4.1.1 Darwin: natural selection
The Greeks and Romans seriously studied fossil remains and were fascinated with the remains of giant creatures (mastodons, mammoths, rhinoceroses, giant bears). They had museums where the bones were displayed and they examined the fossils with the proven techniques of comparative anatomy, skeletal reconstruction and paleogeography. They concluded that there was a former time when monsters as the griffin (this was proceratops) existed, fought by gods and giants1. In later times, comparative anatomy was abandoned and resumed again only 1500 years later. During this lull, fossils were interpreted as the equivalent of the biological world in mineral terms. It was assumed that animals were represented by their equivalents in mineral form (e.g. oysters, corals) and in vegetal form (e.g. anemones), and conversely. As a result, fossils had accumulated, whose true meaning was elusive: e.g. a stalactite was considered to be a mineral priapus, and a brachiopod fossil, whose physical external appearance is that of a mussel, was assumed to be a mineral vulva. The tendency is strong to anachronically judge this interpretation stupid in retrospect. If we indulge in this arrogance, we run the risk and must be prepared to be deemed ourselves stupid in two hundred years from now.
Linnaeus recognized in 1735 that all the species of living organisms are groupable on the basis of their anatomical characteristics. He grouped them in genera; the genera were assembled into families, themselves into orders, and orders into classes. He failed to see that this pattern resulted from a common descent: in his days, fossils had accumulated and were as seriously examined as in ancient Greek times but the science of paleontology was rudimentary and it was not realized that members of the same class share a common ancestor, as do members of the same order, which share a more recent common ancestor, ending up with members of the same genus, which share an even more recent common ancestor.
By the 1850s, deep theoretical and ideological transitions (to which Linnaeus had himself potently contributed), rooted in profound social and philosophical changes of the society, allowed a reappraisal of the true nature of fossils and it became clear that the biological world was subject to evolution. Various ideas floated in the early- to mid-19th century air. Jean-Baptiste de Lamarck had proposed already in 1809 a theory of evolution in which life forms adapt to a continuously evolving universe: evolution occurs by the inheritance of traits acquired during lifetime. The strength of Darwin’s theory, presented in 1859 2, was its strict scientific basis, resting on the two observations that in every species individuals vary among themselves in heritable characteristics and, second, that in every generation more individuals of each species are born than ever survive to reproduce themselves. This leads to the concept of natural selection, by which the individuals best adapted to their environment multiply more successfully and pass their favorable characteristics on to their offspring. According to Darwin, natural selection is differential reproductive success mediated by the environment. With this scientifically solid theory, Darwin enabled humans to be viewed as part of nature and provided a theoretical platform for rejecting the notion of a special creation. This theory was offensive at the time it was proposed because it ran counter the entrenched belief, grounded in a Christian theology that never repudiated the Old Testament, that species are static, remaining as designed by the Creator.
This assumption that natural selection is the driving force of evolution remained contentious until our own days because evidence from nature was lacking. It is now possible to watch evolution in action. In flies, hereditary changes in the length of wings are observable within the span of 20 years and, as importantly, the changes are predictable3. In fishes trapped in coastal lakes formed by the retreat of glaciers, natural selection yields new species as a consequence of adaptation, in a repeatable fashion4: the same type of speciation occurred in fishes isolated in three different lakes. House finches that are presently rapidly becoming predominant in the urban environment of many cities of the eastern United States, California, Alabama and Montana thank their success to the capacity of the mothers in controlling the sex of their eggs as they lay them. In Alabama, the mother first lays males, and the final egg laid is female. These first laid males grow faster than females, have wider bills and longer tails. In Montana, it is the opposite: the females are laid first and have the growth advantage. Within a few decades, the finches of Montana looked substantially different from the finches of Alabama. By biasing the sex of the eggs and laying them in a particular order, the mother increased chick survival by 10% to 20% over chicks from eggs laid in no particular order. Thus, adaptation along different categories helped make finches successful in both States. Apparently, parental effects play a crucial role at the initial stages of population divergence by enabling establishment of populations in novel environments5. It is now clear that Darwin was right: natural selection is a major agent of evolutionary change.
4.1.2 Neo-Darwinism: point mutations and reproductive isolation
Darwin formulated his theory in the absence of any understanding of the processes of heredity. By the 1930s, a general agreement on the mechanics of evolution emerged under the rubric of the "Modern Synthesis". This "Neo-Darwinism" theory of evolutionary biology 6 integrated Mendelian genetics, systematics, paleontology and ecology into a coherent theory of evolution that combined the theory of natural selection with the understanding of how genes are transmitted from one generation to the next, to which was added the realization that individual selection is more important than long-term advantages beneficial to the group.
The theory of the gene holds that the evolution of multicellular organisms is obtained through errors in replication in the genetic material. Since proteins are made under the command of a messenger RNA which is a faithful copy of a piece of DNA, a changed base in this DNA may result in a change in the composition of a protein. Yet, errors in replication perpetually change the composition of this DNA. If the change occurs in a germinal cell involved in reproduction and if it results in the formation of a protein that is still functional, the change in the DNA is proven harmless and the mutated offspring will not die as a result of this mutation.
Once multicellular organisms appeared, evolution could work quite easily on anatomical and physiological details, which can adapt readily to modifications of the external environment. The anatomical details of a unicellular organism are much more closely related to the primary structure of the proteins making up the cytological features. Finally, at an even more basic level of organization, there where proteins fulfill fundamental metabolic or protective functions, evolution works directly on the various amino acids that compose the protein chain.
The possibility of constructing a molecular clock by counting replacements separating two sequences and assuming that the rate constant for amino acid replacement is invariant over time has been explored. Several proteins, as histone, cytochrome C, hemoglobin and fibrinopeptides have been analyzed in this view. This is done by analyzing the composition of these proteins in animal species of different orders such as lampreys, fishes, amphibians, reptiles and mammals. Paleontological clues give an indication of the time of appearance of these organisms and this is correlated with the number of changes occurring in the analyzed proteins. These studies demonstrate that the replacement of one amino acid by another is peculiar to each protein analyzed (fig. 4.1).
Figure 4.1. The rate of changes occurring in the amino-acid composition of various proteins such as histone IV, cytochrome C, hemoglobin and fibrinopeptides, varies with each protein. The rate of change is proportional to the steepness of the curve. It is represented by the time required for the amino-acid sequence to change by 1% after two evolutionary lines have diverged. This divergence (vertebrates versus insects, agnates versus fish, etc.) is noted in the upper part of the figure. For histone, this period is 600 million years and for fibrinopeptides, it is 1.1 million years.
In eukaryotic cells, the genetic material is intimately associated with basic proteins. Histone IV is one of these proteins. It is composed of 102 amino acids. An analysis of the composition of histone IV in various animal species reveals that changes in the amino-acid composition occur only very rarely. It requires about 600 million years for histone IV to exchange 1 % of its amino acids with other amino acids. The conservative pressures on histone IV are thus intense: few changes in its composition are allowed under the penalty that the protein be eliminated. This rigor is presumably due to the close relationship of this protein with DNA, that brings it to the heart of the genetic mechanism.
This conservatism is no longer so apparent when one analyses a protein like cytochrome C. Cytochrome C is enclosed in the mitochondrion. It is a protein associated with a heme (see fig. 5.15) and is one of the enzymes that favor oxidation. Cytochrome C evolved early, the respiratory chain having "settled" down for good more than 1.2 billion years ago. A large portion of the surface of cytochrome C is subject to strong conservative selection pressures in order that a proper contact be maintained between itself and the oxidases and reductases with which it reacts, but the rest of the molecule may vary because it will not interfere with its function. In this case, the time required to induce a change of 1% in the amino-acid composition of this protein is reduced to around 20 million years, which is substantially less than for histone IV.
A younger acquisition of Life is hemoglobin, which utilizes 4 heme groups instead of 1 for cytochrome C and thus also 4 iron atoms. Hemoglobin is a very large molecule that interacts in the main only with very small substances such as oxygen and carbon dioxide (see fig. 4.1. and fig. 5.15). As long as these restricted reactive portions of the molecule are apt to function, the rest of the molecule of hemoglobin is allowed to vary in a fairly broad way. The time necessary to induce a change of 1% in the amino-acid composition of such a protein is reduced to 5.8 million years.
A presumably even younger acquisition than hemoglobin by living organisms is fibrinopeptide. Fibrinopeptides are small molecules intercalated between fibrinogen. The presence of fibrinopeptides prevents the fibrinogen from adopting the fibrin configuration. Such an adoption would initiate a coagulation process right within our blood vessels and would be catastrophic. As long as the fibrinopeptide spacers are apt to be excised from the fibrinogen molecule whenever the need arises to start the blood clotting mechanism, these fibrinopeptides will be adequate. One would expect a fibrinogen molecule to tolerate many changes in the spacers, provided these spacers continue to fulfill their simple function, i.e. the changes should not interfere with the disposal of the spacers whenever it is required. Indeed, the rate of appearance of harmless mutations in fibrinopeptides is brought down to 1.1 million years.
The possibility of constructing a molecular clock by assuming that the rate constant for amino acid replacement is invariant over time is illustrated here. Unfortunately, protein behaviors are closely tied to the demands and constraints of natural selection. Amino acid replacement is faster or slower depending on how these change, making protein sequences irregular clocks at best. Likewise, it is difficult to correlate the molecular record for a specific protein family with the paleontological record. Selective pressures make protein clocks "tick" irregularly. Nowadays, most workers examine silent sites in a gene.
According to the neo-Darwinian model of speciation, families can be isolated from a common stock. This occurs usually through a geographical accident such as a continental drift, a mountain chain build-up, a rise in sea level, the appearance of a desert, the retreat of a glacier, etc. Isolated groups will have their genetic material subject to random mutations. If these mutations are not disfavorable, they will be maintained and introduced into the isolated population’s genetic pool. Whenever thereafter the environmental living conditions change, the hitherto indifferent genes may turn out to be favorable. These genes will give a survival edge to their possessors so that these will slowly extend to the totality of the population.
In the course of time, a hereditary patrimony may evolve that will in the end be so much different from that of the other groups, that fertile mating will no longer occur between them under natural conditions. If such matings are per chance fertile (matings of tiger with lion, zebra with donkey, donkey with horse, rat with rabbit, etc.), then the offsprings are themselves sterile, unless inbreeding occurs, i.e. fecundation by a littermate or by the father.
The explanation for the transformation of groups of living beings into races and eventually new species lies thus on mutations followed by selection and sexual exclusion. But is this explanation sufficient by itself? Is the notion of species as restricted as here assumed and have observations been made that do not corroborate with the Neo-Darwinian theory?
4.1.3 The concept of species
A species is defined as a group of living beings whose hereditary patrimony is naturally isolated from that of other groups. A species is a congregation of animals or plants that will intermate and whose offspring is fertile.
The recognition that species are sexually reproductive communities is, however, not enough to explain the concept of species. Sometimes, especially among primitive species, the same organism will bear the ovum and the sperm. In this case, either there is auto-fertilization or else copulation can still occur with another hermaphrodite and give place to an exchange of genes. In other cases, parthenogenesis takes place and the female bears offspring without any contribution from a fertilizing entity, even if copulation occurs. Such cases are found among insects and also among vertebrates. One mode of reproduction used by unisexual vertebrates is gynogenesis, which is the development of an ovum after stimulation by a sperm but without fusion of the male and female pronuclei. This results in offspring genetically identical to the mother. Natural populations of gynogenetic vertebrates have been identified in teleost fishes (i.e. the cyprinodontiforms, such as goldfishes) and salamanders. Sometimes, among insects and marsupials, the father’s chromosomes are lost during the development of the egg.
Among social insects, there are cases where all the males and females of a group but one fecundated female are eliminated. The fecundated female establishes a colony consisting mostly of asexuated workers and soldiers. In this case, all the workers of a colony are sisters, which are closer kin than mother and daughters because males donate all their genes to each daughter. The result of this is that sisters share on the average three quarters of their genes, while the mother shares only half of her genes with her own daughters. Among vertebrates also, cases occur where most males of a group are eliminated, so that only one male will fecundate the near totality of the available females.
The fact that nearly all animals are diploid, carrying two copies of each chromosome, has been taken as evidence for an inherent superiority of the diploid state, as I exposed in the previous chapter. However, counterexamples exist, indicating that the premise upon which the rule is based, is false. There is no process that absolutely prevents animals from developing as haploids. During animal evolution, male haploidy, i.e. males whose somatic cells have half the normal chromosome number, with female diploidy, has arisen at least 17 times among the Rotifers, the Insects, the Acarians and the Nematodes. The reverse, i.e. haploid females and diploid males, has not yet been observed. What has been observed, however, is a unique case of haploid females among the mites (acarians). There exists one mite species (Brevipalpus phoenicis) where haploid females produce only female offspring from unfertilized eggs7. This haploid oddity is traced to intracellular bacteria that infect the eggs and forces the development of only females. With antibiotics, the cured offspring developed into males, indicating that haploid males must have been present in earlier times, among the ancestors of these mite species. Note that there exists a species of toad (Bufo pseudoraddei) in a secluded Pakistani valley that reproduces sexually with three pairs of chromosomes.
Bacterial species are even more puzzling, to the point that one can no longer say what is a species anymore, as shown by the two following cases. Methanosarcina mazei is a methane-generating archaea. Contrary to other methanogens, its single circular chromosome is not 3 million bases rich but is 4.1 million bases long, i.e. similar to E. coli. Of the 3300 predicted genes, a good third, 1100 of them, are eubacterial genes, acquired by horizontal transfer. Another microbe, Rhodopseudomonas palustris, carries circadian rhythm genes, where such genes do absolutely not belong. Although it is a purple nonsulfur bacterium, its genome closely resembles the genomes of the bacteria that fix nitrogen. In addition, this nonsulfur bacterium is able to break down organic matter, which the other non-sulfur bacteria are unable to do, as well as fix nitrogen and produce hydrogen gas. Either this bacterium borrowed a lot of genes from nitrogen-fixing bacteria or else they are genuinely closely related, indicating that we understand little about bacterial species’ definitions.
Coral experts have suspected that many coral species are promiscuous. In a maritime orgy, dozens of coral species release their gametes on the same few nights, once a year. Occasionally, sperm of one species pair with eggs of another, and a hybrid occurs. Interbreeding occurs between the hybrids and the parents, supporting the idea that corals are too intermingled to qualify as separate species.
The concept of species based on sexually reproductive isolation is thus not fully applicable and even totally inapplicable in some cases. One should recognize that all species are not of the same kind and different sorts of species should be understood in relation with different strategies of adaptation to living conditions.
4.1.4 Genetic drift: random fluctuations
The failure of concordance between mutation and speciation is evidenced by the following fact. A constant increase of the size of the brain is observed from Australopithecus to Homo habilis, Homo erectus and Homo sapiens. Yet, within each of these human species, no trend to such an augmentation is observed. The paleontologist is thus inclined to postulate that, whenever a species is replaced by another, several candidates covet the freed place, of which each possesses particular adaptations to distinct environments. These adaptations are not oriented towards any defined evolutive trend but are, on the contrary, distributed at random. If however, as is the case in the hominisation line, a given environment proves to be more favorable to the expansion of the species endowed with a particular adaptation, this species only will subsist. If such a process repeats itself at each new step of speciation and if it is always the same type of environment that proves itself to be the more favorable for the expansion of a species and thus thereby reveals itself to be favoring a particular adaptation, then, the paleontologist will observe, in the succeeding species of the evolutive line, an evolutive trend, as is the case of encephalisation in the hominid line, which, in reality, is due to chance mistaken as a trend.
A type of variation 8 different from point mutations and selection explains this particular chance. This variation, called genetic drift, is a random fluctuation in the frequency of a gene, as it appears in a population made of an exceedingly small number of individuals, from one generation to the next. It is so completely dissociated from natural selection that it seems to promote the predominance of genes that oppose adaptation to the environment rather than favor it.
From a small animal group of 20 individuals, let us postulate that 10 have the genetic trait "red fur" or, in humans, "red hair". The trait is thus fixed in a normal way, at 50%. In such a group, the fortuitous appearance of twin males endowed with this character should be enough to eventually fix the character very rapidly at 100% of the totality of the group. In this case of variation, no selective pressures are applied from the part of the environment on any characteristic. We can safely assume that the drift will apply in the same way to all the genes present in the pool. A gene present in small populations can become fixed at 100% in less than 30 generations. If the generation time is around one year, the fixation of the trait will be realized in 30 years or less. Through genetic drift, speciation may thus occur within a few generations.
4.1.5 Saltations: genetic upheaval
Paleontologists have observed that each new species appears once in a fossil series, which thereafter remains without change during long periods of 5 to 10 million years. Then, suddenly, a given species is replaced without transition by another species. This permanence and stability of species, once they are established, run counter to the neo-Darwinian models, which postulate that minimal changes introduced through mutations constantly and gradually modify the populations representative of a species. In fact, a species is defined by the capacity of its members to exchange their genetic material; the birth of a new species from an old species consists in the formation of two reproductive communities where there existed initially only one. It is thus clear that if natural selection changes the adaptive characters of species, it is not this adaptation that creates new reproductive communities: the constitution of the reproductive isolation of a population vis-à-vis an initial reproduction community does not represent an adaptation but is an event due to an upheaval of the physical or psychological environment of a species. This upheaval fragments the original continuum of the species. Such an upheaval is implicit in the observations of paleontologists who define new species in different geological strata. This physical fragmentation can be constituted among higher organisms by psychological pressures that induce an animal group to consider at one time a given territory as its exclusive property and forbids its access to others. Humans do so by creating national boundaries. It is thus clear that the processes of speciation and of adaptation are not perfectly in tune.
If an animal group is very small, several incidences bearing on the fate of genetic traits are to be considered. For example, in many species, fathers are prone to fecundate their own daughters, while others kill their sons. This may occur also in the human race (e.g. Abraham, who was ordered to slay his son, yet did not do so). Sons fecundate their sisters but are forbidden to fecundate their mothers. This is considered a mortal sin, as experienced by Oedipus, who married his mother. A break-down of the psychological block that somehow forbids the mating of human beings who have been in close social contact during infancy can occur also in human populations and leads to consanguineous procreation, as was practiced with extreme frequency in ancient Egypt, Babylon and by the Incas. Such a case occurs still now among the Amish living in the US and Paraguay. The Parsees of India, of whom some customs are those of the people of Cytal Hüyük living about 10,000 years ago in Turkey, engaged until about 60 years ago in closely consanguineous marriages with extreme frequency. Small animal groups have thus many chances to be consanguineous to a high degree.
In the most extreme case, a single endogamous bred female fertilized by her father or twin brother, carrying only a small proportion of the total genetic variation of the parental population, will become isolated and establish a new population through filial inbreeding during one or several generations. This process brings about a large genetic upheaval. The population descended from the founder, rapidly expanding in a virgin environment, is thereafter restructured by natural selection. This selection operates by now on a profoundly changed genetic pool and usually in an altered environment.
This process may explain the appearance of human races. According to the Genesis, the Moabites and the Ammonites stem from the fecundation of the two daughters of Lot by their father; the origin of the Arabic race is to be found in the fecundation of Agar, the Egyptian servant of Abraham, by her son, Ishmael, begotten from Abraham, after their isolation in the desert where Abraham had banished them. The Bible also extols incestuous love between brother and sister (read without exegesis the Song of Songs). Greek mythology and tragedies are saturated with stories of closely consanguineous procreations (e.g. Zeus, who married his sister, was also the son of his grandfather; Oedipus married his mother).
There may thus be more than one process of speciation. The first involves a widespread population that changes through adaptation leading ultimately to speciation. This synthetic theory (neo-Darwinism) holds that small fluctuations within a large population lead to a gradual selective change and, eventually, to speciation. The second mode of speciation entails a reproductive isolation and speciation that precede differential adaptiveness. In this case, speciation is devoid of a biological function until an adaptation arises, that follows this speciation. This neutralist theory rejects the idea of a constant selective pressure and advances that inconsequent mutations within a small group extend rapidly to the totality of the members of the group. These mutations are initially neither good nor detrimental for survival. Finally, the saltationist theory, observing the general stability of species extending over periods of about 10 million years, claims that evolution works on a single or a few reproductive individuals totally isolated from the common stock, haphazardly thrown into inclement survival conditions that enforce a genetic adaptation.