Coenonympha tullia and evidence

[Kondla, Norbert FOR:EX]

Thanks to all for demonstrating that there is a great deal of interest in
this topic and reinforcing the view that Coenonympha needs more work - I
agree completely with this view.  So my closing thoughts on ringlets (for
now), in response to the questions raised below, are: I have consulted the
papers which are referenced and with due respect to those who hold a
different view; they do not in my view provide evidence relevent to the
example I provided eg. published evidence of intergading taxa in the Great
Basin does not alter the lack of published evidence of intergrading taxa in
the northwest part of our continent; the matter of defining species (or any
other taxonomic unit) and then thinking that everyone will agree with the
definition (of course there is more than one) and further the challenge of
defining agreeable criteria for assigning, for example, something to a
subspecies or species rank -- well, that goes to the heart of why there has
been and will continue to be discussion and debate. I do not expect there
will ever be full agreement. I am personally very uncomfortable about using
a single diagnostic character for the purposes of taxonomic decision making.
Single characters do often work well for identification purposes but even
then I am uncomfortable relying on one character. This holds true regardless
of what the character is - color/pattern, genitalic feature (so far as I
know the lock and key hypothesis has never been proven) or chemical, etc.  I
have no conceptual difficulty with an arbitrarily huge number of species (or
even an arbitrarily small number of species)-- because I do not think that
decisions are arbitrary in most cases. People most often have reasons for a
particular decision, even it is only a preference as to how many or few
species one wants to cram into a single genera for example.  I have no
preconceived notions about the "right" quantity of any taxonomic unit. Which
reminds me of another taxonomic issue that has provided me with intellectual
fodder: do we really have only one species, Boloria chariclea(can't remember
the generic name that the generic splitters like), in North America? or is
there more than one.  The most recent taxonomic interpretation seems to be
based on only one genitalic character -- gosh I wonder :-) Anybody out there
bold enough to share some thoughts on that one ??

-----Original Message-----
From: Felix Sperling [mailto:Felix.Sperling at ualberta.ca]
Sent: Tuesday, April 04, 2000 8:48 PM
To: leps-l at lists.yale.edu
Subject: Coenonympha tullia and evidence


Norbert,

	As you work toward trying to assign a name for the Coenonympha
that you are seeing in se BC, I recommend that you read the following:
1. Brown, F.M. 1955. Studies of nearctic Coenonympha tullia (Rhopalocera,
Satyridae). Bull. American Museum of Natural History 105: 363-409.
  - this paper presents evidence for continuity of populations across
much of North America, and the introduction deals at some length with
the philosophical problems of attaching names to such populations.
2. Austin, G. and R.E. Gray. 1998. The Coenonympha tullia (Muller)
complex (Lepidoptera: Nymphalidae: Satyrinae) of the Great Basin region.
pp. 587-612 in Systematics of Western North American Butterflies.
Mariposa Press, Gainsville, FL. (ed. by T.C. Emmel)
- more documentation of continuity across the Great Basin, in spite of
plenty of geographic variation

	Also, before you proceed farther in arguing for or against the
species status of "californica" or "inornata", I think you should define
what you mean by "species". If it is that two populations in different
places have "obvious and consistent differences", then are what are
these differences in Coenonympha and what proportion of the populations
have these differences? How do you account for the clear statements in
the literature (for example above) that these differences are not
consistent? For allopatric populations, how does the extent of these
differences compare to the differences between related sympatric species
pairs, like C. haydeni vs C. tullia? If you are relying on a single
"diagnostic" character to justify elevating allopatric populations
as species, how would you prevent the problem where the number of new
species you recognize becomes arbitrarily huge and depends primarily
on the amount of, say, sequencing or microsatellite survey that you could
afford?

	Felix Sperling