In all human populations, people usually select mates non-randomly for traits that are easily observable. Cultural values and social rules primarily guide mate selection. Most commonly, mating is with similar people in respect to traits such as skin color, stature, and personality. Animal breeders do essentially the same thing when they intentionally try to improve varieties or create new ones by carefully making sure that mating is not random. When they select mates for their animals based on desired traits, farmers hope to increase the frequency of those traits in future generations. In so far as the discriminated traits are genetically inherited, evolution is usually a consequence. However, the results are not always what farmers expect. The reasons why will be explained shortly.
Even without the intervention of farmers, most animals select mates carefully--they do not mate randomly. Charles Darwin noted this fact in his 1871 book Descent of Man and Selection in Relation to Sex. He suggested that mate selection is a powerful force of evolution similar in its effect to natural selection. This idea was widely rejected in Darwin's time, but later research showed that he was correct.
Why Animals Mate Non-randomly: Tale of the Peacock
video clip from PBS 2001 series Evolution
requires RealPlayer to view (length = 4 mins, 2 secs)
In order to understand the effect of non-random mating patterns, it is useful to first examine the results of random mating. As Hardy and Weinberg demonstrated in the early 20th century, the gene pool of a population that is mating randomly and is not subject to any other evolutionary process will not change--it will remain in equilibrium. If mating is entirely random, there will be nine possible mating patterns for a trait that is controlled by two alleles (A and a).
AA X AA Aa X AA aa X AA AA X Aa Aa X Aa aa X Aa AA X aa Aa X aa aa X aa
In a population which has 50% of each of these two alleles, the expected offspring genotype frequencies with random mating will be 25% homozygous dominant (AA), 25% homozygous recessive (aa), and 50% heterozygous (Aa), as shown in the table below. They will remain in this ratio every generation that random mating continues and no other evolutionary mechanism is operating.
The number of children
is what would be
expected by chance
if each mating pair has
You can work this out
yourself by creating a
Punnett Square for
each set of parents.
Random Mating Possible parent
Expected offspring genotypes AA Aa aa
AA X AA
AA X Aa 2 2 AA X aa 4 Aa X AA 2 2 Aa X Aa 1 2 1 Aa X aa 2 2
aa X AA
aa X Aa 2 2 aa X aa 4 Total 9
( 25% )
( 50% )
( 25% )
Positive Assortative Mating
The most common non-random mating pattern among humans is one in which individuals mate with others who are like themselves phenotypically for selected traits. This is referred to as positive assortative mating . The term "assortative" refers to classifying and selecting characteristics. An example of positive assortative selection would be tall slender people mating only with tall slender people. Taken to the extreme, positive assortative mating results in only three possible mating patterns with respect to genotypes for traits that are controlled by two autosomal alleles--homozygous dominant with homozygous dominant (AA X AA), heterozygous with heterozygous (Aa X Aa), and homozygous recessive with homozygous recessive (aa X aa).
The net effect of positive assortative mating is a progressive increase in the number of homozygous genotypes (AA and aa) and a corresponding decrease in the number of heterozygous (Aa) ones in a population, as shown in the table below. Each generation that there is positive assortative mating, this polarizing trend will continue in the population.
Positive Assortative Mating Possible parent
Expected offspring genotypes AA Aa aa
AA X AA
Aa X Aa 1 2 1 aa X aa 4 Total 5
( 42% )
( 17% )
( 42% )
Negative Assortative Mating
The least common non-random mating pattern among humans is one in which people only select mates who are phenotypically different from themselves for selective traits. This is referred to as negative assortative mating . It would occur, for instance, if tall slender people mated only with short stout people.
In terms of genotypes, there are six possible negative assortative mating patterns for traits that are controlled by two autosomal alleles, as shown in the table below. The net effect is a progressive increase in the frequency of heterozygous genotypes (Aa) and a corresponding decrease in homozygous (AA and aa) ones in a population. In other words, negative assortative mating has the opposite effect as positive assortative mating.
Negative Assortative Mating Possible parent
Expected offspring genotypes AA Aa aa AA X Aa 2 2 AA X aa 4 Aa X AA 2 2 Aa X aa 2 2
aa X AA
aa X Aa 2 2 Total 4
( 17% )
( 67% )
( 17% )
Evolutionary Consequences of Non-random Mating
Like recombination, non-random mating can act as an ancillary process for natural selection to cause evolution to occur. Any departure from random mating upsets the equilibrium distribution of genotypes in a population. This will occur whether mate selection is positive or negative assortative. A single generation of random mating will restore genetic equilibrium if no other evolutionary mechanism is operating on the population. However, this does not result in a return to the distribution of population genotypes that existed prior to the period of non-random mating. A comparison of the 2nd and 5th generations in the table below illustrates this fact.
Effects of non-random mating on a population’s gene pool Generation Parent mating pattern Offspring genotype frequencies Effect on
AA Aa aa 1 random 50% 30% 20% equilibrium 2 random 50% 30% 20% equilibrium 3 negative assortative 45% 40% 15% change 4 negative assortative 40% 50% 10% change 5 random 40% 50% 10% equilibrium 6 random 40% 50% 10% equilibrium 7 positive assortative 43% 45% 12% change 8 positive assortative 48% 34% 18% change
NOTE: genotype frequencies in an actual population may differ somewhat from those in
this table, but the direction of change from generation to generation will be the same.
Plant and animal breeders usually employ controlled positive assortative mating to increase the frequency of desirable traits and to reduce genetic variation in a population. In effect, they try to guide the direction of evolution by preventing some individuals from mating and encouraging others to do so. By doing this, farmers, in a sense are acting in the place of nature in selecting winners and losers in the competition for survival. This method has been used to develop purebred varieties of laboratory mice, dogs, horses, and other farm animals. The amount of time it takes for this process can be much shorter than one might imagine. If brothers and sisters are mated together every generation, it will only take 20 generations for all individuals in a family line to share 98+% of the same alleles—they essentially will be clones, and breeding results will be close to those resulting from self-fertilization. Commercially sold laboratory research mice have been mated brother to sister for 50-100 generations or more. The downside of this practice is that positive assortative mating results in an increase in homozygosity of harmful alleles if they are present in the gene pool. The high frequency of hip dysplasia , epilepsy , and immune-system malfunctions in some dog varieties are primarily a result of inbreeding. The reduction in viability and subsequent loss of reproductive potential of purebred varieties has been referred to as inbreeding depression. In contrast, animals that have been crossbred with mates from very different genetic lines are more likely to have lower frequencies of homozygous recessive conditions. Subsequently, they are liable to be more viable. This phenomenon has been referred to as hybrid vigor or heterosis .
Human mating rarely is as consistently positive assortative as is the case with purebred domesticated animals. As a consequence, inbreeding depression is rarely a problem except for some reproductively isolated small societies and subcultures. The Old Order Amish are an example. This relatively small population centered in Pennsylvania and Ohio has been self-isolated by their religious beliefs and lifestyle for more than two centuries. They mostly select mates from within their own communities, which results in positive assortative effects on their gene pool. The Amish population has a comparatively high frequency of Ellis-van Creveld syndrome , which is a genetically inherited disorder characterized by dwarfism, extra fingers, and malformations of the arms, wrists, and heart. The majority of the known cases in the world of this rare syndrome have been found among the Amish, and 7% of them carry the responsible recessive autosomal allele.
Consanguineous mating , or inbreeding, is the sexual union of closely related individuals, such as brothers, sisters, or cousins. It is an extreme form of positive assortative mating since close relatives usually are genetically more similar than are unrelated people who share a few traits. When siblings mate together, it is in effect positive assortative mating for many genetic traits. Half of the alleles of brothers and sisters are likely to be shared. If they mate together, their children would be expected to have a quarter of those alleles in common. Therefore, when consanguineous mating occurs, the result is significantly less genetic diversity among the descendants than if the parents had mated with someone who was not closely related but was like them in terms of selected traits such as skin color or stature.
It has long been assumed by the general public in western nations that children of inbred parents inevitably have a high probability of inheriting mental retardation and other serious genetic defects. This is not necessarily true. If a harmful allele is present in a family, it will show up at a higher than normal rate among inbred children. If inbreeding continues to be the common mating pattern in a family line, it is likely that homozygosity will increase in frequency and the family will experience a progressive rise in the genetic load of the deleterious allele. On the other hand, if the allele is not present in the family line, inbred offspring are not likely to have a higher than normal risk of inheriting a mutation for it. Inbreeding also could potentially increase the odds of a child inheriting desirable traits. If the family genetic line has alleles that contribute to advantageous characteristics, such as intelligence, health, or what their culture defines as beauty, they are more likely to show up in children resulting from inbreeding if the parents have these characteristics. Consanguineous mating also may be an advantage for women who are Rh negative because it would increase the chances that their children would be Rh negative. As a consequence, there would be a lower risk of erythroblastosis fetalis in the children. You will learn more about this potentially fatal condition and its connection with Rh blood types in the next tutorial of this series.
NOTE: The word "consanguineous" comes from Latin and literally means "with the blood". In other words, consanguineous mating is between "blood relatives". The term "blood relatives" goes back to the time when people mistakenly thought that what passed between parents and their children was blood. In fact, it is DNA rather than blood.
The closer two mates are in generational distance from their common ancestor, the greater the likelihood of positive assortative effects on the genomes of their children. Based on statistical data from 38 populations in South Asia, Africa, Europe, and South America, it has been determined that the increased risk for “significant birth defects” among the offspring of first cousins is only 1.7-2.8% above the risk for the general population. The predicted risk for the children of brothers and sisters or parents and their children is 6.8-11.2% above that of the general population. Based on these numbers, it would seem that while the risk is high for very close biological relatives, it is relatively low for first cousins and more distant kinsmen. In fact, there is a high probability that the children of first cousins will not have significant birth defects. In addition, there does not appear to be a statistically significant increase in the frequency of gross chromosomal anomalies, such as trisomy-21 (Down syndrome), in the children of consanguineous unions.
While first cousin marriages are extremely rare in North America, Europe, and East Asia, they are very common in some parts of the world due to long existing cultural traditions. Roughly a third of the marriages in rural India are between first cousins. In the Arabian Peninsula, the rate is 50% or higher. In both areas, there are ongoing nationwide educational campaigns to discourage first cousin marriages. The hope is that if they are successful in reducing the number of such unions, it will cut medical costs for the nations. So far there has been some success in changing the marriage patterns of more educated urban Saudis and Indians, but there have been little inroads into rural areas where most first cousin marriages occur.
A recent statistical study of 165 years of genealogies for 160,000 couples in Iceland has shown somewhat surprising results. Married couples who were third cousins (they shared a great-great-grandparent) had more offspring than did couples who were less closely related. In other words, marrying third cousins resulted in greater reproductive success. However, couples who were first or second cousins had fewer offspring and those children died at a younger age.
Is Love in Our DNA--how do you choose a mate? (Take a poll to find out.)
This link takes you to an external website. To return here, you must click the
"back" button on your browser program.
We have seen in this and previous sections of the tutorial that evolution can result from any of four main processes operating independently or together. In addition, there are two ancillary contributing processes.
Main processes Ancillary processes 1. mutation 1. recombination 2. genetic drift 2. non-random mating 3. natural selection 4. gene flow
Mutation is the ultimate source of new genetic varieties. However, gene flow can be responsible for the introduction of new alleles into a population and very rarely into a species. Generally the most rapid and dramatic evolution is due to natural selection. Recombination and non-random mating can change the frequencies of genotypes which in turn can be selected for or against by nature. Genetic drift can also result in rapid evolution of the gene pools of small, reproductively isolated populations. It is highly likely that our ancestors lived in such small populations for 99+% of the last 2.5 million years during which our genus Homo was evolving. Subsequently, genetic drift and other small population size effects must have frequently been a major factor in our evolution along with natural selection.
Primary source for human birth defect statistics related to inbreeding: Bennett, R. L. et al., "Genetic Counseling and Screening of Consanguineous Couples and Their Offspring: Recommendations of the National Society of Genetic Counselors", Journal of Genetic Counseling (2002) 11(2):97-119. Source of data on Icelandic couples: "An Association Between the Kinship and Fertility of Human Couples" Science (2008) 319: 813-816.
Copyright © 1997-2012 by Dennis O'Neil.
All rights reserved.