PDA

View Full Version : Wallace Arthur's Biased Embryos and Evolution: A Review


James Brody
October 2nd, 2004, 09:36 PM
"There are things that are facts, in a statistical sense, on paper, on a tape recorder, in evidence. And there are things that are facts because they have to be facts, because nothing makes any sense otherwise." Philip Marlowe
---------

The 20th century dominance of evolutionary theory by population geneticists took a bullet in the 1980s when biologists discovered Hox genes, transcribers that organize the head-to-tail and back-to-belly of every multicellular creature. The impressive fact is that Hox not only occur widely but also with great similarity in their DNA sequences, similar enough that Pax-6 taken from a mouse or squid will organize the eye in a fly and do so whether transplanted to a head, wing, or leg. And not just an eye but one that generates electrical activity! Haeckel breaths...

Species vary in their number of Hox sets and sometimes in the number of Hox within a set, but for plants, clams, flies, mice, and Meg Ryans, Hox unpack everyone's bags and similar Hox make similar arrangements of whatever was in them. Thus, Wallace Arthur, who started in population genetics and moved into evolutionary-developmental biology (evo-devo), suggests but not very forcefully that (1) natural selection is fundamental but cannot shoulder all evolutionary outcomes, (2) "developmental bias" influences the outcomes that may emerge from selection (for example, of 3000 centipede species, not one has an even number of legs), and (3) there's a lot to say about historical contingency in evolution but not in this book.

He advocates that developmental bias provides both constraints and options for evolutionary changes. It serves us as a hypothetical construct that acts between a mutation and the spread of that mutation through a population: developmental bias both serrates and creases the folding of a species's paper but, like a generative grammar, also provides hooks for an infinity of small utilities called adaptations. (Waddington could have been the ghost writer!) Further, the evolution of an organism requires changes in the birth, growth, and maturation of that same organism: evolution shapes not only adults but also gametes, fetuses, neonates, and adolescents and, therefore, variations in phenotypes must arise from changes in the development of individuals. (Arthur also suggests a three-story analogy: selection catalyzes population changes, mutation makes genetic ones, and "developmental reprogramming" is nominated to be the operator for those in single organisms.) It follows that evolutionists must consider not only the scatter of genes in bean crops but also changes in the unfolding of individual beans. Ontogeny both repeats phylogeny and leads it.

There is still extraordinary flexibility as to when, where, and how developmental biases become active. Arthur illustrates these possibilities with his thoughts on segments and appendages or, in his terms, "stripes" and "spots." For example, the Hox gene for "spots" has a role in butterfly decorations, insect legs, and primate arms. Furthermore, "stripes" appear to have evolved independently three times but orchestrated each time by the same genes. Segments and appendages often occur together but it is also true that many more species do not have segments than have them and many more species have appendages than do not. (A clue?)

Arthur wants to bring back the organism but admits that heads and limbs, and even the left- and right-side versions of them, may follow somewhat independent trajectories across time of whatever duration. (According to Arthur, A. R. Wallace noted the same possibility in 1887.) Symmetry lapses, therefore, may not reflect pathogens or accidents but developmental experiments! After all, our mind works as it does because it's asymmetrical and we don't yet thank or curse a virus.
Arthur also believes there are multiple drivers in evolution (and cites his affinity to Darwin on this point) and would legitimize orthogenetics if he could. He acknowledges roles for molecular selection and random drift but not as major stars in the big game. I appreciate his brief comments on cladistics, on the simultaneous importance and neglect of behavior genetics, and on the forward and backward evolution that pivots around our phylotypic stage. (Humans and chickens, for example, look very different as eggs but soon look very much alike at the phylotypic stage before, still later, they either nurse or peck and again look very different from each other.) Also, I like Arthur's recognition that evolution is a digital event and may take steps of various sizes; things are added on usually in smaller increments than when something is transplanted, disinhibited, or taken away.

Each of the 14 chapters almost stands as a separate essay and Arthur keeps his arguments short. Unfortunately, Cambridge resurrected Polonius as the final copy editor. Arthur backs into most sections and minuets through preambles, asides, and qualifiers. He introduces developmental bias as if debuting his awkward youngest daughter at a cotillion, defined by Webster as an intricate Victorian dance, of elitists and perverts. I mumbled harsh words and threw his book back in my bag several times before coming to enjoy his frequent moments of nonlinearity.

A GUT?

We dreamers want one Grand Unified Theory that we can ride from quanta to our forming the clubs that manipulate them. Unlike Marlowe on a case, Arthur neglects some fundamentals in the "mystery of mysteries" (but, then, he wrote a short book).

My favorites are:

1) Statistical physics (emergent networks, power laws, phase transitions, and oscillation phenomena) supplies our primal environment: we find in biology the phenomena that we find not only in NY commuters but also in gases and liquids, magnetism, conductivity, satisfiability problems, networks, and oscillators. Further, life has the squishy equivalents of tuners, capacitors, resistors, amplifiers, filters, diodes, timers, and batteries as part of its evolutionary heritage.

It should be, therefore, no surprise that the emergent networks for social patterns in humans, wolves, dolphins, and termites accomplish the same efficiencies that we find in ecological webs and neural organizations. A creative high school term paper has one to three major points and lots of smaller ones and might be described by a power law! (Would we get better themes if we wrote through doodles rather than outlines? This method certainly worked for Feynman and perhaps for Arthur!) My point? The payoffs for emergent, nonrandom organizations are so much a part of our nature that we "just do it" with the same instinctive efficiency that we manage oxygen. You cannot ignore your social instincts but you can develop without them and, without the rest of us helping you along, you will leave behind only a few dry, stained bones.

2) Margulis loathes neodarwinians but has genuine gifts, if we would accept them, in her thoughts about symbiosis: partnerships speed evolution along. That is, strategies for the lateral transfer of information take many forms and may be busier and more agile than we imagine. (After all, partnerships and lateral transfer give their participants access to a more varied genome!) Nature, for example, reports that eubacteria and archaebacteria probably each contributed to eukaryote genomes such as ours: relationships are not linear but circular. Further, humans have DNA sequences that resemble not only those from chimpanzees but also from gorillas. Symbiosis hustles at Darwin's pool table.

3) Arthur neglects evolvability and exploratory systems, two ideas from Kirschner and Gerhart that provide impressive support for the thesis of developmental bias. The metazoan bounty of species rests on conserved foundations that accept very specific modifications, probably millions of them. Thus, a mutant form, even a tiny one, that serves a purpose is apt to be copied. Introduce positive feedback to an organization and self-interest can follow. Every living creature becomes a self-managing explorer, one that pushes boundaries and discovers uses for whatever traits that it has; costs and baggage sometimes become assets instead of mistakes. Cooling systems took one form in bees and another in birds and man but in all three cases led to flight. The dynamics of exploratory systems now guide our probes to Saturn.

4) Lamarckian concepts apply to environments that are first constructed and then inherited and refined by the next generations. These environments may themselves exert selective effects. Such is consistent not only with Darwin but also with Per Bak's idea of self-organized criticality: environments and occupants stabilize each other, big changes occur early and the small stuff comes later.

5) Receptors may be considered as "strange attractors" for evolutionary development. No matter the varied or peculiar look of our appendages, we orbit around similar events and STKE (Signal Transduction Knowledge Environment. Obviously.) may yet enclose a macrocosm. That is, organisms in a stable niche develop ever finer detection of opportunities to scavenge free energy and new ways to make a living. Receptor mechanisms may be difficult to trace across time but are probably more stable than the thoughts, guts, muscles, and bones that attain them.

Symbiosis with the Gene People

"...the adjectives and such things were frank and fair-minded and straightforward and did not shuffle; it was the Verb that mixed the hands, it was the Verb that lacked stability, it was the Verb that had no permanent opinion about anything." Mark Twain on "Italian with Grammar."
---------

If you believe that selection accomplishes instruction by environments, this section might throw you! Twain's comment applies to many verbs and nouns in evolutionary biology. Doubtful? Survey "gene," "fitness," or "epigenetic." Further, Arthur could have scribbled alongside John and Leda in 1992: neither the mind nor evolution are domain-general systems! Oddly, most of us embrace modularity for the development of cognition in individuals but not for the deep phylogeny of that same cognition. In matters of brains, legs, or wings, we believe that stochastic events in the history of species grew prescriptive organizations that today make individuals! To the contrary, arguments against Lockean instruction for the emergence of individual minds surely apply to the emergence of species by means of selection from random variation. For example, when we consider the careers of species, there is no amount of time that allows what we have unless selection chooses from the outcomes of strategies established long ago, first in physics and much later in biology and repeated today.

Unfortunately, some neodarwinians endorse beliefs that evolution, like constipation, must be incremental, tedious, unpredictable, and driven by discomfort or else it couldn't truly be evolution and that Dr. John Bowlby made up a great, useful story about our Pleistocene mother. Bah! These are ALL stories that Marlowe wouldn't believe after three drinks from his pocket flask and neither would the L.A. cops: neither do I. In contrast, biophysics, evo-devo, and genetics hint that evolution can be quicksilver and that its products may be less uncomfortable and possibly happier than many of us suspected as recently as five years ago. The forces of nature are not hostile but indifferent and mismatch is often a belly pain projected by a few mildly depressed, obsessive academicians.

Fortunately, duplication, divergence, specialization, and cooption apply to the evolution not only of genes but also ideas. Credos detach easily from facts and facts enter new partnerships within a different set of credos even without someone's dying. In the meantime, our extraordinary, wealthy niche supports unique additions to evolutionary theory and many of us throw gossamer threads that hang us from twigs in the Origin. Also, many of our students will read evo-devo texts, Arthur might come to an HBES meeting and speak if invited, and Rudolf Raff should be a keynote speaker every year for the next five!

Eventually, there will be symbiosis. It's a natural thing...

General References

Arthur, W. (2004) Biased Embryos and Evolution. NY: Cambridge University Press. $32 in paper. Index, glossary, 233 pp. See also the review by Leroi, A. (2004) Nature. 430: 294.
Kauffman, S. (2000) Investigations. NY: Oxford.
Kirschner, M. & Gerhart, J. (1998) Perspective: Evolvability. Proceedings of the National Academy of Science. 95(15), 8420-8427.
Marlowe, P. (1958) in Playback by Raymond Chandler. NY: Ballantine, p. 137.
Odling-Smee F.J., Laland, K.N., & Feldman, M.W. (2003) Niche Construction. The Neglected Process in Evolution. Monographs in Population Biology. Princeton, NJ: Princeton University Press.
Raff, Rudolf (1996) The Shape of Life. Chicago, IL: University of Chicago Press.

Detailed references available on request: jbrody@compuserve.com

Copyright, 2004, James Brody