Clarifying tetrapod embryogenesis, accurately

I’ll add here to the post accepted by PZ Myers. The comments will stay off, trackbacks open, the discussion belongs to Pharyngula.

This last paper of Fleury was accepted for publication in a journal of biology, in the section « Development and reproduction biology ». The journal [en] is an official publication of the French Academy of sciences [fr] (Organe officiel de l’Académie des sciences). Neither the editor or the reviewers of the manuscript have ever seen an illustration of a bilaminar human embryo?
If you go thought the text of the article, and I’ll give support for my take below, you have to exclude that a life scientist, or a physicist, or anybody competent on the subject, have reviewed it. You may even be prone to exclude that the reviewers are still able to use basic logics and spot Fleury’s long list of non sequitur.

I have been accused from a few French science bloggers for being harsh, strident, uncivil, harassing and not using the right channels to contest Fleury’s theory: a letter to the Editor.

So, this time I did things their way, on first instance. I contacted Dr Stuart Edelstein, one of the two listed Editors-in Chief, the only one with an e-mail address easy to find. He informed me that he was not acting anymore as EiC and assured me that he was transmitting my message to Dr Jean Rosa.

My message wasn’t something I asked to be published as a response to Fleury. You don’t respond to a guy who haven’t read his textbook in a scientific journal. I just suggested them to check a few points and the names of two people competent in the field from my readings: Brad Martinsen and Mike Levine. A month later, without even a receipt for my message I’m going back to my usual.

I’ll be a little bit more technical below, not much, as I’m not an expert on the subject, but enough to question the quality of the review process of C.R. Biologies, and conclude that either they did a gross mistake in this particular case or that the journal is not worth of trust based on this particular example. Depending of their action, retracting or not the paper, we will be able to chose between punctual error of generally flawed review process.

Poor Cephalochordata

During evolution of vertebrates a sequence of events is empirically observed: first, animals are bilateral, but they have no heart, no head, and no surrounding bag during development (these primitive animals are called cephalochordates [1])

Cephalochordata would be headless vertebrates. What kind of life scientist could read this phrase (the first one of the introduction) and don’t correct it? And no, Fleury doesn’t use mistakenly the word « head », you can read the same in French about the absence of « tête » about the amphioxus. A subphylum named after its anatomical particularity, the extension of the notochord into the head

Sugarcoating bullshit

In terms of darwinian evolution, it is clear that these structures might confer a positive advantage for animal development [3]

And the particular reference is the whole Origins! And the reviewers haven’t asked for a precise passage where this is discussed. Name dropping to sugarcoat that:

[…]the selection of tetrapods by evolution is bullshit[…]

Where the heart lies

I know that Fleury have real difficulties dealing with more than two dimensions at a time and connecting dot. He also have tremendous difficulties when it comes to know where the lateral mesoderm plates come from. But any scientist interested on the avian heart formation must have seen at least one of these schemas showing the origin of cardiac precursor cells populations.


This one is from a review article: Fashioning the vertebrate heart: earliest embryonic decisions, Fishman MC & Chien KR, Development 124, 2099-2117 (1997). Stages are according to Hamburger & Hamilton, the view is ventral and the colors show the region of the heart to which precardiac cells will eventually contribute.

How can one conclude that:

It is remarkable that the heart lies ahead of the dorsal, median axis, as regards the direction of the vector field of formation of the body axis, […] [emphasis mine]

A small portion migrates in front of the neural tube in formation as seen in the diagram for stage 7. And the heart never lies ahead of the neural tube (and that’s for avian embryos, I could get the same for a few other types).
What is really remarkable is that only an pig avian ignorant person could write this sentence, the introductory remark of Fleury’s « Discussion and conclusion ». And only as much ignorant reviewers could let it be accepted in a scientific article.

Where « entire » is redefined as « as small portion of »

The entire tissue field was analyzed, both in vector direction, and in magnitude.

If you have never, ever, see images of the chick development you may read the paper and accept, on the basis that is a scientists who makes the assertion, that he really analyzed the entire tissue field. But this is Fleury and one should certainly not accept anything he is asserting without doubting it (and you should crosscheck everything, whoever is talking).
If you have read a textbook about chicken development you know that he analyzed part of it. He didn’t got the ovoid picture. Or worse, he didn’t wanted to present it, as there are a few elements of it he can’t address with his model.
Patten’s The Early Embryology of the Chick. (1920, Philadelphia: P. Blakiston’s Son and Co.) have a much more detailed description of the entire tissue filed. In pages 68 through 72 he give a nice description of the heart formation, without omitting the contribution of the lateral folding during the primitive endocardial tubes migration toward the midline (fig 26) and clearly depicting the phenomenon from a ventral view (Fig 27). In Chapter XI, dealing with the extra-embryonic membranes, he provides a nice description of the lateral folds of the embryo and through figures 30 to 32 he gives a view of the entire egg and some very nicely illustrated details of the embryo’s development. Well, no time-lapse videos, but who need them when a trained embryologist is on the bench?
Fleury avoids to discuss the lateral folding of the embryo, around an axis oriented anterio-posteriorly, parallel to what he thinks induce the cephalic and caudal folds. Or the ventral extension of the somatopleure.
One can’t have the necessary expertise to review Fleury’s manuscript without knowing that a great part of the chickens developments have being simply omitted. Anyway, if the reviewers had to refresh their memories about chicken development they could have browse through Patten’s illustrations at UNSW Embryology pages.

When it is flat, with a modest furrow

However, in the context of this article, I have filmed with precision the developmental process just prior to the formation of the ventral and dorsal folds, in order to confirm that the heart and head folds occur perpendicularly to the developmental movement of the embryo at the blastula stage. Fig. 6 shows the vector field of development analyzed by Particle Imaging Velocimetry on a chicken blastula, at the stage when it is flat, and it exhibits only a modest furrow along the presumptive dorsal axis.

Unfortunately, despite the otherwise detailed section of « Methods », Fleury forgotten (and the reviewers didn’t noticed) to give any information about the software and the method used for PIV. The same guys where satisfied with the staging of the embryo « when it is flat » and with « a modest furrow ». I would say that this is a HH4-5 embryo with the primitive streak and Hensen’s node apparent and the images are frames of Movie 3 (gif, heavy).

The left panel gives « the global pattern of movement over one large area of the embryo« . Enlarge, check. Always check.


This horizontal arrow pointing to the left would be due to the two tiny specks encircled in red? Come on! Fleury just imagined the contraction here, it’s inconsistent even with the data he displays.

Worse(?), the right panel. Fleury focus around Hensen’s node during the regression of the primitive streak and (breaking news, from cutting edge research!, I’m kidding) demonstrates that the primitive streak regress! Textbook stuff. Certainly, the quality of his movie isn’t really great and somebody who isn’t familiar with chicken’s gastrulation may be lured. It’s so easy to subtract some background noise and enhance contrast to improve it that one wonder how Fleury didn’t spent a few minutes to do so.

Here, frames #1 and #36 of Movie 3, treated in ImageJ, background noise reduced by Rolling Circle, radius 20, and Auto contrasted, annotated: PS primitive Streak ; HN Hensen’s Node ; NC Notochord (visible through the transparent epiblast.

And here a juxtaposition of a detail around HN of frames #1, #36 & #71 showing its displacement.

I’m no expert on PIV, as Fleury remarked already, but I’m intrigued by the performances of the software he used. The embryo isn’t immobilized, at least not enough to make it easy to perform any PIV measures with the software I’m using to learn. I color coded the timing of Movie 3, and with a window of 20 frames I « z-plotted » the maximal intensity, in ImageJ. The resulting video show what a mess it is. Maybe Fleury will complete his « Methods » in the comments.

Somehow, the reviewers didn’t noticed that the left panel of fig 6 contains a gross error (the left pointing arrow) and that the right panel just confirms primitive streak’s regression. And forgotten to ask for details of how the PIV was performed. Lazy guys.

Where an ellipse behave like a circle

Back to fig 3 for a remark. Panel « a » shows the closure of the amniotic folds, over 8 h. At 0′ you can see the opening, a nice ellipse with a major to minor axis ratio of ~1.5. Somehow Fleury managed to measure

the speed of closure of the chorion hole (rate of variation of the radius) is found to be 1.66 µm•sec-1

Rate of variation of what radius? I suppose a physicist would be particularly interested to understand that point, but a biologists is also intrigued. They should have asked for an explanation in the « Methods » section, this is quite new!
A side remark concerning figures 2 and 3. Seems like the measures where made on just two embryos, one per figure. No numbers of independent experiments are given, no error bars are present on the graphs. That’s weird for a comparison of the two speeds, isn’t it?

If you haven’t read it don’t use it as a reference

The way Fleury reports the works of others in his articles is quite peculiar. You never know if the guy misunderstood what he read, didn’t read really the paper but use it anyway, if he intentionally distorts what others have shown because it doesn’t fit in his theory or even worse, contradict it. About his invited review pile of errors I heard one of the silliest excuses one could give: that’s a lot of papers cited here, some mistakes are acceptable.

The model presented in his last paper ask for symmetry, not only bilateral, but also cephalo-caudal. After renaming two folds cardiac folds and presenting them in a rather weird manner on a single cartoon despite the fact that he is aware that they are separated over time by 20-23 h (unexplained delay and what kind of reviewer haven’t ask for some explanation?), Fleury goes after the aneural caudal heart of the hagfish, to prove the power of prediction of his model. One may be sure that neither Fleury or the reviewers of the manuscript have read Jensen’s article, and not even the introduction of it. From The aneural heart of the hagfish, Jensen D., Ann N Y Acad Sci. 1965 Sep 8;127(1):443-58, doi: 10.1111/j.1749-6632.1965.tb49418.x–Introduction :

Anatomically and physiologically, the hagfishes possess four independently functioning pulsatile organs or hearts to propel their blood.

[…]three of these organs are completely independent of one another in their operation and each serves to circulate the blood in a specific area of the body. FIGURE 4 illustrates schematically the relationship among three of the hearts.

Situated in the caudal region are a pair of flattened sacs between which lies a cartilagenous plate. Lateral bending of this cartilagenous plate by rhythmic contractions of those skeletal muscles which are attached to it causes alternate filling and emptying of the lateral sacs: thus, the caudal heart propels the blood forward from the large subcutaneous sinuses, as the hagfish circulation is partly open. Retrograde flow is prevented by valves. Incidentally, the caudal heart is innervated, and its intermittent rhythmic beat is initiated by impulses from the central nervous system11, 15 (vide infra). [Emphasis mine]

Such a little animal, so many hearts and not enough folds to explain them. Unnerving.

Fleury is blindfolded by his obsession with the essence of tetrapods in the platonician space of forms and he didn’t read or understood what he was reading. That’s already a big problem for a scientist and I pity him for his condition. But what about the reviewers? Are they in the same case as Fleury? Or where they chosen between Fleury’s friends, just to please somebody, or because of the editors laziness to seek for experts?

Now what?

I don’t expect Fleury to do the honorable thing, retract the paper himself. But I ask to be surprised.

The Editor-in-Chief, Dr Rosa, ignored a letter pointing some of the papers deficiencies (I must admit haven’t presented the human bilaminar embryo, had to give him the opportunity to do better than me) and the article is still available.

How about a letter to the members of the editorial board or C. R. Biologies to signal this post and ask their opinion ? I’ll give Fleury a week and do that. Anybody interested to join me for this action?

All the original material (images and videos) are the production of Vincent Fleury and used according to his instructions for credit : Si vous utilisez ces films ou images, vous pouvez citer la source (Vincent Fleury/CNRS/Université Paris-Diderot), Those I produced are available for reuse under a Science Commons license.

%d blogueurs aiment cette page :