Human life expectancy.
YEARLife Expectancy

2000 BC 18
500 AD 22
1400> 33
1790 36
1850 41
1900 50
1946 67
1991 76

Almost all living organisms pass through a sequence of changes, characterized by growth, development, maturation and senescence. Up until last year the oldest person ever to live was thought to have been a Japanese man who reached 120. However, it was discovered that his birth record was not accurate and, therefore the oldest person ever to live is now living--a woman in France who was born in 1875. The isolated villages in Russia where many people were supposedly very old turned out to be a myth. None of the 500 or so people who said they were between 120 and 170 years old could produce reliable document of birth, marriage, education or military service.

People are living longer. For example, in 1900 only 25% of U.S. population survived to age 65. Now almost 70% survive to this age. Currently about 30% attain age 80 and within a few years 50% will attain age 80. It is speculated that there no person attained the age of 100 before 1800 and no one lived to over 110 prior to 1950.

Three Primitive Questions--The Biology of the Finitude

  1. Why do people grow old? This is concerned with the process of senescence.
  2. Why do people live as long as they do? This question is concerned with the biology of longevity.
  3. Why do people die? This question is concerned about the transition from illness to death.

A Model of Aging (from Gavrilov and Gavrilova, 1991)

The reduction in endogenous causes of death (cancer, heart disease) may have little effect on prolonging life because the speed of transition from the healthy state to the state of "unspecific vulnerability" (also referred to as the "one-foot-in-the-grave" state) determines the size of the death quota. This quota is then distributed among "causes" of death. It is argued that this explains why elimination of individual endogenous causes of death cannot significantly alter the magnitude of the biological component of mortality--organisms will simply die of another competing risk once in the vulnerable state. The future of research on aging and life span prolongation belongs to a strategy founded on explaining the mechanisms which determine the organism's resistance to a broad spectrum of harmful factors and not in eliminating specific causes, per se.

Concept of Maximal Length of Life

One of the most compelling concepts in human biology and demography is maximal length of life or the age limits of the human species. The validity of this concept is viewed by many as self evident because:

  1. Different species exhibit different life expectancies
  2. All individuals eventually die before the age of infinity; and therefore
  3. Each of the different species must possess unique and finite maximal ages.

"I don't want to achieve immortality through my work. I want to achieve it through not dying." (Woody Allen, Film Director)

The problem with this concept is that everything we know about the nature of mortality makes it difficult to accept the notion that there is an age--one single year--which some may reach but no one has any chance of surviving. The only valid alternative is the existence of an asymptote to which the probability of dying tends and that may or may not be near unity. Gavrilov and Gavrilova (1991) also reject the notion of a human-specific upper age limit and argue that people die before the age of infinity, not because they cannot pass bounding age, but because the probability of a person riding out the ever present risks of death for that long is infinitesimal. These authors cited over 50 references from the demographic, gerontological, and biological literature containing 28 different estimates of the biological limit to the length of human life ranging from 60 through 200 years. They noted that these values were not measured but prescribed. For example, Fries (1980) based his hypothesis that the maximal life expectancy of humans is around 85 years on the assumption of "ideal societal conditions" but does not specify these conditions (what diet? what patterns of work, leisure, family life?).

The most obvious evidence that the length of human life is limited is that we have not observed extremely long-lived people, say over 130 years old. However, the assumption of a genetically fixed upper bound to life expectancy and life span ignores the likelihood of strong genetic-environment interactions because these interactions likely had relatively little manifest effect until recently. This may account for the upward historical trend of oldest ages. Some demographers have speculated that no one may have lived to age 100 before 1800 or to 110 before 1950. Wilmoth (in press) analyzed data from five countries (Sweden, England and Wales, France, Japan and the United States) and found that the maximum age attained by humans has been increasing for at least 130 years. He suggested that the relevant limit of human life span may be in the pace of improvement rather than in the ultimate level to be attained.

An Exceptional Case of Human Longevity, Jeanne Calment
(Gerontological Society of America, New Orleans, Nov. 23, 1993)

Centenarian Divorce. The French group of demographers and gerontologists studying Madam Calment was also studying the life style and habits of centenarians. A centenarian couple in their study decided to seek a divorce. When asked why they had waited so long to make this decision they responded, "We waited out of consideration for our children. Thus we postponed the decision until they all had died." (story told by K Ritchie over dinner at Gerontological Meeeting in New Orleans, Nov., 1993).

The oldest age to which any person on record has lived was a French woman, Madam Jeanne Calment, who was born in the small town of Arles on February 21, 1875 and died August 4, 1997 at age 122 years, 164 days.

"The plausibility of certain reported cases of longevity" V. Kannisto. To test the plausibility of a reported extreme age, we first have to determine the universe to which the person belongs. A woman who has reached the age 118 any time after 1950, reached the age 100 between 1931 and 1975. We estimated that in ca. 30 countries with reliable information (combined population 600 million), there were 57,000 such women. Applying and extrapolating the most reliable survival data we find that there was one chance in twenty that one of them would reach the age 118. This is 4 to 20 times the customary confidence limits in statistics and the case is entirely plausible. Using the same method we find it extremely unlikely (probability 0.002) that a man, as has been claimed (Japanese), would have reached the age 120.

"I see badly, I hear badly, I can't feel anything, but everything's fine." (J. Calment on 120th birthday, Sacramento Bee, March 14, 1995)

"Validation of the age of Jeanne Calment" J. M. Andrieux. It is our great advantage that Jeanne Calment was born and lived in the ancient Roman city of Arles (South of France), with its long history of civil administration. It has thus been possible to validate her age by tracing the course of her life via civil and religious documentation and population census data. Not only has it been possible to verify the age of Jeanne Calment with a high level of certainty, but also to provide international guidelines for the validation of cases of extreme longevity.

"Present health status and medical history of Jeanne Calment". B. Forette. The present physical health status of Jeanne Calment is not normal, it is very good, as indicated by all standard clinical parameters. Everyday functioning is surprisingly good as measured by ADLs (activities of daily living). The three principal difficulties with which she is confronted are bilateral cataracts, deafness and an environment which confines her to a chair. The medical history is astonishing by its absence, up until an elbow and hip fracture at the age of 114.

"Smoking History" Madame Jeanne Calment began smoking when she married in 1896 at the age of 21. She smoked no more than 2 cigarettes per day. We do not know whether she inhaled nor do we know what brand she smoked. She quit smoking at about age 113-114 when she broke her hip. She had to quit because her cataracts prevented her from following the cigarette into her mouth.

LONDON (Reuter—August, 1997). The Guinness Book of Records on Thursday confirmed 116-year-old Canadian Marie Louise Febronie Meilleur as the world's oldest person. She takes over the record from Frenchwoman Jeanne Calment, who died last week at the age of 122. Guinness said it confirmed Meilleur's age after checking her birth and baptism certificates, census records and two marriage certificates. ``We have an extremely stringent process of verification to ensure that only someone whose age can be proved beyond any shadow of doubt can be recognised as the record holder,'' said Guinnesse Keeper of the Records Clive Carpenter. Meilleur, who now lives in a nursing home in Corbeil, north Ontario, has 300 descendants from two marriages. Her family said the secret of her longevity was an active life and hard work. She used to enjoy fishing and still loves the outdoors. Guinness also confirmed that the world's oldest man is 114-year-old Christian Mortensen. The cigar-smoking centenarian emigrated from Skaarup, Denmark, to the United States in 1903 and now lives in San Raphael, California.

Christian Mortensen died April 25, 1998 at age of 115 yrs and 252 days.


Longevity is Moderately Heritable in Twins.

The evidence of human longevity was investigated in a sample of 218 pairs of monozygotic (MZ) and 382 pairs of like-sex dizygotic (DZ) Danish twin pairs. Findings included: (1)Twin similarity for age at death was significant for MZ twins but nonsignificant for DZ twins; (2)The average age difference at deaths for two identical twins, two non-identical twins and two random individuals was 14.1, 18.5 and 19.2 years, respectively; (3)Genetics plays a moderate role in determining a person's longevity. (from McGue et al., 1993 J. Gerontological Soc. Amer. 48, B237)

  1. The Hayflick Limit (Replication Limit)--states that cells have a limited capacity for division. Steps: i)obtain tissue (e.g. fetal lung tissue); ii)culture in laboratory until cover laboratory dish; iii)subsample laboratory culture and inoculate two new dishes; iv)allow these to grow until cover dish; v)repeat. Findings: only can repeat this about 50 times at which point the cells do not keep growing. Conclusion: the number of doublings is finite and can be used as a model of cell aging. Organisms, like cells, are not immortal but have a limited life span. Criticism of this theory includes the concept that there exists two kinds of cells in cell cultures--post mitotic cells which do not have the capacity for division and cells with the properties of stem cells with an unlimited capability for division. They note that scientists determine the death of a cell culture in a highly arbitrary way (i.e. if cell numbers do not grow within 1 to 4 weeks) and that cell death is operationally defined using their definition of cell culture death. What occurs [in experimental cell cultures] is not the disappearance of dividing cells, but only the dilution of the culture with postmitotic cells. Consequently there may be no grounds for speaking of the limited capability of all the cells in the culture for division.
  2. Stochastic Theories (Errors Theory): i)Somatic Mutation Theory--mutations will produce functional failure eventually resulting in death; ii)Error Catastrophe Theory--protein molecules containing errors will eventually be turned over and replaced by correct copies. However if an erroneously synthesized protein is a protein involved in the synthesis of the genetic apparatus, then this original error will produce faulty molecules in the genetic apparatus, which will then produce more error-containing proteins.
  3. Developmental-Genetic Theories (Midnight Hour Theory): i)Neuroendocrine Theory--body contains a master timekeeper; ii)Intrinsic Mutagenesis--each species is endowed with a specific genetic constitution which regulates the fidelity of the genetic material and its replication. The degree of fidelity regulates the rate of appearance of mutation or errors and thus the life span; iii)Immunological Theory--fidelity of the immune system declines with age as evidenced by the striking age-associated increase in autoimmune disease.
  4. Rate of Living Hypothesis (Gas Tank Theory)--a long standing hypothesis that the duration of potential lifespan varies inversely with metabolic intensity. What I refer to as the 'gas tank' model--the young has a full tank of gas which runs out as they age depending upon their rate of living. People die at different ages partly because they are endowed with different sized tanks (i.e. live fast, die young strategy).
  5. Negative Pleiotropy Hypothesis (Pay Later Theory)--genes are selected on the basis of advantages to fitness early in life but harmful later in life. The deleterious genes are not selected out because selection decreases with age. For example, a gene that increases fitness at age 20 when a person is at a peak reproductive age will be passed on to the next generation. However, if this same gene has a deleterious effects later on in life such as an increase in the risk of acquiring cancer at older ages, it will not be selected against since it occurs at an age when reproduction is low or nonexistent. This theory is embraced by many evolutionary biologists.
  6. Longevity Assurance (Spaceship Model). The concept of longevity assurance is that senescence did not evolve, rather the ability to survive to a certain age evolved and that senescence is simply a process of falling apart after the primary mission of reproduction is achieved. For example, a space ship designed for a planetary fly-by must be built sturdy enough to make it to the planet. But because it was build sturdy enough to make it there also means that it is likely to keep on going, perhaps to a new solar system. The strategy is to design a species to ensure that they continue to regenerate their populations. This can be done by either: i)low survival rate but producing large numbers of individuals; or ii)high survival rate but smaller numbers produced. The first strategy is that individuals are cheap and thus disposable whereas the second strategy is that individuals are well build (Mercedes) and thus neither cheap nor disposable.


    LONG-LIVED MAN. Dr. John W. Rowe, president of the Mount Sinai School of Medicine in New York, provides a thumbnail sketch of the man least likely to have coronary heart disease (the leading cause of death in the United States): "An effeminate (women live longer) municipal worker or embalmer (low stress) with no drive, never has to meet a deadline of any kind, underfed, subsisting on a diet laced with fruit, vegetables and cod-liver oil, with a full head of hair (baldness is a risk factor), hates smoking, avoids TVs and automobiles, works out with his puny muscles, low in income (wealthy people are at higher risk) and blood pressure, who has undergone a prophylactic castration." (from Sac. Bee Supplement Ap. 23, 1989).
  1. Increased disease, morbidity, mortality with age. At age 11 the probability of dying prior to age 12 is 1 out of 6,000 whereas at age 100 the probability of dying prior to age 101 is about 1 out of 3. Increased vulnerability to disease with age.
  2. Physical size. Both standing and sitting height decreases with age beginning in males around 40 and females at 43. Weight increases in the middle years and decreases in old age. Men achieved a maximum average weight of 172 pounds in 34 to 54 year range but fell to 140 pounds by age 75.
  3. Progressive deteriorative changes. Skin shows dramatic changes with age including discoloration, wrinkles and deterioration. Hair growth decreases and hair loss. Sense of hearing changes with age in most people, particularly a decrease in ability to hear the higher frequencies. Starts as early as late 30s. Contrary to common belief, there does not seem to be a loss in the number of taste buds in older people. However some deterioration in taste bud cells and ability to detect the four primary tastes (salt, sweet, bitter and sour) decreases with age. Sight--lens becomes thicker with age reducing the ability to focus on close-up objects.


Table 1. Increments of chronic disease (from Fries and Crapo, 1981).

Age Stage Arthero-
Cancer Oestero-
Diabetes Emphysema Cirrhosis

20 I. Start elevated

obesity smoker drinker
30 II. Discernible small
plaques on
slight joint

liver on
40 III. Subclinical larger

50 IV. Threshold leg pain
in situ
GI hemo-

60 V. Severe chest

70 VI. End stroke,
disabled blindness;

The concept in Table 1 is that chronic diseases develop through a long series of stages. People will live longer and healthier if we can develop ways to either prevent the onset of some of these diseases or reduce the rate of their progression.

The compression of morbidity thesis notes that the species life span if finite while the onset of chronic disease is relatively easilty delayed (Fries 1988). Thus the period from onset of chronic infirmity to death may be shortened, with benefit to both individuals and society.

Vigorous Exercise and Mortality. Study published in J. Amer. Med. Assoc. (April, 1995) found that fates of 17,300 middle-aged men depended upon the amount of vigorous exercise they engaged in. Examples of activity needed to reduce by 25% the chance of dying over a 26-year period included: walking at 4-5 mph for 45 mins five times/wk, playing one hour of singles tennis 3 times/wk, swimming laps for 3 hours/wk, cycling for one hour 3 times/week or jogging at 6-7 mph for 3 hours/wk.

Fries (1988) characterizes the basic syllogism of the compression of morbidity as follows:

  1. If morbidity may be defined as that period from the onset of the first irreversible chronic disease or aging marker until death
  2. If the date of occurrence of that marker can be postponed until later in life
  3. If the rate of such postponement can be greater than the rate of increase in adult life expectancy
  4. Then morbidity for the average person can be compressed into a shorter period of time.

Finite Human Life Span Evidence (1983, Milbank Mem. Fund Qrt. 61, 397):

  1. There are no exceptions to the declining numbers of individuals present at successive ages.
  2. Gompertz's law, as published by the English actuary Benjamin Gompertz in 1825, appears to hold in all populations and assures an exponentially increasing mortality rate and, therefore, death for the entire population within a decade or two past the age of 100.
  3. There has been no historical change over several centuries of observation with regard to maximum life potential, as underscored by studies of centenarians; this observation has been repeatedly made in the United States with good data since 1939 or earlier. Life expectancy at age 100 has changed at most 0.7 years over 80 years and much of this improvement must have been due to reduction of premature death, not change in life span.
  4. There is no biological reason to assume that any change in genetic longevity characteristics should have occurred merely because we have improved infant mortality, cleaned up water supplies, or invented penicillin.
  5. The difference in species life span among animals is a commonplace daily observation; for example tortoise =175, man=115, horse=46, chicken=30, cat=28, mouse=4.
  6. Anthropological analyses suggest a formula by which mammalian life spans may be predicted by the brain size/body weight ratio; such models suggest an approximately constant life span for the human species for the past 100,000 years.
    Death, is that you? When a new disability arrives, I look about me to see if death has come, and I call quietly, "Death, is that you? Are you there?" and so far the disability has answered, "Don't be silly. It's just me." (F. S. Maxwell quoted in "The Fragile Species, Lewis Thomas, 1992. Macmillan)
  7. The linear decline in organ reserve, repeatedly the subject of physiologic observation, mandates a point at which function must be inadequate to support life, that point apparently being when organ reserve is reduced to approximately 20 percent over that function required for the maintenance of basic life processes; reserve of this magnitude is required for daily functions outside of bed.
  8. The increasingly rectangular curve demonstrates the barrier to immortality.
  9. We have the important phenomenon of a priori aging, the daily evidence of our senses. People do grow older, with changes which are apparent to all of us, as we age. And these changes--from hair color to hearing--are not the result of disease as we usually define it.

Mortality Rates for Centenarians

The mortality rates for centenarians were presented in a workshop on Oldest Old Mortality sponsored by the National Institute on Aging held at Duke University, March, 1993. Note the following: i)the mortality rates are extremely high--30 to 50% or more die at each age; ii)the males and females are relatively similar; and iii)the rates do not increase according to the Gompertz model.

The Baby Boomers and Aging

The 76 million people born in the 19 postwar years from 1946 to 1964 are considered the Baby Boomer generation. Their numbers are 150% of those born in the previous 19 years. The first Boomers will reach 50 in the year 1996 and the last cohort of Boomers in 2014 and will reach age 80 in 2026 and 2044, respectively. Some believe that the Boomers will be candidates for trendsetters as fiftysomethings and beyond (Rukeyser, 1995):


Anderson, W. F. 1995. Gene therapy. Scientific American Sept. 1995. 124-128.

Caskey, C. T. 1993. Presymptomatic diagnosis: a first step toward genetic health care. Science 262:48-49.

Collins, F. and D. Galas. 1993. A new five-year plan for the U.S. human genome project. Science 262: 43-46.

Graubard, S. R. (Ed.). 1986. The Aging Society. Daedalus 115 (Winter, 1986)

Finch, C. 1990. Longevity, Senescence, and the Genome. The University of Chicago Press, Chicago.

Fries, J. F. 1980. Aging, natural death, and the compression of morbidity. New England J. Med. 303:130-135.

Fries, J. F. 1983. The compression of morbidity. Milbank Mem. Fund, Qrt. 61:397-419.

Gavrilov, L. A. and N. S. Gavrilova. 1991. The Biology of Life Span: A Quantitative Approach. Harwood Academic Publishers, Chur, Switzerland.

Hamilton, J. B. & Mestler, G. E. (1969) Mortality and survival: comparison of eunuchs with intact men and women in a mentally retarded population. Journal of Gerontology, 24, 395-411.

Hayflick, L. 1994. How and Why We Age. Ballantine Books, New York.

Hazzard, W. R. (1990) The sex differential in longevity. Principles of Geriatric Medicine and Gerontology. 2nd Edition. (eds. W. R. Hazzard, R. Andres, E. L. Bierman and J. P. Blass) pp. 37-47. McGraw Hill, New York.

McGue, M., J. W. Vaupel, N. Holm and B. Harvald. 1993. Longevity is moderately heritable in a sample of Danish twins born 1870-1880. J. Gerontology 48:B237-B244.

Nesse, R. 1995. Why We Get Sick. Westminster, Maryland: Random House.

Rukeyser, William. 1995. Let's do the hobble. Atlantic Monthly, July, pp16-18.

Sacher, G. A. 1978. Longevity and aging in vertebrate evolution. BioScience 28:497-501.

Siebert, C. 1995. The DNA we've been dealt. NY Times Sunday Magazine. 50-57, 64, 74, 93-95.

Williams, G. C. and R. M. Nesse. 1991. The dawn of Darwinian medicine. Quarterly Review of Biology 66:1-22.

Data from surveys on attitudes and intentions are usually poor predictors of behavior. A woman may do well in predicting her future fertility given her present circumstances; she may not, however, clearly foresee her future circumstances, which of course would have greater bearing on her future fertility than would her present situation (Bennett, 1983).


Reproduction involves three necessary steps: i)intercourse ii)conception and iii)completion of gestation. Therefore the following set of intermediate fertility variables were proposed by Davis and Blake (1956):

  1. Intercourse Variables--factors affecting exposure to intercourse including age of entry into sexual unions
  2. Conception Variables--factors affecting exposure to conception including:
  3. Gestation Variables--factors affecting gestation and successful parturition including:

The three components of birth interval:

  1. Postpartum Infecundable Interval--ovulation and menstruation is absent; primarily a function of breastfeeding.

  2. Waiting Time to Conception--affected by natural fecundability and use of contraception.

  3. Full-term Pregnancy--this birth interval segment is basically constant at 9 months.


Sperm24 to 48 hours
Ovum12 to 24 hours

  1. Only 2 to 3 days in which can conceive due to life of egg and sperm
  2. Nearly a 50% probability of becoming pregnant if inseminated during 2 days prior and 1 day after ovulation.
  3. Only a maximum of 10% probability of becoming pregnant on any given day from 8 to 25 of the 28 day menstrual cycle. The reason is that ovulation does not always occur at mid-cycle. Can occur anywhere from day 8 to day 25.


  1. Hormonal contraception (the pill)--disrupts ovulation

  2. Barrier methods--prevents sperm from reaching egg (condoms, diaphragms and caps, spermicides, coitus interruptus and periodic abstinence)

  3. Intra-uterine contraception (IUD)--relies on a local inflammatory effect that stops implantation

  4. Abortion--kills fetus

  5. Sterilization (tubal ligation)--physically blocks passage of egg to uterus.

CONTRACEPTIVE METHODS (after Djerassi 1989):

  • New spermicide with antiviral properties--drug needs to be developed that is effective both as a spermicide and one that will kill AIDS virus particles.

    Hellin's Rule--the ratio of of twin births to single births is 1 out of 86, of triplet births is 1 out of 862 (= 1 out of 7,396), quadruplet births is 1 out of 863 (= 1 out of 636,056) and so forth.

  • 'Once-a-month' pill effective as a menses-inducer--instead of the current pill which is taken daily for most of all of each month, the menses-inducing pill would be taken by a woman only during those months when she had unprotected coitus. Instead of waiting to see whether she had missed her period, a woman would take the pill to induce menstrual flow at the expected time. The single most important advantage of such a method is that the decision to contracept is made postcoitally. RU484 known as the "abortion pill".

  • A reliable ovulation predictor--as human sperm has a fertile life span in the women's reproductive tract of up to 3 days, couples wishing to have unprotected intercourse during the first half of the menstrual cycle need to be able to predict the onset of ovulation by approximately 3 days beforehand.

  • Easily reversible and reliable male sterilization--vasectomy would have to be guaranteed reversible before young men without children would opt for such a method of fertility control.

  • Male contraceptive pill--problem is verification of safety only through long term epidemiological studies. This is due to long reproductive period of men.

  • Antifertility vaccine--would radically change our perception of human fertility if teenage males or females or both were vaccinated so that they would be infertile until a conscious step was taken to achieve fertility. To accomplish prompt restoration of fertility a method would be needed that actively reversed the immunological infertility.





    An estimated 65,000 children are born each year as a result of artificial insemination and other methods using stored sperm. However, human in vitro (test tube) fertilization (IVF) is surprisingly unsuccessful (Winston and Handyside 1993). In the U.S. the overall birth rate per IVF treatment cycle is 14% from 16,405 oocyte retrievals. Superovulation hopefully leads to fertilization of several oocytes, and it is common to transfer several embryos to the uterus, anticipating that at least one will implant. Pregnancy resulted from 13% (184 out of 1436) of transfers when three or fewer embryos were transferred and around 25% when four to six embryos were transferred. Of triplets and quadruplets born after IVF, from 65 to 75% required admission to intensive care. The quality of both the embryo and the uterine environment affects success.

    In Vitro Fertilization Research Trends (from Winston and Handyside 1993).

    In 1990 in the U.S. nearly 26,000 stimulation cycles were reported to the IVG-Embryo Transfer Registry, from which there were 5,150 pregnancies and slightly less than 4,000 live births. Assuming that around 10% of couples suffer infertility involuntarily, there may be upwards of 2 million women in the U.S. that will be candidates for IFV treatment.


    Sex preselection represents a potential extension of fertility control. Numerous studies have attested to the presence of sex preferences and thus it seems reasonable to suppose that in societies practicing extensive family limitation and birth spacing, many couples would be inclined to take advantage of techniques of sex preselection.

    Wives tales:

    Two types of goals are commonly recognized:


    Couples who deliberately interrupt their childbearing when a minimum number of sons, or a minimum number of daughters are obtained may be said to be following a "stopping rule." The question is whether stopping rules affect the sex ratio at birth which is normally 105 boys to 100 girls or 51.2% boys and 48.8% girls. If couples prefer to have boys, then stopping rules will result in skewing the sex ratio slightly toward girls. For example, if a set of couples are following the stopping rule of no more children after the second son is born, the effect is to lower, not raise the sex ratio. This result follows because couples predisposed to have daughters will have to bear more children on the average in order to bear two sons than will couples predisposed to have boys.

    But even when no couples are predisposed to having more daughters than sons, the consequences of the stopping rule of 2 sons has a minimal effect on the sex ratio. The table below presents the results of a computer simulation where 4,995 families were created using the stopping rule of 2 sons. The simulation was based on the assumption that there is a 50:50 chance of having a boy and that the couple continues having more children until the 2 boys are born. Note the following: i)51% of all families consist of either 2 children (2 boys) or 3 children (2 boys & 1 girl); ii)18.5% of all families consist of 2 boys and 2 girls. Thus a total of 69.5% of all families consist of 2 girls or less; iii)however, the remaining 30.5% of families consist of 5 or more children and thus contain more girls than boys. One of the nearly 5,000 families required the birth of 15 children to obtain their 2 boys; iv)of the 19,675 children born, there were 9,990 boys (2 boys/family X 4,995 families) and therefore 9,685 girls (19,675 children minus 9,990 girls). Thus the sex ratio of the population of 4,995 families is 50.8% boys and 49.2% girls. The results reveal that the ratio is nearly 50:50 despite a rule (stop at 2 sons) whose effect which, at least on first approximation, would appear to skew the population toward mostly males.

    Table 1. Results of computer simulation of 4,995 families using a stopping rule of 2 boys.

    No. Children
    in family
    Total number
    of families
    Total number
    of children
    of all families
    No. girls
    in family
    No. boys
    in family

    2 1,255 2,510 25.1% 0 2
    3 1,295 3,885 25.9 1 2
    4 926 3,704 18.5 2 2
    5 629 3,145 12.6 3 2
    6 380 2,280 7.6 4 2
    7 224 1,568 4.5 5 2
    8 135 1,080 2.7 6 2
    9 77 693 1.5 7 2
    10 38 380 .8 8 2
    11 16 176 .3 9 2
    12 11 132 .2 10 2
    13 5 65 .1 11 2
    14 3 42 .1 12 2
    15 1 15 .0 13 2
    4,995 19,675 100.0

    Woman only produces X gametes (XX zygote; homogametic sex); Men produces X and Y gametes (XY zygotes; heterogametic sex).

    Approaches to Sex Preselection:

    Old Wives' tales. Couples who pay to choose the sex of their child may be wasting their money, say Britain's sperm experts. Sex choice clinics claim that "male" sperm bearing the Y chromosome swim faster through viscous solutions of human albumin and so can be separated from "female" sperm bearing an X chromosome. But, according to John Aitken of the Universisty of Edinburgh, there is no good scientific data to confirm that X- and Y-bearing sperm do swim at different rates. Clinics that claim to select the sex of the sperm with an accuracy of over 70% use their own data. Researchers at the University of Edinburgh ended up with 50:50 male and female sperm using the same techniques.(New Scientist, 30 January, 1993).
    But there may be hope for the future. Methods for producing sex-selected cattle semen are close to becoming commercially available—XY, Inc. of Fort Collins, Colorado. The method involves sorting sperm by flow cytometry, in which cells are stained with a flourescent dye and then passed through a laser beam. Cattle X sperm contain about 4% more DNA than Y sperm do, so the dye makes the X sperm glow more brightly (1997 Science 276: 1503).

    1. Sperm separation in vitro and subsequent artificial insemination. Barlow and Vosa (1970) developed method to stain stem end of Y-chromosome. Procedure useful in determining efficiency of any sperm separation technique under development. Two types of sperm separation technique: i)chromotography; and ii)sedimentation using centrifuge. Both methods are extremely inefficient.

    2. Timing of coitus relative to ovulation within menstral cycles. This is based on the observation that male progeny conceived early and late in cycle while female progeny conceived in mid cycle. The biological basis for this is unknown. In order for this to work need to predict ovulation. Reason this is difficult is that i)calendar method is only an approximation; and ii)basal body temperature (temperature immediately upon waking) is difficult to measure precisely and detect the .5 to 1 F change in temperature. Problem: in natural insemination boys are conceived early and late in cycle but in artificial insemination girls are conceived early and late. In all cases the shift is very minor. (James, W. H. 1980. Time of fertilization and sex of infants. Lancet 1124-1126).

    3. Selective abortion after gender indentification of fetus. This is based on amniocentesis or sonograms which are techniques which allow for identification the sex of the fetus. Does not try to alter conception probabilities but only to eliminate those preganancies fated to result in an unwanted outcome.

    Consequences of Sex Pre-selection {(2,1) indicates 2 boys and 1 girl; B-G indicates the birth order of boy then girl}:

    1. Change Fertility Rates. Consider a couple whose first six choices for completed family compositions are, in order of preference: (2,1), (1,2), (1,1), (2,0), (0,2), 3,0), (0,3). In other words, this couple would like to have a family of three children as long as the three were not of the same sex. Even though this couple most prefers a family of three children, if their first two children were of the same sex they might well stop at two rather than take the risk of having a third of the same sex. If perfectly reliable sex selection were available, however, they could have a third child without taking that risk. Thus the availability of sex selection could serve to increase fertility.

    2. Change Sex Ratio. Consider a group of couples who all want to have three children regardless of sex composition but who most prefer the composition (2,1). With sex selection the preferred composition (2,1) could be achieved giving a sex ratio of 2. For this group the sex ratio would have changed without a change in fertility rate.

    3. Change Within-family Birth Order of Sexes. Suppose a group of couples all want two children regardless of sex composition but who most prefer (1,1) with the boy occupying the first-born position. With sex selection the order B-G could be achieved by all couples.


    A man who agrees to help a woman bear a child by donating sperm has a constitutional right to claim fatherhood, according to the Supreme Court. Under past Supreme Court rulings, unwed fathers who had an extended relationship with a woman were deemed to have a constitutional right to claim fatherhood of any child that is produced. These rulings involved children conceived through sexual intercourse, but Oregon courts said that the same principles apply to artificial insemination. In 1986, the California state courts issued a similar ruling. Though the state law said sperm donors do not have paternity rights, the state courts said this law covered only artificial insemination supervised by a doctor. A man who agreed to help a woman become impregnated without a doctor's involvement retained the right to seek paternity in court. (Sacramento Bee, April 24, 1990)


    Alexander, N. J. Future contraceptives. Scientific American Sept. 1995. 136-141.

    Bongaarts, J. and R. G. Potter. 1983. Fertility, Biology, and Behavior. An Analysis of the Proximate Determinants. Academic Press, New York.

    Djerassi, C. 1989. The bitter pill. Science 245:356-361.

    Ellison, P. T. 1994. Advances in human reproductive ecology. Annual Review of Anthropology 23:255-275.

    Maranto, G. 1995. Delayed childbearing. The Atlantic Monthly. June issue. pp. 55-66.

    Roush, W. 1996. Fertile results: bringing up baby (eggs). Science 271:594-595.

    Westoff, C. F. and R. R. Rindfuss. 1974. Sex preselection in the United States: some implications. Science 184:633-636.

    Winston, R. M. L. and A. H. Handyside. 1993. New challenges in human in vitro fertilization. Science 260:932-936.

    Return to the Course Syllabus