrations,
references, synonymy, geographical data, CITATION OF SPECIMENS,
discussion and comments").

Thus, under the ICBN if I describe a new species, citing a holotype
and if appropriate one or more isotypes, but also include a list of
say 30 additional specimens examined, ALL those 30 specimens
automatically become paratypes. There is no way of getting round it
by demoting some to a "lower category" as suggested by Reveal; if
they're in the protologue, they're paratypes whether you like it or
not!

It seems odd that no-one seems to have found this a problem before.
Since I received the original message this a.m. and sent my initial
reply directly back to Tim I have leafed through all our old copies
of the Code in the E library, going back to 1935. The wording of the
paratype definition has been virtually unchanged since it first
appeared (1952) but from about 1956 until (and including) the Sydney
Code, the definition was moved to a separate section of the Code
called "Guide for the Determination of Types", where (quoting
Leningrad Code 1978, Guide Types c as an example) the definition read
"a paratype is a specimen cited in the protologue other than the
holotype, isotype(s), or syntypes." Before c.1956 the sentence
briefly had the status of an Article. Article status was restored in
the Berlin Code of 1988 which had the only material change in the
text I have traced since 1956: "a paratype is a specimen OR
ILLUSTRATION cited in the protologue that is neither the holotype
nor an isotype, nor one of the syntypes ... (etc)." Why "an
illustration" was inserted for the first time in the Berlin Code I
don't know; the words have been removed again. The rest of the
Article's wording is extremely similar to that of 40 years ago,
therefore why the sudden fuss? Surely many taxonomists would have
spotted the problem by now? I think it may be because our methodology
has grown ever more scientific with more and more emphasis on citing
all or most specimens examined. I myself have numerous new plant taxa
in course of publication, many of which are based on sizeable suites
of specimens. I do not really want all of these to be treated as
paratypes, but under the ICBN I have to treat them as such, even if
they are listed as "Other specimens examined" and not "Paratypes".
The only way to get round the problem, unless the ICBN can be amended
so that we botanists can cite specimens without them becoming
paratypes as seems to be possible under the ICZN, is to publish two
papers every time:

a. "Ten new species of Bloggsia" which would contain ONLY the
absolute minimum to validate the names (Latin diagnosis and citation
of a holotype).
b. "A revision of the genus Bloggsia" which would contain the full
works: English descriptions, discussion, full citation of all
specimens, geographical distribution, other comments.

This does not seem to be a good way of doing taxonomy even though
you get two papers for the price of one on your c.v. Occasionally I
have had to resort to it in an emergency in order to get a name into
circulation in advance of a Flora in which the name will be used,
e.g. in March 1996 I will be publishing two new Boraginaceae taxa
with only the Latin and the holotype, in advance of a more detailed
revision of the group concerned. These two species are relevant to
this problem as one is based on the study of 33 gatherings
(represented by more than that number of specimens as some
gatherings are duplicated in 2 or more institutes); the other is
based on 13 gatherings. Thus, one species will be based on upwards
of 50 specimens. Had I validated the names in the revision currently
in progress, all of these (other than the holo. and isos.) would
have been paratypes!! The fact that I had to get the name published
quick has relieved them all of a status I did not really want to
award them.

Paratypes of course do not normally have significant nomenclatural
status since the application of the name is fixed by the holotype
assuming one is designated (which we all have to do now). Only if
some calamity resulted in the destruction of the holotype and all
its isotypes (unlikely if there are isotypes conserved in more than
one institution, becomes even less likely when the institutions are
in more than one country or continent) would it be necessary to use
a paratype to designate a neotype or lectotype. The corollary of
this is that one should if possible distribute duplicate material of
type  specimens (isotypes) as widely as possible.

The unimportance of paratypes is suggested by the fact that for much
of the last forty years they were relegated to a small-print
footnote in the Guide to the Determination of Types. It is only
since Berlin Code 1988 that the status of Article has been restored
to them. Like Tim, I do think the present Code inflates their status
and forces us into taxonomic practices, and awarding status, that we
did not intend.

As a point of interest, at E we put holotypes, isotypes, lectotypes,
neotypes and syntypes in red-edged type covers but not paratypes. No
decision has yet been taken regarding epitypes!

Best wishes

Robert Mill

      ********************************************************
      (Dr) ROBERT R MILL
      Royal Botanic Garden Edinburgh
      20a Inverleith Row, EDINBURGH EH3 5LR, SCOTLAND, U.K.

      Electronic Mail:   R.Mill at rbge.org.uk OR robert at rbge.org.uk
      Telephone:         + 44 131 552 7171 exts. 240 or 449
      Facsimile:         + 44 131 552 0382

     ***********************************************************