Snake Classification

Cal, From what I can tell from the computer-garbled cladogram you posted (in the Dino-Bird thread), the Eurasian and American Elaphe species should be in different genera. Below I am assuming the type species is Eurasian, thus restricting the genus Elaphe to clade 3. If this hasn't been done yet, I assume the cladogram you gave is somewhat controversial. Lampropeltis is clearly a paraphyletic genus, and I would suggest Stilosoma and Cemophora probably should be subgenera of Lampropeltis. I could code them as full genera, but from what I can tell from the cladogram, I don't know why this would be necessary. This way Lampropeltis would only be paraphyletic with respect to one exgroup (Arizona, Pituophis, and the American 'Elaphe' species). Anyway, from what I can make out from your cladogram, I would code these genera as follows (main clades numbered in the order that they split off, and subclades are lettered):

1 Heterodon B Diadophis 2 Gonyosoma B Drymarchon C Masticophis D Coluber 3 Elaphe (Eurasian forms) 4 Lampropeltis (sensu lato)*** _1_ Arizona 2 Pituophis 3 Elaphe (part)++

I suggest that you get a hold of at least one of these papers and try again. Hate to criticize someone's classification when he cannot see the actual topology of the cladogram.

Cal, I would be interested to know where one can connect Maxson. But it seems pretty clear to me that his phylogeny must be controversial, or Elaphe would have almost certainly have been split into two or more genera by now. According to the phylogeny you gave, it is either polyphyletic, or so paraphyletic as to be as useless as Condylarthra. If there is some good reason to maintain such extreme paraphyly, I would consider it, but I have a feeling there may be an unfortunate clinging to tradition working here. Reptilia is one thing, but Elaphe looks to be even less useful and more paraphyletic (more so than even I will accept). But perhaps Schulz and others do not think Elaphe is as paraphyletic as Dowling and Maxson's phylogenies would indicate?

Linda Maxson is dean of Liberal Arts and Sciences at the University of

various reptiles was at Penn State.

: But it seems pretty clear to me that his phylogeny must be : controversial, or Elaphe would have almost certainly have been split : into two or more genera by now. According to the phylogeny you gave, it : is either polyphyletic, or so paraphyletic as to be as useless as : Condylarthra.

: If there is some good reason to maintain such extreme paraphyly, I : would consider it, but I have a feeling there may be an unfortunate : clinging to tradition working here. Reptilia is one thing, but Elaphe : looks to be even less useful and more paraphyletic (more so than even I : will accept). But perhaps Schulz and others do not think Elaphe is as : paraphyletic as Dowling and Maxson's phylogenies would indicate?

The methodology used by Dowling and Dowling and Maxson may be part of the

Sean, Thanks for the information. I'm trying to get a copy of Lopez and Maxson. However, they apparently don't include Bogertophis species and other subgenera/genera in this complex, so not sure a good classification is possible based only on their work. Since a broader genus Elaphe is still being recognized, I assumed that their results must be pretty controversial. Even traditional eclectic classifications cannot properly handle situations where there is a lot of controversy. However, once I have several different viewpoints on these snakes, I bet I can come up with fairly decent classification that is neither too eclectically vague, nor cladistically oversplit. By the way, can someone tell me what is the type species of Elaphe??

No, the North American species of Elaphe definitely form a clade. It is not polyphyletic. According to allozyme data obtained in the early 1980's, all of the species of N. American Elaphe and their closest relatives, such as Lampropeltis, are descended from a single species of Elaphe of Old World origin. And no, paraphyletic taxa are not useless. The degree of paraphyly is a reflection of the different rates of morphological evolution within and among lineages. The more disparate the rates of evolution are, the greater the degree of paraphyly. Darwin himself is cognizant of the fact that evolutionary rates differ, and he has proposed that classification should take into account these differing evolutionary rates by placing morphologically disparate species in new genera, families or other higher taxa. Darwin therefore does not have any problem recognizing 'paraphyletic' taxa. Neither do Darwinian taxonomists such as Ernst Mayr and G. G. Simpson et al. and the overwhelming majority of the greatest herpetologists past and present.

Absolutely. Paraphyly is the 'inevitable result of the process of evolution' (Carroll 1988, Vertebrate Paleontology and Evolution, p.13). If there is no such thing as evolution, then there would be no such thing as paraphyletic taxa. A species that never gives rise to another species will never become paraphyletic, and the same is true of any higher taxon. For example, if birds and mammals never evolved, then Reptilia would not be paraphyletic. Recognizing paraphyletic taxa means recognizing the 'inevitable result of the process of evolution'. That is why the school of taxonomy that recognizes paraphyletic taxa is known as 'evolutionary taxonomy.' That is why the cladists, who do not recognize paraphyletic taxa, have either abandoned evolutionary models or they have been accused of having abandoned evolutionary models, according to the cladist Gareth Nelson.

Chopping up, say, North American Elaphe into numerous genera in order to eliminate paraphyly would indeed be useless. Dowling and his colleagues have already removed several morphologically distinct species of North American Elaphe and placed them in new genera such as Bogertophis and Senticolis, but there are some who contend that even these species may not be distinct enough to merit generic recognition! The remaining species of Elaphe therefore are quite homogeneous morphologically. Since they do not form a polyphyletic group, there is really no scientific justification to split them into different genera.

The responses you have given demonstrates that the Kinman system, because of its intolerance of paraphyly, is an unsuitable replacement for evolutionary taxonomy, which has proven useful for biologists of all disciplines for centuries. Splitting the currently recognized North American species of Elaphe into two or more genera is useless and scientifically untenable. Equally untenable is the cladistic alternative of lumping the descendant genera of Elaphe, namely Lampropeltis, Pituophis, Cemophora, Stilosoma, Arizona and Rhinocheilus back into Elaphe, because the resultant taxon would be too heterogeneous to be useful. It is noteworthy that the cladistic system provides only two possible solutions for the classification of Elaphe and its closest relatives, and these two solutions occupy opposite extremes in a spectrum of possible classifications: either one all encompassing Elaphe for all North American ratsnakes and their descendants (including Stilosoma, Pituophis, Rhinocheilus, Arizona and Cemophora) or splitting the currently recognized species of North American Elaphe into several different genera.

Sean, Thanks for the information. Very interesting. If Bogertophis was proposed that long ago, is the cladogram that Cal King presented just

Cal, Just saw your last post. The Kinman System is not intolerant of paraphyletic groups. I am perfectly willing to recognize a paraphyletic Lampropeltis giving rise to a single genus of North American species (including Arizona, Pituophis, North America Elaphe, and

Boertophis

There is a kernel of truth in that statement. Due to the popularity of cladism, and due to the cladists' disdain for distance data, many systematists have drifted away from distance techniques. But things have not always been this way. For example, evolutionary biologist S. J. Gould writes in his book 'Dinosaur in a haystack:'

'Buss and colleagues sequenced part of an important gene that codes for ribosomal RNA, and they found 108 'phylogenetically informative' positions-an enormous increase in the number of useful characters for classification. They developed a matrix of similarities for all pairs of comparisons among twelve species of hermit and king crabs and a thirteenth more distant relative, chosen to anchor the tree (and called an 'outgroup' in our jargon; they used the brine shrimp Artemia salina). They then applied a variety of standard tree-building techniques to this matrix of relative similarities. They achieved the same basic result with either of the two most common methods for tree-building: distance analysis, which works only with measured degrees of overall similarity; and parsimony, which constructs trees with a minimal number of evolutionary steps.'

In fact, the other method that was commonly used, namely parsimony analysis, has lost some popularity. Many systematists are starting to use maximum likelihood techniques.

Nevertheless, immunological techniques have served science very well and it would be a naïve mistake to simply ignore immunological data because they have been supplanted by more precise molecular techniques. In fact, many of the findings using immunological techniques have been corroborated by more precise molecular methodologies and of course by morphological analysis.

The Lopez and Maxson paper I cited uses mtDNA data, not immunological techniques, so your apparent prejudice against immunological techniques, whether justified or not, is irrelevant.

Dowling et al.'s data indicate that the common ancestor of the North American ratsnake clade (which includes Elaphe, Lampropeltis, Rhinocheilus, Stilosoma, Cemophora, Pituophis and Arizona) arrived in the New World some time in the Miocene, which the evolutionary history of the N. American ratsnake clade is well within the 5-50 million years time frame. So once again, there does not seem to be any justification for ignoring Dowling et al.'s or Lopez and Maxson's data.

Maxson is aware of this sort of attacks on her work and she has rebutted them in print. Once again, it is also irrelevant because Lopez and Maxson's tree is constructed using mtDNA data. Further, their mtDNA data corroborates Dowling and Maxson's immunological data. Does it mean that mtDNA data is also unreliable? Dowling and Maxson's trees show that Elaphe is closely related to Lampropeltis. Is that unreliable? After all, who among the herpetological community would deny such a close relationship, which had long ago been inferred from morphological data? Unfortunately, Mr. Barry fails to point out just which part of the tree he finds objectionable, because I do not see how he can possibly dispute, for example, the close relationship between Elaphe and Lampropeltis, and the close relationship between Elaphe obsoleta and E. guttata, which are indeed part of Dowling and Maxson's findings!

If Mr. Barry thinks that the high degree of paraphyly in Dowling and Maxson's trees is a good reason for doubting the validity of their data, then he is simply imposing his own assumptions and believes on scientific data, because such a high degree of paraphyly is not at all unusual. See, for example, the cladograms constructed using parsimony analysis and mtDNA data in Bickham et al.,1996, Molecular Systematics of the genus Clemmys and the intergeneric relationships of Emydid Turtles. Herpetologica 52(1):89-97. Their analysis of emydid turtle relationships did not utilize supposedly unreliable immunological techniques. Nevertheless it still shows a high degree of paraphyly.

Yet another case of hidden paraphyly revealed by molecular techniques was narrated by S. J. Gould in his book 'Dinosaur in a haystack.' Using mtDNA data, Buss and colleagues find that the king crabs were nested within a single genus of hermit crabs. Again, no immunological technique was used, but paraphyly was revealed nonetheless.

In fact, Dowling and Maxson's tree showing Cemophora and Stilosoma nesting deeply within the genus Lampropeltis is not even remotely as remarkable as the king crabs being nested within the single hermit crab genus Pagurus.

accepted.

I am not quite sure how Mr. Barry arrived at this particular conclusion. Dowling and Maxson are not cladists. They have not proposed a radical reclassification of snakes based on the fact that many of the taxa in their analysis, including Lampropeltis, Coluber and Elaphe, are paraphyletic. In fact, they explicitly state that biochemical paraphyly does not concern them. Therefore one cannot simply tell whether their phylogeny is widely accepted by judging how many herpetologists have adopted their taxonomic proposal, since they did not make any. Only an opinion poll can tell us whether many herpetologists are even aware of their findings but I know of no such poll. Therefore Mr. Barry is merely using anecdotal information. I myself have quite different experiences than Mr. Barry, since I personally know of some herpetologists who specialize in snake systematics but who are unaware of Dowling and Maxson's data. Hence it is unlikely that many herpetologists who are not even specialists in systematics would have any informed opinion of Dowling and Maxson's findings.

Dowling and Maxson did not propose a revision of the genus Elaphe based solely on their molecular data, and neither did Lopez and Maxson. Dowling and colleagues (without Maxson or Lopez), however, did indeed revise the genus Elaphe by transferring some species of Elaphe to the new genera Senticolis and Bogertophis. They did so using a wide variety of data, including scale counts, scale morphology, biogeography, hemipenial morphology and molecular distance data, unlike some systematists, who for example split Rana pipiens and Dicamptodon ensatus largely, if not solely, on the basis of molecular distance data. Whether Schulz accepts Bogertophis or Senticolis as valid taxa or not is therefore not an indication as to whether Schulz accepts the validity of Dowling and Maxson's phylogeny based largely on immunological data.

: indicates that immunological techniques are inappropriate or only : marginally appropriate for most phylogenetic analyses (and considered : 'appropriate' only in the 5-50 mya time frame), and that isozyme : analysis, restriction analysis, and direct sequencing of genomic or : mitochondrial DNA are far more informative over a broader range of time : scales and for other types of molecular-based studies.

Oh, I meant to add that the Maxsons authored the chapter on immunological techniques in the Hillis and Moritz book, which is pretty much the standard reference for technique applications and interpretation in molecular systematics (it's even used by traditional Darwinian sytematists). I don't think that the editors, Hillis and Moritz, intended to denigrate the utility of immunological techniques in systematics, just to put them in proper context with other available and apparently more broadly applicable techniques.

Sean Barry

Elaphe is the paraphyletic genus that has given rise to the other genera, including Lampropeltis. It is 'massively paraphyletic.' If the Kinman system believes that such 'massively paraphyletic' taxa is useless, then it is indeed intolerant of paraphyletic taxa. Of course you can prove me wrong by coming up with a classification that will recognize Elaphe as a valid genus even though it is 'massively