As far as I understand it, that’s not possible because the microspecies (clone lines) are reproductively isolated due to being incapable of sexual reproduction. Unless some of the introduced species are sexual species, which I’m not aware of. European cities could have endemic microspecies though since they have the sexual species that are producing these clone lines. I was reading something about Rubus taxonomy from Europe the other day and it said that clone lines with ranges less than 50 km shouldn’t be described as species, I guess because then things would get out of hand.
It’s not completely clear to me how Flora of North America defines T. officinale which makes it hard to say how accurate it is, but it’s roughly synonymous with section Taraxacum, which is how European/international taxonomies define it.
FNA lists 3 introduced Taraxacum sections (Taraxacum, Erythrosperma, Palustria) and 3 native sections (Spectabilia, Borealia, Arctica).
The BC paper added Borea, Boreigena, Celtica, Hamata, and Naevosa as newly discovered introduced sections, with Celtica and Hamata apparently being relatively common in southern BC near the coast. The macrospecies names for these (e.g. if recognized by FNA) would be T. celticum and T. hamatum.
Right, well the microspecies may exist in some taxonomy treatments but what I’m also saying is there’s no value in trying to use them on inat since they basically can’t be identified and thus can’t be tracked here. Interesting from an evolutionary biology standpoint but not useful for this website.
I mean, they’re probably easier than plenty of fungi or insects that need to be dissected to inspect obscure hairs or spores to identify sometimes even to genus, but yes. Hence the focus on sections.
Theoretically a dandelion observation that shows a good view of the leaves should be identifiable to section/macrospecies, and if it shows leaves and the bracts under the flower it could be identifiable to microspecies. The only issue is that there’s only like one person in North America who’s confident in identifying to that level. And without knowledge of even which sections exist in which states it’s hard for others to start. There’s maybe 2-3 people on iNat in Europe, but they’re not big identifiers.
If the images wouldn’t be sufficient to distinguish T. officinale and T. erythrospermum, which are both accepted by FNA, then it should probably be at genus level anyway given that both are common and widely distributed.
Maybe we need to set inat to automatically go to research grade in section (with two IDs)for some of these. But I also think it’s a horrible idea to do microspecies on inat. That being said I’ve already posted plenty in here so am going to leave it at that for now rather than have a long debate that will be tedious to others.
I don’t think doing this indiscriminately to all dandelions is a good approach. It depends on location, the identifying features shown in the pictures associated with the observation, and the keys available for that location. For example, the floras for the southeastern US only have two species listed (with a note that more may exist), T. officinale and T. erythrospermum, and a key how to distinguish them. If an observation shows the characters noted in the key, I don’t think it’s appropriate to bump it back to genus just because there are different keys published for other locations or based on the assumption that things are more complicated and current keys are insufficient. However, if the pictures show none of these features (e.g. no leaves or seeds), that’s different as it cannot be keyed out beyond genus, at least not using the key in the southeastern US flora. It seems more work needs to be done on sorting this out for the eastern US but that’s way over the heads of the average iNat observer or identifier. Meanwhile, I think using existing keys should be acceptable and not be bumped back to genus.
If we applied dandelion logic more broadly, every cultivar of edible banana would be its own (micro)species. And how many (micro)species of breadfruit would there be?
I’m not principally opposed to this solution or anything; it seems pretty reasonable to me. Given that a species concept can be finer or broader for different fields the only real difference is what name we give to each level (and the fact that iNat decides only some levels become RG automatically, if that’s a valid concern for taxonomy decisions).
The problem with applying it for iNaturalist is that it means iNaturalist curators would be creating a new taxonomy for Taraxacum that no one else has done in primary or secondary literature, which as far as I know is not supported by the site.
But the other options that keep T. officinale are not having the microspecies at all by lumping them in (which European users won’t like at all) or having two contradictory systems in place at once (also not supported by the site).
It is mostly the section Taraxacum (=Taraxum officinale) that is such a mess due to the apomicts. But determination to the section level is quite reasonable. And should be doable. Marking the other sections as T. officinale is wrong.
BTW Seek apparently tries to identify some dandelions as T. erythrospermum even when it is clear that there is no point - no details, wrong season, bad photo… That is an actual problem. Not people who try to go into detail.
Similar to CV making people to mark all Alchemillas as Alchemilla millefolium. Even in the autumn season when they really cannot be distinguished. The Complex taxon exists on iNat, but CV never suggests it.
Both of those have a descendant species that did meet the criteria and are included in the model. As far as I know, higher taxa (“leaf nodes”) are only included if none of their descendant species meet the qualifications.
It’s a pity, because likely most of thousand+ of Leptidea sinapis are random ids as those species are ided by genitals and complex is what should be proposed by cv, so we need someone to review all of those and then in a next model complex would be added, right?
They really should add the symbols / text for “this taxon is included in the model” or “pending” to the about pages for all taxa included in the CV model whether they’re species or a higher leaf taxon. When I visit the Chrysoperla carnea complex page it has no indication that it’s included in the CV model, meaning we just have to infer that after the fact like this. That’s too opaque.
Ok and thanks, but this really doesn’t seem as hard as he’s making it out to be (copied below). I’d respond there but that thread is closed.
Hopefully once we figure out how to display it in a way that increases clarity without adding confusion, we’ll add a label to indicate when genus or family is in the model because none of its children are. We just haven’t quite worked that out yet.
It’s not like this is a matter of translating the Septuagint or writing a Supreme Court decision, we just need an automatic tag that says “this taxon is included in the CV model”. If people want to learn more about what that means for the precise taxon level they’re at, it could link to the page explaining the CV so they can dig into it. I’d guess only a small fraction of users actually even care about the nuts and bolts of how this works. Or drafting a 2 sentence explanation blurb that explains the status like they did for the species level included-vs-pending tags would take 15 minutes max, not 3 months.
Now, we’re all busy, and I’ve certainly also got tasks on my long-term to-do list that would have only taken 15 minutes to clear. So I mean no ill will to anyone in this. (Somebody recently jumped on me for questioning the narrative coming from the admins on here so I’m trying to stay out of trouble…)
I do not understand this, I get CV suggestioms for higher taxons all the time. Saying we are pretty sure it is this genus or family and then a list of possible species, often not all of them belonging to the taxon it was so sure about.
Complexes could be shown in the same way.
Even better, some in-complex species suggestion could be suppressed and the complex shown instead. That would make much better sense in many cases. For some taxons the map coverage is a joke because the determination is completely unreliable. The only consolation is, that the plant with the ID’d name probably also grows somewhere near anyway.
Based on my understanding, if the CV knows 1 species in the genus/complex, then it will never suggest the genus and always suggest the species. If the CV knows 2 or more species in the complex/genus/family/etc., then it can suggest both those species and the higher levels (if it can’t decide between those species).
The issue here is that there are plenty of cases where there is one distinctive species in a genus and many species that are mostly very similar to that species but not identifiable (or none are identifiable but field guides just use one name to refer to all of them, or none are identifiable in one location but only one species exists at another location, etc.).
(maybe this discussion starting from raymie’s comment should be moved to this thread or a new one?)
