Search This Blog

Friday, December 19, 2008

Shall Darwinists fire up the big OZ head again?

My last couple of posts were so (apparently) utterly devastating to the concept of macroevolution and anthropogenic climate change that no naysayers could come up with any counter arguments. I therefore urge you to read the last couple of posts before this one to review them if you have not read them previously.

The attempt has been made to therefore change the subject. I am going to address that attempt, but let us be clear on something: Macroevolution has now been tested and found to not ever happen. We've analyzed and poked and prodded enough multiple thousands of generations of bacteria to discover that it simply never occurs.

First, I will put an exclamation point on the complete failure of macroevolution to show up on the bacterial field of play. I will do so using a post from a scientist who is seemingly willing to consider the possibilities of macroevolution having happened and the idea of an old Earth. He starts from that point of view as he looks at and presents the evidences he sees in today's scientific world and the present state of research. Second, time to have a dialogue with Creeper.

What Can Evolution Really Do? How Microbes Can Help Us with the Answer

Ralph Seelke, Department of Biology and Earth Science, UW-Superior

Outline of Euonia Presentation

Hello- I’m Ralph Seelke, and I’m a microbiologist who’s obsessed with answering one basic question: What can evolution really do??? By this I mean: What can we really SHOW it to be capable of doing? I want to cut through the theory and inference and speculation, and get down to what it has been shown capable of doing, and what it has been shown to be incapable of doing IN THE LABORATORY.

In this talk I want to introduce you to the notion of experimental science, as opposed to historical science, and then convince you that actual, real-time experiments can help us understand the capabilities and limitations of evolution. Then, we’ll take a tour of some of the things researchers have discovered about evolution through experimentation.

A small warning, and a bottom line- do not expect to be overwhelmed by its capabilities.

A one number bottom line: TWO. This appears to be the number of things evolution cannot do, when both of the events are required for evolution to occur.

Let’s start by covering some of the distinctions between historical science and experimental science

A. Historical Science:

When you think of the science behind evolution, you are normally thinking about historical science. If it is science that could be helped by having a time machine, then it’s probably historical science. Evolution, geology, astronomy, all fall into this category. It is science that deals with an observation, and then uses facts, reason, and logic to deduce the story of what happened in the past. In fact, we use this type of science all the time. We see that the gas gauge on the car is on E, and had half a tank two days ago. Then we remember- our teenage daughter is home.

In Geology, we see layers in the sandstone rock formation, and explain it by saying that each layer represents a long period of time, when the sand was deposited and then turned to rock.

We see fossils in the rocks- more primitive ones lower down, more advanced ones higher up- and infer that the lower ones were the ancestors of the younger ones.

We may include other facts to help support our story: in the case of evolution, facts from genetics and from anatomy have been typically used to support the standard story of evolution. We may even do things that resemble experimental science to support our story, such as determine the DNA sequence of genes from two different organisms, or examine their anatomy to determine similarities or differences.

Yet, in all the work that we do, the one fact remains: we weren’t there to observe what actually happened.

B. Experimental Science

In experimental science, on the other hand, you ARE there to observe what happened, because you’re the one who made it happen! It’s the sort of thing you often see on product advertisements- the BIG tomatoes produced when you use Miracle-Grotm , and the small ones when you don’t. You are asking a specific question, preferably with a yes/no answer (Does Miracle-Gro produce bigger tomatoes?), and devise an experiment to answer the questions. In the case of Miracle-Gro, it’s a fairly simple experiment: you grow plants under exactly the same conditions, except one set gets the M-G, the other doesn’t. Then you wait for the results.

How might that work with evolution? Let’s start with a ridiculous example, and then go to one that could really work:

Take a Population and give it an “evolutionary task”:

Have a shrew evolve into a bat

Bats appeared quite suddenly in the fossil record, about 50 million years ago; one of the suggested ancestors was the tree shrew.

Put the population in an environment where it can live, BUT that favors evolution:

Use tree-climbing shrews, with predators and flying insects! If it evolves into a BAT, it will avoid the predators and catch the flying insects BETTER

Wait for evolution to occur; look for bats in your tree shrew population.

II. Wait a minute- you can’t do experiments on things that take millions of years! Well, yes you can

The example, of course, is ridiculous- you’d need too much time, and too many shrews, to do the experiment. (Also, the critic of your experiment would maintain that bats didn’t really evolve from shrews, but from an ancestor common to both- shrews can’t readily back down the evolutionary pathway). Anyway, the fact that you didn’t evolve any bats wouldn’t hurt the theory of evolution.

Even though this type of experiment wouldn’t be possible, you can do similar experiments with microbes- in particular, bacteria.

A. Evolution isn’t really about having enough time- it’s about having a large enough population and enough generations.

So- what WOULD you need to make experimental evolution work?

Lots of organisms and Lots of generations.

Lots of organisms allows rare events to occur. Winning the PowerBall, for instance, is a rare event- but it does happen. Why? Because there are hundreds of millions of tickets sold. If a mutation that is favorable only occurs in one out of every 10 billion organisms (which would be rare), then it will occur in a population of 100 billion.

Lots of generations allows cumulative advantageous mutations to occur. If evolution is gradual, then the population will collect multiple advantageous mutations.

So, if we have billions and billions of organisms, and thousands and thousands of generations, then we can begin to ask what evolution can really do. With what kind of organism can you have billions of individuals, evolving for thousands of generations?


B. Microbes excel on both counts

1. They reproduce rapidly to produce immense numbers of generations and individuals.

You can produce immense numbers of microbes: In a quart container, you can grow, overnight, from one bacterium, a trillion bacteria. When you compare that to human populations, this is many times more than the number of people who have ever lived on earth in all of time!

You can also produce many generations of bacteria as well- A very common way to produce multiple generations is to do serial transfer- you grow up your microbe in an environment that lets you find out if evolution has occurred. This can be as little as a teaspoon or tablespoon of growth medium. When the microbe has grown for the day, and reached its maximum number of organisms, you than transfer 1% of the culture to a new tube. This part that you transferred then grows and increases 100-fold from the number that you put in, until it reaches saturation- the maximum number that it can grow. As it does this, it produces over six generations of growth. With bacteria, each time the population doubles is considered a single generation- so one becomes two becomes four—8—16—32—64—when they’ve all doubled over six times the culture has increased 100-fold, and you’ve made the microbes reach their maximum number.

The number of generations that you produce in this way is 6.64- if you consider a human generation to be 25 years, this is the equivalent of 166 years for a human population!

What this means is that it is possible to get THOUSANDS of generations in real time:

46 generations per week

Almost 400 per month

Over 2400 in a year

24,000 in ten years!

Richard Lenski has been doing this sort of experiment with 12 separate cultures of the laboratory bacterium Escherichia coli. At this point, these 12 cultures have evolved for approximately 40,000 generations in a broth that has extremely low levels of nutrients available. This is equivalent to a million years of evolution in human terms. We’ll talk about what Lenski found in a little bit.

Granted we are not bacteria- for one thing, bacteria are asexual, and so the contributions of sex to evolution are eliminated. Still, it means that we can get answers to what happens when organisms evolve for thousands of generations.

2. They have lots of complex structures and activities that are well-studied

Bacteria are amazingly complicated. They can make a new copy of themselves in as little as 20 minutes. They can search for food in their environment. Most can make all 20 amino acids, the building blocks of the thousands of proteins that a cell contains- we would starve if we weren’t supplied with 8 of these in our diet.

Many of these activities would be considered irreducibly complex. Most of you are familiar with this term, with the bacterial flagella being the poster child for an irreducibly complex structure. But let me give you a few that might be a harder to visualize, but are more useful for studying evolution.

Here’s one example of an irreducibly complex system. This one allows bacteria to use milk sugar- lactose- as a source of food. It turns out that for lactose to be used for food, bacteria have to do two things- they have to bring it into the cell, and they have to break it apart. They have two genes that code for two proteins that do these functions. One codes for a permease that allows the lactose to enter the cell. The other is an enzyme that breaks the lactose into two smaller sugars, glucose and galactose. The common name for this enzyme is lactase- you can buy it in a grocery store, since it helps people who are lactose intolerant.

What this means is that for the bacteria to grow using lactose, it must have both functions. If it is able to break lactose down, but can’t bring it into the cell, it will starve while surrounded by lactose. If it can bring it into the cell, but can’t break it down, it will still starve.

I could list literally dozens of examples of bacterial processes that take multiple steps, and if any one of those steps is disabled, the whole process ceases to work.

Of course, the problem with either producing an irreducibly complex function to begin with, or restoring one that has been disables by mutation, boils down to probabilities-

When you require two events to take place, and both are required before anything happens- what statisticians call independent events- your probabilities become much worse. If you need two events, and both are one in 10 million occurrences, then you’re probability of both happening become one in 10 million times 10 million, or one in 100 trillion. And as Michael Behe has pointed out, the simplest cell is loaded with irreducibly complex structures, and most require MANY more than two independent events to produce a new function.

3. We can find mutants – the result of evolution- easily by selection

The WONDERFUL thing about bacteria is that I can find evolution simply by LOOKING FOR GROWTH.

Let’s say I want to find out if a microbe that is unable to use lactose for food, can evolve the ability to do this.

I can look for evolution simply by putting a billion or 10 billion, or 100 billion… bacteria on an agar plate or plates. The only food source (technically, source of carbon and energy) in the plate is lactose. If only one of them evolves to be able to grow on lactose, then it will form a colony on that plate.

Or let’s say that you think I’m not being fair- it may take several steps to make the microbe able to grow using lactose- I might have to accumulate small mutations to be able to do this.

Then I can grow the bacteria in liquid culture. I’ll give them a little glucose, so they’ll be able to grow, but I’ll also give them a LOT of lactose as well. If the bacteria evolve to be able to use the lactose, then I’ll again find them. Being able to use the lactose will produce a large advantage- the bacteria that evolve will continue to grow and multiply, while the others won’t. I can transfer my culture every day- and produce hundreds or thousands of generations. And if evolution occurs, I will know- the culture that evolves will be much more turbid that the one that doesn’t.

What I want to do now is go back to our question- what can evolution really do? To do that, we will look at a case where evolution has resulted in a microbe gaining a function, and what happens when microbes are allowed to evolve for a long period of time under selective conditions. We will also look at similar examples, where evolution fails, and try to draw some conclusions about what evolution can really do.

Please note that I’m sticking to laboratory examples of evolution in microbes- there are certainly other examples, primarily antibiotic resistance that could also be covered under this topic. Those, however, aren’t the sort of controlled experiment that we are describing here.

III. Case Studies: Examples of evolution from the world of microbes.

A. Gaining the ability to utilize a new source of food

Let’s start with an example of evolution that is occasionally discussed in ID circles. This is work that Dr. Barry Hall, now retired, but formerly at the University of Rochester, did.

As we’ve said, bacteria can utilize a number of different compounds for food. One of the things E. coli can “eat” is lactose- milk sugar. In order to do this, it has to make two proteins, and so it has the genes that code for these two proteins. One is a permease, that brings the lactose in; the other is the lactase itself. Now the microbe is quite sophisticated in how it controls production of these two proteins. Under normal conditions, it only makes them in quantity when lactose is present. If the cell is growing without lactose, you would find very little lactase or permease around. But if you then add lactose, the little bit of permease and lactase that is there acts to stimulate the production of MORE lactase and permease, allowing the cell to make these proteins only when they are needed.

Hall took a strain of E. coli that had was missing the lactase gene- it was GONE! He then showed that E. coli had in it a spare gene that could, through evolution, serve as a lactase gene. His evolved bacteria looked like this- small colonies on top of a larger colony that was unable to use lactose. This gene (called the ebg gene, for evolved β-galactosidase gene; βgal is a more technical name for a lactase) was stimulated by the presence of lactose (i.e., more of the ebg gene product was made). However, the ebg enzyme was TERRIBLE at splitting lactose. Even when the enzyme was around, it split lactose so poorly that the cell could not grow on lactose as a food source. However, under the right circumstances E. coli could, through evolution, turn the ebg gene into a useful lactase gene. The right conditions involved, growing cells in the presence of lactose where lactose could get into the cell, and finding mutants that had evolved the ability to use lactose and were thus more fit than the other E. coli in the population. In Hall’s experiments, the evolved bacteria showed up as mini-colonies, growing on lactose from a large colony that had grown on other food, but could not grow on the added lactose.

Once this gene was activated enough to allow E. coli to grow on lactose, then further evolution could take place, and the microbe evolved to become better at growing on lactose.

I’m going to come back to this example later, but this shows what evolution can do.

  • A hidden gene for breaking down lactose was present in E. coli. It was activated by the presence of lactose

  • When the gene was activated , the product that it made worked so poorly that the cell still couldn’t grow on lactose

  • All it took was one mutation to make it able to break down lactose and slowly grow on it

  • After that first mutation, other mutations made the cell able to grow even better on lactose.

Gaining the Ability to use a new source of food

Over 10 cases of bacteria evolving the ability to use a new food source have been observed-

What are the common themes?

  • A gene is present in the microbe, but is either silent (not expressed) or expressed but producing an inactive product.

  • A single mutation may activate the gene (no longer silent) or result in an active product.

  • Further mutations can then make the microbe better at using this new food source.

  • Not exactly earth shaking

B. LONG term evolution: Richard Lenski’s 35,000+ generation evolution experiment.

Richard Lenski, Michigan State U., has been evolving E. coli for over 16 years, obtaining the 6.64 generations per day that we discussed earlier. He and his assistants have followed the evolution of 12 cultures of E. coli for about 40,000 generations. Again, this is a relatively simple experiment.

You grow start 12 flasks of bacteria, growing in a medium that has very little glucose in it. The bacteria multiply for the first 6 hours or so, and then are in a state of starvation for the next 18. You then transfer 0.1 ml of the culture to a fresh supply of this medium- about 2 teaspoonfuls, and let them grow up again. You do this every day for the REST OF YOUR LIFE.

How do you measure evolution? By measuring fitness- you essentially run bacteria races-

Comparing the culture that has evolved with its ancestor. The neat thing about bacteria is that you can COMPARE A MICROBE WITH ITS ANCESTOR FROM 30,000 GENERATIONS AGO!!!

You mix equal amounts of the ancestor with the microbe that has evolved

You let them grow together.

You then determine the number of bacteria of each type, by diluting and plating. The evolved culture and its ancestor have a “marker” that doesn’t affect evolution, but allows you to tell them apart- one is red on a plate, and the other is white.

What have we learned from this type of experiment?

1. Bacteria become more fit.

2. They become bigger

3. Most of the gain is in the first 2000 generations

4. Most of the gain comes from five different genes that have mutated.

5. After 20,000 generations, his group sequenced 918,700 bases from 50 isolates- they found 10 changes, all in ones with a “mutator” phenotype.

6. These bacteria are still very much E. coli.

IV. Case Studies: When evolution fails- confirmation of the problem of irreducible complexity

Let’s return to the case of “evolution in action” that was observed by Barry Hall. He observed the evolution of the “evolved β-galactosidase” gene in E. coli, allowing bacteria missing the normal lactase gene to grow using lactose as a food source.

Now, if you or I tried to get this E. coli strain to evolve- it wouldn’t work. Even if we used selection to allow trillions of cells to evolve, you would not be able to observe evolution.

So how come Barry Hall got them to evolve? The answer tells us about both the capabilities and limitations of evolution.

It turns out that when you delete the lactase gene, you produce not one, but two problems for the cell to overcome. Remember that it is the permease that brings the lactose into the cell. Well, it turns out that that in order to get the permease to be made (the one that brings in lactose), you need the original lactase gene to be active. It turns out that the lactase, in addition to breaking down lactose, is part of a feedback control loop- the fact that it is working tells the cell that there’s lactose around, and that in turn stimulates the cell to produce the permease, bringing in MORE lactose. So, E. coli that’s missing its lactase gene has TWO problems- it can’t break down lactose, and it can’t bring it in.

Sound familiar? The cell faced a problem of restoring an irreducibly complex system.

Now, in order to observe this evolution, he had to trick the microbe. It turns out that you can artificially stimulate the permease gene, causing it to become active. When you do this, you’re problem goes from evolving two functions to evolving one function.

What happens if you take a bacteria with a missing lactase gene, and an ebg gene, and grow it in the presence of lactose?



Lactose can’t get in

No lactose inside= no ebg gene stimulated

No ebg gene stimulated = can’t find mutant ebg genes that break down lactose!!!


What did Hall do? He “tricked” the microbe- gave it a compound called IPTG that tricked it into making the permease. Now lactose was in the cell, and the ebg gene could evolve to produce the “evolved β-galactosidase” to break down the lactose.

Now- what happens if you try to evolve E. coli, without “cheating”- requiring E. coli to both bring in the lactose and make the new β-galactosidase?

NOTHING HAPPENS- you don’t get any evolution- the requirement for two steps- both of which are needed- stops evolution.

There are at least three other examples of this same principle- the requirement for two steps seems to stop evolution.

This should not be so surprising. Mutations are rare events- often only occurring once in 10- 100 million cells. If you need two events to occur- the probability that BOTH will occur becomes one in 10 million X 1 in 10 million, or one in a hundred trillion. Most studies simply haven’t looked that hard for evolution, and so haven’t found it.

Now, here’s where things get a bit tricky. You MIGHT find evolution in bacteria taking two events- that’s because there are SO many of them out there- I have heard one estimate of the total number of bacteria on the earth to be around 1027 .

But lets go back to our shrew trying to evolve into a bat. How many shrews are there? Certainly not a hundred trillion. How many events need to occur, to turn a shrew into a bat? Certainly more than two, and most need to occur at the same time.

Long fingers

Webbed skin

Changes in muscles

Light bones.

NOW the requirement for multiple events to all occur becomes a hurdle

My work on trpA.

Now, this sort of study, where researchers fail to find evolution, are not easy to track down. People don’t trumpet their failures. They were side-results from other studies. As I looked in the literature, I could find no indication that anyone had deliberately tried to find what the limits of evolution were. So, I decided to do it myself.

My approach was to take a well-studied gene- the α subunit of tryptophan synthase. This is the gene that performs one of five steps needed in making the amino acid tryptophan. This is a well-studied gene, and there were a number of point mutations that were known to inactivate the gene- single changes in the DNA sequence that resulted in single amino acid changes in the protein, resulting in an inactive tryptophan synthase. Now if a microbe does not have a working version of this gene, then it won’t grow unless you provide it with tryptophan. But what happens if it evolves- regains a functional trpA gene? Then it will have an ENORMOUS advantage- it will keep growing, even after the medium has been depleted of tryptophan. Pretty soon, it will completely dominate the culture in which it is growing.

I then proceeded to introduce a series of changes- mutations- into the trpA gene. If multiple steps are the problem that other experimental evolutionists have shown it to be, then my mutant trpA genes with only one mutation should evolve just fine. However, those with TWO, THREE, or FOUR mutations should have trouble evolving- a lot of trouble evolving.

Let’s say my bacteria has only one mutation in its trpA gene. Now, if I have 10 million cells- this amount would fit into a drop of water easily- I will have one evolved bacteria, able to make its own tryptophan.

Now let’s say that my bacteria has TWO mutations in its trpA gene. Now my problem has grown immensely.

BOTH changes have to be restored by random processes. In order to have, on average one evolved bacteria, now I’ll need 100 trillion bacteria- to grow this many bacteria, I’d need about 100 liters of medium- a fair amount, but I should be able to find it.

However, what about three mutations? I would need a billion trillion bacteria. Now, I’m into some large numbers. I would need a container that held a billion liters- a cube 100 meters on each side- BIG.

What about four mutations? Now I’d need a 10,000 trillion trillion bacteria- and a container that held ten thousand trillion liters. Actually, there’s one that just about that size- Lake Superior, the largest freshwater lake in the world, which holds 12 thousand one hundred trillion liters.

The only ones that we have tested so far are the mutants with 1,2, and three mutations. As expected, our mutant with 1 mutation evolves readily. We’ve checked about a 100 billion cells, and at this point have yet to see our gene with 2 mutations evolve.

CONCLUSION: When evolution requires two steps, and nothing happens unless both take place…nothing happens.

V. Why isn’t this approach used more often?

I think it is because, being convinced of the truth of evolution, the scientific community does not want to be bothered by the details, and would prefer to let the subject of how complex structures were produced by random processes rest with the past, or be answered by untestable theories.

A few years ago, I submitted a grant proposal to NSF; one of the experiments that I wanted to try was the evolution of a “difficult” trait- specifically the evolution of lactose utilization without benefit of the tricks that Barry Hall used. I was not funded, of course. But the response of some of the reviewers was informative.

1. Can long-term evolution demonstrate the evolution of “difficult” traits?

On [this] question (evolution of “difficult” traits), we certainly know that long-term evolution (really long-term) has created "difficult" (complex) traits such as photosynthesis, DNA replication, protein synthesis, cell division, nitrogen fixation, transformation, toxins and many more.

Another reviewer:

What can be said if the answer is no?

VI. Conclusion: What can evolution really do? Not much when you ask it to do two things in order to succeed. And this, I am convinced, severely limits what it can do in nature.


Creeper in blue and Radar in black

Hi Radar, it's always amusing when you take a foray into the global warming/cooling debate, since it is so completely at odds with your YEC beliefs. How you manage to maintain both these positions (YEC and just about any position on global warming/cooling that is supposedly based on science) in your head at the same time is truly baffling. The problem is this: All climate research*, whether it argues for or against global warming, is based on data that presumes an old Earth, i.e. one that is more than 6,000 years old. All these quotes that you so adoringly cut-and-pasted above... are all based on research using methods that you, Radar, think are completely wrong. Your position on the global warming and cooling should not be on either side, but a huge protest that it's all based on nonsense, i.e. the presumption that the world is far more than 6,000 years old.

Ad nauseum argument. You are simply wrong. Climate research includes records that have been strictly kept for many years, recorded by man. There are also accounts and records within the historical timeframe of mankind in which conditions and observations tell us the basics of what climates were like at different times in mankind's history. Evidences from the last few thousand years based on human observation and documentation yield all the information I need and depend on to make conclusions about how the climate works.

As soon as you say that you agree with the validity of either side, you are implicitly acknowledging that dating methods indicating an old Earth are correct.

Since you deny the validity of dating methods that indicate an old Earth, the only plausible conclusion you can draw based on the knowledge that your belief system allows you to accept is that you just don't know - the world could be cooling or it could be warming.

Actually, if you actually read my article about global cooling, you will see that current scientific measurements based almost entirely on readings we have taken in modern times tell us that a time of cooling is happening right now. I don't depend on forensic/historical "science" making guesses about ice cores, I depend on evidences from the last three thousand years or so that show us that it is the Sun's activity that determines whether we will get warmer or cooler. It also tells us that we are finding that the Earth is a sophistcated system in which the results of global cooling trigger reactions that help warm the earth and vice versa, all of which take place completely independent of the doings of mankind. Your premise is wrong and so your conclusion is also wrong.

If you live most places in the USA, you simply have to look out of your window to remind yourself that we are in a period of global cooling rather than warming. Record low temperatures, record snows, major ice storms, glaciers growing all over the globe and the Antarctic ice growing like mad are not signs that global warming is a problem right now. Go back and look at the actual temperature readings around the globe for the 21st century and you will see a decrease rather than an increase.

Furthermore, there is not enough and cannot be enough CO2 in the atmosphere to make much of a difference in the global climate. People tend to miss this part of the equation and climatologists who want grant money don't mention it. It is not even possible for CO2 to cause any significant warming of the planet.

Furthermore, records kept over the last several hundred years indicate that CO2 levels follow temperature rather than the other way around. Yes, Virginia, there IS a correlation between CO2 levels and global temperature change. But it is backwards. CO2 levels chase temperature change rather than the other way around.

(* I may be wrong - maybe there is a YEC outfit that has somehow found a scientific YEC explanation for, say, ice core data that also yields some kind of useful information regarding global warming/cooling. If so, please point me to it and I'll be happy to stand corrected. But it seems extremely unlikely if not impossible to me.)

Dating methods are particularly problematic for you, not only with regard to this inconsistency re. your global warming/cooling stance, but also with respect to dendrochronology and ice core data. You claim to have answered these questions, but I've shown quite clearly that that wasn't the case - that this is an argument that you are still running away from (even though, IIRC, you once announced that you were going to do a "series" on dendrochronology - but when faced with some critical questions, you abandoned that after the first post; I'll try to find the post for you).

I'll be happy to pick up both of these arguments again, since you fled the scene of the debate both times.

Okay, well we can and will address ice cores or dendrochronology and you can pick which one first. Present your evidences that you want me to refute so I can address them. But you will first have to abandon your assertion that I depend on Old Earth arguments and data to defend my positions on climate, because it simply is not true.

Yes, I do use data from scientists some scientists who think the Earth is much older than a few thousand years old. But I use their data from recent times and ignore their theoretical data from a supposed and unproven distant past. In other words, I depend upon observational science and experimental science but not historical science to teach me how the climate works.