This reliance on use and disuse as a source of variation, without any effect on his main argument, is the only part of Darwin's demonstration which has had to be abandoned, and he would have welcomed the reasons for it. If only Darwin had realised it, the solution to all these difficulties was being provided by Gregor Mendel, but his results remained unknown until 1900, eighteen years after Darwin's death.

The Mendelian theory of the gene, as worked out by T. H. Morgan and his colleagues with an unprecedented wealth of experimental evidence from the breeding pen and from cytological studies on the structure of the cell and its chromosomes, has established as firmly as Newton's laws of motion or the atomic theory that hereditary resemblances are determined by discrete particles, the genes, molecules situated in the chromosomes of the cells, which are transmitted to offspring in accordance with the mechanism of germ-cell formation and fertilization, and conform to distributional patterns known as Mendelian inheritance). The researches of C. D. Darlington and others on the structure and behaviour of the chromosomes have reached such a degree of refinement and precision that each step in the mechanism of Mendelian inheritance can actually be seen under the microscope.

The genes preserve their separate identity; they collaborate in the production of the characters of the individual that possesses them, but they never contaminate each other; they remain constant for long periods but, from time to time, they undergo a change known as mutation and involving a change in the characters which they control, after which they remain constant in their new condition until they mutate again. It has been conclusively proved that the theory of the gene applies to all plants and all animals investigated, and that the mutation of genes is the only known way in which heritable variation arises. The modifications resulting from good or bad food supply, or from the climatic conditions in which plants and animals live, are not inherited and therefore without significance in evolution.

The history of the reception of Mendelian genetics after its discovery has been peculiar. The earliest mutations discovered, often called 'sports', were usually deleterious and showed marked and discontinuous steps instead of the gradual and continuous variation which Darwinian selectionists looked for as the raw material of evolution. Selectionists therefore rejected Mendelian genetics as the source of variation. On the other hand the Mendelian geneticists, knowing that their mutations were the only source of heritable variation, thought that as they showed wide discontinuous steps and arose suddenly, ready-made and apparently without long-continued selection, selection was inoperative in evolution and they rejected it.

With the progress of knowledge it gradually became obvious that each of the two schools of research objected to the other for reasons which were baseless. As more and more genes were identified and their effects studied, it became clear that the wide and discontinuous mutations first observed were the more easily detected extremes of a range in which the majority exert only slight effects. For the same reason, these mutations were deleterious because organisms are delicately adjusted systems, more likely to be upset by large and discontinuous changes than by small and gradual steps.

The Mendelian geneticists also had to learn two lessons. On the one hand they discovered that although individual genes are associated with particular characters, their control of those characters is also affected by all the other genes, which constitute an organized gene-complex. As a result of previous mutations, gene-complexes of plants and animals in nature contain many genes and these are sorted out and recombined at fertilization in astronomically numerous possibilities of permutations. These recombinations have been shown to bring about gradual and continuous changes in the characters under the major control of individual genes. Sir Ronald Fisher demonstrated the significance of this by showing that a mutant gene that now exhibits the quality known as dominance has gradually become dominant from a previous intermediate condition. This is what has happened to those mutations that confer a benefit on their possessors, and in their case there has been a selection of gene-complexes in favour of those which accentuate the effects of a favourable mutant, so that these effects are manifested even if the mutant gene is inherited from only one parent, which is the definition of dominance. Conversely, with genes that place a handicap on their possessors, there has been a selection of gene- complexes in favour of those which suppress the effects of such genes so that they are manifested only when the mutant gene is inherited from both parents and the gene has gradually become recessive; or suppress them even further when the effects of such a gene are obliterated and the gene becomes what is known as 'modifier' without major control over characters. It has even been demonstrated by E. B. Ford under rigorous experimental conditions that one and the same mutant gene can be made to become dominant in one strain and recessive in another simply by selecting as parents those individuals whose gene-complexes accentuate or diminish the effects of the gene.

The Mendelian geneticists' second lesson was the realisation that although the effects of the mutations which they first observed appeared to be clear-cut, they were already the results of selection in past gene- complexes. For these mutations have occurred before, and the gene- complexes have become adjusted to them. The fact that a single gene may now act as a switch controlling the production of one or other character-difference does not mean that this character- difference originally arose at one stroke by one mutation of such a switch-gene, because it has probably been built up gradually as a result of past selection in the gene-complex.

It is therefore clear that mutations and recombinations of genes provide the supply of variation on which selection acts to produce evolution exactly in the way Darwin's theory requires. Its requirements are exacting, for as T. H. Huxley pointed out, some organisms have evolved slowly and others have evolved fast, and he saw that natural selection was the only mechanism that could satisfy both those requirements. It is able to do so because Mendelian inheritance is capable of producing both diversity and stability. As E. B. Ford has said, an immense range of types must be available for natural selection to act upon, and this is provided by mutation and recombination of genes; yet when a favourable gene-complex has been achieved it must not be dissipated and broken down, and this is provided against by the facts that the genes do not blend or contaminate one another, and that they mutate only rarely.

Ecological genetics describes the experimental study of evolution and natural selection carried out by means of combined field-work and laboratory breeding. The field work needed in these investigation is of several kinds. It involves detailed observation, usually conducted on successive generations, having strict regard to the ecology of the habitats. It often requires long-continued estimates of the frequency of genes, or the characters controlled by genetic mechanisms, together with population size. The research also involves founding of new colonies in nature, not the more familiar technique of establishing them in the closed laboratory.

A major experimental finding is that it is the independent appearance of new mutations in each population, and the independent course of genetic change within populations, that together make up the driving force within the organism which tends to make each isolated population gradually become different from every other. The force outside the organism that aids the process is simpler. Natural selection acts to adapt the population to its surroundings. But no two patches of woodland, no two freshwater ponds, will be absolutely identical, even if they lie in the same area of country. They may differ in the precise nature of their soil or water, in their range of temperature, or their average temperature, or in the particular species of animal or plant that may become unusually rare or unusually common in that locality. Since each population has to adapt to slightly different conditions, the two populations will gradually come to differ from one another.

The history of a patch of sunflowers living in a ditch in the Sacramento Valley of California provides a good example of the way in which all these forces can gradually make two populations become quite different from one another. The population consisted of natural hybrids between the two annual sunflowers of California, Helianthus annum and H. bolanderi. To begin with, the original population gradually became split into two by a drying-out of part of the ditch, the dry section being colonized by grasses among which the sunflowers could not survive. Over the space of five years, the dry grassy patch widened until it had pushed the two separated sub-populations of sunflowers over 100 metres apart. One of these was now in a deeper part of the ditch, which remained wet until late spring, while the other grew in a shallower, drier position. The two sub-populations became different in a number of characteristics, such as the shape of the flower head as a whole, the number of sterile floret rays surrounding the head, the shape of the base of the leaf, and the length of the hairs on the stem and leaves. Even though bees could easily fly from one population to the other, so that some cross-pollination between them must have taken place, observations over the next seven years showed that the differences between the two populations did not disappear. Their environments differed sufficiently to ensure that natural selection preserved the distinctiveness of the two populations.

Exactly the same process takes place in animal populations though, because they can move, the separate populations may each cover a larger area. For example, two species of Carpenter Bee, Xylocopa diversipes and X. nobilis, live in the large East Indian island of Celebes. Three different colour patterns of each species are known on Celebes, each making up one or more separate populations. Another three colour patterns of X. nobilis are known to exist on neighbouring islets.

Once populations have started to diverge in their genetic adaptations in this way, the foundations for the appearance of a new species have been laid. If two divergent populations should meet again when the process has not gone very far, they may completely hybridize and merge into one another. The further, vital step towards the appearance of a new species is when hybrids between the two independent populations do appear, but only along a narrow zone where the two populations meet. Such a situation suggests that, though continued interbreeding within this zone can produce a population of hybrids, these hybrids cannot compete elsewhere with either of the pure parent populations. This seems to be the situation with the woodpecker-like flickers of North America. The eastern yellow-shafted flicker, Colaptes auratus, does mate with the western red- shafted flicker, C. cafer, along a narrow 2OOO-mile-long stretch where the two meet, but this zone of hybridization does not seem to be spreading.

The principles revealed by this kind of research that includes the analysis of variation and selection in the most diverse forms within a population, show that they apply in higher plants, butterflies, snails, or men. So much so indeed is this true that by applying these fundamental concepts it has been possible to make genetic predictions from one group to another and subsequently to find them verified.

One of the most far-reaching results of recent work on ecological genetics is the discovery that unexpectedly great selective forces are normally operating to maintain or to adjust the adaptations of organisms in natural conditions. Selective advantages of 20 or 30 per cent and more are common and indeed usual: a general conclusion derived from the study of a wide range of forms whatever the control of the variation investigated may be. That consideration forces us completely to readjust our thoughts on evolution and to recognize that a population may adapt itself very rapidly to changing conditions.

The direction of selection may sometimes alter with time. In these circumstances, a particular character may have a total advantage of small magnitude when considered over a number of generations, yet swift adaptations to changing conditions are possible since the opposed selective forces are generally great.

If ever it could have been thought that mutation is important in the control of evolution, it is impossible to think so now; for not only do we observe it to be so rare that it cannot compete with the forces of selection but we know that this must inevitably be so. Mutation is originally responsible for the diversity of the genetic units, but living organisms are the product of evolution controlled not by mutation but by powerful selection.

Among the aims of ecological genetics, the study of evolution in wild populations is outstanding. Yet that project is possible only when the process takes place rapidly. This it has been found to do

(i) when marked numerical fluctuations affect isolated communities;

(ii) when polygenic characters can be studied either (a) in populations inhabiting ecologically distinct and isolated areas or (b)even in the absence of isolation if subject to very powerful selection- pressures;

(iii) when a species successfully invades a new country or territory;

(iv) in all types of genetic polymorphism.

It proved essential to define the latter condition precisely. Once that had been done, its analysis could be carried out in species belonging to very distinct groups; a procedure which has disclosed two of its fundamental properties.

First, since the co-operation of several major-genes is often necessary to promote advantageous adjustments, the evolution of close linkage between them is favoured, so holding together the appropriate characters for which they are responsible. Consequently polymorphism is often controlled by means of a super-gene.

Secondly, when a major-gene spreads in a population, owing to changes in the environment or in the gene-complex, we have seen that it will tend to develop heterozygous advantage. It and its allele will therefore be stabilised at the best attainable frequencies: that is to say, polymorphisms are then generated, so that they must be very widespread and important phenomena.

That is one of the reasons why genetics is playing an increasing part in medicine. For, owing to the powerful selection-pressures now known to operate, polymorphisms must be associated with considerable, though balanced, advantages and disadvantages, the existence of which they indeed advertise. Thus, however trivial the characters may be by which we detect their alternative phases, these must have reached their observed frequencies because the genetic unit controlling them is of importance to the organism. The fact that a quarter of the population in western Europe is unable to taste phenyl-thio-urea for long seemed a negligible matter to the medical geneticist. Its association with thyroid disease has altered that point of view now.

The advances brought about by a combination of field-studies and laboratory genetics can be appreciated by looking to the past. Charles Darwin expressed his belief that by choosing the right material it might be possible actually to detect evolutionary changes taking place in natural conditions at the present time. For this purpose he said that long-  continued investigations and careful records would be needed, extending over a period which he estimated at perhaps fifty years in species reproducing annually. As usual, Darwin was right, but on this occasion he was too pessimistic. The techniques of ecological genetics have made it possible not only to observe evolutionary changes in three or four generations in such forms but to evaluate the force of selection in wild populations and to analyze its immediate effects.