Simulating Evolution

I recently read an interesting article on simulating Evolution that was posted by ordinarygirl.

This IS a very interesting article! Way cool!

I've seen experiments in simulated evolution before, but this one is one of the most surprising yet.

However, I want to share my own opinion that many of these simulations of Evolution operate under an incomplete view of Evolution. They see Evolution as just natural selection through some kind of "survival of the fittest" mechanism.

That is definitely one aspect of Evolution, and it is an effective mechanism to keep a species adapted to a gradually changing environment.

However, natural selection is not the only driver in Evolution. The other is natural potential.

If you have a pair of mice, and give them easy access to unlimited food and resources and territory, you will have an exponential population explosion of trillions of mice that will cover the entire surface of the Earth with a solid layer of mice in under 5 years.

The fact that our planet isn't covered in mice, even after millions of years, is a reflection of their limited access to resources, and the "virtual" death toll of trillions of unborn potential mice. This outweighs the death toll caused by maladaptive traits, and so is potentially a larger driver of Evolution than merely "survival of the fittest".

If a small number of species can find a way to exploit an untapped resource, or develop a unique adaptation to a given environment, then that variety will temporarily experience that explosive growth potential that leads to the biggest leaps in Evolution.

While this new variety rapidly grows and flourishes, the old variety continues to maintain itself in the manner it always has before. As natural selection continues to put pressure on both varieties to maximize their adaptation to their own environments, this causes the two varieties to pull apart until they become distinct species.

Dog breeding, for example, produces a WIDE variety of dogs. But it doesn't produce a new species. This is because the only mechanism at work is selective breeding (semi-natural selection). There is nothing fundamentally changing in the dogs environment. A dog is still a dog and does doggy things and eats doggy food provided to it by doggy people. :)

But if a group of dogs were able to adapt to a different environment with a different manner of survival...one in which they had different interactions with species around them, and yet had little competition and thus could survive while they adapted to their new life...then you would start to a true split into a new species, and a wildly successful one at that.

To use an analogy: Imagine a large and established corporation. Such an organization is generally only capable of gradual changes as needed to meet changes in it's environment. A large corporation that gets caught up in rapid change usually goes extinct. (Just like well established animals cannot survive rapid changes in their environment.)

But now imagine some of the employees from this large corporation go off and start a new business of their own, taking advantage of some untapped or newly developed market. They might struggle at first, but if they have no immediate competition, and if they are successful, they will grow like crazy with a completely different "corporate DNA" than the parent company. Since this company started small, it had the ability to make rapid fundamental changes as it grows. Established corporations can't do that.

Rapid progress in any industry comes from small companies blooming, not from from killing off (Naturally Selecting) failed companies.

Likewise, having a small population of animals allows favorable genetic changes to spread more quickly and not be drowned out by the masses.

So, a true simulation of Evolution would include simulating a vast and changing multi-variable environment, with an incalculably large number of possible interactions and dependencies. This kind of environment with such vast numbers of possibilities is needed to give all species a chance to stumble into something that is new and yet survivable. And this is what is needed for a new species to arise.

Most of what I've said here is in Darwin's original Origin of Species book. Darwin quite rightly figured out that the Evolution of animals was a far more complicated phenomenon that simply "survival of the fittest". Darwin did not use that phrase in his book. It was coined by someone else later.

So, the whole point of my whole posting is this: It is very tempting to want to simulate evolution with a computer, and see what kind of virtual species you can create. But understand that such a program, if it is to be accurate, would have to be far more complex than any software written to date. Simply trying to create new species by selective breeding is not going to mirror reality. You indeed might create new and surprising species, but it won't be in exactly the way nature would have done it. Such programs tend to succeed by having far more "open-ended" possibilities for viable mutations than what nature actually provides.

If anyone is interested in Evolution, I highly recommend reading Darwin's original book cover to cover. It is a far more complete and insightful coverage of the theory of Evolution than I have found in any other book. (And surely more than I have found in any science class.)

Modern books on Evolution try too hard for the dumbed down approach to explain Evolution in simple and obvious terms. But you miss out on something important if you don't follow all of Darwin's research and reasoning from beginning to end.

Oh, and if you are going to read Origin of Species, you need to find a reproduction of the FIRST edition. There were six editions total, and unfortunately, in each edition, Darwin kept adding crap or tinkering with part of the book to better address questions that were raised in those days. The first edition is best at standing on it's own.