|
dividido en |
@k8thegr8 @paul_dennehy Would you take a look at the Atlases (linked above the map, next to the output taxa, in Green) and see if they would handle all the umbrata records correctly? The ranges are highlighted in green, not the red squares.
When this split is committed, umbrata IDs on a record falling within only the pulverea range will change to pulverea, those falling only within the umbrata range will remain umbrata, and those falling within the range of both or outside the range of both will be bumped up to the umbrata group and will need individual reidentification. I don't know which species is in Arizona, Mexico, and Central America.
Neither species occurs in Arizona; that was a misidentified H. averna, which I just corrected. Neither is known from Central America, and the group hasn't been resolved in Mexico, so I bumped those back to genus (I doubt most of them were even this pair). All the observations left should be taken care of by the atlases by the looks of things.
I don't buy it. The DNA barcoding doesn't back this split.
You can go on NCBI yourselves and compare FASTA sequences: . The umbratas from Canada and Florida share 99.7% of their nucleotides. It's just not possible that they're different species. I briefly ran distance tables in Geneious between H. guttivitta, varia, cubana, umbrata (FL), and umbrata (Canada) 89%-92% similarities for other species of Heterocampa. Again, the Canadian and Florida umbrata specimens shared 99.7% identity.
NCBI database of Heterocampa: https://www.ncbi.nlm.nih.gov/Taxonomy/Browser/wwwtax.cgi?mode=Undef&id=214162&lvl=3&keep=1&srchmode=1&unlock
an example accession of Florida and Canada umbrata respectively: https://www.ncbi.nlm.nih.gov/nuccore/HQ985928, https://www.ncbi.nlm.nih.gov/nuccore/KJ377493
https://auth1.dpr.ncparks.gov/moths/view.php?MONA_number=7990.00&ref=1
Conclusion is H. pulverea, named in 1867, 12 years after umbrata in 1855, is most likely a subspecies of umbrata.
Morphological differences, as much as taxonomists love to dissect insects and look at the shapes of genitalia, aren't enough to indicate speciation.
Less likely explanations are still possible - i.e. the samples from Florida were all "pulverea" so we still need to get some mitochondrial DNA from a real life true umbrata
I made a very quick and dirty tree to illustrate: https://drive.google.com/file/d/1TfcyFhlDMWPHjtkxhBRHNSNC4_vqsdvM/view?usp=drive_link
The cluster at the bottom are all the "umbrata" samples (including a few from FL and GA - used state acronyms and Can meaning Canada) in the barcoding of life project and they are all much less than 1% different (0.008 line in the chart represents 8/10ths of 1%). You'll notice the other species do but none of the "umbrata" separate by more than a fraction of 1%. I'm learning as I go so plenty of grains of salt to be taken from this but I'm stubborn such that I won't be convinced I'm wrong until someone shows me the goods...
assorted references:
https://bmcbioinformatics.biomedcentral.com/articles/10.1186/1471-2105-11-206
https://pubmed.ncbi.nlm.nih.gov/22684951/
http://www.boldsystems.org/index.php/Public_RecordView?processid=LPSOD363-09
https://auth1.dpr.ncparks.gov/moths/view.php?MONA_number=7990.00&ref=1
A relevant quote from James Adams regarding DNA barcoding for separating species, and I'm paraphrasing here- "if the barcodes are different, that's useful information; if the barcodes are similar, that doesn't mean much at all".
As I'm sure you know, the BOLD barcodes are based on a single 658 base pair sequence in a mitochondrial cytochrome oxidase gene. It's usually a pretty informative section of the mitochondrial genome, but having a unique COI sequence is by no means a necessity for a population to be considered a unique species. A 99.7% similarity in that one 658 bp section of one gene, or even a 100% similarity, could easily occur between two distinct species. There are other examples of moth groups where the barcode sequence just isn't informative, as speciation has occurred too recently for unique barcodes to develop; this pair is far from the only place on the BOLD tree where distinct species overlap in their barcodes. Lumping two populations based on that evidence alone would be akin to lumping them based on similar wing pattern, genitalia, etc. If the populations exhibit consistently different COI sequences, that's decent evidence that there's no gene flow between the populations. But if they show the same sequences, that could mean they are interbreeding, or it could just mean they speciated recently and have conserved that particular sequence in their separate populations. The MONA fascicle based the separation of the species on a number of differences, namely:
-The forewing lines of umbrata are consistently more poorly defined than umbrata, and the degree of black overscaling is different
-The genitalia are distinct and easily separate individuals from the two populations
-Umbrata is smaller than pulverea
-The final instar larvae of the two populations are distinct, including shape and color of the dorsal knobs on T1, the color of the antenna bases (bright yellow in umbrata), the abdominal saddle shape, the head capsule pattern, etc.
-The above differences are consistently seen between populations are different geographic locations
Any one of these features by themselves would not be convincing evidence for separating the species, but in the aggregate, they suggest that two species are involved. If the two populations have consistently distinct morphologies in their patterns, size, larvae, genitalia, and ranges, the fact that they don't also differ by a few more nucleotides in a single mitochondrial gene does not, in my mind, alter the conclusion that they are distinct species. I'd like to see hybrid rearing experiments to really get a feel for how distinct the species are; if an inbred population of each "species" can persist for more generations than a population of hybrids, I'd say would pretty much confirm that two names are warranted (that's the evidence that was used to split Antheraea oculea/polyphemus, and I found it convincing). In the absence of that, I'd say tentatively splitting them is fine based on the current evidence.
I agree in principal that "Morphological differences, as much as taxonomists love to dissect insects and look at the shapes of genitalia, aren't enough to indicate speciation" but I would also argue that lack of nucleotide differences in one gene absolutely does not indicate lack of speciation. If it did, there would be some very strange synonymies being proposed.
Thanks Paul that's really informative! I did cursory searches for the LepF1 /R1 primers and COI region for leps but obviously hadn't found what you're saying here.
Today I looked again briefly and found this:
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5453547/
They made a tree just like I did! https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5453547/bin/pone.0178548.s014.pdf
CTRL-F "umbrata" to see theirs and how there's no discerning between the regions you find the moths, of course. So your argument is this newly elevated subspecies, "pulverea" should be accepted despite the barcode sharing.
For Notodontidae an average 0.59% instraspecific variation and 4 out of 125 species "sharing barcodes." So it's possible... although maybe unlikely.
In a case like this the above mentioned morphological differences would have to be extremely convincing. I appreciate you listing out the reasoning as I can't access Miller 2021 chapters so I am left with only this strange unlikely genetic occurence and nothing to prove the speciation. As we all know insects can adapt to their environments very quickly meaning color, size, an even shapes aren't indications of speciation but often temperature/humidity/diet/physiological/hormonal effects. If you consistently see smaller, less hormonal larval instars, etc in the same geographies that makes sense to me, since they could be responding to humidity change, etc. It is making me further dubious.
But moths aren't my thing. In my limited experience in vector-borne pathogen genomics, this type of identity isn't possible between species as this is a highly variable region chosen for this purpose. So when I searched and read through a few articles on the subject saying <2% is the cutoff I assumed I had found something more interesting.
Edited throughout answering my own questions wasting more of my time but trying not to waste yours!! Hahah
Thanks again
Don't commit this taxon split! The atlases are not yet finalized.