> I believe the formula D. intermedia x D. rotundifolia can be construed as
> different from D. rotundifolia x D. intermedia... isn't the convention to
> place maternal species first?
This is a very useful (and for some purposes essential) convention of
horticultural practise, where single clones are distinguished from
each other.
However, for biological or at least taxonomic questions, this
convention does not apply. Here, it is immaterial in which direction
and how frequently hybridization occurred. The product is always a
+/- recognizable mixture of the parent species. The only matter of
importance is the introduction of some part of the genome of one
parent species into the genome of another. All plants (i.e. incl.
different clones) resulting from such events are (i.e. should be
sensu ICBN) labelled by the same name.
> My question is, if a hybrid plant is found in nature, or for any reason
> one doesn't know which of the two species served as the maternal parent,
> how does one designate the cross without conveying false information?
A consequence of what I have written above is the commutability of
the position of the names in a bastard formula for scientific
purposes. By convention, taxonomists usually cite the alphabetically
first name first in bastard formulae. I do not follow this convention
in my synonym list in those cases where the maternal species is
known (i.e. usually in cultivars), because no information is lost for
taxonomic purposes (but some info is gained for horticultural
purposes!) if the maternal species is cited first. Only in typified
natural hybrids where the maternal species is unclear, I have
followed the "taxonomic" (alphabetical) usus.
A horticulturist "recreating" a natural hybrid following the
alphabetical formula and finding that the result does not correspond
to the original description (whereas the opposite hybrid does so)
will most probably wish to designate the different products by
cultivar names, and this is the appropriate way to deal with the
different clones. In many cases, the hybrids resulting from either
procedure will not notably differ from each other, however.
Kind regards
Jan