This paper is published in Proceedings of the Genetic and Evolutionary Computation Conference (v 2, p 1444), W. Banzhaf et al., eds., Morgan Kaufmann Publishers, San Francisco, 1999 [PDF]; and was presented at that conference, GECCO-99, in Orlando, Florida, 13-17 July 1999.

In Real or Artificial Life, Is Evolutionary Progress in a Closed System Possible?       What'sNEW   |   Lenski et al.

Assumptions are more striking than ideas — Alexander Hiam

Evolutionary progress in life on Earth is evident in the long series of steps that lead from prokaryotic life almost four billion years ago to the variety of multi-celled eukaryotic creatures with specialized organs, tissues, systems and features that exist here today. It is driven by the accumulation of new genes, the encoded instructions for life. We would like to understand this progress.

Energy reaches Earth from the sun, of course, but encoded instructions do not. We have long believed that evolutionary progress takes place in a biologically closed system, because we thought, until recently, that space was a perfect barrier to life, making our whole planet a closed biological system.

Today, however, we know that space is an imperfect barrier to life. We now know that cells can survive in space and could be delivered in viable form to Earth's surface (NASA, 1999). We now know that dormant bacterial spores can remain viable for at least 25 million years (Cano and Borucki, 1995); it is reasonable to suppose that they are immortal (Postgate, 1994). There is growing evidence that Mars once harbored bacteria, and that rocks containing them have reached Earth (McKay et al., 1996). It is no longer certain, nor even likely, that Earth's biological system is closed.

Closed-system demonstrations of evolutionary progress in biology are not difficult in principle, but they have not been convincingly done. The most ambitious demonstration to date is a series of experiments on E. coli that have cumulatively run for 24,000 generations. Although mutation and recombination were rampant, no new genes or suites of genes with new functions were reported to have evolved. Only microevolution or sideways adaptations by mutations that enabled, disabled, or slightly changed existing genes took place (Papadopoulos et al., 1999; Vulic et al., 1999).

Meanwhile, biologists are finding more and more evidence, like viral genes in humans (Sverdlov, 1998), indicating that the lateral transfer of genes is a ubiquitous process. The biological means to make evolutionary progress in an open system are becoming well known (Lake et al., 1999).

At this point, the case for evolutionary progress in a biologically closed system depends heavily on the remotest evidence of all, the new perfect barrier to life, the big bang. If the whole universe is a permanently closed system that began in a lifeless state a finite time ago, then evolutionary progress, including the origin of life, must have subsequently happened in it. But the big bang theory is plagued with frequent surprises (e.g. Glanz, 1998). In some versions, big bangs are preceded by other big bangs ad infinitum (Guth, 1997), and ways for life to persist through big bangs have been proposed (Frautschi, 1982; Krauss and Starkman, 1999). In any case, to understand evolutionary progress biology should be able to cite firmer and more immediate evidence than the big bang!

With its basis insecure and under revision, and with an alternative becoming apparent, the theory that life makes evolutionary progress in a closed system needs additional support.

Computers, like life, rely on encoded instructions. They also exhibit evolutionary progress. Accumulated improvements have made commercial software far more powerful today than only fifteen years ago. Of course, this evolution has occurred in an open system, because people installed the improvements. But computer experiments that attempt to model evolutionary progress in closed systems are under way (e.g. Ray, 1996). The work is called "artificial life" and various other names, and the experimental environment is not restricted to conventional software. Obviously, a closed-system model that exhibited lifelike, sustained evolutionary progress would have profound importance for biology. In fact, many closed-system computer models exhibit surprising behavior or solve preestablished problems. But in spite of much honest effort, none has achieved ongoing, open-ended evolutionary progress. They all remain confined within their original parameters.

Nevertheless, computer scientists are confident that an unquestionable demonstration of evolutionary progress in artificial life is imminent, because they think they are only trying to model a phenomenon already proven in biology. Many biologists, on the other hand, are under the impression that computer models have already corroborated evolutionary progress in a closed system.

Yet the phenomenon has not been unequivocally demonstrated in either medium. Until it is, one can reasonably doubt that evolutionary progress in a closed system is possible, in real or artificial life.


Thanks The author thanks Max Garzon, Chris Langton and Dan McShea for their advice and encouragement.


  • Cano, R. and M. Borucki, pp 1060-1064 v 268 Science, 1995.
  • Frautschi, Steven, pp 593-599 v 217, Science, 1982.
  • Glanz, James, pp 2156-2157 v 282 Science, 1998.
  • Guth, A., The Inflationary Universe, Addison-Wesley, 1997.
  • Hiam, A., "Obstacles to Creativity," The Futurist, October, 1998.
  • Krauss, Lawrence M. and Glen D. Starkman, pp 58-65 v 281, Scientific American, November 1999.
  • Lake, J. et al., pp 2027-2028 v 283 Science, 1999.
  • McKay, D. et al., pp 924-930 v 273 Science, 1996.
  • NASA's "Astrobiology Roadmap," 4 January 1999:
  • Papadopoulos, D. et al., p 3807 v 96, PNAS, USA, 1999.
  • Postgate, John, The Outer Reaches of Life, Cambridge University Press, 1994.
  • Ray, Thomas, "Artificial Life," 15 July 1996:
  • Sverdlov, Eugene, pp 1-6 v 428 Federation of European Biochemical Societies - Letters, 1998.
  • Vulic, M. et al. p 7348 v 96, PNAS, USA, 1999.


    01 Jan 2020: By 2020, sustainable evolutionary progress, will not have been demonstrated. (Online prediction, 2002.)
    08 Apr 2019: Helpful genetic mutations can be induced by environmental stress....
    11 Nov 2016: George Church says Prove it.
    Szamecz B, Boross G, Kalapis D, Kovács K, Fekete G et al., "The Genomic Landscape of Compensatory Evolution" [html], doi:10.1371/journal.pbio.1001935, 12(8): e1001935, PLoS Biol., 26 Aug 2014.
    Erick Chastain et al., "Algorithms, games, and evolution" [link], doi:10.1073/pnas.1406556111, Proc. Nat. Acad. Sci., USA, online 16 Jun 2014. What algorithm could possibly achieve all this in a mere three and a half billion years?
    Scientists find 'holy grail' of evolving modular networks by Anne Ju, Cornell Chronicle Online, 30 Jan 2013.
    23 May 2012: Proving Darwin by Gregory Chaitin [book review.]
    S. Ciliberti et al., "Innovation and robustnes in complex regulatory gene networks" [link], p 13591-13596 v 104, Proc. Nat. Acad. Sci., USA, 21 Aug 2007. Robustness and the Ability to Innovate Are Negatively Correlated.
    29 Jul 2007: The Limits of Organic Life in Planetary Systems
    John Doyle and Marie Csete, "Rules of engagement" [text | Editor's Summary], p 860 v 446, Nature, 19 Apr 2007. New genes and pathways... can plug-and-play, as long as they obey protocols.
    The Golem Project by Hod Lipson and Jordan B. Pollack, last updated 3 Sep 2001. "...We have concluded that merely more CPU is not sufficient to evolve complexity: The evolutionary process appears to be hitting a complexity barrier that is not traversable using direct mutation-selection processes...."
    Wolfgang Banzhaf et al., "From artificial evolution to computational evolution: a research agenda" [abstract], doi:10.1038/nrg1921, p 729-735 v 7, Nature Reviews Genetics, Sep (online 8 Aug) 2006.
    15 Apr 2006: The Evolution Prize will be launched hopefully at ALife X.
    Roger Brent and Jehoshua Bruck, "Can computers help to explain biology?" [text], doi:10.1038/440416a, p 416-417 v 440, Nature, 23 Mar 2006.
    5 Jan 2006: "Evolution in Action" was the number one "Breakthrough of the Year" according to Science.
    Brig Klyce solicits Tom Ray's current opinion on this question in an open email, 31 Dec 2005.
    31 Aug 2005: Americans think public schools should teach creationism alongside evolution.
    10 Dec 2004: Evolution versus creationism was the topic on CNN last week.
    26 Nov 2004: The evolution of a new fruitfly gene...
    26 Sep 2004: An article promoting Intelligent Design...
    19 Jun 2004: Darwinism's ability to produce evolutionary progress has not been demonstrated....
    2002, December 31: No evolutionary progress in a closed system!
    Scientists exposed as sloppy reporters by Hazel Muir, New Scientist, 14 Dec 2002. Is this how biologists and computer modelers became misinformed?
    2002, May 9: Bet on sustainable evolutionary progress?
    2001, November 21: The University of Oklahoma will probe for evolutionary progress in closed systems.
    2001, October 9: Funds available for research.
    A Scientific Critique Of Evolution by Dr. Lee Spetner — A creationist writer makes an informed case that Darwinian processes have not been shown to produce evolutionary progress. 25 February 2001.
    2000, November 23: ...The Concept of Progress in Evolutionary Biology — book review.
    Essay by Jon Richfield and exchange with Brig Klyce, mostly about progress in evolution (sometimes acrimonious), in "Replies," February-March 2000.
    1999, November 22: GECCO-2000, July 8-12, in Las Vegas, NV.
    1999, November 18: NASA's Center for Computational Astrobiology inaugurated today.

    What'sNEW: Lenski et al.

    "Changing fitness effects of mutations through long-term bacterial evolution," by Alejandro Couce et al.,
    Science, 26 Jan 2024. The numbers of beneficial mutations rapidly tailed off during long-term passage.... The authors found nonessential genes that became essential and essential genes that became nonessential in all lineages.
    "Evolutionary innovation through transcription factor rewiring in microbes...," by Matthew J. Shepherd et al., PLoS Biol, 23 Oct 2023. ...we reveal a hierarchy among transcription factors that are rewired to rescue lost function, with alternative rewiring pathways only unmasked after the preferred pathway is eliminated.
    ...Taking a Closer Look at Experimental Evolution by Casey McGrath, Genome Biology and Evolution, Sep 2022; re:
    Rapid Adaptation Often Occurs through Mutations to the Most Highly Conserved Positions of the RNA Polymerase Core Enzyme by Yasmin Cohen and Ruth Hershberg, Genome Biology and Evolution, Sep 2022.
    Legendary bacterial evolution experiment enters new era by Ewen Callaway, Nature, 14 Jun 2022.
    10 Mar 2021: New genes can be acquired only via HGT....
    18 Dec 2018: Horizontal Gene Transfer (HGT) is the only source for all of the evolutionary innovations observed among all of the bacterial clades examined.
    Innovation in an E. coli evolution experiment is contingent on maintaining adaptive potential until competition subsides by Dacia Leon et al., PLoS Genet., uncorrected proof online 12 Apr 2018.
    ...experiment with horizontal gene transfer shows that recombination can sometimes overwhelm selection by Rohan Maddamsetti and Richard E. Lenski, doi: 10.1371/journal.pgen.1007199, PLoS Genet., 31 Jan 2018.
    Molecular evolution: No escape from the tangled bank by Joshua B. Plotkin, Nature, online 18 Oct 2017. many genes, mutations quickly provide their maximum possible fitness gain for the cell. After this, further mutations in the gene hold no adaptive benefit.
    Mutator genomes decay, despite sustained fitness gains, in a long-term experiment with bacteria by Alejandro Couce, Larissa Viraphong Caudwell et al., doi:10.1073/pnas.1705887114, PNAS, 10 Oct 2017. These findings suggest the need to reexamine current ideas about the evolution of bacterial genomes, and they have implications for other hypermutable systems such as viruses and cancer cells.
    Specificity of genome evolution in experimental populations of Escherichia coli evolved at different temperatures by Daniel E. Deatherage et al., doi:10.1073/pnas.1616132114, PNAS, 15 Feb 2017. These findings demonstrate that genomic signatures of adaptation can be highly specific, even with respect to subtle environmental differences, but that this imprint may become obscured over longer timescales as populations continue to change and adapt to the shared features of their environments.
    Tempo and mode of genome evolution in a 50,000-generation experiment by Olivier Tenaillon, Jeffrey E. Barrick, Richard E. Lenski et al., doi:10.1038/nature18959, p 165-170 v 536, Nature, 11 Aug 2016. ...We analysed complete genomes of 264 clones from 12 populations across 50,000 generations of the long-term evolution experiment (LTEE) with E. coli. These populations have evolved in a defined medium with scarce resources since 1988. Mean fitness measured in competition with their ancestor increased by ~70% in that time. After 50,000 generations, average genome length declined by 63 kb (~1.4%) relative to the ancestor....
    Contrasting effects of historical contingency on phenotypic and genomic trajectories during a two-step evolution experiment with bacteria by Jessica Plucain et al., doi:10.1186/s12862-016-0662-8, BMC Evolutionary Biology, 23 Apr 2016. It is somehow surprising that the historical contingency detected at the phenotypic level was not related to parallelism at the genomic level.
    Richard E. Lenski et al., "Sustained fitness gains and variability in fitness trajectories in the long-term evolution experiment with Escherichia coli" [html], doi:10.1098/rspb.2015.2292, Proc. R. Soc. B, 16 Dec 2015. "...Fitness was measured competing a population sample against a reference strain...."
    Jeremy W. Fox and Richard E. Lenski, "From Here to Eternity—The Theory and Practice of a Really Long Experiment" [html], doi:10.1371/journal.pbio.1002185, PLoS Biology, 23 Jun 2015.
    Elizabeth A Ostrowski, Charles Ofria and Richard E Lenski, "Genetically integrated traits and rugged adaptive landscapes in digital organisms" [html], doi:10.1186/s12862-015-0361-x, n 83 v 15, BMC Evolutionary Biology, 12 May 2015. "Finally, our results illustrate the richness of evolutionary dynamics in digital systems and highlight their utility for studying processes thought to be important in biological systems, but which are difficult to investigate in those systems."
    4 Mar 2015: Our results demonstrate that natural selection can rapidly rewire regulatory networks in very few, repeatable mutational steps.
    12 Aug 2014: ...The set of known underground reactions has a significant potential both to increase fitness in existing environments and to exploit new nutrient sources.
    Goldsby HJ, Knoester DB, Ofria C, Kerr B, "The Evolutionary Origin of Somatic Cells under the Dirty Work Hypothesis" [html], doi:10.1371/journal.pbio.1001858, 12(5): e1001858...; and commentary: Chase JM, "A Fool to Do Your Dirty Work?" [html], doi:10.1371/journal.pbio.1001859, 12(5): e1001859, PLoS Biol., 13 May 2014.
    Leiby N, Marx CJ, "Metabolic Erosion Primarily Through Mutation Accumulation, and Not Tradeoffs, Drives Limited Evolution of Substrate Specificity in Escherichia coli" [html], doi:10.1371/journal.pbio.1001789, 12(2): e1001789, PLoS Biol., 18 Feb 2014.
    Cooper VS, "The Origins of Specialization: Insights from Bacteria Held 25 Years in Captivity" [html], doi:10.1371/journal.pbio.1001790, 12(2): e1001790, PLoS Biol., 18 Feb 2014.
    Erik M. Quandt et al., "Recursive genomewide recombination and sequencing reveals a key refinement step in the evolution of a metabolic innovation in Escherichia coli" [abstract], doi:10.1073/pnas.1314561111, Proc. Nat. Acad. Sci., USA, online 30 Dec 2013. "...We identified a key mutation that converts a rudimentary form of the innovation into a refined trait...."
    Michael J. Wiser, Noah Ribeck and Richard E. Lenski, "Long-Term Dynamics of Adaptation in Asexual Populations" [abstract], doi:10.1126/science.1243357, Science, online 14 Nov 2013; and commentary: Elizabeth Pennisi, "The Man Who Bottled Evolution" [summary], doi:10.1126/science.342.6160.790, p 790-793 v 342, Science, 15 Nov 2013. "[Citrate metabolism] was the biggest event in the entire E. coli experiment."
    Arthur W. Covert III et al., "Experiments on the role of deleterious mutations as stepping stones in adaptive evolution" [abstract], doi:10.1073/pnas.1313424110, Proc. Nat. Acad. Sci., USA, online 5 Aug 2013.
    Thomas Bataillon, Paul Joyce and Paul Sniegowski, "As it happens: current directions in experimental evolution" [abstract], doi:10.1098/rsbl.2012.0945, Biol. Lett., online 31 Oct 2012.
    20 Sep 2012: Richard Lenski's research group has analysed the evolution of aerobic citrate metabolism among cloned bacteria.
    Nicholas Leiby, William R Harcombe and Christopher J Marx, "Multiple long-term, experimentally-evolved populations of Escherichia coli acquire dependence upon citrate as an iron chelator for optimal growth on glucose" [abstract], doi:10.1186/1471-2148-12-151, 12:151, BMC Evolutionary Biology, 21 Aug 2012. "The strains we examine here have evolved specialization to their environment through apparent loss of function."
    Welcome to the E. coli Long-term Experimental Evolution Project Site, Richard E. Lenski, Michigan State University.
    Mickaël Le Gac et al., "Ecological and evolutionary dynamics of coexisting lineages during a long-term experiment with Escherichia coli" [abstract], doi:10.1073/pnas.1207091109, p9487-9492 v109, Proc. Nat. Acad. Sci., USA, 12 Jun 2012.
    23 Feb 2012: Experimenters with a virus and its bacterial host in a quarantined system report a breakthrough.
    Hsin-Hung Chou et al., "Diminishing Returns Epistasis Among Beneficial Mutations Decelerates Adaptation" [abstract], doi:10.1126/science.1203799, p1190-1192 v332, Science, 3 Jun 2011.
    Aisha I. Khan et al., "Negative Epistasis Between Beneficial Mutations in an Evolving Bacterial Population" [abstract], doi:10.1126/science.1203801, p1193-1196 v332, Science, 3 Jun 2011.
    29 Apr 2011: An analysis of long-running evolution experiments has been done by biochemist Michael Behe.
    13 Apr 2011: Nothing yet. That's what we observe from an experiment at Michigan State University.
    Tanguy Chouard, "Evolution: Revenge of the hopeful monster" [html], doi:10.1038/463864a, p864-867 v463, Nature, 18 Feb 2010.
    J.E. Barrick and R.E. Lenski, "Genome-wide Mutational Diversity in an Evolving Population of Escherichia coli," p119-129, Evolution: The Molecular Landscape (Cold Spring Harbor Symposia on Quantitative Biology LXXIV), Bruce Stillman, David Stewart and Jan Witkowski, eds., Cold Spring Harbor Laboratory Press, 2009.
    An Evolve-By Date, online commentary by Olivia Judson from The Times, London, 24 Nov 2009.
    29 Oct 2009: 40,000 generations of E. Coli have been monitored in a long-term experiment.
    5 Jun 2008: Cloned bacteria evolved an unexpected feature in a long-running experiment led by Richard Lenski.
    Christoph Adami, "Digital genetics: unravelling the genetic basis of evolution" [abstract], doi:10.1038/nrg1771, p 109-118 v 7, Nature Reviews Genetics, Feb 2006.
    Carl Zimmer, cover story: "Scientists at Michigan State Prove Evolution Works," p 28-36 v 26, Discover, Feb 2005 [Zimmer's transcript]. "Avida is not a simulation of evolution; it is an instance of it."
    Chet Raymo, "Evolution was, and is, a great notion," The Boston Globe, 17 June 2003. "As these computer experiments continue — and this is not the first — the insights of Darwin and Wallace become ever more compelling."
    Origin of complex functions? — Brig Klyce's letter to Nature, 17 May, and a response, 22 May 2003.
    2003, May 11: Computer model evolves complex functions?
    2003, February 4: The latest results from a closed-system biological experiment.
    2003, January 26: "Evolving Inventions," in Scientific American.
    Claus O. Wilke et al., "Evolution of digital organisms at high mutation rates leads to survival of the flattest"[abstract], doi:10.1038/35085569, p 331-333 v 412 Nature 19 July 2001. "...Competition ...favour[ed] the genotype with the lower replication rate. These genotypes, although they occupied lower fitness peaks, were located in flatter regions of the fitness surface and were therefore more robust with respect to mutations."
    Survival Of The Flattest,, 23 July 2001.
    Contacting Richard Lenski, in August 2000, was not fruitful.
    2000, April 14: Comments from Nobel laureate Joshua Lederberg seem relevant.
    1999, September 26: Experimental Evolution with Microbes and Molecules
    1999, August 12: New computer model of evolution
    1999, July 15: A Recent Issue of Science....
  • COSMIC ANCESTRY | Quick Guide | Next | by Brig Klyce | All Rights Reserved