recognizing "species group" or "species complex" suffixes as indicators of infrageneric groupings

[dimus]

created by @mtholder at https://gitlab.com/gogna/gnparser/-/issues/55

(First off: thanks for all the work on gnparser! It is shockingly efficient, precise, and complete.)

This may well be out of scope, but I am working with some data that uses names like

"Pelomedusa subrufa species complex"

between the genus-level and the species level.

In gnparser "v0.7.4" (running on ubuntu 18.10), this " species complex" is just treated as a part of the unparsed tail, and the name info reflects just the info for "Pelomedusa subrufa"

[dimus]

created by @dimus at https://gitlab.com/gogna/gnparser/-/issues/43

This is a murky region of ICZN, so I tried to postpone it until someone has need for such name strings :)

I guess this issue should be part of other cases like:

• Aus bus species group
• Aus bus species complex
• Aus (bus) cus #this sadly sometimes is just a typo of subgenus, and sometimes a super-species grouping as I understand.

I am trying to wrap my head around how to implement this in a general way.

For example for subgeneric names we have two different formats:

• Aus subgen. Bus
• Aus (Bus)

Both are normalized to the same canonical form:

short version: Bus

full version: Aus subgen. Bus

One problem with names like Aus bus species group is a 2-word rank, which will create trouble for canonical forms.
So one suggestion would be to treat such names like this:

Aus bus species group becomes:

normalized: Aus supersp. bus

canonical simple: Aus bus

canonical full: Aus supersp. bus

in details rank: species group

Aus bus species complex is treated similar way, except in details rank will be species complex

Aus (bus) is treated the same way as Aus bus species group

Aus (bus) cus becomes:

Canonical simple: Aus cus
Canonical full: Aus supersp. bus cus

details: species cus gets field superspecies: bus

Does it make sense?

[Archilegt]

Recommendation: Follow the ZooCode article 6.2 to the extent possible.
Article 6. Interpolated names.

6.2. Names of aggregates of species or subspecies. A specific name may be added in parentheses
after the genus-group name, or be interpolated in parentheses between the genus-group name and the
specific name, to denote an aggregate of species within a genus-group taxon; and a subspecific name
may be interpolated in parentheses between the specific and subspecific names to denote an aggregate
of subspecies within a species; such names, which must always begin with a lower-case letter and be
written in full, are not counted in the number of words in a binomen or trinomen. The Principle of
Priority applies to such names [Art. 23.3.3]; for their availability see Article 11.9.3.5.

Recommendation 6B. Taxonomic meaning of interpolated names. An author who wishes to denote
an aggregate at either of the additional taxonomic levels mentioned in Article 6.2 should place a term
to indicate the taxonomic meaning of the aggregate in the same parentheses as its interpolated
species-group name on the first occasion that the notation is used in any work.

Example. In the butterfly genus _Ornithoptera_ Boisduval, 1832 the species _O. priamus_ (Linnaeus,
1758) is the earliest-named member of an aggregate of vicarious species that includes also _O. lydius_
Felder, 1865 and _O. croesus_ Wallace, 1865. The taxonomic meaning accorded to the _O. priamus_
aggregate may be expressed in the notation "_Ornithoptera_ (superspecies _priamus_)", and the members
of the aggregate by the notations "_O._ (_priamus_) _priamus_ (Linnaeus, 1758)", "_O._ (_priamus_) _lydius_
Felder, 1865", and "_O._ (_priamus_) _croesus_ Wallace, 1865".

[Archilegt]

Treat interpolated names of subgenera (Art. 6.1) in a consistent way to the extent possible.

[Archilegt]

The key question for Pelomedusa subrufa species complex is not how to handle the tail but how to handle the expression itself.
A species complex is:
In biology, a species complex is a group of closely related organisms that are so similar in appearance and other features that the boundaries between them are often unclear.
https://en.wikipedia.org/wiki/Species_complex
Which does not automatically equate to a name used for an aggregate of species of subspecies which is denoted by a term such as "superspecies" in the sense of the ZooCode.

aggregate, n.
A group of species, other than a subgenus, within a genus; or a group of species within a
subgenus; or a group of subspecies within a species. An aggregate may be denoted by a
species-group name interpolated in parentheses [Art. 6.2].

The authors referring to a "species complex" are not necessarily aware of the rules of the ZooCode governing the availability, priority, and correct formation of a name for an aggregate of species, e.g., 6.2, 10.4, 11.9.3.5, 23.3.3, and the related definitions in the Glossary.
Therefore, "species complex" can be taken off of the name and just retrieving the stated name is a good result. Same treatment to be applied to Aus bus species complex.

[Archilegt]

For the example for subgeneric names with two different formats:

• Aus subgen. Bus
• Aus (Bus)

Normalization to canonical form:
short version: Bus
full version: Aus (Bus) #Parentheses are important here as per article 6.1 of the ZooCode.

For the example of aggregate of species:
Aus bus species group becomes:
normalized: Aus (superspecies bus) #Parentheses are important here as per article 6.2 of the ZooCode.
canonical simple: Aus (bus)
canonical full: Aus (superspecies bus)
in details rank: superspecies or aggregate of species #Avoid using species group, which is defined as:

species group, n.
In the zoological classification, the lowest-ranking group of taxa the names of which are
regulated by the Code. The species group includes all taxa at the ranks of species and
subspecies [Art. 45.1].

For the example of the species name with the interpolated superspecies name:
Aus (bus) cus becomes:
Canonical simple: Aus cus
Canonical full: Aus (superspecies bus) cus #Parentheses are important here as per article 6.2 of the ZooCode.

details: species cus gets field superspecies: bus

[dimus]

Thank you @Archilegt for detailed explanation!

Aus subgen. Bus and Aus (Bus) are already covered:

Id,Verbatim,Cardinality,CanonicalStem,CanonicalSimple,CanonicalFull,Authorship,Year,Quality
0f831955-2e8b-52f0-8b05-aa3a825c629e,Aus subgen. Bus,1,Bus,Bus,Aus subgen. Bus,,,2
784a3c10-e635-5cdb-a4bb-0f29c1a167ce,Aus (Bus),1,Bus,Bus,Aus subgen. Bus,,,2

[dimus]

@Archilegt and @proceps, do I understand it correctly that names like

Aus bus species complex
Aus bus species group
Aus bus superspecies
Aus (bus)

all can be converted to:

normalized:           Aus (superspecies bus)
canonical simple:  Aus (bus)
canonical full:       Aus (superspecies bus)

?

Can it be that for names iike Aus bus species group people actually do mean coordinated name according to ICZN?

[Archilegt]

@dimus, "species complex" should not be converted. See this comment.
The rest can be converted, but I have often found that the authors don't refer to ssuperspecies anywhere in the text, which is what should be happening after 2000. I have also found cases where the authors state that their species group corresponds to whatever subgenus but they still use "species group". In those cases, converting "species group" to superspecies instead of leaving it blank could result in confusion. But it may be that for most cases will be fine. You can set gnparser to convert "species group" to superspecies.

[dimus]

@dimus, "species complex" should not be converted.

Oups my bad. So I guess we can continue to treat species complex the same as before?

http://parser.globalnames.org/?format=html&names=Aus+bus+species+complex&with_details=on

Parser now ignores 'species complex' and creates a warning about an unparsed annotation.

And for all kind of superspecies, as I understand, no warning is needed, and we would trust the author of a database or a checklist that they took care to write down the name as it supposed to be written. The same way as we trust that 'Aus bus' meant a binomial of genus 'Aus' and specific epithet 'bus' even if it is not a real name.

http://parser.globalnames.org/?format=html&names=Aus+bus&with_details=on

I mean that parser does not try to verify if a name is real, just tries to figure out what differnt parts of a name-string would mean in case if such string would be a scientific name. If the structure of a string is compatible with rules, its components will be parsed.

[proceps]

@dimus I believe it should be treated as superspecies, but the problem is that authors not always follow the ICZN rules for coordinated names, and many taxonomists believe that 'species group' and 'species complex' are informal grouping. 'Superspecies' is not clearly defined by ICZN, neither do complexes and groups. The only thing that ICZN says, that different ranks are permitted, and if the rank falls within species group rank, it should follow the rules for this group. So, I would probably agree, that it is safer not to treat complexes as superspecies, just because the taxonomists do not follow those regulations consistently. At the same time, I know that the new edition of ICZN will probably reduce the significance of species complexes even more.

[dimus]

Thanks @proceps, very helpful comment