Diplacus dilemma in California area
The common ex-species Mimulus aurantiacus that occurs in California and Baja California was split into several species of Diplacus under the new taxonomy. However, all of the many M. aurantiacus observations are now classified as D. aurantiacus when in reality most of them are one of the other species. See this post: https://www.inaturalist.org/posts/18612-observation-of-the-month-red-sticky-monkeyflower-diplacus-puniceus-phrymaceae as well as this paper: http://www.phytoneuron.net/PhytoN-sectDiplacus.pdf and other references linked in the post above. While some may know i am an unapologetic lumper when it comes to plant taxonomy, i don't have a problem with these species divisions, but am worried about how we deal with them.
I may have missed out on some of the procedure here, in terms of sensu latu vs sensu stricta but I'm concerned about cases where a large 'species' is split, in that i don't think we should just be lumping all on the observations into one species. Tossing them back to genus is very problematic as well, though. I am wondering... is there a way to move the previous M. aurantiacus into a separate place until the community can review them, either 'D. aurantiacus sensu latu' or else a grab bag subgenus? Otherwise... we've created a huge slug of wrong IDs, and because of how community ID works it's a huge slog requiring multiple people to get these swapped over to the correct ID, especially in cases of multiple IDs by no longer active users.
Thoughts? @loarie ? I am not sure who else is working on California taxonomy? @bouteloua I know this isn't your geographic area of expertise but you've been pretty involved with this process. Anyone else? @jdmore ? Is the policy to toss the splits into one species like this? Or are there just very few of them so it's a special case? Apologies if this was already addressed, i haven't had as much time to devote to this as I'd like...
Comentarios
I'm not familiar with this group but given the swaps of multiple infraspecifics of M. aurantiacus that were raised to Diplacus species, definitely seems like M. aurantiacus should have been split instead of 1:1 swapped. Does the below look correct?
M. aurantiacus -----split into---->
D. aurantiacus
D. aridus (M. aurantiacus var. aridus)
D. × australis (M. aurantiacus subsp. australis)
D. grandiflorus (M. aurantiacus var. grandiflorus)
D. × lompocensis (M. aurantiacus subsp. lompocensis)
D. parviflorus (M. aurantiacus var. parviflorus)
D. puniceus (M. aurantiacus var. puniceus)
D. longiflorus (M. aurantiacus var. pubescens)*
*Not sure about D. longiflorus. M. aurantiacus var. pubescens is treated as a synonym of the nominate D. aurantiacus on POWO, though it's treated as a jr. synonym of D. longiflorus in Tulig & Nesom 2012.
From a kingdom over, a similar problem happened when a mishandled taxon changed caused the misclassification of thousands of elk (Cervus canadensis) and red deer (Cervus elaphus) observations (https://www.inaturalist.org/taxon_changes/3993). The only real solutions required atlasing and administrative involvement...
Yeah, that looks right cassi. Unfortunately (at least from this angle), this is a bunch of narrow california endemics that do intergrade on the edges on a county-scale basis, so atlasing won't totally solve it. Also with such a new division I'd worry any generalization of which one was where by location may end up burying small scale range variations that we would like to know about. I am also unsure what to do about it when others have already added IDs in a big push to try to fix some of them. Maybe it's an admin-level issue at this point? A bunch of the ones in san Diego got changed to x austrailis which i guess is the correct name here.
As mentioned in one of these threads i'd favor creating a 'holding bin' or subgenus for all of these at least until the community could add the real IDs, but I think that approach got vetoed, right? Too bad they weren't treated as subspecies or vareties considering they apparently freely intergrade and many of them are hybrids anyway. but unfortunately that's how botany is going these days and we have to figure out how to get iNat to adapt to that and still follow the mission of connecting people with nature and allowing them to collect and share data. I personally don't think we should bump them all to genus either considering all the other Diplacus species on iNat. https://www.inaturalist.org/observations?nelat=49.330225572096936&nelng=-44.811039716005325&place_id=any&swlat=8.501849359193972&swlng=-161.88135221600533&taxon_id=157750&view=species But i know others are more involved with this taxonomy push than I am...
Sorting them into their sections would help some.
yeah, good point, that was really helpful with the sedges, i use those a lot...
This is a subset of the problem that iNaturalist hasn’t figured out a good way to handle cases where you know observation X is one of a small number of species in a large genus.
I run into that a lot with aphids. Observation X is one of two species in a genus of 200. Pretty much all I can do is tag as the genus then put on a “holding bin” tag. It seems like I should be able to do better but all the proposals I have seen to fix this turn into a complicated mess.
I suspect a lot of these monkeyflower observations were previously identifiable to species but now only to genus which makes it worse.
@bouteloua I looked at the key for D. longiflorus (M. aurantiacus var. pubescens) and I don't like the fit.
This might be one of the most fluent species, new papers indicate more changes. Probably too many working on these plants. And they have to produce a new paper.
Maybe @jrebman can help us out.
I don't like it that they got rid of the var. that makes it that much harder to ID from a picture. So sadly enough you might go back to the Genus level after this change.
M. aurantiacus var. pubescens https://www.calflora.org/cgi-bin/species_query.cgi?where-taxon=Mimulus+aurantiacus+var.+pubescens
Shows an awfull lot of alternate Names.
I guess we could consider ignoring POWO for this and shoving them back into Mimulus but I think a lot of people wanted some of the changes, as with Erythranthe, and i don't think there's a way to do that with just one genus within this group?
If we accept for the sake of discussion that the split of M. or D. aurantiacus into several species is valid, then the existing IDs were just as wrong (or right) under Mimulus aurantiacus as they now are under Diplacus aurantiacus.
Therefore to me the mechanics of the taxon swap in iNaturalist is not the issue here. Splitting a species when new species are recognized will always be a labor-intensive effort on the part of the iNat observers and IDers involved (unless there is correlation with previously recognized and IDed subspecies or varieties). Whether it happens in Mimulus or in Diplacus is beside the point.
As @charlie already knows, I am for recognizing whatever species the science supports, whether the taxonomic consequences end up being lumping or splitting. Of course, as the science develops more powerful tools (or humans just become better observers - see https://scholarship.claremont.edu/aliso/vol30/iss1/6), the trend is to reveal species that had previously gone unnoticed by less powerful tools, resulting in splits. Usually correlated morphological markers are also revealed that aid with identification of the new taxa. Usually, but not always, and then the resulting impacts to practical application of new species units become a real and thorny issue.
Correct or not, the M. aurantiacus species concept included all of these new species, where as D. aurantiacus is more narrow and tossing them all into this bin implies precision we don't have because D aurantiacus is not the same thing as M aurantiacus at all, it's a much narrower concept...
I don't mean to re-ignite my lumper-splitter discussion here, so I'm just assuming they should be split for the sake of fixing the other issue which is a bunch of things having been moved to a new species when most of them don't match the criteria for such. If we leave them as is, it makes it really hard to track the actual D aurantiacus and also makes it very difficult to override existing IDs to fix the other ones. In terms of what taxonomists should do i could go on and on about that but will spare this working group from being cluttered by that.
Points well taken @charlie. As I understand it (not being an iNat Curator), the former Mimulus aurantiacus was a "non-leaf" taxon, containing infraspecies that mostly corresponded to the newly recognized "leaf" taxa now being ranked as species under Diplacus.
So essentially the change is forcing the community to refine our IDs out to the "leaf" taxa, which wasn't necessary before because one could use Mimulus aurantiacus (sensu lato) as a catch-all.
Now that Diplacus aurantiacus is a narrower leaf taxon, Diplacus section Diplacus would need to be used as the next-broader catch-all if and when needed. But it looks like options for infra-generic ranks currently do not exist in iNaturalist.
No doubt this has been discussed somewhere else, but do we need to provide for subgenus, section, and subsection ranks for taxa in iNaturalist to be consistent with the Code of Botanical Nomenclature, like we already are above the rank of genus?
Sections are definitely an option available on iNat. If someone can provide a source that lists Diplacus species and their sections, or copy a list here, I'm happy to sort them into their sections.
@jdmore not sure what you mean about consistency
@bouteloua Apparently with my limited view into iNaturalist I was under the misunderstanding that no infrageneric ranks are available for taxa. If subgenus, section, and subsection (in that hierarchy) are available, then that would be consistent with (in conformation to) the ranks permitted in the Code of Botanical Nomenclature.
I'll see if I can find a comprehensive sectional classification of Diplacus (and maybe Erythranthe too) for you!
@bouteloua I'll do you one better - here is the whole family Phrymaceae with species assigned to their sections as of 2012:
http://www.phytoneuron.net/PhytoN-Phrymaceae.pdf
There may be a few species newly described or recognized after 2012 that are not included, in which case tag me and I'll hunt down the literature supporting their sectional placement if possible.
Oh very neat @jdmore and very interesting stuff. it sounds like putting them all in sect diplacus for now would allow them to then be classified to species by the community when possible. I'm not sure how that plays out with cases where people have already added other IDs though...
@charlie If one just replaced the Mimulus aurantiacus --> Diplacus aurantiacus swap with a Mimulus aurantiacus --> Diplacus sect. Diplacus swap (if that is possible), would that not just replace all the old M. aurantiacus IDs without affecting any other IDs that had been added before or since?
One would need to make sure that only species-rank Mimulus aurantiacus IDs are replaced. IDs at infraspecific rank under M. aurantiacus should be swapped directly with their new species-rank IDs under Diplacus.
makes sense.... i'm not sure though, does the taxon swap register as a new disagreement? It seems to reset the ID
Diplacus is sorted! https://www.inaturalist.org/taxa/157750-Diplacus
and Diplacus sect. Diplacus species are atlased.
Ideally this taxon swap could be reversed and a new taxon split, along the lines of my first comment here, could be made in its place. IDs from the split would be automatically assigned to species based on the atlases, which are currently refined at county level. When atlases overlap (i.e. 2 or more species occur in the county the observation is in), the common ancestor ID is assigned instead (Diplacus sect. Diplacus).
*"does the taxon swap register as a new disagreement?"
no
You can read more about splits here: https://www.inaturalist.org/pages/curator+guide#changes
Thank you @bouteloua for all the hard work!
If reversing and replacing the taxon swap with your split can be done, that sounds like the way to go. I would just suggest swapping any infraspecific IDs under Mimulus aurantiacus to their species equivalents in Diplacus first, to preserve those users' IDs, even if they disagree with the atlas limits. Maybe that's what you already had in mind...?
I definitely think reverting and going with that split is best like @bouteloua said. I'm a little reluctant about the atlases since these are pretty newly divided species and it may be we don't really know their range well enough to do that. But... given the thousands of observations, it may still be our best option
The infraspecific M. aurantiacus were all already swapped into their respective Diplacus spp.
In southern California theres so much overlap of atlases that most will go to section anyway.
Only iNat staff can do reverts on taxon changes. @loarie is it too late to do so here? (see my comment 3 up from this one)
Unfortunately reverting isn't perfect (see https://github.com/inaturalist/inaturalist/blob/master/app/models/taxon_change.rb#L274 ) and the imperfections become worse the more time that elapses between the commit and the revert. Since I made that swap several weeks ago now, I worry that reverting will create more problems than it will solve.
That said, there's nothing wrong with making a new inactive D. aurantiacus (sensu stricto) and splitting the current active D. aurantiacus (sensu lato) into:
D. aurantiacus (sensu stricto), D. aridus, D. × australis, D. grandiflorus, D. × lompocensis, D. parviflorus, D. puniceus, D. longiflorus
The downside is that there will be more non-current and new generated current IDs from this path of
M. aurantiacus (sensu lato) -swap-> D. aurantiacus (sensu lato) -split-> D. aurantiacus (sensu stricto), etc. than had I had the foresight to do:
M. aurantiacus (sensu lato) -split-> D. aurantiacus (sensu stricto), etc. directly
But my gut tells me that this will be less dangerous that getting to cavalier with the emergency revert tool. What do folks think about that - will it achieve whats intended?
I'm ok with it as long as it's really clear it's sensu lato and there is some sort of note in there or something to explain what is going on. I'd prefer the revert but if we can't do it, we're stuck with it i guess.
what a mess taxonomy is!.
well the split will generate updates/new IDs with links to the split which can include more information about what went down and a link here.
Our philosophy with iNat and taxonomy at the moment is that: taxonomy will always change so we should embrace it and try to build a system that can accommodate lots of change. I think we have an emerging issue where all the accumulated made-not-current IDs and newly generated IDs are becoming a bit of a visual clutter problem which we can/should be able to clean up on the display side. But its not yet a database size issue yet in terms of the ID table getting to big or anything. Re-IDing things is a pain, but iNat is a bit more streamlined than sites (e.g. eBird where I think one staff member is responsible for manually updating every observation on the site when they do their annual taxonomy updates)
yeah, there are always going to be changes and overall the system works as well as anything does. I don't think I'll ever embrace the changes, though this mimulus one is more interesting than annoying really.
https://www.inaturalist.org/taxon_changes/39636 can y'all take a look and see if we're missing anything here
looks good to me cassi
Couple of minor notes:
Mimulus aurantiacus var. pubescens still needs to be dealt with. Currently in iNat the name is mapped to "Diplacus aurantiacus var. pubescens" which, as far as I can tell in IPNI, is a non-existent combination. To stay consistent with the rest of the sect. Diplacus taxonomy, I suggest we deviate from POWO and follow Tulig & Nesom in assigning it to Diplacus longiflorus. I suspect it was just an oversight on the part of POWO that they did not already do this.
The common name Bush Monkeyflowers no longer applies to the genus Diplacus overall, but only to sect. Diplacus, and should be moved there.
Looks like you covered everything in the split Cassi -- thanks again for all your work on this!
Oh, but also note the statement in Tulig & Nesom: "Plants cited and mapped as Mimulus aurantiacus var. pubescens by Thompson (2005) from Fresno, Tulare, and Kern cos. are identified here as Diplacus calycinus."
This would be the concept of var. pubescens that was followed in the Jepson Manual, with the name apparently being misapplied to the inland portion of its distribution.
So looks like var. pubescens should actually be managed as its own split: inland plants to Diplacus calycinus, and coastal plants to Diplacus longiflorus.
@bouteloua at a glance that looks good, but did we figure out a way to make the aurantiacus say sensu latu or something? Sorry if i missed that part of the discussion
@jdmore Misapplied taxonomically or misidentified? I don't think taxon splits should deal with the latter. But I guess the two could be blurry here.
@charlie sensu stricto? Common names shouldn't hold that information (see policy "just the names please") but I can add that info to the taxon split.
@jdmore So looks like var. pubescens should actually be managed as its own split: inland plants to Diplacus calycinus, and coastal plants to Diplacus longiflorus.
I wouldn't go that far, the number of plants on the paper is not very high.
Inland (Anza-Borrego Desert) we have (Jepson key)
Calyx tube glabrous ..... var. aurantiacus D. aridus
40' Calyx tube puberulent to hairy ..... var. pubescens ?? D. longiflorus
They don't seem to be well separated, but we've not seen enough to be sure.
To determine D. calycinus you need some measurements that can't'be done with a picture.
It states D. calycinus as a form of D. longiflorus, I hate it not being a var. as it was before that's much obvious and easier.
@efmer @bouteloua by "inland" I meant the distribution Tulig & Nesom referred to in Fresno, Tulare, and Kern counties. Sorry for any confusion.
I'll leave it to you Curators as to whether var. pubescens should be managed as a split or not. But it looks to me like the two taxa that have been included previously under that name in the Jepson Manual treatment (the inland plants erroneously according to Tulig & Nesom -- misidentified, if you like) are geographically separated and "atlas-able."
I don't know about the 'just the names' policy @bouteloua does that affect how we deal with duplicate name of subspecies? I guess that's another issue. In terms of the sensus, i guess it's better to use the section diplacus entity anyway.
If we can skip splitting things off that can't be distinguished by photo, i'd always lean towards that. It doesn't get us anything and only does harm. Not sure why some of these aren't varieties anyway. Splitters.
Regarding Mimulus (Diplacus) aurantiacus var. pubescens, I was about to propose it just be swapped 1:1 into D. longiflorus. But Tulig & Nesom 2012 suggest that D. longiflorus flowers are salmon-colored and literally none of the observations of var. pubescens on iNaturalist are salmon colored. So considering that and the past confusion in Jepson, probably best to do the split after all, which will kick them up to Diplacus sect. Diplacus.
@bouteloua Something is wrong with that key.
The description states paler yellow than all the other.
Another paper from a later date http://www.phytoneuron.net/2013Phytoneuron/66PhytoN-SubgDiplacus.pdf
Shows one that is pale YELLOW and SALMON on page 7.
So the key should probably be Corollas salmon (pale yellow San Diego county)
Taxon change #1: Diplacus aurantiacus var. pubescens split into Diplacus
Taxon change #2: Diplacus aurantiacus (sensu lato) var+subsp. aurantiacus merged into Diplacus aurantiacus (sensu stricto)
Taxon change #3: The Diplacus aurantiacus split
Link to help identify rank=section Diplacus observations
On a different, but related topic: can any of you suggest the best way to distinguish Diplacus puniceus from Diplacus x australis? There is so much variation in D. x australis (being a hybrid classification and all) that I'm not sure what the cut-off should be, particularly in San Diego, Orange, and Riverside counties.
@flower_prof If answers to your question are not in one or more of the Phytoneuron papers linked in the previous comments above, then I would not know where else to look, at least until the upcoming Flora of North America treatment is in print.
probably they shouldn't be different species at all. Can we just treat them as subspecies?
just kidding.
ok kinda not kidding. sigh.
someone put a proposed deviation in spreadsheet were in the process of replacing with taxon frameworks that says "Mimulus aurantiacus var. pubescens ----> Diplacus aurantiacus" and references this post. Right now both iNat and POWO seem to match with monotypic Diplacus aurantiacus. Is that what is intended? Deviations should describe how we are (and are intending to maintain) deviations from POWO. If we're matching POWO no need for a deviation. If we intend to deviate from POWO, the protocol would be to make the change and then update the Taxon Framework Relationships as needed
POWO fixed the incorrect synonymy for that variety on their end.
awesome thanks - so looks like we're all good
Yep, and both concepts of D. aurantiacus--iNat and POWO--are sensu stricto.
Mimulus aurantiacus var. pubescens = Diplacus longiflorus
Mimulus aurantiacus var. aridus = Diplacus aridus
I don't see this paper referenced in the above discussion, but it puts several of the taxa raised to species back into varietals of D. aurantiacus: http://www.phytoneuron.net/2019Phytoneuron/03PhytoN-DiplacusVars.pdf
it was published after these taxon changes were committed. probably better to flag the taxa and start a new conversation if you're proposing a deviation from Plants of the World Online
a lumper was allowed to publish? i'm so excited
We should impose a rule not change the same plant more than once every 5 years.
All these unnecessary changes from one synonym to another drives everyone crazy and will introduce errors.
Just wait until the dust settles.
Since the names were not completely sunk, just moved to a different rank, I don't see the harm in leaving them at their current rank in iNaturalist for a while. As has been noted in at least one Forum thread, choice of rank can be fairly arbitrary at times. Although in this case I think the author justifies his choices of varietal rank quite well, it's likely to be a while before they become incorporated into widely used secondary sources.
Agregar un comentario