Response from Ralph Seelke to David Hillis Regarding Testimony on Bacterial Evolution Before Texas State Board of Education, January 21, 2009

TO: Texas State Board of Education

FROM: Ralph Seelke, Professor of Biology, University of Wisconsin-Superior

SUBJECT: Response to David Hillis’ comments on my research

DATE: March 23, 2009

This is in response to Dr. David Hillis’ comments at the January 21 meeting of the Texas State Board of Education concerning my research on evolution. As you may recall, I have been testing a very simple (but important) aspect of evolution: the ability of evolution to produce a new function when two changes are both needed at effectively the same time. My research has basically verified that yes, indeed, a requirement for two changes effectively stops evolution, in a population of trillions of bacteria and over thousands of generations. Dr. Hillis essentially dismissed my work as being “nothing in that that you could construe as being a weakness of evolution,” and ridiculing the thought of students being presented with this as evidence that evolutionary theory had weaknesses (his exact words were “that evolution didn’t occur”). He also referred to the work of Dr. James Bull, indicating that his work had shown evolution to be capable of what I have shown it incapable of doing, i.e., of producing a function when 2 independent changes are effectively required simultaneously. He finished by saying that, even if evolution had to occur one step at a time, that this could not be construed as a weakness in evolutionary theory.

Dr. Hillis is an articulate devotee of evolution, but in this case he is wrong. In fact, there is a general consensus that a requirement for two independent events is a barrier to evolution. And, while Dr. Bull has made important contributions to our understanding of evolution, his work has not addressed the issue of multiple independent steps being a barrier to evolution. Dr. Hillis cited no specific papers by Dr. Bull, but based upon my personal expertise in this field, and my analysis of Dr. Bull’s research, it seems that Dr. Hillis was mistaken when citing Dr. Bull’s research as a refutation of my claims.

Let me begin by telling you about a paper that should have rocked the skeptics of evolution. In 1991, Dr. Barry Hall published a paper in the Proceedings of the National Academy of Sciences.(1) In this paper, Dr. Hall described an observation that appeared to solve what he called “a very general problem of molecular adaptive evolution: How is an advantageous phenotype selected when it requires multiple mutations, none of which are advantageous until all are present?” In other words, Hall was asking how does evolution take place when multiple events have to happen, and nothing advantageous to the organism takes place until both or all have occurred? (This is precisely the question that I am addressing.) He went on to describe this problem as producing a barrier to evolution that was difficult to overcome when two mutations are required and insuperable (my emphasis) when more than two are required. In that paper, Hall indicated that he was able to isolate bacteria which appeared to have evolved while faced with this problem. He thought he had found a case in which evolution had overcome the problem of having to do two things in order for evolution to occur. But it turned out Hall was probably mistaken: in 1993, he published a second paper in the Journal of Bacteriology, in which he indicated that the evolution that he observed was probably a case of sequential evolution—one event happened that was selective, followed an increase in the population of that mutant, followed by a second selective event.(2)

Now, the Proceedings of the National Academy of Sciences is a prestigious journal. The reason Dr. Hall’s article was accepted was because of the importance of the issue that he was addressing. The problem of having to do two things for evolution to occur is, indeed, a very general problem. It does not go away because Dr. David Hillis, or anyone else, declares it to be solved. It has not gone away in the past 18 years. It will not go away in the next 18 years. And we could give another name to an unsolved problem for a theory; we could call it a weakness to that theory.

A requirement for multiple mutations is a problem because of simple probabilities. Evolution can readily occur when the probability of an event is one in 100 million. Bacteria are so numerous that this number is readily found in a drop of bacterial culture. The human population is so large that sixty or more people on earth would carry a mutation that was that rare. However, when two mutations are required, the probability of both events happening now becomes one in 10,000 trillion. Even for evolution, this is a large number. It is a large number for bacterial populations, and an unimaginable number for organisms that are more complex and less numerous, such as plants and animals. And the problem, (or weakness) in evolutionary theory is that it still has no mechanism to account for the creation of new functions when multiple independent steps are involved. As Darwin himself said, “if it could be demonstrated that any complex organ existed which could not possibly have been formed by numerous, successive, slight modifications, my theory would absolutely break down.”(3) Darwin then said that he could find no such case, but the list of functions that meet this description has grown uncomfortably large in the last 150 years. Michael Behe’s book Darwin’s Black Box only scratches the surface of this problem.

We actually take advantage of the fact that a requirement for multiple events is a barrier to evolution. Our current strategy for treating HIV infections involves treating patients with three different antiviral drugs. Why is this strategy successful in most cases? As the the text for my Microbiology course states, “The effectiveness of HAART (the multidrug antiviral treatment) probably stems from the fact that each of the medications act on the replicating virus at different parts of the virus life cycle, and despite the high mutation rate of HIV, it is much less likely that any one mutant could develop resistance to all the medications at the same time.” (ref. 4, p. 751) Thus, the barrier that multiple independent mutations works to our advantage (unfortunately it is not 100% successful, primarily because of the high mutation rate of HIV and its ability to survive in patients receiving therapy). It is why a three-drug combination is the treatment of choice for the treatment of tuberculosis (ref. 4, p. 563). Again, the use of multiple drugs is in direct response to the barrier it creates to the evolution of multi-drug resistant bacteria. In fact, Barry Hall has proposed that, with mutagenic testing, we can actually predict whether evolution will be able to produce microbes resistant to an antibiotic. The test involves determining how many mutations are required to produce resistance; if multiple steps are all required, then the development of resistance will be unlikely.(5)

Dr. Hillis made reference to the work of James Bull at UT-Austin. Dr. Bull’s lab been studying viral evolution for over 15 years, and Dr. Hillis has been a co-author with Dr. Bull on some of his many publications. Dr. Bull’s lab has discovered a number of interesting features of evolution. His most recent paper illustrates the types of experiments that he has done.(6) This study involved deleting an important but not essential gene from the T7 virus. This deletion makes the virus less fit; it was a “crippled” but not “dead” virus. A second gene, with a similar ability but different function then evolved to overcome this problem. It is interesting science; it shows the flexibility of some genes—their ability to take on related functions during evolution. However, the evolution that Dr. Bull observes relies on the sort of thing that evolution does best—the ability of mutation and selection to produce the slight, successive modifications needed to change a protein into one that is slightly different. His work and mine are complementary, not contradictory. He has done excellent work showing the capabilities of evolution when it can take one step at a time. I have used a different approach to show the difficulties that evolution encounters when it must take two steps at a time. So while similar, our work has important differences, and Dr. Bull’s research has not contradicted or refuted my own.

A key question is: to what extent is the sort of evolution that James Bull has shown to occur in viruses a common occurrence? And to what extent is the absence of evolution (due to a requirement for two or more changes being needed) a common occurrence? These questions will only be answered by further research, but there is much evidence that an inability to evolve new functions is the rule, rather than the exception. This is a weakness of the theory of evolution, and, yes, students should know about it.


[1.] Hall, B. 1991. Adaptive evolution that requires multiple spontaneous mutations: Mutations involving base substitutions. PNAS 88: 5882-5886.

[2.] Hall, B. 1993. The role of single-sutant intermediates in the generation of trpAB double revertants during prolonged selection. J. Bacteriol. 175 (20): 6411-6414

[3.] Darwin, Charles. 1859. On the Origin of Species by Means of Natural Selection, or The Preservation of Favoured Races in the Struggle for Life. Bantam Classics Edition, 1999, p. 158.

[4.] Nester,E., Anderson D., Roberts C.,Pearsall N, and Nester M. Microbiology, a Human Perspective. 4th ed., 2003. Prentice Hall.

[5.] Hall, B. 2004. Predicting the evolution of antibiotic resistance genes. Nature Reviews/Microbiology 2:430-435.

[6.] Heineman,RH, Bull JJ, Molineux IJ. 2009. Layers of evolvability in a bacteriophage life history trait. Mol Biol Evol. 2009 Mar 5. [Epub ahead of print]