Sceoporus undulatus complex map

[alexb0000]

Hi everyone, those of us who identify a lot of Sceloporus lizards are well acquainted with the uncertainty surrounding the ranges of the four fence lizard species that used to all be part of Sceloporus undulatus. These species were split back in 2002, but there’s still a lot of uncertainty about the exact boundaries of the species. The map included in the original paper outlined general species ranges, but didn’t investigate where the species meet, and didn’t include any features beyond state borders, making identification in these untested regions tricky.

I decided to create a Google map to plot out all the samples in the below papers to hopefully create an easier map to reference when trying to identify lizards in this complex. The original 2002 paper doesn’t include exact latitude/longitude coordinates for its samples so I had to estimate where the samples came from based on the information it did provide. In most cases this doesn’t really matter since the samples are very widely spaced, so a “good enough” location still serves its purpose. The supplemental 2007 paper sampled a lot more S. cowlesi and S. tristichus specimens, and did include exact coordinates and one known intergrade zone, so it allowed for a more precise mapping of those two species.

These are the referenced papers:
Adam D. Leaché and Tod W. Reeder, Molecular Systematics of the Eastern Fence Lizard (Sceloporus undulatus): A Comparison of Parsimony, Likelihood, and Bayesian Approaches
https://academic.oup.com/sysbio/article/51/1/44/1631347

Adam D. Leaché and Charles J. Cole, Hybridization between multiple fence lizard lineages in an ecotone: locally discordant variation in mitochondrial DNA, chromosomes, and morphology
https://www.semanticscholar.org/paper/Hybridization-between-multiple-fence-lizard-in-an-Leaché-Cole/3c8676752ff9830ff504ccf34982849f67f8c510

And here is the map itself:
https://www.google.com/maps/d/edit?mid=1k0sbMWkNjZfbH5Qtmk6UcQ4r1XLetGLB&usp=sharing

The overall S. undulatus complex range was made by basically tracing iNat’s rangemaps for the four species, with some added areas for places with confirmed observations outside of the accepted range. The areas labeled “tested range” are a conservative connect-the-dots around the sampled areas. with the lighter-colored “estimated ranges” places around those tested sites where everyone seems pretty confident what the locally-occurring species is.

The biggest unknown area is most of Mississippi, there’s a known population of Prairie Lizards in the bottom of the state along the Louisiana border but it’s unknown where that population meets the Eastern Fence Lizard. Further north it’s generally accepted the Mississippi River separates the species, although there’s only once tested specimen near the river itself, near St. Louis.

In the west the most contested area is northern New Mexico, though once I plotted everything out on the map it actually proved to be better-tested than I thought. The mountains around Santa Fe and Taos and the top of the state tested pretty strongly as Plateau Fence Lizard territory, though there’s still a large unknown area north of Albuquerque extending almost all the way west to Flagstaff. There was also no clear tested boundary between the Southwestern Fence and Prairie Lizards in eastern New Mexico and western Texas, and a smaller unknown range between the Plateau Fence and Southwestern Fence Lizards northeast of Tucson, but neither of those regions seem to produce as many observations on iNat as that patch of northern New Mexico.

I hope this map can help others identify lizards in this complex. If anyone knows of any other papers that would provide good data to add please let me know. I’m aware of Leaché’s other paper at https://academic.oup.com/sysbio/article/58/6/547/1633600, but the samples he references in that one are the same as in the 2002 paper.

[cthawley]

@alexb0000 This looks great thanks for making it!

This matches up with what I’ve seen as well for the current best knowledge of scelop distributions. The areas in LA and MS are currently the most problematic to me; there doesn’t seem a way to be sure of what species is actually present in those locations, though I wouldn’t jump in and knock back all the scelops in MS to genus level IDs.

One potentially interesting thought is that the phylogeographic break for many species in AR and MO is not the Mississippi itself but Crowley’s Ridge. Leache et al’s map seems to follow the ridge (though maybe this was just to avoid overplotting the range map lines so close to the Mississippi).

Also linking this similar discussion from a month or so back: https://forum.inaturalist.org/t/different-sceloporus-species-range-in-mississippi/18212/6

[jnstuart]

Nice map. I haven’t looked at the specimen records for these species in the Southwest for several years but the ranges as depicted in NM are a decent approximation. I’d point out that S. consobrinus in eastern NM is probably confined to the grasslands and does not extend into the mountains as depicted, but we really don’t know where the contact zones (if any) exist in the state, unlike in Arizona where there is some information. (All three species in NM – cowlesi, consobrinus, tristichus – seem not to be habitat specialists, occurring in various vegetation communities, which is not helpful in identification.) The former S. undulatus has been well-documented from every county in NM so it seems likely that there are such contact or overlap areas.

Another reference is the book "Lizards of the American Southwest: A Photographic Field Guide by L.L.C. Jones and R.E. Lovich (2009, Rio Nuevo Publishers, Tucson). Although the shaded range maps for S. cowlesi, consobrinus, and tristichus are not particularly detailed, the species accounts were co-written with A. Leache so I assume the maps are a reasonable approximation of distributions based on limited data.

There are parts of northern NM where I personally would still hesitate to put a species-level ID (although I’ve done it on some iNat records). Someday we might have better data and then such IDs might be possible with better confidence.

[alexb0000]

@cthawley, thanks for the info. I’m not very familiar with Midwestern geography, I’d never heard of that ridge before. If that was the divide it would certainly why the Mississippi didn’t prove to be a barrier at the delta where the river’s the widest, but allegedly was further north. Of course the wandering of the river over time might have also cut off a population at one point, I know the Mississippi wanders a lot but I’m not sure if the drainage has moved over time.

@jnstuart thanks for the info, I’m not very familiar with eastern New Mexico, you’re talking about the mountains up in the north of the state near Taos correct? I’ll update the estimated ranges to make the mountains more S. tristichus territory. I’m most familiar with the Western Fence Lizard so it’s always weird for me to think of their analogues as specialists, since Western Fence Lizards are such generalists.

[jnstuart]

I’m not convinced the former undulatus species are habitat specialists — I suspect they aren’t — but from what little we understand about consobrinus in eastern NM it seems to be associated with the grasslands and plains and not mountains. The mountains to the west (Sangre de Cristo) presumably have tristichus. But without better genetic data that’s really an assumption that might be inaccurate.