Flag for Taxon: narrowleaf wild leek (Allium burdickii)
| Flagger | Content Author | Content | Reason | Flag Created | Resolved by | Resolution |
|---|---|---|---|---|---|---|
| pdabell | loarie | narrowleaf wild leek (Allium burdickii) |
Allium burdickii is a synonym of the accepted name, A. tricoccum var. burdickii. |
Feb. 17, 2023 17:22:00 +0000 | rynxs |
see comments |
Comments
@pdabell I just changed the source to: NatureServe Explorer: An online encyclopedia of life. Version 7.1
Again, the accepted name in POWO is based on the faulty and outdated treatment in Flora of North America. And IPNI lists both names without rejecting either. POWO is not the only naming authority.
We should probably email POWO about this, they are receptive to changes.
@zihaowang , although Nature Serve is more authoritative than a master's thesis, I don't think the organization is a botanical authority on taxonomic nomenclature, and, although I might be mistaken, I don't think iNaturalist bases its use of accepted taxonomic names on what Nature Serve uses.
@wildlandblogger , I agree. Perhaps some of the botanical authorities, of which I am not, on this issue will do so.
I do agree that they appear to be separate at the species level. (While on the topic, I definitely need to lobby to get Allium mutabile recognized as a full species)
I've added a deviation for now, linking back to this flag. Please report back on POWO's response.
There may be four species in the A. tricoccum complex, so this swap to species level should probably be reversed and postponed the other two elements are split and described.
How do the potential four species split out? If two come from the burdickii group and two come from the tricoccum group, I'd say keep as is, but if it's three from tricoccum group then I'd do as you suggest.
As is, Eastern Onions are all over the place on here.
We have, in the canadense complex, Allium fraseri as a separate species, Allium canadense var mobilense as a variety (not as its own species, but it is valid as one under the name mutabile), Allium canadense var lavandulare (also a separate entity), Allium canadense var ecristatum (which lacks a valid species name), and Allium canadense var hyacinthoides (valid as Allium hyacinthoides).
Allium oxyphilium and Allium alleghenense are split out as deviations from POWO on here (in my opinion, validly so), and there's at least four other entities in the Allium cernuum complex that are undescribed (the Nashville Basin entity that's occasionally called Allium stellatum by people because it shares characteristics with that species, Allium neomexicanum, the coastal Carolina entity, and the Cumberland Escarpment entity).
POWO generally follows the very conservative FNA treatment from 2002. I would deviate from that inadequate treatment personally and separate out Allium canadense's varieties as full species, pull back on the burdickii treatment until that's sorted out, and hold off on changing anything about Allium cernuum's complex until someone does proper taxonomic work on that group. I will be flagging Allium canadense with a copy of this post.
@wildlandblogger one comes from the burdickii group, the other is pretty much perfectly intermediate
Oof, an intermediate- yeah I agree with you then.
Here's the paper discussing them: https://etd.ohiolink.edu/apexprod/rws_etd/send_file/send?accession=ohiou1534064390052709&disposition=inline
Page 38 is probably the best one for those just wanting to see what the undescribed entities look like at a glance. "Highland green ramps" is apparently identical to A. tricoccum sensu strictissimo when vegetative, aside from completely lacking anthocyanin. "South green ramps" is apparently identical to burdickii when vegetative, apart from having deeper set bulbs (lol, we're may have to dig 'em up to ID them).
The burdickii-like entity may not pan out, but from the analysis "Highland green ramps" may be a monophyletic group.
@rynxs You want me to revert https://www.inaturalist.org/taxon_swaps/122627? @zihaowang please curate in the direction of POWO for vascular plants
Yes, please. As mentioned prior, this is another group that's going to get messy in the future, and for this specific instance I believe having these entities under the same species for the time being will be substantially more beneficial in future actions (mostly splits). The inability to separate these entities without difficulty will probably lead to the addition of an A. tricoccum complex in the future, as mentioned in the paper.
Good call, yeah. Onions are only gonna get messier from here on out. I'll reach out to POWO in a bit on the Allium canadense complex and see what they say before doing anything about that.
Perhaps I'm mistaken but when changing var. burdickii from a variety of A. tricoccum to a its own Allium species, shouldn't the other variety of A. tricoccum, A. tricoccum var. tricoccum cease to exist as a variety and drop its infraspecific name? I ask this because A. tricoccum var. tricoccum remains after this taxon swap.* Since l currently don't agree with this taxon swap, I'm not suggesting the same is done to var. tricoccum. I'm simply pointing out what seems to me to be an inconsistency and redundant.
*https://www.inaturalist.org/taxa/633217-Allium-tricoccum-tricoccum
@pdabell yes, that is typical procedure. I am quite glad the nominate var was not swapped.
@pdabell @loarie If POWO continues to follow the inadequate treatment of Allium in FNA, POWO is NOT the authority and should not be used to guide the naming of Allium species in North America. And there is nothing wrong in citing a master's thesis, which was detailed study of the taxa in the Allium tricoccum complex. I agree more work is needed to separate the four potential species in the complex. But that also means A. burdickii cannot simply be reverted to A. tricoccum var. burdickii. Like @rynxs said, the addition of an A. tricoccum complex should be considered.
Unfortunately we are supposed to continue to use POWO as the basic authority for iNat's plant taxonomy, but I personally would deviate from POWO's treatment for tricoccum, cernuum and canadense, and their fellow complex members.
@zihaowang , POWO actually follows 3 botanical authorities for its accepted name. Interestingly, FNA sites Almut. G. Jones (1979) seminal study of Allium burdickii as its sole "Select reference" but Jones splits it from A. tricoccum and doesn't lump it in with it as as a variety. It is 1st those botanical authorities that need to make some corrections before we make the change here.
Have you noticed that the Nature Serve Explorer source that you based your taxon swap on claims that Allium burdickii is "critically imperiled (S1)" in Nova Scotia? Jones (1979), FNA and POWO don't mention NS in its range. It also describes the species habitat as "Rich upland woods." Really? Not according to FNA, which is one of its references. See https://explorer.natureserve.org/Taxon/ELEMENT_GLOBAL.2.155701/Allium_burdickii .
There's a lot of inconsistent information for this species in apparent botanical authorities. Too numerous to mention here.
@zihaowang , I neglected to add the FNA reference and description of A. burdickii's habitat as " Dry soil in upland woods," (see http://www.efloras.org/florataxon.aspx?flora_id=1&taxon_id=242102170), which is largely the same--in this case--as Jones (1979) description for the habitat as ""rich upland woods tending toward somewhat drier ground, as compared to A. tricoccum." (See Jones, Almut G. “A Study of Wild Leek, and the Recognition of Allium Burdickii (Liliaceae).” Systematic Botany, vol. 4, no. 1, 1979, pp. 29–43. JSTOR,
https://doi.org/10.2307/2418663.*) Nature Serve Explorer captures part of that; just not the significant "tending toward somewhat drier ground" portion. But, then, NSE doesn't reference Jones (1979). Again, lots of inconsistent information for A. burdickii. Or is that A. tricoccum var. burdickii? ;)
*Accessed 9 Mar. 2021.
@pdabell I originally referenced Bina Swasta Sitepu's paper as a source, but for some reason I thought an online database is needed as a source for a species' name, and changed it to NatureServe. That is changed back now. The inconsistency you mentioned only highlighted why taxon swaps like this should not be based on "authoritative" databases which are merely collecting names from other references. Taxonomically difficult groups like this should be treated on iNat on a case-to-case basis.
Regarding the record in Nova Scotia, NatureServe used the data from Atlantic Canada Conservation Data Centre (http://accdc.com/en/ranks.html). Allium burdickii is listed in their database as S1, as of 12/03/1999. Whether or not that's a result of misidentification, you will need to take a look at the original specimen.
I did find this paper, which supports A. burdickii in Nova Scotia: https://link.springer.com/article/10.1007/s006060170038
@zihaowang and @rynxs , I am familiar with the paper and have discussed it with Nature Serve through the Atlantic Canada Conservation Centre. I will elaborate later on my opinion as to why I think the study is based on an erroneous understanding of A. tricoccum var. tricoccum in Nova Scotia.
I fully realize that it was bold, and risky, to claim that the above-mentioned study is based on an erroneous understanding of Allium tricoccum var. tricoccum but the verifiable evidence for A. tricoccum var. burdickii is either scant or, from what I can see, non existent. If an herbarium has occurrences of preserved specimens, they are not listed with GBIF (which includes iNaturalists RG observations, save for the most current ones) or VASCAN (Database of Vacular Plants of Canada). The Atlantic Canada Conservation Data Centre lists (only a listing and not an observable record) it is in the province but it isn't even certain if its there since its at risk status of S1 is followed by a "?."
The rare variety is mentioned in a handful of other webpages or downloadable documents available at them, but being mentioned is not evidence.
A. tricoccum var. burdickii is mentioned in Nova Scotia Plants but its description isn't, well, very descriptive. According to the field guide/manual, its only unique (to the variety) characteristic is that it is the smaller of the two varieties in the province.* That characteristic certainly isn't very diagnostic, especially on its own.
I am perhaps mistaken on this, but this issue matters to this taxon swap because until the A. tricoccum var. tricoccum that is misidentified as A. tricoccum var. burdickii (especially in regions such as Nova Scotia that are distant from var. burdickii's natural range further west in eastern North America) is correctly understood as, at minumum, not being A. tricoccum var. burdickii and just smaller A. tricoccum var. tricoccum, our understanding of true Allium burdickii (use of the synonym intended) will be muddied with a sizable number of incorrectly IDed A. tricoccum (also intended to not include the infraspecific taxon of var. tricoccum).
To give L. Vasseur's abovementioned study credit, it is not the author's fault that the author identified the Allium tricoccum of the author's research as A. tricoccum var. burdickii given the scant knowledge of its unique characteristics. And the studies findings are equally as useful when correcting the variety to var. tricoccum. Given the apparent variation in var. tricoccum across its (natural) range, it would be useful to our understanding of it to simulate the research in other disconnected populations across its range.
*See "Nova Scotia Plants; Marian C. Munro, Ruth E. Newell & Nicholas M. Hill; Published 2014" at
https://ojs.library.dal.ca/NSM/pages/view/Plants ;
https://ojs.library.dal.ca/NSM/issue/view/509 ;
https://ojs.library.dal.ca/NSM/article/view/5472 ; and then download the chapter with key for liliaceae at
https://ojs.library.dal.ca/NSM/article/view/5472/4933 ; or the whole Nova Scotia Plants as a single file at http://nsplweb.library.ns.ca/botany/Print%20Nova%20Scotia%20Plants%20complete%20manuscript.pdf (note: an unsecure http website and not an https one).
The link to the ACCDC webpage that I mentioned above but forgot to add: http://accdc.com/webranks/NSvasc.htm.
Apparently POWO has changed their taxonomy to accept A. burdickii. However, var. tricoccum still exists, so it needs to be swapped into A. tricoccum and A. tricoccum needs to be split in order to fix the absurd number of disagreeing IDs. I don't want to take the heat for a change I did not want, though. @loarie, would you mind fixing this or finding someone who can, please?
@rynxs , POWO references Bina Sitepu's master's thesis mentioned above (the original primary reference for this taxon swap) and R. Goevert's (1995) World Checklist of Seed Plants and sites FNA for the alternative taxonomy of A. tricoccum var. burdickii.* Why it doesn't base its accepted names on Almut G. Jones (1979) 'A Study of Wild Leek, and the Recognition of Allium burdickii (Liliaceae)'** is bewildering.
There's an "absurd number of disagreeing IDs"? After going through an absurd number (pretty close to all) of Allium tricoccum (when there was one recognized species with two accepted varieties) observations, and seeing similar activity by other iNat users such as @ephouston , I'm confident that the vast majority of A. (tricoccum var.) burdickii has been separated out of A. tricoccum into A. (tricoccum var.) burdickii or, at minimum, had the species-level ranking dropped and moved up a taxon to be IDed no lower than genus. And if you search for dissenting observations in the the recently-created Pre-Maverick project, you won't find many. Even if you query the whole Allium genus, which currently gives 54 results, of which 3 or possibly 4 are of A. tricoccum (var. tricoccum). Although it is implied by this, all the A. tricoccum currently IDed on iNat no lower than the species level are simply A. tricoccum that haven't been fully agreed to be A. tricoccum var. tricoccum; what POWO now recognizes as a distinct species from A. burdickii and without any varieties.
*See https://powo.science.kew.org/taxon/urn:lsid:ipni.org:names:8704-2
**In Systematic Botany, 4(1), 29–43 at https://doi.org/10.2307/2418663 at the time of my adding this comment.
I'm happy to help, but I would benefit from a very simple articulation of the issue and the steps that need to be taken.
@rynxs , BTW, Allium tricoccum var. tricoccum is no longer listed at POWO. At least at the time of my leaving this comment. See https://powo.science.kew.org/taxon/urn:lsid:ipni.org:names:8906-2 , which does not lost any infraspecific taxons.
Also, I realized after posting my above comment that the Pre-Maverick project won't capture the observations with dissenting suggested IDs since burdickii and tricoccum are in the same genus. However, add "burdickii" to your search under "Allium tricoccum" and there won't be many results
@ephouston 's project, Potential Allium burdickii should help with this.
@pdabell you can use iNat URLs to see observations identified as A. burdickii. You will not find the now genus-level observations of contested A. tricoccum vs. A. burdickii observations by searching for any one species, outside of using the ident_taxon_id URL parameter.
https://www.inaturalist.org/observations?place_id=any&subview=map&ident_taxon_id=158169
There are 1,004 observation identified as A. burdickii, and only 434 observations currently sitting at a community taxon as A. burdickii. Doing a search limiting the lowest possible taxon rank to genus, 561 observations are above the rank of genus, implying that more observations are currently stuck at genus due to the swap to species than you have successfully gotten down to species. I would definitely call over half of the existing observations being stuck at genus "absurd."
I'm not sure what you mean by stating var. tricoccum is no longer listed by POWO. I am aware of this; I think this is a misunderstanding. As I stated prior, it needs to be swapped into the species.
@loarie please create an A. tricoccum complex with A. tricoccum and A. burdickii and source it to this paper: https://etd.ohiolink.edu/apexprod/rws_etd/send_file/send?accession=ohiou1534064390052709&disposition=inline. Then, A. tricoccum needs to be split. I'm not sure what @pdabell would prefer to base the atlases on, but I suppose we can just base it off of states with existing observations of A. burdickii if POWO and other sources do not have accurate range information. Then, var. tricoccum needs to be swapped into the species. Thank you in advance.
@loarie , since Allium burdickii and A. tricoccum are now the accepted species names of POWA, the now-formerly accepted A. tricoccum varieties of A. tricoccum var. burdickii and A. tricoccum var. tricoccum are redundant. This is addressed on iNat with the taxon swap for var. burdickii but it is not (yet) addressed for var. tricoccum, which should be merged with its parent species, A. tricoccum. Although I don't think it will cause a mess due to the issue @rynxs points out in his last comment (see my rationale in my 2 last comments), not everyone would agree with me, so I am hesitant to suggest the merger.
That being said, one idea to possibly consider is to 1st swap the A. tricoccum that is not IDed to its variety (in this case, A. tricoccum var. tricoccum) into a new taxon, Complex Allium tricoccum and then 2ndly, merge A. tricoccum var. tricoccum into a the A. tricoccum taxon. The complex would leave observations that are RG with that intact and include both the 'old' A. tricoccum and a small number of observations in which their suggested IDs are disputed as possibly being A. burdickii. And the A. tricoccum taxon consisting of the A. tricoccum that was IDed as A. tricoccum var. tricoccum would include only the A. tricoccum in which there is agreement that the observations are A. tricoccum. I'd prefer there to not be a complex but this approach might be a good temporary fix until the observations in a Complex A. tricoccum are sorted into A. burdickii, A. tricoccum or, when there is still uncertainty, moved up a taxon to being IDed no lower than their genus.
Although not "a simple articulation of the issue and the steps that need to be taken," I don't think the issue is simple and, as a result, the next steps are not yet clear.
@rynxs , after posting my last comment, I saw yours. I can't respond to it in depth right now but I am in general agreement, as I think is clear since some of what I suggested you also did.
@pdabell what you are describing will accomplish exactly what I described prior, but will take more work and time, and be more destructive. Your path will bring IDs of A. tricoccum to complex for all IDs everywhere, but splitting will only do so for observations in the shared range of A. tricoccum and A. burdickii, as determined by atlases.
@pdabell sounds good. I'm going to stop responding temporarily so that we're not writing out responses at the same time.
@rynxs , I was intending for the known A. tricoccum (ones currently IDed as A. tricoccum var. tricoccum) to not be included in Complex Allium tricoccum and only include ones that have agreement to the level of species but no lower. But if that is not possible, I have misunderstood what a complex is.
Can the A. tricoccum that have currently only been IDed to the species level and no lower be marked in someway (i.e. in an observation field or tag) that separates them from the known A. tricoccum while being in the same taxon (A. tricoccum)? If so, and if they were marked before a merger, I suspect that would achieve the results I am intending. But, perhaps I'm mistaken again.
I fully realize that there are numerous known A. tricoccum currently only IDed to the level of their species and not their previously-accepted variety. I'm just trying to suggest a way to make it simpler to eventually separate out the ones that are A. burdickii and still in the A. tricoccum taxon and to separate out the observations that there isn't certainty for IDing as either species (IDing them then to the level of their common genus and no lower).
The short explanation: this will accomplish exactly what you describe you want to happen. What you describe as "known" A. tricoccum will become A. triccocum, and identifications currently at species-level will be caught by the complex and become complex-level, effectively marking them in the way you have asked for.
The very long explanation:
I think some things have to be defined first. I understand how difficult it is to understand how taxon changes work without access to them.
A "swap" is a curatorial action which replaces the IDs of one taxon with another, inactivating the singular input taxon.
A "merge" is a curatorial action which combines IDs or two more input taxa into one output, inactivating the input taxa (not relevant here).
A "split" is a curatorial action which divides IDs of the input taxon through a geographic filter. Relative locations of output taxa determine what happens to IDs on each observation. This has one of four possible effects on identifications on observations, described in relevant terms to this situation below. It does not affect subordinate taxa.
You have misunderstood what a complex is on iNat. Complexes are monophyletic groups that include all the species described as members, so it is not possible to create the complex and exclude A. tricoccum var. tricoccum (A. tricoccum sensu stricto), nor would the effects of omitting it be preferable to including it. Use of the complex allows for searching of both A. burdickii and A. tricoccum sensu stricto simultaneously, as one group, similar to how our old taxonomy allowed for searching for var. burdickii and var. tricoccum simultaneously by searching for A. tricoccum.
I will flush out what will happen when the path I described is committed.
To be clear, splits only affect identifications of the taxon being split. Allium tricoccum is a separate taxon from var. tricoccum. Identifications of variety tricoccum will be unaffected by the action of splitting the species A. triccocum. Following the split of A. tricoccum sensu lato, var. tricoccum will be swapped into A. tricoccum, so current IDs of var. tricoccum will function as species-level IDs of what you call "known" A. tricoccum. There is no need to mark any observations, as those affected by the split will have their identifications simply be elevated to and caught by the A. tricoccum complex, effectively treating IDs of the old A. tricoccum as "either A. burdickii or A. tricoccum."
1 - The creation of a complex will group together observations of A. tricoccum (and by extension var. tricoccum) and A. burdickii.
2 - A. tricoccum will be split, causing one of four possible effects on observations: (1) an observation of A. tricoccum exists in the shared range of A. tricoccum and A. burdickii, so IDs of the species A. tricoccum will be elevated to complex. (2) an observation of A. tricoccum exists in the exclusive range of A. tricoccum, so IDs of the species A. tricoccum will be unaffected. (3) an observation of A. tricoccum exists in the exclusive range of A. burdickii, so IDs of the species A. tricoccum will be swapped to A. burdickii. (4) an observation of A. tricoccum exists outside the atlased range of A. tricoccum, so IDs of the species A. tricoccum will be elevated to complex.
3 - A. tricoccum var. tricoccum will be swapped into the now-split A. tricoccum, preserving A. tricoccum sensu stricto IDs.
So, following this process, iNat will achieve exactly what you are asking for, @pdabell. Identifications of A. tricoccum sensu stricto (previously var. tricoccum) and A. burdickii will take precedence over previous IDs of A. tricoccum sensu lato, all while keeping observations of both taxa organized together into a complex.
Here is a list of every potential outcome for identifications (prior -> post):
An ID of Allium tricoccum var. tricoccum -> An ID of Allium tricoccum
An ID of Allium tricoccum in the shared range of A. burdickii and A. tricoccum -> An ID of Allium tricoccum complex
An ID of Allium tricoccum in the exclusive range of A. tricoccum -> An ID of Allium tricoccum
An ID of Allium tricoccum in the exclusive range of A. burdickii -> An ID of Allium burdickii
An ID of Allium tricoccum where neither A. burdickii nor A. tricoccum are known to occur -> An ID of Allium tricoccum complex
@rynxs , thank you for explaining the process to me. Very helpful! I think your suggested approach to the taxon changes captures what I think is needed for the taxon changes. Although I suspect you already have accounted for this, I suggest that swap of A. tricoccum var. tricoccum be done following the splits to its parent taxon or, I assume, the observations that are currently agreed to being var. tricoccum will be swapped into the A. tricoccum complex.
What are the atlases that are used to determine the ranges? If they are the distributions shown in POWO, which is the source for the taxon swap from A. tricoccum var. burdickii to A. burdickii and which I presume will be the source for these taxon changes, the ranges aren't actual but geopolitical. All of Ontario, for example, is shown to be in the range of both tricoccum and burdickii. But, if burdickii does exist--or did--in Ontario, there are only collection and disputed observation records for it in a few locations in the most southerly areas of its range (a little north of Lake Erie and Lake Ontario). If the whole province is recognized at burdickii's range, it would then be the same as the range for A. tricoccum sensu stricto, which would result in the overwhelming majority of A. tricoccum being elevated to the A. tricoccum complex. But, if the distributions shown in FNA, which are based on records (whether accurate or not), are used, this problem will not occur.
Despite my disagreement with some (most in some region's, such as Quebec) identifications on GBIF and at the SERNEC Portal, I suggest the distribution of their records should be included when determining the tricoccum and burdickii ranges.
In the FNA key for var. burdickii, it states, "With its overlapping characteristics and distribution, Allium tricoccum var. burdickii is doubtfully separate from var. tricoccum." When looking at records through the SERNEC Portal and GBIF, and observations here on iNat, the findings would largely be in agreement. But one notable exception is the few disjunct populations on iNat of A. burdickii in western Arkansas. Being outside of the known range for A. burdickii and based on your proposed treatment for "A. tricoccum where neither A. burdickii nor A. tricoccum are known to occur," that would then do the same for A. burdickii where neither of the two species occure, would these then be elevated to the complex? If so, that would pose a problem, as outliers such as these should be recognized in the lowest taxon they are known to be in (formerly as a variety and now as a species).
Before any further changes to the A. tricoccum and A. burdickii taxons are committed to, hopefully other iNat users will weigh in with their thoughts on this.
If you want a Google Map showing the known locations of SERNEC's records for A. tricoccum, A. tricoccum var. tricoccum, A. burdickii and A. tricoccum var. burdickii, please let me know.
@pdabell - @loarie just needs the most specific list of geopolitical boundaries in which both species occur. We have control over what regions are included individually, and can even go down to the county level if necessary. Please provide @loarie with the most specific list of iNat places representing the ranges of each species.
@loarie and @rynxs , compiling a list of where both species occur is simple enough since, even for obscurred observations, file downloads of iNat data include what county or similar region (e.g. regional municipality) they are in but the three sources of which I think there is general agreement that they are perhaps the most authoritative references on the range of Allium burdickii--A.G. Jones (1979) study and the distributions mapped in POWA's and FNA's keys*--don't show what county/region A. burdickii has been observed in. If I only include the counties/regions where observations have been added to iNat, some (depending on the juristiction, potentially a large number) will be omitted. I could approximate what counties/regions are within Jones' and FNA's boundaries but that wouldn't be accurate given the scale and limited features to work with (useful references) on the maps. If we only use the counties where A. burdickii has been IDed on iNat that wouldn't represent the typically larger ranges in most states/provinces recognized by Jones and FNA.
I doubt anyone wants to unnecessarily elevate all A. tricoccum sensu stricto that currently is recognized on iNat at no lower than the level of species to an A. tricoccum complex when querrying "Allium tricoccum in the shared range of A. burdickii and A. tricoccum." Please correct me if I am wrong, but without being able to accurately identify the counties and regions within the ranges mapped by Jones and in FNA, the querry would be state- or province-wide. That wouldn't be a problem in states in which boundaries are approximately equivalent to A. burdickii's range (most often within A. tricoccum's), such as Ilinois, Ohio and Indiana, but they are the exception, especially if Jones' more conservative range is sited (although FNA's range maps are, according to its "Selected reference,"** based on Jones' (1979) study, oddly, its range for A. (tricoccum var.) burdickii is slightly different).
Although making these changes is necessary now that POWO recognizes the former A. tricoccum varieties as separate Allium spp., which I support, I do wonder if it would have been simpler to 1st have made the necessary changes under their formerly accepted names.
*See:
1) Jones, A. G. (1979). A Study of Wild Leek, and the Recognition of Allium burdickii (Liliaceae). Systematic Botany, 4(1), 29–43. https://doi.org/10.2307/2418663 (accessed at https://www.jstor.org/stable/2418663 on March 9, 2021);
2) Flora of North America online at http://www.efloras.org/object_page.aspx?object_id=8444&flora_id=1 (although based on A.G. Jones (1979) above study, the range is--oddly--slightly different); and
3) Plants of the World Online at https://powo.science.kew.org/taxon/urn:lsid:ipni.org:names:8704-2#distributions.
**http://www.efloras.org/florataxon.aspx?flora_id=1&taxon_id=242101410
@pdabell individual IDs of Allium tricoccum would be elevated in the shared range, so IDs of var. tricoccum would end up as IDs of the species. Unfortunately, there really isn't a better way of doing this other than getting the most specific range information possible for both species.
@rynxs and @loarie , I created a spreadsheet showing the range of Allium burdickii as shown in A.G. Jones (1979) study mentioned above. It lists the counties (in the US) and other administrative regions (in Canada) where Jones's map shows A. burdickii to have been. Although over 40 years old, since Flora of North America's key is apparently based on Jones's study but oddly differs from it slightly to somewhat significantly, and because Plants of the World Online's distribution only shows the range to the level of state and province, I think it is reasonable, especially for these purposes, to consider it most authoritative. Whether it is most accurate can only be known as this species and the Allium tricoccum complex are further understood. Please let me know how you'd like me to send it to you and also who you might suggest to first review it for accuracy.
@loarie and @rynxs , can either of you explain the rationale for including Nova Scotia in the range for A. burdickii?* Yes, Nature Serve Explorer lists it but POWO, FNA, GBIF, SERNAC--all at the time of my adding this comment--and Jones's study mentioned above all do not. Although I am doubtful of its ever having been in Ontario and Quebec, and even the New England States but at least there are preserved specimens that the rationale is based on, whether or not the IDs are correct. In the case of Nova Scotia, until preserved specimens that are IDed as A. burdickii--correctly or not--are made available, claims that it is there or historically were are unfounded and should not result in iNat, or other relevant authorities, including it in the species' range. iNat users might also agree to suggestions of the observations of A. tricoccum in NS are instead of A. burdickii but I'm close to certain that not even one observation there has been suggested to be A. burdickii. Yes, L. Lasseur's (2001) study, 'Allozymic diversity of Allium tricoccum (Ait.) Solander var. burdickii Hanes in isolated populations of Nova Scotia (Canada)' claim it is there but if any of the material referenced in the study claim that A. burdickii is there, it would be imperative to know what 1st hand evidence those claims are based on.
@pdabell if you're talking about the current display for iNat's map, I'm not really sure what's caused it but I imagine it might have been inherited from a range map for A. tricoccum as A. tricoccum var. burdickii before it was swapped, maybe? There's no atlas for it right now, so any display range is not something that was created for it.
I checked out the spreadsheet and it looks good. As mentioned in my PM, we should be able to cross reference those counties with existing observations once the atlas is done. Thank you for doing this, @loarie
I'm not sure what I'm supposed to do - I'm not familiar with this taxon.
Is this something you can do since you're more familiar with this group. If you have a specific question about how to curate taxon on iNat as relates to https://help.inaturalist.org/support/solutions/folders/151000147064 happy to help explain how to do what you want to do.
My preference would be for you to do what you're trying to do. but If you want me to do it, I'm happy to help, but I'll need to know what you're trying to do via very simple articulation of the issue and the steps that need to be taken (e.g. no more than 5 lines of text). I get easily overwhelmed and confused by such long threads
@loarie please create an A. tricoccum complex with A. tricoccum/A. burdickii and source it to this paper: https://etd.ohiolink.edu/apexprod/rws_etd/send_file/send?accession=ohiou1534064390052709&disposition=inline.
Then, A. tricoccum needs to be split based on the county map @pdabell has put together. Then, var. tricoccum needs to be swapped into the species. Thanks in advance
Am I understanding that these are the steps you want to take?
1) create complex Allium tricoccum descending from Subgenus Anguinum
2) move species Allium tricoccum from Subgenus Anguinum to complex Allium tricoccum
3) move species Allium burdickii from Genus Allium to complex Allium tricoccum
4) create draft split with species Allium tricoccum as inputs and species Allium burdickii, species Allium tricoccum as outputs
5) create atlases for Allium tricoccum and Allium burdickii
6) commit draft split
7) swap Allium tricoccum var. tricoccum into species Allium tricoccum
If so, would @rynxs mind doing steps 1-5, then report back here and I'll take a look before step 6? Re: step 5, I don't think these need to be at the county level, atlases don't do well with more than 50 or so atlas places, but I if someone wants to spend making county level atlases they should do so directly as I'd prefer not to middleman that.
If you need me to do steps 1-5, comment here and I'll do it and report back here before step 6
Finished steps 1-3, but I do not want to do 4 or 6 because I don't want my name on the split. I am OK with doing 7 once the split is done, though, and I can work on 5. I think state boundaries should do fine for most states, the county data will be most useful for the edges and holes in A. burdickii's range, although I'll investigate it further as I work on the atlases.
OK I did Step 4 here https://www.inaturalist.org/taxon_changes/124543
please report back here once 5 is done so we can review before 6
Atlas is done for A. burdickii, but there are three observations outside of it: one in Missouri, one in Arkansas, and one in Cumberland County, Kentucky.
@pdabell do you have a preferred range delineation for A. tricoccum?
Also preemptively changed the common name of A. tricoccum to "wide leek" and added the common name "small white leeks" to complex Allium tricoccum.
In reponse to @rynxs's question and a few issues I initially have:
1) the range delineation for Allium tricoccum is covered by a number of botanical authorities but since it and A. burdickii have often been treated as varieties of one species, it's difficult to know if the mapped A. tricoccum ranges are for both varieties or just A. tricoccum var. tricoccum. To be conservative, I suggest delineating the 'new' A. tricoccum's range based on the states and provinces that are within the range mapped by A. G. Jones in the author's abovementioned paper. iNat does not currently include New Brunswick in the range for A. tricoccum var. tricoccum, which will be swapped (merged?) into A. tricoccum, but I think that was an oversight when mapping the state and province based locations.
A posible exception to adding the whole state within the range would be in SW Illinois, in W Iowa and SE South Dakota, where about 6 populations of A. burdickii are shown in Jones's study to have occur (or to have historically occured) outside of the range of A. tricoccum sensu stricto at the time. Manitoba is included in the known range on iNat for the A. tricoccum that will become Complex Allium tricoccum but I am not familiar with any occurrences listed on GBIF, SERNAC, VASCAN--not even ones listed without images.
Similar to what I wrote in my initial comment with this flag, I'd prefer that Complex Allium tricoccum not be based on a master's thesis but instead on an accepted taxon of POWO or, at minimum, a peer reviewed study such as Jones's, even if, in that study's case, it's over four decades old. But, it won't be significant issue so long as the new A. tricoccum is sourced to POWO.
2) Looking at the proposed taxonomic split 124543, if I understand it correctly, I largely support it. Though, a few items I'd 1st consider are: i) the split shouldn't be finalized until the atlas is complete, which I think is intended to first be done; ii) if A. burdickii's range includes whole states for ones in which the edge of the species' range is within, and if I understand the swap correctly, a large number of A. tricoccum that are not IDed as A. burdickii and not currently IDed as A. tricoccum var. tricoccum (to become A. tricoccum sensu stricto) that are outside of Jones's mapped range will be IDed no lower than at the level of the complex. For example, Allium tricoccum in N Minnesota that are currently IDed no lower than the level of species (not as A. tricoccum var. tricoccum) will be IDed as Complex Allium tricoccum. There certainly might still be some observations in that region of A. burdickii that are only IDed as A. tricoccum and weren't agreed to be A. tricoccum var. burdickii before the taxon swap of this flag, but they are few; and iii) similarly, how will A. tricoccum in provinces such as Ontario be treated? There are zero RG observations of A. burdickii in the province, but, all the A. tricoccum, which has a enormously larger claimed range than that of A. burdickii in Ontario, would be IDed to no lower than their common complex with A. burdickii.
3) Although I don't think that this is something that can be changed when creating Complex Allium tricoccum without reference to at least one recognized (by iNat) botanical authority, I don't understand why tricoccum is in the complex's name, unless burdickii is included, as well, since A. tricoccum and A. burdickii are equally part of the complex. Referring to it as an A. tricoccum complex might result in a number--potentially a significant number--of observations of A. tricoccum being IDed at no lower than the level of its complex. I suspect this has been occurring to a lot of A. tricoccum sensu stricto that hasn't been getting IDed as A. tricoccum var. tricoccum (soon unnecessary, of course) since I suspect many users didn't understand that A. burdickii wasn't accepted as a separate Allium sp.
Thank you @loarie and @rynxs for your work on this.
Please note that, although I would like to participate in this discussion and process more, I won't be able to until next week, so if I don't reply over the next few days, it's not because I have bowed out.
@pdabell
1) The complex isn't sourced in the same way as POWO, we basically just need evidence that the group is accepted and used in literature, which the master's thesis provides.
2) The split occurs across the geographic ranges of both species, for better or for worse. At least the complex will "catch" these observations and make it easier to find them.
3) The name of a complex on iNat is determined by which member species was described first. If an observation cannot be identified past complex, then it should sit at complex. I don't really see an issue with a user IDing to complex if they aren't sure of species. There can't have been many IDs of the complex since I just created it the other day, if that's what you're saying.
I am also currently traveling, so I suggest we postpone this for a few days.
Allium tricoccum 55634 needs an atlas
@rynxs and @loarie , returning to this flag, I'm curious if anything more has happened regarding the changes mentioned above (steps 1-7) and what you might require of me, such as possibly carrying out step 6. More details are needed, though, before my possibly committing to that.
I was looking at the atlas for Allium burdickii. The default view still shows Manitoba and Nova Scotia as yellow-shaded provinces, and the provinces aren't just without any observations in need of ID, as mentioned above, there aren't any occurrences in them listed with GBIF, VASCAN (Canadensys) or at the SERNEC Portal, and they ar not included Jones's (1979) range (referenced above), nor even FNA's similar but slightly different one that is apparently based on the one in Jones's study. That being said, I noticed that when zooming in a couple of times, the range appears to be based on the counties and similar regions (in Ontario and Quebec) that with Jones's range, which then excludes Manitoba and Nova Scotia. Some counties are also yellow-shaded but have RG level observations in them. Should those be shaded green or is that purposeful?
Regarding the common name additions of small white leeks of wide leek to Complex Allium tricoccum and Allium tricoccum, where are those common names, well, common? Not that it is a botanical authority but the Wikipedia article that iNaturalist references doesn't even mention those names. At least for the complex, wouldn't ramp, ramps or wild leek be more common and befitting since the complex includes Allium burdickii, which has a common name that is quite the opposite of wide? Wild leek would include wide leek and narrowleaf wild leek, or wide leek as the common name of A. tricoccum would also be wild leek since the complex name, as @rynxs pointed out above, is based on "the member species [that] was described first"? Also, since A. tricoccum with red pigment to some degree is more common that "albino," (Michigan Flora's term for pigment free A. tricoccum) pigment free, ones, "white" shouldn't be in the common name. In any event, I'm pretty sure that small white leek isn't a commonly used common name.
@pdabell as stated prior, I need a range delineation for A. tricoccum sensu stricto. Is POWO's range sufficient?
I don't know why those provinces show, as I stated prior I think it's inherited from Allium tricoccum. Those provinces are not atlased, so they will not affect the split. I do not know why a RG observation would not cause the county to become green on the atlas, but if you have any examples I would appreciate it. From a quick look I was unable to find any, most appear to either not be RG or be obscured and showing up in another county.
I don't know what the Canadian common name for Allium tricoccum was set to, but before I changed the common name for A. tricoccum to "wide leek" (preemptively for the narrowed sense following the split) it was "small white leek," so that's what I named the complex. You are free to add any sourced names you would like for Canada. The common names for complexes that I prefer are specific to the group, so my ideal common name for it would be "American wild leeks," if we can find a source substantiating that or something similar. I'm not sure what your point is with "wide leek," since the name is only applied to A. tricoccum, which has wide leaves.
@rynxs , sorry, I thought I suggested POWO's distribution by state and province be used for the iNat atlas for Allium tricoccum sensu stricto, but I see that I was unclear. I don't always agree with POWO's species distributions (for example, POWO doesn't include Ontario in its range--either indigenous or introduced--for Acer negundo, which is without doubt in the province. (See https://powo.science.kew.org/taxon/urn:lsid:ipni.org:names:781412-1#distributions , which illustrates this at the time of my posting this comment.)
Regarding my comment about there being some counties that are shaded yellow but have research grade observations, I was incorrect. For example, I looked closer at the 4 RG observations of A. burdickii that appear to be in Grundy County, Illinois, and they are not observations there but ones in neighboring counties with obscured coordinates, which leads to them initially appearing to be in Grundy County.
And my point about "wide leek" not being a fitting common name relates to the fact that there are often clumps and larger populations of A. tricoccum that have have narrow leaves (all or most of the leaf blades less than 4 cm wide at their widest). As a complex includes 2 or more species, I presume that, unlike for the common name for a species, the name can, and perhaps should be, written in the plural form. Complex Allium tricoccum would then be referred to as "wild leeks" or "ramps," which are common names in both Canada and the USA. As A. ursinum, at least according to iNaturalist, is not commonly referred to as "wild leek," "American" would be an unnecessary addition to the name for Complex Allium tricoccum. Of course, iNat might not have the common name of "ransoms" correct and if in fact "wild leek(s)" or even "ramp(s)" are more common names for A. ursinum, it wouldn't matter if one of them was applied to the complex since the same common name is often applicable to a number of species.
Worth noting that POWO's range for A. tricoccum does not include Arkansas, Oklahoma, Manitoba, or Nebraska, but those places appear to have been added to lists for these places at some point (or some other way of having them appear on the atlas before one was created). I've left them in for now.
@loarie all set to go ahead with the split.
@rynxs , the last I checked, the species of the disjunct populations in Arkansas in Complex Allium tricoccum are A. burdickii, and the counties where they were observed are included in the atlas for it. The observation of A. tricoccum in Nebraska is a similar outlier but since it hasn't been identified--yet--to the level of it formerly accepted variety, in the taxon swap it will be elevated to the complex. And in addition to Oklahoma and Manitoba not being included in the range for A. tricoccum by any of the aforementioned botanical authorities, unless I'm mistaken, there aren't even any observations on iNat in either of them. I might be mistaken about Oklahoma but I am certain about Manitoba. So they shouldn't be included in the atlas. If a different shade can be added for jurisdictions with unsubstantiated claims of there being A. tricoccum (it's mentioned in field guides as being in Manitoba), then perhaps that is a way to add it to the atlas.
OK, I removed Arkansas, Manitoba, Nebraska, and Oklahoma.
@rynxs , thanks for making that change.
I don't know if it is because the taxon changes aren't finished yet and if the same has occurred outside of A. burdickii's range used for the atlas but in Ontario, three RG observations with IDs of A. tricoccum sensu stricto that are outside of A. burdickii's range for the atlas have been moved to Complex Allium tricoccum. Those observations are in Huron County, the Regional Municipality of Waterloo and (the City of) Toronto (just "Toronto" on iNat) and I understood that observations of A. tricoccum outside of the atlases range for A. burdickii would remain as A. tricoccum. That being said, Huron County is adjacent to Lambton County, which is a boundary county in the atlas. And the line in A. G. Jones (1979) study could possibly be argued to cross through the extreme south portion of Huron County.
Can you link those observations? Everything I'm finding at complex level is because someone IDed it as A. burdickii as well, unless you're asking if A. burdickii IDs will be affected by the split?
If you're asking if those A. burdickii IDs will be affected, then no, they will not. Only IDs of the species A. tricoccum will be affected by the split. Other identifiers will have to come in and sway the community taxon.
@rynxs , the 3 observations I mentioned in my last comment can be linked to here:
https://inaturalist.ca/observations?lat=44.625182133746144&lng=-79.81603128817024&locale=en&lrank=complex&place_id=any&preferred_place_id=57637&radius=188.91461994317254&subview=map&taxon_id=1457593 . Check them out and I'll leave it for you to decide whether they should be treated as A. tricoccum sensu stricto or more generally in Complex Allium tricoccum following the split. On seeing what taxon each changes to, I'll understand the process better.
I'm not nearly as familiar with the group as you are so I can't make an ID either way, but observations currently sitting at complex level with both A. burdickii and A. tricoccum var. tricoccum IDs will end up at complex level with A. burdickii and A. tricoccum IDs post-split.
@rynxs , if I understand the process correctly, using Ontario as an example, the 6 observations that are IDed strictly in Complex Allium tricoccum will, regardless of where they are in the province, remain with that ID. Observations of Allium tricoccim sensu stricto within A. burdickii's atlased range will be elevated to the complex, and observations of the same outside of that range will remain with the A. tricoccum sensu stricto ID (or stated differently, be swapped into the new, of sorts, A. tricoccum without varieties); and all observations with IDs of A. tricoccim var. tricoccum, regardless of where they are in Ontario will be swapped into the 'new' A. tricoccum..
IDs of A. tricoccum within the atlased range will become an ID of the complex. An ID of var. tricoccum, regardless of place, will become an ID of A. tricoccum.
Relevant to this flag and discussion, see the discussion regarding the proposed split of Allium tricoccim sensu lato into A. tricoccum sensu stricto and A. burdickii at https://www.inaturalist.org/taxon_splits/124543#activity_comment_c2c9fcbc-92c9-470f-9935-03bcade07bba .
You all have worked very hard on this, however it's resolved, and I thank you for caring. I support the "complex" idea as we wait for resolution. Systems thinking makes me wonder if some of this would have been more efficient to do with some phone calls or meetings. I see no reaching out to NatureServe for example. No reaching out to Alan Weakley. No reaching out to Sean Blaney re Atlantic Canada. Questions could have been answered more directly I suspect. Somehow in this digital age we forget about the value of analog dialog.
Finally, Canadian common names I believe shoud be bi-lingual, and Gray's 8th manual is full of Canadian common names. I see "ail des bois" offered and it is clearly used. It seems to mean "garlic of the woods." However this is all resolved, no one can say it wasn't thoroughly discussed.!
Hi Folks,
Just to close the loop here I will say that I have just revised the Nova Scotia provincial status rank from S1? to SNA (reported but unconfirmed). This change will eventually make its way to the NatureServe Explorer maps.
Cheers,
Sean Blaney
@seanblaney , this is good to hear, and ultimately a win for the species.
@seanblaney , wow, is Nature Serve ever slow with making what would seem to me to be a simple change. Allium burdickii is still listed as S1? in Nova Scotia but it's at least good to see that they updated the species' status in Vermont from SH to being not listed and blank on NatureServe Explorer (see https://explorer.natureserve.org/Taxon/ELEMENT_GLOBAL.2.155701/Allium_burdickii ; conservation statuses are accurate at the time of my posting this comment).
@seanblaney and @rynxs , further to the final few commemts above, earlier in the month, NatureServe updated the provincial status rank for Nova Scotia on its NS Explorer. If you click on the link in the immediately-preceding comment, you will see that the range map no longer includes the province (s1? to SNA). Thank you, @seanblaney , for updating the SRank with NS. Other changes to the SRanks are also shown for Ontario (S1? to SU), Quebec (S2 to SU) and Vermont (SH to unlisted (possibly SNA ?)).
The taxon swap from Allium tricoccum var. burdickii to A. burdickii is based on a master's thesis and not the accepted name listed by POWO or IPNI. For the POWO and IPNI entries, see https://powo.science.kew.org/taxon/urn:lsid:ipni.org:names:8704-2 and https://www.ipni.org/n/8704-. Although I agree that it is a distinct Allium sp. and think that master's thesis has much to contribute to our understanding of the "Allium tricoccum complex," it is not the taxonomic naming authority for the species and its currently-accepted varieties.