My personal preference is a lot more permissive than the official iNat policy, which I suspect was written long ago and perhaps in response to a particular situation at the time. I am fine with the iNat taxonomy including hybrid plant taxa in these scenarios:
Natural hybrids with identifiable characteristics occurring with moderate frequency (i.e. we don’t need to have a hybrid formula in the taxonomy for every possible combination ever recorded, but it’s worthwhile for any identifiable thing an iNat user is reasonably likely to find in nature.)
Horticultural hybrids that are known to naturalize, e.g. Crocosmia × crocosmiiflora.
Widely grown horticultural hybrids that are commonly added by novice iNat users. I’m not suggesting we distinguish cultivars or add many of these taxa. But putting a precise name on Iris × germanica provides knowledge to the observer, and helps separate out these hybrids from their parent taxa in regions where both may occur.
In all those cases, I think iNat users and the dataset benefit and the cost is only a small increase ion the size of the iNat taxonomy.
I was going to make a post about this but it looks like you guys have a good chat going. My inquiry was about introgression. If we have introgression, this is valuable information, but the specimen may only be 10% species A, and 90% species B. Imagine there existed species A, a rare native species. An alien species (B) in the same genus was introduced in proximity to the last disjunct population of species A, in 1824 and started crossing with A. This was left undocumented for decades, until a biologist noticed pedunculate fruit on a species that has sessile fruit. He visits a local herbarium and finds specimens of species A, that were misidentified, proving they were there in the 1830s. How would you label the spcimens that remain growing wild? they are on average 90% invasive (species B) but the small amount left is visible in some specimens, and making sure the species A signal is described is important data. Would this be labelled a hybrid? is this enough DNA to be considered a hybrid or would it be labelled as the next higher taxon? Thanks.
This is a good example of why the Curator Guide was pretty conservative about when to add hybrids as taxa.
When I see an observation that I can tell is introgressed as part of a hybrid swarm, and there is no existing name for the hybrid, I will do one of two things:
If the genetic influence of each parent looks about equal, I will identify at the next ancestor level, and leave a comment about the introgression I am seeing.
If the genetic influence is clearly dominated by one parent, I will identify as that parent, and leave a comment about the influence I am seeing from the other parent.
If one of the parents involved is of conservation concern, and needs to be tracked for all introgressants, I might add an observation field for that purpose.
Why not just identify all introgressants involving a rare taxon as that rare taxon? Because iNaturalist will then treat all photographs in those observations as photos of the rare taxon, even if the genetic influence is slight, thereby misleading both humans and Computer Vision (CV) about what “good” examples of the rare taxon look like, and making it even harder for CV to distinguish between the parent taxa and (if hybrids are ever allowed in CV) their introgressant “taxon.”
I would think that if any hybrid ancestry can be identified, then it would be better to place in the hybrid bin rather than the parental species bin. In the example where the plant is 90% species B and 10% species A, then I would still ID as the hybrid rather than species B, assuming the hybrid taxon existed on iNat. Obviously, this could be taken to a philosophical extreme - what if it was only 1% species A, what if there was only 1 bp of DNA left from species A, etc, but in practice I think any field-identifiable hybrid ancestry would warrant going in the hybrid taxon bin. (In practice there would be cases where identifying hybrids wouldn’t be practical, like cases where a large population of species B is known from genome sequencing to have 5% ancestry from species A due to ancient introgression, but they are still considered species B and have no discernible hybrid traits)
I think I am going to go with your stance. Using Quercus as an example, I will treat any oak with traits from two species as the hybrid of those parents. If 3 species are known locally and all three are visible, then I will treat it as the section it is in and make a note. For example, a hybrid of: ‘Quercus alba, Q.macrocarpa, and Q.bicolor, could be identified as Section Quercus, the white oaks, with a note in the description. I have examples of these. Or maybe identify it as the dominant hybrid and add the 3rd species in the notes secton?
Undoubtedly artificial in origin as Aesculus hippocastanum and Aesculus pavia come nowhere near each other in their natural ranges. Therefore it had to originate when humans planted the two species next to each other. Did it originate as a deliberate cross or did it originate through accidental natural pollenation? Good question.
However it now breeds true and can reproduce. It’s in iNaturalist and a great many of its observations are cultivated plants. I’ve marked some as such myself. However since it breeds true and given how common it is there is every chance there are some genuinely naturalised specimens around.
How do we treat it if “artificial” hybrids are excluded?
Hybrids of this type are different than cultivars. Indeed Aesculus × carnea itself has multiple cultivars with the most famous probably being Aesculus × carnea ‘Briotii’
My own thoughts? Cultivars should never be an option for wild organisms. If they are to be supported at all they should only show up for organisms marked as cultivated. A low priority feature to say the least.
Hybrids on the other hand should be more widely supported, at least for plants. Even if they are “artificial” in the sense that they require human intervention to plant the parent species next to each other they can still arise spontaneously through natural pollenation of those parents. So iNaturalist should support most of them for plants.
I agree with this when there is a hybrid taxon bin available in iNaturalist. I was addressing how to deal with varying degrees of introgression in iNaturalist when
I tend to follow your approach of ID to section (or as far down as possible) if 3 species seem somewhat co-dominant, if there seems to be a clear dominant pair, I’ll ID as that hybrid and add all three parents into an observation field to note which species are apparent in the specimen.
I tend to be somewhat expansive in my use of hybrids, but I think 90:10% would be a little far to go. At least for Quercus, most literature stops calling them hybrids at the F2/F3 level (25 and 12.5% minor parent respectively) and switches to calling them “introgressed” or having an “affinity” to a 2nd species.
12.5% is pretty darn close to 10%, but I think it makes more sense to consider in terms of “simple” hybrids and backcrosses in terms of answering this question, though it also is surely more messy in practice. If we split the difference of the percentages, we’d stop at 18.75% as a cutoff, which rounds nicely to 20%, which feels right to me as well.
Near 10% for most taxa I think it would be much harder to be sure by morphology and definitely more difficult to come to a consensus opinion based on photos. With DNA barcoding on the rise, the nuance probably increases.
Of course any cutoffs are totally arbitrary, but I tend to favor my semi-systematic approach to it here. :)
I agree it’s arbitrary, and in many cases individuals with 10% introgression would no longer be detectable as hybrids by morphology. 10% hybrid ancestry can be just 2 generations of backcrossing (recombination leading to variation from the 12.5% expectation) so I would definitely still consider them to be hybrids - though I agree in a publication I would refer to them as a “later-generation backcross” or “having hybrid ancestry” rather than just calling them a “hybrid” in order to be more precise. I think in practice, since it usually isn’t possible to know the %, it would be best to put them in the “hybrid” bin if they can be detected as having any hybrid ancestry (eg., slightly introgressed Glaucous-winged Gulls getting ID’d as Olympic Gull hybrids if their wingtip colour is detectably a shade too dark) unless established conventions are different for a given taxon (eg maybe Quercus where there has been so much introgression), or, in cases where an entire taxon is introgressed, to reserve the hybrid bin for individuals that have noticeably greater hybrid ancestry than average for their taxon. (That would mean the exact % cutoff would be different for different taxa since some will be more detectable than others, but that’s of course already the case even for first-generation hybrids). I think in most cases that’s what would make the data most useful - marking the introgressed individuals as hybrids so they don’t get lost and are more easily accessible to researchers. At least, as someone studying hybridization that’s what would be more useful to me - making it easier to eg. get an idea of the geographic extent of hybrid zones (with the caveat that morphology won’t show the full story).
How can one distinguish introgression from infra-species variation? Example: some S. California bush monkeyflowers may be hybrids or pollinator-driven variants, or even separate taxa; there are clearly differences. And, while the taxonomy is leaning towards hybrids, the genetic science points toward variation. It seems like jumping on hybrid naming obscures this, while it at least can capture some of the observed variation. Unstructured comments seem problematic for searches; maybe tags?
Good question, and the distinction may not always be meaningful. To me, hybridization and introgression are phenomena of secondary contact between species that had been fully independent lineages at some point in the past. The area of contact and hybridization tends to be limited to a very small area relative to the range sizes of the parent species, and the parent species maintain their differences and independent trajectories outside the area(s) of secondary contact.
Infraspecific variation, on the other hand, is generally considered to be “primary contact” between partially differentiated populations of a single species.
In some cases, secondary contact and extensive introgression could lead to complete breakdown of separation and re-merging of the parent species. Depending on when during that process the populations are observed, it might be impossible to distinguish them from a case of infraspecific variation.
Introgression is honestly my greatest fear when it comes to the Populus that are native to my area. It’s the topic I want to study the most so I’m really glad this topic is getting some discussion!
The non-native P. alba clone that is planted heavily all over north America hybridizes readily with P. tremuloides and P. grandidentata. I’m concerned the hybrids will eventually introgress to the point where their alien genetics bury their way into the native Populus. Thankfully the morphology for all involved species are relatively easy to tell apart from the hybrids. However, that may just be a sort of ‘survivorship bias’ in terms of how easy they are to identify!
I also believe prominent or common cultivars should absolutely be on iNat. There are planted cultivars that hybridize similarly to the above. P. nigra ‘italica’ being one that is present on iNaturalist, another being P. tremula ‘erecta’, which isn’t. I’m not suggesting every cultivar of rose be added, but I think something like Acer platanoides ‘King Crimson’ could be added. It’s certainly common enough and observed often.
As for a solution, The best I can think of for undocumented, or missing hybrids could be something under the Data quality or annotation. Maybe “Potential hybrid” or something. At the very least it might be easier for people to find them!
If it’s an identifiable cultivar then by definition, unless it’s literally the origin plant of the cultivar which is a wild individual, someone must have planted it at some point. Cultivars do not generally reproduce true as the cultivar from seed after all.
That’s also true of clonally suckering plants like some types of Populus. The suckering cultivar’s origin plant was planted by a person.
Backcrossing of cultivar stock into natural species stock is a different matter. To use Populus nigra as an example, in the UK that would likely be Populus nigra ‘Italica’ backcrossing with Populus nigra subsp. betulifolia. However such backcrossing would no longer be the cultivar but simply Populus nigra.
Having the possibility to ID DQA-marked cultivated plants as a particular cultivar would be useful. However it’s an extremely low priority thing given the focus and mission of iNatutalist.
As @cthawley pointed out, and I discovered on tha FAQ; clonal growth is considered wild. I agree with this, as regardless of cultivation status, growing outside of where individual was planted and intended to stay makes them wild.
I feel my specific situation is a little unique, the exception, not the standard.
There is a cultivar hybrid of the non-native P. nigra and native P. deltoides: P. x canadensis. It also occurs naturally where the two grow together. The same goes for P. alba
I’ve had it suggested I should mark all instances, of P. x canadensis and P. alba as cultivated. But as I said above, I use that designation only in instances where its very obvious or I can be certain. Sort of like “innocent until proven guilty”! I generally do a little bit of digging each time I’m uncertain to see if there was an old homestead, or historic reason for it to be there!
I’m confused. You cut out the part of that quote where I said it also occurs naturally. The hybrid occurs both artificially and naturally. That was my point
When you state this:
If it’s an identifiable cultivar then by definition, unless it’s literally the origin plant of the cultivar which is a wild individual, someone must have planted it at some point. Cultivars do not generally reproduce true as the cultivar from seed after all.
That’s also true of clonally suckering plants like some types of Populus. The suckering cultivar’s origin plant was planted by a person.
I interpreted that as you suggesting any cultivar should be marked as captive/cultivated regardless. I was chiming in with that notion that you can’t always be certain a cultivar species is currently cultivated. Multiple cultivars of the hybrid only compound that fact.