Chapter 34 Chapter 18 Various theories about genes-2

The expression of almost any gene (especially one with quantitative effects) can be modified (altered) by other genes.The gene that modifies fur color in the Kessel mouse experiment is a typical example.Modifier genes are particularly important in evolutionary change because they readily respond to natural selection and provide the necessary flexibility for a population to cope with its environment. sudden change.The essence of multifactorial (polygenic) inheritance is that a single component of a phenotype (a single trait) can be controlled by several independent genes (loci).Examples of multifactorial inheritance have been found very early in the history of genetics, starting with Mendel's experiments with bean hybridization.A famous example is the chicken's walnut crown, which Bateson and Punnett demonstrated in 1905 to be the result of an interaction between the bean-shaped and rose-shaped crest genes; they also found it in sweet pea polygenic phenomenon.But there has been considerable resistance among evolutionists to accepting the multifactorial hypothesis of continuous variation.From their point of view, this seems to be a rather subjective special hypothesis to cover up the shortcomings of Mendel's theory.

Although the phenomenon of multifactorial inheritance has been discovered repeatedly since 1905, I think it should be attributed to the Morgan school who used it to deny the early Mendelian one gene one trait (ie unit trait) theory.Negating this theory makes it possible to more clearly distinguish transmission genetics from physiological genetics.It excludes some of the preformist influences of the early Mendelian theory and thus can be explained in terms of molecular genetics ("the genetic program") virtually without modification at all. Multifactorial inheritance, also known as polygenic inheritance, is not the only example of the interaction of different genes.In fact the possible kinds and extent of gene interactions and interactions between different kinds of DNA as is now understood are becoming more and more apparent.This importance was already recognized by some early Mendelians, notably Bateson, who attached great importance to "epistatic" (Bateson's term) interactions between different genes.To give a simple example, a white gene can suppress several color genes to produce pigment.The Soviet geneticist Chetverikov (Chetverikov, 1926) first clearly stated that all genes provide the genetic background or conditions for other genes.This idea is important in both physiological genetics and evolutionary genetics.

A special kind of gene interaction is pleiotropy: the phenomenon in which a single gene can affect several traits (that is, different components of a phenotype).Understanding gene pleiotropy is particularly important for determining the selective value of genes of interest.All progress described above.Including the discovery of polygene and pleiotropy, it makes people's point of view that all hereditary phenomena can be explained by nuclei-independent genes more clear. Since then, genetics has begun to analyze the continuous variation of the biostatistical school and prove that it is consistent with Mendel's theory.This started with the originality analysis of Fisher (1918), followed by Mather (1949) and some animal breeders (Lerner, 1958).Quantitative genetics has come a long way since the 1940s (Falconer, 1960; Thompson and Thoday, 1979; see also Part II).

In the 1920s and 1930s, what Morgan and other geneticists called genes articulated hard inheritance.The only way a gene can change is through mutation, where a previously fixed gene is transformed into another gene in a single step.One might think that the confirmation of the fact of mutation will declare the death of all theories of soft inheritance, but this is not the case.In fact, soft inheritance is not easy to eradicate.There are many reasons for this.One of these is that the early Mendelians (De Vry, Bateson, Johnson) who were the first to espouse hard inheritance had an unacceptable view of evolution.Their opponents mistakenly assumed that accepting hard inheritance required acknowledging the manifestly invalid Mendelian theory of evolution.

In addition, the laws of heredity have been suggested by means of the study of abnormal (if not obviously morbid) traits (albinism, polydactyly, structural defects, etc.).Naturalists believe that soft inheritance is still needed to explain gradual changes in evolutionarily important traits (as opposed to Mendelian catastrophe) and to account for adaptive geographic variation (climate laws, etc.).The stronger the evidence in favor of hard inheritance, the harder the neo-Lamarckians sought evidence for the inheritance of acquired traits. By the 1930s and 1940s the evidence in favor of hard-form inheritance was accumulating and so convincing that even the last geneticists to insist on some form of non-Mendelian inheritance either changed their tune or remained silent.Thirty years later, a firm believer in soft inheritance was occasionally to be found among non-geneticists (Mayr and Provine, 1980), but it had come to an end as a valuable scientific doctrine.

Perhaps the decline of soft genetics can be attributed to three reasons.The first is that all attempts to find experimental evidence for the existence of soft inheritance have failed (see above).The second is that all studies of genes show that genes are completely stable (except for occasional mutations).Finally, it seems that all phenomena that need to be explained by soft inheritance, such as continuous variation and climatic laws, can finally be explained by Mendelian genetic factors (genes) and natural selection.Although it was no longer needed at the time, the death knell for soft genetics was not sounded until molecular geneticists demonstrated in the 1950s that the pathway from nucleic acid to protein was a one-way street.

Genetics has advanced by leaps and bounds in the 50 years since the rediscovery of Mendel's principles.Almost every aspect of pass genetics was resolved during this time.Here I provide a brief summary of the discoveries made around 1950 to help track subsequent genetic advances. (1) The genetic material is particulate (granular), consisting of units called genes, which have long-term stability ("hard inheritance"). (2) A specific trait is the product of a stator (gene) located at a certain locus (site) on a chromosome. (3) Genes are "linked" on chromosomes in a certain linear order, but this linkage can be cut off by exchange; the farther the gene loci on the chromosome are, the higher the exchange frequency (except reversed or restored by double exchange) .

(4) In individuals of sexually reproducing species, each gene is usually expressed twice, with one of the two homologous units being derived from the male parent and the other from the female parent (called the diploid principle). (5) Mutation is a discontinuous change of gene. (6) The genotype (genetic material) and phenotype must be strictly distinguished. (7) Several genes may express a single "trait", that is, a component of a phenotype (polygenicity), and a single gene may affect several traits (pleiotropy). By around 1920, the basics of Mendelian inheritance had been roughly understood, and genetics began to specialize.Population genetics took off in the 1920s and made great strides especially in the 1930s and 1950s (see Chapter 13).Physiologists and embryologists also began to realize that the phenomena they studied must eventually be traced back to genes, so the study of gene function became a more important branch of genetics.However, there are still many questions that are not fully understood in transmission genetics, such as: What is the nature of genes?How many "morphologies" does a gene have?What kind of molecule or group of molecules is it?How big are genes?How do different genes differ chemically?Are all genes basically the same, or are there different kinds of genes?Many similar questions about the true nature of genetic material remain unanswered, and the attention of various schools of thought has been focused on these questions in an attempt to find answers.

It is difficult to describe the history of transmission genetics from 1920 to 1960, because the problems studied during this period (such as piebald) were very technical and some of them have not been understood until now.In fact, these questions can only be answered after the structure and function of eukaryotic chromosomes can be explained.Great efforts were made during this period to elucidate the nature of the gene, but these efforts seemed insignificant or irrelevant when the structure of the DNA molecule was discovered in 1953.This period also saw no significant progress in establishing new concepts.In fact, most of the new concepts proposed at that time, such as the gene particle (genomere) hypothesis and Goldschmidt's field theory of the gene (field theory of the sene), at least in their original form must be abandoned.This period also lacks a critical historical record that can reconcile the discoveries and controversies of the time based on the results of molecular genetic research.For example, Muller and Stadler (L.J. Stadler) often come to different conclusions when explaining their radiation experiment results.Could the apparent contradictions between some of their discoveries be resolved in light of modern knowledge of the fine structure of chromosomes and the constituent enzymes and regulatory genes?There is still a lot of work to be done in historical analysis, and my exposition below is only tentative, and may require considerable revision.

In order to help understand the debates and doubts in the forty years from 1920 to 1960, I will first introduce the classic gene-chromosome theory very briefly.Chromosomes are like a string of beads, each bead representing a different gene. Each gene is thought of as a single particle, never changing (except for extremely rare mutations) through generations, independent of and unaffected by neighboring genes (except for rare position effects).Genes are believed to have three capabilities: (1) each gene controls (or affects) a trait (genes as functional units), (2) each gene mutates independently of other genes (genes are mutational units), (3 ) Each gene can be separated from its nearest neighbor on the chromosome by the process of crossing over (genes are considered as recombination units).Mutations are considered to be small changes in the gene molecule that result in new alleles.Exchanges are considered to be purely mechanical breaking of the beads followed by re-melting (fusion) of corresponding "bead fragments" of homologous chromosomes.

The notion that a gene is independent of its neighbors and that a gene's position on a chromosome is purely by chance seems to be strongly supported by the findings of the Morgan school.Neighboring genes on Drosophila chromosomes often control completely unrelated traits, and genes affecting a single trait (such as eyes) tend to be widely distributed across all chromosomes.The close proximity of genes is generally thought to be nothing more than a historical artifact of a break in the original chromosome.The fact that there are as many linkage groups as there are chromosomes is also consistent with this theory. In addition, if the genes are well-defined particles, their approximate size should be calculated by various technical means, so as to estimate the number of genes that can be accommodated on the nuclear dyed filaments.Mill was (some would say "of course") the first (1922) to make this calculation, and he later (1929) revised it.He calculated the total number of genes in the common fruit fly to be between 1,400 and 1,800 based on a number of metrics, including mutation frequency and some data on crossing over.Later, other scholars used the radiation method to calculate that the X chromosome alone contained 1300-1800 genes, so all chromosomes of Drosophila contained more than 14,000 genes. Cytological studies also seem to support the idea that chromosomes are a string of beads, and it is even possible to count the beads.Indeed, nuclear material often assumes the shape of strings of beads during the leptotene phase of meiosis, and these beads are called chromomeres by cytologists.Some cytologists believe that each chromomere is a different gene. Belling (1931) counted about 2500 pairs of chromosomes in the nucleus of Lilium.Other cytologists have demonstrated that certain chromomeres contain several genes. The once-sensational advances in cytology also seemed to confirm this chromosome theory. In 1933, Heitz and Baner rediscovered the giant banded (striped) chromosomes in the salivary glands of Drosophila. Painter and Koltsov believed that these bands were equivalent to a series of chromosomal particles in the polytene chromosomes (Polytene chromosomes), and the order of the bands was equivalent to Gene sequence.Bridges (Bridges, 1938) counted at least 1024 bands on the X chromosome of the common Drosophila salivary gland and believed that there were the same number of genes.Gene size can be estimated by measuring the volume of the chromosome.However, such estimated values ​​often differ by one or two orders of magnitude. Later research on microbial genes showed that genes have no fixed size; the size of different genes in the same individual may differ by one or two orders of magnitude. The discovery of the salivary gland chromosome is far more important than determining the number and size of genes in other problems of genetics.Microscopic examination of salivary gland chromosomes can often directly determine the genotype without complicated and detailed breeding or breeding tests.It can also show the presence of chromosomal mutations (rearrangements) inferred by genetic analysis.It is now easy to study phenomena such as inversions, deletions, duplications, and translocations in Diptera.At the same time, the complexity of the band patterns provided reliable evidence for the first time for the complexity of eukaryotic chromosomes and the heterogeneity of chromatin substances. At first all the facts of inheritance seemed to fit the bead-on-a-string model of genes and chromosomes, but inconsistencies and contradictions were later discovered. The first serious contradiction arises from the discovery of position effects by Stefante (1925).There is a dominant gene called "rod eye" on the X chromosome of common fruit flies, which can make the eyes of fruit flies become narrow rods instead of round. This gene can be mutated to be narrower (super stick eye) or to return to a round shape.Further analysis reveals two noteworthy aspects of this situation.The first is that the rod eye phenotype is not simply due to genetic mutations but due to changes in the structure of chromosomes.The normal Drosophila salivary gland chromosome has six bands (S) in this locus segment, but these six bands are repeated (SS) in the rod eye Drosophila, that is, twelve bands, and the super rod eye Drosophila has three repeats in the same segment ( SSS) is eighteen strips. Normal round-eyed flies produced by the rod-eyed mutation have only six bands.This structural change can only be explained by unequal exchange, which was demonstrated by Striest's studies of the behavior of the mutated gene on both sides of the rod-eye locus.Detailed analyzes of other genes in Drosophila and other organisms have finally shown that unequal crossing over is by no means uncommon, in other words that the recombined units are not necessarily genes.This is the first gap in the triple-competence theory of genes. Perhaps more noteworthy is the second aspect of the rod eye gene.The effect on the number of ommatidia (compound eyes) in Drosophila eyes is not the same when two rod eye genes are next to each other on the same chromosome than when the two genes are opposite each other on two homologous chromosomes. Te Fant called this the position effect.Therefore, the case of the rod eye gene proves that the function of the gene and its influence on the biological phenotype can be changed only due to the change of the arrangement of the genetic material on the chromosome, without mutation and without changing the content of the genetic material. The traditional concept of genes has become more complex due to the discovery of pseudo-allelic phenomena.What particularly struck the Morgans in their early discoveries was that adjacent genes generally seemed to be functionally unrelated to each other, going their separate ways.Genes affecting eye color, wing vein formation, setae formation, body immunity, and more may all be located next to each other."Genes" that have very similar effects are generally nothing more than alleles of a single gene.If genes were exchange units, recombination between alleles would never occur.In fact, Morgan's students failed in the early days (1913; 1916) when they tried to find the exchange of alleles at the white eye gene locus. It was later learned that it was mainly due to the small number of experimental samples.However, since Stylvant (1925) proposed the theory of unequal exchange of duplications of the rod eye gene and Bridges (1936) supported this theory based on the evidence provided by the salivary gland chromosomes, attempts have been made again to identify alleles that appear to be The time is ripe for restructuring. Oliver (1940) was the first to succeed, finding evidence for unequal exchange of alleles at the rhomboid locus of the common fruit fly.The heterozygote with two different alleles (Izg/Izp) spliced ​​together by the marker gene reverts to the wild type at a frequency of about 0.2%.Recombination of marker genes demonstrates that exchange between "alleles" has occurred. Exchanges between genes in close proximity can only be observed in extremely large numbers of test samples and are called pseudo-alleles because of their normal behavior as if they were alleles (Lewis, 967).Not only are they functionally similar to the true allele, but they can also produce mutant phenotypes after transposition.They are not only present in Drosophila, but have also been found in maize, especially in certain microorganisms with considerable frequency. Molecular genetics has provided many explanations for this problem, but because the gene regulation in eukaryotes is still not well understood, it is still not fully understood. The discovery of the position effect had profound implications.Dubzhansky once made the following conclusions in a review article: "A chromosome is not only a mechanical aggregate of genes, but also a unit of a higher structural level... The nature of the chromosome is determined by its structural unit determined by the nature of its genes; yet the chromosome is a harmonious system which not only reflects the history of the organism but is itself a determinant of that history” (Dobzhaansky, 1936: 382). Some people aren't content with this mild revision of the "beaded concept" of genes.Since the early days of Mendelianism some biologists (such as Riddle and Chiid) have cited what appears to be ample evidence against the granular theory of genes.The position effect happens to work in their favor. Goldschmidt (1938; 1955) then became their most eloquent spokesperson.He proposed a "modern genetic theory" (1955:186) to replace the (genetic) particle theory.According to his new theory, there are no localized genes but only "a certain molecular pattern on a certain segment of the chromosome, any change in this pattern (position effect in the broadest sense) changes the function of the chromosome component and manifests itself as a mutation." Body." Chromosome as a whole is a molecular "field" and what is customarily called a gene is a discrete or even overlapping region of this field; mutation is a recombination of a chromosomal field.This field theory contradicted a great deal of the facts of genetics and was not accepted, but the fact that such a seasoned and well-known geneticist as Goldschmidt took it so seriously shows how shaky the genetics theory is still.Many theoretical articles published from the 1930s to the 1950s also reflect this (Demerec, 1938, 1955; Muller, 1945; Stadler, 1954). In some of the earliest genetic studies, de Vry discovered in 1892 that the offspring of a variety with red striped flowers of the snapdragon (Antirrhinum mains) showed a wide variety of variegation, from small spots to wide and narrow stripes to large red fan.Different flowers or flowers on different branches of the same plant may vary in the form of the flower class. Since this initial discovery, such unstable genes have been found in many plants and animals and various explanations have been proposed, such as dominant transfer, the emergence of "senomeres" or subgenes of highly complex genes, and so on.Because of its extreme or absolute granularity and gene field theory, this point of view can be said to be the exact opposite.According to the gene granule theory, some (all?) genes are thought to be composed of different particles that are not equally distributed during mitosis (Weismmann's ghost!).Collins, E. G, Anderson, Eyster, and Demerec all once supported the gene grain hypothesis, but they gave up this hypothesis in the early 1930s due to increasing negative evidence (Dererec, 1967; Carlson, 1966). Deme6ec ultimately attributed this instability to "chemical instability of the gene," which of course explained nothing but shifted this annoying phenomenon from the domain of biologists to that of chemists. After a period of silence, unstable genes are gaining prominence again, and their behavior is thought to be the result of genetic or chromosomal interactions.I am referring here to the work of McClintock (1951), who showed that the introduction of a structurally unstable chromosome g into certain genotypes of maize can "mutate" many genes on chromosome g And make other chromosomes become unstable recessive.This apparently involves reversible inhibition of the expression of these genes.Although the true significance of this "abnormal" finding (as someone pointed it out at the time) was not generally recognized until its rediscovery in microbial genetics 12 years later, clearly indicating a "mutation" at a locus Can be mimicked by regulatory activity at a different locus.In other words, the phenotypic expression of a gene can be altered by other genes, while the gene itself remains completely unchanged.It is unclear whether such "pseudomutations" due to epistatic gene interactions occur frequently.In the past 50 years, many scholars have spent a lot of time and energy on the study of unstable genes, and believe that the elucidation of this instability will make an important contribution to the understanding of the essence of genes.Unfortunately, it was later discovered that this phenomenon was not caused by some properties of a single gene but was the result of the operation (interaction) of the entire gene system. The period from the 1930s to the 1950s, when geneticists did their best to actively pursue the study of the nature of genes, also had considerable setbacks.Microscopy does not provide a clearer picture of genes than genetic analysis alone.This is true even for the giant salivary gland chromosomes, which show a bewildering assortment of bands (stripes) which, and any function of the genes inferred to be on or adjacent to them, are not closely related.Since genes cannot be observed directly, they can only be understood inferred.In fact, knowledge about genes can only be obtained by studying the changes that occur in genes through mutations. Although the study of changes in the chemistry of gene products caused by mutations (especially in microorganisms) has progressed rapidly (led by the excellent work of Beadle and Tatum), since these studies have been consciously Enzymes, so it is not very helpful to understand the structure of the gene itself. By 1920 it was beginning to appear that it was impossible to learn more about the nature of genes by hybridization experiments alone.Completely new knowledge must be obtained by other means.Before 1944, biochemistry and biophysics were still immature in terms of concepts, and biochemistry could not solve genetic problems with the proficiency of experimental techniques.In this case, some scholars consider that causing mutations through experimental methods may be a way to understand the essence of genes.Muller was the first to realize that the haphazard way in which some scholars study mutations, even experimental ones, never leads to definitive results.Therefore, he believes that certain necessary conditions must be met in order to obtain definite conclusions, especially: (1) the genetic purity of the test material, (2) the number of individual samples in the experimental group and the control group must be large in order to test the experimental results statistically (3) research and development of new methods, especially strains or strains of special composition (with a suitable lethal factor, marker or marker factor, exchange inhibitor) in order to test different gene structure hypotheses.This particular class of fruit fly stocks—covered in genetics textbooks—enables Müller to calculate the actual frequency of new mutations.This is especially important because many mutations are recessive, and it is difficult to determine when recessive mutants first appeared.In addition, many mutations are lethal in the homozygous case, that is, when they occur on two homologous chromosomes. Homozygous lethal ones certainly do not appear in the offspring.Three steps are particularly important in Müller's method: placing a marker gene on the chromosome for unambiguous identification; deploying a set of crossover suppression mechanisms on the chromosome; and pairing the marker chromosome with another chromosome suitable for displaying mutant changes .Once Mueller had finished these preparations, he irradiated his flies with varying doses of X-rays. He used a stock of female fruit flies to mate with a male fruit fly with a lethal mutation on the X chromosome. As a result, all male flies in the F2 generation died.Therefore, if one of the irradiated male flies produces only female offspring in the F2 generation, this indicates that a lethal mutation has been induced on the K chromosome of this male fruit fly. When a normal non-irradiated male mated with this stock female, only about one in a thousand matings resulted in all females in the F2 generation.This means that the chance of a lethal mutation spontaneously occurring at any one locus on the normal X chromosome is one in a thousand (0.1%).This is the natural or spontaneous rate of mutation.When the male flies are irradiated with about 4000 roentgen units of X-rays, there are only about 100 females in the F2 generation in about 100D mating times, so the mutation rate of the X-ray-treated fruit flies is about 100 higher than its natural mutation rate times.Almost at the same time as Muller, plant geneticist Stadler (L.J. Stadler, 1896-1954) also conducted research on artificial mutations in barley and corn (1928). Muller's discoveries, and especially the remarkable method he developed, opened up a whole new field of research.It places mutation studies on a quantitative basis (e.g. the relationship between mutation rate and X-ray dose). "The whole field of mutation research is dominated by Müller's ideas and experiments. He provided the conceptual structure, posed decisive questions, formulated excellent experimental techniques, and at various stages directed the interpretation of the growing body of experimental data to a A rigorous and complete theory. Many of the views and opinions he put forward could not be verified at the time, but they were later proved to be correct” (Auerbach, 1967). It was finally proved that not only radiation but also many chemical drugs have mutagenic effects.Mustard gas was one of the first mutagens.Robson, a British surgeon, keenly found that mustard gas burns are very similar to those caused by x-rays, so he suggested to geneticist Auerbach whether mustard gas can be used to cause mutations. Through experiments, she indeed proved the mutations that Robson predicted in 1941. effect. Rapoport also independently discovered the mutagenic effect of formaldehyde in the Soviet Union.The mutagenicity (induction of mutations) of many compounds has been identified since the 1940s (Auerbach, 1976).Each mutagen can cause a wide range of mutations, not just specific effects on specific genes. But the frequencies of certain mutations caused by chemical mutagens tend to be different than those caused by radiation.Another particularly noteworthy finding is that some (many?) mutagens are also carcinogens.The finding has prompted suggestions for a quick way to screen for possible cancer-causing chemicals: exposing bacteria to the chemical's influence to check for increases in mutation rates. But more importantly, Muller believes that artificially induced mutations will provide an explanation for the essence and structure of genes.If the genes are particles of a certain size, bombardment with ionizing radiation (electrons or short-wave radiation) will hit these particles and the damage will be manifested as mutations.This is the mutation's "hit theory" (hit theory) or target theory, physicist K. G． In a classic article jointly published by Zimmer, geneticist Delbrück and Soviet geneticist Timofeeff-Ressovsky in 1935, it was elaborated in detail. However, the target theory did not yield consistent results (Carlson, 1966: 158-165), and thus did not provide a better explanation for genes.In addition, it was found that even irradiation of the culture substrate can increase the mutation rate, and many chemicals (such as mustard gas, phenol, etc.) can also induce mutations like irradiation.Anything that interferes with the normal process of gene replication can cause mutations.This has led some scholars to adopt "any error in gene duplication" as the definition of mutation (studies in recent years have shown that this definition cannot be applied to all mutations). Radiation technology, however, also encountered more fundamental difficulties.It is not the isolated genes that are irradiated but the chromosomes, the genes and the matrix in which they are embedded.Both the gene and the chromosomal matrix are susceptible to x-ray damage, so studies of phenotypes that have mutated after irradiation make it difficult to distinguish whether gene mutations or matrix (chromosomal) mutations are involved.Cytological studies often reveal slight (often very small) defects in chromosomes that can be clearly identified as chromosomal mutations. Muller (for fruit flies) and Stadler (for maize), two of the most prominent scholars in the study of X-ray mutations, disagree on the frequency of true genetic mutations produced by X-ray treatment.Stadler only admitted that the new mutants could revert to the pre-irradiation traits after irradiation to be regarded as mutations.This is extremely rare, at least for corn.In all other cases Stadler is skeptical of the production of unstable genes and of chromosomal damage.As he stated in his last article (1954), "A mutant may meet the requirements of various tests of genetic mutation, but if it fails to revert to mutation, there is reason to suspect that it may be due to loss of (chromosomal deletion), however, if it is able to reverse the mutation, there is reason to suspect that it may be due to expression effects (unstable genes).” Not everyone (Müller especially) shares the same opinion about radiation effects pessimistic attitude.Even with the best of intentions, however, there are severe limits to the information that can be gleaned from radiation experiments. Two facts were indisputable during this period: the first was that (contrary to initial impressions) genes with the same function were sometimes very close on chromosomes (gene complexes; Lewis; 1967).The second is that genes must have structural complexity ("morphology") in order to have partial independence in function, mutation, and recombination.This complexity must be at the macromolecular level.It is increasingly apparent to geneticists that they are facing a wall that cannot be crossed with their genetics-cytology equipment. Another observation in radiation experiments is puzzling.The earlier the mutation rate is determined after irradiation, the higher the rate.It appears that damaged chromosomes have the ability to "heal" or "repair" (at least in part) or restore the knocked-off segment.Subsequent research revealed that there is indeed a formal repair mechanism for damaged genes and chromosomes (Hanawalt et al, 1978 Generoso et al, 1980).The observed mutations can therefore be seen as lapses in repairing genes. Despite the dedication and extensive work of mutation researchers from the 1920s to the 1940s, they did little to help us understand the nature of genes.As Demerec (1967), one of the most active geneticists in this field, observed in his review, "During the first half-century of genetics, our conception of genetic structure remained more or less stagnant. state.” Substantial progress was only made after new methods and different experimental materials were employed. Eukaryotic chromosomes are so complex that even today little is known about their organization and how genes are integrated within these chromosomes (Cold Spring Harbor Symposia, 1978).It has now become clear that in the first half century it was absolutely impossible to understand genes through the chromosomes of eukaryotes.It was not until the object of analysis was changed from eukaryotes such as mice, fruit flies, corn, etc. to prokaryotes such as bacteria (E. coli) and viruses that the situation changed and made great progress.Because prokaryotes have no chromosomes, their genetic material has a much simpler organizational structure, and DNA can be directly understood without the interference of the chromosome matrix. The most important knowledge gained from the study of eukaryotic chromosomes is negative.Unequal crossover suggests that functional genes are not necessarily recombination units.Mutation analysis, especially in microorganisms, has demonstrated that there may be several distinct mutation loci within a single functional gene.Position effects (cis versus trans differences) suggest that genes are not necessarily functional units.The naive view that genes are simultaneously recombination units, mutation units and functional units must be abandoned.Because of these contradictions, Benzer (1957) made a radical suggestion that the word "gene" should be dropped altogether and replaced by three different words: "muton" (muton), as the unit of mutation, "recombinant" (recon), as a recombination unit (determined by the exchange site), "cistron", as a gene functional unit (cis-trans difference by position effect).Of these three words, cistrons are closest to the traditional concept of a gene, since in general a gene's characteristics are determined by its effects.但是“基因”这个词最终仍然按Benzer为顺反子下的定义被普遍采用，而“突变子”和“重组子”这两个词则一直未流行。 从1900年到1950年代遗传学家们究竟持有哪些基因概念很难确定。即使我们只着眼于承认颗粒性基因、不考虑相信基因场论和能扩散的连续性遗传物质的学者，情况也是这样。由于还没有任何一位历史学家作过分析，我将试图在这里提供一点初步的说明。 因为有一些着名的遗传学家在其一生之中几度改变他们的观点，所以事情就更加复杂化。 我在这里介绍的关于基因的四种观．点决不排除还可能有其它的看法。 可能最古老的观点是将基因本身看作是生物的结构物质。达尔文的微芽学说可能接近这一观点。德弗里（1889）对这学说多少作了一些修正，他认为泛子从细胞的细胞核移向细胞质，而细胞就是生物有机体所含有的组织和器官的结构物质。这一观点有时还默认基因由蛋白质组成。 广泛流行的是第二种观点，即认为基因是酶（或像酶一样起作用），作为体内化学过程的催化剂。这一观点在主要原则方面可以追溯到Haberlandt（1887）和魏斯曼（1892）。因为后来证明酶是蛋白质，所以这将意味着基因也是蛋白质（Fruton，1972）。染色物质是由核蛋白（如果不是完全由核酸组成）构成这一发现对酶学派的这一观点并没有产生什么影响。 当核酸的重要性开始被人们认识时，基因被看成是能量传递的一种手段。在Avery及其同事论证了DNA是转化因素三年以后，穆勒于1947年提出了核酸的化学功能可能是为基因反应提供能量的观点：“核酸以聚合形式可能将能量导向基因结构的特殊复杂模式中或使基因作用于细胞。”就基因的作用而言，穆勒认为“如果基因的主要产物不像……基因本身……那么基因必然作为酶来生产它们”（1973；另见Carlson，1972）。 然而穆勒又认为“断言基因或它的主要产物的确是，或通常是像酶那样起作用还为时过早。”穆勒还提起基因可能“产生和它本身相同的或其一部分的组成相同（或互补）的更多分子，”这些基因产物“实际上将会在它们即将参与的反应中被消耗掉。”穆勒提出的这两种看法都偏重代谢方面。 最后一种观点是把基因看作是特殊信息的传递者。这一观点以模糊助形式早就四处流传。一般都会认为在1953年以前某些学者必定会谈到它。然而我有意识地查阅文献资料却没有发现这一类的假说。除了其它概念因素而外还需要承认遗传型和表现型是完全分开的。自从发现了DNA的结构及其在合成蛋白质中的作用（转录和转译）之后，基因作为信息单位的概念当然已经成为现代的标准概念。 在上述的四种基因概念中，各自都就基因的化学组成及其功能作出了某些设想。然而在大约1950年以前，基因的化学在确定基因实质上是特别重要的这一点并没有被人们充分认识。