Cheaper than printing it out: buy the paperback book.

Out of Control

The bodies that genes wear play an incredible role in the gene's evolution. When two chromosomes recombine in sex they do so not in nakedness but clothed inside a gigantic egg cell. The overstuffed egg has a great deal of say in how the genes are implemented. The yolky cell is chock-full of protein factors and hormonelike agents, and controlled by its own nonchromosomal DNA. The egg cell directs the chromosomal genes as they begin to differentiate, guiding them, orienting them, and orchestrating the construction of their baby. It is no exaggeration to say that the final organism reproduced is partly under the control of the egg cell, and out of the control of the genes. The state of the egg cell can be affected by stress, age, nutrition, etc. (There is one claim that Down's Syndrome, common in babies born to older women, happens because the two chromosomes responsible for the birth defect become physically entangled by lying so close to each other for so many years in the mother's egg cell.) Even before you are born -- indeed from the moments of conception onward -- forces outside of your genetic information form you genetically. Hereditary information does not exist independently of its embodiment. The origin of an organism's inheritable body, or morphogenesis, is due then to a partnership of nongenetic cell material and hereditary genes -- body and genes. Evolution theory, and in particular evolutionary genetics, cannot understand evolution in full unless it remembers the complicated morphology of life. Artificial evolution will only take off when it is embodied.

Each biological egg cell, like most nucleated cells, carries several libraries of DNA information outside of the chromosomes. Most disturbing to standard theory, the egg cell may be constantly swapping bits of code within itself, between the files of its in-house DNA and the files of inherited chromosomal DNA. If information in the house DNA could be shaped by the experience of the egg cell, then transmitted to the chromosomal DNA, it would transgress the stern Central Dogma, which states that in biology information can only flow from the genes to the cellular body -- not vice versa. That is, there is no direct feedback from the body (phenotype) to the gene (genotype). We should be suspicious of any rule such as the Central Dogma, Darwinian critic Arthur Koestler pointed out, because "it would be the only example found in nature of a biological process devoid of feedback."

There are two lessons in morphogenesis for creators of artificial evolution. The first is that changes in an adult organism are triggered in embryos indirectly through the environment of the mother's egg, as well as directly by genealogy. There is plenty of room in this process for unconventional information flow from the cell (the mother's cell) to the genes via control factors and intracellular DNA swap. As German morphologist Rupert Riedl puts it, "Neolamarckism postulates that there is direct feedback. Neodarwinism postulates that there is no feedback. Both are mistaken. Truth lies in the middle. There is feedback but it is not direct." One major route for indirect feedback to the genes is the very early stages of embryonic growth, the hours of incarnation when the genes become flesh.

During these hours, the embryo is an amplifier. Hence the second lesson: Small changes can be magnified as development unfolds. In this way, morphogenesis skips Darwinian gradualism. This point was made by the Berkeley geneticist Richard Goldschmidt, whose ideas on nongradual evolution were derided and scorned throughout his life. His major work, The Material Basis of Evolution (1940), was dismissed as near-crackpot until Steven Jay Gould began a campaign to resurrect his ideas in the 1970s. Goldschmidt's title mirrors a theme of mine here: that evolution is an intermingling of material and information, and that genetic logic cannot be divorced from the laws of material form in which it dwells. (An extrapolation of this idea would be that artificial evolution will run slightly differently from natural evolution as long as it is embedded on a different substrate.)

Goldschmidt spent a unrewarded lifetime showing that extrapolating the gradual transitions of microevolution (red rose to yellow rose) could not explain macroevolution (worm to snake). Instead, he postulated from his work on developing insects that evolution proceeded by jumps. A small change made early in development would lead to a large change -- a monster -- at the adult stage. Most radically altered forms would abort, but once in a while, large change would cohere and a hopeful monster would be born. The hopeful monster would have a full wing, say, instead of the half-winged intermediate form Darwinian theory demanded. Organisms could arrive fully formed in niches that a series of partially formed transitional species would never get to. The appearance of hopeful monsters would also explain the real absence of transitional forms in fossil lineages.

Goldschmidt made the intriguing claim that his hopeful monsters could most easily be generated by small shifts in developmental timing. He found "rate genes" that controlled the timing of local growth and differentiation processes. For instance, a tweak in the gene controlling the rates of pigmentation would produce caterpillars of wildly different color patterns. As his champion Gould writes, "Small changes early in embryology accumulate through growth to yield profound differences among adults....Indeed, if we do not invoke discontinuous change by small alterations in rates of development, I do not see how most major evolutionary transitions can be accomplished at all."