There seems to be a systematic issue with identifications of glowworm beetles, genus Phengodes, from the United States. There are about ten species, and some named subspecies in this area. The distribution of these species is perhaps poorly known, and species-level identification likely requires specimens and microscopic examination.
Also, keys may only exist for males. Larvae, and the females, which are larviform, may not be keyed in the literature, though I am not 100 percent sure of this. The state-of-the-art seems to be a 1975 publication which is behind a paywall. I have read the summary, and I believe it keys only adult males. Again, I have not seen the whole thing:
Wittmer W. The genus Phengodes in the United States (Coleoptera: Phengodidae). Col. Bull. 29: 231-250, 1975. https://www.jstor.org/stable/3999661
In particular, some observers have been assigning many images of Phengodes larvae to…
Phengodes laticollis
This is a common species, apparently, but I just do not see any documentation on how the ID’s are being made. Adults are being assigned to this species as well, and I am not seeing any justification for that either. Many confirmations of IDs are coming from one user who does not appear to have a background in the group, and is not leaving any explanatory notes–I feel vital for such a difficult group. There are over 100 observations, many research grade, many verified by that one user.
I sent an email about the issue to an acquaintance who is a beetle taxonomist (active on BugGuide, but not iNaturalist). I quote his succinct response: “Phengodes is too much of a mess, I wouldn’t even try…” If the professional taxonomists are not able to sort out the species in the genus, I do not think citizen scientists are likely to be successful simply by looking at images. AFAIK, different species of (both adults and larvae) in this group are visually similar and are not, in general, identifiable based on images.
IMHO all of these ID’s should be backed off to genus-level. I am not sure how to proceed. There are rather too many to correct with individual comments. I have made a few, giving documentation, and have gotten some people to agree, others not.
The site does not support, not does it have any tools to bulk add or replace identifications. The only option is to add individual identifications on each record
Yeah, that’s what I thought, and that is normally a great feature. This appears to be a systemic issue in this genus, and part of it is due to an enthusiastic, but perhaps overconfident, user. (I’ve been there many times myself–no disrespect intended!)
My thought would be that people knowledgeable in Coleoptera give some feedback here, in case I am missing something. Then perhaps several people could go in and start pulling ID’s back to genus level, if that is the consensus. Perhaps everyone suggesting the ID correction could reference this discussion?
One caveat is that there is a book (below) that includes this group. The author, Lynn Faust, has made one ID of a larva as laticollis on BugGuide. However nobody on iNaturalist is referencing her work that I see. I have not seen her publication, perhaps someone else here can comment?
-Fireflies, Glow-worms, and Lightning Bugs: Identification and Natural History of the Fireflies of the Eastern and Central United States and Canada (2017). ISBN-13: 978-0820348728
You can prepare a message about how group is too complex to be ided from photos with some references, it can be then used by you and everyone else reiding, users are learning by such messages pretty quickly and amount of new ids will get lower.
Great, I will start working on it, and will run it by this group for edits. Also hoping to get some volunteers to work in parallel and, perhaps, some expert commentary on the group.
You can subscribe to Jstor as an independent researcher for free, and read a limited number of papers. This particular one i have checked, and will not help much for photos, Wittmer’s keys are best used under a microscope… Maybe the book is more user-friendly.
Thanks for the tip. I’ll try to get a look at the book.
Well, here is another thought, as this is, I feel, an issue with a number of taxa. Would it be possible to add a curation feature to iNaturalist such that a particular taxon could be marked as “not typically identifiable from images”, and this would warn users not to attempt ID at that level, but to choose a higher level, typically genus instead of species? Seems like some sort of warning or flag would be valuable. These flags could be suggested by users through the “mark for curation” feature. I am happy to post this idea in a different forum if people feel that is more appropriate.
It’s being discussed quite often, there was a proposal to add notes to taxa about which parts are needed to be pictured for an id. In fact for prevention more focus should be on iding etiquette, if people knew they shouldn’t reach for species no matter what, many situations wouldn’t happen.
There are a lot of institutions through which people can access JSTOR with unlimited downloads. Happily, my public library (Austin Public LIbrary) has an institutional subscription. I sign in through them with just my library card number. You might see what is available to you regionally. Go to the main JSTOR page at,
https://www.jstor.org
Click on “Log in” in the upper right and then click on “Find My Institution” to bring up a list of cooperating entities. It’s a long list!
Again, thanks, I downloaded the paper and looked through the descriptions and key. As I suspected, the key is for adult males and does not even mention the females or larvae that I can see. As you noted, the key for the adult males is based, largely, on fine measurements of a specimen in hand. Many of the species-level ID’s on iNaturalist are of larvae, which are not keyed, or described, at all in that reference. Unless there is some good description of species ID for larvae elsewhere, I am not seeing any justification for assigning those images of larvae to particular species. Some of the images of males may be assigned correctly, but this should likely get a critical look as well.
I will try to get a look at Faust’s book in case there is something useful there.
Going forward, I feel that the community should make an effort to bring identifications in that group to where there is a decent level of confidence in them.
This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.
Thread reopened at my request. Thanks too the forum moderators!
Update. I looked at Lynn Faust’s book Fireflies, Glow-worms, and Lightning Bugs (University of Georgia Press, 2017). There is not a lot of information about Phengodes species. There is, I believe, just one photograph (adult male, I think), and it is not identified beyond genus level. There is no discussion of species-level identification within the genus.
So I can summarize:
- The only key I can find for species within this genus is Wittmer W. The genus Phengodes in the United States (Coleoptera: Phengodidae). Col. Bull. 29: 231-250, 1975. (link in discussion above). The key covers adult males, and the characters used are careful measurements of various structures. These would only be accessible on specimens in hand. AFAIK, there is no published key for larvae or for the adult females, which are larviform in this group. Only male Phengodes are potentially identifiable to species, and then likely only if a specimen is available.
- I can find no other source that has reliable information on differentiating species via photographs, be they of adult males, adult females, or larvae.
- There appear to be several species in eastern North America. There are several other species in the Southwest, including Arizona.
- There are a number of observations of Phengodes larvae here on iNaturalist identified and verified at species level. Images thought to be of P. laticollis larvae, for instance, are commonly identified as research grade. Most, if not all, of these identifications appear to be over-specified. They are not based on definable characters in the photographs, only on resemblance to other images.
- I feel that most of the species-level identifications of Phengodes here on iNaturalist should be reclassified as identified to genus-level only. This would include all images of larvae and of females. (Note that differentiation of a larva from an adult female requires a good view of subtle characters, including various abdominal setae, presence of a gonopore, and compound eyes.) Edit. This would also include most images of adult males. Absent reference to a keyed specimen, it seems species-level ID is seldom justified for those observations either. (end edit)
Some references:
-BugGuide family account: https://bugguide.net/node/view/9382
-BugGuide record of a larva, possibly identified as P. laticollis: https://bugguide.net/node/view/492216/
-Texas Entomology: http://texasento.net/Phengodes.htm
-University of Florida, “Featured Creatures” https://entnemdept.ufl.edu/creatures/misc/beetles/glow-worms.htm
Demotion completed for observations made in the U.S. Can you do the same in BugGuide?
Wow, great job. I will look through BugGuide later this week. I’ll put a discussion thread there as well. There are a couple of adults there identified via museum work, and one larva Lynn Faust says is likely laticollis. Certainly she has expertise there!
Well, I only thought I had got all of them…several slipped through. Will continue working as time permits.
that paper is free through jstor, as others have mentioned, but for what it’s worth many papers that are actually behind paywalls are accessible for free through sci-hub. doesn’t hurt the author(s) since they don’t get the money anyways lol. science is for everyone, whether everyone has like 45usd to pay for 24-hour access to a paper or not (the timed access ones drives me crazy!) :)
@cotinis, if anyone needs specimens from Texas, I have a dozen or more adults collected as bycatch in a Malaise trap–stored in EtOH. Here’s one of them–first one identified as this species on iNat and BG (assuming I’m correct): https://www.inaturalist.org/observations/120599112
I’m guilty of a few of these species-level IDs, which I’ll try to find and correct. I’m glad another user directed me to this discussion.
BugGuide is a fabulous resource, and I usually try to check out its references before plunging into species-level IDs, but as cotinis pointed out, it’s sometimes hard to track them down and they may hide behind paywalls. I’ve gotten a lot more cautious about cross-checking references as I’ve gained more experience on iNat.
Edit: I just did a search and could only find one, which I backed off. Apparently I didn’t get as carried away as I feared!