eric horvind blog picture

40 failed predictions of Evolution

Evolutionists often claim that they have made many successful predictions that show evolution is true. Let us investigate 40 predictions and expectations based on cosmic evolution (cosmological, geological, and biological evolution).

Important note: if a prediction/expectation is born out, it proves nothing, as it is the fallacy of affirming the consequent. If predictions are falsified, however, it is a problem for a scientific hypothesis, because this constitutes disproof, according to the rules of formal logic. For example, consider the premise ‘All horses are brown’. No matter how many brown horses you observe, unless you know that you have observed all horses, this can never be proven true. It only takes the observation of one white horse to show it is false (disprove it).

In the case of evolution, however, evolutionists do not accept a falsified prediction as disproving evolution, or even many failed predictions. Rather, they invent secondary hypotheses to ‘explain’ the contradictory evidence and continue to accept that evolution is true. We see many cases of such invented ‘rescuing devices’ in the examples below. This shows that evolution is not a true scientific theory but a quasi-religious or philosophical worldview that persists despite the evidence.

It is interesting that the entrenched belief in ‘deep time’ frequently undergirds the failed predictions/expectations.

Note: original sources are cited directly or in the linked articles.


    1. Prediction: the cosmic microwave background radiation (CMBR) should be lumpy, but it is extremely smooth. So, evolutionary cosmologists invented a miraculous period of ‘inflation’ (the rescue device) to try to account for this.1
    2. Prediction: there would be no preferential plane or axis or position in the universe (the cosmological principle; the universe is homogeneous and isotropic). However large-scale galaxy surveys point to our galaxy being in a preferred position; this is commonly referred to as the ‘axis of evil’, because it is such a problem for evolutionary cosmology.2
    3. Prediction: the universe contains a lot of ‘dark matter’, which is mysterious stuff that is very difficult to detect but nevertheless creates a lot of gravitational force. Galaxies seemed to need this to explain the rotational speeds of stars. The big bang also needs this to have a ghost of a chance of forming any stars. The case for dark matter was considered so strong that all that was left to do was find it; hence a multibillion-dollar industry developed to build ever more sensitive particle detectors, which in 40 years have turned up nothing. Now it seems that a thorough application of Newtonian mechanics to galaxy star orbits explains them without dark matter.3 That leaves the hypothetical big bang cosmic evolution story as the only reason for keeping the notion of dark matter.
NGC 4414 spiral galaxy. The James Webb Space Telescope is finding loads of spiral galaxies where evolutionary cosmology (big bang) predicts that they should not be.
  1. Prediction: a lot of ‘dark energy’ exists in the Universe. This is more mysterious stuff needed by big bang cosmology whereby the red-shifted radiation from distant objects is interpreted to mean that they are accelerating away from us. This acceleration needs energy: hence ‘dark energy’. There is no evidence for such dark energy, despite looking ‘hi and lo’ for it.4
  2. Prediction: distant galaxies would not be spiral galaxies—because they would be ‘young’ and not have had enough time to wind into spirals.5 In similar vein, astronomers were surprised to discover a very distant (= ‘young’ in stellar evolutionary terms) galaxy with heavier elements present, which were not expected based on cosmic evolutionary theories (big bang, etc.).6 The findings of the James Webb Space Telescope have seriously compounded this problem.7
  3. Prediction: there would be no magnetic fields on planets that should be cold and dead due to their size and supposed billions-of-years age. They have magnetic fields.8


  1. Prediction: paraconformities, or ‘flat gaps’,9 should be rare due to due to the extended timeframe of deposition of the layers, which should mean that there is ample time for erosion of the surfaces of underlying formations. However, the boundaries between layers often show no signs of erosion. Flat gaps are common.10
  2. Prediction: there should be little evidence of large-scale folding of soft sediments involving many strata. Again, this expectation arises from the extended timeframe assumed, which means that the lower layers at least should be solid rock at the time of the bending. However, extensive folded sedimentary rocks around the world argue for their being soft before folding.11
  3. Prediction: sedimentary strata should have limited geographical extent due to the localized nature of depositional processes (i.e. no global flooding cataclysms). However, continent-wide sedimentary formations are common, even extending between continents.12 There are six sequential ‘megasequences’ in the Phanerozoic (fossil-bearing strata) that are global in extent.13
  4. Prediction: there should be no carbon-14 in coal, oil, or diamonds that are supposed to be many millions, or even billions, of years old, but it is consistently detected above background levels.14 That’s because carbon-14’s half-life is only 5,730 years, and it should have decayed below the detection limit after about 100,000 years. Not only that, but coal samples ranging in evolutionary ‘age’ from 37–318 millions of years all proved to have the same amounts of carbon-14 (consistent with all the coal samples coming from plants buried during Noah’s Flood).15
  5. Prediction/claim: regarding the age of planation surfaces:16 “most of it is no older than Pleistocene”17 (less than two million years). This was based on the observed pervasive effect of the rate of erosion processes today that would not leave any flattish surfaces flat for long, especially when there is a variety of rock types differing in hardness. And yet, based on the evolutionary deep-time paradigm, there are now said to be many ‘very old’ planation surfaces.18 For example, the Kimberley Plateau of north-west Australia is said to be older than 540 Ma! Encyclopaedia Britannica notes: “There has been much scientific controversy over the origins of such surfaces.” Indeed so. The global biblical Flood in Noah’s time would explain them nicely!
  6. Expectation: bioturbation (mixing of sediments due to the activity of marine organisms) should be evident throughout the major fossil-bearing strata (Phanerozoic). This expectation arises out of the experimental observation that marine burrowing animals, which are evident throughout the Phanerozoic, thoroughly mix sediments to 10 cm depth or more within hours. Thus, if sediments were deposited at the very slow rate needed to match the deep time dating of the strata, they should all be thoroughly mixed; there would be almost no layering evident. This is not the case at all; a major fail.19


  1. Prediction: “no organism wholly soft can be preserved”. Darwin said this in Origin of Species. This prediction has failed repeatedly and spectacularly. For example, exquisitely preserved jellyfish fossils are found in multiple places.20
  2. Prediction: the fossil record, according to evolution, should show diversity first, developing into disparity, but the record is the opposite: disparity comes first (with all the major phyla appearing in the ‘Cambrian explosion’) and then comes diversity within those major categories.21 This is a serious failure.
  3. Prediction: there should be found in the rocks many finely graduated organic chains of transitional fossils showing the transitions from one major kind of organism to another, but such are absent. Transitions between major body plans should be the most amply documented, but the fossils show no such transitions.22
  4. Prediction: fossils should not cut across many strata representing millions of years. The supposed slow and gradual rate of formation of the rock layers (or with episodes of deposition separated by millions of years) means that any organism would decay before being completely buried and preserved. However, such ‘polystrate’ fossils are common.23
  5. Expectation: sea creatures and land animals would not be commonly found fossilized together, again due to the localized, slow formation of rock layers in lake beds or the seabed. However, mixtures of sea and land fossils are common.24
  6. Prediction: “… no original protein and/or DNA fragments can be recovered beyond ca. 100 kyr [100,000 years] …”.25 However, they have been found in fossils ‘dated’ at many 10s of millions of years. The desperate efforts made to overturn Dr Mary Schweitzer’s findings on dinosaur bones (proposals of bacterial biofilm to explain flexible blood vessels in dinosaur bones, contamination for proteins and DNA, preservation by iron compounds, etc.) underline just how problematic this is for the evolutionary storyline.26
  7. Prediction: fossils of organisms found in geological layers separated supposedly by many millions of years should not be similar and certainly not apparently identical. But similar fossils and lack of evolutionary change over many geological ‘eras’ are the rule; they are known as ‘living fossils’.27 Again, a rescuing device, called ‘evolutionary stasis’ was invented to try to accommodate these uncomfortable facts. Evolutionary stasis is an oxymoronic term (evolution = change; stasis = no change) pretending to be an explanation. Another rescue device: the living and fossil creatures only appear to be similar; their genomes ‘would be’ quite different! This is special pleading, and not falsifiable since we do not have access to the genomes of the fossilised organisms.28 And there are many living fossils!


    1. Prediction: vestigial organs as ‘left-overs’ of evolution. Evolutionists predicted vestiges of past evolution that are now useless organs. They identified many candidates, such as the human appendix, tonsils, ‘tailbone’, etc. Because claimed vestigial organs have been progressively shown to be functional,29 some evolutionists have resorted to redefining what ‘vestigial organ’ means—an organ that has lost its original function and now has a different or reduced function. This is yet another rescuing device that shows how evolution is not falsifiable in the minds of those who have a need to adhere to naturalism.
Famous evolutionist Haldane said that a wheel would never evolve, but extremely sophisticated ‘wheels’ in the form of rotary motors have been discovered, such as the bacterial flagellum.
    1. Prediction: no wheels would be found in living things. The famous evolutionary population geneticist, J.B.S. Haldane, proclaimed in 1949 that mutations and natural selection (neo-Darwinian evolution) could never produce “various mechanisms, such as the wheel and magnet, which would be useless till fairly perfect.” But ‘wheels’ far more sophisticated than Haldane could have imagined, such as subcellular rotary motors, have been discovered.30 A related prediction: there would be no magnets in living things, for similar reasons. Haldane was wrong about that too.31
    2. Prediction: similarities, being due to common ancestry, would show a clear pattern of phylogeny (evolutionary ancestry), tree of life, etc. This is not so; there are numerous ‘homoplasies’, which are similarities that do not fit any pattern of common ancestry, or phylogeny. Homoplasies are so common that evolutionists invented the rescuing device of ‘convergent evolution’.32 A comparison of the genes involved in bat and dolphin sonar found 200 similar genes. Since there is no possible sonar-equipped common ancestor of both, these similarities must have evolved independently, by chance mutations.33 This stretches ‘convergent evolution’ to breaking point. Another rescue device is horizontal gene transfer, which creationist Walter Remine predicted would be invoked by evolutionists.34 E.g., a key gene regulation system known as citrullination is said to have been introduced into vertebrate animals by horizontal gene transfer from cyanobacteria!35
    3. Prediction: independently originating similarities should not exist. That is, convergence is not predicted by evolutionary theory. Evolution is ‘contingent’, as Stephen Jay Gould emphasized, so if the evolutionary experiment were run again, it would have different outcomes.36 So, the evolution of two very similar creatures with entirely separate phylogenies, would be so unlikely that it would not happen. And yet ‘convergence’ abounds.37
    4. Prediction: there would be little genetic resemblance between extant and ‘primitive’ life forms (biochemical homology). Being separated in deep time, every locus of every gene would have mutated multiple times. Thus, Ernst Mayr stated in his 1963 book Animal Species and Evolution “the search for homologous genes [derived from the same ancestor] is quite futile except in very close relatives.”38 This was a strong prediction, but it has been falsified repeatedly. One example: humans share a gene involved in eye formation with flies. Walter Gehring, University of Basel scientist, remarked: “Much to our surprise, the same gene causes eyeless[ness] in the fruit fly. That came as a total surprise, because we thought that the fruit fly eye was in no way a homologous, a similar structure as in humans.”39 (emphasis added). By non-homologous, they meant that the insect compound eye and the human eye could not possibly have arisen from an eye in a common ancestor. It was a “total surprise” because it was not expected in evolutionary theory, which holds that insect and vertebrate eyes evolved separately. Another failed expectation.
    5. Prediction: Richard Dawkins explicitly predicted that all living creatures share the exact same genetic code and this is ‘proof’ of evolution. After all, switching from one code to a different one would be like switching keys on a keyboard, and scrambling the messages. However, organisms with different genetic codes have been catalogued since the 1970s. This is a massive fail under Dawkins’ own criterion.40
    6. Prediction: it was proposed in the early 1940s that genes would each code for one protein or enzyme.41 This applied in bacteria. And the realization that the formation of any functional gene coding for one protein by evolutionary processes was improbable reinforced this view. Thus, it was applied to all organisms. It was not anticipated that a gene could code for more than one protein, or other functions as well as protein production. Hence it was thought that humans would have over 100,000 genes. This is not the case; we have ~23,000 genes, but we produce many more than 100,000 different proteins. This is achieved by a given DNA sequence being multi-functional, coding for more than one protein.42
    7. Prediction: genes are linear, where a control sequence will be next to the gene controlled, and others genes involved in the same biochemical pathway will be next on the DNA strand (the lac operon in E.coli is like this). However, genes in eukaryotes such as humans are divided into exons that are separated by stretches of DNA called introns. Different exons combine in modular fashion for multiple different functions. In humans, exons from up to 33 different genes on as many as 14 different chromosomes combine to code for the sequence for a specific protein.43 Furthermore, gene control often comes ‘from afar’, a long way from the gene involved, even located on a different chromosome.44 However, the controls often turn out to be close to each other because of the 3D-arrangement of the chromosomes, also controlled by DNA coding. These discoveries of stupendous complexity go against evolutionary expectations.
    8. Prediction: there must be lots of junk DNA. Evolution needs lots of non-functional DNA for three reasons: a) Being a messy process, evolution could never produce a high proportion of functional DNA; b) Evolution needs lots of non-functional DNA to experiment with so that evolution can be ongoing; c) Most mutations are harmful, if only slightly so on average, and there are many of them, so if most of the DNA is functional this means that these mutations would inevitably cause genomic degradation (extinction), not progressive evolution. When the ENCODE project found that at least 80% of human DNA is functional, evolutionists went into overdrive to criticize the ‘dangerous’ notion that there was little if any junk DNA. This is a major failure for evolution theory that has hampered scientific progress (why study something that is junk?)!45
    9. Prediction: pseudogenes are functionless. Pseudogenes look like protein-coding genes but do not code for proteins. They were said to be ‘broken’ genes. Evolutionists have repeatedly claimed that, because they have ‘no function’ to constrain their sequences, shared pseudogene sequences (e.g. in humans and chimps) are evidence of evolution (common ancestry).46 The very notion has discouraged research to discover pseudogene functions, but functions are being discovered because diseases are associated with mutations in pseudogenes. An example is the β-globin pseudogene, which is involved in the production of red blood cells.47
    10. Prediction: the first living cell must have been quite simple (it must be to get started by purely natural processes). Researchers expected that a cell could be found that worked on ‘only’ about 20 genes (which would still be an impossible hurdle for its origin by natural means). However, the proposed LUCA (last universal common ancestor) is getting increasingly complex. The minimal viable cell now has over 400 genes/proteins! The prediction was stupendously wrong.48
    11. Prediction: there can be no global gene switching network system in genomes because such could not evolve—an evolutionary response to Prof. John Mattick’s suggestion that such might exist,49 since borne out by the ENCODE project. Mattick scathingly denounced the junk DNA claim (see #28), saying it “will go down as the biggest mistake in the history of molecular biology.”50 Even after becoming CEO of Genomics England, Mattick reaffirmed that his “most important professional achievement” was recognizing that the so-called junk DNA “specifies a massive hidden layer of regulatory RNAs that organise our development and provides the platform for brain function.51 A sophisticated gene switching network controls the construction of the bacterial flagellum, for example.30
    12. Prediction: a low mutation rate in complex organisms—it must be to avoid extinction over millions of years. The human mutation rate was assumed for deep time evolutionary reasons to be < 0.3 per person per generation. However, it has been measured at over 200 times that. As the evolutionary geneticist Alexey Kondrashov said, “Why aren’t we extinct 100 times over?” Indeed. Geneticist Dr John Sanford, co-inventor of the ‘gene gun’, has highlighted this evolutionary failure.52
    13. Prediction: mutations are random. This has been a core assumption of evolutionary theory since mutations were discovered and adopted as the source for new traits (neo-Darwinism). Evidence is mounting that mutations are not random, but that core genes are protected and that more mutations are permitted in regions of the genome that are not critical to survival.53 The findings are said to “radically change our understanding of evolution”.54 Indeed so! Could it be that such ‘directed’ mutations are part of a designed mechanism for adaptation?55 Such of course would be anathema for evolutionists.
    14. Prediction: under Richard Dawkins’ evolution-inspired ‘selfish gene’ hypothesis, stepparents should be less devoted to child rearing than biological parents. Not so. A comparison of parenting of children conceived naturally, through IVF or donor insemination (DI) showed that the quality of parenting with IVF and DI exceeded that in well-functioning families with natural conception.56 Fail, Dr Dawkins!
Darwin speculated that a peacock’s tail came about because peahens preferred showy males, but the story failed to survive experimental testing.
    1. Prediction: male birds have colourful plumage because the females preferred ostentatious males, and so the males with more colourful plumage passed on more of their genes. This was Darwin’s idea of ‘sexual selection’. However, colourful birds are more prone to predation and extinction.57 More importantly, experiments showed that peacocks with a tail, the prime example of sexual selection, are not preferred by females over males with their tails cut off.58 As an aside, even if sexual selection operated in the peacock, it would not explain the origin of the intricate design of the feathers.59
    2. Prediction: ‘kinship’ theories of cooperation explain colony formation in eusocial animals. The naked mole rat and Damaraland mole rat (which is hairy) are eusocial—the colonies are organized like a honeybee or ant colony with a ‘queen’ and several males breeding and the rest of the colony caring for the young. “This behaviour—like that of termites and ants—is found in very few mammals, and it has remained a puzzle for natural selection.” With the naked mole rat, the colony is a virtual clone, so helping raise others ensures one’s own genes survive. So, the evolutionist reasons from kinship theory for the maintenance of such eusocial behaviour. However, the Damaraland mole rat colony is much more genetically diverse because the colony seems to prefer a replacement queen to come from somewhere else if their queen dies, contrary to kinship theory.60
    3. Prediction: hybridization between different species would be unusual/rare, due to the assumption of deep time separating the species (they would be expected to be too different genetically). For example, evolutionary biologists were greatly surprised by the discovery of hybrids between different species of iguanas, and between finch species in the Galápagos islands.61 There are many hundreds of examples of hybrid mammals62 and birds (in 18 orders).63 This is not at all surprising in a biblical creation scenario, because the species are only separated by thousands of years at most.
    4. Prediction: patterns in mitochondrial DNA ‘barcode’ sequences across many species would reflect their origin in deep time. Not so, and the consternation expressed by evolutionary authors shows just how much this result was unexpected and a problem for evolutionary theory.64 Indeed, creationist research predated evolutionary research on this issue.65 The number of mutations needed to account for the differences is consistent with measured mutation rates and the Bible’s timeframe.39
    5. Prediction: microbes separated by 165 million years in evolutionary time would be quite different: “Given the large geographic distances separating the subsurface sampling sites, we hypothesized that CDA66 genomes should be genetically divergent. Further, because of the differences in the physicochemical conditions among the sampling sites, we also anticipated divergent adaptations to the local environments, i.e., that the evolutionary trajectories of the CDA populations would be analogous to those of Darwin’s finches.”67 However, the genomes had more than 99.2% average nucleotide identity. The authors invoked a ‘rescue device’: “High-fidelity DNA replication and repair mechanisms are the most plausible explanation for the highly conserved genome of CDA.” However, this would then bring into serious question the whole molecular clock paradigm, which assumes a given rate of mutation to estimate the date of origin and divergence of different organisms. It also seriously contradicted by measured mutation rates in a wide range of microbes (none are known to have such a low mutation rate, anything remotely like that needed). (CDA also lives in a hot (60°C/140°F) and alkaline (pH 9.3) environment, both very bad for DNA stability.)
  1. Prediction/claim: inheritance of acquired characters does not happen. This relates to the Weismann barrier, named after a German evolutionist who first posited that only germ cells in organisms pass on their genes—and thus changes to body cells cannot be inherited. Weismann strongly rejected the Lamarckian view of the inheritance of acquired characteristics. This was a ‘central dogma’ of Neo-Darwinism (the Modern Synthesis). There is now a growing list of examples of epigenetic inheritance that breach the Weismann barrier.68 Epigenetics: epi: above genetics: genes; hence ‘above genes’. A major mechanism involves the addition of a methyl group to a DNA base in a gene. This can block the gene’s activity. The environment (e.g., diet) can cause methylation of germ cells and the methylation pattern can be inherited such that the offspring have the same feature as the parent. It appears that such an epigenetic inheritance mechanism is involved in the rapid adaptation of fish to living in dark caves, with loss of functional eyes.69


How many falsified predictions should it take before the grand schemes of stellar evolution, geological evolution, and biological evolution are recognized as failed hypotheses? But the evolutionary ideas will not be rejected because the idea is sacrosanct. It is an item of faith because the alternative is unthinkable. Leading evolutionary population geneticist D.M.S. Watson (1886–1973) commented years ago:

“evolution [is] a theory universally accepted not because it can be proven by logically coherent evidence to be true, but because the only alternative, special creation, is clearly incredible.”70

Leave a Reply

Your email address will not be published. Required fields are marked *