Climbing Mount Improbable Page 11
There is something that could come to the evolutionary aid of the freak macro-mutant, namely the fact that the effect of a given gene depends upon the other genes that are present in the same body. The effect of a gene on a body, its so-called phenotypic effect, is not written on its surface. There is nothing in the DNA code of the achon-droplasia gene that a molecular biologist could decode as ‘short’ or ‘dwarf. It has the effect of making limbs short only when surrounded by lots of other genes, to say nothing of other features of the environment. A gene's meaning is context-dependent. The embryo develops in a climate produced by all the genes. The effect that any one gene has on the embryo depends upon the rest of the climate. R. A. Fisher, whom I've already quoted, expressed this long ago by saying that some genes act as ‘modifiers’ of the effects of others. Notice that this doesn't mean genes modify the DNA code of others. Certainly not. Modifiers simply change the climate in such a way as to modify the effects of other genes on bodies, not the DNA sequence of the other genes.
As we have seen, it is not (quite) totally inconceivable that a parent with a six-inch, tapir-like proboscis could have produced a mutant child with a five-foot elephant-like trunk in a single generation, due to a one-gene change — a macro-mutation. It is very unlikely that the new nose would immediately behave like a good, working trunk. This is where ‘modifier’ genes, and the notion of a ‘climate’ of other genes, could theoretically come to the rescue. So long as the macro-mutation is at least approximately good for something, so that individuals possessing it don't die out, subsequent selection of modifier genes can clean up the details and smooth off the rough corners. Think of the advent into the population of the major mutation as equivalent to a cataclysmic challenge, like an ice age. Just as a new ice age causes a whole collection of genes to be selected, so also does a drastic muta-tional change to the average body, such as a sudden elongation of the nose.
Genes that ‘clean up’ in the wake of a new major mutation do not work only on the major gene's most obvious effects. They may act on some unexpectedly distant part of the body to compensate for, mitigate the ill-effects of, or enhance the possible benefits of, a major {104} mutation. In the wake of a greatly enlarged nose, since the trunk increases the weight of the head, the bones of the neck will need to be strengthened. The balance of the whole body may change, with further knock-on effects, perhaps on the backbone and the pelvis. All this consequential selection works upon dozens of genes affecting many different parts of the body.
Although I introduced the idea of ‘cleaning up afterwards’ in the context of major macro-mutations, this kind of selection is certainly important in evolution whether or not there are any macro-muta-tional steps. Even micro-mutations arouse consequences such that ‘cleaning up afterwards’ is highly desirable. Any gene can act as a modifier of the effects of any other. Many genes modify each other's effects. Some authorities would go as far as to say that, of those genes that have any effect at all (lots don't), most genes modify most other genes’ effects. This is another aspect of what I meant when I said that the ‘climate’ in which a gene has to survive consists largely of the other genes of the species.
At the risk of spending longer on macro-mutations than they deserve, there remains one possible source of confusion that I must anticipate. There is an expertly publicized, and not uninteresting, theory known as ‘punctuated equilibrium’. To go into detail would take me beyond this book's scope. But, because the theory is heavily promoted and widely misunderstood, I must just stress that the theory of punctuated equilibrium does not have — or should not be represented as having — any legitimate connection with macro-mutation. The theory proposes that lineages spend long periods in stasis, undergoing no evolutionary change, punctuated by occasional rapid bursts of evolutionary change which coincide with the birth of a new species. But, fapid though these bursts may be, they are still spread over large numbers of generations, and they are still gradual. It is just that the intermediates usually pass too quickly to be recorded as fossils. This ‘punctuation as rapid gradualism’ is very different from macro-mutation, which is instantaneous change in a single generation. The confusion arises partly because one of the two advocates of the theory, Stephen Gould (the other is Niles Eldredge), also independently happens to have a soft spot for certain kinds of macro-mutations, and he {105} occasionally underplays the distinction between rapid gradualism and true macro-mutation — not, I hasten to add, miraculous Boeing 747 macromutation. Eldredge and Gould are rightly annoyed at the misuse of their ideas by creationists who, in my terminology, think that punctuated equilibrium is about huge, 747-type macro-mutations which, they are right to believe, would require miracles. Gould says:
Since we proposed punctuated equilibria to explain trends, it is infuriating to be quoted again and again — whether through design or stupidity, I do not know — as admitting that the fossil record includes no transitional forms. Transitional forms are generally lacking at the species level, but they are abundant between larger groups.
Dr Gould would lessen the risk of such misunderstanding if he more clearly emphasized the radical distinction between rapid gradualism and saltation (i.e. macro-mutation). Depending upon your definition, the theory of punctuated equilibrium is either modest and possibly true or it is revolutionary and probably false. If you blur the distinction between rapid gradualism and saltation you may make the punctuation theory seem more radical. But at the same time you offer an open invitation to misunderstanding, an invitation that creationists are not slow to take up.
There is a supremely banal reason why transitional forms are generally lacking at the species level. I can explain it best with an analogy. Children turn gradually and continuously into adults but, for legal purposes, the age of majority is taken to be a particular birthday, often the eighteenth. It would therefore be possible to say, ‘There are 55 million people in Britain but not a single one of them is intermediate between non-voter and voter.’ Just as, for legal purposes, a juvenile changes into a voter as midnight strikes on the eighteenth birthday, so zoologists always insist on classifying a specimen as in one species or another. If a specimen is intermediate in actual form (as many are) zoologists’ legalistic conventions still force them to jump one way or the other when naming it. Therefore the creationists’ claim that there are no intermediates has to be true by definition at the species level, but {106} it has no implications about the real world — only implications about zoologists’ naming conventions.
To look no further than our own ancestry, the transition from Australopithecus to Homo habilis to Homo erectus to ‘archaic Homo sapiens’ to ‘modern Homo sapiens’ is so smoothly gradual that fossil experts are continually squabbling about how to classify particular fossils. Now look at the following, from a book of anti-evolution propaganda: ‘the finds have been referred to either Australopithecus and hence are apes, or Homo and hence are human. Despite more than a century of energetic excavation and intense debate the glass case reserved for mankinds hypothetical ancestor remains empty. The missing link is still missing.’ One is left wondering what a fossil has to do to qualify as an intermediate. In fact the statement quoted is saying nothing whatever about the real world. It is saying something (rather dull) about naming conventions. No ‘missing link’, however precisely intermediate it was, could escape the terminological force majeure that would thrust it one side of the divide or the other. The proper way to look for intermediates is to forget the naming of fossils and look, instead, at their actual shape and size. When you do that, you find that the fossil record abounds in beautifully gradual transitions, although there are some gaps too — some very large and accepted, by everybody, as due to animals simply failing to fossilize. In a way, our naming procedures are set up for a pre-evolutionary age when divides were everything and we did not expect to find intermediates.
We've taken a preliminary look at Mount Improbable and seen the difference between the forbidding cliffs on one side and forgiving slopes on the oth
er. The next two chapters look carefully at two of die peaks beloved of creationists because their cliffs seem particularly itcep: first wings (‘what is the use of half a wing?’) and then eyes (‘the eye won't work at all until all its many parts are in place, therefore it can't have evolved gradually’). {107}
<<<br />
* * *
>>
CHAPTER 4
TO FLY HAS FOR SO LONG BEEN A HOPELESS DREAM OF humanity and, when we achieve it, we, do so with such difficulty that it is easy to exaggerate how hard it is. Flying is second nature to the majority of animal species. To modify an aphorism of my colleague Robert May, to a first approximation all animal species fly. This is mostly because, as he actually said, to a first approximation all species are insects. But even if we take just warm-blooded vertebrates it is still correct to say that more than half the species fly: there are twice as many bird species as mammals, and a quarter of all mammal species are bats. Flying seems formidable to us mainly because we are large animals. There are a few larger than us like elephants and rhinos and we are naturally very aware of them but to a first approximation all animals are smaller than us (Figure 4.1).
If you are a very small animal, the conquest of the air is no problem. When you are very small, the harder challenge may be to stay on the ground. This difference, between large animals and small, follows from some inescapable principles of physics.
For objects of a given shape, weight increases disproportionately with (specifically as the third power of) length. If an ostrich egg is three times as long as a hen's egg of the same shape, it will not be three times the weight but 3 × 3 × 3, or twenty-seven times the weight. Until you are used to it, this can be rather an arresting thought. If one hen's egg is breakfast for a man, one ostrich egg is breakfast for a {108} twenty-seven-man platoon. Volume, and hence weight, goes up as the third power (cube) of the linear dimension. Surface area, on the other hand, goes up as the second power (square) of the linear dimension. It is easiest to demonstrate this with cubical boxes, but the rule applies to all shapes.
Figure 4.1 Living things vary in size over about eight orders of magnitude. To help sort the variation out, generation time is plotted against size (they are strongly correlated, for reasons not discussed here). Both axes are drawn on a logarithmic fcale, otherwise paper 1,000 miles across would be required to accommodate a redwood tree on the same scale as a bacterium. {109}
Imagine a large cubical box. How many smaller boxes of exactly half the edge size could you fit in it? You can see quickly, by sketching the boxes, that the answer is eight. The big box can hold not twice as many apples as one of the smaller boxes but eight times as many apples; not twice as many tins of paint but eight times as many can be crated in the box. But if you want to paint the surface of the big box, how much more paint do you need than to paint the outside of one small box? Again, you can quickly verify by sketching the scene that the answer is neither two times nor eight times, but four times as much paint.
The difference between surface area and volume becomes more dramatic when you look at objects of very different size. Suppose a match-manufacturer, for advertising purposes, constructs a man-sized matchbox, two metres high when lying flat on the ground. A standard matchbox is two centimetres high so a stack of 100 matchboxes would just reach the height of the crate. A line of 100 matchboxes would just stretch from one end of the crate to the other. And a row of 100 matchboxes would span the width. So, if you filled the crate with matchboxes, how many would it hold? The answer is 100 × 100 × 100 or one million. In one sense the crate is only 100 times as big as an ordinary matchbox, and a naive human eye may estimate that it is about 100 times as big. Yet in another sense it is a million times as big, and it will hold at least a million times as many matches (actually more, because relatively less space is occupied by cardboard).
If we assume that the giant matchbox is made of the same kind of cardboard as an ordinary small matchbox, what is the relative cost of the cardboard? This depends, not on the volume, nor on the linear dimensions, but on the surface area. The giant box would need, not a million times as much cardboard but a mere 10,000 times as much. The surface area of the standard, small matchbox is hugely greater for its weight than the surface area of the giant matchbox. If you cut up a small matchbox, you could only just stuff the folded cardboard in another small matchbox. But if you cut up our giant matchbox, the folded cardboard would hardly be noticed lining the bottom of another giant matchbox. The ratio between surface area and volume is a very important quantity. For every cubing of volume, surface area is {110} merely squared. You can express this mathematically as the statement that, if the shape is uniformly scaled up, the ratio of surface area to volume goes up as the two-thirds power of length. The surface area to volume ratio is larger for small objects than for large objects. Small objects are more ‘surfacy’ than large objects of the same shape.
Now, some important things in life depend on surface area, other important things depend on volume, other important things depend on linear dimension, and yet other important things depend on various combinations of the three. Imagine a perfectly scaled-down hippopotamus, the size of a flea. The height (or length, or width) of the real hippo is perhaps a thousand times that of the flea-hippo. The weight of the hippo is then a billion times that of the flea-hippo. The surface area of the hippo is a mere million times that of the flea-hippo. So the flea-hippo has 1,000 times greater surface area for its weight than the large hippo. It feels like common sense to say that a scaled-down miniature hippo would find it easier to float in the breeze than a full-sized hippo, but it is sometimes important to see what lies behind common sense.
Of course big animals never are just scaled-up versions of little animals, and we can now see why. Natural selection does not allow them to be simply scaled up, because they need to compensate for such things as the change in the ratio of surface area to volume. The hippo would have about a billion times as many cells as the hippo-flea, but only about a million times as many skin cells in its outer surface. Each cell needs oxygen and food and needs to get rid of waste products, so the hippo would have about a billion times as much stuff to pass in and out of itself. The flea-hippo could use its outer skin as a significant part of the surface over which oxygen and waste products pass. But the outer skin of the full-sized hippo is, relatively, so small that it needs to increase its surface area, very substantially, in order to cope with its billion-fold greater cell population. This it does with a long folded gut, with spongy lungs and micro-tubular kidneys, the whole irrigated by a massively divided and re-divided network of blood vessels. The result is that the internal area of a large animal is spectacularly more than the area of its outer hide. The ftnaller an animal is, the less does it need lungs or gills or blood {111} vessels: the outer surface of the body has a large enough area to cope, unaided, with the input — output traffic of its relatively few internal cells. A less precise way to put this is to say that a small animal has a higher proportion of its cells touching the outside world. A large animal like a hippo has such a tiny proportion of its cells touching the outside world that it has to increase that proportion with area-intensive devices like lungs, kidneys and blood capillaries.
The rate at which substances can pass in and out of a body depends upon area, but it isn't the only important thing that does. So also does the tendency to catch the air and float. The flea-hippo would be lifted by the lightest puff of wind. It could be carried high on a thermal and then float delicately back to earth where it would land softly and without injury. The real hippo, if dropped from the same height, would plummet to a terrible crash landing; and if dropped from a scaled-up proportional height it would dig its own grave. For the real hippo, flying is an impossible dream. The flea-hippo would hardly be able to help flying if it tried. To make a real hippo fly, you'd have to strap a pair of wings on it so large that ... well, the project is doomed from the start because the mass of muscle needed to power those gigantic wings would be too heav
y for the wings to lift. If you wanted to make a flying animal, you wouldn't start with a hippo.
The point is that if a large animal is going to leave the ground it has to grow large area-intensive wings, for the same kind of reason as any large animal needs surface-rich kidneys and lungs. But for a small animal to leave the ground it hardly has to grow anything at all. Its whole body is already surface-rich. There is a so-called aerial plankton, consisting of millions of small insects and other creatures floating high in the air and spreading around the world. Many of them, to be sure, have wings. But the aerial plankton also contains numerous tiny wingless creatures which float in spite of having no specialized aerofoil surfaces. They float simply because they are small, and a very tiny animal finds floating in air about as easy as we find floating in water. Indeed, the comparison goes further, for even when a tiny floating insect has wings it flaps them not so much to keep up as to ‘swim’ through the air. The reason ‘swim’ is an appropriate word is {112} that other strange things happen when you are very small. At that scale, surface tension is such an important force that, to a small insect, the air would feel all treacly. Flapping its wings must feel, to a little insect, rather as swimming through syrup would feel to us.