I've been reading a strange book by Stuart Pivar, LifeCode: The Theory of Biological Self Organization ^{(amzn/b&n/abe/pwll)}, which purports to advance a new idea in structuralism and self-organization, in competition with Darwinian principles. I am thoroughly unconvinced, and am unimpressed with the unscientific and fabulously concocted imagery of the book.

There exists a real difference of opinion between two approaches to biology, the functionalist and structuralist views, and it influences how we look at evolution. The functionalist position is the one most people are familiar with: populations contain organisms with differing fitnesses, selection promotes the individuals with the better functional adaptations, and the properties of populations change over time to improve their functional status.

Structuralism is harder to get across. A key figure in historical structuralism is D'Arcy Wentworth Thompson and his book, On Growth and Form ^{(amzn/b&n/abe/pwll)}; Thompson summarized it most tersely in this aphorism:

Everything is the way it is because it got that way.

The emphasis in structuralism is on process and history and interactions with the environment; what Thompson means is that form is a consequence of developmental and evolutionary processes (although no Darwinist, Thompson was entirely on board with evolution), and that the understanding of why organisms look as they do lies in understanding how they are assembled. Stephen J. Gould, in his forward to On Growth and Form, explained Thompson's guiding principles this way:

This hybrid theory of Pythagoras and Newton argues that physical forces shape organisms directly (with "internal" and genetic forces responsible only for producing raw material, admittedly in gradients and programmed timings, for construction under principles of physics)—and that the ideal geometries beloved by classical Athens pervade organic form because natural law favors such simplicity as an optimal representation of forces.

As you might guess, Thompson is very popular with developmental biologists. It's not just that he privileged our discipline with importance, but that he was a fascinating writer, extremely erudite, and his book is rich with observations and insight. He didn't just claim that physics and geometry dictated aspects of form, but he carefully documented it with examples. Cells tend to be round, for instance, not because they need explicit genetic instructions to assume that shape, but because extant physical forces configure it that way in the absence of other specific internal efforts to change it.

I mention Gould and Thompson intentionally, because as it turns out, Pivar claims the mantle of both Gould and structuralism. In LifeCode, he lays out his theory of morphogenesis based on topological distortions of cells. He was a personal friend of SJ Gould's, and has recently made friends among the anti-evolutionists and Intelligent Design creationists by claiming that Gould did not believe that natural selection played a major role in evolution, that he was suppressed by "anti-antidarwinist forces", and that heterochrony was "Stephen Jay Gould's theory of evolution." It's all very peculiar. I'm not going to bother with these claims—they've been dealt with elsewhere—except to say that I think they are ludicrous.

Instead, Pivar sent me a copy of his book to review, so I'll focus on that. I am a fan of Gould's and I rather intensely despise creationists, so taking that tack would not predispose me to evaluate it charitably. However, I also appreciate the ideas of structuralism (especially as expressed in that quote by Gould above), think Thompson was a demi-god in the pantheon of developmental biology, and favor the hybrid perspective of West-Eberhard, so instead I sat down with it as I would to a new book by that better known structuralist, Brian Goodwin^{(amzn/b&n/abe/pwll)}—critically, but with an inclination to enjoy some new ideas.

I was disappointed.

I will say this: it is very pretty. Pivar has a website for the book, with some of the artwork on display. It's visually striking stuff.

Unfortunately, it has almost nothing to do with reality. His theory (which is also explained on a website) is all about topological manipulations of embryonic forms, and he uses the artwork to show his models for how the embryo is distorted by physical forces to generate various structures…and they simply do not bear any relationship to cell and tissue movements in any embryo I've ever seen. Here, for instance, is his explanation for the development of the skull.

Cheetah Skull and Brain. The brain is an enclosed membrane within the brain case. With expansion, it bulges and extrudes through six fenestrae, extruding to form nasal, optic, and auditory sacs.

The head forms from the polar cap, which consists of three flat, broad segments of the polar sphere appearing as three concentric, ringed zones. The segments are divided into four quadrants. Each of the resulting twelve parts develops a circular hole at the center. The six fenestrae of the future dorsal side will form the nasal, optic, and auditory capsules. The ventral half produces the jaw. The architectural scheme of the skull results as concurrent, deep, invaginating infolds along the dorsal and ventral midlines fuse, sealing off enclosing cavities in each hemisphere. These meet internally and form the floor of the brain case and the roof of the mouth. These cavities are analogous to the structure in plants that produce the anthers.

This is a remarkable story, especially since, try as you might, you will see absolutely none of the described features present in the developing embryo. The jaw arises as a caudal condensation, for instance, and extends forward during development. There aren't any of these regular fenestrae, there isn't a ventral invagination, and most importantly, the vertebrate head never looks like any of his intermediate stages. These are pure confabulations.

He attempts to explain many phenomena in development, all with the same stunning failure to attempt to look at real organisms, favoring instead the geometric purity of his imagination. His explanations of gastrulation, while more abstract, are notably lacking in any association with actual data. I've watched gastrulation occur; I have friends who study cell and tissue movements in zebrafish gastrulation; I've written software that is used to analyze changes in embryo shape and cell migration. Let me tell you, nothing in this book is even an approximation of what we actually see.

I'm afraid all of these artistic inventions are in the service of a theory that is unappealing, uninteresting, and without the slightest bit of predictive power. Here is his idea, briefly: all cells and embryos are donuts, and how they turn themselves inside out determines their future morphology. To be fair, I'll transcribe it as he has written it:

The shape assumed by the surface enveloping the primordial germ plasm is a geometric or topological form called the sickle torus, a kind of torus that can be described as one spheroidal surface inside another, interconnected at the poles by a narrow funnel vortex. Its cross section is sickle shaped as opposed to the circular cross section of simpler doughnut-shaped torus.

The toroidal egg cell surface is fluid. It is in constant axial flow like smoke rings or magnetic fields but substantially slower.

Toroidal streaming imposes on the membrane sheet the relentless cycle of alternate compression and expansion suffered as the sheet passes through the narrow vortices. The membrane becomes empatterned by radial and circumferential subdivisions. This pattern guides the expansion of the membrane to the mature egg-cell plasma membrane and later, as separate cells, guides the formation of the standard adult phyletic body.

Doesn't help much, does it?

Again, none of this—sickle torus, rings of axial flow, toroidal streaming, radial and circumferential subdivisions—is in evidence. How it generates the complex form of an organism is not obvious. How his theory can be reconciled with a large body of embryological evidence that directly contradicts virtually all of it is not clear, and Pivar has chosen not to address any of it. And a book full of geometrically interesting sketches neither explains nor supports Pivar's theory.

I have to add another compliment for the book, though. In addition to the lovely artwork, it's an extremely high quality print; well bound, on heavy stock, and looking to last a thousand years. It seems no expense was spared getting it published, which is in contrast to the content, and is unusual for such flagrant crackpottery. It may well be popular among creationists, who can always be trusted to favor glossy superficialities over substance.

To Mr Pivar, I would suggest a simple rule. Theories are supposed to explain observation and experiment. You don't come up with a theory first, and then invent the evidence to support it.