The Evolution of Aging

T. C. Goldsmith May, 2002    Revised October, 2002

A longer, more recent version of this article is available at .


What causes aging?  Until the causes are understood there is little hope of devising effective anti-aging treatments.  In fact our whole attitude towards aging is affected by our perceptions of the cause.  If aging is caused by a “fundamental” limitation of life, then essentially by definition, there is no hope of ever devising a treatment for the “root” cause and we must be satisfied with treating “symptoms”.  Research on the causes of aging are only of academic value. 

If aging is not a fundamental limitation but more like a universal disease, then a different approach may yield more comprehensive solutions and treatments.   Research on aging may indeed lead to dramatic improvements in health.

There are many, incompatible, theories of aging which have evolved over the last 150 years and seem to divide into two schools of thought. 

One school holds that aging is caused by some fundamental process, "an inescapable biological reality",  that causes increasing deterioration over time.  For example, nuclear background radiation could cause aging by causing increasing cumulative tissue or genetic damage.   Other sources could cause an "accumulation of random damage to the building blocks of life".   Aging has no benefit to an organism and could not be an evolved characteristic.  Aging is not genetically programmed.  Aging exists despite the best efforts of nature.

The other school holds that aging is an evolved beneficial characteristic of an organism which is a result of evolution just like femur length or number of teeth.  Aging is genetically programmed.  Organisms are designed by nature to age.

Both schools have problems.

Aging as non-evolved: Some people think that living organisms eventually "wear out" in the same way that any mechanical device wears out.  But living things have the capability of constructing themselves from raw materials and also have at least some capacity to repair themselves, unlike mechanical devices.

Perhaps there is a fundamental limitation, a biological equivalent to the mechanical wear out process, that affects living things.  There could be accumulated errors in the process of copying DNA, causing accumulated random genetic damage, or accumulation of some toxic condition (free radicals), etc.

If aging is caused by some fundamental process, how do we explain why average life span varies so dramatically from species to species?  Some insects live for a matter of days, some live 17 years.  Why do men live so much longer than mice?  Mammal life spans vary from less than a year to at least 100 years.

Some theorized that it had to do with body mass.  Life spans in animals tend to be loosely proportional to size.  Bigger fish live longer than small fish.  Larger mammals tend to live longer than small mammals.  But there are obvious gross exceptions: horses only live for about 30 years, parrots live to be 75, etc. 

Could aging be a matter of metabolism, where animals with slower metabolisms have longer life spans?  Again there were difficulties, does a parrot have a slower metabolism than a horse?

Various "evolutionary" theories attempt to explain why aging so resembles an evolved characteristic (large variations between species, etc.)  without actually being an evolved characteristic.

If evolution "doesn't care" what happens after an animal stops reproducing, then aging could be the result of the accumulation of random bad genes that only affect animals later in life (mutation accumulation theory).  

Aging could also be the result of side effects of some characteristics that help an organism in its youth but accidentally, coincidently, hurt it in later life (antagonistic pleiotropy theory).  The beneficial characteristics would evolve and the side effects, though adverse, would not affect evolution greatly because they only affect individuals after they had some opportunity to reproduce.   

Despite the "inescapable biological reality" some reptiles and fish exhibit little or no aging.

Aging as evolved: Aging does seem to share properties with evolved characteristics.  Aging, like femur length, seems to vary somewhat between individuals of the same species, and, like femur length, varies to a greater extent between species.  Aging characteristics of animals are at least somewhat inheritable. 

Darwin and his contemporaries thought longevity was evolved.  Weismann and others proposed in the late 1800s that aging  was an evolved,  genetically programmed characteristic beneficial to species.

However, if aging is an evolved characteristic, then Darwin’s theory of evolution says that aging would have to have some benefit to the organism.  But what possible benefit could aging be to a species? 

Even if one assumes that a limited life span would somehow be of benefit to a species, the mechanics of evolution, “survival of the fittest individual” would appear to prevent the evolution of a characteristic which resulted in aging.  An animal which had such a characteristic would tend to “survive” for a shorter period and breed less than an animal which did not have the characteristic.  Therefore natural selection would select the animals which did not have the aging characteristic (or had it to a lesser extent) and it would eventually disappear.   This problem was actually cited by a Darwin contemporary as an argument against Darwin's theory.

This paper presents arguments to the effect that the aging mechanism is indeed a beneficial evolved characteristic.  

Darwin’s Theory of Evolution

As we shall see, Darwin's theory is central to the "evolution" of aging theories over the last 150 years.

Darwin’s theory of evolution was published in his book The Origin of Species in 1859 and was immediately controversial and a popular best seller.  His later book, The Descent of Man (1871) applied the theory specifically to Man’s evolution from earlier primates.  

Many of Darwin’s conclusions were based on comparison of the generic characteristics of wild plant and animal species with those of domesticated plants and animals.  The wild species were the result of natural selection or “survival of the fittest” while the domestic species were partly the result of selective breeding by humans.

Even the slowest reproducing species, would, if allowed to breed in an uncontrolled manner, occupy the entire planet in a relatively short period of time.  The growth of the populations of all species is therefore checked by a variety of factors such as predators, diseases, food supply, and environmental conditions so that in a stable population each individual has an average of only one progeny which survives to produce one progeny and so on.  All wild species are therefore in competition for survival.

Darwin deduced that normal variation of characteristics between individuals was more important to evolution than larger mutations only affecting a single individual because natural selection working on variations would be working in the entire population simultaneously where an individual with a (very infrequent) beneficial mutation might not survive to reproduce, the mutation might not appear in its surviving progeny, and in any event the change would only spread slowly through the population.  For example, if conditions changed such that increased height was a survival advantage for a particular animal species, then all the taller individuals would have a survival advantage, live longer, breed more, and the average height of the population would increase relatively rapidly until it was again optimum for that particular situation.  Variation in characteristics between individuals in a population is thus a required property of life to support Darwin’s evolution.  (Some contemporaries thought new species were created instantaneously by massive mutations in a single individual.)

In addition to physical characteristics, Darwin included instincts and inherited behavior patterns in characteristics which evolved through natural selection.

Few scientists of the time would have argued against the idea that natural selection could cause a species to evolve.  After all, humans had for thousands of years been causing domesticated species to change by selective breeding.  If small dogs were bred with small dogs for a long time men could evolve a Chihuahua.   If fast dogs were bred with fast dogs for long enough a Greyhound would result.  If you go back far enough, Chihuahua and Greyhound are descended from the same dog.  The argument was whether all the species that now exist could have evolved from a single original primordial species (probably on the order of pond scum) simply by the effects of natural selection acting in slow small increments on individual variation.  It was a much greater leap to believe that humans evolved from pond scum than to believe that fast dogs evolved from slow dogs.

Development of different varieties was driven by geographic separation and differences in conditions.  If a mammal lived in an area which contained both mountains and lowlands, it could not be optimized for both areas.  The animals in the mountains might tend to develop characteristics which would favor survival in the colder, higher areas such as increased fur.  The lowlands animals might develop characteristics favoring their conditions.  Since they were somewhat geographically isolated the two groups would tend not to interbreed (genetic isolation).   The two new varieties are both more effective at surviving in their habitat than the original variety and so would probably replace the original variety.  After a long time the two varieties could evolve to be so different they would become separate species.  Eventually, the number of new species and variations being produced would be more or less matched by old species becoming extinct. 

Besides natural selection which is based on survival, Darwin recognized that sexual selection also played a role in evolution.  Sexual selection would involve advantages that an individual might have that did not affect its survival but did represent an increase in its probability of breeding such as ability to attract the opposite sex.  Darwin considered that sexual selection was weaker than natural selection.

Darwin noted that in domestic breeding, genetic diversity seemed of itself to be of value.  Mongrels tend to be healthier and more vigorous than inbred animals.

Humans would be classified in Darwinian terms as “domesticated” as opposed to “wild” animals because humans have probably not existed under “wild” conditions for at least a hundred thousand years.  Humans have also been selectively breeding themselves to an extent not generally seen in “wild” animals.  For example, one would expect the incidence of genetic diseases among humans to be higher than in a wild species because the effects of civilization and medical intervention allow individuals with adverse mutations to survive and propagate in a way not possible in a wild species.   Use of human data (such as actuarial data) in an attempt to prove or disprove theories based on natural selection (such as aging theories) is therefore highly suspect although commonly done. 

In later editions of Origin, Darwin provides a chapter, Miscellaneous Objections to the Theory of Natural Selection, in which he responds to objections raised by contemporary scientists.

If Darwin’s theory was correct, there would have existed at some time in the past individuals possessing all of those little variations extending from that original organism to each current species.  One objection was that the fossil record didn’t seem to support this as there were times when new species seemed to suddenly appear.  Darwin presented extensive geological arguments showing that the geological fossil record itself had gaps that would explain the non-discovery of some intermediate forms.

Another objection had to do with the intermediate survival value of various organs and structures.  Darwin agreed that his theory would be defeated if a single case could be found where an organ or structure did not have increased survival value in all of the incremental intermediate forms needed to evolve that organ or structure.  For example, the eye is a complex structure which might not appear to have value without all its complex parts including retina, cornea, iris, etc.   If this were true, there would be no incremental evolutionary path from no eye to complete eye.  Darwin was able to show that even in existing living species, there are examples of a continuum of minor variations of optical organs all the way from a light sensitive spot on a worm to an eagle’s eye.   He was also able to show that eyes evolved down at least two different evolutionary paths.

Another objection was the relative absence of longevity.  Since longevity was of value to any organism, wouldn't natural selection (if true) result in ever increasing longevity?  This was the first statement describing the conflict between Darwin's theory and longevity / aging.  Darwin's extremely brief response suggested that longevity was determined by natural selection but in a complex way involving (as had been noted by others) "the amount of expenditure in reproduction".

As we all know, life is mainly a matter of luck.  Whether any one animal survives or not is much more a matter of random chance than genetic complement.  Evolution works because the fates of billions of animals "average out" the effect of luck.  Therefore any discussion of evolution involves speaking of various probabilities.


Darwin's Dilemma

Prior to Darwin, there was little or no reason to distinguish longevity / aging from any other characteristic of an animal.  Whatever caused a rat to have curved teeth, beady eyes, and a long tail presumably also caused it to have a certain average lifespan, which, like the teeth, eyes, and tail, was different from other species.  Darwin and contemporaries therefore assumed longevity and aging were evolved characteristics.  

However, as indicated above, it was immediately apparent (ca 1859) that there was a conflict between the theory of natural selection and observed characteristics of animals with regard to longevity and aging.  Since longevity was clearly beneficial to the survival of animals, one would expect that longevity would evolve through natural selection to eventually become infinite.  Conversely, since aging was manifestly an undesirable characteristic, one would expect that aging would be "selected out" and disappear as a result of natural selection.  This conflict eventually led to the development of the various "non-evolved" theories in which aging was not an evolved characteristic but a result of other forces.  

Most non-evolved theories have difficulty explaining the wide variation in aging characteristics between different species.  This in turn led to the development of the various "evolutionary" non-evolved theories of aging in which longevity does indeed evolve to the point where animals live long enough to attain physical maturity and at least begin reproducing but then fall victim of other forces which cause aging.  Longevity is evolved but aging is not evolved, a subtle but very important distinction.  Aging characteristics, although undesirable, could result from random factors and be retained in an animal's genome because natural selection would only weakly tend to select them out.  This is so because aging in such a case would only weakly affect an animal's ability to reproduce.

At the same time it was recognized that reproduction was critical to the natural selection process.  Fitness (we will say individual fitness) is a term of art used to describe an animal's ability to not only survive, but also to successfully reproduce.  

Could a characteristic be evolved which actually slightly reduced an animal's probability of survival if it created a compensating increase in the probability of successful reproduction?  In fact, there are numerous examples of just such evolved characteristics.  One of the most obvious is "protection of young".  Animals that protect and support their young evolved from animals that did not.  An animal that protects its young is individually less likely to survive and have additional progeny than one that does not.  Apparently the increased probability of progeny survival compensates for the decreased probability of the individual surviving.  Protection of young would be considered an improvement in fitness because the net effect of the characteristic is to improve the ability to successfully produce progeny.  Fitness involves the idea of tradeoffs between individual survival and reproduction.  Fitness encompasses individuals and their progeny.

A related, similar, concept was that of group selection which theorized that tradeoffs could exist between individual survival or even individual fitness and the survival of a group such as a herd, colony, etc.  Group selection could explain observations such as animals protecting the young of other animals, hive behavior, herding, spiders eating their mates, etc. 

The fitness concept can be applied to species.  There have been a nearly infinite number of species in the succession of species between pond scum and any present day animal species.  Each of these species inherited its genetic structure from its ancestor species.  The modifications to genetic content during the lifetime of any given species are negligible compared to the total genome of the species.  We can consider species fitness as describing the probability of any given species giving rise to descendent species.  An increase in the probability that a species would have descendents would obviously be beneficial.  A characteristic benefiting a species could be passed to descendent species.  

Could a characteristic evolve which improved species fitness even though it had a negative effect on individual fitness?  Could a situation exist analogous to that of individual fitness where a characteristic which had a sufficiently small negative effect on individual fitness could be compensated by an increase in species fitness?  Alternately, could the group selection concept be extended to species?  Could a species level group selection characteristic exist?

In 1890, Weismann proposed that "programmed death" was a genetically programmed, evolved characteristic, and that this characteristic had evolved through natural selection because it conveyed a benefit to the species even though it had a negative effect on individual fitness.  Weismann's theory was subsequently largely ignored because it did not seem to adequately address two questions:

1) Is it really feasible for a species fitness characteristic (or species level group selection characteristic) having a negative effect on individual fitness to evolve?  Some critics felt that the mechanics of evolution would preclude this.  Wouldn't an individually adverse characteristic "select out" on a time scale that was short compared to the time required for a species characteristic to "select in"?  Wouldn't an even very minor individual fitness disadvantage override an even major species advantage?

2) Is there really a species benefit to aging and/or death?  Weismann's proposal that death, per se, was the species desirable characteristic, was attacked on the grounds that animals in the wild essentially never lived long enough to die of old age and that therefore the alleged benefit of programmed death could never have been realized and could not have driven evolution.

Because aging has a negative effect on individual fitness, these questions also apply to any other theory of aging as an evolved characteristic and will be discussed in detail.

A factor lending credibility to the feasibility of species fitness characteristics is the existence of mating rituals.  Mating rituals, like aging, have a negative impact on individual fitness.  While it is at least plausible that aging could be caused by random accumulated mutations (or similar), it certainly strains credulity to believe that a complex mating ritual is somehow the result of random forces and not an evolved characteristic.  Mating rituals therefore represent a problem for those claiming that species benefit characteristics with negative individual fitness effect cannot evolve. 

Comparison of Theories

The difficulty in reconciling the variation in aging characteristics between species with the simpler "random accumulation of damage" theories led to the development of the non-evolved "evolutionary" theories of aging which are now widely respected (See Reading - Gavrilov).  In these "evolutionary" theories, characteristics which support longevity, and are therefore individually desirable, do in fact evolve, but only up to a point.  The point at which longevity ceases to evolve is approximately the age at which the animal has at least begun to reproduce.   Characteristics that cause aging are genetically propagated.  Because some reproduction has or could have occurred, the negative effect of aging on individual fitness is small.  There is no a priori "fundamental" limitation on longevity.

The alert reader will notice that so far the non-evolved "evolutionary" theories are identical to an "evolved" theory such as Weismann's.  The only difference is in the mechanics proposed for the genetic propagation of the individually adverse aging characteristics:

In Weismann's or other "evolved" theory, the aging characteristics are retained because they have a positive "species" benefit which compensates for their minor negative impact on individual fitness.

Issues: Feasibility of "species fitness", existence of specific "species" benefit(s)

In the mutation accumulation theory (Medawar, 1952), individually adverse aging characteristics are gradually selected out because of their minor negative effect on individual fitness but are replaced by new, random mutation, aging characteristics at a rate which is in equilibrium with the rate at which old adverse characteristics are selected out.  

Issues: Evolution is said to occur "in tiny steps" by means of the accumulation of tiny positive mutations.  The functional evolutionary effect of the minor individually adverse aging mutations is the same as any mutation that has an equally minor negative effect on fitness.  If it is possible to accumulate mutations with a minor negative effect on individual fitness would there not be an equilibrium between the accumulation of all positive and negative mutations which would stop evolution at some early point, maybe the "amoeba" level?

Aging has a general character (weakness, loss of agility, reduction in reproductive vigor, multiple tissue deterioration, increased susceptibility to disease, physiological changes) that is quite similar between species, however the timing (age of onset) and aggressiveness of aging varies dramatically between even similar species.  If the accumulation and out selection occur on a time frame that is short with regard to a "species lifetime" then why is the general character of the random (presumably different) accumulated mutations so similar in different species?  If the accumulation occurs on a long time frame relative to "species", then why is the timing of (presumably the same mutations) so different? 

In the antagonistic pleiotropy theory (Williams, 1957), individually adverse aging characteristics are not selected out because they are inextricably associated with beneficial characteristics which have a compensating positive effect on individual fitness.  For example, a single gene might promote a process that was beneficial or essential during youth but damaging at older ages.

Issues:   This theory depends on the idea that a characteristic which is good at some point in an animal's life is bad at another point because of some change in circumstances between the two ages.  Are there really that many circumstances that change between mature and older mature as opposed to all the changes that occur between conception and mature?  Why wouldn't antagonistic pleiotropy be more of a problem during an animal's development than during a relatively static period in an animal's life?  

Some very similar species (eg. salmon) have very different aging characteristics especially regarding timing.  Why would the inextricably linked bad characteristics of similar "good" genes be so different?

Comment: Since evolution would try to find characteristics which had the same positive effects without the negative (aging) effects, this theory depends on the idea that certain essential positive effects cannot be obtained without the associated aging effect, a very pessimistic view in connection with potential treatments for aging.  This theory is functionally a cross between an evolved theory and a "inescapable random damage" theory.


In addition to the issues raised above, the non-evolved "evolutionary" theories still do not completely fit the wide species variations observed in the aging of similar species.

It is important to note that the negative effects of aging on individual fitness are identical in all three theories.  Some proponents of the non-evolved theories make much of the "impossibility" that characteristics that have negative effects on individual fitness could evolve while simultaneously claiming that the "accumulated mutations" (or whatever) have negligible or even zero effect on fitness!  


Evolution Promoting Characteristics

As will be explained, evolution is not something that necessarily just happens.  Many factors could enhance or degrade the rate at which evolution could take place in any particular species as well as the general effectiveness of natural selection in selecting beneficial  characteristics.  Characteristics that could facilitate the rate and quality of evolution could therefore be of value. 

Suppose an animal had a characteristic which promoted, facilitated, accelerated, or otherwise enhanced the natural selection of other characteristics.  Suppose that this promoting characteristic had, of itself, a negative effect on individual fitness.  Suppose also that the other characteristics that were promoted were beneficial and had a positive effect on fitness.  A tradeoff then exists between the negative fitness effects of the promoting characteristic and the positive effect of the promotion of the beneficial characteristics.  If the negative fitness effect of the promoting characteristic is sufficiently small, and if the positive fitness effect of the promotion is sufficiently large, then the net effect of the promoting characteristic and the promoted characteristics on individual fitness would be positive and the characteristic could evolve.  In summary, a promoting characteristic can itself have a negative effect on individual fitness and still be an evolved characteristic.  Since promoting characteristics result in a positive net effect on individual fitness this concept does not require a belief in "species fitness", "species level group selection", or even involve species or speciation.   Although we can refer to a promoting characteristic as having a "species benefit" it really has to do with evolution, not species.

Mating rituals are an example of just such a promoting characteristic.  As one of the most spectacular examples of a mating ritual, consider the Bighorn Sheep (Ovis Canadensis) which live in the Rocky Mountains of North America.  Bighorns (See Reading) reach sexual maturity in two years, mate only in November and December, have a gestation period of 6 months, bear one or two young, and live about 15 years.  The Bighorn have evolved an instinct (lets call it the head-butting instinct) that leads them to have head butting contests to determine which rams are to mate with the females.  Such contests have been known to last as long as 24 hours.  To support the head-butting mating ritual, the Bighorn have evolved extremely large and heavy horns that weigh as much as 10 percent of the entire animal’s weight.  Increased skull size, spine, and muscle mass needed to support the horns probably represent another 10 percent.

Although sexually mature at 2 years, the average male does not mate until 7 years of age because the mating ritual requires animals to be older and stronger to mate.

The head-butting mating ritual instinct has a negative effect on individual fitness since an animal which had the instinct would be less likely to breed than one that did not.  The probability of an animal breeding is severely reduced by the mating ritual because it has to survive on average for an additional five years (after sexual maturity) in order to breed, and because it has to "pass" the "test" imposed by the mating ritual.

In addition, the excess size of the horns is apparently a significant disadvantage to the Bighorn as it relates to its universe of food, predators, and environment.  The horns have little value in resisting predators and have no value in helping to obtain food.  The extra weight of the horns is a disadvantage for both.  The head-butting contests are noisy and attract predators.  

However, the mating ritual promotes the evolution of beneficial characteristics as follows:  Presumably the "test" imposed by the mating ritual selects animals with desirable characteristics such as strength, stamina, and agility.  In addition, by delaying mating until animals are older and stronger, the mating ritual allows generic natural selection more time to work.  Animals will have to pass a longer "life test" to breed.  Less competitive animals have a greater chance of having died prior to breeding.

The beneficial effect of the mating ritual on individual fitness resulting from the promotion of beneficial characteristics exceeds the direct negative effect resulting from restricting breeding.  The beneficial effect is very immediate.  If we removed the mating ritual from the sheep (eg. forced random breeding in captivity) we would expect the strength and other "beneficial" qualities of an average sheep to decline in the first generation.  

The second problem with Weismann's theory involved the absence of a plausible "species" benefit to aging.  As will be described the beneficial effect of aging is very similar to that of a mating ritual.  The following sections suggest four specific evolution promoting benefits of aging.   

Evolution Rate and Selection Rate

If evolution depends on natural and sexual selection then it follows that the maximum rate at which evolution can proceed is dependent on the rate at which selections occur.   According to Darwin, natural selections are accomplished by deaths of organisms that are less fit.  In other words, the natural selection rate is dependent on the death rate.

The rate at which a species could evolve could well determine whether a species will produce descendent species or become extinct without descendents.  Although some species have evolved relatively slowly, it is clear that the capability to evolve more rapidly is of value to other species.  The ability to evolve rapidly is more important for more complex species because more complex species are usually more specialized in their adaptation to their environment and therefore more sensitive to changes in their environment.  (We have seen previously that speciation involves increased specialization.)  If a species or variety could not adapt to changes in its prey which made the prey more adept at evading capture, or changes in predators which made them more effective hunters, it may well become extinct without leaving descendents.  Similarly, if a species was able to evolve more rapidly than its prey or predators, its population would reasonably be expected to increase.

At the same time, evolution becomes progressively more difficult as organisms become more complex.

More complex organisms tend to be larger and require more resources than less complex organisms.  Their populations therefore tend to be smaller.  Their death rates and selection rates are therefore smaller.

More complex organisms also tend to take more time to develop.  Therefore they require a longer lifetime in order to reach full maturity.  Because their lifetimes are longer, their death rates are proportionately less even for the same size population.


Evolution and Genetic Diversity

Another factor which affects evolution effectiveness is genetic diversity.  Matings between relatively dissimilar individuals would tend to enhance evolution relative to matings between relatively more similar individuals because natural selection is selecting among the differences and the greater the differences the more opportunity for beneficial selection.  As an illustration of this in the ultimate limit case, consider what would happen in a population of animals consisting entirely of identical clones: evolution would be impossible because there would be no differences between the individual animals for natural or sexual selection to select!


Focus of Natural Selection

The effect of natural selection tends to be greatest at (focused at) a specific portion of an animal’s development.  To illustrate this consider a hypothetical fish which lays 5000 eggs only ten of which (on average) survives to adulthood and only one of which survives to breed.  Most of the fish die as fry, a few make it to the fingerling stage, yet fewer survive to become adults, one breeds.

Many of the fry could have latent characteristics which would benefit the adult fish but which are not apparent at the fry stage.  Since these characteristics are not expressed at the fry stage in the fish’s development they would not affect which fry survived and which did not.  Natural selection in this case is focused on the fry stage since that is where 99+ percent of the selection is happening. 

Such an animal would have difficulty in evolving into a more complex creature since more complexity is going to have to involve enhancements and evolution of the adult animal.  In effect we could expect to rapidly evolve a more perfect fry but very much more slowly evolve the adult stage of this fish.  Characteristics which only appear in the adult are not evolved by selections which occur in some other stage of an animal’s life cycle.

The major difficulty for the evolution of this fish is that it has to simultaneously have good survival characteristics in the fry stage (2 cm long) and in the adult stage (30 cm).  The characteristics of prey / food, predators, and habitat are likely to be substantially different for a 30 cm fish than for a 2 cm fish.  Its situation would be similar to a mammal attempting to simultaneously adapt to living in mountains and lowlands.

A characteristic which tended to focus selections towards the most developed stage of an animal’s life cycle would therefore be an aid, both in evolving to a more complex form, and in being able to rapidly evolve the characteristics of the adult animal to handle a changing environment. 

The head-butting instinct in the Bighorn focuses natural selection on the adult period in the Bighorn’s life.  Although Bighorn are sexually mature at two years of age, most don’t breed until they are seven years old because of the head-butting mating ritual. 


Evolution and Protection of Young

An instinct which leads an animal to protect and support its young is an important aid to evolution because it greatly reduces the problem of having to have good survival characteristics at all stages of an animal’s development.  In a species that protects its young, the survival characteristics of the immature animal are relatively less important.  The probability of any particular animal surviving to become an adult is now more dependent on the survival characteristics of the parents and less on the characteristics of the immature animal.  Therefore, in species that protect their young, natural selection tends more to focus on the characteristics of the adult animal  which aids in the evolution of adult characteristics.

Another major advantage conveyed by the protection-of-young feature is a tremendous reduction in the number of progeny required per mating.  If each mating is the result of a selection then some mammals have only one progeny per selection as opposed to 5000 in the case of our hypothetical fish.  Each progeny can therefore be much more highly developed at the time of birth.  Deaths (selections) are therefore much more likely to occur in the adult animal.


Evolved Characteristics which Regulate Breeding

According to Darwin, the stable population of any wild species is determined by how the species relates to its external universe.  If a species develops characteristics which make it able to better deal with food, predators, environment, habitat, or other external conditions its population would increase.  If it is less able, its stable population would decrease, otherwise it would stay the same.

Short term variations in population size frequently occur as a result of hard winters, disease outbreaks, floods, droughts, and other short term conditions.  It would be important for a species to rapidly reproduce in such an event in order to avoid having its territory taken by other species.

The average age of animals in any stable wild population is heavily dependent on their breeding characteristics.  The death rate has to equal the birth rate in a stable population.  Therefore, the fewer limitations there are on breeding, the earlier an average animal would breed, and the lower the average lifespan must be.   

Mammals have a large number of characteristics which restrict breeding:

All of these characteristics have a negative effect on individual fitness because they restrict breeding and reduce the probability that an individual animal will have progeny.  Many of them also have compensating positive effects on individual fitness by increasing the probability that progeny will survive.  All of them also have promoting benefits because they increase average lifespan (see next section) thus focusing natural selection on the adult animal and allowing natural selection a longer period in which to work.

The Aging Mechanism and Evolution

An aging mechanism is an essential addition to the above list of evolved characteristics which limit breeding, focus selection and promote evolution of most species.  Mechanism is used here to denote a system of elements which together implement a necessary life function (like the digestive system). 

To illustrate this lets consider three different aging scenarios in a hypothetical wild mammal:

Case 1, optimum aging mechanism:  We consider a  mammal which is usually born in a litter of one, reaches sexual maturity in 2 years, is fertile in the month of November, has a mating ritual which usually delays mating until after the animal is fully mature at 3 years, has a gestation period of four months, and protects its young for 8 months.  This animal has an aging mechanism which begins to weaken the animal at 4 years and few animals pass the mating ritual and mate beyond 5 years.  Although under zoo conditions, the animal lives for an average of 20 years, in the wild a typical animal lives for 7 years because of deaths caused by predators, intra-species warfare, disease, accidents, environmental conditions, and inability to obtain food.  Aging drastically increases death rates for older animals because aging causes weakness, lessened agility, and more susceptibility to disease and adverse environment.  Aging therefore (not coincidently) increases the probability of death from all of the causes listed above.  (The evolutionarily significant things about aging are the age at which it begins to weaken the animal, and the rate at which deterioration proceeds beyond onset, not the typical lifespan under zoo conditions.)

The probability of an older animal having progeny is reduced because aging reduces the probability of it surviving to reproduce.  In addition, aging affects the probability of reproduction because it reduces an animal's reproductive vigor and reduces the animal's probability of passing mating rituals, independently of survival.

Notice that the mating ritual and annual fertility periods combine to promote selection of beneficial characteristics because a superior animal might pass the mating ritual a year younger than normal.  Likewise, the aging mechanism promotes selection because a superior animal is more likely to be able to survive and mate at an older age despite the effects of aging.  Aging acts as a kind of challenge for potential beneficial characteristics.  Superior animals are not only more likely to mate but are likely to be able to mate over a longer period.

The birth rate is affected by puberty age, fertility period frequency and length, and mating rituals.  The death rate is controlled by the aging mechanism, predators, disease, warfare, environment, and availability of food.  If all these things are optimally balanced, a stable population will result and selections will be maximized and maximally focused on adult animals.

Case 2, no aging mechanism:   Now lets modify our hypothetical mammal by eliminating the aging mechanism.  Older animals would therefore be physically and functionally identical to younger animals beyond the age of full maturity.  Now the probability that an animal would die in any given time period would be nominally constant after it reached full maturity instead of increasing with age.  Animals would still die of predators, disease, warfare, environmental conditions, accidents, and starvation but the probability of such death would not depend on age beyond full maturity.  

The probability of such an animal breeding would also be constant instead of decreasing with age.  The number of mature animals of any given age in the population would follow an exponential "half-life" decline with age but, statistically, the number of progeny an animal would have would increase with age.  Although there are indeed fewer older animals than younger ones, the older ones have disproportionately more progeny than in case 1.  The effect of aging on individual fitness is not negligible!

These animals have some intelligence and are capable of learning from experience.  Older animals will have more skills and knowledge regarding dealing with food, predators, environment, and competitors than younger animals.  In addition, once an animal obtains some position in the "pecking order" it would tend to retain that position for at least some period because these animals can remember their position and the position of other animals in the pecking order.  The effect of these additional factors would result in a situation where  animals would have a probability of dying which decreased with age (after maturity) and a probability of breeding which increased with age.  Older genetically inferior animals would use their (non-genetic) superior knowledge and skills to out-survive and out-breed younger, genetically superior competitors.  This is an obviously evolutionarily bad outcome and suggests that an aging mechanism would be more important for more advanced, more intelligent species.

Similarly, older animals would be less likely to die of an infectious disease because they would be more likely to have been already exposed to that disease and have some immunity.

The median lifespan of these animals would actually be lower than in the first case.   Since some few animals would live very long lives and have very many progeny, the remaining animals must be having fewer progeny because to have a stable population which is neither increasing or decreasing, the average birth rate must equal the death rate.  These other animals must be dying prior to reaching breeding age.  In case 2, the population would increase to a new stable size where increased population pressure would result in increasing compensating deaths of young animals before they breed from causes like starvation, infant mortality, disease, and warfare.

Genetic diversity would be very poor because a few animals are having most of the progeny.  Focus of selection would be poor since most deaths would occur in immature animals.  This group would be at a serious evolutionary disadvantage relative to the first group.

Case 3, Longer life span:  Finally, lets modify the aging mechanism of the animals so that they typically live for 14 years (twice as long), and adjust puberty age and other breeding characteristics such that birth rate equals death rate under those conditions.  Now the focus and genetic diversity would be fine but the death rate, selection rate and thereby the evolution rate would be half that of the first case, a severe disadvantage.


Life Cycle Characteristics

We have seen from the foregoing that an aging mechanism is one of a family of evolved characteristics including puberty age, length and spacing of fertility periods, mating rituals, etc. which regulate life cycle.  The life cycle characteristics must be closely coordinated to achieve optimum evolution rate.  A change in any single characteristic would require corresponding changes in other characteristics to continue to result in an optimum cycle.  Changes in external conditions (eg predators) might affect these characteristics.  Evolution to a more advanced animal which required longer to mature would require changes to life cycle characteristics.  We would therefore not be surprised to observe (as we do) that life cycle characteristics varied dramatically between otherwise fairly similar species or even varieties.  

There have been observations to the effect that the same species has displayed different aging characteristics depending on external conditions such as predators.  There are even observations to the effect that individual animals can adjust life cycle characteristics (including aging) in compensation for external conditions such as availability of food.

It would appear that an optimum mammal life cycle would provide time for development to maturity, a period for selection of adults, breeding, and rearing of young.  Any longer life cycle would put the variety at a disadvantage because of reduced evolution rate.

Age at puberty has some interesting tradeoffs which are similar to those of other life cycle characteristics.  Age at puberty, especially in males could physically be higher or lower than it is in any given mammal species.  (Some human conditions result in puberty occurring as young as four years of age.)  Any increase in the age at puberty represents a potential decrease in individual fitness because animals would have to survive longer in order to breed and therefore are individually less likely to have progeny.  So why doesn't male age at puberty evolve toward zero?  One obvious compensating factor is that older animals would be better able to protect and support their young, a fitness benefit.  

But there are other tradeoffs.  Because a higher age at puberty requires that animals survive longer prior to breeding, it is more likely that less competitive animals will have died prior to breeding.  Animals, in effect, have to pass a longer and therefore more severe "life test" in order to breed.  The quality of the progeny will therefore be higher in animals with a greater age at puberty.  Since the quality of the progeny can be passed to their progeny, and so on, accumulating and evolving, this is a "species" benefit.  Another "species" benefit of higher puberty age is that pressure to evolve adult characteristics (focus) is increased.  Age at puberty is therefore an example of a characteristic which has both individual and species fitness benefits.

A mating ritual has a "species" benefit because it increases the quality of progeny.  The ritual, of itself, presumably selects animals with superior qualities to breed.  In addition, it also allows generic natural selection more time to eliminate less competitive animals before breeding because of the effect of a mating ritual in delaying breeding until animals are older and stronger.  

In this second regard, a mating ritual is more flexible than age at puberty.  While age at puberty imposes a more or less fixed reproduction delay, a mating ritual is population sensitive.  In an area of low population, the mating ritual could result in little delay and therefore reduced impediment to breeding because there would be fewer males to compete.  In an area with a higher population density, competition would be greater resulting in a larger average delay.  This aids the species in rapidly repopulating after an event such as a famine while concentrating on "quality" once a substantial population has been achieved.  A mating ritual is a population sensitive restriction on breeding.

If aging has a "challenge effect", increases genetic diversity, increases focus, and decreases the negative effect of learning and disease immunity as outlined above, or if it even has as many as one of these evolution promoting benefits, it would also improve the quality of progeny.  The beneficial effect of aging on the quality of progeny would therefore appear to be remarkably similar to that of a mating ritual.  As an illustration, suppose that a wild sheep is stronger than average.  Because it is stronger, this sheep can pass the mating ritual and breed a year younger than an average animal thus having more progeny.  Because of the challenge effect of aging, the stronger sheep can also pass the mating ritual and breed a year older than an average animal thus having more progeny.  Aging and the mating ritual act to promote the beneficial characteristic "stronger".

Mating rituals can have a more severe negative effect on individual fitness than aging.  (In the Bighorn, the mating ritual is said to result in an average delay in males breeding of five years in an animal that reaches sexual maturity in two years and only lives to be 15!)


Diseases and the Aging Mechanism

Many diseases have increasing probability of occurrence with age.  Some diseases have mechanics and etiology which are plausibly related to the simple passage of time.  For example, cancer is thought to be caused by multiple mutations separated by cell multiplication, a process which is time dependent. 

However, if the aging mechanism has an evolutionary purpose and benefit as described above, it is reasonable to believe that many age related diseases are caused by, enhanced by, or in effect parts of the aging mechanism.   Major evidence that this is in fact the case is provided by several conditions where a genetic malfunction causes accelerated aging. 

There is a rare human disease, progeria, in which aging is greatly accelerated to the point where individuals usually die by age 14.  Progeria has an incidence of about one in four million births and is thought to be caused by a dominant single gene mutation in each victim.   A variant, Werner's syndrome, causes somewhat less accelerated aging resulting in victims dying usually by age 50.  Patients with these conditions have many age related conditions and diseases including atheriosclerosis and other heart disease, baldness, gray hair, joint disease, skin conditions, cataracts, osteoporosis, some cancers, and diabetes, some at ages younger than 10!  Needless to say, research on progeria victims could lead to important insights into the operation of the aging mechanism.


Aging and Caloric Restriction

Researchers have observed that caloric restriction (CR), the feeding of a nutritional but reduced diet, in many animals including primates causes a dramatic increase in lifespan (~50 percent) under zoo conditions.  The lifespan increase is accompanied by a delay in the appearance of features generally associated with aging such as increased susceptibility to diseases, weakness, etc.  (ie. The animals stay "younger", more active, and healthier longer, not just live longer.)   CR can apparently reverse effects of aging even if applied only late in an animal's life.  An especially interesting finding is that the CR animals appear to have lower rates of some cancers.  

The existence of the CR phenomenon provides an important tool for development of treatments:  A method might be found for simulating or stimulating the anti-aging effect of CR without experiencing CR itself.  A CR mimetic, D2G, has already been found (see Reading -  Lane) which mimes the effect of CR in animal trials.  (D2G is considered too toxic for human use but searches are underway to find additional mimetic agents.)  

"Gene chip" studies are being performed (Reading - Spindler) comparing gene expression in CR animals to control animals to find differences which might lead to identifying CR related genes and thereby potentially aging related genes.  This could lead to identifying a reliable indicator of "aging" so that effectiveness of potential anti-aging treatments could be rapidly evaluated.  (Currently researchers have to "wait for some rats to die" in order to have a generally accepted indication that an anti-aging treatment is effective.  Human trials using this approach could take a very long time for each trial!)

Could the relaxation in the aging mechanism be an evolved response to starvation.  One would think that if the cause of starvation was overpopulation then a logical response would be a reduction, not an increase in lifespan, so that the population would be more rapidly brought under control.

On the other hand, if starvation was the result of an event such as drought, a reasonable response might be a relaxation of the aging mechanism combined with a reduction in birthrate.  Maintaining an adult population would require less resources than producing and supporting young and would position the species for rapid reproduction once the event was over.  


Sex, Salmon, and Elephant Teeth

There are reports that people having more sexual activity tend to live longer.  It is hard to establish cause and effect here because healthier people would be expected to be more sexually active.  However, it is known that sexual activity tends to increase levels of certain hormones.

Salmon are interesting in that they display one of the most spectacularly aggressive aging mechanisms.  A five year old salmon essentially falls apart within days of spawning although some varieties can survive after spawning to spawn in a subsequent year.  Some varieties only live two years.  Some varieties live a variable number of years before spawning but always die almost immediately after spawning.  The aging mechanism onset is apparently triggered by reproductive activity.

Sexual activity may extend the lifetime of humans but is definitely bad for salmon!

There have been other observations to the effect that reproductive activity in various species affects aging.

Humans have two sets of teeth.  In prehistoric times, loss of many teeth in the second set presumably led to weakness and increased mortality.

Elephants, which as herbivores eat almost continuously, have 6 sets of teeth each of which gradually wears out and is replaced by new teeth.  When the last set wears out the elephant starves.

Why don't humans and elephants have more sets of teeth?  Are teeth part of a "self destruct" mechanism which would be functionally similar to aging?  Are teeth controlled by the aging mechanism?

Some animals including sturgeon, some rockfish, and some turtles apparently do not age.  These animals have little or no observed increase in mortality rate or disease rate or decrease in reproduction rate or decline in physiological capacity (strength, agility) with age.  

Since there are only a few non-aging species amid many similar aging species, the non-aging animals must be descendents of aging species.  They appear to have lost the ability to age.  Is this an example of species group selection in progress?  If we came back in a hundred million years would the non-aging species have disappeared without leaving descendents?  

Non-aging animals present an enormous research opportunity.  We should be able to identify genes and associated processes and mechanisms that are unique to the aging animals or unique to non-aging animals.  

Non-aging animals represent an obvious difficulty for the non-evolved school:  How did they manage to "un-accumulate" their inherited accumulated aging genes?  How did they escape the "accumulation of random damage to the building blocks of life"?  How did their inextricably linked aging characteristics get un-linked?  How did they escape the "inescapable biological reality"?

Conclusions and Speculation

Aging is an  evolved, genetically programmed, characteristic which has the following "species" beneficial effects by virtue of promoting the evolution of other individually beneficial characteristics:

Aging would tend to be more necessary in species which are rapidly evolving because it enhances evolution rate.

Aging would be more necessary in more complex organisms which otherwise have lower evolution rates.

Aging would be more necessary in more intelligent species because of the otherwise adverse evolutionary effects of experience.

Aging seems to be a member of a family of closely interrelated life cycle characteristics which need to work together to maintain optimum evolution rate for a species.

The most significant parameters of an aging mechanism are: age of onset of aging related symptoms, and aggressiveness or rate at which aging proceeds once onset age has passed. 

Aging (like other life cycle characteristics) needs to be among the most “adjustable” of evolved characteristics.  Aging characteristics vary dramatically between even closely related species (eg various varieties of salmon). 

All of the life cycle characteristics involve “biological clocks”.

All of the other life cycle characteristics involve hormones.

In the same species there are apparent relationships between sexual/reproductive activity and aging.  

In the same species there are apparent relationships between external conditions such as the availability of food and prevalence of predators and aging.

So an obvious inference is that aging is very likely to be mediated by hormones.  (Are there any other biological functions involving biological clocks and multiple tissue types that do not involve hormones?)  

Is it possible that there are aging and/or anti-aging hormones that have yet to be discovered?  Human hormones are still being discovered.  The most recent of about 50 known human hormones (ghrelin) was discovered in 1999.

The levels of many hormones are known to vary with age. 

Aging is most likely to be mediated by a complex combination of hormones (possibly including currently unknown hormones).  Hormones associated with reproduction are likely to be involved because of the observed relationships between reproduction and aging.

Based on the effects of progeria, (accelerated incidence of diseases),  it appears that some heart disease, some cancers, and many other age-related conditions would be delayed or ameliorated by a treatment which delayed the onset age or reduced the aggressiveness of the aging mechanism.  Other age-related conditions (possibly including some cancers) will be found to be independent of the aging mechanism and therefore not affected by such a treatment.  Some cancer rates could actually be increased by an anti-aging treatment which extended cell reproduction.

Research towards treatment of aging and aging related conditions is obviously heavily influenced by the perceived "cause" of aging.  Researchers of the evolved school will be looking at genes, hormones, and other elements of typical evolved biological mechanisms (e.g. How does CR affect hormone levels and gene expression?).  Researchers of the non-evolved school (currently substantially in the majority) will be looking for much more direct effects (e.g. How does CR affect free radicals?).

People belonging to the non-evolved school tend to be very pessimistic about the prospects for major anti-aging treatments.  Aging is, after all, a problem which 3.5 billion years of evolution has been unable to solve.  Aging is an "inescapable biological reality".   

The majority of researchers and health professionals have been taught that the non-evolved theories are "generally accepted" and are not aware that there are significant alternative theories or that none of the "generally accepted" theories have been proved.

The public also generally sees aging as approximately as immutable as the restriction on travel faster than light.  Research on fundamental causes of aging is therefore often seen as about as useful as study regarding how many angels can fit on the head of a pin, a foolish endeavor, a "search for the fountain of youth".  This view is reinforced by public pronouncements from the non-evolved school such as the recent article in Scientific American (See Reading - Olshansky).

If, on the other hand, aging is an evolved, genetically programmed characteristic, the prospects for major anti-aging treatments are dramatically better.  Aging is then essentially a universal genetic disease.  The task is now to somehow interfere with any one part of a presumably complex biological aging mechanism, without interfering with any of the other myriad mechanisms that we need to live and function.  This is a familiar problem.  Tools and techniques already developed and rapidly improving for treating genetic and even infectious diseases are applicable.

The two schools share a major commonality:  Those in the evolved school believe aging exists as a beneficial evolved species fitness characteristic because it has a sufficiently minor negative effect on individual fitness.  Many members of the non-evolved school believe that aging exists because of accumulated adverse mutations (or similar) that exist because they have a sufficiently minor negative effect on individual fitness.  Because of the commonality, much of the available evidence fits either set of theories.  Evolution is a difficult science.  The scientific controversy between the evolved and non-evolved schools has endured for 150 years and may continue for that much longer.  Successive generations of scientists toil to adapt their theories to new discoveries while remaining faithful to the basic tenets of their school.  It's almost like theology.

Recently, "aging genes" have been discovered in the nematode, fruit fly, and mouse.  Disabling these genes, which have no other known function, has resulted in lifespan increases of between 30 and 100 percent.  These species are among the most heavily genetically mapped and studied of all species so such genes are probably widespread.  Some evolved characteristic people see this as a "smoking gun" proving the correctness of their school.  The non-evolved school disagrees.

Here are some excerpts from the 2001 budget of the U.S. National Institutes of Health (NIH):





Heart, Lung, and Blood




Aging (inc. Alzheimer's, etc.)






Library of Medicine


In contrast, Americans spend about $2 B on chewing gum annually.

Imagine how these numbers would change if people believed that there actually was a reasonably short term possibility that a major treatment for aging was possible and that such a treatment would reduce or delay the incidence of heart disease and other aging related disease.  The anti-aging budget might exceed the bubble gum budget!  



The Origin of Species - Charles Darwin, 1859

The Descent of Man - Charles Darwin, 1871 - A collection of information and links on ageing, and gerontology by Joao Magalhaes at The University of Namur in Belgium -- supports non-evolved characteristic school

No Truth to the Fountain of Youth - Olshansky, Hayflick, and Carnes, Scientific American June 2002 - Provides warnings against common ineffective anti-aging remedies.  Aging is an "inescapable biological reality" caused by "the accumulation of random damage to the building blocks of life".   (51 scientists endorsed this position paper to the effect that aging is not and cannot be an evolved characteristic.)

Aging is a Specific Biological Function Rather than the Result of a Disorder in Complex Living Systems: Biochemical Evidence in Support of Weismann's Hypothesis , V. P. Skulachev  Moscow State University -- A paper supporting beneficial evolved characteristic school

Whence Cometh Death?  - J. Mitteldorf  University of Pennsylvania -- Good discussion of group selection and evolved vs non-evolved controversy -- supports beneficial evolved characteristic school.

Evolutionary Theories of Aging and Longevity - L. A. Gavrilov et al   University of Chicago Center on Aging  -- Links to many articles by this team - Excellent overview of "evolutionary" theories of aging including Weismann (evolved) and mutation accumulation / antagonistic pleiotropy (non-evolved) -  Describes negative impact of some aging theories on research - Cautions that all aging theories are just theories and should not unduly influence research - favors non-evolved school.

Shattered: Medawar's Test Tubes and their Enduring Legacy of Chaos - J. Bowles, Quarterly Review of Biology 73:3-49. (2000) -- Presents extensive criticism of Medawar's 1952 paper which is the basis of most subsequent non-evolved theories of aging.  Supports evolved school.

Reversing the Negative Genomic Effects of Aging with Short-Term Calorie Restriction - S. Spindler, University of California,  Scientific World October 12, 2001 -- see overview here

Progeria Research Foundation

Werner's Syndrome Overview - Information on Long-lived Animals with  "Negligible Senescence"

Longevity Records: Life Spans of Mammals, Birds, Amphibians, Reptiles, and Fish   - Max Plank Institute, ISBN 87-7838-539-3 -- The oldest lake sturgeon caught so far was 152 years old

The Serious Search for an Anti-Aging Pill - Scientific American Aug 2002, M. A. Lane, et al -- Describes experiments with D2G to simulate the effect of Caloric Restriction Information on Bighorn Sheep

Keywords: longevity, mortality, gerontology, lifespan, fitness, senescence, anti-aging

Copyright  ã May, July, October 2002 Theodore Goldsmith

Send comments to the author at:  

Web publication of this document provided by Azinet LLC at