Natural selection does not explain all traits of all organismsby John Tyler Bonner / August 21, 2013 / Leave a comment
Published in September 2013 issue of Prospect Magazine
It is hard to explain the huge variety of diatoms—a microorganism that has 100,000 species—in terms of natural selection, says Bonner There are good reasons why natural selection has become widely accepted as an explanation of evolutionary development. When applied to mammals and other large animals, it fits perfectly. But we cannot assume that all evolutionary steps arise from selection, particularly when looking at smaller animals. The reason for natural selection’s great success is that it provides a satisfying explanation of how evolution might have occurred: individual organisms vary and if those variations are inherited, the successful ones will survive and propagate and pass down their desirable traits to succeeding generations. But this process alone does not explain all of evolution; all it can claim is that it could do so in theory. To argue for its total universality one would have to prove that each instance of a trait that characterises such a step in evolution arose through selection. This is an impossible task. It has been shown that indeed this has occurred in many instances that have been examined—the assumption that it is true for all traits of all organisms is more problematic. What happened in the last century was a deserved blossoming of Darwin and his theory of natural selection, which led to a popular form of ultra-Darwinism: the idea that natural selection did everything. This became the only view—I know this because, along with every other biologist, I succumbed. Evolutionary biology differs substantially from other branches of biology, including my own area of expertise, developmental biology. I do not mean that the objects of study differ; I mean that the attitudes and psychology of the practitioners is different. One generally assumes that all scientists work on the basis of an identical “scientific method,” whereby a scientific statement is accepted or dismissed according to confirmation by experiment. But they do not. Biology is riddled with assumptions, hypotheses, guesses and models, as we try to paint as rounded and complete a picture as possible. Scientists fill in the unknown gaps with conjecture. In today’s literature, one of the most common phrases is, “these data suggest…” which is followed by a guess. All science is expanding and it grows with a mixture of new facts and evolving speculations. The view that natural selection accounts for every feature of all organisms has been much discussed in the literature and has been given many names. For instance, the term “pan-selectionist” denotes those scientists who believe that every trait of an organism is the product of selection. Another class of scientists is termed “adaptionist,” which refers to a similar but subtly different view that all features of an organism are adaptations. This word is used liberally in a well-known essay by two evolutionary biologists at Harvard, Stephen Gould and Richard Lewontin, entitled “A Critique of the Adaptionist Programme.” In the essay, the two vigorously argued that we simply often do not know whether a trait has arisen by selection. One cannot just assume it without becoming a kind of an extremist. The position that Gould and Lewontin attack is sometimes referred to as ultra-Darwinism and is the theory that everything must pass through the filter of natural selection. It is unclear that Darwin himself took this extreme position; he was very sensitive to exceptions. For instance, he puzzled that some lowly microorganisms, such as the foraminifera (left), with their calcareous skeletons—shell-like structures—had remained to some degree unchanged for millions of years as their fossil record shows. (The rock used for building the pyramids of Egypt is made up almost entirely of fossilised deposits of foraminifera.) I began to deviate from the standard ultra-Darwinian position due to my work on microorganisms. The question that I began to ponder was whether these small organisms followed the same rules as larger ones. Almost all scientific literature, from Darwin to the present day, is on the subject of large complex organisms, where natural selection as the agent of change fits like a glove. Was this also the case for very small organisms? I slowly became convinced that the answer to this was no, despite the objections of many friends—some of them first-rate evolutionary biologists—who felt I was either a heretical troublemaker or just plain wrong. My reasons for arriving at this conclusion were strongly influenced by my background in developmental biology. Development consists of a series of sequential biochemical steps. These are often cumulative, meaning that step B cannot occur unless step A was successfully completed. To build a mammal, or a flowering plant, may involve many such steps, perhaps many thousands. Think of all the parts of the human body and the steps that make up an eye, or a heart, and one can see what must go on in each generation. But then, in comparison, consider a unicellular alga whose development consists of the short period between one cell division and the next. There still may be many biochemical steps, but they will be a fraction of that of a developing mammal. The consequence of this is that when a mutation arises in a small organism it will immediately show itself; by comparison in a large organism, only those mutations that occur close to maturity, such as brown eyes or red hair, will be visible and viable. Mutations that affect earlier developmental steps are likely to be lethal, because so many of the subsequent steps will be dependent upon the successful completion of earlier ones. This was pointed out by Lancelot Law White, the Scottish polymath who half a century ago coined the term “internal selection.” Selection occurred not only on the adult in Darwinian natural selection, but also in the embryo by internal selection. This always seemed to me an important point and I never understood why it disappeared into oblivion. Being small makes it more likely that a mutation will directly bring change to an organism. If that change is not immediately rejected by natural selection, then it would remain fixed in the population. Some of the variation in nature, at least in the case of microorganisms, might be due to haphazard random mutation and not natural selection. Is there any evidence that would support such a contention? None that would prove it, but there are facts that are consistent with such a contention. For instance, a number of microorganisms build skeletons or shells that are mineral. Both diatoms and radiolaria build shells of silica and their variety is astounding, not only in the vast differences in their shapes, but also the number of different forms they produce. There are estimated to be roughly 50,000 different species of radiolarians and 100,000 of diatoms. It is very hard to make a convincing argument that they are all separate events that have arisen through natural selection. It is much easier to think that the forms differ through random mutation. This kind of argument hardly comes close to proof, but it does give one pause. And then there are the foraminifera, which like radiolarians live in the sea and have skeletons made of calcium carbonate—there are roughly 250,000 species. All these facts are most easily explained by random mutation followed by little or no natural selection. This in no way denies the great importance of natural selection. It simply says that not all features of all organisms are the product of natural selection, and at least in very small organisms the variety of shapes may be the result of direct, random mutation followed by an absence of selection. The larger an organism the more internal steps leading to its final construction of maturity. Building an organism is an immense enterprise and the bigger the animal or plant the longer it takes. Development can be compared to human constructions: the larger the building, or a boat, the longer it takes to complete all the construction steps. So an outhouse or a rowboat can be built in a day, but it will take closer to a year or more to build a skyscraper or a battleship. Furthermore, in constructing a skyscraper the work cannot begin anywhere, but the construction must be in a particular sequence: one cannot install the floors and the ceilings before putting the steel girders in place; one cannot put in the windows before having the outside walls, and so forth. This means the size of an organism is directly correlated with its ability to avoid natural selection. The smaller the organism, the fewer developmental steps and therefore the greater the chance of producing novel morphological variants, and if those variants are ignored by natural selection they will persist. They are random events that persist because each of the resulting organisms is as fit as the other. Small organisms, with their short developments, will have their mutations near the surface and they will be visible almost immediately. This makes it possible not only to produce great numbers of different shapes, but also to swamp the field with shapes that are not touched by natural selection.