Evolution: An Introduction by Stephen C. Stearns and Rolf F. Hoekstra

Survival is only important if it contributes to reproduction. (p. 2)

Natural selection and history are the two great themes of evolutionary biology. (p. 6)

Compare with RNA, DNA, with its double helix, is similar to sexual reproduction. Is that right?

It is a common misconception that evolution is slow and gradual. In fact, the rate of evolution can be either slow or fast. … Microevolution can be very fast, much faster than one infer from fossils, and it is fastest in large populations for traits with a great deal of genetic variability under strong selection. (p. 52)

The type of selection experienced by a gene uin a genome, or a trait in an organism, or an organism in a group, results from an interaction between the environment, the focal gene or trait or organism, and the other genes, traits, and organisms contributing to survival and reproduction. Selection thus has both external and internal causes. It always involves the thing selected and its environment. (p. 52)

On neutral evolution:

There seems a lot of redundancy in DNA and in the coding of DNA to amino acids. This redundancy is largely explained away as historical accidents. However, the real level of redundancy could be much smaller once we take the need for accuracy in information transmission. Language can and do change very fast. However, languages contain large amount of redundancy. The high level of redundancy is necessary for accurate information transmission.

Selective neutrality of mutations in introns and pseudogenes is supported by molecular evidence that introns evolve faster than the translated arts (exons) and that pseudogenes evolve faster than functional genes. (p. 60)

If such amino acid substitutions have an effect on reproductive successes, it is probably very small: they are approximately neutral. (p. 60)

Even an effect is small, evolution operate over a long period of time. The total effect may not be neutral. It may be because the subtly of the effect has not been detected by the researchers yet.

Some characters show no, or very little, phenotypic variation, despite considerable environmental and genetic variation. These are called canalized characters because the final phenotypic outcome is kept constant, as though developments were confined with a canal that didn’t allow deviation from its course. … Because of developmental canalization, … such mutations are neutral as long as they are not expressed. (p. 61)

Not expressed does not mean that it is neutral. For example, most people act obediently to bosses. But some more willingly, others less willingly, which causes internal stress and harm their health. Therefore, same phenotypes may not have same fitness level.

Mutation do not occur at random with respect to their location in the genome. Some genes mutate more frequently than others. (p. 64)

Recent molecular data on human genetic diseases suggestshigher point-mutation rates in males than females, in some genes. (p. 95)

‘Small organisms are usually small not because smallness improves fecundity or lowers mortality. They are small because it takes time to grow large, and with heavy mortality the investment in growth would never be paid back as increased fecundity’ (Kozlowski 1992). (p. 156)

Reproductive effort models were also confirmed in the field manipulation experiments doe by Reznick and his colleagues on guppies living in shallow streams in Trinidad (Reznick et al, 1990; this example was introduced in the Prologue). At some sites their main predator is a cichlid fish that can eat large sexually mature guppies and cause high mortality rates in all size classes. At other sites, their main predator is a killifish that eats mostly small, juvenile guppies and cause lower mortality rates than the cichlid in all size classes. In 1976, before the experiment began, guppies from cichlid sites mature earlier, made a lrge reproductive effort, and had more, smaller offspring than did guppies from killifish sites. The differences were heritable, and they fit the predictions of reproductive effort models. To demonstrate that predation caused the pattern, predation pressure was manipulated to decrease the mortality rates on all age classes.

After 11 years, or 30-60 generations, significant evolution was observed, as predicted (Table 8.2). (p. 162)

Table 8.2 Divergence of guppy life histories after manipulations

Life history trait	Control (cichlid)	Introduction (killifish)
Male age at maturity (days)	48.5	58.2
Male weight at maturity (mg wet)	67.5	76.1
Female age at first birth (days)	85.7	92.3
Female weight at first birth (mg wet)	161.5	185.6
Size of first litter	4.5	3.3
Offspring weight (mg dry) litter 1	0.87	0.95
Offspring weight (mg dry) litter 2	0.90	1.02

It is not immediately clear why organisms grow old and die, and why different species have different maximum life spans. … However the most striking puzzle is not why clams live longer than lobsters, or why humans live longer than chimpanzees. The most striking puzzle is why your gem line is so well maintained that it is potentially immortal, connecting us through an unbroken sequence over 3.7 billion years long to the origin of life, while we are so poorly maintained that we age and die, even we are protected from accidents and given optimal conditions. Our gem line is part of our own body, determined by the same genes, built with the same biochemistry. Why can it survive, apparently forever, while we must die? The answer is one of the triumphs of evolutionary thought. (p. 164)

Then the authors continue with the details. Our theory can also give a clear understanding of the life span and aging problem.

In humans in industrial countries, the selection pressure to improve survival hasdropped almost to zero by the time one is 50 years old, a fact that does not bring cheer to the middle aged (Figure 8.5). (p. 165)

Sex allocation can also vary as a function of social rank (Trivers and Willard 1973). In polygynous species, one male control access to and mate with several females. Low ranking females in poor condition should have female biased litters or clutches. High-ranking females or females in good physiological condition should have male biased litters or clutches. … at least one example of adaptive variation in the sex ratio of offspring is found in red deer, where high-ranked females produce more sons and low-ranked females produce more daughters (Fig. 8.9). Part of the reason is that sons of subordinate females suffer higher juvenile mortality than sons of high-ranked females, and part is that most stags with high reproductive success are the sons of dominant mothers (Clutton-Brock and Iason 1986). (p. 173)

Mate competition will be stronger in the sex with the greater reproductive potential, which compete for the sex with the lesser reproductive potential. Usually males have greater reproductive potentials than females; thus males usually compete and females usually choose. (p. 179)

Reliable, honest signals must be costly. If they were not costly, sick cheaters could imitate the signal, deceive the partner, and produce mistakes that would eliminate the reason for choosing partners on this basis. (p. 180)

Female northern elephant seals protest against copulation by emitting loud calls and moving their hind quarters, attracting dominant males who chase off smaller males. Females protest less when approached by dominant males. (p. 184)

Organisms should be careful in choosing a mate, but not too careful. Excessive carelessness will result in hybridization with other species and offspring of low fitness; excessive discrimination will take so much time that opportunity to mate will disappear before the choice is made. (p. 184)

Where food was limited, males engaged less in courtship, females fought over males, males were more discriminating in their choice of mates, preferring large, fecund females, and males invested more per reproductive attempt than did females. In contrast, when extra food was supplied, many males courted, females were more discriminating in their mate choice, and females invest more than males. … Thus the sex that experience greater competition for mates is the one that invests less in each reproductive attempt. (p. 190)

It seems most of the history of the human evolution is accompanied by scarcity of food. That is why females use cosmetics while males don’t.

In wealthy countries, when resources are less limiting, male compete for females and hence females have high social status. In poor countries where resources are highly limiting, females compete for males and females have low social status. Therefore, the status of woman is really a reflection of wealth level.

Consider, for example the evolution of the virulence of a pathogen infecting a host population. The myxoma virus infects rabbits. Its replication rate within a rabbit determines the virulence of the virus --- fast replication leads to more severe disease and more rapid death of thee rabbit than slow replication. When there is genetic variation among viruses within a rabbit, selection with rabbits increases virulence. However, selection on the virus for transmission between rabbits favors viruses that have the greatest probability of being transmitted to new hosts, i.e., those living in rabbits that survive for along time. Evolution at this level reduces the virulence of the virus. Thus there is a conflict between two levels: selection among viruses within hosts increases virulence, and selection between hosts reduces virulence. The outcome is an intermediate level of virulence. (p. 201)

The same is true for human beings. Ecological system provides resources to human beings. Those who can extract resources fastest dominate the human society and are selected for. However, over the long term, the sustainable society need long extraction rates. The outcome will be a compromise between the two.

Several plasmids have evolved a remarkably clever mechanism to enhance their stable maintenance in the bacterial host population. They make their host cells addicted to their presence by producing both a toxin and its antidote. The antidote molecules are less stable than toxin molecule. As long as plasmid is present in a cell, there is no problem, for continuous production of both molecules guarantees that toxin is ineffective. But if by chance a copy of the plasmid fails to get transmitted to a daughter cell, this cell dies because the antidote has disappeared while toxin is still present. Thus once a cell has acquired a plasmid, its descendants remains addicted to it. Although this addiction mechanism helps the plasmid to maintain itself, it would not guarantee its maintenance in strictly asexual bacteria with only vertical transmission, where the host cell could lose in competition with uninfected cells. Sex with horizontal transmission overcomes, for the plasmid, the consequence of negative fitness effects on the bacteria cell. (p. 205)

A gene enhancing the rate at which a mammalian embryo extracts resources from the mother may be selected in males but not in females. In the mouse, insulin-like growth factor 2 (Igf2) is expressed in the fetus and promotes the acquisition of resources from the mother across the placenta. The paternal copy of the Igf2 is expressed, but the maternal copy is inactive. There is another gene, the insulin-like growth factor 2 receptor (Igf2r) which appears to inhibit the action of Igf2. This gene shows the reverse pattern: the maternal copy is expressed and the paternal copy is inactivated. … If females have offspring from several males, then it is not in the interest of a male to let the female hold resources in reserve for future offspring fathered by other males. Thus paternal genes are expected to manipulate the mother to supply more nutrients to fetus, while maternal genes are expected to protect her from overprovision that would compromise future reproduction. … Selection for these two genes acts in opposite directions in males and females, thereby creating a remarkable genomic conflict between paternal and maternal genes whose effects occur in the offspring.

The long-term outcome of this conflict may well be the situation. Mice with an inactivated paternal copy of Igf2 are only 60% of normal size at birth, while an inactivated maternal copy of Igf2r are born 20% larger than normal. (p. 212)