Hi! Im am wondering about what species that should be left at the genus level due to only pictures of the organisms and the organisms need to be studied in high detail such as periwinkle (plena and checkered). Please give me all the species that need to be left at the genus level so I can add them to my genus list for identifying in the future. YFI, I am no expert on any of these identify processes so I am going to only focus on species that I can identify just by the photos, such as plants and mammals.
1 Like
To judge by most discussions about identifying, you’d get the impression that this is most taxa. Slugs, Millipedes, Moths, Mushrooms, you name it, someone on the Forums has taken on the project of bumping them all back to genus because “these cannot be identified to species without dissection.”
7 Likes
There is a list here of some of the species that the computer vision has had difficulty with in the past, that may be helpful: https://forum.inaturalist.org/t/computer-vision-clean-up-archive/7281. In general, if you are not sure of a species-level identification, identify only to the level at which you are confident, such as genus or family.
5 Likes
There are some well known examples in the fungal world, for instance, red Russulas with white gills and stipes generally can’t be distinguished without microscopy to examine the spores. Same with some of the fruiting bodies of Chlorociboria.
5 Likes
Geometer moths in the genus Xanthotype can only be reliably ID’d to species via dissection.
So they typically get RG at the genus level.
3 Likes
This list will vary greatly depending on geography. There might be only one species from a particular genus known from Greenland, but two hundred from the same genus in Guatemala. So in Greenland you’d have a good guess what species you have, while further south you’d need to dissect.
Given this, I suggest you specify a geographic location and assemble a list for that location.
5 Likes
No they haven’t.
IDers don’t find perverse pleasure in bumping observations back to genus with the sole purpose of frustrating observers. We don’t enjoy saying that observations need to be left at genus either. We would also find it much more satisfying to be able to provide a species ID. But often we can’t. It isn’t for lack of trying.
Many IDers of taxa that traditionally require specimens for ID are working hard on finding ways to use other criteria to allow for photo ID.
The taxa that are difficult to ID from photos come up frequently in the forums because they create obstructions in the IDing process – IDers have to deal with challenges like cleaning up a taxon that thas been persistently mis-IDed and figuring out how to decide when to bump back an observation and when to mark it as “ID cannot be improved”, and how to manage the “needs ID” pile when RG cannot be used for guidance about whether it is likely an observation has been looked at by someone knowledgeable or not. These things come up in the forums because many people find it useful to discuss problems as part of working through them.
For taxa that can generally be ID’d based on photos, these problems do not occur; when things are running smoothly, people do not generally post to complain or look for solutions.
16 Likes
What organisms are you interested in and what region do you generally ID in? I ask because, as other people have said above, it all depends on what would help you out. It takes time to assemble, say, a list of moths for my state (Massachusetts) that, as far as I understand, cannot be IDed to species from a simple photo, but there’s no point in putting in that time if you’re interested only in plants or mammals.
2 Likes
Plants can be difficult, too. Lupines, many of the Poales, and Asteraceaes in the Pacific Northwest aren’t super easy to id by photo, and unless the user has provided measurements and/or photos of important characteristics (like for Lupine, the banner, keel, and wings all separated), I won’t try to id to species. Flora of the Pacific Northwest, 2nd ed, Hitchcock & Cronquist is still pretty good for our region. I’d suggest trying to key things out until you get a feel for, “This is the level to which I am confident identifying this taxon.”
And don’t just trust the CV! It’s left off rare plant species for me more than once. It cannot detect subspecies, either, or forms like Actaea rubra f. neglecta. For fungi, people iding photos with common “forager cheats,” like calling anything vaguely orange and jellylike “Witches’ butter” - Tremella mesenterica - in the Pacific Northwest has screwed up the CV. Dacrymyces and Naematelia aurantia get incorrectly called Tremella mesenterica by the CV for that reason. The CV is comically bad for spiders, too.
I took mycology classes in uni, but I usually leave fungi at genus, too. I think I’d need a barcode to feel confident with most of them lol. Functionally, all I care about is, “Is this a chanterelle?” and, “How am I going to cook these delicious chanterelles?” and, “How can I avoid anyone finding my wonderful chanterelle spot?”
5 Likes
Great question Therese. I can see a few dimensions to consider, most of them echoed by others in this thread. Let me give a few examples from what I’ve learned over the years, mainly from others on iNaturalist
- Species that can be identified only by dissection - one example that comes to mind is subgenus Arion, a group of slugs. Thanks to @pileated I learned that “Arion ater and Arion rufus can only be separated by dissection and comparing genitalia. The black morph of Arion rufus is not uncommon.” I also learned about the concept of a species flock - a group of closely related species all living in the same ecosystem.
- Bees in Pyrobombus require details to disambiguate to species. A few examples where I learned: Bombus flavifrons, thanks to @greenmother, Bombus vancouverensis ssp. nearcticus with help from @rainhead, his profile has lots of useful bee ID info. Washington Native Bee Society project is also very useful, thanks to @wenatcheeb for introducing me to this project.
- There are some species that require noting certain attributes to differentiate them from others. For example, in Washington State. Flora of the Pacific Northwest 2nd edition is a great reference BUT one needs to learn the details. If the observation you are looking at lacks a key detail, it will be ambiguous. For me, that’s been a trial and error process. I have found the simplified keys by @jhorthos to be very useful, for example: Lomatium key, Astragalus Identification (Washington State) and Ivesia tweedyi (non-technical)
- I have learned that if I’m not certain then ID at genus or family level rather than species. I make mistakes but I will always do my best to correct my errors and learn from them. I have done a lot of learning over the years!
6 Likes
Lupines are a particular challenge for me. Sometimes I will get observations where I know that it’s a Lupinus species but do not key out to species. I am interested in the ecology of an area and lupines play an important role, in particular as nitrogn fixers. In aloine areas, I am often hiking with a group of friends and so I observe and then run to catch my group.
2 Likes
The biggest Pyrobombus example I encounter is B. vosnesenskii vs B. caliginosus. Sometimes you can rule out one or the other, but often not.
Don’t forget Castilleja spp! I identify paintbrushes to species only when they present in very typical ways, in regions where I feel like I know all the possibilities.
2 Likes
For fungi ID I like to use the Audubon Mushrooms of North America guide because when there are many look alikes it labels them as a group. To pick a random example opening to page 399, I find the Hycrocybe conica group.
I would like iNat to guide me in a similar way to a group ID between the genus and species when it’s possible to be more specific than the genus but the species is unknowable, not easily determined or cryptic. If it could be made easy for observers who are not expert to get to the right level of ID, I would very much appreciate that!
2 Likes
Yes, great point. The genus Castilleja has a lot of variation. I use the same ID algorithm as you, I know the possibilities in areas like Mount Rainier and the Olympics, even to the subspecies (in most cases). As I get further afield I usually have to say Castilleja sp. and then I seek help from @markegger
This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.