How to make evolution look bad

About a year ago
Darren Naish, at his Tetrapod Zoology blog, launched an all out (as opposed to specific pinpoint) attack on As you’ll recall, and beyond all logic, he peppered his attack with monstrous renderings from other artists. In the same vein, Naish also included his own version of the large reptile tree (while virtually ignoring the real one, Fig. 2). His self-titled “highly simplified version of the David Peters tetrapod tree” version (Fig. 1) included only 14 taxa. And over half of those were new (not included in the original large reptile tree taxon list of 325.)

So once again,
rather than using real data and real evidence from the website, Darren Naish made up his own tree (Fig.1) and taxa. And by labeling it a “version of the David Peters tree,” Naish set it up (passively in this case) to be ridiculed. Here’s how: Covering such a wide gamut with so few taxa gives the impression that none of these taxa are closely related to one another, or evolve from one another. Which is true! By omitting hundreds of transitional taxa (nodes shown in pink, Fig. 1), Naish’s tree becomes a misrepresentation of the large, robust and fully resolved tree that depends on those hundreds of transitional taxa to show close relationships and evolutionary pathways.

Making up your own evidence and planting it on someone else is typically inadmissible in court. Withholding evidence (omitting hundreds of transitional taxa, ) is likewise frowned upon. Yet Naish still feels justified and proud of his invented journalism.

from Darren Naish, 2012, his take on the large reptile tree, illustrated with seven taxa that are not included in the large reptile tree. The pink numbers represent the number of nodes Naish omitted between nodes he included. None on these taxa look like they are related to one another because all the gradual changes between them have been wiped out.

Figure 1. Click to enlarge. From Darren Naish, 2012, his take on the large reptile tree, illustrated with eight taxa (more than half) that are not included in the large reptile tree. The pink numbers represent the number of nodes from the large reptile tree Naish omitted between his own highly simplified nodes. None on these taxa look like they are related to one another because all the gradual changes between them have been omitted. Key: If a + sign follows a number, the number refers to nodes between basal taxa (eg. basal crocs and basal dinos). If the + sign follows a number it refers to an unknown number of taxa leading to the derived taxon Naish employs to represent the clade.

The failure of prior trees, including Naish’s tree
is in their small size. Too few taxa (14 in Naish’s case) cannot hope to relay the wonders and blends and relationships within the wide gamut of reptile evolution. You need more taxa (at least a magnitude more). And in Naish’s case, you need more basal taxa (no Iguanodon, please).

Naish chose to portray (Fig.1) a derived Cretaceous pterosaur rather than a basal Triassic one. He chose to portray a derived sauropterygian, ichthyosaur, dinosaur, croc, mammal, turtle and frog, none of which are found in the large reptile tree. So there’s massive misrepresentation here. This is not the Peters taxon list. He wanted the taxa to look unrelated, and he succeeded. He wanted to make the reptile evolution tree look bad.

The large reptile family tree.

Figure 2. Click to enlarge. The large reptile family tree. 34o in the main tree, another 32 or so in the therapsid tree. 

If only Naish had listened to the good angel on his other shoulder…
Naish could have shown the actual reptile evolution tree, or segments of it, but he didn’t.  He could have focused on a recovered clade and demonstrated problems within it, but he didn’t.

(Notice, I am using Naish’s own tree in my attack here. It wasn’t that difficult to do.)

As an analogy, Naish’s tree gave us the colors blue, red and yellow without including the gradual spectrum of color transitions between them — which is the whole point of any study of evolution, right?

On a sarcastic note
Using Naish’s “highly simplified” version of evolution (Fig. 1), let’s also show the evolution of humans starting with a sponge, a fly, a starfish, a lamprey, a swordfish, Eryops, a snake, Dimetrodon, Burnetia, a platypus, an elephant, a cat, a rabbit and a tarsier. That’s the same number of taxa that Naish used and in the correct phylogenetic order. But the evolutionary transitions are just too difficult to understand because basal taxa are not used.

This is how Darren Naish made evolution look bad.

By contrast —

This is how to make evolution look good.
If you are interested in any genus listed on the large reptile tree, you can find it at Look it up and you can link to any number of predecessors and descendants (if there are any), relatives and offshoots. You’ll see that related taxa share a very large suite of traits. Unrelated taxa don’t share as many traits. You’ll be able to read what traits are new and what traits have become vestiges. Sadly, such rich details are missing from Naish’s “highly simplified” tree. In large gamut evolutionary studies 14 taxa cannot provide the same value as 340. To “highly simplify” evolution is to misrepresent it and make it unintelligible and laughable. Naish’s version of simplification has the same effect as removing 20 or more letters from the alphabet.

The power and value of
is in the number of included taxa, each one a slight variation from its closest relative. The whole point is to get closer to these transitions so you can begin to understand the natural selection trends and the gradual evolutionary processes that occurred back then. Here you’ll get higher resolution by removing the evolutionary distances between related reptiles. And you do this by increasing the number of included taxa.

More is better!!!!

You just can’t get such high resolution and increased understanding from a mere 14 taxa in a “highly simplified” tree.

A year later Naish still doesn’t understand his error. He thinks he did the world a great favor by exposing the “sham” behind Hopefully someone someday will point out to him that, in this case, his over “simplification” was the real sham.

23 thoughts on “How to make evolution look bad

  1. Very interesting: this post clearly demonstrates that you have no idea of what a cladogram means. The branching pattern shown in Naish simplified diagram is the same as in your “reptile tree”. Do you understand that a cladogram is always the simplified representation of a branching pattern, not a literal depiction of the evolutionary process? When you compare the two cladograms, what matters is the lineage pattern, not the single species used as “example” of each lineage.

    Do you know what “phylogenetic relationships” means? Based on your own words: no.
    Do you understand what “simplified” means? Based on your own comments to Naish diagram: no.

    For example, is it true that your “reptile tree” places pterosaurs phylogenetically closer to squamates than dinosaurs? Yes, it does. And is it true that Naish diagram shows the same pattern? Yes, it does. And the same happens with any other lineage depicted in Naish tree, because it was based on Peters tree.
    The taxa depicted in Naish tree are just examples of species belonging to the main lineages shown in the simplified tree. But they are not relevant to Naish argument: what matters is the branching pattern among the lineages, not the species/genera illustrated. The Naish tree pattern is identical to the main pattern resulted in your tree.
    In short, Naish tree is your own tree.

    How could you reply to Naish if you are unable to understand what he wrote one year ago?

    • Andrea, you misunderstand the whole point. I understand that Naish’s tree topology was essentially identical to my own, reduced from 300+ taxa to 14. To reduce it, however, is to remove its value. I show gradual accumulation and removal of traits. Naish’s tree does not. To nest mammals with plesiosaurs and Iguanodon invites ridicule, which was Naish’s motivation. Details matter.

      Darren’s reply was trashed as it contained no specific fixes, instead was another blackwash.

      • You’re failing to understand A) the point that everyone is trying to make, and B) the point of Dr Naish’s summary tree. I will try and explain a bit less angrily than everyone In the hope you get it.

        Naish did NOT remove taxa or add in taxa which you never tested. He did what is called collapsing nodes. This is a perfectly legitimate practice when creating a summary tree. It is, I repeat, not removing taxa, it is collecting the taxa which are suggested by your own tree to form a monophyletic grouping and displaying only the name of that grouping. The important point is that as long as all the taxa within the node form a monophyletic grouping, collapsing neither changes the information in the tree or constitutes the removal of taxa. Nor does collapsing nodes change the time of splitting illustrated (see above).

        Your biggest misunderstanding is that you think Naish’s tree shows something different to yours. Remember what a cladogram (for that is what you are really making,) is: NOT a depiction of the evolutionary series, but an illustration of the relative time of the splitting of lineages (hypothesised). Also remember , it is not possible to test for ancestral relationships using the methods you have used, or for ancestral character states (the former is extremely debatable whether it is even possible, the latter requires extra analyses which you do not do). So to accuse Naish of not showing the evolutionary sequence you show is giving yourself credit not earned. You do not show evolutionary sequences; such is not possible for a cladogram.

        To actually say Naish mis-represented your phylogeny, the summary tree would have to suggest a different ORDER of splitting of the collapsed lineages to the large reptile tree. That is, if you collapsed the same nodes as Naish did, you would get a tree suggesting a different order of splitting. You would have to be able to say “my tree suggests that clade A is more closely related to clade B than clade C, whereas the Naish tree suggests it is closer to C.” Find me a single example in the Naish tree where you can say that. The clause “than clade C” is the key bit: again I say you are not showing an evolutionary sequence, but a relative order of splitting of branches

        Naish could have displayed your large reptile tree, but what would be the point? Remember, he was writing for a lay audience, many of whom would not be familiar with the taxa you have tested. If they were, they can go for your blog. By collapsing nodes, Naish was able to show the more familiar names that people would recognise, WITHOUT changing the order of lineage splitting suggested by your tree (and therefore not changing the info at all). The pictures were also for this point: not to illustrate the evolutionary series, but to illustrate what each clade looks like for a wider audience. Yes, maybe a basal synapsid would be more scientifically accurate than a badger, but frankly who cares? its an example, nothing more. I say once again: neither your tree, nor Naish’s, suggests that the taxa illustrated evolved into the others. Putting a badger is not intended to suggest that badgers evolved into anything: its a sample synapsid.

        I hope you understand what I have said. The fact that you didn’t understand it before you even started your tree, let alone after however many years of phylogenetic work, demonstrates your lack of research and education in the subject of phylogenetics. You seem to think its an easy subject, yet every post speaks of your lack of understanding (I still laugh inside when I remember you writing that phylogenetics was “guaranteed to give the correct answer”). Do you not agree that learning what a phylogeny is and what it shows should go before making your own and telling everyone else they are wrong? This post shows you have not done that

      • To put it simply, I understand what you’re saying. You need to understand that I’m not writing for a lay audience. You also need to understand the negative context in which Darren’s (my) tree was presented. To your point, every cladogram shows the relative order of branch splitting, but it also sheds light on evolutionary sequences, so I disagree with you there. If not in a phylogenetic tree, how does one present evolutionary sequences with data to support it? I hope you can provide an answer. Otherwise you’ve painted yourself into a corner.

      • I do get that it was presented in a negative context, but that does not make the method he used wrong. he showed the same relationships you did, using a perfectly reasonable method to summarise. The criticisms of missing taxa and changing relationships are unfounded

        As for how you tell evolutionary sequences it depends what you mean. If you want ancestor descendant relationships, a lot of people would say you can’t. There are some suggested ways, all of which come with extremely heavy caveats (e.g. taxa which appear before their sisters and have no autapomorphies may be inferred to be ancestors). But most people say we don;;t have the methods to do it with confidence.

        If you mean evolutionary sequence of characters, that actually is possible, although again comes with caveats. There are programs like mesquite and TNT which will do it for you. In effect you choose a model of character state transformation (the two common ones are DELTRAN and ACCTRAN; minimising reversals and minimising homoplasies respectively) and the program will work it out with parsimony. There are BAyesian and likelihood models as well, which are designed for molecular work but have been applied to morphological trees

        So yes a cladogram can give a clue about evolutionary sequences, but it does NOT give the sequence. You can’t tell the sequence just by looking, and trying to line the sister taxa up in an order. You have to analyse it

      • Trees do give the sequence: vertebrate (fish) – tetrapod (amphib) – reptile (synapsid) – mammal, eutherian, primate, hominid.
        And, where have you been? I do analyze the taxa after the computer puts them in an order. That’s the whole point. Others, who nest pteros with dinos and mesos with pareiasaurs are not doing proper analyses.

      • “I do analyze the taxa after the computer puts them in an order.” That one sentence shows you have paid no attention at all to what I said. The computer does not, repeat NOT put the taxa in order. As I clearly said in my last post, that is considered at best extremely difficult and at worst impossible. The computer does not even put the characters in order during phylogenetic analysis (at least not in a way that is visible just by looking at the phylogeny). As I said that must be done in an entirely separate analysis. Why do I bother typing out these basic concepts when you don’t even read them!!! This is phylogenetics at its most basic, most fundamental! You should have made sure you understood them before doing the analysis

        The sequence you gave (fish, amphibian synapsid,mammal…etc) is not a sequence given by phylogenetic analysis: it is a sequence given by analysis of a phylogeny AFTER a phylogenetic analysis!!!!!! This is not difficult!!! Phylogenetic analysis gives a relative date of branching events, not a sequence of evolution! This is what I spent my last two posts trying to get across: the sequence of character acquisition and evolution comes from post hoc analyses.

        As for comments you made not in reply to me:
        “Too few characters is not a specific error, especially when the number of characters used produces a completely resolved tree, as discussed earlier.”

        Drivel. As I (and many others) have tried to explain before, a single tree is not an indication that it is right. The less characters you have, the less morphological variation you are taking into account. How can you think that that is not important? This is not difficult!!!

        Yes, you have more taxa, a great thing and something which should be encouraged. But your character list takes into account the tiniest proportion of the morphological variation available! Benson (2012), just as an example, used 30 percent (ish) more characters that you to analyse a tiny fraction of the tree! how can you think that more taxa can be reliably analysed by taking into account less morphological variation?? The very idea is beyond stupid!!
        Adding more taxa is fine and dandy, but the raw data are the characters as well. And if the raw data is poir, the result will be. Your taxon list is great, your character list is not. Before you code a single character, your dataset is handicapped: your character list is too small, full of redundancy (e.g. the character which deals with the presence of palatal teeth redundant with characters dealing with presence of teeth on specific palatal bones) , poor understanding of polarity (Taxa for which a character is inapplicable are coded instead of left blank, which will alter the polarities of characters e.g. taxa without teeth are coded for tooth shape characters as not having the shape, implying the morphology is the same as those who have teeth but not with the shape), poorly defined aspects (e.g. fenestreae are not defined by the bones which border them, making it tricky to test homology) and attempts at analysing multiple aspects of morphology with a single character (e.g. character 96 refers to both the presence AND the morphology of the vomerine teeth).

        Again, these are basic issues which need to be sorted before an analysis is even run! And to deny their importance and say that your analysis is still better because of more taxa makes me think you consider the characters of secondary importance in phylogenetic analysis. To the contrary they are the most important aspect.

        If you don’t listen to me, listen to your own data: a character consistency index of 10%. Only 10% of your own data supports the hypothesis you put forward. Analyses have been done assigning characters at random that have shown stronger support for a tree than that.

      • Neil, I can never add enough characters in your philosophy. If I add 1000 you’ll say I need 3000. This will not raise the consistency index. The consistency index is low because several taxa lack legs, several clades develop upper temporal fenestra and antorbital fenestra and elongated caudals, etc. etc. The large gamut of taxa is the largest contributor to the low consistency index. As a test of this, do as I have done: reduce the number of taxa to just a few clades, like just the captorhinidae or just the scleroglossa then run the analysis. You’ll find the CI will rise.

        If the computer does not create the Most Parsimonious Tree(s) then what does the computer do? The separate analysis of putting characters in order is done in MacClade, right?

        Granted some to many of my characters are not perfect. However, it’s up to you, or the next generation to show where the topology would be better with a different set of characters. The present topology makes sense in a more parsimonious fashion, which is, after all, the goal, right? More sense than the present analyses that nest long fingered pterosaurs with short fingered archosaurs, etc. etc. I’ve improved on prior analyses. Others are free to improve upon mine.

      • I see you’ve added a new blog post about the CI, so will deal with that there. For now I’ll stick to the phylogenetic issues.

        The computer creates most parsimonious trees. I never said it didn’t. However That is all it dows. A most parsimonious tree does NOT suggest ancestor/ descendant relationships. As for putting characters in order, there are lots of different ways of doing it. I believe MacClade can (although I’ve never used it). The point I was trying to make is the sequences of taxa you show ccan’t be derived just by oggling the tree. The ancestral states may not always be what seems apparent from the tree, and will only very rarely be represented in the entirety by specific taxa. But its the issue of trying to create an evolutionaryt setries from a cladogram that gets to me. You can suggest ancestor/descendant relatuionships if the ancestor is paraphyletic (e.g. fish/amphibian) but not if the tips are monophyletic and species. But I do wonder at this point if we are talking at cross purposes and are actually arguing differnent points. The trouble with arguing on the interweb instead of in person

        As for the character issues: if you acknowledge there are problems, why are you so certain you are right? That seems to me to indicate a fundamentally flawed mindset

  2. How can you seriously moan about an ‘all out attack’ from someone else (which is really just pointing out your errors) when you have written a whole series of ‘scathing reviews’ on what is regarded by everyone else as the definitive guide to pterosaurs and then deleting comments you have no reply to?

    • When specific errors are shown, they are fixed. I have been very specific in the scathing reviews. I delete comments that are general blackwashing, that do not offer specific solutions.

  3. If you are going to make broad claims such as “more is better!!!!” and that the whole point of evolutionary study is to show gradual transitions, you must be prepared for those broad claims to be disputed, and not artificially limit debate to “specific solutions.”

    • What possible debate counters: “more is better” and “gradual transitions?” Please advise. Specific disputes will be addressed. List the taxon. List the problem. We’ll deal with it. I’m trying to keep you guys scientific, apolitical and to scrub the emotion out.

      • What counters “more is better” and “gradual transitions”? My background is astrophysics, so let’s go back to one of your own analogies, the electromagnetic spectrum. You criticize Darren for talking about red, blue, and yellow when you include every shade of violet, magenta, azure, and chartreuse. Obviously “better”, right?

        Well, no. Suppose we want to look at radio waves from a distant galaxy. Now we’re out past the infrared, in wavelengths of millimeters or centimeters rather than nanometers. It hardly even makes sense to mention visible light on our spectrum at all, let alone preserve the gradations between all the colors of the rainbow. At the same time, we know where all those wavelengths are, over yonder — their relationship to radio waves hasn’t changed.

        Now we detect a gamma ray burster in that galaxy, and we’re off in the other direction, past visible light to ultraviolet, X-rays, and then gamma rays. Do we need to keep the radio wavelengths on our spectra now? That doesn’t make sense either.

        Back to evolution. You give an example of ordering “a sponge, a fly, a starfish, a lamprey, a swordfish, Eryops, a snake, Dimetrodon, Burnetia, a platypus, an elephant, a cat, a rabbit and a tarsier” as though that were an obviously ridiculous thing to do. But it isn’t at all. That’s actually a fine set of taxa to outline the broad strokes of animal (and specifically primate) evolution. We don’t need every known species of fish on a cladogram showing the relationships of humans to other animals — zoom in on the hominins, and maybe other primates, and then plant stakes in the ground to show the direction to the rest of the tetrapods and beyond.

  4. It is a fine list of taxa, but it is ridiculous to promote them as your own and damaging if you attribute them to someone else. As you know, Darius, we have much better, more direct, ancestors and descendants, so we use them to show the paths that evolution took. That’s the whole point. We don’t show just the cities. We show the highways between them.

    We don’t need every known species. True. That’s why only 357 are shown. I hope you see the logic in that.

    • I feel you misunderstand the reasoning behind Darren’s simplification of your cladogram. He didn’t omit taxa to make it appear as though you placed obviously unrelated taxa next to each other. Indeed, for the most part, his tree aligns with (for want of a better term) orthodoxy. The main divergence is putting pterosaurs over with the squamates, which is also your main divergence; his simplification makes that easier to see. And pterosaurs aren’t even obviously ridiculous over there — a layperson wouldn’t bat an eye. Reptiles are reptiles, right?

      In contrast, a layperson would look at your “large reptile tree” and their eyes would immediately glaze over. “Only” 357 species? That’s a lot of trees getting in the way of seeing your forest. More is not necessarily better. We can talk about the forest without discussing every tree.

      Oh, and in case your reference to “Darius” wasn’t just a riff on my pseudonym — my name isn’t Darius, it’s Mark Bradford. My degrees are in physics and astrophysics, and my trade is computer support at a major scientific institution, but I have a lifelong amateur interest in paleontology.

  5. Thank you, Mark. I had no idea what your real name was until now. I’m not blogging for the lay person. And by not using images from my site to criticize the site, yet crediting me, more or less, Darren was biasing the argument for his lay readers, basically creating editorial cartoons. In this case, “more ” further cements every relationship as each new taxon provides new opportunities for nesting sites. With full resolution and so many taxa firmly nested that reduces the possibility of error. Some readers eyes may glaze over, but I am reaching others’ that do not.

    • The non-layperson is even less likely to bat an eye at Darren’s simplified cladogram, because they understand that it’s the branching relationships that are important, not everything in between. You are persistent with the notion that full resolution means correctness, but I’ve seen many professionals try to correct you on that.

      What you’re doing, ironically, is the inverse of what some creationists do. When a new transitional fossil is identified, they proclaim “That’s two more gaps!” You’re piling more taxa into your tree as though eliminating every gap eliminates every possibility of error. But if the fundamental basis for your tree is incorrect, the whole thing unravels at once, however tightly-knit it may have seemed.

      You say you are reaching the eyes of those whose eyes do not glaze over. But when real, working, professional paleontologists come here to engage with you, you reject their conclusions and delete their comments. It’s a shame to see you throwing away opportunities for engagement that didn’t exist in the pre-Internet age.

      Off to the day job now.

  6. It amazes me that I’m doing exactly what others do, creating phylogenetic trees, only on a larger scale, with a larger gamut of taxa. I solved several mysteries, the tree makes sense, taxa share traits and yet commenters throw cold water on it — without showing errors or better nestings.

    I delete comments from professionals that are not professional in nature, that only insult.

    • Are you, in fact, “doing exactly what others do”? Let’s explore that for a moment.

      Given that I am an observer in the field and not a participant, I must rely on what I have seen described. I’m sure you will correct me as needed. But right away you state that commenters do not “show errors” — whereas I have seen commenters describe your use of too few characters as an error, and your reliance on photographs of specimens as an error, and (in this very post) your interpretation of Darren’s tree simplification as ridicule as an error.

      As for showing “better nestings,” certainly other researchers show their phylogenetic trees, which differ from yours — but if you are the arbiter of what constitutes a “better” nesting, then naturally you would not consider those to be “better.”

      Your use of DGS to determine the characters you use is certainly not “doing exactly what others do.” And your claim to have “solved several mysteries” relies on others finding your solutions to be well-supported explanations. But when others find your solutions to be dubious or non-reproducible, it is not surprising that they are not widely accepted.

      As for insults… disagreement, even vocal and heated disagreement, is not necessarily equivalent to insult. In Witton’s text about pterosaur affiliation, he observes that your protorosaur hypothesis has some merits, before leaning toward the archosaur hypothesis. And when he rejects the squamate hypothesis, he calls it the “most unlikely” hypothesis; that is not the language of insult. Your own characterization of his work in your “scathing book review” is at least as intemperate as anything I’ve seen in criticism of you.

      At any rate, I suspect you’ve heard others say things similar to what I’ve just said many times, so I have no particular expectation that you will find my comments illuminating or convincing. I’m an optimist, though, so I felt I had to try.

      • Errors for specific parts of specific taxa are dealt with. Too few characters is not a specific error, especially when the number of characters used produces a completely resolved tree, as discussed earlier.

        Everyone relies on photographs and, more often, line drawings of specimens as their data. No one has every specimen on their desk in front of them when they conduct phylogenetic analysis. Often they rely only on previous data matrices and skip seeing even drawings of the fossil, but accept the data as reliable. Not saying that that’s bad…

        With regard to most of your comments, like better nestings, those have to be dealt with on an individual basis, but my tree offers a larger gamut of nesting opportunities that smaller studies do not. The taxa find their own sites, not always in traditional places.

        Witton’s pinned the label squamates on my alternatives for pterosaur ancestry when squamates (Iguania + Scleroglossa) are not my alternatives. This was covered in an earlier post. And it is you who are labeling this as an insult. I considered it bad Science and deliberate misinterpretation of my reportage.

  7. “No one has every specimen on their desk in front of them when they conduct phylogenetic analysis.”
    True, but most people that construct trees travel and look at the specimens and measure them in person so as to avoid the problems that may occur due to use of photos or line drawings only (or primarily). Researchers do often trust that other researchers are doing the same thing when taking their data and building new trees. Does that mean mistakes don’t happen? Of course not. You (and Mickey, and various others) have pointed out mistaken codings in trees and it happens in the published literature all the time. I’ve had it happen to me. In general, though, the data are thought to be relatively reliable since other scientists are checking and attempting to reproduce results (the essence of science). I think your comment struck a chord here since you seem to be dismissing others’ work since they don’t have “every specimen on their desk” but in fact most of them have had the specimens in question in front of them while coding and recording data…maybe just not all at one time.

    • Rob, Virtually all of those studies are introducing new fossils to the world. So the new fossil is indeed in front of them, but often the analyses, as you know, are pulled from the literature and incrementally added to. That is standard operating procedure. Rarely, if ever, do workers attempt reconstructions or initiate precision tracings (in pterosaur studies). I’m attempting an experiment that demands quantities of data from a larger gamut of taxa. It’s a different sort of ball game, but only to the effect that I’m NOT depending on data from other analyses, but go straight to the primary literature. Many specimens I -have- examined in person, by the way, with trips to Kansas, China and Germany and Spain, while Russian specimens came here to St. Louis. Lucky for me!

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.