Bruce Potter

This article is the result of more than a year of investigation on my part into the books, technical papers, and simulation programs on macroevolution. It is intended to be read in the context of my other articles on Evolution Overview and Why Abiogenesis Is Impossible. If you want to read this article, but need a primer on genetics, see Genetics Home Reference primer or National Geographic’s genetics overview.

Macroevolution is generally defined as the process of one species evolving into a completely different species, and by extension, the theory that all current species evolved from an original single cell organism. At a high level, all of the arguments for macroevolution that I have read are either extrapolation from microevolution, very high level statements saying that given enough chances mutation and natural selection can accomplish macroevolution, or studies showing that 1 small isolated aspect of macroevolution is possible. None of these constitute solid evidence for macroevolution, and in fact there are serious problems with the theory. Let’s look at the specifics to explain why.

The Overall Pattern In The Fossil Record Is Not Consistent With What Is Predicted By Evolution

We do not see the gradual marching from simple to complex organisms that the theory of evolution predicts. For over 3 billion years, the biological realm included little more than bacteria and algae (Brocks, 1999). Then, beginning about 570–565 million years ago, the first complex multicellular organisms appeared in the rock strata, including sponges, cnidarians, and Ediacaran biota (Grotzinger, 1995). Forty million years later, the Cambrian explosion occurred (Bowring, 1993):

The “Cambrian explosion” refers to the geologically sudden appearance of many new animal body plans about 530 million years ago. At this time, at least nineteen, and perhaps as many as thirty-five phyla of forty total, made their first appearance on earth within a narrow five- to ten-million-year window of geologic time. Many new subphyla, between 32 and 48 of 56 total, and classes of animals also arose at this time with representatives of these new higher taxa manifesting significant morphological innovations. The Cambrian explosion thus marked a major episode of morphogenesis in which many new and disparate organismal forms arose in a geologically brief period of time…. Studies of modern animals suggest that the sponges that appeared in the late Precambrian, for example, would have required five cell types, whereas the more complex animals that appeared in the Cambrian (e.g., arthropods) would have required fifty or more cell types. Functionally more complex animals require more cell types to perform their more diverse functions. New cell types require many new and specialized proteins. New proteins, in turn, require new genetic information. (Meyer, 2004)

Everything about the macroevolution theory suggests that it should be a gradual, consistent march from simpler to more complex organisms, and yet that is not what the fossil record shows. And this discrepancy is not limited to the Cambrian explosion. Evolutionists have seen this pattern of relatively sudden appearance and disappearance of species without the multitude of stepping stones between species that one would expect: “The fossil record of an evolutionary progression typically consists of species that suddenly appear, and ultimately disappear, in many cases close to a million years later, without any change in external appearance.” (Punctuated Equilibrium:Transitional Fossils; Campbell, 1990; Gould, 1977; McCarthy, 2005)

“Over evolutionary time, one would expect species to slowly diverge, eventually creating the higher taxonomic categories. Instead, the largest differences (kingdoms and phyla) appear first.” (Silvestru, 2014, 46%)

There is a lot of debate among evolutionists themselves about how to explain the fossil record, and many different sub-theories have been proposed, for example, Punctuated Equilibrium. In fact in recent years the classic tree taxonomy diagram that had been used for decades has been discarded and other diagrams (e.g. cladograms) have been proposed to try to account for what is seen in the fossil record. “The reason for this is that the field of genetics has discovered that there is no tree. Instead, diverse organisms share different genes, seemingly at random.” (Silvestru, 2014, 46%). The overall pattern of the fossil record does not provide evidence for macroevolution.

Examples of Transitional Fossils Are Not Nearly Sufficient

Transitional fossils are fossils that appear to be between 2 other species, and therefore supposedly show the evolutionary transition between the 2 species. Evolution is founded on the premise that it takes a huge number of miniscule steps for 1 species to evolve into another, because the transition needs to happen 1 (or a few) nucleotides at a time. (A nucleotide is the basic building block of DNA and there are millions or billions of nucleotides in most organisms.) But the fossil record does not show huge numbers of miniscule steps. This list of prominent examples of transitional fossils, or this longer list of examples only show 1, or a handful, of intermediate fossils between species. But that is not nearly enough to substantiate that millions of evolutionary steps occurred between the two species.

The transitional fossils all focus on the similarity of its physical form with that of 2 of its neighbors, the implication being that there are a manageable number of physical changes necessary to get from 1 to the other. But evolution takes place at the genetic level, 1 mutated nucleotide at a time. This means it requires millions of changes to get from 1 species to another. For example, there is a difference of approximately 150 million nucleotides between chimps and humans. So it would take close to 150 million mutations to transition from the supposed common ancestor to either chimps or humans (occasionally frame shift mutations change multiple nucleotides at once). Granted, some other species have a smaller genetic code and the 2 species that were supposedly transitioned between might be a little closer genetically, but we are still talking about millions of mutations over time. Even with an incomplete fossil record, why do we not see 1000’s of slightly different fossils between 2 species? I am with Darwin when he asks: “… why, if species have descended from other species by insensibly fine gradations, do we not everywhere see innumerable transitional forms? Why is not all nature in confusion instead of species being, as we see them, well defined?” (Darwin, 1859) Scientists have been searching for these missing transitional fossils for over a century. If they were as numerous as the process implies, I would think they would have found a lot more of them.

A more plausible explanation for the relatively small number of what appear to be transitional fossils is that they were species that existed all along, that happen to be similar to other species, and then they died out. Finding only 1 or 2 fossils that seem to be morphologically in between 2 species is not sufficient to know which direction the progression went. It is similar to doing a connect-the-dots drawing in which you see a few of the dots, but thousands of dots have faded until they are invisible. You will end up with a picture composed of a few straight lines. Is that the picture that was originally intended? Possibly, but more likely the original picture was much more intricate. And yet evolutionists declare (often with great fanfare) fossils to be transitional under similar circumstances.

Natural Selection Has Critical Limitations

On the surface, it is easy to think that given enough time (chances), mutation and natural selection can accomplish anything. I think this idea is the basis of why the theory of evolution has been accepted by so many people. But if you dig a little deeper, you see that there are limits to natural selection, some boundaries that it cannot cross.

Natural Selection Has No Foreknowledge

Changes (mutations) are made at the genetic level 1 nucleotide at a time. The only way an individual change will be selected is if it is beneficial enough to increase the organism’s fitness (chance of reaching the reproductive stage and reproducing), and the benefit is strong enough to not get lost in the statistical noise of other random variables like predators, availability of food, etc. This means each tiny genetic change on the road to a functional improvement (e.g. wings) must be nontrivially beneficial in and of itself, or it will be lost. Natural selection cannot look ahead and know that this small change is good to keep, even though it is not a benefit on its own, because it will eventually lead to something good. What would actually be needed to make natural selection an effective evolutionary engine “is not just a source of variation (i.e., the freedom to search a space of possibilities) or a mode of selection that can operate after the fact of a successful search, but instead a means of selection that (a) operates during a search—before success—and that (b) is guided by information about, or knowledge of, a functional target.” (Meyer, 2004)

Natural Selection Is Limited By The Percent Of The Population It Can Keep From Reproducing

Natural selection functions by not allowing most of the inferior organisms in the species to reproduce while allowing the improved organisms to reproduce. This mechanism is limited by the number of excess organisms in the population (the number that can be prevented from reproducing each generation and still allow the population to thrive). For something like bacteria this is not as much of an issue, but for higher level organisms like mammals, with low reproduction rates and long generation times, this is a real limitation. It significantly slows down natural selection, and therefore it takes much longer for the improvements to occur than evolution predicts (more on this later).

Deleterious Mutations Are Much More Common Than Beneficial Mutations

Research has shown that beneficial mutations are incredibly rare, compared to harmful mutations, likely close to a ratio of 1:1,000,000 (Sanford, 2014, p. 35). This is because a gene is so complex and precisely tuned at multiple levels in the genome that a single bad nucleotide mutation can hinder the gene function, but it takes many simultaneous, perfectly placed, good nucleotide mutations to improve the gene function (Montanez, 2013). For example, human DNA is polyfunctional in these ways (Sanford, 2014, p. 140-141):

  • Most DNA sequences encode for 2 different RNAs that read the sequence in opposite directions.
  • Some sequences encode for different proteins, depending on where the reading begins and ends, or based on alternate mRNA splicing.
  • Some sequences serve as both a protein-coding sequence and as an internal transcriptional promoter.
  • Some sequences serve as both a protein-coding sequence and protein-binding region.
  • The three-dimensional folding architecture of the DNA is critical so that when the DNA is folded in the cell the proper genes are accessible in the necessary places. The instructions for folding the DNA are encoded in the DNA itself.
  • The DNA changes over time as the organism develops, cells differentiate, and in response to certain stimuli (Carter, 2014, 20–22%).

With each section of DNA containing so much overlapping information, the odds of a random mutation of a nucleotide resulting in a beneficial change in some of the information levels, without causing damage to other levels, is minute. It is analogous to a random one-letter printing error in a book that improves the book. First, that changed letter and the letters next to it must spell an actual word. Then the new word must make sense in, and improve, the sentence. Next, the new sentence must make sense in the paragraph, etc. Now imagine the book contains a secret code (which must be preserved) when you read every other letter, and another secret code when the entire book is read backwards. You start to see how improbable (although not impossible) it is to get a random improvement.

Deleterious Mutations Overwhelm Beneficial Mutations

The effect of most single nucleotide mutations (good or bad) is so subtle that it is not directly selected for or against, because other variables like food supply, predators, etc. outweigh it. Combined with the previous points, this means that when a rare good mutation comes along it is in an organism that has already accumulated 1000’s of small bad mutations. Over time, the overall effect is a degradation of the genetic code, not an improvement (Sanford, 2014). To compound the problem, once a good mutation occurs on a DNA sequence that also includes many bad mutations, natural selection can not normally separate the good from the bad, because nucleotides are inherited in large sections called genetic linkage blocks, not on an individual nucleotide basis. The process is similar to a textbook that is repeatedly reprinted, often with single letter changes. A few wrong letters in a 300 page textbook will not affect its usefulness because the students will know what those words were supposed to be. At the next reprinting, the couple of additional wrong letters will not make it any less useful compared to the previous version. But continue this cycle for a long time, and eventually you have a textbook that is hard to read because multiple words per sentence are misspelled. If every 100,000th printing you injected 1 additional useful sentence, the overall progression of the textbook would still be deterioration. Even if you threw out some of the least readable books after each printing, the harmful changes would still be accumulating much more quickly than the beneficial ones.

There Is Not Enough Time For The Number Of Mutations Required For New Physical Features

Most people think in terms of physical features when thinking about one species evolving into another, and from this perspective there is a manageable number of features that must be changed or created. But what they forget, or do not realize, is the vast amount of genetic information that is required for those features, and that evolution (mutation and natural selection) has to work at the genetic level, not the level of physical traits that we can see. Thus evolution has to make many orders of magnitude more changes than we normally consider.

Calculating the actual speed of evolutionary progress for a particular species is virtually impossible because there are too many unknown variables. But putting a bound on it (by assuming values for many of those variables that are extremely favorable to evolution) is much easier, and illustrative. Imagine a population of 100,000 of the ape-like ancestors of humans. Suppose that a male and a female both received a mutation so beneficial that they out-survived everyone else; all the rest of the population (all 99,998 of them) died out. And then the surviving pair had enough offspring to replenish the population in one generation! And this repeated every generation (every 20 years). None of these assumptions could be anywhere close to being true, but let’s continue with the calculation. This goes on for 10 million years, more than the supposed time since the last common ancestor of humans and chimps. That would mean that 500,000 (10 million/20) total beneficial mutations could be added to the population. Even with all of these unrealistically good assumptions, which maximize evolutionary progress, only about 0.02% of the human genome could be generated. Considering that the difference between the DNA of a human and a chimp, our supposed closest living relative, is at least 5%, or 150 million nucleotides (Britten, 2002), the evolutionary progress is still deficient by a factor of 250 (Batten, 2014, 14%). In other words, under ideal circumstances it would take 2.5 billion years to generate the necessary DNA differences. Under more realistic conditions, that number would be several orders of magnitude higher (at least). Given that the the first land mammals appeared on earth 220 million years ago, there is an obvious timeline problem, and we have not even accounted for the other 95% of the genome!

No Evidence of Creation of Genetic Information

Creation of new genetic information (as opposed to changing what is already there), is essential to the theory of macroevolution because life on this planet had to go from the small genome of the first, rudimentary organism to the incredibly complex genetic information that exists in humans. And yet there have not been any cases demonstrated in the lab or observed in the wild in which mutation has resulted in new genetic information. All the known cases in which a mutation has resulted in an organism being more “fit” have actually been a loss or corruption of genetic information (Sanford, 2014, p. 27).

There are many documented examples of this, just one of which is the stickleback fish. The sticklebacks come in two forms: a saltwater form that has prominent body spines and numerous armor plates (which help protect the fish from predators), and the freshwater form that generally has shorter dorsal and pelvic spines and substantially fewer armor plates (some with no pelvic spines or armor plates at all) (Catchpoole, 2009). Biologists have observed that when saltwater stickleback fish are introduced to freshwater lakes, over a number of years the amount of body armor and size of spines reduces. This is because a number of factors make sticklebacks with armor and spines less fit in lakes: the lack of larger predators, the cost of making the armor in the calcium-depleted lake water, and the presence on the lake bottom of predatory dragonfly larvae, which use the pelvic spines to grab onto sticklebacks that swim over them. Sticklebacks with reduced armor and lacking pelvic spines are clearly more fit to survive in the lake environment and natural selection operates to increase the number of fish with less armor and spines. “What is behind these changes? Has some new feature been invented by ‘evolution’? Geneticists have located a mutated genetic switch that affects the expression of a gene called Pitx. In the pelvic region, the corrupted genetic switch prevents spines forming in that area. … In the anniversary ‘Year of Darwin’, 2009, Nature journal honoured the stickleback as one of ’15 Evolutionary Gems’. One high-profile evolutionist, Sean Carroll, called it, ‘One of the most compelling case studies of evolution.’ ” (Batten, 2014, 13%; Catchpoole, 2009) And yet it was really a breaking of an existing gene, not creating new genetic information. Was this a case of microevolution? Absolutely. Was this the type of evolving that is required for macroevolution? Not at all.

Another example of a destructive mutation that happens to be beneficial under specific circumstances is one that most people have heard about: drug resistant bacteria. Because of the large populations and fast reproduction cycles of bacteria they can mutate and be naturally selected in time frames that are observable. In some cases, bacteria have become resistant to various forms of antibiotics. However, this is not due to the creation of new genetic information or new capabilities of the bacteria either. For instance, Mycobacterium tuberculosis (the cause of TB) has an enzyme which (in addition to its other useful functions) changes the antibiotic isoniazid into a form which destroys the bacterium. A mutation causes the loss of that enzyme and helps the pathogen withstand isoniazid (Zhang, 1992). To give another example: the 4-quinolone antibiotics attack the enzyme DNA gyrase inside various bacteria (Lewin, 1992). An informationally insignificant mutation which results in the substitution of one amino acid by another destroys the enzyme/antibiotic interaction (Wieland, 1994). It is significant that when these antibiotic-resistant bacteria are placed in a non-antibiotic environment, they die out because they cannot compete with the non-defective bacteria (Wieland, 1997).

Other cases which are sometimes used as examples of evolution, but are also actually loss of function that happens to be beneficial in a particular environment, include: loss of sight in cave fish (Wieland, 2000) and cave salamanders (Sarfati, 2008), loss of functional wings in beetles on a windy island (Wieland, 2003), and a defective gene in tomcod fish that helps them survive in waters polluted with PCBs (Wieland, 2011).

At a theoretical/mathematical level, it has been compellingly shown by Sanford that it is infeasible for even a single gene to be created by mutation and natural selection in the time frame allotted by evolution (Sanford, 2014, p. 131–152). If there is no creation of genetic information, there is no macroevolution.

Body Plans

Body plans are the higher level structure of how our bodies are arranged, and microbiologists do not know yet if body plans can mutate successfully. They do not yet fully understand what controls/directs the body plan, but it does not appear to be completely embodied in the DNA. “Instead, other factors—such as the three-dimensional structure and organization of the cell membrane and cytoskeleton and the spatial architecture of the fertilized egg—play important roles in determining body plan formation during embryogenesis.” (Meyer, 2004) Even the aspects of the body plan that are controlled by DNA are in genes that are expressed very early in development and are highly interrelated with other aspects of development, with the result being that modifying any one of them seems to always result in development shutting down (Meyer, 2004, “Novel Body Plans”).

This problem has led to what McDonald has called “a great Darwinian paradox”. McDonald notes that genes that are observed to vary within natural populations do not lead to major adaptive changes, while genes that could cause major changes—the very stuff of macroevolution—apparently do not vary. In other words, mutations of the kind that macroevolution doesn’t need (namely, viable genetic mutations in DNA expressed late in development) do occur, but those that it does need (namely, beneficial body plan mutations expressed early in development) apparently do not occur. According to Darwin natural selection cannot act until favorable variations arise in a population. Yet there is no evidence from developmental genetics that the kind of variations required by neo-Darwinism—namely, favorable body plan mutations—ever occur. (Meyer, 2004; McDonald, 1983)

This is one reason why extrapolating evidence of microevolution to macroevolution (which is often done) is not meaningful, because the genetic mechanisms are very different.

Another aspect of body plans that does not line up with macroevolution is that “developmental biologists have observed a small set of genes coordinating organismal development of body plans— and these are present across the multicellular kingdom, in the various phyla and classes.… In short, the genes that control body plans had to have originated when there were no bodies.” (Silvestru, 2014, 46%) But in that case natural selection would not have preserved those genes, because it had no advantage for those organisms!

No Realistic Proposed Scenario

A theory that is so universally accepted that it is considered essentially a scientific fact must include a realistic scenario of how it happened. And yet there is not a single specific hypothetical scenario proposed for how 1 species evolved into another. The millions of changes that would be necessary both at the genetic level and the body plan level, must have had a specific sequence to them, must have overcome many cases of irreducible complexity, would have needed a certain number of steps, and (based on all of these factors) would have taken a certain amount of time to accomplish. I have not seen anyone seriously tackle this and propose a feasible scenario, but that is exactly what is needed to show how the theory of macroevolution can overcome the serious problems it faces (many of which are cited in this article).

Computer simulation has the potential to much more accurately represent and model the process of evolution. Since there are so many mechanisms in effect simultaneously, the evolutionary process is virtually impossible to model accurately on paper with mathematical formulas. For a review of the currently available computer simulations, see Sanford, 2012, p. 4-8. Despite the potential of computer simulation in this field, there appears to be only 2 simulation programs in existence today that are even close to being biologically realistic.

Avida

Avida was developed at Michigan State University and demonstrates digital evolution as an analogy for biological evolution (among other areas). It is described in detail in a paper by Lenski, et al. (Lenski, 2003). Avida creates many small software programs (digital organisms) that run in virtual environments and run a series of basic instructions (similar to assembler code) that can sometimes combine together to perform logic functions (AND, XOR, NAND, etc.) and organism replication. Replication sometimes introduces copy errors (changes to the list of instructions) to the child, and the resulting digital organisms that perform more complex logic functions with its list of instructions are considered more fit and their replication rate is increased.

This simulation certainly demonstrates an evolving system, but the magnitudes used are not at all representative of a genomic system. Avida has just 26 possible instructions that the “mutations” choose from, and it only takes a handful of instructions to perform one of the accepted (beneficial) logic functions. Contrast that with the fact that the average human gene contains 50,000 nucleotides and the total genome contains 3 billion nucleotides. When an avida organism performs one of the logic functions, its replication rate is multiplied by anywhere between 2 and 32 (depending on the complexity of the logic function). These replication rate increases are a couple orders of magnitude higher than any real biological system (Sanford, 2014, p. 172–173). When the replication rate is made more biologically realistic, no new logic functions become fixed (firmly established) in the Avida population (Nelson, 2011).

Mendel’s Accountant

Mendel’s Accountant was developed by a distributed team of scientists and engineers, led by John Sanford. It is a simulation program that models genetic change over time. Each mutation that enters the simulated population is tracked from generation to generation to the end of the experiment. The model takes into account many realistic genetic factors, including: mutation rate, percentage of good, bad and near-neutral mutations, fitness benefit, selection method and strength, linkage blocks, genetic drift, genome size, and population size. It is currently the most biologically realistic evolution simulation software available. Even so, some simplifying assumptions must still be made about the incredibly complex process of genetic evolution, but the authors are careful to make assumptions that are clearly generous to evolution. Even with these assumptions for the hard to model factors, the simulations representative of higher order organisms conclude that selection is not able to filter out the far greater number of deleterious mutations and promote the rare beneficial mutations. The result is that the modelled genome always degrades (Nelson, 2013).

Mendel’s Accountant has also been used to model a specific interesting scenario: how long it would take for a mutation of a specific length string of nucleotides (that presumably benefits fitness) to happen and become fixed (firmly established) in the hominin population (humans and our supposed ape-like ancestors). (This is the same case for which mathematical calculations were done earlier in this article.) Using factors for population, reproduction rate, generation time, etc. that are consistent with what we know about this population in the fossil record, they have run repeated simulations for different length nucleotide strings. The result of the simulations is that for evolution to generate any meaningful string would take far longer than the real time frames available. It predicts that generating a specific string of only 5 nucleotides would take on average 2 billion years (Sanford, 2015). (The supposed time for our ape-like ancestors to evolve into humans is less than 10 million years, and there are over 100 million different nucleotides between the two.)

Scenario Summary

When you dig into the details of the macroevolution process you realize how overwhelming it is to demonstrate a potential scenario of how this could happen by chance. In fact, the simulation program that comes closest to modelling the genetic process realistically concludes that it is not feasible. I can sympathize with the problem that macroevolution scientists are up against to show a realistic scenario, but if you cannot show a realistic scenario of how this can happen, you cannot claim that macroevolution is essentially a scientific fact.

Conclusion About Macroevolution

Given the discrepancy of the fossil record, the serious limitations of natural selection, the lack of evidence of creation of new genetic information, the problem with evolving body plans, the absence of realistic scenarios, and the problem of not enough time shown by both calculations and simulations, it seems clear that macroevolution is a tenuous theory with many serious, unanswered, problems.

When Darwin first proposed his theory it was plausible that what they knew of biology at that time could have been created through mutation and natural selection. But much of what we have learned since about the incredible complexities of our genome and cells makes it a much more challenging task for macroevolution, and in fact has made it possible to calculate and simulate that mutation and natural selection are not sufficient to have produced all of it.

References