Q: How do you arrive at a family tree of reptiles?
1. You can look at hundreds of taxa and score them for hundreds of characters and let maximum parsimony recover the family tree, as I did here. This is following in the paths of dozens of others who have created their own subsets of the reptile family tree with their own taxon and character lists. Because morphology is an expression of genetics, morphology is really the only genetics that counts. Morphology helps you survive in your niche and enhances secondary sexual characteristics to help you get laid to create replicants of your own naturally selected self.
or
2. You can look at the molecular DNA of several distinct specimens (all currently living) and let maximum parsimony recover a much more unfocused family tree leaving out hundreds of extinct taxa with no known living counterparts. Of course some of this DNA doesn’t express anything. Some of it jumps around the chromosomes. And DNA cannot be recovered from the vast majority of extinct taxa, so we’re left with huge ghost lineages, during which “things” happen (see below). Even the sex chromosomes are different between birds, mammals and lizards.
Male mammals have the XY combination while females have XX. Male birds carry what’s known as the ZZ pair and females have the ZW pair. The green anole X chromosome is a microchromosome. Yes, it’s nothing compared to having a skull, tail, four feet, five toes and relative constants like those when it comes to DNA.
BTW I freely admit to not knowing much about molecular studies except that they sometimes don’t replicate morph studies.
The DNA problems don’t end there…
We’re talking about reptiles here: DNA study results do not replicate morphological study results. Often DNA study results do not replicate other DNA study results. DNA is changing even when morphology does not (Hays 2008). This alone could be the reason for the discrepancy in DNA and morphology among reptiles (and not among mammals and humans involved in paternity suits).
Is this why reptile DNA studies don’t match morph studies?
A recent article on genetic sequencing of an anole lizard DNA reports, “We’ve now sequenced a lizard genome for the first time ever. The anoles shed light on non-coding sequences of genes. What they might be are the husks of special DNA sequences known as transposons. These can only be described as “jumping DNA”, able to actually move through the genomes and copy and paste themselves elsewhere. Transposons can give any genome that carries them great agility and resilience in dealing with unexpected environmental challenges.”
Lizard egg proteins
From the same article, “It appears that, as far as egg genes are concerned, reptiles are in a constant state of evolutionary flux, with the proteins revealing clear signs of rapid evolutionary change.”
Figure 1. The Tuatara (Sphenodon)
The Hays et al. 2008 abstract
“The tuatara of New Zealand is a unique reptile that coexisted with dinosaurs and has changed little morphologically from its Cretaceous relatives. Tuatara have very slow metabolic and growth rates, long generation times and slow rates of reproduction. This suggests that the species is likely to exhibit a very slow rate of molecular evolution. Our analysis of ancient and modern tuatara DNA shows that, surprisingly, tuatara have the highest rate of molecular change recorded in vertebrates. Our work also suggests that rates of neutral molecular and phenotypic evolution are decoupled.”
I take this to mean DNA genetic evolution is decoupled from morphological genetic evolution in the Lepidosauria (but not necessarily decoupled in other living things). With that hanging over our collective decision making processes, maybe morphological genetic studies should trump DNA genetic studies in non-mammalian reptiles.
Reptile DNA studies might be interesting, but let’s not hang our hats on them. Let’s stick with fossils and phylogenetic analysis. That covers all the bases down to the specimen.
As always, I encourage readers to see specimens, make observations and come to your own conclusions. Test. Test. And test again.
Evidence and support in the form of nexus, pdf and jpeg files will be sent to all who request additional data.
References
Hay JM, Subramanian S, Millar CD and Mohandesan E 2008. Rapid molecular evolution in a living fossil. Trends in Genetics, 24(3):106-109.
The Pterosaur Heresies 
Thanks for writing this in response to my questions, but I don’t think you really get a lot of important ideas.
“Because morphology is an expression of genetics, morphology is really the only genetics that counts.”
First, there is a LOT more to morphology than just the skeletal features we score for paleo-based reptile analyses. Soft tissue, cell structure, etc.. And these are all based on genes too.
Second, what “counts” in a phylogenetic analysis are any genetic changes passed on to descendants. These include phenotypic features that help survival, yes. But they also include neutral changes that get passed down and often don’t influence survival, like retroposons. And in fact, tracking changes that are neutral is better, since they won’t be affected when an animal changes ecology.
“And DNA cannot be recovered from the vast majority of extinct taxa, so we’re left with huge ghost lineages”
I’ve told you twice now that this doesn’t matter for retroposons. Retroposons have no known mechanism for reversal and are highly unlikely to be inserted at the same point in two genomes convergently. So no reversals or convergences can happen in fossil taxa’s retroposons either, thus extinct taxa couldn’t disrupt a retroposon-based tree, even in principle.
“Often DNA study results do not replicate other DNA study results.”
I’ve also told you you can’t use this as a talking point. Well supported (Baysian liklihood, etc.) nodes in one molecular study are almost never contradicted by well supported nodes in another study. If you have specific exceptions in mind, please state them. Certainly this happens far less than in morphological analyses.
“DNA is changing even when morphology does not (Hays 2008). This alone could be the reason for the discrepancy in DNA and morphology among reptiles”
Exactly! Or at least when skeletal morphology isn’t changing. A retroposon inserting into a certain position in the genome will often have no effect on morphology, for instance. But it still gets passed down the reptile’s lineage. But this isn’t a problem at all. The point of phylogenetic analyses is to examine the distribution of inherited characters, and what’s actually inherited are nucleotide positions. As you said above, some of these contribute to skeletal anatomy. But most don’t, and those are at least as good for determining phylogeny.
I don’t see the point of your paragraph quoting about anole transposons. As I’ve been saying, transposons (which include retroposons) are great for determining phylogeny, and contradict your morphological tree (and Gauthier et al.’s, and Conrad’s…).
““It appears that, as far as egg genes are concerned, reptiles are in a constant state of evolutionary flux, with the proteins revealing clear signs of rapid evolutionary change.””
Yes, some genes are poor for phylogenetic analyses since they evolve too quickly. But molecular phylogeneticists know this and can check rates of change (you might have seen tests for noise saturation in papers). So this doesn’t affect genes they do use, like RAG1 and C-mos. Similarly, us morphological phylogeneticists are smart enough not to try a reptile analysis based on color.
“I take this to mean DNA genetic evolution is decoupled from morphological genetic evolution in the Lepidosauria (but not necessarily decoupled in other living things). With that hanging over our collective decision making processes, maybe morphological genetic studies should trump DNA genetic studies in non-mammalian reptiles.”
Ah, but why would decoupling mean morphology should trump molecules? After all, it’s the DNA that is actually changing and being inherited, with some of that translating into the morphological changes, even if most of it is having no morphological effect. There isn’t some separate non-molecular source of inherited morphological change.
“Reptile DNA studies might be interesting, but let’s not hang our hats on them. Let’s stick with fossils and phylogenetic analysis.”
Not only have you not made that case at all, you haven’t answered my two questions that inspired this post-
1. Molecular scientists say retroposons have no known mechanism for being reversed and are basically impossible to converge. How do you account for your phylogenies requiring retroposon homoplasy?
2. Different genes get us phylogenies which disagree with yours in the same ways, and agree with retroposons too. Even if these genes were “in constant flux” or decoupling somehow made them less reliable, why would the different genes and retroposons give the SAME wrong phylogeny? Surely if they weren’t reliable, nuclear genes would give well supported nodes different from those found using mitochondrial genes, and neither would match retroposons. And yet instead, the phylogenies match. What mechanism could cause that?
Thanks Mickey, and I appreciate your position. I’m no molecule expert, but when molecules create “strange bedfellows” that share few to no traits, then you’ve got to figuring out what went wrong. Ultimately your answer will have to be enlightening, showing us where to find gradual changes reflecting evolution as it happened. Where molecules do well, in mammals, we get gradual changes, as would be expected in evolution. But not always. We still need fossils and morphological trees to confirm the insights molecules give us. More taxa, even those made of stone, is still the answer. If you want to hang on to molecules for reptiles, then you’ll have some ‘splaining to do. I will let the reconstructions do the talking and add details in the captions.
I have ‘splained how morphology can go wrong- convergence and reversals caused by similar adaptive needs. We know how often this happens. Just look at Asher et al.’s (2003) mammal morph analysis. Ungulata, pangolins with anteaters, bats with colugos…
What you haven’t ‘splained is how multiple genes can all go wrong in the same way, and how this can match retroposons and how the retroposons are wrong when molecular biologists tell us they almost can’t be. I really want to know your answers since by now it looks like you’re dodging the questions.
Not to be insulting, but the situation reminds me of how Young Earth Creationists doubt dating methods. Sure tree rings (nuclear DNA) could be wrong. Sure ice cores (mitochondrial DNA) could be wrong. And it’s technically possible radiocarbon dating (retroposons) could be wrong, even though physicists (molecular biologists) tell us it’s basically impossible. But what is the liklihood of all of these methods being wrong in the same way so that their results match each other? That’s the problem you’re faced with, and ignoring the question or claiming your method (morphological analyses / Bible) trumps all the other methods without valid reasoning doesn’t make you look good to scientists.
Mickey, I appreciate and understand your comments. I won’t be blogging on molecules any more because I haven’t studied them. This is all I know. I don’t understand how molecules don’t replicate morphology except that they don’t. That we agree on. That’s where the large reptile tree comes in.
You mentioned reversals and convergence, but you haven’t pinpointed where those reversals and convergences are in the large reptile tree. If you could, that would be go a long way toward helping me revise the tree, if valid. I’m eager to hear it. I don’t doubt the results of molecules. So, how can we reconcile them with morphological results? They have to reconcile at some point because there is only one tree, not one for morphology and one for molecules. Let’s find it together.
Young Earth Creationists find every excuse for why evolution did not happen. I’m just the opposite. I’m looking for every clue to the path evolution took. That’s why I built the large reptile tree out of specimens. I didn’t preplan it. I didn’t shift branches to match established paradigms, whether molecular results or traditional results. I simply let the results speak for themselves.
Your input is important. Criticisms bring out problems that need to be resolved. Please continue, but let’s turn this into a construction project, not a demolition.
PS. If you see any points of reversal or convergence that are not currently represented on the large reptile tree, please feel free to point those out too so that by moving the branches we can arrive at a new tree that matches molecular results. The large reptile tree enables all sorts of testing. So let’s do it.
You don’t doubt the results of molecules, yet you consistantly say your morphological results which contradict them are correct? You’re not going to blog on DNA anymore, yet you are looking for every clue to phylogeny? You seem to contradict yourself.
It’s indeed very clear you haven’t studied DNA phylogenetics, as seen in my original comments here. My advice would be to learn about it if you’re going to make statements about the phylogeny of living taxa. If you continue claiming your morphological results are correct and fail to mention the grave problem that they contradict multiple independent sources of molecular evidence, it doesn’t reflect well on you to other scientists. I know how I would feel about a paper claiming Ungulata was real without reflecting on all of the evidence for Afrotheria. Even Gauthier et al. (2012) had an extensive discussion of reasons morphological and molecular phylogenies for squamates mismatch so badly, though I don’t think they did much of a better job than you and basically end up saying they think the morph tree will win out in the end. Yet I can’t think of a single time where well supported molecular results from several sources were overturned in the scientific consensus for morphological results. You and Gauthier are on the losing end of history.
As for where convergence is present in your tree, I’ve pointed out aquatic habits make your Enaliosauria plausibly caused by convergence. Or since Lanthanotus (not Lanthanotes as your site always says) and Heloderma both burrow they are probably converging on burrowing snakes in your phylogeny. Or how the arboreal features of bats, colugos, tree shrews and primates are probably helping them clade together. As you say, your supposedly related animals more often than not have similar body shapes caused by similar ecologies but differ in details (the minutia you haven’t studied like braincase foramina), and this should make us very suspicious.
I see you are still not able or not willing to break the large reptile tree to fit your molecules. Please be specific as I have been in the large reptile tree. Your convergence hypothesis is just so nebulous. It’s not just the body shapes, but every detail of their anatomy down to the teeth and toes that the tree is built on. Your pronouncements about the losing end of history are political not scientific. And your timing on Lanthanotus is perfect as I just finished fixing that typo throughout the site.
Pingback: Researching dragons | midnightwritersblock