Peripatus Home Page pix1Black.gif (807 bytes) Evolution >> Evolution - Introduction and ProcessesUpdated: 13-Jan-2017 

Evolution - Introduction and Processes


Notes on the process of evolution. Keywords: evolution, variation, selection, natural selection, speciation, macroevolution


Living things are observed to change in important and sometimes fundamental ways over the course of many generations. If all we are seeking is a definition for a word, then this is what evolution means.

But a deeper understanding must at least attempt to account for the detailed patterns of change observed on all timescales, including the very long timescale of the fossil record, to provide a mechanism for such change, and explain how and why it works.

xxxx something about Darwin’s original notions xxxx

‘Descent with modification’ is the unifying theory of all biology. In the famous words of Theodosius Dobzhansky, “nothing in biology makes sense except in the light of evolution” (ref).

Evolution by natural selection leads to improved adaptation: not to perfection. There is an old joke about two explorers suddenly confronted by a lion. One says, resignedly, that there is no point in running, because the lion can easily outrun both of them. But the other, who has already taken to his heels, shouts back over his shoulder, “I don’t need to outrun the lion. I only need to outrun you.”

And that is how that works. To obtain an advantage, one does not need to be perfect; only better than the ‘competition’.

Related Topics

Further Reading
  • R. Dawkins (1989): The selfish gene (second edition).
  • S.J. Gould (1977): Ever since Darwin.
  • E. Mayr (2002): What evolution is.
  • Related Pages Other Web Sites

    Patterns in Evolution

    A number of evolutionary “patterns” are revealed by both paleontological and neontological studies. A complete theory of evolution must adequately explain them all.

    Patterns Observed in the Fossil Record

    From Thompson 1998, p. 331: “Because of the nature of the fossil record, most of the comparitive estimates across different timescales have used rates of change based on morphological characters. These show an inverse relationship between the measured rate and the timescale on which the rate is based – fast evolution over short timescales and slow evolution over long timescales.”

    Ibid.: “... rapid evolution over thousands of years is invisible in the fossil record. Organisms that evolved this rapidly would probably not be recognized as the same species in different geologic strata.”

    Ibid.: “Values for long term rates are for sustained directional selection. Estimates of short-term rates also evaluate directional selection, but the observed changes might often be part of a longer term pattern of fluctuating natural selection.”



    It must be emphasised that stasis in the fossil record means morphological stasis. Nobody knows what is happening to the genes, although it is expected that genetic mutations are continually accumulating, at a more or less steady rate (the neutral theory). This accumulated genetic change remains invisible at a phenotypic (morphological) level for the most part. If it does manifest itself, the phenotypic consequences, more often than not, are non-adaptive (or even fatal) and the mutation does not establish itself or spread. It has long been conventional theory that the overwhelming majority of such mutations are quickly extinguished by natural selection, and that only once in a very great while does a phenotypic variation arise which is successful enough to become “visible” in the fossil record. This version of events, which is entirely consistent with both the neutral theory of mutation, and the observation of stasis in the fossil record, has been around for decades – certainly long preceding 1972.

    However, we mustn’t become so fixated with stasis as to believe it is the only game in town. Stephen Gould’s (1980, p. 181) observation that the “extreme rarity of transitional forms in the fossil record persists as the trade secret of paleontology” simply isn’t true in all cases – and maybe not even in most cases. Again, part of the problem lies in the lack of rigour in which the debate has traditionally been framed. There is no objective benchmark upon which to base our definitions; this time of what is an “insensible gradation”. Two forms which an expert might immediately recognise as two different species may appear virtually indistinguishable to a non-specialist. [A couple of Beu’s molluscs in a sidebar?] And, in a good many cases, two such species might equally be challenged (e.g. as regional variants) by a different expert, as well. It simply isn’t cut and dried: interpretation and personal preference (e.g. for splitting or lumping) play significant roles. The very existence of the colloquial terms ‘splitter’ and ‘lumper’ tells us that.

    Patterns Observed in the Field and Laboratory

    From Kryazhimskiy et al. 2011, p. 1160: “Microbial evolution experiments in a simple, constant environment reveal a characteristic pattern: At fi rst, a population rapidly acquires benef icial mutations, but then adaptation progressively slows so that thousands of generations pass between subsequent benefi cial substitutions [Elena & Lenski 2003].”

    Ibid.: “Epistasis describes how the fi tness consequence of a mutation depends on the status of the rest of the genome. In one extreme example, called sign epistasis, a mutation may be benefi cial if it arises on one genetic background, but detrimental on another.”

    Evolutionary Processes

    Evolution is a two-phase process, comprising variation and selection.


    Life will be easier if we forget about meiosis for the moment; we’ll return to it all in good time.


    Darwin didn’t know anything about genetics, but his theory requires that selected variations are inherited. Selection acts upon variations no matter how they arise – developmental, nutritional, environmental – whatever. But only if the variable characteristic is passed on can it give rise to a systemic drift in response to the selecting pressure.

    The unit of selection is typically the individual: this is the “thing” which lives, dies, and possibly reproduces. However, various traits are “bundled” in the individual. Traits are not selected separately: the whole organism lives or dies, reproduces or not. My genes for outrunning lions (or at least other explorers) will not be passed on to my offspring if my heart is defective and I die at birth.

    Although the unit of selection is the individual, it is not the individual that is passed on to subsequent generations. It is their genes which are passed on (or their sister’s genes, in the case of social insects such as ants and bees). This point seems very obvious, but many of the implications are subtle and were rather poorly appreciated prior to the publication of Richard Dawkins’s The Selfish Gene in 1976. Not all of the contentions in this book are universally accepted, but there can be no doubt that this aspect of inheritance is no longer passed over so readily.

    Paleontology generally concerns itself with morphospecies: groups of individuals which are morphologically indistinguishable. (Nobody pretends we know the genetic basis for these units, except in a very few, special cases.) Perhaps because of this, evolutionary studies have tended to revolve around the concept of a species also. This approach is mistaken, in my view, because the very concept of what is a species is an artificial, human construct. Certainly, most of the time it is easy and unambiguous to say a person is a person is a person, but what about a duck? Lots of animals look like ducks, waddle like ducks, and quack like ducks, but does that mean they are one species? Well, that all depends upon how we define “species” and that is not nearly so obvious.

    The Species Concept

    Three main conceptual themes: morphological, biological, phylogenetic.

    Despite difficulties and practical constraints, “many practising taxonomists try hard to ensure that the species they recognize are the products of evolution, although this is often not made explicit” (Gornall 1997, p. 184).

    According to Davis & Nixon 1992, phylogenetic species delimited as the smallest groups that can be recognised on the basis of unique combinations of attributes can never be monophyletic in the sense of Hennig 1966, partly because there is no discoverable subordinate hierarchy, and partly because there are difficulties concerning the concept of the most recent common ancestor at this level. “A properly argued extension of the Hennigian definition of monophyly to the species leval and below is clearly needed” (Gornall 1997, p. 184).

    “Although not strictly a problem for the phylogenetic species concept itself, the occurrence of hybridization may present problems in regard to the cladistic methodology and analysis which the phylogenetic approach often involves. There is considerable evidence that very many [plant] species may have a hybrid origin and that much plant evolution may be reticulate rather than cladistic. ... [A] particularly impressive example of reticulate evolution involving allopolyploidy is shown by american members of the Polypodium vulgare complex (Haufler et al. 1995a, 1995b). It is also notable that many allopolyploids have evolved more than once, sometimes in well separated places ... and therefore may be described as polyphyletic when considered at a population level. Since it has been estimated that 70-80% of angiosperm species are of polyploid origin (Goldblatt 1980; Lewis 1980), it is a mode of speciation which is not only sympatric but also possibly represents the most common form in flowering plants.” (Gornall 1997, p. 184-185).

    Include the link to the University of California at Davis Species Concept page.

    Allopatric and Sympatric Evolution

    include some very simple definitions

    include the bit from the AAAS ‘Decade’ magazine, p. 43

    The Fossil Record

    We do not need fossils to ‘prove’ evolution. The biochemical and structural commonality of life is obvious in every organism, and at every scale, and at every level of complexity. Darwin knew a fair bit about geology, but he deduced his famous theory from living things.

    Fossils are essential to evolutionary studies, however. Without them, we could not be aware of the true diversity of life; of extinct forms. No amount of gene sequencing will ever reveal there were once such things as dinosaurs, or trilobites, or giant lycopod trees, or even the strange, umbrella-shaped protists we call Kakabekia which lived more than two billion years ago.

    Without accurately dated fossils, nor could we know when today’s lineages diverged. (Much nonsense is spouted, today, about molecular dating. The substitution rates from which these dates are initially calibrated are derived from fossil evidence, and fossils provide reality checks to test whether calculated divergence times are credible. Doubtless these molecular techniques will improve greatly in time to come, but the track record so far has been spotty, to say the least.)

    Include something about how molecular clock dating has been over-hyped – perhaps Rebecca Cann’s drivel from Scientific American – she deserves a degree of ridicule for that.

    In the words of Richard Fortey, “Fossils provide evidence of when, how, and why history happened” (Fortey 2011, p. 280).

    Living Fossils

    The principal misunderstanding to deal with, here, is that no living thing is “more evolved” than any other. Whether bacterium or human being, both share a common ancestor at least 3.4 billion years ago, and both have been evolving ever since.

    Truly, there is a fascination about certain organisms – a brachiopod called Lingula, horseshoe crabs, coelacanths, the New Zealand tuatara, ... – which seem little changed from their fossilised ancestors, but this is really just a conceit. To quote from Richard Fortey’s book, Survivors “... I have been rather cautious about using the tag ‘living fossil’ too readily. Just as it is untrue that evolution erases the past, it is also the case that no organism remains completely unchanged through long periods of geological time. The horseshoe crab may carry an ancient carapace on its back [he’s speaking figuratively here, people], but it has still moved with the ages. Even the tuatara has evolved at the genomic level for all that its obscurely smiling visage seems to speak directly of the Triassic” (Fortey 2011, p. 280-281).

    The Pace of Evolution

    There is an interesting and revealing passage in Stephen Gould’s (1980) book, The Panda’s Thumb, where he remarks upon “The conflict between adherents of rapid and gradual [evolutionary] change...” (chapter 17, p. 180). I am perplexed as to why Gould, a man of no small intellect or learning, should use the word ‘gradual’ as an antithesis to ‘rapid’. Surely the opposite of rapid is ‘slow’? The word gradual means, or can mean, in small increments, and a writer as adept as Gould would surely have been aware of the subtle trick he was pulling.

    We shall employ no such sleight of hand, and will explicitly recognise that we have two quite separate and independent things to consider: the speed of evolutionary change per se, and also whether this change occurs in discrete steps or is smoothly continuous.

    The Speed or Tempo of Evolutionary Change

    Nobody doubts, now, that in some circumstances evolution can proceed very rapidly indeed. Many studies have been published describing significant phenotypic change in a wide range of organisms, including bacteria, gastropods (Seeley 1986), guppies, birds, and introduced plants (Buswell et al. 2010), in timescales as litte as 100 years. Moreover, rapid evolutionary changes can have important ecological consequences as well (Thompson 1998).

    Continuous or Quantised?

    Again, today there remains little doubt about the basic facts: genes are made of molecules and DNA can’t incorporate half a molecule. Substitions involve whole molecules so, by definition, they are quantised. But what does this really mean? After all, data on the familiar CD or DVD is also quantised (we call it ‘digital’) and yet the effect we perceive is one of smooth continuity.

    Virtually every birth of sexually reproducing organisms represents at least a unique combination of alleles. No organism has such a simple genome – one so lacking in possible combinations – that any significant duplication comes about through chance alone.

    A major mutation, producing an offspring which is significantly different from its parents, is called a saltation. It is not known what contribution, if any, saltations have made to evolution, overall. Certainly they are unlikely to play any significant role in the evolution of sexually reproducing organisms, because any major genomic departure from the parent population is likely to result in reproductive isolation: the individual cannot reproduce alone, and, being unique, there is no partner with which it will be fertile.

    So far we have talked about genetic change. What about morphological change? At this point it is probably worth recalling to mind the oft-cited relationship between genetic and morphological differences in chimpanzees and humans: We differ by only about 1% in the nucleotide bases that can be aligned between our two species [??and the average protein differs by fewer than two amino acids??]. But a large amount of non-coding material is either inserted of deleted in the chimpanzee genome, as compared to the human, bringing the total difference in DNA between the two species to about 4%. (AAAS ‘Decade’ magazine, p. 42, and maybe hunt down the primary sources from 2005 as well) Even so, 4% is rather little to effect what we would consider to be substantial change: to appears that a little genetic mutation can go a long way.

    Anagenesis, Speciation, and Extinction


    Maybe better under ‘Pace’: Some evolutionary biologists believe that morphological breaks in the fossil record represent speciation events, and suggest that natural selection plays only a minor role in these rapid transitions. Others maintain that natural selection may cause rapid transitions as well as slow and gradual change and that no new mechanisms need be invoked to explain punctuated morphological evolution (adapted from Seeley 1986, p. 6897).

    “These data support the view that morphological transitions that appear abrupt in the fossil record may be a product of classical Darwinian selection and should not be assumed to represent speciation” (Seeley 1986, p. 6897).



    It is popular nowadays to emphasise the role of blind luck (’contingency’) in extinction. To an extent, this is both undeniable and trivial, especially in the context of mass extinctions. If a meterorite falls out of the sky and wipes out a localised taxon, that surely is bad luck: no amount of adaptation is going to cope with that.

    But it should not be forgotten that most organisms have gone extinct in other circumstances. Yes, a meteorite may have been instrumental (though just how much is still debated) in wiping out the last of the dinosaurs. However, it should not be forgotten that the great majority of “the dinosaurs” evolved and went extinct long before the end of the Cretaceous. Jurassic forms – Allosaurus, Apatosaurus (=”Brontosaurus”), Brachiosaurus, Stegosaurus, and all the rest – were 80 million years dead before the end-Cretaceous event, whatever it was.

    Most organisms appear to go extinct because they are out-competed or fail to adapt to climate change. Consider the panda, for example. Certainly, human beings may well be responsible for hurrying the creature along to an eventual demise, but any analysis of the creature’s ecology shows an organism with little capability to cope with any perturbation in its environment.

    Outstanding Problems

    Punctuated Evolution


    However, punctuated equilibrium does not advocate any important role to saltation: “In describing the speciation of peripheral isolates as very rapid, I speak as a geologist. The process may take hundreds, even thousands of years; you might see nothing if you stared at speciating bees on a tree for your entire lifetime. But a thousand years is a tiny fraction of one percent of the average duration for most fossil invertebrate species – 5 to 10 million years. Geologists can rarely resolve so short an interval at all; we tend to treat it as a moment” (Gould 1980, p. 184).

    Some evolutionists imagine that the recognition of stasis is a relatively new development, many crediting the concept to Eldredge & Gould 1972. Nothing could be further from the truth. Biostratigraphers have been utilising the concept of stasis since before the publication of Origin. Indeed, almost all of the finer divisions of the geological time scale (ages) are based upon zones defined by the persistent presence of one or another taxon over an extended time range. For example, the age/stage division of the Cambrian through Silurian Periods was completed by Sedgwick and Murchison in the 1830s. Indeed, one of the preferred characteristics for a good index fossil is that its stratigraphic range should not be too long. Even the great period divisions themselves may implicitly recognise the persistence of higher taxa; who does not know that the Cretaceous-Paleogene boundary brought an end to the “age of dinosaurs?”

    Stasis is an old idea.

    The Role of Determinism in Evolution

    It is common to describe evolution as completely ‘random’. Authorities as diverse (in every sense of the word) as Matt Ridley and Stephen Jay Gould make statements such as “Evolution has no pinnacle and there is no such thing as evolutionary progress” (Ridley 1999, p. 24). But, of course, that is nonsense.

    In the sense that it has a direction (actually several), as opposed to being guided by some supernatural agency, then evolution is indeed directed. It does, for example, proceed from simple to complex – from primordial slime to hormonally-fuelled teen angst – though this may not always seem like an “advance” in the usual sense. This is elementary reasoning: complex things do not spring into existence, fully formed, without antecedents. They are built from simple components.

    But some kind of direction or polarity isn’t the only form of determinism we expect from evolution. We obviously expect the overall adaptation of any lineage to a given environment, to improve over time, until it reaches an equilibrium point where further accommodation to the environment is offset by the biological costs. Writ large, this means that some evolutionary innovations are not only predictable, but are likely to recur in diverse lineages. Mobility, multicellularity, heterotrophy, symbiosis, the acquisition of skeletal elements for support, and the cooption of hard parts for defence, vision, ... the list goes on. And not only is it possible to make such a list, the sequence in which such innovations must arise is also predictable.

    In the foregoing, the word ‘predictable’ is intended in a probabilistic sense. I do not resile from it; I sincerely believe that it was inevitable that some form of vision, for example, should have evolved not just once, but many times, independently, among vagile organisms. Not only that, but the broad range of wavelengths that animal eyes respond to is readily explicable to any physicist. The exact when, where, and why, however, are beyond us, at least for the present: there are too many subtlties. For example, a study of the common bacterium Escherichia coli found that two clones having the beneficial mutation which eventually took over the experimental population, initially showed significantly lower competitive fitness than other clones having mutation which later went extinct. Repeated experiments, replaying evolution many times, led to the belief that the initially less competitive clones eventually prevailed because they had a greater potential for further adaptation (Woods et al. 2011).

    My preferred analogy for evolutionary “progress” is the two-gases-in-a-divided-box demonstration of entropy: although each individual interaction is purely random, in aggregate the result is as inevitable as it is irreversible. Certainly there are examples of simplification – a flea has lost its wings, a tapeworm has lost its gut – but it is not accidentally that we call such characters ‘derived’ and claim that possession of wings or gut are ‘primitive’ for these organisms. Simplifications can represent progress, too. Moreover, in many such cases, it is uncertain that the organisms’ genomes are “simplified” in any meaningful sense, even if their bodies seem to be.

    Re-read and reference Arthur’s Creatures of Accident.

    However, this most emphatically does not mean that evolution somehow “knows” where it is headed. It is quite possible for evolution to “steer” an organism into a blind alley which ultimately proves to be non-adaptive.

    I am often tempted to think that these people say such silly things largely in reaction to the ‘Scientific’ Creationism brigade. However, this misnamed movement is really only of any significance in the United States, and then only on account of its political influence; hardly through any intellectual virtue. I feel one would have better luck addressing oneself to the Taleban than to bible-belt America, so, if our American colleagues must continue fighting this fight in order to secure their funding streams, then we must surely sympathise with them. But, for the rest of us, the evolution debate was fought and won decades ago, and it is time to move on.


    Sometimes macroevolution is defined simply as evolution of taxa above the level of species. Aside from being worryingly imprecise (see the discussion of species concepts, above), this is not very edifying. In the words of Jeffrey Levinton, it omits the most interesting stuff.

    The really good question, here, is whether macroevolution is simply the outcome of “ordinary” evolution (microevolution) carried on for a sufficiently long time, or whether, as some have suggested, there are actually some different processes at work.

    In the interests of full disclosure, I’ll declare my hand immediately: I don’t think there are. Some examples of “higher selection” commonly advanced seem flawed to me:

      It is sometimes suggested that species with a broad geographic range are less vulnerable to extinction; if they are wiped out in some part of their range, they are more likely to survive in some refuge or another if they are very widely distributed. I feel this argument is flawed because it assumes an essentially random (“field of bullets”) extinction scenario, rather than an event which attacks an ecological niche or food source required by the organism, and also because it is unclear to me how a “species” behaves any differently to a population or any other grouping of individuals in this context.

      It is also sometimes suggested that organisms with “short haul” dispersal mechanisms are more likely to speciate frequently, thereby making the clade as a whole less vulnerable to extinction. This model assumes allopatric speciation which, for some reason, seems to have become the almost exclusive fixation of contemporary evolutionary theorists ever since the dubious contribution of Eldredge & Gould 1972. Annoying though that may be, it does not invalidate the example, because nobody doubts that allopatric speciation is important, at least for multicellular, sexually reproducing organisms (by no means the majority of life on Earth). I am more concerned about the arbitrary definition of “the” clade under consideration. Don’t misunderstand me: unlike “species” or any other taxon, clades are real; they are natural phenomena. I just don’t see the relevance of saying this or that particular clade survived an extinction because it is highly diverse. Most evolution theorists believe that all presently living things are descended from a single common ancestor: be we elephants or pond scum, we are all one clade. Some bits went extinct; the rest of us are still here.

      Another supposedly “species-level” property is the potential for further adaptation. The study in which initially less competitive clones of Escherichia coli eventually took over the experimental population (Woods et al. 2011) may seem like an example of this. Again, I am unconvinced. The potential for further adaptation lay in the mutations carried by individual members of the two eventually successful clones; not by the clones as a whole.

    However, Wallace Arthur points out an argument arising from a consideration of germline development, which must be answered:

    “This approach has important consequences for the neo-Darwinian view that major morphological differences between organisms from different higher taxa have been produced through the gradual accumulation of very small changes. Specifically, while in a non-developmental approach it seems plausible that many small changes can indeed accumulate to give a larger one, in a developmentally explicit approach it is clear that many late changes can not accumulate to give an early one. Thus if taxonomically distant organisms differ right back to their early embryogenesis, as is often the case, the mutations involved in their evolutionary divergence did not involve the same genes as those involved in the typical speciation event, where usually the early embryogeneses of the daughter species concerned are virtually identical.

    “There are, of course, all sorts of complications to the argument, including the fact that some development-controlling genes make diffusible morphogens. The morphological effects of such genes will be much more widespread than those of cell autonomous genes” (Arthur 1997, p. 22).



    The general tenor of this discussion has been to support the sufficiency of orthodox evolutionary theory.

    The phenomena which need to be explained by a credible theory of evolution include anagenesis and stasis in the fossil record, as well as the sometimes abrupt appearance of new taxa which are notably different from any antecedents. I contend that the foregoing discussion of the pace of evolution, which encompasses both stasis and change, sometimes at genuinely rapid rates, together with a clear understanding of how sediments form, and a sensible acknowledgement that no package of rocks represents every moment of its deposition, fully explains the fossil record. There is no need to appeal to mysterious (or at least abyssmally articulated) alternative theories (such as PE) or speculative selection mechanisms (such as species sorting). There may yet be something in these ideas, but I believe the burden remains on their proponents to demonstrate their usefulness; not upon conservative Darwinian theorists to “disprove” them.


    No reference data for [Gould 1980] .

    No reference data for [Lewis 1980] .

    No reference data for [Seeley 1986] .

    No reference data for [Thompson 1998] .

    No reference data for [Ridley 1999] .

    No reference data for [Fortey 2011] .

    No reference data for [Kryazhimskiy et al. 2011] .

    No reference data for [Woods et al. 2011] .

    Arthur, W. 1997: The Origin of Animal Body Plans. Cambridge: 1-338.

    Davis, J.I.; Nixon, K.C. 1992: Populations, genetic variation, and the delimitation of phylogenetic species. Systematic Biology 41: 421-435.

    Dawkins, R. 1989: The selfish gene (second edition). .

    Eldredge, N.; Gould, S.J. 1972: Punctuated Equilibria: An Alternative to Phyletic Gradualism. In Schopf, T.J.M. (ed.) 1972: Models of Paleobiology. Freeman, Cooper and Co., San Francisco. : 82-115.

    Elena, S.F.; Lenski, R.E. 2003: Evolution experiments with microorganisms: the dynamics and genetic bases of adaptation. Nature Reviews Genetics 4: 457-469.

    Goldblatt, P. 1980: Polyploidy in angiosperms: monocotyledons. In Lewis, W.H. (ed.) 1980: Polyploidy: biological relevance. Plenum Press, New York. : 219-240.

    Gornall, R.J. 1997: Practical aspects of the species concept in plants. In Claridge, M.F.; Dawah, H.A.; Wilson, M.R. (ed.) 1997: Species – the units of biodiversity. Systematics Association Special Volume Series 54. : 171-190.

    Gould, S.J. 1977: Ever since Darwin. .

    Haufler, C.H.; Soltis, D.E.; Soltis, P.S. 1995a: Phylogeny of the Polypodium vulgare complex: insights from chloroplast DNA restriction site data. Systematic Botany 20: 110-119.

    Haufler, C.H.; Windham, M.D.; Rabe, E.W. 1995b: Reticulate evolution in the Polypodium vulgare complex. Systematic Botany 20: 89-109.

    Hennig, W. 1966: Phylogenetic systematics. University of Illinois Press, Urbana.

    Mayr, E. 2002: What evolution is. .

     Peripatus Home Page pix1Black.gif (807 bytes) Evolution >> Evolution - Introduction and Processes