Genus and type designation question

[Doug Yanega]

A colleague here (not on the Net) brought me a confusing problem today, and
I told him I'd try to solicit opinions from you all;

The story begins with a genus of Noterid beetles, Suphis, originally known
from Africa and South America. Crotch, in 1873, discussed another genus,
Colpius, which he felt was likely to be a junior synonym of Suphis, but he
stated that if this were true, the American species included in Suphis
(_Noterus bicolor_ Say, _Suphis lineatus_ Horn, and _Suphis puncticollis_
Crotch, which he described therein) "might receive the name Suphisellus".

This hypothesis proved correct, with Colpius becoming synonymized under
Suphis later, and Sharp in 1880 recognized that Crotch was correct in
pulling the American species out of Suphis, but did not recognize
Suphisellus as a valid name (he concluded it was informal), instead
including the three species above into the new genus _Canthydrus_ Sharp.

Zimmermann in 1921 came along and moved the American species of Canthydrus
into the genus Suphisellus, and described four new species in addition, the
first being _S. variicollis_ Zimm. - no mention was made of Crotch's work.

Guignot in 1946 came across this and designated variicollis as the genotype
of the genus _Suphisellus_ Zimm.

The final published word came from Leech in 1948, who missed Guignot's
paper and instead recognized that Crotch's original proposal of the name
was valid, and therefore designated _Noterus bicolor_ Say as the genotype
of the genus _Suphisellus_ Crotch.

My colleague has concluded that Leech was essentially correct, and the name
Suphisellus should be credited to Crotch, although he feels that Guignot's
designation of variicollis Zimm. as the genotype takes precedence over
Leech's assignment.

Myself, I'd have to pull out the book to make sure such an informal
designation as Crotch's would be valid - after all, he never *formally*
placed any species in _Suphisellus_ - though the name was put in print
there first. At any rate, assuming that the genus name *is* credited to
Crotch, then, *I* would have assumed that since variicollis was not one of
the three species Crotch originally included in the hypothetical genus, it
could not be designated the genotype, and I would consider Leech's
designation of bicolor to stand. If, however, the name is credited to
Zimmermann, then Guignot's designation of variicollis would. What do folks
think about this??

One additional thing...in *either* case, do we presume correctly that the
species Crotch described would be listed as Suphisellus puncticollis
(Crotch), given that he described it in the genus Suphis?

Thanks,

Doug Yanega       Illinois Natural History Survey, 607 E. Peabody Dr.
Champaign, IL 61820 USA      phone (217) 244-6817, fax (217) 333-4949
 affiliate, Univ. of Illinois Urbana-Champaign, Dept. of Entomology
  "There are some enterprises in which a careful disorderliness
        is the true method" - Herman Melville, Moby Dick, Chap. 82