life span

1. the longest period over which the life of any organism or species may extend, according to the available biological knowledge concerning it.

2. the longevity of an individual.

[1915-20]

* * *

Time between birth and death.

It ranges from a mayfly's day to certain trees' thousands of years. Its limit appears to depend on heredity, but such factors as (in humans) disease, natural disasters, war, diet, and habits such as smoking reduce it. Maximum life span is theoretical; more meaningful is average life span, which life-insurance companies and actuaries analyze and tabulate. Long-lived progenitors tend to beget long-lived descendants. A very-low-calorie diet appears to prolong life. Reduced infant mortality and improved sanitation and nutrition account for much of the increase since с 1800

from about 35 to over 70 years in most industrialized countries. The oldest well-documented age reached by a human is 122 years.

* * *

Introduction

 the period of time between the birth and death of an organism.

      It is a commonplace that all organisms die. Some die after only a brief existence, like that of the mayfly, whose adult life burns out in a day, and others like that of the gnarled bristlecone pines, which have lived thousands of years. The limits of the life span of each species appear to be determined ultimately by heredity. Locked within the code of the genetic material are instructions that specify the age (human aging) beyond which a species cannot live given even the most favourable conditions. And many environmental factors act to diminish that upper age limit.

Measurement of life span

      The maximum life span is a theoretical number whose exact value cannot be determined from existing knowledge about an organism; it is often given as a rough estimate based on the longest lived organism of its species known to date. A more meaningful measure is the average life span; this is a statistical (statistics) concept that is derived by the analysis of mortality data for populations of each species. A related term is the expectation of life (life insurance), a hypothetical number computed for humans from mortality tables drawn up by insurance companies. Life expectancy represents the average number of years that a group of persons, all born at the same time, might be expected to live, and it is based on the changing death rate over many past years.

      The concept of life span implies that there is an individual whose existence has a definite beginning and end. What constitutes the individual in most cases presents no problem: among organisms that reproduce sexually the individual is a certain amount of living substance capable of maintaining itself alive and endowed with hereditary features that are in some measure unique. In some organisms, however, extensive and apparently indefinite growth takes place and reproduction may occur by division of a single parent organism, as in many protists, including bacteria, algae, and protozoans. If these divisions are incomplete, a colony results; if the parts separate, genetically identical (clone) organisms are formed. In order to consider life span in such organisms, the individual must be defined arbitrarily since the organisms are continually dividing. In a strict sense, the life spans in such instances are not comparable to those forms that are sexually produced.

      The beginning of an organism can be defined by the formation of the fertilized egg in sexual forms; or by the physical separation of the new organism in asexual forms (many invertebrate animals and many plants). In animals generally, birth is considered to be the beginning of the life span. The timing of birth, however, is so different in various animals that it is only a poor criterion. In many marine invertebrates the hatchling larva consists of relatively few cells, not nearly so far along toward adulthood as a newborn mammal. For even among mammals, variations are considerable. A kangaroo at birth is about an inch long and must develop further in the pouch, hardly comparable to a newborn deer, who within minutes is walking about. If life spans of different kinds of organisms are to be compared, it is essential that these variations be accounted for. The end of an organism's existence results when irreversible changes have occurred to such an extent that the individual no longer actively retains its organization. There is thus a brief period during which it is impossible to say whether the organism is still alive, but this time is so short relative to the total length of life that it creates no great problem in determining life span.

      Some organisms seem to be potentially immortal. Unless an accident puts an end to life, they appear to be fully capable of surviving indefinitely. This faculty has been attributed to certain fishes and reptiles, which appear to be capable of unlimited growth. Without examining the various causes of death in detail (see death) a distinction can be made between death as a result of internal changes (i.e., aging) and death as a result of some purely external factor, such as an accident. It is notable that the absence of aging processes is correlated with the absence of individuality. In other words, organisms in which the individual is difficult to define, as in colonial forms, appear not to age.

Peter W. Frank

Plants

       Maximum ages for some seed plantsPlants grow old as surely as do animals; however, a generally accepted definition of age in plants has not yet been realized. If the age of an individual plant is that time interval between the reproductive process that gave rise to the individual and the death of the individual, the age attained may be given readily for some kinds of plants but not for others. The Table (Maximum ages for some seed plants) lists maximum ages, both estimated and verified, for some seed plants.

Problem of defining age

      An English oak that has 1,000 annual rings in the trunk is 1,000 years old. But age is less certain in the case of an arctic lupine that germinated from a seed that, containing the embryo, had been lying in a lemming's burrow in the arctic permafrost for 10,000 years.

      The mushroom caps that appear overnight last for only a few days, but the network of fungus filaments in the soil (the mycelia (mycelium)) may be as old as 400 years. Because of important differences in structure, the life span of higher plants cannot be compared with that of higher animals. Normally, embryonic cells (that is, cells capable of changing in form or becoming specialized) cease to exist very early in the life of an animal. In plants, however, embryonic tissue—the plant meristems (meristem)—may contribute to growth and tissue formation for a much longer time, in some cases throughout the life of the plant. Thus the oldest known trees, bristlecone pines (bristlecone pine) of California and Nevada, have one meristem (the cambium) that has been adding cells to the diameter of these trees for, in many cases, more than 4,000 years and another meristem (the apical) that has been adding cells to the length of these trees for the same period. These meristematic tissues are as old as the plant itself; they were formed in the embryo. The wood, bark, leaves and cones, however, live for only a few years. The wood of the trunk and roots, although dead, remains a part of the tree indefinitely, but the bark, leaves, and cones are continually in the process of dying and sloughing off.

      Among the lower plants only a few mosses (moss) possess structures that enable an estimate of their age to be made. The haircap moss (hair-cap moss) (Polytrichum) grows through its own stem tip each year, leaving a ring of scales that marks the annual growth. Three to five years' growth in this moss is common, but life spans of 10 years have been recorded. The lower portions of such a moss are dead, though intact. peat moss (Sphagnum) forms extensive growths that fill acid bogs with a peaty turf consisting of the dead lower portions of mosses whose living tops continue growing. Mosses that become encrusted with lime (calcium carbonate) and form “tufa” beds several metres thick also have living tips and dead lower portions. On the basis of their observed annual growth, some tufa mosses are estimated to have been growing for as long as 2,800 years.

      No reliable method for determining the age of ferns (fern) exists, but on the basis of size attained and growth rate, some tree ferns are thought to be several decades old. Some club mosses (club moss), or lycopsids, have a “storied” growth pattern similar to that of the haircap moss. Under favourable conditions some specimens live five to seven years.

      The woody seed plants, such as conifers (conifer) and broadleaf trees (tree), are the most amenable to determination of age. In temperate regions, where each year's growth is brought to an end by cold or dryness, every growth period is limited by an annual ring (growth ring)—a new layer of wood added to the diameter of the tree. These rings may be counted on the cut ends of a tree that has been felled or, using a special instrument, a cylinder of wood can be cut out and the growth rings counted and studied. In the far north growth rings are so close together that they are difficult to count. In the moist tropics growth is more or less continuous, so that clearly defined rings are difficult to find.

      Often the age of a tree is estimated on the basis of its diameter, especially when the average annual increase in diameter is known. The source of greatest error in this method is the not infrequent fusing of the trunks of more than one tree, as, for example, occurred in a Montezuma cypress in Santa María del Tule, a little Mexican village near Oaxaca. This tree, described by the Spanish explorer Hernan Cortés in the early 1500s, was earlier estimated on the basis of its great thickness to be 6,000 years old; later studies, however, proved it to be three trees grown together. Estimates of the age of some English yews (English yew) have been as high as 3,000 years, but these figures, too, have turned out to be based on the fusion of close-growing trunks, none of which is more than 250 years old. Increment borings of bristlecone pines have shown specimens in the western United States to be 4,600 years old.

Growing season of seed (seed and fruit) plants

Annuals (annual)

      Plants, usually herbaceous, that live for only one growing season and produce flowers and seeds in that time are called annuals. They may be represented by such plants as corn and marigolds, which spend a period of a few weeks to a few months rapidly accumulating food materials. As a result of hormonal changes—brought about in many plants by changes in environmental factors such as day length and temperature—leaf-producing tissues change abruptly to flower-producing ones. The formation of flowers, fruits, and seeds rapidly depletes food reserves and the vegetative portion of the plant usually dies. Although the exhaustion of food reserves often accompanies death of the plant, it is not necessarily the cause of death.

Biennials

      These plants, too, are usually herbaceous. They live for two growing seasons. During the first season, food is accumulated, usually in a thickened root (beets, carrots); flowering occurs in the second season. As in annuals, flowering exhausts the food reserves, and the plants die after the seeds mature.

Perennials

      These plants have a life span of several to many years. Some are herbaceous (iris, delphinium), others are shrubs or trees. The perennials differ from the above-mentioned groups in that the storage structures are either permanent or are renewed each year. Perennials require from one to many years growth before flowering. The preflowering (juvenile) period is usually shorter in trees and shrubs with shorter life spans than in those with longer life spans. The long-lived beech tree (Fagus sylvatica), for example, passes 30–40 years in the juvenile stage, during which time there is rapid growth but no flowering.

      Some plants—cotton and tomatoes, for example—are perennials in their native tropical regions but are capable of blooming and producing fruits, seeds, or other useful parts in their first year. Such plants are often grown as annuals in the temperate zones.

Longevity of seeds

      Although there is great variety in the longevity of seeds, the dormant embryo plant contained within the seed will lose its viability (ability to grow) if germination fails to occur within a certain time. Reports of the sprouting of wheat taken from Egyptian tombs are unfounded, but some seeds do retain their viability a long time. Indian lotus seeds (actually fruits) have the longest known retention of viability. On the other hand, seeds of some willows lose their ability to germinate within a week after they have reached maturity.

      The loss of viability of seeds in storage, although hastened or retarded by environmental factors, is the result of changes that take place within the seed itself. The changes that have been investigated are: exhaustion of food supply; gradual denaturing or loss of vital structure by protoplasmic proteins; breakdown of enzymes; accumulation of toxins resulting from the metabolism of the seed. Some self-produced toxins may cause mutations that hamper seed germination. Since seeds of different species vary greatly in structure, physiology, and life history, no single set of age factors can apply to all seeds.

Lawrence Kaplan

Animals

      Much of what is known of the length of life of animals other than man derives from observations of domesticated species in laboratories and zoos. One has only to consider how few animals reveal their age to appreciate the difficulties involved in answering the apparently simple question of how long they live in nature. In many fishes, a few kinds of clams, and an occasional species of other groups, growth is seasonal, so that annual zones of growth, much like tree rings, are produced in some part of the organism. Among game species, methods of determining relative age by indicators such as the amount of tooth wear or changes in bone structure have yielded valuable information. Bird bands and other identifying marks also make age estimation possible. But one of the consequences of the fact that animals move is that very little is known about the life span of most species as they exist in nature.

Maximum and average longevity

       Maximum longevity of animals in captivityThe extreme claims of longevity that are occasionally made for one species or another have consistently been proven false when subjected to critical scrutiny. Although the maximum life span that has been observed for a particular species cannot be considered absolute, since a limited number of individuals at best has been studied, this datum probably provides a fair approximation of the greatest age attainable for this kind of animal under favourable conditions. Animals in captivity, which provide most of the records of extreme age, are exposed to far fewer hazards than those in the wild. In the accompanying table (Maximum longevity of animals in captivity) of maximum longevity, particular species have been so selected as to encompass the known range of longevity of other members of the taxonomic group to which they belong.

Environmental influences

      Life span usually is measured in units of time. Although this may seem eminently logical, certain difficulties may arise. In cold-blooded animals in general, the rate of metabolism that determines the various life processes varies with the temperatures (temperature) to which they are exposed. If aging depends on the expenditure of a fixed amount of vital energy, an idea first proposed in 1908, life span will vary tremendously depending on temperature or other external variables that influence life span. There is considerable evidence attesting at least to the partial cogency of this argument. So long as a certain range is not exceeded, cold-blooded invertebrates do live longer at low than at high temperatures. Rats in the laboratory live longest on a somewhat restricted diet that does not permit maximum metabolic rate. Of perhaps even greater significance is the fact that many animals undergo dormant (dormancy) periods. Many small mammals hibernate; a number of arthropods have life cycles that include periods during which development is arrested. Under both conditions the metabolic rate becomes very low. It is questionable whether such periods should be included in computing the life span of a particular organism. Comparisons between species, some of which have such inactive periods while others do not, are dangerous. It is possible that life span could be measured more adequately by total metabolism; however, the data that are necessary for this purpose are almost entirely lacking.

      Length of life is controlled by a multitude of factors, which collectively may be termed environment (ecosystem), operating on a genetic system that determines how the individual (population) will respond. It is impossible to list all the environmental factors that may lead to death. For analytical purposes it is, however, useful to make certain formal separations. Every animal is exposed to (1) a pattern of numerous events, each with a certain probability of killing the individual at any moment and, in the aggregate, causing a total probability of death or survival; (2) climatic and other changes in the habitat, modifying the frequency with which the various potentially fatal events occur; and (3) progressive systemic change, inasmuch as growth, reproduction, development, and senescence are characteristics intrinsic in the organism and capable of modifying the effects of various environmental factors.

Patterns of survival

      Consider a group of similar animals of the same age. Although no two individuals can have precisely the same environment, let it be assumed that the environment of the group remains effectively constant. If the animals undergo no progressive physiological changes, the factors causing death will produce a death rate that will remain constant in time. Under these conditions, it will take the same amount of time for the population to become reduced to one-half its former number, no matter how many animals remain at the beginning of the period considered. The animals therefore survive according to the pattern of an accident curve. This is the sense in which many of the lower animals are immortal. Although they die, they do not age; how long they have already lived has no influence on their further life expectation.

      Another group of animals may consist of individuals that differ (polymorphism) markedly in their responses to the constant environment. They may be genetically different, or their previous development may have caused variations to arise. Those individuals that are most poorly suited to the new environment will die, leaving survivors that are better adapted (natural selection). The same result can also be achieved in other ways. If the environment varies geographically, those individuals that happen to find areas in which existence can be maintained will survive, while the remainder will die. Or, as a result of their own properties, animals in a constant environment may acclimate in a variety of ways, thus adjusting to the existing conditions. The pattern of survival that results in each of these cases is one in which the death rate declines with time, as illustrated by the selection–acclimation (acclimatization) curve.

      In the absence of death from other causes, all members of a population may exist in their environment until the onset of senescence (aging), which will cause a decline in the ability of individuals to survive. In a sense they can be considered to wear out as does a machine. Their survival is best described by individual differences among members of the population that determine the curvature of the survival line (wearing-out curve). The more the population varies, the less abrupt is the transition from total survival to total death.

      Under the actual conditions of existence of animals the three types of survival (accident pattern, selection–acclimation pattern, wearing-out pattern) above all enter as components of the realized survival pattern. Thus in animals that are carefully maintained in the laboratory, survival is approximately that of the wearing-out pattern. Environmental accidents can be kept to a minimum under these conditions, and survival is almost complete during the major part of the life span. In all known cases, however, the early stages of the life span are characterized by a noticeable contribution of the selection–acclimation pattern. This must be interpreted as a result of developmental changes that accompany the early life of the individuals and of selective processes that operate on those organisms whose genetic constitutions are ill fitted for that environment.

      In some of the larger mammals in nature, the existing evidence points to a similar survival pattern. In a variety of other animals, however, and including fishes (fish) and invertebrates, mortality in the young stages is so high that the selection–acclimation curve predominates. One estimate places the mortality of the Atlantic mackerel during its first 90 days of life as high as 99.9996 percent. Since some mackerel do live for several years, a mortality rate that decreases with age is indicated. Similar considerations probably apply to all those animals that have larval (larva) stages that serve as dispersal mechanisms.

      When the postjuvenile portion of the life span is considered by itself, a number of animals for which such information has been gathered—including primarily fishes and birds—have survivorship curves that are dominated by the accident pattern. In these species in nature, death from old age apparently is rare. Their chance of surviving to an advanced age is so small that it may be statistically negligible. In modern times, human predation is a large factor in the mortality of these species in many cases. Since deaths from fishing and hunting are largely independent of age, once an animal has reached a certain minimum size, such a factor only makes the survival curve steeper but does not change its shape. One consequence of such increased mortality is that fewer old and large individuals are noticed in a population.

      More complex survival patterns, such as the hypothetical one illustrated, undoubtedly exist. They should be looked for in those species in which extensive reorganization of the animal is part of the normal life cycle. In effect, these animals change their environment radically, in some cases several times during a lifetime. The frog offers a familiar example. During its period of early development and until shortly after hatching, the animal is subject to major internal, and some external, change. As a tadpole it is adjusted to an aquatic, herbivorous life. The metamorphosis to the terrestrial, carnivorous adult form is accompanied by varied physiological stresses that must be expected to produce a temporary increase in mortality rate. In some insects (insect) the eggs, larvae, pupae, and adults are exposed to and respond to quite different environments, and a survivorship pattern even more complex than that described by the composite curve may exist.

      The same species will exhibit changed survival in different environments. In captivity an animal population may approach the wearing-out pattern; in its natural habitat survivorship may vary with age in a quite different way. Although one can assign a maximum potential life span to an individual—while realizing that this maximum may not be attained—it is impossible to specify the survivorship pattern unless the environment is also specified. This is another way of saying that life span is the joint property of the animal and the environment in which it lives.

Human life span

      The exact duration (human aging) of human life is unknown, although there is presumably a maximum life span for the human race established in the genetic material. At first thought, this statement seems irrational. Surely no human being can live 1,000 years. Even though all may agree that the likelihood of an individual living 1,000 years is infinitesimal, there is no scientific proof that this statement is or is not true. The indeterminacy of the maximum limit of human life is made more comprehensible if one chooses a number that may appear to be a more reasonable limit.

      Since there is no verified instance of a person having lived 150 years, this number may, for purposes of illustration, be arbitrarily accepted as the maximum limit of the span of human life. But if the possibility is admitted that an individual may live exactly 150 years, there is no valid reason for rejecting the possibility that some other individual may live 150 years and one minute. And if 150 years and one minute is accepted, why not 150 years and two minutes, and so on? Thus, based on existing knowledge of longevity, a precise figure for the span of human life cannot be given.

Studies on longevity

      Much information concerning the inheritance of longevity has come from the study of genealogical (genealogy) records of nobility and landed gentry. The early genealogical studies were criticized on the grounds that the downward trend in the death rate (attributable generally to scientific advancements) introduced a spurious correlation in statistics derived from records extending over long periods of time. It was argued that in some instances records were included of persons who, at the time of the study, had not had the opportunity of living out their possible life span. The general finding of such investigations was that the expectation of life of sons of long-lived parents (i.e., those living to age 70 years or older) was greater than that of sons of shorter-lived parents (i.e., those having attained less than age 50 at the time of death).

      An American biostatistician attempted to avoid the defects of genealogical studies by collecting records of the family histories of 365 nonagenarians (90-year-old persons) and of a comparison group of 143 individuals of varying ages, selected because all of their six immediate ancestors were dead. The study introduced the concept of “total immediate ancestral longevity,” or TIAL—the sum of the ages at death of the two parents and the four grandparents of a given person—as a measure of longevity. This number is unlikely to be greater than 600 or less than 90. The average TIAL of the nonagenarians and centenarians definitely exceeded that of the comparison group. This held true not only for the six immediate ancestors as a group but also for each category—father, mother, paternal and maternal grandparents. In the same study, investigators also computed the expectation of life for sons of fathers as classified in three groups by age at death: (1) under age 50, (2) from age 50 to age 79, and (3) age 80 or over. The expectation of life for the three groups at birth was 47.0, 50.5, and 57.2 years, respectively. The same relative ranking continued through the lifetime of the sons, their expectation of life at age 40 being 27.3, 28.9, and 32.0 years, respectively.

      While certain doubts have been raised about the validity of these as well as earlier studies, taken at their face value, these data show clearly that long-lived persons had parents and grandparents who lived longer than the parents and grandparents of shorter lived persons.

      Since longevity is important in life insurance underwriting, several studies have been made of the relationship between heredity and the life span by an analysis of life insurance records. Such analyses showed that policyholders both of whose parents were living when the policy was written live longer than those whose parents were dead when the policy was written. These results are in conformity with those obtained from genealogical records and family histories.

      Each of the various types of studies of the inheritance of longevity—genealogical records, life insurance records, and family histories of the general population—has limitations that restrict the applicability of the findings. The principal studies indicate, nevertheless, that the children of long-lived parents are more likely to be long-lived than are the children of short-lived parents. Conversely, the immediate ancestors—parents and grandparents—of long-lived persons on the average are older at death than are the immediate ancestors of persons who die at a relatively young age. These studies support the conclusion, mentioned earlier, that longevity is determined in part by heredity.

Actual versus possible life span

      It should be observed that this conclusion relates to the inheritance of longevity—the observed expression of the span of life—and not to the span of life itself. The actual length of life itself is shorter than the possible life span, since the former reflects the effect of unfavourable environmental factors. In the absence of any biological data from which the maximum limit of the span of life can be determined precisely, an estimate of the limit must be obtained from observation of the actual length of life of persons who already have died. But such observations cannot establish a fixed limit for the span of life.

      The estimation of the length of the span of life from observed data is a form of sampling from a large but incomplete population. The tabulation of the ages at death of a large number of persons from a large general population of the United States will give an asymmetrical frequency distribution with two modes, or peaks, of highest frequency: the first at age less than one year and the second between ages 75 and 80 years. The frequency distribution is bounded by age zero at the lower limit but there is no boundary at the upper limit. The number of deaths of persons whose length of life is near the upper limit of this frequency distribution (e.g., 100 years or more) varies from year to year. The age of the oldest person dying also varies from year to year.

      The number of deaths of centenarians (100-year-old persons) depends in part upon the number of deaths counted. Ages at death are frequently unverified, so that the true numbers of centenarians almost certainly deviate from those given in official vital statistics. Moreover, only a very small proportion of the deaths that have occurred throughout the history of the human race have been registered. The potential number of future deaths greatly exceeds the number that already has occurred. Statistical theory supports the expectation that as the total number of deaths continues to increase, the death of a person whose length of life will be longer than that of any person previously known will be recorded.

      Observation of the length of life of persons who have died can show that it is possible for a human being to live to the oldest age recorded as of any specified date and can provide an estimate of the relative frequency or probability of that event. But such observations do not provide a logical basis for fixing any age as the maximum possible limit of the life span.

      The continuation of the worldwide decline in the death rate will naturally result in an increase in the number of persons who live until age 100 years or more. Since the number of persons who may live to an advanced age, such as 110 or 115 years, is directly related to the number of persons who live to age 100, an increase in the latter number will increase the probability that the death of an individual attaining some greater age (e.g., 115 years) will be recorded at some future date.

      Many instances of persons alleged to have died at an age considerably greater than 100 years have been recorded. Statements concerning the age at death of biblical characters such as Methuselah can be dismissed, since scientific verification is impossible. Three of the most frequently cited cases of more recent times are: Thomas Parr, who died in November 1635 at the alleged age of 152 years; Henry Jenkins, who died in December 1670 at the alleged age of 169 years; and Catherine, countess of Desmond, who died in 1604 at the alleged age of 140 years. William Harvey (Harvey, William), a famous English physician, performed an autopsy on Thomas Parr and the account of the autopsy was cited for many years as evidence that Harvey—in his paper—had confirmed Parr's age. Quite apart from the fact that it is impossible accurately to determine the age of a person by an autopsy, Harvey made no attempt to verify Parr's age but merely referred to the current estimates. Subsequent investigations have revealed that no proof exists of the age at death of any of these three individuals and that their reported ages were based solely upon hearsay.

      An example with more definite documentation is that of Christian Jacobsen Drakenberg, stated to have been born on November 18, 1626, and to have died on October 9, 1772, aged 145 years and 325 days. Although the authenticity of his age was attested to by many persons, including two celebrated Scandinavian actuaries, later investigations cast doubt upon the record. It is difficult to accept the statements concerning Drakenberg's age at death, since this age is more than 30 years greater than the next oldest verified age at death—a difference that in itself casts doubt on its authenticity.

      Of eight individuals for whom records substantiate the fact that each had lived more than 108 years, seven were females. Six of the eight were more than 110 years old at death. The oldest was Pierre Joubert, who was born July 15, 1701, and died November 16, 1814, aged 113 years and 124 days. Discounting the Drakenberg record, this is the oldest age at death that has been generally accepted as authentic.

      It may be concluded that the span of human life is at least 114 years, but that this is not the maximum upper limit. This does not mean the span of life of each individual now living or to be born in the future is at least 114 years. The span of life, since it is determined by heredity, varies from one individual to another as do other genetically determined traits.

      A significant proportion of human embryos and fetuses die before birth. Other infants at birth have defects (childhood disease and disorder) that limit their span of life to a few years. Some malformations (e.g., certain cardiovascular defects) are developmental rather than genetic in the strict sense of the word and can be corrected so that the length of life of such persons is extended.

      In the past the length of life of most individuals has been considerably shorter than their possible span of life because of unhealthful environmental factors. As these factors are increasingly brought under control or eliminated, the actual length of life will approach more closely the span of life. At the end of the 18th century the expectation of life at birth in North America and northwestern Europe was about 35 or 40 years. By 1970 it exceeded 70 years, and at some future date the death of a person at an authenticated age of more than 114 years can be expected.

      There is no evidence that the span of human life has increased since the beginning of recorded history. Neither is there any evidence that the death rate of centenarians has decreased. The expected increase in the number of centenarians results from a decrease in the death rate at ages under 100 years and not from any demonstrable increase in the maximum length of the span of life. The remarkable increase in the average length of life during the past 2,000 years—from 20–25 years to 70 years under favourable conditions—has increased the likelihood that a person may live to the maximum limit of his span of life.

Peter W. Frank

Additional Reading

Studies of longevity include Leonid A. Gavrilov and Natalia S. Gavrilova, The Biology of Life Span: A Quantitative Approach, rev. and updated ed. (1991; originally published in Russian, 2nd ed., rev. and updated, 1991); and Joep M.A. Munnichs et al. (eds.), Life-Span and Change in Gerontological Perspective (1985), a report of diverse research on behavioral development across the life span. United Nations Demographic Yearbook includes mortality statistics for all countries, showing the influence of economic, social, and climatic factors on mortality.

* * *