Reptile & Amphibian Forums

Can someone point me in the direction of a dendrogram (or something simular) that illistrates many of the popular species of snakes and their linage. Are there any free ones out there that show broad catagories such as pythons, boas, rats, etc down to individual species? I imagine this would be very big, but I was wondering what was out there.
Thanks
jesse
-----
1.0 Corn Snake
1.0 Ball Python

Some of the hypotheses of relationships among snakes can be found if you scroll down to the older messages in this forum. Other hypotheses of relationships are often buried within the pages of scientific journals, although sometimes these articles are made available by the authors on the world wide web. An important thing to remember is that these hypotheses are by no means established scientitic truth. They are subject to constant revision as scientists find better characters with which to analyze relationships.

Phylogenetic history cannot be directly observed, as George Gaylord Simpson pointed out, and must be inferred from the available evidence, some of which is not directly phylogenetic. For example, some scientists have attempted to infer phylogenetic relationships from dietary preference, even though diet is often strongly influenced by the availability of prey species and may differ geographically within a species' range. For now, there is no general consensus concerning phyletic relationships among animals. This is because the character evidence used by many scientists are often subject to convergence or parallelism. Scientific theories are only as good as the evidence that support them. For now, the quest for better character evidence will continue.

Molecular evidence tends to be superior to morphological characters in general, but even molecular data often disagree with each other. In the case of the origin of whales and the relationships among their relatives, the artiodactyls, the use of SINE characters have shown great promise. If similar sorts of characters can be found in snakes, then there is a strong possibility that a robust theory of snake relationships will emerge. For now, do not treat any single hypothesis or diagram of relationship as absolute truth, although there are agreements among scientists on some subsets of relationships. It is generally agreed that the lampropeltine genera such as Pituophis, Arizona, Bogertophis, Stilosoma, Cemophora and Lampropeltis are descended from a species of Elaphe which originated in the Old World. It is also generally agreed that the "advanced snakes" of the family Viperidae, Elapidae and Colubridae are all descended from a single common ancestor. Relationships that are more specific than these tend to be controversial as scientists are still attempting to work out the branching orders among the living snakes, often by using less than ideal characters.

Yes, I guess, I wasn't looking for a really specific diagram, but just something to look at that showed general linage between the major families of snakes. IE, the genetic distance between Boidae, Colubridae, Elapidae, Viperidae etc. and maybe some of these families well known species. Knowing that nothing is etch'ed in stone and specific species are debatable.
I just thought there might be somethign out there of such.
Thanks!
jesse
-----
1.0 Corn Snake
1.0 Ball Python

Here is a hypothesis of relationshp which I believe to be quite robust, although it too is subject to revision as evidence from better characters can be found.

The diagram depicted below was published in the article:

Philip J. Heise, Linda R. Maxson, Herndon G. Dowling, and S. Blair Hedges 1995. Higher-Level Snake Phylogeny Inferred from Mitochondrial DNA Sequences of 12s rRNA and 16s rRNA Genes. Mol. Biol. Evol. 12(2):259-265.

Wonderful! Exactly what I was looking for. Thanks!
jesse
-----
1.0 Corn Snake
1.0 Ball Python

One thing that strikes me as suprising is that Boiga is placed closer to Elaphe than Gonyosoma and much closer than Coluber. I probably should look at that paper to get the details.
John

I wouldn't worry to much about that paper; there are newer papers with more data and better, more sophisticated analyses. Check out Slowinski and Lawson in Molecular Phylogenetics and Evolution (2001 or 2002?) and a more recent paper in Systematic Biology (2003 but can't remember the author).

>>I wouldn't worry to much about that paper; there are newer papers with more data and better, more sophisticated analyses. Check out Slowinski and Lawson in Molecular Phylogenetics and Evolution (2001 or 2002?) and a more recent paper in Systematic Biology (2003 but can't remember the author).

The most recent papers on the subject are:

Slowinski, J. B. and R. Lawson. 2002. Snake phylogeny: evidence from nuclear and mitochondrial genes. Mol. Phylogenet. Evol. 24:194-202.

Vidal, N. and S. B. Hedges. 2002. Higher-level relationships of caenophidian snakes inferred from four nuclear and mitochondrial genes. C. R. Biologies 325:987–995.

Vidal, N. & S.B. Hedges. 2002. Higher-level relationships of snakes inferred from four nuclear and mitochondrial genes. C.R. Biologies 325: 977-985.

Kelly, C. M. R., N. P. Barker and M. H. Villet. 2003. Phylogenetics of advanced snakes (Caenophidia) based on four mitochondrial genes. Syst. Biol., 52:439-459.

And before out trolling but ignorant friend starts to whine about ignoring older evidence, let me point out several of these papers incorporate the exact sequencees used by Heise et al., but come to different conclusions by virtue of different methods of analysis and the inclusion of more taxa (the position of Atractaspis is one of them).

Cheers,

Wolfgang
-----
WW Home

Thanks to all for the references. I'll have to get my hands on the various papers when I have a chance. But something required to really evaluate the results is an understanding of the statistical analysis of the data. Typically this seems to consist of the use of a software package. Is this software readily available and is it expensive?

Regards
John

>>Thanks to all for the references. I'll have to get my hands on the various papers when I have a chance. But something required to really evaluate the results is an understanding of the statistical analysis of the data. Typically this seems to consist of the use of a software package. Is this software readily available and is it expensive?

It's not so much a matter of statistics as simply generating the tree, which is very memory consuming.

The software for that is widely available. Some is free (e.g., MEGA from www.megasoftware.net, primarily for distance analyses and Neighbour-Joining approaches; MrBayes for Bayesian analysis; Phylip for a whole bunch of things), some is available for more or less reasonable prices (PAUP* is probably the industry standard for many types of analysis).

Joe Felsenstein has a great page with links to all the different programs that are currently available (link below). The problemw itha lot of them is that (i) you need to know what you want to do before you start using them, and (ii) the documentation that comes with many is pretty sketchy, so learning to use them by yourself is pretty time-consuming and frustrating. The first thing you really need to do is to pick up a book on phylogenetics (most evolution and even general biology textbooks will have a chapter on the basics) and molecular phylogenies ot get a general grasp of the basic underying ideas.

Cheers,

Wolfgang
Joe Felsenstein's Phylogeny Programs Page

-----
WW Home

NT

can be found in the EMBL databse website:
EMBL database - snake phylogeny

-----
WW Home

Thanks for the internet addresses. I have linked this post to one of the trees found in that web page, and the image below is a tree from Vidal and Hedges. Interestingly, according to them, Calabaria reinhardtii is basal to a large number of boid snakes. This cladogram strongly conflicts with that of Kluge, which shows Calabaria clustering closely with Lichanura trivirgata and Charina bottae. There are several possibilities:

1. Calabaria is not closely related to Charina and Lichanura. It only resembles these two animals superficially. In other words, Kluge was fooled by convergent similarities between Calabaria, Charina and Lichanura.

2. Vidal and Hedges' characters and/or their analysis are flawed.

3. Both 1 and 2.

Unless Kluge is absolutely correct in his analysis, L. trivirgata, C. bottae and Calabaria probably cannot be maintained in the same genus without also including some or all of the species of boids in Vidal and Hedges' tree. In that case, the name of this genus is probably not Charina. If Boa constrictor is really part of the Lichanura-Charina-Calabaria clade, then the name for this clade would probably be Boa because of priority. Lichanura trivirgata would then become "Boa trivirgata" if Vidal and Hedges are correct and if one were to adhere to Hennigian taxonomic practice. Alternatively, one would have to rearrange the boids and erycines in wholesale fashion (similar to what Utiger et al. did to the ratsnakes) if one were to slavishly follow the dictates of Hennig, who required his followers to only recognize a taxon as one ancestor and all of its descendants. The taxonomic chaos generated by the cladists (which have been predicted by the opponents of the cladists) is upon us.

WW wrote:

"The most recent papers on the subject are:

Slowinski, J. B. and R. Lawson. 2002. Snake phylogeny: evidence from nuclear and mitochondrial genes. Mol. Phylogenet. Evol. 24:194-202.

Vidal, N. and S. B. Hedges. 2002. Higher-level relationships of caenophidian snakes inferred from four nuclear and mitochondrial genes. C. R. Biologies 325:987–995.

Vidal, N. & S.B. Hedges. 2002. Higher-level relationships of snakes inferred from four nuclear and mitochondrial genes. C.R. Biologies 325: 977-985.

Kelly, C. M. R., N. P. Barker and M. H. Villet. 2003. Phylogenetics of advanced snakes (Caenophidia) based on four mitochondrial genes. Syst. Biol., 52:439-459.

And before out trolling but ignorant friend starts to whine about ignoring older evidence, let me point out several of these papers incorporate the exact sequencees used by Heise et al., but come to different conclusions by virtue of different methods of analysis and the inclusion of more taxa (the position of Atractaspis is one of them).

Cheers,

Wolfgang"

Me:
WW seemed to have me in his crosshairs when he wrote his message. He seemed to realize that he in fact prefers newer data or newer analyses of old data and he attempted to obviate my criticism by his comments. Let me point out that it is often difficult to decide which one of two or more different contradictory hypotheses of relationships, if any, represents objective truth. That is because, once again, as George Gaylord Simpson pointed out, phylogenetic history cannot be directly observed and must be inferred. If a different method of analysis produces different results, it is probably because one or more of the following is true:

1) The characters are not any good.

2) The methods of analysis are not any good.

3) The authors picked the wrong tree from among many produced by the computer.

And as J.D. Lazell (1994 Herpetol. Rev.) said, "I question that making consensus cladograms out of X-1 certainly wrong trees and maybe one right one has much to do with science or reality." Therefore it is a bad idea to classify animals strictly according to the topography of even consensus cladograms, which have little to do with science or reality, as WW and his cladistic colleagues insist we must do. Classifications that are subject to the constant changes in the preferred hypotheses of branching order are useless for scientists, since the only useful classification is a reasonably stable one.

I have seen the trees in Vidal and Kelly and I am not impressed. They seem to require a lot of additional work to make them more believable. As to Slowinski and Lawson, their trees from the nuclear and cytochrome b genes also conflict strongly with one another and with Kluge's tree, which is derived from morphological characters. Specifically, in one tree Charina bottae and Lichanura trivirgata cluster with Boa constrictor, whereas in another C. bottae and L. trivirgata cluster with Exiliboa and Ungaliophis. Since Calabaria is not included in Slowinski and Lawson's analysis, there is no clue as to where Calabaria may lie in those trees. Slowinski and Lawson's two different trees can be used to produce vastly different classifications. Another author with a different set of trees and a newer publication date (the presumed tie-breaker) presumably can result in yet another completely different classification. For those people who like to play around with names, that is probably acceptable or even "cool." For scientists who need a stable taxonomy for information retrieval and effective communication, the constant stream of new phylogenies based on different characters or even different methods of analyzing the same old characters can be most annoying if classifications must be based strictly on branching order. An unstable taxonomy is nothing to cheer about for most scientists.

All of these papers that WW cited give us good reason to ignore Kluge's lumping of Charina bottae, Lichanura trivirgata and Calabaria reinhardtii into the same genus, if the morphological disparity among these taxa have not provided good enough reason before. Kluge's tree is contradicted by Slowinski and Lawson's trees and any classification of the dwarf boas on the basis of a strict application of Hennigian taxonomy is premature.

the specifics of cladistics. For a first approximation I compare it to the textual analysis that classical scholars use to establish the lineages of ancient texts. They analyze them for identifiable descent from older texts. They even have the equivalents to geographic isolation and subsequent remixing that might occur with climate and geologic changes to biological species.

For biology I assume there is a much less subjective methodology than in textual analysis. With texts it is as much art as science. But in both cases I imagine that experience and individual skill have a strong bearing on the lasting or ephemeral nature of the individual's impact on the field.