Progress towards complementarity in genetics

Johannes Wirz

Forschungslaboratorium am Goetheanum, Hügelweg 59, CH-4143 Switzerland


The appearance of adaptive mutations in bacteria raises basic questions about the genetic theory of spontaneous mutation and hence the concept of the generation of biological variation. Adaptive mutations were observed in bacteria exposed to selective conditions during the stationary phase of growth in the absence of DNA replication. Both anabolic and catabolic traits were affected. None of the classical explanations, which depend on errors and irregularities during the replication process, is able to account for these mutations. Various observations suggest new mechanisms for the generation of genetic variation. The theory of adaptive mutations paves the way for the introduction of complementarity in modern genetics.

Theories of adaptive mutations elaborated before the era of molecular genetics argue strongly for holistic approaches to life and heredity. They make a revision of the current concepts of reductionist biology necessary. A synthesis is presented that considers the function of spontaneous as well as adaptive mutations in the development and evolution of organisms. Both forms of mutations reflect the fundamental quality inherent among all living beings; i.e. self-relation and world-relation.


According to modern theories of heredity and evolution the tremendous variation amongst living organisms comes about in two ways, namely through spontaneous mutation and through chance hybridisation during sexual reproduction. An overwhelming number of publications provides evidence for chance variation. Because of this chance variation and DNA molecular replication (doubling) processes, which produce changes in the genetic make-up, spontaneous mutations pass undisputed as the driving force of variation and thus speciation. According to this view, in a second step, choice or selection determined by the environmental conditions sees to it that only the most fitting forms survive, thus limiting the variation which arises.

In spite of the many confirmations of the theory based on spontaneous mutation this article aims to outline and provide support for another possible theory, one in which the environmental conditions do not merely select, but direct and bring about variation. This is not intended to cast doubt on the reality of spontaneous or chance mutation, but rather to challenge its claim to absolute and exclusive validity.

The current situation in modern genetics is like that which prevailed in physics at the beginning of the 20th century. Just as at that time wave and particle theories of light were shown to be complementary views, it will be demonstrated that the present theory of chance evolution of organisms must be enlarged to include a complementary one, namely directed evolution. The theory of spontaneous mutation is placed beside that of adaptive or selection-induced mutation. Which of the two types of genetic change is realised depends on the physiological circumstances and the environmental conditions. These two types of change require different concepts for describing the relation of organism and environment and are dependent upon different molecular processes. Whether complementarity in genetics will have paradigmatic consequences for the overall understanding of living nature or whether, like complementarity in physics, it remains without effect on a wider public, remains to be seen.

There has been no shortage of attempts to develop concepts of variation other than that of spontaneous mutation. The best known goes back to Lamarck1. His was the first attempt in a modern scientific approach to evolutionary theory to explain how organismic variety arises. Lamarck's idea of inheritance of acquired characteristics, as discussed in more detail by Lefèvre2, formed an important though not central support for the theory. Whether it is justified to treat 'inheritance of acquired characteristics' and 'adaptive mutations' as synonymous is discussed in more detail below. Both Darwin3 and Haeckel4 embedded the inheritance of acquired characters. Because of this, Haeckel's biogenetic law was largely rejected (c.f. De Beer5).

As controversial as adaptive mutation is amongst modern biologists, as certain does its underlying evolutionary principle render service to convinced Darwinists (e.g. Mayr6) as an explanation for cultural evolution. Cultural advance is unthinkable without the passing on of acquired characteristics. Experiences are received inwardly and as capabilities are passed on to others (descendants). This principle is essential to the evolution of human communities. If one asks which quality is fruitful for this kind of evolution, the answer has to be cooperation. But the same question posed of Darwinian evolutionary theory gives competition as its answer. The demonstration of adaptive mutations in modern genetics is a contribution to a new understanding of nature. At the same time it leads to a humanising of natural science in that in this kind of genetic change the central human evolutionary principle finds expression in organic nature.

Spontaneous mutations

To understand the discoveries which have led to the concept of adaptive mutations, it is necessary first to be clear about the premises which gave rise to the theory of spontaneous mutation. This also means dealing with molecular interpretations.

Although genetic research was initially confined to plants and animals, bacteria soon played a significant part in answering the questions which arose. Procedures for producing pure cultures of totally different strains as well as for characterising toxin or viral resistance genes were a precondition for genetic experimentation. The other precondition comes from the bacteria themselves. Short generation times and large cell numbers made experiments possible which with other organisms would have lasted years and taken up a vast amount of space. In addition, as bacteria, having only one chromosome, are haploid, genetic changes usually show up phenotypically immediately after they have occurred.

Despite these advantages interpreting genetic changes proved to be difficult because the results were not reproducible. Whilst it is true that the phenomenon of bacterial virus resistance could be observed on repetition, the number of resistant cells in each replicate experiment exhibited wide variations.

Luria and Delbrück7, from studies for which they later received a Nobel prize, suspected that it was just this observed variability which might explain how virus resistance comes about in bacteria. They neatly hypothesised that if resistance is acquired by contact with the virus the number of resistant bacterial cells should be proportional to the total number of cells used in the experiment, provided that the probability of cells becoming resistant is the same for all cells. A series of identical parallel experiments would thus allow one to expect a Poisson frequency distribution of resistant cells. But if the mutations occur spontaneously in bacterial cultures before contact with the virus, then the number of resistant cells should be independent of the total number of cells used in the experiment - provided that the mutation event is very rare - and would simply depend on the time elapsed between the appearance of the mutation and contact with the virus. If the mutation occurs long before virus contact, the number of resistant cells will be large. If it occurs only a short time before contact, the number will be correspondingly small. The frequency distribution of resistant cells from parallel experiments is clonally determined. All resistant cells come from one and the same parent cell. Testing the variance or fluctuation can thus allow a conclusion to be drawn as to the kind of mutation which has arisen.

In their experiments Luria and Delbrück inoculated between ten and twenty tubes containing nutrient broth with 50 to 500 cells of a virus sensitive strain. After a few hours incubation the cell densities rose to about 109 cells/ml. 0.1 ml of each liquid culture was spread on petri dishes containing culture medium treated with a large number of bacterial viruses (ca. 1014). After overnight incubation, resistant cells formed colonies visible to the naked eye. Almost all the bacteria plated-out (ca. 108) were destroyed (lysed) by the virus and died.

In accordance with expectations, the results were unequivocal. The fluctuation in the number of resistant cells in the cultures tested in parallel was very great. In one experimental series there were petri dishes with no colonies and some with more than 500. The distribution of resistant cells clearly showed itself to be clonal. The mutation event most probably must have arisen before virus contact had taken place and must therefore be spontaneous or 'chance'. The virus simply selected the resistant cells.

This result was in total agreement with the hypotheses of Darwinian evolution. The resulting excitement was so great that Delbrücks warning at a conference in 1947 not to generalise from his discovery went unheard (see Stahl8). After the presentation of a paper by Ryan et al.9, in which it is shown how the number of genetic changes in a metabolic mutant increased in a matter of days, he said 'In the case of mutations of bacteria ... to phage resistance ... the phage does not cause themutations. In your case of mutations permitting the mutants to utilize succinate... as a sole carbon and energy source ... it is an obvious question to ask whether this particular medium had an influence on the mutation rate.... One should keep in mind the possible occurrence of specifically induced adaptive mutations'.

Another milestone in the development of a theory of spontaneous mutation was reached when Lederberg and Lederberg, using their replica-plating method, managed to isolate from a virus-sensitive bacterial strain cells which were resistant to the virus without having come into contact with it10. This showed that resistance mutations arise spontaneously, that is without contact with the selecting agent.

The discovery of the double-helical structure of DNA by Watson and Crick11 and the biochemical investigations of the replication events in the material of inheritance (c.f. Alberts et al.12) made possible an explanation of spontaneous mutation. In principle perfect replication of the material of inheritance is guaranteed by the physico-chemical conditions of its molecular structures. A host of proteins participate in this synthesis and minimise the errors which arise during replication. Such errors can manifest as mutations and are interpreted as the reason why evolution happens at all. It is also clear that the faithfulness of replication of DNA is directly proportional to the size of the genome (the quantity of the substance of inheritance) (c.f. Maynard Smith13). The smaller the genome the higher the mutation rate. Put another way, a text with thousands of words can be transcribed many times without distorting the meaning when one wrong word is substituted in every ten thousand. If errors were to occur with the same frequency in a text with a hundred thousand words, ten words would be altered at each transcription. With frequent transcription, distortion of the meaning could not be ruled out. To avoid unacceptable changes, the transcription accuracy would have to be increased.

If during replication of the material of inheritance the mutation rate is too high it could have catastrophic consequences for the organism concerned. But if DNA replication were absolutely perfect, undirected 'chance' evolution of living organisms would be rendered impossible. Spontaneous mutations are an essential component or instrument of the evolution of all living beings on earth. Such mutations are not determined by environmental conditions but arise mainly through replication of the material of inheritance.

Adaptive mutations: the concept clarified

In order to deal properly with adaptive mutations it is necessary first to clarify a misunderstanding and a conceptual confusion. Equating the concepts 'inheritance of acquired characteristics' and 'adaptive mutations' is often criticised (c.f. Lenski et al.14). The first explicitly emphasises the fact that characteristics must first be formed before they can be passed on. But the second concept implies that known mutations are seen to revert to the wild type. After a reversion event the cells concerned exhibit characteristics that were shown by their ancestors prior to the mutation. Reversions provide modern genetics with a tool that allows phenotype and genotype to be kept equally in view. I will use the two expressions 'inheritance of acquired characteristics' and 'adaptive mutations' synonymously, because in both cases it is true to say that there must be an effect on the material of inheritance directed from the environment and the living organism. Furthermore, new characteristics that must be inherited can manifest only through modification of already existing heritable material.

Another difficulty concerns the view that the theory of adaptive mutations is 'Lamarckian' (c.f. Marx15, Symonds16, Mayr6 ). There are several objections to this. As already mentioned Darwin and Haeckel include the inheritance of acquired characteristics in their theories, although they have both expressly countered Lamarck's teleological evolutionary theory. The term 'adaptive mutations' expresses the fact that the constraints of life and the environmental conditions not only work selectively on preformed characteristics, but also can determinenew ones. Such characteristics can be described as 'goal-directed' without, like Lamarck, presupposing an evolutionary goal. Even Darwin3 coined an expression for this: 'Effects of habit and the use or disuse of parts'.

Early supporters of the theory of adaptive mutations

Since Mendel, adaptive mutations became a topic of increasing interest and was described in reputable journals. One of the most outspoken representatives of the theory was Kammerer. On one of his trips to the USA he was even heralded by the newspapers as the 'new Darwin' (Koestler18). In many publications and using a wide variety of animals he sought to demonstrate the existence of the inheritance of acquired characteristics (c.f. Kammerer19,20). He described them for the midwife toad Alytes obstetricans. By raising the temperature of its surroundings the animal can be made to depart from its usual behaviour of reproducing in water. Under the new conditions the male forms 'nuptial pads'. These thumb-like structures occur on the forelimbs of many amphibians that reproduce under water. It is thought that they help the males get a better hold during copulation. After copulation on land, the male carries the strands of spawn containing the fertilised eggs around with him wrapped round the hind leg until the larvae hatch. Under the new conditions the spawn remains in water. The tadpoles which have undergone their embryonic development in water exhibit external gills similar to the larvae of other toads and frogs.

Both nuptial pads and external gills can be regarded as an expression of an adaptation to the new conditions. Both features also appear in subsequent generations even when the animals are returned to normal living conditions. They appear to be genetically fixed.

More convincing were the experiments with the sea squirt (ascidian), Ciona intestinalis. Kammerer described them as providing the most significant evidence for adaptive mutations. After repeated amputation of the terminal tubes which are used for feeding and excretion, these organs grow extremely long. Specimens with long tubes give rise to long-tubed offspring thus giving rise to the supposition that inheritance of acquired characteristics is involved. To exclude the possibility of prior chance mutation causing the long tubes, Kammerer removed the gonads. After regeneration of these organs long-tubed specimens once again developed out of the newly formed germ cells. Thus it seemed that clear evidence for acquired characteristics had been obtained.

Kammerer's experiments are clearly described and from their methodical structure withstand critical appraisal today. Nevertheless alternative explanations such as cytoplasmic or maternal effects that could bring about developmental modifications without changing the DNA would nowadays have to be excluded. In view of the tragic circumstances of Kammerer's death, which is interpreted as admission of his scientific fraud, Koestler18 emphasised the need for a repeat of these experiments.

In Russia, Mitschurin (see Sankjewitsch21), using the most varied cultivars investigated the questions of environmental influence on seeds and rootstock on fruit. He too observed environmental influences which were genetically fixed. But his work fell into disrepute and oblivion probably through the political polemic from and surrounding Lysenko and his unsuccessful wheat vernalization experiments.

Waddington22 and Piaget23 reported theoretical considerations, suggestions and descriptions regarding experiments on adaptive mutations which will be discussed below. At the level of molecular genetics, the phenomenon of adaptive mutations has been reported for flax (Marx15, Cullis24).

Adaptive mutations since 1988

The publication of evidence for adaptive mutations by Cairns et al.25 brought about a change. The standing of both the author, as former director of the respected Cold Spring Harbour Laboratory, together with that of the journal Nature in which the work was published, left littledoubt as to the scientific quality of the work and sparked-off discussion and controversy which has lasted to this day. Many 'main stream' geneticists felt obliged to take positions and carry out further experiments. Since then there have been a considerable number of publications describing adaptive mutations for various microorganisms and cellular anabolic and catabolic processes. Furthermore some authors tend to the view that this form of inheritance also plays a part in tumour formation (for reviews see Foster26,27).

Cairns' group investigated the frequency of reversion of a well known and genetically characterised metabolic mutant lac in E.coli. Cells with this mutation can no longer use lactose and are dependent for their growth on glucose or another sugar in the growth medium. The reversion of the mutation to lac+ can easily be demonstrated by plating out the cells onto a medium containing lactose and a colour indicator. Revertant cells form red colonies.

In an experimental design based on that of Luria and Delbrück7, analysis of the frequency distribution of sixty cultures prepared in parallel showed that spontaneous reversions must have taken place before selection. However, others appeared to have occurred adaptively only after contact with lactose the selecting agent. Further observations showed that the number of reversions increased when the petri dishes containing lac- bacteria were incubated for several more days. Obviously in the course of time more revertants were generated. Control experiments showed that reversions only occurred when the growth medium contained lactose. If this sugar was missing, or only sprayed on the bacteria after one or more days, the number of lac+ colonies remained unchanged with longer incubation. Finally, it was shown that with mutations such as valR, which are not selectable, no reversions occurred. Increase in the reversion rate only resulted when it was 'useful' for the multiplication and growth of bacteria. They were without doubt adaptive, or, as Cairns' group put it, directed.

The results stood in contradiction to the theory of spontaneous mutations. The reversions occurred only during selection and in appropriate environmental conditions. Lactose had to be present. The medium appeared to 'entice' out the reversions. Particularly noteworthy is the fact that they only took place during the stationary phase when DNA replication errors cannot occur. None of these observations were new. In 1961 Ryan's group had already published work suggesting mutation events without replication (Ryan et al.9 and Symonds28), but this received little attention amongst geneticists. The Cairns' group managed only to publish once more in their entirety the most important observations evidencing non-spontaneous mutations.

The Cairns work is also noteworthy for another reason. Since 1943 bacterial genetics has concerned itself with cells in the exponential growth phase and investigated many phenomena which determine the life and death of bacteria. But adaptive mutations occur only when cells are not dividing and even then only when the genetic change is choosing between growth/division or rest. For this reason it is possible to speculate as to the significance of adaptive mutations for natural conditions. From the still young science of the genetics of the stationary phase, there are reports which suggest that adaptive mutations occur also under natural conditions (Kolter29).

Hall's work

Adaptive mutation research was greatly extended in variety and scale by Hall, a microbiologist based in Rochester (USA). Working intensively with the conceptual problems of the new theory, he investigated several organisms and catabolic processes as part of his interest in reversions of point mutations (substitution of individual base pairs) and deletions. The conclusions he drew from this were uncertain and provisional. Where observed changes were at first adaptive (Hall30), they later were explained as spontaneous (Hall31,32), or occasionally in the following paradoxical way 'Spontaneous point mutations that occur moreoften when advantageous than when neutral' (Hall31). These he called 'selection induced mutations' in a later publication (Hall33), and he ultimately reached the conclusion that there is indeed a phenomenon of adaptive mutation, but there is no explanation for it (Hall34).

Hall's initial work was on the double mutation in the bgl operon in E.coli (Hall30). This operon codes for the necessary enzymes for the catabolism of glucosides. The individual reversion rates experimentally determined for the two mutants is 4x10-8 and <2x10-12 per cell division. Assuming that the two mutations are independent from one another, in bacterial strains with both mutations the reversion rate, given by the product of the two individual reversion rates, is 8x10-20. Such an event would never be observable under experimental conditions because at least 8x1020 cells would need testing, thus requiring at least 100,000 litres of liquid culture. Bacteria incubated for two to three weeks in petri dishes formed colonies of revertant cells able to catabolize glucosides. The reversion rate of was 2x10-8, far higher than expected. Here too reversion managed to take place in the stationary phase and only when glucosides were present in the medium.

Further work investigated point mutation behaviour in the tryptophan operon in E.coli (Hall31,33,35). Once again the reversion rate was far higher than was expected on the basis of spontaneous mutation and appeared under conditions of selection. The author also demonstrated that reversion was independent of DNA replication and increased according to the length of time cells were in contact with the selective substrate. Control experiments ruled out the possibility that cryptic growth of cells or retarded division of preexisting revertants determined the reversions. Experiments with baker's yeast Saccharomyces cerevisiae (Hall36) showed that adaptive mutations can also be demonstrated for eukaryotes.

Objections and attempts at a molecular explanation

Critical and partially justified objections to the idea of the existence of adaptive mutations were not slow in appearing. Several experiments were repeated with more stringent controls. The mobilisation of the bacterial virus Mu which the Cairns group25 observed and interpreted as a directed mutation proved to be a spontaneous mutation (Mittler and Lenski37). The high reversion rate which Hall30 had observed with double mutants was explicable in terms of the growth of intermediary genotypes (Mittler and Lenski38). Finally it was shown that the difference in reversion rates between two independent mutations (Cairns et al.25) could be ascribed to known physiological processes (MacPhee39).

The criticism had the result that in subsequent work the necessary control experiments were carried out. Thus in his investigation of the reversion of mutations in the tryptophan operon Hall34,35,36 was able to rule out that adaptive effects were arising through intermediary growth or death of cells. Both possibilities would have given a deceptive nominal increase in the mutation rate thus allowing spontaneous mutations to appear as adaptive events (Mittler & Lenski40). The criticism as to the reality of adaptive mutations eventually led to their experimentally verified acceptance.

Still unsolved was the question of how adaptive mutations could occur. The search for an explanation based on the underlying molecular processes was linked to the hope that phenomena which would not fit in could nevertheless eventually be interpreted 'classically'. The lynch pin in the structure of modern genetics is still its central dogma which states that 'information' flows only from the material of inheritance to the protein (DNA>RNA>protein). This underpins the idea that heritable changes are never determined by protein. The phenotype has no influence on the genotype. Since the discovery of retroviruses, whose viral RNA chromosome after successful infection is transcribed into DNA, the dogma is only partly valid. Adaptive mutations now threaten to overthrow it completely.

To explain adaptive mutations, various working hypotheses were formulated (summarised in Koch41) which, under selective conditions and with known molecular mechanisms would have allowed a raised mutation rate to be assumed. The postulates of three most important hypotheses are stated here.

Hypermutability: The basic mutation rate in bacteria under stress conditions is significantly raised (Symonds42, Hall31) and that amongst many chance mutations some also occur which are selected.

Increase in the mutation rate through reverse transcription (Stahl43): In cells in the stationary phase there are always transcription processes going on, i.e. DNA is transcribed to RNA. It is known that in these processes the transcription accuracy is relatively small and thus the mutation rate is increased. RNA molecules arising in this way which enable the synthesis of a protein necessary for growth can, after being changed to DNA, replace the original chromosomal sequences.

Slow repair (Stahl43): Under stationary phase conditions small pieces of DNA are broken down and resynthesised. The repair mechanisms which normally replace wrongly inserted nucleotides are not active.

All hypotheses were experimentally tested and had to be rejected. With hypermutability the frequency of the adaptive reversions in the trp operon signified a mutation rate of 0.04 per base pair (Hall32). Thus on average every 25th base pair would have to be substituted. Such a high rate would without doubt have been lethal for the bacteria. Hall investigated the relevant gene locus by sequencing to determine whether, in the neighbourhood of the necessary reversion, other substitutions had taken place. But he was without success. Such a 'directed' localised increase in the mutation rate would however have only postponed the crisis of finding an explanation.

The second hypothesis also had to be rejected (Hall31) because with some bacterial strains which exhibit adaptive mutations no reverse transcriptase activity has so far been demonstrated.

The slow repair hypothesis failed because as well as the expected selective mutations, independent mutation events in other genes would also have had to occur (Hall32). In no case could these be detected.

That the molecular basis of adaptive mutations is of a non-classical kind was revealed by a series of unexpected results, which, however, in retrospect an unprejudiced observer would hardly wonder at. Adaptive mutations always appear in the bacterial stationary phase. DNA turnover is minimal. Mutation events are time dependent. But spontaneous mutations occur by maximal DNA turnover in the phase of exponential growth and are dependent on replication.

A first indication of the difference at the molecular level in the occurrence of the two types of mutation was given by the analysis of the spectrum of reversions under selective (adaptive) and non-selective (spontaneous) conditions (Hall44). Thirteen strains with different mutations in the same gene (lac) were used to compare reversion rates during exponential growth with those during the stationary phase. The rates were as much distinguished by the two culture conditions as by the individual strains. Base pair substitutions, insertions and deletions are dependent on the physiological state of cells and the type of change in the environment.

Unlike spontaneous mutations, adaptive mutations are dependent upon various components of the recombination system (RecA, RecBCD, Harris et al.45). Under normal conditions, this system mediates homologous recombination between chromosomes and enables insertions and deletions of DNA sequences in the bacterial chromosome. If the proteins of the RecBCD system are lacking, adaptive mutations no longer take place. These findings have been described as progress towards the understanding of genetic intelligence (Thaler46).

Another piece in the jigsaw was the discovery that not only the recombination system but also intercellular DNA transduction, the transfer of genetic material during bacterial conjugation (a kind of primitive sexual pairing), participates in the appearance of adaptivemutations.

Conjugation proved significant is several ways: for bacterial strains which had the selective gene on the chromosome rather than on the transduction plasmid, the reversion rate was 25 to 50 times smaller (Radicella et al.47, Galatski & Roth48). Removal of the conjugation apparatus with detergents or additional mutations in the enzymes of the transfer function reduced the adaptive mutation rate to the same extent.

According to Shapiro49, these results have far reaching consequences for evolutionary theory, although he also holds that they make the hypothesis of 'directed mutation' superfluous. The transfer of the transduction plasmid is dependent on DNA replication, which is why mutations associated with chromosomal replication can occur by 'chance'. Even so the results show that the rate of meaningful mutations can be significantly increased by selection and that by transduction, which can be regarded as a primitive form of intercellular communication, meaningful mutations can be passed on. Recombination and plasmid transfer are cellular functions which allow an active reaction to its environmental conditions. A significant component of genetic variation is without doubt no longer attributable solely to chance events in the replication of the material of inheritance, but can only be understood by considering the relationships between living organism and the world in which it lives.

Non-molecular concepts of adaptive mutation

I hope to have shown in the foregoing that modern genetics has reached a turning point. But true insight as to the significance of adaptive mutations cannot be gained through describing molecular processes. This is because, by reducing the phenomena to molecular processes, the fundamental and qualitative differences between spontaneous and adaptive forms of inheritance are overlooked. The description gets lost in detailing DNA-protein interactions. But the differences lie in the possibility of manifesting in the most varied of ways relationships and interconnections between organism and environment and of making these available to the next generation. They are of course dependent on molecular processes, but they are not determined by them. Thus molecular genetics points to the necessity of looking beyond its current paradigm for alternative concepts and approaches to organisms and their inheritance. Paradoxically this leads first to the rediscovery of theoretical foundations which have been forgotten. I shall illustrate this with reference to the work of three individuals.

I turn first to a pamphlet essay by Steiner50. In it the 'inheritance of acquired characteristics' is seen as a consequence of Haeckel's biogenetic law. Steiner emphasises that without this law a monist evolutionary theory has no validity. The essay is in essence against the last vestiges of vitalism and the preformation theory associated with Weismann.

Monistic evolutionary theory signified a big challenge to understand 'being' and 'appearance', requiring one to grasp the organic as a process which takes place as much from top downwards (from idea to world of the senses) as from bottom upwards. In this process both aspects - the ideal in the type at work in the organism and the real as its appearance in the world of the senses - undergo changes and metamorphoses in reciprocal interdependence.

These ideas are substantially developed in an earlier essay by Steiner51 in which the relationship of the Goethean idea of type (archetype) to the organism which actually manifests is clarified and discussed. It is the essence of all living organisms that they respond inwardly to the experiences they undergo in their development and thus eventually pass them on to their offspring.

Waddington22 offered a further theoretical principle. According to him there are two distinct possibilities for genetic variation. One concerns isolated features which are altered accidentally. Industrial melanism in the peppered moth Biston betularia is a textbook example of this and for modern evolutionary biology provides irrefutable evidence for theoccurrence of spontaneous mutation. The other possibility for genetic variation concerns features which are embedded in the totality of the organism and its environment. Waddington's example for this is the forequarters of the gibbon and pangolin (scaly ant-eater): the gibbon's forelimbs point in their slenderness, length and exceptional mobility to activities such as climbing and hanging, whereas the form of forelimbs of the scaly ant eater exhibit rigidity, shortness and compactness of bone formation which are easily comprehensible in terms of a digging function. Both animals reproduce in their entire bodily make up the specific orientation of their different activities and modes of behaviour. The logical construction of the entire form is unmistakable. According to Waddington it is extremely unlikely that the extremities have come about by a large number of accidental individual changes. It seems more plausible that they were formed through the specific behaviour of the animal in its respective habitat during the course of the evolution of the species.

Waddington used Drosophila experiments to develop the concept of organismic totality and adaptive reaction to specific qualities in the surroundings. He described short term physiological changes that had become genetically fixed 'genetic assimilation'. Such changes as take place over a long period, he described as 'evolutionary adaptation'.

Piaget23, the Swiss developmental psychologist, provides a third fundamental consideration. The idea of adaptive mutations was a logical outcome of his investigations into cognitive processes and their biological basis. It is an undoubted fact that the human being gains knowledge by constantly taking in experiences and as a result of this process is able to pass on faculties. According to Piaget the act of cognition is only possible through (subjective) receptivity and (objective) external stimulus. It produces a relationship between object and subject. Thoughts are contoured by percepts and determine our intentions. Intentionality fixes what is extracted from the sense perceptible world and determines the framework of elements of observation. As zoologist, I look primarily at animals and through an interest in morphology further restrict observations to their form.

Both aspects, thinking and perceiving, reciprocally determine and alter one another. Because cognitive processes have and must have a biological basis, this in turn must have a functional structure like the cognitive process itself. Organic regulatory processes follow the same laws as those of cognition: they undergo adaptive change and development, not only physiologically but also genetically. Piaget's morphological investigations of the water snail (Lymnaea stagnalis L.) in a wide range of habitats appeared to confirm the hypothesis of adaptive mutation.

The discussion of these three authors provides more than a foundation for a theory of adaptive mutation. All three overcome the materialistic tendency in the modern view of heredity. With Steiner, the overcoming is quite explicit. Development and heredity can only be grasped through a combination of sensory and supersensible processes. All living organisms are an expression and result of ideal-material processes. The unity of matter and spirit is the basis of Steiner's monist evolutionary theory.

With Waddington the overthrow of the materialistic view of heredity is reflected in the idea of organismic wholeness, which is not to be thought of as solely material. Relationships and interactions with the environment belong just as much to the organism as to its organs, cells and molecules. The basis of his evolutionary theory is unified life of organism and surroundings.

Finally, Piaget postulates unity of cognitive and living processes. As psychologist he did not doubt the former, nor as biologist the latter. One could describe his theory of development as a monism of the soul, keeping consciousness and body together.

Complementary genetics and enlivening of the concept of heredity

To quote Steiner50: 'the essence of monism is the idea that all occurrences in the world, from the simplest mechanical ones to thehighest human intellectual creations, evolve themselves naturally in the same sense, and that everything which is required for the explanation of appearances, must be sought within that same world.' In relation to adaptive mutation, this view means, as we have seen, that genetic changes must be understood as an expression of an interrelationship between living organism and its habitat. Steiner, Waddington and Piaget have shown approaches to such an understanding. Bockemühl's work on groundsel Senecio vulgaris provides striking examples of such an understanding52.

Spontaneous mutations also have a part to play in a monist evolutionary theory. They show that genetic changes can also occur in the relation of the organism to itself. They are complementary to adaptive mutations. The polarity of world relation and self relation and its overcoming through the organism itself are hallmarks of the living. I have attempted to elaborate this in studies on developmental processes in amphibians53. Similar polarities have come to light in other studies (Pankow et al.54, Schad55, Suchantke56,57). Spontaneous and adaptive mutations are not causes of the variation in form and function, but results of a variety of organic processes. These thoughts will be extended in a further article. Furthermore, I shall report on experiments investigating the existence of adaptive mutations in Drosophila.


I thank all my colleagues in the Goetheanum Research Institute and especially Wilfried Gabriel for our many fruitful conversations and Norbert Pfennig for drawing my attention to the work of Piaget. I thank Hans Christian Zehntner, Jochen Bockemühl and Wilfried Gabriel for their critical comments on the manuscript and Birgit Althaler for the stylistic improvements. I gratefully acknowledge support over many years' work from the Rudolf Steiner-Fonds für wissenschaftliche Forschung.


1    Lamarck, J. B. (1809): Philosophie zoologique. Paris.

2    Lefèvre, W. (1984): Die Entstehung der biologischen Evolutionslehre. Frankfurt, Berlin.

3    Darwin, C. (1859): The Origin of the Species

4    Haeckel, E. (1866) Prinzipien der generelle Morphologie der Organismen., Berlin

5    DeBeer, G. (1958): Embryos and Ancestors. Oxford.

6    Mayr, E. (1984): Die Entwicklung der biologischen Gedankenwelt. Berlin, Heidelber, New York, Tokyo.

7    Luria, S.E. and Delbrück, M. (1943): Mutations of bacteria from virus sensitivity to virus resistance. Genetics, 28, 491.

8    Stahl, F. W. (1990): If it smells like a unicorn. Nature 346, 791.

9    Ryan, F., Nakada, D. and Schneider, M. (1961): Is DNA Replication a Necessary Condition for Spontaneous Mutation? Z. Vererbungsl. 92, 38.

10    Lederberg, J. and Lederberg, E. (1952): Replica plating and indirect selection of bacterial mutants. J. Bacteriol. 63, 399.

11    Watson, J.D. and Crick, F.H.C. (1953): Molecular structure of nucleic acids: A structure for deoxyribose nucleic acid. Nature 171, 737.

12    Alberts, B., Bray, D., Lewis, J., Raff, M., Roberts, K. and Watson, J.D. (1989): Molecular Biology of the Cell. New York.

13    Maynard Smith, J. (1989): Evolutionary Genetics. Oxford, New York, Tokyo.

14    Lenski, R. E., Slatkin, M. and Ayala, F.J (1989): Mutation andselection in bacterial populations: Alternatives to the hypothesis of directed mutation. Proc. Natl. Acad. Sci. USA 86, 2775.

15    Marx, J. L. (1984): Instability in Plants and the Ghost of Lamarck. Science, 224, 1415.

16    Symonds, N.D. (1991): A fitter theory of evolution? New Scientist 21, 30.

17    Mayr, E. (1994): War Darwin ein Lamarckist? Natw. Rdsch. 47. Jhrg. Heft 6, 240.

18    Koestler, A. (1971): The Case of the Midwife Toad. New York.

19    Kammerer, P. (1923): Breeding experiments on the inheritance of acquired characters. Nature 111, 637.

20    Kammerer, P. (1924): Neuvererbung oder Vererbung erworbener Eigenschaften. Stuttgart-Heilbronn.

21    Sankjewitsch, E. (1950): Die Arbeitsmethoden der Mitschurinschen Pflanzenzüchtung. Stuttgart.

22    Waddington, C.H. (1959): The Evolution of an Evolutionist.

23    Piaget, J. (1974): Biologie and Erkenntnis. Über die Beziehungen zwischen oganischen Regulationen und kognitiven Prozessen. Frankfurt a.M.

24    Cullis, C.A. (1988) Control of variation in higher plants. In: Evolutionary Processes and Metaphors; Mae-Wan Ho and Sydney W. Fox, Chichester.

25    Cairns, J., Overbaugh, J. and Miller, S. (1988): The origin of mutants. Nature 335, 142.

26    Foster, P. L. (1992): Directed Mutation: Between Unicorns and Goats. J. Bacteriol. 174, 1711.

27    Foster, P. L. (1993): Adaptive Mutation: The Uses of Adversity. Annu. Rev. Microbiol. 47, 467

28    Symonds, N.D. (1993): Francis Ryan and the Origins of Directed Mutagenesis. Mutation Res. 285, 9.

29    Kolter, R. (1992): Life and Death in Stationary Phase. ASM News 58, 75.

30    Hall, B. G. (1988): Adaptive Evolution That Requires Multiple Spontaneous Mutations. I. Mutations Involving an Insertion Sequence. Genetics, 120, 887.

31    Hall, B. G. (1990): Spontaneous Point Mutations That Occur More Often When Advantageous Than When Neutral. Genetics 126, 5.

32    Hall, B. G. (1991a): Is the Occurrence of Some Spontaneous Mutations Directed by Environmental Challenges? The New Biologist, 3, 729.

33    Hall, B. G. (1991b): Increased Rates of Advantageous Mutations in Response to Environmental Challenges. ASM News 57, 82.

34    Hall, B. G. (1991d): Adaptive evolution that requires multiple spontaneous mutations: Mutations involving base substitutions. Proc. Natl. Acad. Sci. USA 88, 5882.

35    Hall, B. G. (1993): The Role of Single-Mutant Intermediates in the Generation of trpAB Double Revertants during Prolonged Selection. J. Bacteriol. 175, 6411.

36    Hall, B. G. (1992): Selection-induced mutations occur in yeast. Proc. Natl. Acad. Sci. USA 89, 4300.

37    Mittler, J. E. and Lenski, R. E. (1990): New data on excisions of MU from E. coli MCS2 cast doubt on directed mutation hypothesis. Nature 344, 173

38    Mittler, J. E. and Lenski, R. E. (1992): Experimental evidence for an alternative to directed mutation in the bgl operon. Nature 356, 446.

39    MacPhee, D. G. (1993): Directed mutation: paradigm postponed. Mutation Res. 285, 109.

40    Mittler, R.E. and Lenski, R. E. (1993): The Directed Mutation Controversy and Neo-Darwinism. Science 259, 188.

41    Koch, A. L. (1993): Genetic Response of Microbes to Extreme Challenges. J. theor. Biol. 160, 1.

42    Symonds, N.D. (1989): Evoltuion: Anticipatory Mutagenesis? Nature 337, 119.

43    Stahl, F. W. (1988): A unicorn in the garden. Nature 335, 112.

44    Hall, B. G. (1991c): Spectrum of mutations that occur under selectiveand non-selective conditions in E. coli. Genetica 84, 73.

45    Harris, R.S., Longerich, S. and Rosenberg, S.M. (1994): Recombination in Adaptive Mutation. Science 264, 258

46    Thaler, D.S. (1994): The Evolution of Genetic Intelligence. Science 264, 224.

47    Radicella, J.R., Park, P.U. and Fox, M.S. (1995): Adaptive Mutation in Escherichia coli: A Role for Conjugation. Science 268, s. 418

48    Galitski, T. and Roth, J.R. (1995): Evidence that F Plasmid Transfer Replication Underlies Apparent Adaptive Mutation. Science 268, 421

49    Shapiro, J.S. (1995): Adaptive Mutation: Who's Really in the Garden? Science 268, s. 373

50    Steiner, R. (1900): Haeckel und seine Gegner. In: Methodische Grundlagen der Anthroposophie, GA 30, Dornach 1989.

51    Steiner, R. (1891): Über den Gewinn unserer Anschauungen von Goethes naturwissenschaftlichen Arbeiten durch die Publikationen des Goethe-Archivs. In: Methodische Grundlagen der Anthroposophie, GA30, Dornach 1989.

52    Bockemühl, J. (1980) Eine neue Sicht der Vererbungserscheinungen. In: Lebenszusammenhänge erkennen, erleben, gestalten. Published by the author. Dornach, 2nd Ed. 1986.

53    Wirz, J. (1990): Schritte zu einem neuen Ansatz in der Entwicklungsbiologie. Elemente der Naturwissenschaft. 53(2), 3.

54    Pankow, W. et al. (1991) The significance of Mycorrhozas for Protective Ecosystems. Experientia 47, 391.

55    Schad, W. (1983) Zur Biologie der Gestalt der mitteleuropäischen buchenverwandten und ahornartigen Bäumen. In: Goetheanistische Naturwissenschaft, Bd. 2. Stuttgart.

56    Suchantke, A. (1974) Biotoptracht und Mimikry bei afrikanischen Tagfaltern. Elemente der Naturwissenschaft, 21, 1

57    Suchantke, A. (1976) Biotoptracht bei südamerikanischen Schmetterlingen. Elemente der Naturwissenschaft, 21, 1


Translated by David J. Heaf from 'Schritte zur Komplimentarität in der Genetik', Elemente der Naturwissenschaft, 64(1), 37-52 1996 with permission.

English translation published in Archetype (ISSN 1462-8775), September 1998, No. 4, pp 21-36.

Author's email address (for enquiries about this article):

For a more recent article on adaptive mutation please see Craig Holdrege's article on this site entitled 'Genes Are Not Immune to Context: Examples from Bacteria'