Evolution ("by non-random natural selection")

The reason I put quote marks in the title is that "evolution by non-random natural selection" is a phrase that Richard Dawkins uses pretty often, so I'm quoting it and thus need quote marks.

The brackets are there to point out one important aspect of evolution and of this discussion, but the main topic is evolution in general so brackets were one way to highlight that.

I've posted a great deal about this subject, so I have a temptation to go through some of my recent (and not so recent) posts in many different threads and collect quotes until I have an OP, but I'll probably produce a much better text if I write it from scratch.

I'm going to define "evolution" and "natural selection," using arguments and examples from Darwin and Dawkins. They said it all so well that I think it's probably impossible to say those things better. When I'm explaining evolution, I'll probably at some points touch some subjects that deserve to be addressed; so I'll leave a series of notes at the bottom of this post.

Evolution means "change over time." One proper use of the term would be that the US Constitution was born of a set of previous ideas and social, political and geographical circumstances and has then evolved to its current form, with each amendment a step on the evolutionary process.

The term "evolution" is usually used in the context of the evolution of living organisms. In Darwin's theory¹ of evolution by natural selection, evolution is postulated as the mechanism by which life has reached its past and present diversity. Unlike its portrayal by Creationist propagandists (such as Kent Hovind,) the theory of evolution doesn't claim to explain the history of the universe or of our planet except for life itself, and neither does it attempt to explain how life emerged in the first place.²

Evolution, in the sense that the word is used in biology (which from this point on will be the only sense of the word I'll be using in this post,) works like this:

• First, you need an imperfect replicator. A "replicator" is anything that automatically makes copies of itself. Self-replicating structures are found in nature, and not all of them involve life. An imperfect replicator is something that makes copies of itself but makes small mistakes every now and then.

• When you have an imperfect replicator making imperfect copies of itself, there is going to be variety in the copies. Some of the variations (the aforementioned "mistakes," or "mutations") are going to be harmful to the replicator, some are going to have no statistically significant effect (this type of mutation is called a "neutral" mutation) and some are going to be beneficial.

• In the environment in which the replication occurs, there are many variables, such as temperature, the composition of the atmosphere, the types of terrain (or air or water) in which the replication occurs, etc. The environment also changes sometimes.

• As mentioned, some mutations are harmful. What is meant by that is that some mutations make the new replicator less likely to succeed at copying itself or even surviving until it can copy itself. Neutral mutations have no significant effect either way. Beneficial mutations make the replicator more likely to copy itself successfully and to survive in functioning condition.

• These different varieties of the replicator are part of the environment in which they "live." In other words, what they do changes their environment (e.g. by making resources more scarce by using them up or by changing the shape of the terrain.) Thus, a variety that was previously average at copying itself might be more or less well suited to another environment, and thus it might become a "weaker" or "stronger" replicator without changing itself, as a result of the change of the environment.

• When there are different varieties with different characteristics, there is competition. The same environment serves as the background for the "life" of all types of replicators, so when resources are scarce or threats emerge, some are going to lose out, while others will be more capable of survival and successful replication. This is natural selection.³

• Now that there is selection among the different types of replicators, the statistical trend will be that newer replicators are going to be on average stronger (where "stronger" means simply "more capable of survival and successful replication") than their ancestors. From this stronger generation, new generations with new mutations will emerge, and they will also be subject to competition and selection. Note that it is by no means random which individuals get "selected" for survival and which are doomed to die without sufficient offspring to keep the line alive. Instead, it is the environment that determines which traits are beneficial, and thus the development (or "evolution") of the replicators over time is not a random process.

• Sometimes, some lineages (or "families") will be separated from the main group, by migration or by some natural events (such as a river changing course as a result of a flood.) These "families" will then go on to "live" in their own environment, evolving (by random mutation and non-random natural selection) into a direction that will usually be different from the one that their cousins in the main group are taking. This is called "speciation," though that term is mostly relevant when discussing "species."

Now that I've laid out the foundations of evolution and natural selection, you should be able to see how a very primitive replicator, one that we wouldn't even consider "life" at all, could evolve into a more elaborate kind of replicator, one small evolutionary step (that is, mutation followed by selection) at a time.

At this point, I'll try to prevent some misunderstandings and answer some arguments that are likely to occur.

If at any point a mutation had occurred that would be beneficial to the organism in a different environment, or if there had been one that requires many other mutations to work, that individual would be selected against, as it would be spending resources on traits that aren't helpful and thus its competitors would be able to breed faster and at lowed expense, eventually drowning out the "weak" one. This is the point of the failed quest for "irreducible complexity" undertaken by the Intelligent Design movement.

The only way that a complex feature can emerge is by a cumulative build-up of small evolutionary changes. It is possible that one complex feature requires many mutations to evolve, and that those mutations alone would be harmful and thus should not be able to survive until the build-up to the "reward" of the complex feature is completed. This is true, except that some of the steps in the build-up can be neutral (in which case they would basically lie dormant in the genome of the family until a mutation that co-operates with the neutral one occurs and gives the yield of a complex feature,) or they can be beneficial, as is the case with the gradual evolution of the eye.⁴

Thus, there is no giant leap from simplicity to complexity, but a slow, cumulative process by which complexity is approached from simplicity.

If it was possible to travel back in time and pick up all of your direct ancestors, there would be no point in the line where you could point out an incident of speciation (or branching of a line that leads to a different species from the main branch of the ancestors.) Each animal would be able to breed with its offspring and its parents. (When two animals are unable to produce viable offspring, they are classified as different species.)

But if you were to pick two samples from the line, hundreds of thousands of years apart, you would immediately be able to tell that they're related, but not the same species.

All life on Earth is related. We all share a common ancestor, and by the process of speciation, that original family has branched into a glorious tree of life. More than 98% of all species that have ever lived are long extinct, but there is vast diversity in life on Earth still, and new species are born all the time - it's just that the process is so slow that it takes a long time to detect it.

---

1: The word "theory" has two very distinct meanings. The one that is used in common parlance is synonymous to "conjecture." That sense of the word "theory" is very weak, in the sense that not much weight is placed on the credibility of that theory.

In science and academia, "theory" has a very different meaning. To quote Wiktionary, a scientific theory is "[a] coherent statement or set of ideas that explains observed facts or phenomena, or which sets out the laws and principles of something known or observed; a hypothesis confirmed by observation, experiment etc."

It is that sense in which evolution is referred to as a "theory," just like the theory of gravity, Einstein's theory of general relativity and the germ theory, which is the model that describes how microscopic organisms, such as viruses and bacteria, cause diseases.

When an opponent of evolution asks why we don't call it the "fact of evolution," the correct answer is two-fold. 1: Yes, we do; and 2: we also don't call "germ theory" "germ fact," because the word "theory" refers to the entire framework, so "theory" is a more precise word for this function. If it had no evidence to back it up, it would be called "the evolution hypothesis" or "the evolution conjecture."

2: "Abiogenesis" is the term used for the emergence of life from non-life. Similar to how there is no distinct point at which one species can be said to have split from its ancestor (as any pair of organisms from that lineage would be able to interbreed if they were a sufficiently low number of generations and mutations apart,) there is no clear point where "non-life" became "life."

Instead, it's a gradual process and the only reason we can make a distinction is that we can compare life at an advanced state (such as a cell) to replicators at a low level of complexity. Where we draw the line is a matter of definition, and it will never be very clear.

3: Darwin was the first to use the contrast between "natural" and "artificial" selection as a tool for explaining how natural selection works and what the time-frame is like.

"Artificial selection" refers to any situation where there is an intelligent agent deciding which individuals in the sample (or collection of organisms) get to breed and which individuals they are paired with.

If you want a huge cow, you need a normal flock and time. What you do is, you pick the largest ones and allow them to breed (well, not with cows, but you know.) The small ones, you remove from the process.

If you keep doing this, generation after generation, you'll eventually end up with cows that are on average much larger than the ones you first had, as you will have artificially created an environment where the size of a cow is the primary trait by which it gets selected.

Humans have been doing this with cows for over 10 000 years (which is longer than the age of the Earth, according to Young Earth Creationists.) Similarly, dogs have been domesticated for possibly as long as 33 000 years and at least for 15 000 years, and all current varieties have been produced via artificial selection from wolves.

Natural selection differs from artificial selection in that there is no plan and no intelligent agent doing the selection. Instead, selection can be driven by other organisms or by other features of the environment.

A change in the environment, such as an ice age, can cause extinctions. In the surviving species, they causes sometimes dramatic changes, when traits that in the past were very successful can become harmful, and traits that were neutral or harmful in the past suddenly (or gradually) start surviving better and are thus selected for. When the environment changes again, so do the conditions for survival, and again the species changes, possibly back the direction it used to be.

A more direct means of natural selection is provided by other organisms: parasites and predation. If, for instance, there was an ant with a red, round gaster, it might sometimes be mistaken for a berry by birds. Thus, the ants with the most berry-like gaster would be selected against, and so the species would on average evolve a smaller gaster with a different colour.

4: The eye is often chosen as the prime example of "irreducible complexity," mainly because it serves as a platform for a delicious quote from Darwin, which is always mined without context. In The Origin of Species, Darwin gives this quote, followed by a detailed explanation of how the eye evolved. Quote mining target in teal, explanation in black:

The evolution of the eye is now well understood, and it is indeed a gradual process. The question, often asked by Creationists, of "what use is half an eye?" is staggering in its stupidity. It assumes that evolution claims that a half an eye or a complete eye evolved, in a single leap, from something that wasn't an eye at all.

The correct question would be "what use if half an eye compared to 49% of an eye?" and then the evolutionist could go on to explain what kind of advantages might result from an organism having a slightly better eye than the rest of its kind.

The eye began as a small collection of nerves that could detect light very faintly, without making a distinction for which direction the light was coming from. Such "eyes" exist in some life-forms today.

From there, the next step would be to develop a cavity, at the bottom of which the light-sensitive cells would be spread. This, too, is easy to imagine as a natural process. This gives a limited ability to detect the direction from which the light is coming, and is thus slightly more useful than having those cells on a flat platform. The deeper the cavity, the better the ability to discern the direction from which the light arrives.

After a deep, round cavity is formed, the cavity starts to round out. The result is a large, hollow ball with a "pinhole" on one side and light-sensitive nerves spread out on the inside walls. This gives good ability to detect which direction the light is coming from.

The next step is to develop a transparent membrane over the hole, to prevent objects from entering it. This also allows the development of means by which the water inside the eye can be controlled to produce the best possible results, and every step of this process can easily be imagined as a small step that gives a tiny but significant advantage over the competition.

Now that there's a membrane over the pinhole eye, the next step is to develop a lens. This part of the process seems varied and kind of technical, so I'll just give you a Wikipedia link which contains an explanation about the different ways a lens might form. A lens allows the light to be focused and thus gives a huge advantage, though a primitive lens is of course not much more use than no lens at all. To get the huge reward, a long process of gradual steps is needed.

The final steps are the evolution of an iris and a cornea. There are currently living organisms with each kind of eye described here, from just the light-sensitive cells on a flat platform all the way to the eye of the eagle, so it seems that the mutations that result in the next step don't occur in every lineage or that the benefits for that species are so small that they don't result in an evolutionary advantage sufficient for natural selection. Still, the eye has evolved in many different families of species, so it's not an overwhelmingly unlikely chain of events.

---

This thread is for discussion about evolution. I also want to use this thread as a platform on which those who don't know much about evolution and those who don't believe it at all can ask their questions and I'll try to answer them.

Of course, I'm no expert, either, but evolution is such a simple concept that I think even an amateur can defend it, especially as I have read very much about it. There obviously are people on these fora who know much more about evolution than I do, and I'd be pleased to have them here, too.

In this first post, I didn't really discuss the evidence for evolution. I think I'll probably try to talk about different types of evidence as the questions start to build up, as I don't want to inflate this OP any more. The reason I let it balloon to this size is that there are some very common misconceptions that result in the same questions every time: "how come monkeys aren't giving birth to humans?" and the afore-mentioned "what use if half an eye?" and so on. I wanted to get rid of the most predictable problems before we begin.

Note that I haven't proofread every part of this post, as it simply is so overwhelmingly huge that I can't find the energy for it now that I'm done writing it (in one session!) so there are bound to be mistakes, especially as I am not a native speaker of English. Please be so kind as to point out any mistakes and we'll have an OP that's easy to read and interesting.

So, any questions? Anyone wanna elaborate on some of the points I made in my OP? Have you perhaps seen or read something interesting that has to do with the topic? Have at it!

edited 5th May '12 10:53:40 AM by BestOf

Quod gratis asseritur, gratis negatur.

A minor comment:

As far as I know, it is not clear if the changes are always slow and gradual. This is a matter of some debate about evolution researchers; but I heard that there is some support for punctuated equilibrium — the idea that in evolution there can be long periods with very little change, and then relatively short periods of rapid change.

As I said, it is uncertain (as far as I know) if this is really the case; but it seems pretty certain that there have been periods (the Cambrian explosion, for example) in which evolution "moved" much more rapidly than in others.

edited 5th May '12 9:43:55 AM by Carciofus

But they seem to know where they are going, the ones who walk away from Omelas.

Aron Ra is the best You Tube resource I've ever found regarding the subject of biological evolution. His Foundational Falsehoods of Creationism and Falsifying Phylogeny series are just fantastic, although not all those vids have to do with biological evolution, and they are heavily influenced by the political fight he is forced into (he's a phylogenetics researcher in Texas).

Share it so that people can get into this conversation, 'cause we're not the only ones who think like this.

More quickly than usual, but still much too quickly to be detectable over a human lifespan. Punctuated equilibrium is still gradualism, just variable-speed gradualism.

Well, it depends on the lifespan of the creatures. We have observed the development of new metabolic processes in bacteria within human-observable spans of time, after all (I cannot find a link to the article, but it was about metabolizing citric acid if I remember correctly.)

But yeah, evolution takes multiple generations, obviously.

edited 5th May '12 10:11:59 AM by Carciofus

But they seem to know where they are going, the ones who walk away from Omelas.

I think you should put in the full quote from Darwin, even if in another color to distinguish it from the "mined" part.

"And as long as a sack of shit is not a good thing to be, chivalry will never die."

One thing that I have been reading a little about recently is Artificial Life — the study of "lifelike processes" (and, in many cases, of their evolution) in a simulated environment. It's cool stuff.

For example, there is Tierra, in which the "creatures" are self-reproducing programs in a simulated computer, and in which phenomena of parasitism, co-evolution and so on have arisen.

Or there are other programs which simulate the evolution of body plans and movement patterns — here is a nice video on youtube, for example...

EDIT: And here is a video of virtual creatures evolving to catch balls and carry them.

edited 5th May '12 10:40:29 AM by Carciofus

But they seem to know where they are going, the ones who walk away from Omelas.

You're talking about Richard Lenski's experiments. Here.

Quod gratis asseritur, gratis negatur.

Thanks.

EDIT: I'm reading about Schlafly's reaction to Lenski's paper, and about Lenski's reply to Shlafly. Um. Wow.

edited 5th May '12 10:53:24 AM by Carciofus

But they seem to know where they are going, the ones who walk away from Omelas.

A question: Are the various breeds of dog (or any animal in which distinct breeds have emerged thanks to human intervention, for that matter) essentially examples of human-guided evolution?

With cannon shot and gun blast smash the alien. With laser beam and searing plasma scatter the alien to the stars.

To a degree, yet. They're still the same species, in that there are very few subspecies of dog that can't interbreed (I think Great Dane + Jack Russell is still possible, for example). When hybrids between dog subspecies come out infertile then we'll be looking at ring species evolution (A can mate with B, and B with C, but not A with C).

Share it so that people can get into this conversation, 'cause we're not the only ones who think like this.

@pagad: Domestication itself is inherently an example of human-guided evolution. Cows, dogs, chickens, goats, sheep, horses, cats... we have shaped them and continue to shape them.

"It's Occam's Shuriken! If the answer is elusive, never rule out ninjas!"

According to current classification, all domestic dog breeds are not only the same species but the same subspecies, a variety of wolf, Canis lupus familiaris. Any barriers to dog interbreeding are logistical (highly disparate size makes mating unfeasible), not genetic. Not many actual mutations have been involved in the production of dog breeds—instead, selective breeding has reshuffled the existing canine genome to isolate desired features. It is unlikely that humans will effect a true speciation via dog breeding.

God, I sound pedantic.

But can dogs breed with wolves?

Anyway, you're right that as long as a Great Dane can technically (never mind the logistical issues) mate with a poodle, then they're all the same species.

edited 5th May '12 12:08:15 PM by Fighteer

"It's Occam's Shuriken! If the answer is elusive, never rule out ninjas!"

Dogs can absolutely breed with wolves, and are likely to if they are kept together. It's not recommended because the mix of "tame" and "wild" genes messes with the pups' heads and makes them much more dangerous than either a pure dog or a tamed wolf, but it's easily possible.

Indeed, because "docility towards humans" is precisely one of those traits we have encouraged to proliferate in domesticated canines. Most of them, anyway. Fighting/attack dogs keep those traits and are much more dangerous as a result.

"It's Occam's Shuriken! If the answer is elusive, never rule out ninjas!"

First of all, Im not an expert of the topic. And now, following a well known tradition, I will spam a pair of categorical assertions after saying that:

1. It´s really difficult talk about species. We cant breed with dogs or dolphins, despite attempts, but in a lot of times the bonds aren´t clear. Lions and tigers can have fertile descendants and except some african people all of us have some % of Neanderthal heritage.

2. Humans have grown in power and now can take decide his fate better than yesterday. Eugenics it is still a taboo, but we are -and we were-. pressing our evolution and that influence has more weight than in other ages. And we simply dont know where are we going clearly. Or if it matters and all war now it´s in cultural survival. Muted flesh by aesthetic desires. Behaviour changes wich leads to meat ones and vice versa. We have been done before, and we cant blame to enviroment now.

I´ve lost. I´ll go from some coffee liqueur to the fridge.

Edit: Awww, you already have posted my first point before.

edited 5th May '12 12:54:26 PM by Picheleiro

About dogs and wolves: one very interesting hypothesis about how dogs came to be is that wolves domesticated themselves, to a degree.

When the agricultural revolution started, human communities were for the first time able to settle in one place for a very long time. Thus were born the first cities, through various stages (much like in Civilization IV - hut->hamlet->village->town->city.)

As soon as we settled, we started to use land that was set aside for dumping our trash, such as useless bones and rotten bits of meat and so on. And where there's a scrap heap, there's a scavenger.

So, wolves would have feasted on our scrap heaps, but the problem is that those heaps were still close to human settlements, so humans, too, would occasionally interrupt the wolves who were having the time of their life there.

The social aspects of a species are hereditary, either by genetic or by memetic processes. There in fact are genes for social behaviour. Wolves would have had genes that make them likely to flee when approached, so early proto-domestic wolves would have lived close to human settlements and been partially dependant on our scrap heaps, but they would also have been easily frightened and thus they would have fled as soon as humans arrive.

Through mutation, individuals that are more likely to stick around as humans approach would have evolved, and they would have been able to eat longer than the ones who fled, thus becoming stronger in the long term as they had shorter and fewer periods of malnutrition. On the other hand, they would've been more likely to be killed by humans if those humans happened to be willing to kill a wolf for any reason.

Thus, there was an intermediary - a level where a wolf would get just enough food while avoiding humans that get too close. The genes that approach that level would have been selected for, and thus they would have become the dominant form in that pack of wolves.

At this point, humans would start to get accustomed to having wolves around, and perhaps some wanted to see if wolves might be any use, probably for hunting. And that would start a process where humans approach wolves, trying to get wolves accustomed to humans, while wolves are naturally evolving towards greater tolerance and attachment. So as soon as wolves start to willingly live in human communities, and humans are willing to have them, domestication can begin. And while it would have looked to humans like it was their idea, in fact the wolves would unknowingly have been going in that direction for a while already.

Symbiosis is of course a very common phenomenon in nature, and it is very likely that it always emerges as a side effect of some mutual, unguided but non-random development like this.

edited 19th Nov '12 3:17:45 PM by BestOf

Quod gratis asseritur, gratis negatur.

I've seen that theory. I've also seen a corollary to the effect that by the time deliberate domestication started, the beasties in question didn't look like wolves anymore, but more like dingos (or the result of several generations of stray dogs mixing and interbreeding). This sort of change seems to be closely linked to docility in canids, as experiments with foxes bred strictly for tameness have also resulted in their ears drooping and markings becoming similar to collie markings.

Question: I'm not really trying to challenge this concept, but I'd still like it clarified a bit—why do we assume that everything on the planet evolved from a single organism? Is it because everything on the planet shares at least one identical part of our genome? Or do we just assume this because there's no way to prove that we evolved from multiple populations, and so we just pick the simplest solution?

I ask this, because if life evolved from one single tiny organism on an entire planet in all of history, wouldn't that imply that the event of life occurring, even in an environment that has everything it needs to start, is ridiculously unlikely to happen? Because if life is so rare that it only ever happened once, and only once for just one microscopic cell, then that doesn't bode well for, say, our search for life on other planets.

The only reasonable explanation I can come up with is that our ancestral cells simply outcompeted all other forms of life, but even then, you'd think that there would be at least a few simple cell populations alive with ancestral genes that aren't exactly like everything else's, but are still more or less the same in practice.

edited 5th May '12 4:09:40 PM by Vellup

They never travel alone.

Well first of all, to imagine a cell as a primitive organism is a mistake. A single modern cell is very complex, and primitive organisms were much simpler than modern cells.

But the reason we assume that all life evolved from the same ancestor is that we do indeed carry evidence of that in our genome. Every genome that has ever been looked at (and that's a huge number) has had some shared genes with every other genome on the planet, and we can in fact through a meticulous and ridiculously expensive process actually compare the genomes of any two species to see how much they have in common and where the similarities and differences are, though obviously you usually can't compare a chromosome in one species with that of another if they don't have that particular chromosome in common, though there are exceptions. (If one chromosome in species A splits into two, then it's possible to find the larger chromosome in species B and notice where the split occurred.)

Not only is our code shared, but equally importantly, the genome of every species is coded in a very simple language of 5 nucleobases (there are three kinds of base pairs - AT, GC and AU (well, and their opposites TA, CG and UA.) The nucleobase U only occurs in RNA, where it replaces T.) That all life uses those same nucleobases is evidence that they all have shared ancestry; if they didn't, they would probably use other molecules or indeed other types of molecules for the same purpose. If we were ever to find an organism that doesn't use the same nucleobases as us, we would have a very hard time coming up with a scenario in which it was related to us, and of course it probably wouldn't be.

edited 5th May '12 5:43:29 PM by BestOf

Quod gratis asseritur, gratis negatur.

Query: I heard that the scientists were trying to make yeast that consumes crude oil (or something), but they couldn't do it with genetic engineering, so they just put different species in a barrel of crude oil, put the lid on, and wait. Long and behold the yeast evolved to use crude oil as food.

Yet I can't find the paper for this hilarity. You've heard of it?

There's a problem with that thought and that's the concept of horizontal gene transfer. I'm not sure of the exact biochemical mechanisms but essentially there are ways that single cell organisms that normally reproduce by asexual mitosis can swap genes with other single cell organisms. Thus its possible that there were multiple independent lineages that arose independently but be the modern period their ancestors have become thoroughly mixed so sorting out what came from which is a non-trivial task, assuming its possible to get enough information to do it at all.

edited 5th May '12 5:50:30 PM by KnightofLsama

Its the idea of the lottery.

Yes. Its very unlikely for it to happen. But theres a second part. even [peering in just our galaxy. theres a LOT of worls with the same general; life causing conditions as our world. and we know enough to know our world is not unique or unlikely to occur.

But given how LARGE we know the universe is, theres likely TRILLIONS of potentially life friendly worlds in the universe. And as the old saw goes, somebody's gonna win the lottery every now and then.

edited 5th May '12 5:55:10 PM by Midgetsnowman

About the chances of there being life on other planets: here's a thread I made about exoplanets and life in their Goldilocks zone. (If you don't know what that means, head over to the thread and read my epic OP. You'll learn a lot. If you know what the thread's name means, well, you'll learn a lot less but I've got some stuff up there about recent discoveries and calculations about the frequency of "Goldilocks zone" planets in the universe.)

The most relevant bit for this thread is that there are about 100 billion galaxies in the universe, and in our galaxy alone, there are probably tens of billions of planets in the Goldilocks zone.

edited 5th May '12 6:09:12 PM by BestOf

Quod gratis asseritur, gratis negatur.