Just some comments about some of your statements....
Geography: Ranges do not overlap, but are separated by a gap of several hundreds of miles. No populations having intermediate specimens known to date.
The absence of intermediate specimens in a poorly collected area of north American may not mean anything at all.
Furthermore, there are significant gaps within the range of Texas bairdi populations, such as in the Davis Mts and Chisos Mts.
They are only really continuously distributed in the eastern part of their range. I don't think this gap (I assume you are basing this on a thorough knowledge of museum specimens?) has any significance in this light.
How much evidence of gene flow is there between the populations as we know them now?
Ecology: The Texas phase is adapted to a hot and arid environment, therefore it's mostly nocturnal and spends most of the day hidden deep down in rocky crevices. Texas phase bairdi rarely, if ever, climb trees or other kinds of vegetation.
In much of their range, TX bairdi are restricted to cooler heavily wooded hillsides. The only reason we get the impression that they are snakes of open desert country is because most bairdi that are found are found by people hunting for alterna at night. Bairdi are much more common in wooded areas.
The Mexican phase on the other hand is closely linked to much cooler high altitude pine / oak forest, and is known to be a very arboreal snake. Also, it's diet seems to be a bit more focussed on birds instead of rodents or lizards.
If this is true, it may simply be a populational adaptation to the habitat that is available. I don't know that this basis alone would warrant separation of a questionably allopatric population.
Furthermore, where does the habitat transition take place? In Big Bend National Park, bairdi are also largely restricted to the high pine/oak woodlands.
Morphology: The Texas phase bairdi has about the same body shape as any other Pantherophis species. The Mexican phase on the other hand has a laterally compressed bodyshape, a trait shared with many arboreal snakes, giving the animal a much more slender appearance. Also, there are some considerable differences in head structure. The Mexican phase has a more slender mouth, with a prominent ridge between the eyes and the nostrils. The eyes obviously are significantly bigger than in the Texas phase and seem to be directed a bit more forward in the Mexican phase, alowing a higher degree of bifocal vision.
If this is true, is it a character that changes abruptly at some area, or is there a slow clinal shift as you go from one area to the next?
Are you basing this on the few lines of Mexican bairds in captivity, or have you examined real Mexican specimens in collections?
Ethology: The Mexican phase bairdi has a very fierce feeding response compared to the Texas face, and they seem to like birds a lot more better too. Also, they are very arboreal snakes and are significantly more flexibel in their diurnal versus nocturnal activity modus, whereas the Texas phase bairdi is mostly a nocturnal and fossorial snake and rarely, if ever, climb at all.
I suspect that over most of their range, they climb frequently. And this is where the transition doesn't make sense.
If you are proposing that TX pops are more closely related to obsoleta, why don't they show they same arboreal tendencies as obsoleta in central Texas where they come together? I think the probably do. The problem is, we tend to look for them on roads at night. Where are they hunting at dusk or on cooler days?
The juveniles of both phases on the other hand are quite similar, but this similarity could be selectively driven by its shared similarity to another snake which shares their habitat. The Texas phase bairdi occurs in the same area as Crotalus lepidus lepidus, whereas the Mexican phase's natural range coinsides with that of Crotalus lepidus molurus. This kind of mimicry is also known from another species of snake which shares their habitat, Lampropeltis alterna, so this hypothesis really is not as far fetched as it may seem.
Actually, juveniles look more like juvenile obsoleta than they do lepidus or alterna. Furthermore, many saxicolous snakes worldwide (Telescopus, Trimorphodon, C. lepidus, alterna, etc.) have this same pattern. I don't think the mimicry hypothesis is very parsimonious when you look at snakes as different as Telescopus and L. alterna.
Obviously, these two snakes are closely related and share a common ancestor. But are they related to the degree in the true sence of a biological species?
This may be worthy of investigation, but I think it is a little too early to draw a priori conclusions based on any of the data you have put forward here.
Since the Pantherophis bairdi holotype is a Texas phase specimen, I would like to propose Pantherophis angusticeps as an appropiate name for the Mexican phase.
Have you looked into the potential synonimies that already exist within the species? No other names have been applied to the Mexican pops?
The right to propose a name is reserved for those who actually do the work, not those who simply toss around a hypothesis.
My problem with your conjecture is that it is just that, conjecture. Don't get me wrong here, I salute your interest in this problem, but you seem to be drawing conclusions in the absence of data. Get the data, evaluate it properly, and then draw some conclusions.
I could make a similar arguement for eastern and western populations of Crotalus lepidus klauberi -
1. eastern and western pops look different
2. eastern and western pops are found in different habitat
3. eastern and western pops are isolated by inhospitable habitats
I don't think that is a just basis for taxonomic separation. It would take a lot more real data to justify this.
-----
Chris Harrison
