Skeletal Morphogenesis Horizon Problem

I have a question. A biggish question, which takes a little explaining. I apologize in advance for the length... but bear with me. A little analogy may help the medicine go down, so here's a description of the HorizonProblem?:

Look up. You can see the universe stretching away from you for about 14 billion light years (ok a lot less. But if you wore better glasses...) Now look down, through the earth. (use your imagination). You can now see the universe in the other direction for about 14 billion light years, and it looks much the same. The universe is only about 14 billion years old; these parts of the universe (28 billion light years apart) cannot possibly have come into contact. So how come they look so similar? (The common explanation is the InflationModelOfTheBigBang?.)

fractal nature of gravitational expansion - any part of the universe will look similar to any other at some level

Ok. Now look at your left hand. Then take a glance at the right. Amazingly alike, aren't they? Ditto your left hand and anyone else's left hand. Here is another HorizonProblem? (another example of this phenomenon which amazes me on 'planes is the regularity of the distance between cirrus clouds, but I can think of a plausible explanation for that one).

There are only a few plausible explanations for large-scale coordination:

Coincidence. However, there are so many examples of bodies that coincidence causing similarity is unlikely.
Design. If you can believe in an all powerful god then there's not much point in trying to determine causality; you can always <insert god here>. [That is, if you hold a cartoon version of religion, the kind only held today by fundamentalists and uninformed atheists, rarely by serious theologians.] So I'm not going to discuss this option. However, be my guest. You could be right.
(Physical) Connection. ie things look the same because they were connected in the past, and have radiated from that point of similarity. This works well for the universe (ie inflation explains a lot) but not for the body (are you connected to a chimp at birth? So why are we so alike?). This one might explain symmetry though - please correct me, I don't know enough about embryology to say if our two arms split from one, but I doubt it.
Rule (or Logical Connection). There is a generative rule that causes us to look alike and have symmetry. This follows from believing that DNA contains our blueprint, but saying its so doesn't make it so. For example, crystals grow by a rule. They have a coordinated small scale structure in distant locations, but often not a coordinated large-scale structure. At the other end of the scale, sand dunes have a common large scale structure but zero small scale structure (the rule being their critical slope)
(BTW I welcome suggested alternatives!)

Generating body shapes by rules seems like an ok proposition to begin with, but what about the detail? Our bodies contain many structures which are largely similar, but differ greatly in detail. E.g. no two fingerprints are alike, but we (mostly) all have 8 fingers and 2 thumbs. This suggests that the rules are not just rules like those which grow crystals, but that they are largely independent of the small scale detail to achieve their aim. We are living proof that there are billions of near-miss body plans which look much the same at their largest scale. We also have a lot of small scale similarity though, in that cells specialize in the same way in all of us. We don't have liver cells in our hearts or bones in our brains.

We're nearly at the question. It should be obvious by now that I want to know why we are the shape we are, but first I want to point out and reject two kinds of answers to this question. If you do a search for craniofacial morphogenesis (as I'm sure you do at least once a week! ;O) ) then you'll find lots of information for orthodontists, and what they call morphogenesis. Including (so called) theories of c-f-m-g (my contraction). However, these are all entirely driven from data collected from real cat scans. They describe what has been seen. They are, in short, empirical. Secondly, cell biologists do studies of something they call morphogenesis. However it works at the smallest scale, looking at (eg) deposition of calcium to form cylindrical bones. I have yet to find a paper on the web about the 'big picture' - of whole body morphogenesis, which looks at why the body plan is robust to small scale changes. Maybe I'm not looking hard enough, or I'm looking for the wrong thing - do you know?

One thesis I read included a quote from Wolpert to the effect that you should not concentrate on reproducing effects, since many causes can have the same effect; instead focus on proving causality.

Well, bugger causality. If body plans are as robust as they appear to be, then we ought to be able to 'grow' a human in silico using a wide variety of rules, without paying a great deal of attention to small scale processes.

Its worth noting that the body plan is not only robust to small scale environmental effects during growth (people are stunted more often than they grow a third arm) but they are clearly also robust to a great deal of mucking with their DNA; the chimp body plan is not much different to ours, and miscarriages are mercifully rare enough for us to assume that random crossover of our genes consistently produces, well, people. So not only is the thing produced by the rule robust to changes while it acts, the rule itself is robust to change.

Ok so why do you have five fingers? Why do most mammals? Why is your skull the shape it is? Why only one joint in the arm? This is the problem I have with Morphogenesis.

Anyone got some answers? -- BrianEwins

Five fingers? Someone has said that this is about physically maximum for correct function - more makes some fail to function/limited freedom/etc.

One joint in your arm? Odd, I have 3 (shoulder, elbow, wrist)

Skull shape? Why do you consider this shape significant?

saying we evolved to 5 digits for some reason does not answer the question of how the embryo shoots at the target of 5 and hits. That's about the evolution of the rule, not the rule itself. And the problem is that the rule acts on a tiny scale (protein production in cells) and produces consistent, diverse large scale effects despite variation in the small scale. Your joint question is facetious. Saying 'its not one its three' (which you can then extend to lots more to take account of metacarpal bones) does not answer the question of: why always 3. Why is skull shape interesting? Well, consider a starfish with 5 appendages. Now divide the tips of those appendages in 5. Look, its got little hands and feet, and a little hand where its head should be! In other words, the rule for the skull is quite different from our other bony protuberances. Why? Its also interesting in that its composed of many bones which fuse into this shape; complex large scale coordination. And in that the shape is clearly constrained (animal skulls are quite recognizable as skulls, deformed from our own.) -- BrianEwins

Form follows function - all skulls look alike because all skulls are doing the same thing(s) - covering and protecting a brain, housing eyes, ears, and noses, and supporting a maw. Look at the cephalopods for these things done in a different manner with a different (non-existent?) structure.

If you're interesting in morphological development, the best data currently is for the common fruit fly Drosophila. The literature goes into quite some amazing detail about how the body parts form, why they form where they do and what controls it all. You can get to the literature from the NIH's pubmed site (http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed) but this is typically a little dense for the layman. As for the "why" questions, a trivial answer is that development is the way it is because other stuff has been tried and doesn't work as well. For example, the trait of having six fingers and/or toes per limb is dominant genetically, yet examination of the population shows that five digits is the overwhelming actual situation. The simplistic explanation would be that people with six digits per limb have some kind of reproductive fitness disadvantage. This may also be true for four digits per limb -- perhaps not enough. -- AndyPierce

The problem is deeper than that. No vertebrate has more than five digits per limb except for a few early amphibians. It's not so much a matter that the mutation is a disadvantage, as that it doesn't come up. Not true - there are people alive and well today with six digits. Clearly, it comes up; it just isn't very successful from an evolutionary point of view for whatever reason. [Extra claws are quite common among cats. I read somewhere that cats have a dominant gene for "polydactyly", but it doesn't precisely control how many extra claws they get.]

I have no problem with the fact that we evolved 5 (not 4 or 6) fingers. I want to know how the rule that gives us 5 works so well under perturbations. The question is not really why 5. Nor is it why are we so alike (evolution blah blah). It's how do we as individuals grow to be so alike.

So you are asking why every human develops in the same manner? Or something else?

Very briefly, as I understand it, programmed chemical attractors. If you place these around a circumference, you get a sheet, for example. Configuration depends on extended neighbourhood. Initially you have a configuration of spots on a cell wall where other cells can attach. Symmetry is actually approximate and easily messed up. There is also an electronic component to the whole process but this is poorly understood. On the simplest level, growing ends of bones tend to have a higher potential than the shafts. This mechanism reactivates when bones are broken and may be augmented with small batteries. This is hardly a full explanation, but I think it indicates that the complexity of the process is sufficient to do some clever things; like maintain an extended symmetry using only local information. -- RichardHenderson

Five is a Fibonacci number (1,1,2,3,5,8,13, ...). Many flowers have numbers of petals or seeds which are Fibonacci numbers. That's what gives sunflower heads their orderly spirals. Perhaps we have five fingers because we are flower children? -- RobinWilson

The fossil record shows (unless the ones found so far were odd mutants) that early tetrapods had more digits, and ForSomeReason? that got reduced to five before tetrapods started successfully diversifying.