It is curious that through the centuries men seem to have been more interested in studying the heavens than things on Earth. Even when they did begin to investigate terrestrial matters it was the inorganic that they studied first. Exact knowledge about living creatures has come last of all among the sciences. So it has come about that Newton and his successors among physicists and chemists have usually been considered to be the only real scientists. Nor can we say that things have changed much recently. The use of high energies to break up atoms and to blow men to the moon is still acclaimed as the most fundamental form of inquiry. Man seems to have an urge to “get to the bottom of things,” to break them up, and he continually expects to find “ultimate” particles of which all are made. It may even be that there is some special property in our brains that makes it seem so obvious that such analysis is the most profound form of knowledge to which we should all aspire. Or is it the result of cultural and economic influences in the West over recent centuries?

A little thought surely makes one wonder whether this cult of analysis is really as sensible as it seems. Most of the phenomena that we meet around us are not simple things at all but highly complicated systems of interacting parts. Surely the most challenging problems are how such complexes arise and are maintained. Especially for the study of living creatures including ourselves much more than a knowledge of ultimate particles is needed. Yet it is only in the last twenty years that biology has found, through the genetic code, the means for understanding how it is possible for these enormously complicated groups of molecules to maintain their organization for millions of years. The secret is of course that each single mass of living matter is sooner or later destroyed, and then replaced by new individuals, made by embryological processes directed by a new set of instructions.

At least since Aristotle men have speculated about how a wonderfully elaborate creature can emerge from a simple little egg. In Ontogeny and Phylogeny Stephen Jay Gould traces for us the history of the two sorts of explanation of this mystery of development, solutions that depend upon two different approaches to science. If one is to believe one’s eyes the chicken is not there in the egg. Therefore some outside influence must be at work to put in the legs and eyes, the beak and the feathers and all the rest. This is the theory known as epigenesis, and to find the outside force thinkers have invented all sorts of mysterious and mystical agents. Aristotle postulated a sequence of increasingly higher “souls” that enter the human embryo, first the “nutritive,” then “sensitive,” and finally “rational,” an idea that still appeals to some religious controversialists today.

The other theory about development is preformation, which holds that one should not trust one’s eyes, since they are not good enough to show everything that is in the egg or sperm. The chicken must be there all the time and with good enough lenses one would see it. Indeed the very earliest microscopist, the Dutchman Leeuwenhoek, when he saw human spermatozoa for the first time in the seventeenth century, thought he could see a little person in each one of them.

Which of the theories is right; which of them is more truly “scientific”? Here are problems indeed, and as Joseph Needham once said, the history of the discussion of them is almost synonymous with the history of embryology. The epigeneticists seem to be more “objective,” they won’t let us imagine what we cannot see. But they have to postulate unknown forces to make the parts of the body appear. The preformationists seem to avoid that mystery; but if the embyro is “really” there all the time, then it seems that all human history must have been encapsulated in the ovaries of Eve—“the homunculus in the egg, the next homunculus in the egg of the homunculus, box within box within box, to prospective future generations of incredible tininess.” At first thought no idea could be more absurd, and yet for Charles Bonnet in 1794, “This hypothesis of encapsulation is one of the greatest victories that pure understanding has won over the senses.” And perhaps, as we shall see, he was not so far wrong.

What can the answer be? As is often the case the truth lies in between. The preformationists correctly said that the explanation cannot be intervention of mysterious forces like the soul, sent in from outside. Something must be there all the time. But what is it? Until this century no one had realized that what we have to look for is not some tiny complete organism in the egg, but the coded instructions of how to make one. The preformation idea was partly right. The “something” that is there is information, and this has indeed been accumulating in the code of the nucleic acids ever since the origin of life. It needed such inventions as player pianos and computer programs to teach people to look for coded instructions. So the epigeneticists were also right when they said that the organs are not all there in the egg, and their gradual appearance is not an illusion.


Modern discussions mostly imply that the victory was wholly for epigenesis, probably because the thought of Eve’s ovary is so absurd. But the really serious difficulty of the old preformationist theory is that it assumes a perfect identity of replication. It allows no place for change and therefore it died with Darwin and the Origin of Species. But the epigeneticists were left without any explanation of how the organs appear in the embryo or of how changes in this process have occurred during evolution. Those who study embryonic development have wrestled with this question ever since and are still only beginning to glimpse its solution.

The life history of every individual does in fact largely resemble that of its ancestors, rather as the preformationists suggested. In spite of evolution much remains little changed. For example, all mammals have a series of gill pouches when they are fetuses, and indeed our ears are the remains of the first gill chamber, still communicating with the mouth, as we recognize by the pain in our eardrums when an airplane descends. Early embryologists were very much struck by such facts and they developed the doctrine that the life of each individual recapitulates the history of the race. In technical terms, that “ontogeny recapitulates phylogeny.” Taken literally this implies that you have gill slits as a fetus because your ancestor was a fish (true) and that new stages have since been added to the previous fish to make a mammal, but this is false. The question is at what stage in a life history does novelty appear? According to recapitulationists, evolution proceeds by addition at the end. “Higher” creatures have new characters added at the end of their development.

We know now that this is not true and that changes may take place at any stage. The apparent “recapitulation” is the result of the “conservatism” of living things. Making a complex animal from a simple spherical egg is an amazingly intricate process. As evolution proceeds new genes can modify the course of development, but they never wholly alter it. So the old mechanisms such as those that make gills were modified but not completely lost when land animals evolved.

One might think that this was clear enough, but the idea of recapitulation has been extremely powerful and widespread and still persists today. As Stephen Gould says, it had, and still has, an influence on human affairs probably greater even than the theory of natural selection. The great recapitulationist Ernst Haeckel produced in 1899 a book called The Riddle of the Universe (Die Welträtsel), which sold half a million copies in Germany alone. Ostensibly it was based on science but it preached doctrines of racial purity and the right of favored races to dominate, which were such powerful features of the ideology of Nazism. Recapitulation provides a perfect basis for the racist. “Children of higher races (invariably one’s own) are passing through and beyond the permanent condition of adults in lower races. If adults of lower races are like white children, then they may be treated as such—subdued, disciplined and managed.”

There have been many such connections of solutions of political and educational problems with recapitulation, all based on the assumption that evolution consists in the addition of later and better stages to the previous imperfect ones. If ontogeny recapitulates phylogeny every normal child passes through a primitive “savage” phase, and should be treated accordingly. Again, according to the Italian Lombroso, the founder of “criminal anthropology,” delinquents are arrested at a lower stage of mental and even physical development so it is obviously no good trying to educate them. He did however suggest that 60 percent of criminals lack those supposed physical stigmata of primitiveness and their delinquency must be due to circumstance—but the other 40 percent should be segregated as brutes. His ideas were not quite so pernicious as they seem because they were an attempt to reform the prison system by isolation of the really black sheep. But social reforms cannot be based upon unsound biological reasoning. Unfortunately there have been many further recent examples of such reforms.

Both Freud and Jung were firm recapitulationists. Jung’s idea of archetypes is based upon recapitulation, and he wrote that “childhood is a state of the past.” For Freud “each individual somehow recapitulates in an abbreviated form the entire development of the human race” (Introductory Lectures, 1916). He combined this with a Lamarckian belief in the inheritance of acquired characters. Even in his last book, Moses and Monotheism (1939), he says, “I cannot picture biological development proceeding without taking this factor into account.” He knew of course that biologists reject Lamarckism, so what is one to make of a scientist so ready to ignore the evidence? Charitably we can suppose that he probably had not really read or understood the data of modern genetics. The psychoanalyst Sándor Ferenczi carried the idea of recapitulation to absurdity, supposing that the latency period following infantile sexuality recapitulates the ice ages. Gould comments, “Though lest one wonder why we didn’t do ourselves in by declining to copulate during cold times, Ferenczi assures us that the ice ages only redirected some of our genital drives to the development of ‘higher’ intellectual and moral activity.”


But besides these and many other wild attempts at explanation of the human condition, there has been a steady growth of understanding of how development is in fact modified during evolutionary change. In the last century it was pointed out by the German von Baer that all eggs are rather alike. In fact the young stages of animals are more like each other than the adults. Then the British zoologist Walter Garstang showed how new characters may arise at any stage of development. They can come to appear either earlier or later in development by the process known as heterochrony.

The time of appearance of characters can change independently. Thus if the genital system matures relatively faster than the rest of the body, the result may be a sexually mature larva, as in the case of the famous Mexican Axolotl, a species of salamander which in some lakes becomes sexually mature while still in the tadpole stage. The terrestrial part of the life cycle is thus completely eliminated, the condition called neoteny. Garstang expressed this in one of his poems, published posthumously:

But when a lake’s attractive, nicely aired and full of food,
They cling to youth perpetual, and rear a tadpole brood.

There are obviously many possible variations in the times of development of different parts of the body. Earlier maturity can be achieved either by acceleration of growth of the sex organs or by delay of growth of the body and so on. The various combinations have attracted the attention of those zoologists who like inventing new names, as I know to my cost, since my own tutor in Oxford, Gavin Rylands de Beer, was the worst offender, in his book Embryos and Ancestors (1928). Gould spends many pages trying to disentangle the confusion that de Beer introduced and readers may join him in thinking, as we students did, that it was never really necessary. However his mass of Greek names introduced us to a host of fascinating creatures with bizarre methods of reproduction. Change in the relative rate of development of various parts of the body has been one of the most powerful agents in evolution. For example, if a larval environment presents a great abundance of food then a species may exploit it by accelerated development, known as paedogenesis, ensuring that the larvae reproduce rapidly. Everyone knows how quickly greenfly (aphids) multiply on the rose bushes. They do it by parthenogenesis: the eggs develop unfertilized within the mother; then another embryo begins to develop within the first and a third within the second, so that the grandmother comes to contain several generations. This telescoping can theoretically allow one female to produce 524 billions in a year.

Similarly gall midges lay their eggs on mushrooms and to exploit this ephemeral bonanza the larvae develop partheno-genetically in the mother, but when further larvae develop then the first generation dissolve their tissues to provide nourishment so “when the larvae emerge their own parent is no more than a hollow shell.” Gould adds, “Greater love hath no woman.” When the mushroom has finally all been eaten a new sort of larva appears and develops into sexual forms whose offspring fly off to find new food.

Many zoologists believe that several of the major classes of animal have been evolved by similar “paedogenetic” changes. There is some evidence that the ancestors of vertebrates were once rather like the larvae of starfishes or sea urchins. Then by neoteny these larvae became sexually mature adults and finally evolved into fishes, while the star-fish stage disappeared from the cycle altogether. How odd to think that once your ancestor was rather like a sea urchin or a feather star!

Such theories are difficult either to prove or to disprove, but they are of especial importance to us because of the suggestion that man is a neotenous ape. There is increasing agreement that our evolution during the last five million years or so has been largely due to retardation of the time of development of sexual maturity. To put it in the crude terms of the Dutchman Bolk who originated the theory, “man is a foetal ape” or, as I prefer to put it, we achieve manhood by never growing up to ape-hood. Many of our characters as adults are like those of baby gorillas or chimpanzees, for example our large heads and short jaws and sparse hair.

However it is better not to insist too much that an adult man is a fetal ape, which, taken literally, is rather an absurd proposition. The truth is more likely to be that we have evolved by retardation of the time of sexual development. Most mammals become sexually mature shortly after they have finished growing (or even before). But in humans there is a long period of slow growth, followed by a rapid “adolescent growth spurt.” During the period while he or she is small and dependent on the parents the child learns to make use of the large brain with which he or she is endowed. This long childhood has made possible our characteristic social and cultural developments, and has also given us a lifetime longer than that of any other mammals. As Bolk put it, “The life of man runs slowly.”

Stephen Gould follows all the arguments that this theory has provoked, in his book, which is lively and readable, though technical in parts. He explains that large changes in relative rates of development of some organs of the body may be produced by small modifications of the genes that control them. “Alterations in regulation form the major stuff of evolutionary change.” He does not include the exciting evidence that the pineal gland is the source of the hormone melatonin, which holds each individual for so long in a childish state. In animals such as ferrets the sex organs are active in the spring and for the rest of the year they are inhibited by this hormone. There are a few cases of human children whose pineal has been destroyed by a tumor, with the result that they become prematurely sexually mature, perhaps even at four years old.

So it may be that some genetic factor influencing the rate of development of the pineal was the novelty that conferred all the benefits of childhood upon us. In adult humans the pineal no longer produces melatonin. It becomes calcified, and makes a convenient landmark for the radiologist. We have no other use for it since our sexual activity is continuous, not fluctuating like that of ferrets. If the pineal gland has given us the power to achieve all our culture then Descartes was not entirely wrong in calling it the seat of the soul—though for quite mistaken reasons.

This Issue

June 1, 1978