A scientific critique of the theory of evolution

The theory of evolution was formulated in the 19th century, based on natural selection acting on variations of morphological characteristics, such as the size or shape of a bird’s beak. In so far as such variations arise from existing genetic variability (which aspect became evident in the 20th century) evolution works at this level. Such was the success of this evolutionary model, that it was assumed it could also account for the origin of genetic variability, i.e. genes, as well. However, as we began to elucidate the molecular nature and workings of genes, and of biology generally, it became clear that this extrapolation does not work. And the more we discover about the genetic mechanisms of embryonic development, the clearer it is that we can no longer think of evolutionary scenarios simply in terms of morphological variations, but we must consider the genetic and other biochemical implications; and when we do so, such scenarios fail.

The aim of this website is to explain the difference between evolution that does and does not occur. Here is a brief summary; follow the links and use the navigation menu to see more detail.

Evolution works with existing genetic variability

Most species have extensive genetic variability which, in the course of reproduction, enables considerable morphological variation. Some variations are advantageous such that individuals having them in general survive and reproduce preferentially. This is natural selection.

The successive production of variations and selection of advantageous ones can lead to significant changes in the form of a species, such as adaptation to the environment, which is one aspect of evolution.

Also, because different variations can be advantageous in different environments, different variations can be favoured in different parts of a species’ range, which can lead to the production of different forms or races of that species. (This is similar to the production of different breeds through artificial selection.) And in due course this can lead to distinct species, which is another aspect of evolution. Where a species splits in this way it is called speciation.

Development of substantial new features requires new genetic material

Because of the evident success of this evolutionary process at the morphological level it has been assumed that it can go on more-or-less indefinitely, to the extent that it is thought substantial new features can arise through the gradual morphological modification of existing ones. Or, to put it retrospectively, it is thought that features such as wings, eyes or flowers have evolved simply through the gradual morphological modification of preceding structures, such as limbs or leaves. Because Darwin did not know anything about genetics or the biochemical processes through which morphological features develop, it is understandable that he felt able to make this sort of extrapolation.

However, we now know that embryonic development proceeds through the meticulously orchestrated action of 100s of genes and the biochemical pathways they regulate, and that new morphological features must require new genes. Unfortunately these genetic and biochemical facts are usually ignored by contemporary biologists when they propose evolutionary scenarios for the development of novel features. A prime example of this is a much cited model of how eyes could have evolved from a simple light-sensitive patch. But this model is no better that Darwin’s, because it is based solely on how eyes might have evolved through gradual changes in morphology, and completely ignores the genetic and biochemical implications. There is also a lack of clear thinking on the part of evolutionary biologists to distinguish between new variations derived solely from existing genetic variability and variations where new genes would be required (see microevolution and macroevolution).

Functioning genetic material cannot arise through an evolutionary process

Such is the success of the evolutionary model at the level of morphology that it has been assumed it will also work at the molecular level for the production of new genes. And early support for this view came from experiments where mutations were induced in various organisms, notably in fruit flies (Drosophila) which caused changes in their morphology outside of their normal range. At the time (early 20th century onwards) this was seen as evidence of the sort of mutations that could fuel long-term evolution. But we have now realised that these mutations were merely corrupting and inactivating existing genes, not producing new useful genes.

We now know that there are insurmountable obstacles to useful genes arising in an opportunistic evolutionary way. These obstacles can be appreciated by considering the genes required for proteins (which could be structural, enzymes, or transcription factors used in gene regulation).

1. For most proteins their amino acid sequence is very specific, whereas the number of possible sequences for even a modest number of amino acids is truly astronomical; this means that the odds of finding a functioning protein by chance are prohibitively improbable. This is true even taking into account the permissible variation in a protein’s sequence (which improves the chances of finding one that works) and the potential resources available having regard to the size and age of the universe.
2. Because of the success of natural selection at the morphological level, many assume that it can also operate at the molecular level to guide to functioning amino acid sequences (rather than having to rely on chance). However, for natural selection to operate – in this context to favour amino acid sequences with better performance – there has to be at least some useful function. But we know that for most proteins very little variation in its amino acid sequence can be tolerated, i.e. the vast majority of a protein’s sequence must be ‘right’ before it has any function that can be favoured by natural selection; so this first step must rely on chance, and is still prohibitively improbable.
3. Also, although a common view (e.g. presented in text books) is that proteins could have started off as short amino acid sequences (which would reduce the number of possible sequences, and hence improve the odds of finding one that works) in fact there are clear reasons against this happening, based on the need for proteins to fold into a specific 3D conformation in order to function, or that it has happened in the past.
4. Further, putative scenarios for the possible evolution of proteins generally consider only the protein-coding sequence, and overlook the fact that this must be associated with appropriate regulatory sequences to ensure that the protein-coding sequence is recognised as a gene, and that it is expressed appropriately (in the right place, and at the right time). So both a functioning protein-coding sequence and a functioning regulatory sequence must arise together, which clearly is even more improbable than a protein-coding sequence alone.
5. And what completely defies an evolutionary origin for most proteins is that most necessarily function in conjunction with others, e.g. in multi-protein complexes such as in DNA replication, biochemical pathways, or embryonic development.
6. In this context we should remember that whilst natural selection can identify systems that work, it cannot direct or promote their formation. So evolutionary scenarios can rely only on chance for working systems and/or their individual components to arise. What this means is that if a biological function requires two proteins (with neither having a function individually) then both (with their regulatory sequences) must arise together (be available at the same time and place) for natural selection to recognise their utility and preserve them. This clearly compounds the odds against such proteins arising; and as the number of mutually-dependent proteins increases, the odds against them arising increases exponentially.

Another way of looking at this is that most biological systems entail two levels of complexity. They involve a complex (and hence improbable) conjunction of several components, each of which is complex (and hence improbable) in itself. This compounding of improbabilities means that it is totally unrealistic to think that systems such as this could have arisen in an opportunistic or trial-and-error manner.

The fossil record demonstrates sudden, not gradual, appearance of new genetic material

Some might argue that, even if there appears to be a strong theoretical case against new genes evolving, we can see that nature must have found a way because of the evolution we can see in the fossil record.

However, the most striking feature of the fossil record is that new groups (with their new specific genetic material) appear suddenly, without identifiable intermediates from putative ancestors, rather than gradually as we would expect if new groups had evolved progressively from earlier forms. The best example of this is of course the Cambrian explosion when almost all animal phyla appeared abruptly, i.e. without identifiable ancestors, and within a short period of time. And a similar pattern occurs throughput the fossil record.

Following a group’s appearance, some evolution can be perceived within the group, including speciation; but it is limited, and readily accounted for in terms of evolution within the group’s existing genetic variability. The evolution of horses over the last 50 MY is probably the best-known example of sustained morphological change within the fossil record; and an examination of the evidence clearly shows that this evolution is much better accounted for in terms of selection and segregation from an ancestral gene pool, rather than involving the production of new genetic material.

Embryonic development is not consistent with common ancestry

Where groups of organisms have evolved from a common ancestor we expect their early embryonic development to be similar. This is because early development is resistant to change, because subsequent development depends on it, so changing early development is likely to disrupt later development. It has been known since the 19th century that for different classes of vertebrates (e.g. fish, amphibians, reptiles, birds, mammals) their embryonic development passes through what is called a phylotypic stage when they look quite similar to each other. This was seen as clear evidence of their common ancestry, and still is because pictures showing the similarities of early vertebrate embryos are publicised widely.

However, we have now discovered that the embryonic development of different classes of vertebrates before the phylotypic stage are radically different. In fact the earliest stages of vertebrate development are so diverse that it not only completely negates the previously thought evolutionary significance of the similar phylotypic stages, but is clear evidence against a common ancestry of the various vertebrate classes. Further, even the vertebrae – the primary defining feature of vertebrates – form in at least 3 very different ways.

Consequently, although homology is widely presented as evidence of common ancestry, in fact there are major anomalies. And the more we find out about how embryonic development occurs at the molecular level:

• the stronger is the case that embryonic development must be resistant to change (because to change it constructively would require many coordinated mutations, which is too improbable to occur); so
• the stronger is the case that where apparently homologous organs are formed embryonically by significantly different routes, in fact they are not homologous; and hence
• the stronger is the case against common ancestry.

---

Page created February 2017; last revised January 2024.