It isn’t iNat’s place to make judgements about taxonomy, only to follow what the experts say. I have no idea why you think this will kill iNat, when the recent years with many taxonomic changes have been the site’s strongest growth periods.
6 Likes
@eric-schmitty - Haha, thanks, it was fun to write! Under the POWO rubric - which iNaturalist has committed to using - all research-grade T. officinale and T. erythospermum in the United States are inaccurate. So if we’re going to follow what we’ve committed to follow, then all of them - even the seed ones - need to be changed.
There’s clearly some reluctance to follow through - that’s partly why 80,000 observations have sat mostly without review. Aside from the taxonomic uncertainty, really there are some huge, fascinating philosophical questions behind these dandelion issues. For example, they: (1) touch upon traditional “splitter vs. lumper” taxonomic disputes, and (2) theoretically have a big impact on the iNaturalist ethos of ID’ing things to species (you can’t do that if you have a billion poorly described micro-species).
My manifesto, however, more has to do with our collective inactivity resulting from these issues. Let’s ID some dandelions! We’re all capable of ID’ing things - that’s what we do on iNaturalist! Dandelions shouldn’t be an exception!
Right now, I recommend we resolve the low-hanging fruit - generally pushing observations to research grade, where the original observer ID’d at genus level and there are no seeds present. There are 45,000 Needs ID at genus level, so plenty to do!
Then I propose tackling more of the Needs ID observations at species level without seeds, then Needs ID with seeds. Seeds certainly aren’t the end-all-be-all of POWO dandelion taxonomy, and of course, it’s not necessary to strictly follow this order. But it’s somewhat harder, for example, to suggest in an already research-grade observation that the obviously red-seeded specimen is not, despite what the North American flora guides say, Red-seeded Dandelion. Non-seeded specimens are easier to resolve, because even under North American flora guides, lots of them say you need seeds.
Theoretically, at some point we / iNaturalist may do a taxon swap to move everything where it belongs. In the meantime, we shouldn’t just ignore dandelions! For the most part, especially on most of these 80,000 - we already know what should be done!
5 Likes
For those interesting in identifying dandelions in Britain and Ireland, there is some good material to get started here: https://bsbi.org/identification/taraxacum
6 Likes
I, for one, am fully in favor of removing T. officinale from the iNat taxonomy, since it isn’t in POWO.
2 Likes
Theoretically photos showing good views of the leaves (including the base of the leaves) and underside of the flower may be identifiable with better knowledge of what sections and species are present and how to identify them, so it might be worth leaving those as Needs ID?
But many observations show only the top of the flower head and obscured leaves, and I doubt those will be identifiable unless the leaves are very distinctive.
2 Likes
of course it’s our place to choose what taxonomy to accept, plenty of experts aren’t splitters, we just need to not only accept the splitters’ taxonomy.
It’ll kill iNat because there’s no point in tracking and learning the ID of organisms when we can’t actually identify them.
Basically, POWO is no good. We should freeze taxonomy at ~2015 levels and do one big change in 2030 if we decide it’s warranted. Probably half the species that have been changed now will be changed back by then anyway.
i literally stopped adding them most of the time, even when i am on an ‘observe all organisms i see’ visit. What’s the point? I guess genus level IDs of them can still be valuable, kinda.
1 Like
Here are some examples of leaves that are distinctive to section:
- section Hamata: simple hook-shaped lobes, in coastal area https://www.inaturalist.org/observations/68185576
- section Naevosa: leaves with dark pigmentation, in coastal area, https://www.inaturalist.org/observations/194284144
- section Palustria: thin simpler leaves, bracts pressed against flower, https://www.inaturalist.org/observations/167906561
- section Taraxacum: large complexly-lobed leaves with 3-dimensional folds, bracts usually strongly recurved https://www.inaturalist.org/observations/113093298 (note that the small flat winter leaves are simpler and aren’t useful; you need to wait for their full spring leaves)
2 Likes
I see no point in ignoring scientific progress.
If you have a suggestion for another global checklist of plant species to follow, I’d love to hear about it. Although it would be far too late to implement.
I don’t see why species level ID is so important.
4 Likes
As someone big into fungus ID (particularly Amanita), I support the RG genus approach!
10 Likes
That makes it quite easy for me then. Genus it is. I will get to that soon.
1 Like
I read what literature I could find on this issue when it came up last year. I came to the conclusion that the most frustrating thing is that there is something really interesting going on with the reproductive strategy of plants in genus taraxacum, and that the whole ‘microspecies’ debate actively obscures that issue. There is an actual opportunity for scientific discovery that is not just an unresolvable splitters/lumpers issue.
It seems like ‘taraxacum officinale’ is a plant species that has adopted a reproductive strategy of reproducing almost exclusively asexually except for ultra-rare instances of sexual reproduction that enable it to still evolve. The combination seems to let it spread really fast while also adapting to new environments; papers show that different ‘microspecies’ can be dominant in different micro-climates in the same field. Evidence for the ‘one species with an unusual reproductive strategy’ interpretation is that none of the microspecies in ‘taraxacum officinale’ are definitively old; they have all emerged since the last ice age. The expansion of untold hundred of ‘microspecies’ to every country in North America, and every state or province in Mexico, the US, and Canada in less than 400 years speaks to the species ability to both expand and adapt quickly. It seems doubtful that one or a few genetically identical individuals of many species would have enough genetic diversity and robustness to both survive and thrive from Panama to the Northwest Territories.
My feeling is that the ‘microspecies’ taxonomic concept is such a distraction that it diverts effort from studying what is going on. Time spent developing keys for the taxonomic classification of genetically identical individuals and arguing philosophy might be better spent study the ecology of how ‘taraxacum officinale’ has developed a reproductive strategy that allows it to function so effectively.
19 Likes
your error is assuming the splitting is part of scientific progress. Surprisingly, there’s no agreed upon or consistent definition of what a ‘species’ is, and declaring that all cryptic variations should be considered ‘species’ has got nothing to do with science. At best it’s got to do with passionate people in silos, at worst it’s people who want their names on a paper or are trying to manipulate conservation laws. Splitting to species level obstructs science, not the other way around. Study of evolutionary history is important but can be accomplished without constantly changing taxonomy.
i offered a suggestion that i think works well. I think you just don’t like it?
Have you ever done applied field ecology or land management? Without knowing that it’s difficult to know how to answer a question like this. Identifying species is the keystone to most monitoring, management, and conservation efforts. Yes some taxa are more often identified to a higher taxonomic level but with vascular plants species level ID is the standard. By pushing the splittin you are basically saying taxonomic revisionism is so important it warrants literally invalidating all data from before 2024. Because that’s the end result here. If you go to variety/subspecies, older data stays valid. And of course, under the current system you won’t actually be able to collect new data most of the time, as is evidenced by the most common introduced taxa in most of the world now impossible to track.
It sounds silly but it is kind of true: this dandelion situation is what splitters want for ALL taxa.
3 Likes
Yeah! This type of in detail evolutionary history is really valuable to study and this taxa is ideal for it since it’s ubiquitous, has been spread all over, and is evolving rapidly. Unfortunately that’s impossible when you elevate microspecies to a species level in the Linnaean taxonomy. Now instead of tens of thousands of potential data points on iNat you have nothing at all!
Microspecies should be established as a level of the Linnean taxonomy below species, if nothing else.
5 Likes
The hyperbole makes one’s opinion very clear, but in my opinion distracts from the crediblity of one’s arguements. POWO does tremendous good in organizing the huge mass of scientific names, synonymizing large numbers of them, clearing out vast thickets of names that have been published since 1753 and are of no use now. (Check out the list of synonyms for the older accepted names!) POWO isn’t perfect. One may disagree with its policies about accepting newly published names – too quick to adopt multiple splits? too slow to adopt splits that obviously have value? too slow/fast to adopt newly described species that can’t really be called splits from anything? Nonetheless, POWO is the best we have and in my opinion it’s pretty good. And see, I’m distracted from dealing with the basic issue (are the more finely divided species concepts useful or not?) to defend POWO.
The author of the grass treatment for the Jepson Manual (the major flora for vascular plants of California) takes a similar approach for some controversial grass genera, though with an earlier cut-off date. Not an approach I (and most grass taxonomists) agree with. It forces California botanists to hold more names in mind, the Jepson Manual names and the ones currently used by others. However, his approach to the genus Festuca is probably an improvement.
I agree with Charlie. Species level names (smallest-unit names, whatever we call them) are important. They’re important for management, for research, for talking about diversity, for even thinking about diversity!
I also agree that the microspecies debate distracts us from researching and understanding research into the breeding systems, ecology, and evolution of this rapidly diverging group. (Of course, anybody researching the group will need names of some kind for the splinters.) I see the same problem with a rapidly speciating group within Lomatium, where I think calling the whole group a species and using subspecies names for the new groups within it would be better than the current practice of calling them nine or so virtually indistinguishable species.
There’s good reason for this, given the diversity of reality that has no concern with our need for neat, mutually exclusive words. Most of us practical taxonomists use a hierarchy of definitions and try to be consistent with other taxa in the group we study.
I do not share this apocalyptic view of the effect of taxonomic splitting. I do agree it’s frustrating to deal with! Some of it is not useful (a criterion for species delineation, in my opinion). I suspect that the pendulum will swing back on some of the worst of it, and/or iNaturalist will accept a lot of species complexes, long before iNaturalist is destroyed.
I do think that fine taxonomic splitting could put off people who are just getting into the field, especially the enthusiastic amateurs who are essential supporters for all kinds of conservation efforts. The problem can be reduced by emphasizing genus level ID’s, e.g. Dowitchers among the birds.
14 Likes
In general I agree the instability of scientific names and tendency towards splitting based on minute differences is frustrating and unhelpful for learning or for practical application. I feel like this instability is inevitable as we come to terms to how to apply our new abilities to sequence genomes and use that to study phylogeny, and hopefully over time things will resolve as people figure out how to compromise between genetics and phenetics…? I also feel like it’s worth noting that this dandelion taxonomy far precedes 2015.
I’m kind of hoping that as everyone becomes more aware of the dandelion taxonomy chaos, maybe someone will be frustrated enough to come up with a better formal taxonomy for the genus. But as it is, there is a well established taxonomy that just hasn’t been applied in North America before, because people weren’t all using a platform like iNaturalist that requires a globally consistent taxonomy.
7 Likes
well, in my experience, most people who support splitting discard any pushback at all, so i am not that worried about attracting the credibility of splitters any more. I was recently at the actual Kew botanic garden in fact, and it’s wonderful, and they do neat conservation work, i guess i should just say i feel their approach to taxonomy is fundamentally flawed, not that the people there are bad scientists or malicious. But i feel like most people just take it that way anyway…
Well that confuses me, doesn’t constant change, and additional taxa that can’t be parsed out of previous datasets, create this problem more than the taxonomy freeze? Note i am not proposing a forever freeze, just a freeze for a decade or so while people do the research, then one unified jump into the changes, with robust synonomy tables and tracking. Plus, more use of subspecies/variety. I don’t understand the apparent apathy of some scientists on here as to how devastating to field ecology and historic ecology the changes can be. Mostly i just get attacked personally, or it’s handwaved away as not a problem.
yeah it makes sense that there won’t ever be a consistent and agreed upon definition of species, but my point is that being the case, why not use a definition that allos for the most utilization for conservation, understanding, and science?
Well, i hope you’re right! It’s certainly gotten a lot harder for me to use and to be honest i use iNat much less for ‘professional’ ecology than i used to and now use it much more informally and just for fun. Which maybe is for the best, but i wanted this to be a tool to unify both worlds and in the current approach that can’t work in my opinion.
Well, that would basically require splitting the genuses a bit too. Like you I deal with Carex a lot and with that genus, if you’re a wetland ecologist, genus level ID might as well be throwing data in the bin. Most genuses (genii?) are much too large to use that way. I do think complexes and subgenii do help, but it’s a whole lot harder to use those than to just go to subspecies/variety for splits. Oh, and the obsession with which old white dude found the species first needs to end. A name should never be changed just to support a dead person’s ego. It’s meaningless who named it first because countless humans already saw the species before they described it anyway. Nomenclature should follow utility, not ego. No changing names because Linnaeus’s cousin happened to see it six minutes before the established name, and no names like “Schoenoplectus tabernaemontani”, ever. If it doesn’t fit on the datasheet, it belongs in the dustbin of history.
3 Likes
I do work with field ecology and land management. And, sometimes, I see species level ID as not specific enough. Look at Phragmites australis. The two subspecies have wildly different ecologies and appearances. One I want on the land and the other I do not.
If your problem is differentiation, then genera like Carex are difficult, if not impossible to ID most of the time. Dandelions are no different, but their ecologies are generally more similar to each other than many Carex. To be honest, from my land management perspective, I’m not convinced dandelion ID matters at all. I don’t really care if they’re stuck at genus. But, once again, they’re far from the only taxa stuck at genus level, anyway.
Of course splitting is a scientific process. It isn’t perfect, and it’s somewhat subjective, but somethings things have to be imperfect ans subjective. New taxonomy (including both lumping and splitting) helps up better capture the biodiversity on our planet.
Don’t forget that if taxonomy on iNat gets too old, many won’t even want out data anyway. Plus we’d stay behind messing up GBIF.
3 Likes
splitting is totally subjective. Or i guess it’s consistent in ‘any discernable difference no matter how small or indistinct must be entirely elevated to species level’. If you’re able to tell the Phragmites by look though, i’d love to hear how. For sure differentiating those is valuable, but none of our botanists here have found anything consistent other than genetics to parse them out.
I’m not so sure, i think there’s a balance for sure, but if it’s ‘new’ the way the splitters want, no one wants it either. And i don’t think iNat’s policies, split or lump, are anywhere near the biggest issue GBIF has.
3 Likes
You also seem to have this ridiculous idea that “splitters” are these evil people that are “coming” for every genus to suddenly make them hard to ID. No, they just study them and realize the diversity is higher than we though. Many species will never get split.
1 Like
nah, that’s covered above. i say again and again they aren’t evil, just wrong. The harm they do isn’t malicious, but that doesn’t reduce the damage.
3 Likes