“Gradients” and “Fields” in Developmental Biology: A history of the ideas

16 01 2012

Anyone who’s taken a course in developmental biology will have heard of “developmental gradients” or “embryonic fields” or “morphogenetic fields”; I learned these in German, so the English names might be different (I’ve seen those three being used). This post is about the history of these ideas of fields and gradients in developmental biology.

Theodor Boveri was the first to postulate the existence of gradients. Boveri was a star of his day, a pioneer of experimental biology, known for being one of the first to expand on the role of stem cells in the embryo (Boveri, 1892) after Ernst Haeckel (who coined the term “stem cell”); he also left his mark on cancer biology, being the first to discover that extra centrosomes in cells lead to malignancy (Boveri, 2008). He now has a biology institute at the University of Würzburg (where he was Professor of Zoology for most of his career) named after him. Of his students and protégés, the most relevant for this post is Hans Spemann, a name that should be instantly recognised if you’ve done basic developmental biology.

Anyway, in Boveri (1901), he conceptualised the sea urchin embryo as being built of “layers”, which he termed “gradients” in Boveri (1910). This idea was then taken up by Charles Manning Child, a developmental biologist known for precisely this subject of morphogenesis. Due to the idea being discredited for a long time, he’s not nearly as famous as he should be though.

These ideas were based mostly on his experiments on regenerating organisms, namely Hydra and planarians, although he later extended them to all multicellular organisms. It all culminated in the publication of his magnum opus in 1941, Patterns and problems of development.

Basically, his thought went something like this. The developing embryo has “dominant” areas and “subordinate” areas interacting with each other. There are no rigid boundaries between the two types of area; they’re determined solely by physiological activity (i.e. metabolism), and they’re fluid, changing state and forming new areas as development proceeds. This fluidity is best expressed as a gradient, as we’ll see.

The very first producer of these areas is the embryo’s polarity, which, according to Child, is caused by external stimuli. These stimuli cause one side of the embryo to be more active (i.e. have a higher metabolic rate), and this will lead to differentiation as that side produces different quantities of chemicals than the others. This is now known to be nonsense, of course. But the idea of gradients was good, as it was also known that diffusion played a big part.

The idea of “embryonic fields” was first proposed by Gurwitsch (1922). Important additions were made by Gurwitsch (1927) and Weiss (1925). They determined that while these fields are under some sort of stereotypical embryonic control, some areas within the fields can be altered surgically, changing the fates of those specific areas.

Julian Huxley and Gavin de Beer are the two who further carried the torch, by fusing the idea of fields and the idea of gradients together. Huxley needs no introduction: harbringer of the Modern Synthesis, brilliant biologist worthy even of his own post (there’s something for the to-write pile). Gavin de Beer might not be as famous. He was a zoologist and developmental biologist, the first to have published a compilation of Darwin‘s notes in 1960, most known for his 1937 book The Development of the Vertebrate Skull, a book so good that no further research was done on the topic over the next several decades, and his 1962 book Embryos and ancestors. He was one of the first biologists to link developmental biology to genetics. His last book, 1971′s Homology: an unsolved problem, presages the debates about the links between genotype and the phenotype in evolution, debates that are still ongoing.

Based on earlier work (Huxley, 1924; de Beer, 1927), they co-authored The elements of experimental embryology in 1934 (linked is the 1964 edition). In it, they advocated a field-gradient theory, in which the morphogenetic gradients are later punctuated by fields. These fields are also themselves gradients, but instead of being evenly diffuse areas, they’re more like peaks, with the top being an area of high morphogens and the bottom being much more diffuse. This addition of fields is Huxley and de Beer’s breakthrough.

That said, the first comprehensive bringing together of the ideas of fields and gradients into a coherent theory was done by Dalcq & Pasteels (1938). Their theory had two main components: a yolk gradient and a dorsoventral cortical field. However, there was one big problem: they based their entire concept on the study of amphibian embryos. Eckhard Rotmann (1943) took this, as well as other problems, into account, in his very critical look at the gradient-field theory they had developed.

However, the more scathing criticism came from Leopold von Ubisch, a student of Boveri. The critique in von Ubisch (1953) is still as valid today as it was back then, and revolves around the origin of these gradients – he doesn’t deny that they exist, nor does he fail to recognise their importance in morphogenesis. He does, however, criticise the work of previous researchers, referring to their experiments as unsatisfactory with ambiguous results. He especially criticised the lack of consideration of environmental factors – all previous work had focused solely on physiological factors, even though the environment can play a very large role, especially in matters concerning diffusion of molecules.

The real reason why his criticism is still very valid today is because he also stresses the importance of genes, following on from the pioneering thoughts of Morgan (1934) and Waddington (1940). Instead of only considering physiological factors, he calls for the investigation of the interactions between genes and physiology, postulating that as development progresses, different genes get activated, and that these might be the controlling factor behind the gradients and the fields.

This can also be seen as the beginnings of systems biology in developmental biology. We don’t have simple cause and effects anymore, but an entire cascade of events, all of which affect each other. This is attractive to mathematicians, and none other than Alan Turing started modeling these processes. In Turing (1952), he showed that two autocatalytically- and cross-catalytically-interacting substances will spontaneously form a concentration gradient. It’s very easy to follow this on to the idea of gradients and fields of morphogenetic molecules, but the paper was largely ignored for two decades, until Gierer & Meinhardt (1972) picked up the thread by simulating hundreds of regeneration and transplantation experiments on Hydra, using “cybernetic” models.


Boveri T. 1892. Über die Entstehung des Gegensatzes zwischen Geschlechtszellen und den somatischen Zellen bei Ascaris megalocephala, nebst Anmerkungen zur Entwicklungsgeschichte der Nematoden. Sitzungsberichte der Gesellschaft für Morphologie und Physiologie München 8, 114-125.

Boveri T. 1901. Die Polarität von Ovocyte, Ei und Larve des Strongylocentrotus lividus. Zoologische Jahrbücher 14, 630-653.

Boveri T. 1910. Die Potenzen der Ascaris-Blastomeren bei abgeänderter Furchung. Festschrift zur sechzichsten Gebutrstag Richard Hertwig 3, 131-214.

Boveri T. 2008. Concerning the Origin of Malignant Tumours by Theodor Boveri. Translated and annotated by Henry Harris. Journal of Cell Science 121, 1-84.

Dalcq A & Pasteels J. 1938. Potentiel morphogénétique, régulation et “axial gradients” de Child: Mise au point des bases physiologiques de la morphogénèse. Bulletin de l’Académie Royale de Médecine de Belgique (Ser. 3) 6, 261-308.

de Beer GR. 1927. The mechanics of vertebrate development. Biological Reviews 2, 137-197.

Gierer A & Meinhardt H. 1972. A theory of biological pattern formation. Biological Cybernetics 12, 30-39.

Gurwitsch A. 1922. Über den Begriff des embryonalen Feldes. Development Genes and Evolution 51, 383-415.

Gurwitsch A. 1927. Weiterbildung und Verallgemeinerung des Feldbegriffes. Development Genes and Evolution 112, 433-454.

Huxley JS. 1924. Early embryonic differentiation. Nature 113, 276-278.

Morgan TH. 1934. The relation of genetics to physiology and medicine. Nobel Lecture.

Rotmann E. 1943. Entwicklungsphysiologie. Fortschritte der Zoologie 7, 167-255.

Turing AM. 1952. The Chemical Basis of Morphogenesis. Phil. Trans. R. Soc. B 237, 37-72.

von Ubisch L. 1953. Entwicklungsprobleme.

Waddington CH. 1940. Organizers and genes.

Weiss P. 1925. Unabhängigkeit der Extremitätenregeneration vom Skelett (bei Triton cristatus). Development Genes and Evolution 104, 359-394.



3 responses

16 01 2012
Daily Log: 16.01.2012 « Doing Biology

[...] friend led to me pontificating about the history of developmental biology, which led to me writing this post. I find it a bit too technical – no explanation is given as to what morphogenetic fields and [...]

23 01 2012
Miriam English

Lovely post. It certainly isn’t too technical. It is a nice overview of the history. Is it known yet whether the gradients come from fields of chemical diffusion? Or are they “gradients” of information propagated via cell contact like how computer-generated cellular automata create patterns? I imagine the latter could hint at why some animals have difficulty regenerating after damage.

In your log you mention the missed opportunity to mock Sheldrake. That idiot has single-handedly done more damage to thinking than almost anybody I can think of, except perhaps Lysenko. A google search of “morphogenetic fields” returned 13 results on the first page and only one had anything to do with reality — all the rest was Sheldrake drivel.

23 01 2012

As far as I know – and my information is limited – chemicals form the base. But there’s also a big field nowadays that studies the role of physical forces in development: gravity, cell migrations, effects of fluids, etc., and what that shows is that even if the chemical morphogen gradients are important, the raw physical forces can also play a big role.

Agreed on Rupert Sheldrake. A quack, through and through. But he is very useful in one sense, in illustrating one of the patterns that’s common in scientists: the older they get, the more nonsense they spout. If you check out his publications (http://www.sheldrake.org/Articles&Papers/papers/), you’ll notice that his earlier stuff is standard, run-of-the-mill botany. Until you get to the 90s onwards, and suddenly they all become batshit insane morphic resonance and telepathy XD Reminds me of Pauling: revolutionises chemistry and some aspects of biology, then in his old age becomes this pseudomedicinal quack.

I don’t have stats for how common this old-scientists-becoming-crazy pattern is, but it’s definitely there :P Either they become senile or they just feel like they have the authority (and, uh, tenure) to be “creative” with a lifetime of achievements behind them.

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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 )

Google+ photo

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

Connecting to %s


Get every new post delivered to your Inbox.

Join 357 other followers

%d bloggers like this: