v1.2.2 / chapter 13 of 28 / 01 aug 08 / greg goebel / public domain
* The previous chapters have discussed the origins of Darwin's theory of evolution by natural selection, and surveyed what is know about the history of life. It is worthwhile to now provide a closer examination of Darwinism to examine its validity. The proper place to start is with a review of Darwinism's basic ideas, and of competitors to Darwinism.
* As discussed previously, the notion of evolution preceded Charles Darwin. What he brought to the discussion with publication of THE ORIGIN OF SPECIES in 1859 was not only the idea of natural selection, but also a detailed, extremely well reasoned justification for his theory that made it very difficult to dismiss. Had Charles Darwin not existed the theory would have been emerged anyway, Alfred Russel Wallace having almost beaten the meticulously thorough Darwin to the punch, but it is very difficult to think that the brief essay put forward by Wallace would have had anywhere near the same impact as Darwin's hefty argument.
In review, Darwin pointed out:
Darwin's conclusion was that all the species on the planet had evolved from common ancestors through gradual change, and that it had been done though natural selection: those organisms in a population that featured random changes that helped them to survive tended to be more successful at surviving and raising offspring, gradually pushing out those organisms that weren't as well adapted and didn't breed as successfully. Since Darwin's time, his ideas have been fleshed out, by the development of proper genetics, a much improved knowledge of the fossil record, and an ability to provide reliable dates for fossils.
* Darwin did much to make the concept of evolution respectable, but his mechanism, natural selection, proved harder to sell. Sensible people willing to question what they are told are unlikely to accept evolution by natural selection -- Darwinism -- on first sight. The evidence for it tends towards the indirect, the concept has its seeming implausibilities, and some of its implications are unsettling.
However, nobody has been able to suggest any other mechanistic theory for the evolution of life that has stood up to the available evidence. Lamarckism a popular idea even in the scientific community up to about World War II, but it was effectively dead at that late date and now survives only in a ghostly fashion on the scientific fringe. Lamarckism envisioned adaptations as arising in a "directed" fashion due to the "needs" and "strivings" of organisms, with these adaptations passed down to following generations. The giraffe, in stretching its neck, would get a longer neck, and would pass that longer neck down to its offspring.
In terms of what we know about biology today, such notions are completely unrealistic. An organism's capabilities are essentially fixed by its genome at conception. This is not to say that an organism is as fixed in structure as any machine rolling off the production line, only that the organism is dealt a particular set of cards at the outset and its life will be a game played with those cards. The organism has no way of adding cards to the biological hand it is dealt; its progeny will be dealt a new hand of cards, reshuffled in sexually-reproducing species through sexual recombination, and with occasional variations due to mutation. The parent has little or no control over how the cards are dealt, and the hands passed off to its offspring are not necessarily an improvement on the hand originally dealt to the parent.
Even if an organism did acquire new biological features, it has no means of passing them down to the next generation. The mechanisms by which genes code an organism are complicated enough; it's very difficult to suggest in any detail a mechanism that could "edit" the genome in the germ cells of an organism to track changes in form of the mature organism. This "editor" system would have to have some "blueprint" of how the organism was put together, and would have to "scan" the entire organism to see how it conformed to this blueprint. When the modifications due to the efforts of the organism to, say, stretch its neck were uncovered in the scan, the editor would have to then make the appropriate updates in all the genomes of the germ cells. It has to be noted that the genome is far more like a recipe than it is like a blueprint and the alterations would not be straightforward. This system isn't ruled out by the laws of physics, but it would be extremely complicated in its implementation.
A similar observation could be made about the origins of innate instincts. In Lamarckian terms, instincts in animals arose from behaviors learned by their ancestors and passed down, but that would imply some scheme for translating the memory of the ancestor into genetic codes. In any case, if such an editor system existed, there would certainly be some biochemical evidence for it, but no evidence has been found for a trace of it.
Worse, acquired characteristics are not necessarily good; if an animal loses an eye, how would the editor know this was something undesireable? The editor would have to be able to make value judgements, determining what new features amount to improvements and which amount to defects. In Darwinism, the value judgement is built in: if a change improves the odds of survival and procreation, it's retained, otherwise it dies out.
In addition, while it is possible to imagine a giraffe getting a longer neck by its efforts, anyone who's ever worked as a problem-solver knows that problems in themselves do not necessarily suggest obvious solutions, that obtaining solutions may require considerable mental gymnastics, and there may be different possible solutions. Would the editor dream up a list of different possible solutions, and then perform value judgements to figure out which was best? If it just came up with a solution at random, the scheme would begin to sound Darwinian. Carrying this thought further, how could Lamarckism create a complicated organ like the eye, or even a simple eyespot? How could a blind creature "strive" to see? The editor would have to be capable of considerable imagination.
* There's a bit of fine print in this overall condemnation of Lamarckism. There has been some disputed evidence that acquired immunities in a parent can be passed down to offspring by some mechanism, which doesn't seem at all implausible. If it does happen, it would be an example of limited inheritance of acquired characteristics at work. It is definitely known that a mother will pass down a sample of the bacteria in her gut to her baby at birth, and since the mother may have acquired new and "improved" bacteria since her own birth, that would amount to inheritance of acquired characteristics -- at least through a symbiotic "end run".
Of course, memes amount to inheritance of acquired characteristics, with learning acquired by ancestors passed down to descendants -- but this is blindingly obvious and nobody sees it as a challenge to Darwinism. Memes aren't an irrelevant consideration by any means, human culture being a clearly important subject, but in strictly biological terms, they're outside of the domain of concern. Nobody's seriously claiming that Lamarckism is right after all. Even adding these conditions, it's still mostly wrong.
* As far as saltationism goes, it amounts to the idea that an entirely new and viable species, involving significant changes, could arise in a single generation. There is a simple and immediate objection to the idea: where is an organism with sexual reproduction that has emerged in one big step going to find a mate?
Saltationism also suffers from subtler problems. Establishing a new species requires a single major genetic change or a set of minor genetic changes. As far as major genetic changes go, there are certainly "macromutations" that can have highly significant effects, but such macromutations tend to produce freaks -- for example, flies that have legs growing where their parents had their antennas.
Ronald Fisher pointed out that any viable species is a complex, well-tuned organism; the odds of a tweaky little random change -- such as an increment in size or minor change in coloration -- of improving the tuning is, if by no means certain, fair to good. In fact, if a color change due to a mutation can be either lighter or darker and the darker color change provides a selective advantage, then the odds of obtaining an advantage in this particular case are 50:50.
In contrast, the likelihood of a major random change amounting to an improvement, a "hopeful monster", is very low. Completely changing the coloration of an animal at random is unlikely to improve its camouflage. A simple doubling in size, without the needed complementary modifications of the organism, is not likely to give it any advantage -- human giants, the result of thyroid gland malfunction, are impressive but live uncomfortable and short lives, humans not being adapted to be that big. Taking a small step in the dark is likely to work out much better than a big leap.
As a rule, a new species arises through a set of genetic changes. However, these genetic changes have to be tuned by natural selection. The likelihood of having a large number of simultaneous genetic changes that produces a viable organism greatly different from its parent is vanishingly small. As Richard Dawkins famously put it: "However many ways there are of being alive, it is certain that there are vastly more ways of being dead."
Darwin himself rejected saltationism by famously saying chapter 14 of THE ORIGIN OF SPECIES that "natura non facit saltum" -- "nature does not make leaps". Saltationism is like rolling a hundred dice in a shaker and expecting to have them all come up 6. This would happen on the average on an interval that would make the lifetime of the Universe seem like the blink of an eye. Random changes could only produce a viable organism by individual small changes, each tuned by natural selection and accumulating over time. Darwinism is more like rolling hundred dice and setting aside every die that comes up 6, then rolling the remaining dice again until there are no dice left. This might take an hour.
* Incidentally, when Darwin commented that "nature non facit saltum", in his admitted ignorance of genetic processes he took it to something of an extreme. Darwin all but banished mutational processes or "sports" to the margins, believing that natural selection worked primarily on existing variations in a population and that the accumulation of such variations in the population by mutation was a secondary if still necessary process.
The early detailing of mutational processes by John Hunt Morgan and his people led to a dispute with Ernst Mayr and others. Mayr's camp insisted that existing variation in a population was the feedstock of natural selection, while Morgan and his camp emphasized the importance of new mutations in the process. Continued research would generally vindicate Morgan, and today the importance of new mutations in evolutionary change is essentially dogma.
The irony is that Morgan is still widely perceived as being on the blindly wrongheaded side of the argument, thanks to misrepresentations by Mayr's camp that claimed Morgan simply discarded selection completely and asserted evolution resulted strictly from sequences of mutations. In other words, they portrayed Morgan as proposing nothing more than a variation on saltationism, with the lucky set of mutations occurring in sequence instead of all at once. In reality, Morgan never proposed any such thing -- but sneers at his foolish "mutationism" still linger in the literature, even in the writings of Richard Dawkins, despite the fact that it never really existed.
* In any case, as with Lamarckism, there is some fine print to the argument against saltationism. If the probability of a drastic macromutation being beneficial is slight, that means that there's still a long-odds bet that it actually will be. Functional macromutations may well be a very infrequent but potentially significant feature in the evolution of life -- every rare now and then, the Darwinbot will be able to jump and end up on higher ground in the fitness landscape. Some believe that it was such rare evolutionary leaps -- the bigger the leap, the more rare its occurrence -- that were responsible for some of the major innovations in the evolution of life.
There are also some mutations with large-scale effects that aren't all that troublesome and which are clearly demonstrated by the evidence. As discussed later, there are master control genes that direct the construction of an organism from its various structural "subunits", and sometimes they can increase or decrease the number of subunits -- for example add or delete a new set of ribs. Snakes have hundreds of ribs, far more than their four-legged lizard ancestors, and obtained these ribs through mutations in control genes that specified the construction of more rib subunits. Richard Dawkins calls such mutations "stretched jetliner" mutations, along the lines of the way a jetliner is redesigned for higher capacity simply by adding a fuselage "plug" before and after the wings.
As Dawkins points out, in such mutations the basic arrangement of the organism is retained, just as it is in a stretched jetliner, with no requirement for any major reorganization of the organism to compensate for the change. In fact, the discovery of master control genes has suggested that evolutionary modifications along the general lines of "stretched jetliner" mutations are actually more general and important than previously thought,
Saltationism is still mostly wrong in any case -- natural selection remains in effect, in fact the screening of macromutations by natural selection is very severe, with the disastrous macromutations weeded out by immediate doom. Certainly nobody is proposing that one species gives birth to an entirely new species. The importance of occasional macromutations in the evolution of life on Earth remains argued; but even if such a notion was widely accepted, Darwin's theory would be able to accommodate the idea of rare useful macromutations with no major adjustment.
Incidentally, it is true that polyploid plant hybrids are a new species born in one generation, since with their merged sets of chromosomes they are no longer able to reproduce with either of their parent species. However, although much was made of polyploid hybrids by some evolutionists in the past, they are now generally regarded as an interesting but peripheral issue. Saltationism now only survives in comics, with mutations providing impossible super-powers for bands of "X-Men" and their supervillain adversaries. It seems appropriate to keep it there.
* A few other alternative mechanistic evolutionary theories have been proposed but have failed to impress anyone; they need not be discussed here. There's nothing to rule out some new mechanistic theory coming down the road in the future, but it's difficult to discuss a theory that doesn't exist yet. This means that the only active competitor to Darwinism at this time is the non-mechanistic "argument for Design" -- that the organisms on the Earth were actually created by some higher power, a "Designer". Advocates of Design generally associate the Designer with God, but some are willing to leave the Designer(s) undefined -- for example, they might be superpowerful aliens who have taken on the form of white mice.
* A good case can be made for Darwinism. All dogs are genetically wolves; human selective breeding has demonstrated the mutability of species by generating a vast range of forms from the wolf template, ending up with creatures, such as pekinese, that nobody would mistake for wolves. Crop plants have in some cases gone through similar drastic modifications: it is startling to realize that a few thousand years ago, corn looked pretty much like an ordinary sort of grass. Now it features hugely distorted heads with heavy stalks to support them. It can no longer disperse seeds on its own and would die out in a year or two if humans didn't take care of it.
A Russian researcher named Dmitri K. Belyaev worked at a silver fox farm breed more docile foxes, and within twenty years came up with animals that clearly seemed more like dogs than foxes in their appearance and behavior. It appears that a simple change in thyroid action was responsible. Dogs are generally regarded as "neotonous", in other words retaining as adults characteristics of appearance and behavior associated in the cubs of their wolf progenitors, and a similar inability to "grow up" could produce the tame silver foxes. It is said that Belyaev's successors at the farm then tried to breed foxes for savagery, and ended up with paranoid beasts that snarled and leaped to the attack whenever anyone approached them.
* Of course, these are examples of artificial selection, but all natural selection amounts to is saying that the culling of a species by environmental pressures could produce the same sort of results. Species are so mutable that zoo-keepers trying to preserve rare animals find it difficult to maintain captives that really match their wild cousins. Animals that are happy with being captives tend to breed much more easily than those that aren't, and so zoo animals tend to become increasingly tame from one generation to the next -- which still doesn't mean that it's a good idea to walk into a tiger cage.
Consider an experiment: set up a sterile lab environment to prevent contamination, take a culture of harmless bacteria, and then douse it with a toxin that kills the culture off 100% -- no survivors. Repeat this action as many times as desired and get the same results. Now take ten cultures of the same bacteria and administer the toxin in graded doses, from a very small dose to the maximum dose. Take the culture that was given the biggest dose for which some of the bacteria survived, then use these survivors to create ten more cultures, which are given the same treatment. Keep repeating this procedure and eventually the result will be cultures of bacteria that shrug off the maximum dose of the toxin that would have killed any of its ancestors.
This is a lab experiment and still somewhat artificial -- though the only way in which it actually differs from the evolution of bacteria in response to a toxin in the "wild" is that the lab experiment screens out confounding influences and makes the process easier to observe. There is no credible doubt that the same processes occurs in nature, the most notorious example being the emergence of "antibiotic-resistant bacteria". The introduction of antibiotic drugs in the middle of the 20th century provided medicine with a powerful set of weapons against dangerous bacterial infections, but even at the time the inventors of antibiotics knew that bacteria would evolve to defeat the antibiotics as they were, and now we are suffering from an ever-rising tide of bacteria that shrug off drugs that would have killed them off neatly thirty years ago.
Another interesting example was the discovery in 1975 of a bacteria that could digest nylon, using an enzyme of course named "nylonase". Since nylon is a completely synthetic material, it would be hard to argue that this capability had been latent in the bacteria; a chance mutation allowed the bacteria to digest nylon, opening up a new "ecological niche" for the bacteria to grow and prosper. Bacteria were later found that could digest polychlorinated biphenyls (PCBs).
As far as more complicated organisms go, insects are well-known to have quickly evolved resistance to pesticides like DDT. For another example, metal electric power towers that are clad in zinc to resist corrosion will form zinc deposits that kill normal grasses and other plants. In fact, most such plants grow perfectly well around the towers -- but on examination they are strains that can tolerate high levels of zinc. Try to bring in plants that grew up far away from a tower and they will die.
There is also the famous example of the rise and fall of the dark form of the British peppered moth relative to urban industrial pollution. We see evolution by natural selection happening around us all the time.
