Saturday, 3 April 2010

Eye Evolution

A good example of where belief in evolution is maintained because people aren’t prepared to look at the detail is in the supposed evolution of new organs such as the eye.

The eye is the classic example of a highly specialised organ, considered by many pre-Darwinian scientists such as John Ray as incontrovertible evidence of design in biology. Even Darwin recognised that the eye was a challenge to his theory, but in the Origin speculated how it might have arisen progressively from a simple light-sensitive tissue through a series of variations.

In the 1990s a couple of Swedish scientists, Dan-E Nilsson and Susanne Pelger (Proceedings of the Royal Society of London Series B - Biological Sciences, 256:53-9), expanded on this sort of scenario, illustrated as follows:

Starting from a patch of light-sensitive cells (which is a huge presumption in itself, though I can’t expand on that here) it is envisaged that an eye evolves by a flat patch of cells becoming a depression, which gradually deepens into a small pit (a to c), the neck of which then narrows (d). Each of these stages, taking place over several generations, is driven by the advantage of increased optical acuity (better resolution). When this stage has been reached, further improvement can be achieved only by addition of a lens (e), and the authors boldly assert that ‘Even the weakest lens is better than no lens at all, so we can be confident that selection for increased resolution will favour such a development all the way from no lens at all to a lens powerful enough to focus a sharp image on the retina.’

The totally unjustified assumption in this scenario is that if a variation will offer some advantage, then we can be sure that it will arise. No thought whatsoever is given to the crucial question of how those variations will arise. I think this blind spot (!) has arisen for two reasons.

First, before we knew about genetic and molecular mechanisms, it was thought that biological tissues were innately plastic in the sense that variations would arise spontaneously, and favourable ones could then be passed on. However, we now know that the formation of morphological structures – whether it be an eye, feather or leaf – is not by some sort of vague plasticity, but through the closely orchestrated action of many genes. So new structures need new genes. But in the above scenario, all that we have learned in the last 50 years about the biochemistry of tissues and the molecular mechanisms involved in forming tissues is totally ignored.

The second reason arises from the fact that much variation is possible through the mixing of genes that are already available. For example, it’s been known since well before Darwin that domestic varieties of crops and animals can be developed by breeding selectively from those individuals which have the desired variations (which have arisen naturally). But it was also well-known that there are limits to the amount of change that can be achieved this way. Which is why, although artificial selection could validly help to illustrate natural selection, Darwin’s contemporaries also knew that domestic breeding could not support changes such as the evolution of new organs. We now know why: new organs need new genes and molecular mechanisms to construct them - which are not available in the genomes of the original parents.

This oversight is illustrated by the fact that the above-mentioned authors’ calculation of the rate of eye evolution is based on selection from an existing pool of genes. Whereas there can be no doubt at all that the evolution of an eye would require very many new genes – for several proteins used exclusively in the eye, and for the molecular mechanisms that construct the eye in the course of embryological development. So their comment about a lens arising simply because it would be advantageous to do so is just ignorant wishful thinking - scarcely science at all.

There are so many speculative scenarios for the evolution of new structures - whether they be for eye, wing, feather, limb or whatever - but they are no better than those available in the 19th century - because they are based on the assumption that biological tissues are plastic, and completely ignore the genetic and molecular implications. If proponents of evolution want their scenarios to be taken seriously then they really do need to take on board the genetic and molecular detail.


Curt said...

The eye is one of the best cases for an intelligent designer---please see There are actually about nine design design themes that define the basic design concepts of all eyes. It would be extremely difficult to evolve from one design theme to another. Also, the image processing software, sensor design, focusing, light control,and optical-mechanical design are beyond evolutionary capabilities.

David vun Kannon said...

Sorry, David. It is your unjustified assumption that advantageous variation won't arise, not the other way 'round.
The sheets of cells grow by doubling (splitting and growing) individual cells. If the middle of the sheet undergoes a mutation in the cell signalling genes that allows those cells to continue dividing while the outer cells have stopped, the entire sheet is going to be placed in a compressed state. The basic physics of the situation will make the sheet bulge inwards or outwards to relieve the unequal perssures. If the sheet bulges inwards, you have the beginnings of the eye such as we have in vertebrates. If it bulges outwards, you have the beginnings of the eye as it developed in arthropods. All from slightly tweaking how long a growth factor signal is being sent.

Your second point on the limits of variation is important to think about. We can, with atificial selection, make dogs that differ in size from chihuahuas to Great Danes. That is several hundred percent difference in size and weight, just by turning on or off the growth signals for cells. Most of the eye development is the same thing - more cells, not different cells. But cells can change in many ways due to growth signals targeting individual proteins as well. For example cells that eventually become a lens produce more crystallin - a dense, transparent protein. Photosensitive cells would produce more opsin.

Mutations can fine tune these things, but just making more of them is a key advantage to the animals having them.As your reference to artificial selection shows, making more or less is easy.

Mike said...

If God DID design the eye, he did a miserable job! Macular degeneration, cataracts, glaucoma, I could list 20+ diseases that should not exist if the design weren’t a patchwork of evolution. As for eyes themselves, study the octopus eye, it comes from SKIN cells, not brain cells (like humans) but ended up looking about the same. Insects took another path (compound eyes), clams another (blue sensors). In other words, evolution came up with several solutions using different processes. Also, research has found it takes a lot fewer gene changes to make what appear to be big changes – lots of our genes can be found in fruit flys, but do different jobs in humans – or do multiple jobs depending on when/what they are interacting with. Also don’t minimalize the time it took for evolution to occur – it’s not “several” generations, it’s like MILLIONS of generations. Like a scientist, do the simple math – the Cambrian started 542 MILLION years ago, and assume breeding times of 3 months (Mice reach sexual maturity by 4-6 weeks) you have over a BILLION generations. Walking “several” steps is a lot different than walking a billion steps…)

Undisclosed Blogger said...

On David vun Kannon's comment.

I think that Michael Behe (the author of "Darwins' blackbox") is right in pointing towards problems of wishful thinking in macroevolutionary scenarios. In fact, he argues that just mechanically maintaining the curvature of retina tissue (which is a new and supposedly acquired thing if we step from a light sensitive spot to a curved retina tissue) would take a whole raft of new protein-protein interactions. A new element in the structure, such as a lens in the eye, would take a huge gap in biochemical complexity! One might argue that there are mechanisms such as gene duplication whereby a gene "clone" known as a paralogue is freed from the "pressure" of natural selection and, because of this, theoretically it can accumulate any mutations until such times as it says to the cell: "Look, I have become useful in a new way, use me." This is termed as preadaptation. However, there has been research on bio-complexity showing that the abilities of protein preadaptation are extremely limited, which leads to prohibitively low probability of phylogenesis. E.g. Douglas Axe in "The Limits of Complex Adaptation: An Analysis Based on a
Simple Model of Structured Bacterial Populations" shows that for bacteria the number of simultaneous base changes in paralogue genes must not exceed 6 if the changes are neutral to the organism and only 2 if the changes are malfunctions!

I have an engineering background and I find it extremely hard to believe that mactoevolution is probable. But you do not need to take my word for it. I can say that historically, many information theorists, mathematicians and physicists starting from Einstein up until today just could not be reconciled to Darwinian macroevolution. I can't remember the name now but some information theorist in the 60-s proposed treating life as a given such as mass, energy and time. We simply can't scientifically (sic!) explain life. How it started, science can't know. Dawkins type of explanations are rubbish to serious science and therefore can be simply dismissed. He oversimplifies things.

Somebody clever actually showed the number of mutations and the time interval necessary for something as complex as the eye to emerge by random mutation. It is far greater than the number of atoms in the observable universe (10^80). There is a bound for some specific regions of proteins saying that only 1 in 10^63 amino acid residue substitutions can potentially lead to new functionality. What chance are we talking about here?! Of course, one can choose to believe in 0 probability, but I believe in God.

If you decide to rely on chance, you lose out on one important thing, the meaning. I personally prefer theories that provide an insight into the meaning of things.

David Swift said...

Hi. Thanks for your comments which I'll respond to as soon as I can.

Please note that I've dissallowed one comment which merely quoted a verse from the Bible: I want to discuss the science of evolution.

Undisclosed Blogger said...


On another point, it might probably be of interest to you that in mathematics there are rigorous proofs available showing one simple thing: science is incomplete. In other words, no matter how elaborate you theory is, you will need axioms "from outside" (axioms are something you propose as a given). So the more detailed your theory is, the more axioms you will need. E.g. Eucledis axiomatised that two parallel lines will never intersect. This produced Eucledian geometry. Then Lobachevsky came and dismissed that axiom BUT his work resulted in a new and more sohisticated geometry with more axioms.

For the interested, I can quote the Tarsky theorem which states that in any given arithmetic, the notion of "truth" is inexpressible. Also, Goedel's two theorems of incompleteness state that (roughly) you can have a formula that cannot be shown true or false with a complex enough given system of axioms (where "complex enough" just means that it is of any use and not trivial). It has been demonstrated that the Tarsky theorem subsumes the other two as special cases.

To me, this means that there simply is reality that our reasoning cannot grasp. Life is part of it. What is beyond is philosophy and I am afraid of driving off the topic here :)

Llamamom said...

I'm a novice to all of this, but isn't there a law in evolution that states that if something isn't used for a period of time that it goes into disuse and eventually becomes a vestigial organ or falls away completely? So, putting it simply, if you had an optic nerve appear by chance, but waited generations for, let's say, a retina to then appear by chance, you only have a limited amount of time you can wait or the optic nerve will fall into disuse and won't even be there for the retina to attach to? And this scenario has to play out successfully for each new part of the eye over thousands of generations or you just don't get an eye?

Rich said...

Great Post

As Italian geneticist Giuseppe Sermonti, declared, “Science has taken on the wager… and lost.” Sermonti was the editor of the longest running biology journal in the world, an authority on evolution.

In the new book entitled "Evolution, the Extended Synthesis" published by MIT, Massimo Pigliucci and Gerd B Műller, along with the other Altenberg 16, clearly details why Modern Synthesis is in a crisis.

The theory of evolution was once a theory in crisis during the twentieth century, now evolution is in crisis without a theory.

Andrew Chapman said...

Coming back to the initial 'patch of light-sensitive cells' - on which Nilsson and Pelger said:

'Taking a patch of pigmented light-sensitive epithelium as the starting point, we avoid the more inaccessible problem of photoreceptor cell evolution (Goldsmith 1990; Land & Fernald 1992)'.

Goldsmith's article is long and I have admittedly only skimmed it, but there is nothing in the abstract or that I could see from a quick look to suggest that he answers the question of how the first photoreceptor cells supposedly arose. Land & Fernald start with molecules, then have just a short section on photoreceptor cells, before moving on to the whole eye. They focus on the difference on two types of cell, one with their photoreceptive membrane formed from microvilli, one with it formed from cilia. Both papers take it for granted that the eye did evolve, so their focus is on possible phylogenetic pathways, and not on any estimate of whether the whole thing is remotely possible in the first place.

Needless to say, these photoreceptor cells are highly complex. Are we to suppose that a parent without any such gave birth to an offspring with a patch of them? As a way of thinking about this, how surprised would you be if your baby was born with such a patch? Is there any reason why the cells should conveniently arrange themselves on a plane rather than in a clump? Is there any reason why it should be on the head - thinking now of a possible pre-fish ancestor like amphioxus - rather than somewhere else? Indeed is there any reason why it should conveniently be on the surface of the body rather than inside it? And what about the neurons or whatever is needed to convey the information to where it is needed? Is there any reason why these should be conveniently provided along with the photoreceptor cells?