Lecture 16: Mendelian Genetics IV: Sex Linkage breaks 2nd law -Evolution (Revisited)pdf _download pdf _download

Extensions of Mendelian Genetics...


Mendel's 2nd Law DOES NOT ALWAYS APPLY to X or Y -linked chromosomal inheritance!!!

Remember, that ......to equate Mendel with meiosis we had to invoke the role of "chiasmata" an their occurance between two gene loci on adjacent chromatids in paired, "bivalent" chromosomes during prophase I of meiosis.  

We now knowthat this CANNOT occur for X and Y chromosome... OK, but what about ALL the genes on the other autosomal chromosomes. Again, I glibbly stated that, if the probability of a single chiasma forming between two genes is "one" (i.e. a certainty), then the assortment of alleles would, in essence, be the same as if they were on separate chromosomes (i.e. random). 


But, what happens if there are no chiasmata formations?

Then the genes on the same chromosome cannot assort randomly, which would be in marked contrast to Mendelian expectations.  

However, this phenomenon can and does happen.

Such aberrations were first hypothesized to exist by Bateson and Punnett, who observed some curious "asssociations" of heritable traits.  However, it was really verified by an American geneticist, Morgan (1909) who was working on chromosomally-linked genes that resided on the X chromosome of the fruit fly Drosophila melanogaster.

So, what happens if there are NO CHIASMATA formations between homologsous chromosomes once they come together in .... phase of Meiosis?



Morgan explained this apparant anomally by proposing that the two loci were present and "linked" on the same chromosome, and that any variation from the parental "linkage" must have occurred through the exchange of genetic material by some type of "crossover events" occurring between these two loci.....

While not common, this "linkage" phenomenon can also happen -even with some genes on the autosomal chromosomes.

Again, looking back to previous lectures, I glibbly stated that, if the probability of a chiasma forming between two genes is "one" (i.e. a certainty), then the assortment of gene pairs on the same chromosome would (in essence) be the same as if they were on separate chromosomes (i.e. randomly assorted -as predicted by Mendel). 


Such aberrations were first hypothesized to exist by Bateson and Punnett, who observed some curious "asssociations" of heritable traits.  However, it was really verified by an American geneticist, Morgan (1909) who was working on chromosomally-linked genes that resided on the X chromosome of the fruit fly Drosophila melanogaster.

To me it is a curious irony that the vindication of Mendel/Sutton-Boveri's ideas that merged Mendelian genetics with cytological analysis of chromosomes, came about through research on sex-linked genes (that don't obey mendel's 2nd law) because (remember) Mendel's early experiments, were carefully chosen to be sex independent, i.e. he only worked with autosomal genetic traits in a monoecious pea plant

Thus, Morgan (1909) who was working on chromosomally-linked genes that resided on the X chromosome of the fruit fly Drosophila melanogaster also able to confirm that crossing over didn't ALWAYS occur in X chromosomal traits.... if the gene pairs were SO close together that crossing over COULD NOT PHYSICALLY HAPPEN.

Morgan went one stage further and suggested that the frequency of such cross-over events (occurring between two genes that were physically very close together) was actually a function of the genetic distance between the two loci.

He thus defined the unit of genetic distance as being:

one crossover event/100 products of meiosis = one map unit or 1 centiMorgan (cM).

Subsequently, he and his student demonstrated that such linkage could be used to map the relative position of genes that are linked on the chromosome, as they determined that the probability of crossing over between two loci appeared to be additive, which lead to the first "mapping of genes" on specific chromosomes.


So, now we have analyzed two "variations" from the "predictable" Mendelian-type of inheritance,

(a) variations that arise as a consequence of "extensions" to Mendelian genetics, where the function of the genes in question may interact to give different F2 phenotypes.

(b) variations that arise because of "chromosomal linkage" (thus defying Mendel's Second law).

There is a third form of non-Mendelian genetics.....

(c) Cytoplasmic / Maternal Inheritance


Essentially, Mendelian genetics is the genetics of the nucleus, but other cytoplasmic organelles can also carry genetic material.

Mitochondria, chloroplasts, and other plastids possess a small amount of DNA. 

Humans have 20,000 - 25,000 genes in the nucleus, and 37 genes in their mitochondria.

Plastid genomes in plants are five times larger than those of mitochondria.

Thus, any true definition of an organism's genome must include the total configuration of genetic material.

Note: Mitochondria and plastids are passed on by the mother only, as the egg contains abundant cytoplasm and organelles. The mitochondria in sperm do not take part in gamete union.

The hunt for the Mitochondrial Eve??

Is there a male equivalent to this purely female based inheritance?

Well yes, but it's not cytoplasmic......... it is sex-linked... on the Y chromosome

So, Mendel was not always right for a number of reasons.


A haplotype is a set of DNA variations, or polymorphisms, that tend to be inherited together. A haplotype can refer to a combination of alleles or to a set of single mutations or multiple nucleotide polymorphisms (SNPs)/mutations that are found on the same chromosome.

Are the two types of location compatible?


Having now fully understood (?) all the major types of variations from the "normal" Mendedlian type genetic patterns of inheritance how might our newly found insight into Mendelian genetics have an impact upon our appreciation of evolution?

Remember, while "....populations evolve and individuals do not", the gene pool of a population is the summation of all the individual genomes within that population.

To rephrase the question, therefore, How might an understanding of Mendelian genetics (which addresses phenotypic expression of an individual's genes) allow us to understand phenotypic/genetic changes within a population?

Genetic Variation within Populations

To recap (in light of the last few lectures):

For a population to evolve, its members must possess variation, which is the raw material on which "agents" or "forces" of evolution act (genetic variation within a gene pool).

We observe phenotypes in nature: i.e. the physical expressions of genes.

A heritable trait, however, is a genetic characteristic of an organism that is mainly influenced by the organism's genes (we cannot forget totally the influence of environment on this expression.

The genetic component that governs a given trait is called its genotype.
A population evolves when individuals with different genotypes survive or reproduce at different rates.

Genes have different forms called alleles.

A single individual has only some of the alleles found in a population.

The sum of all the alleles in a population is its gene pool, which contains the variation (different alleles) that produce the differing phenotypes, upon which change can come about...evolution.

Most populations are genetically variable.

Natural populations possess inherent genetic variation.

The reproductive contribution of a genotype or phenotype to subsequent generations relative to the contribution of other genotypes or phenotypes in the same population is called fitness.

This "fitness" of any particular genotype is determined by the average rates of survival and reproduction of individuals within that population with that particular genotype;

i.e. the relative reproductuctive contribution of a given genotype.

For example, Man's highly selective preferences for certain edible crops have placed a a selective pressure on the crops that have been and are produced, giving rise to a seemingly wide variety ofimportant crop plants.

Artificial selection in laboratories that have analyzed genetic variation in assorted laboratory organisms, such as Drosophila melanogaster have also revealed genetic variation in these fruit flies.

In Ecological terms a locally interbreeding group within a geographic population is called a Mendelian population.


The relative proportions, or frequencies, of all alleles in this population are a measure of its population's genetic variation.

Population Geneticists can estimate such allele frequencies for any given locus by measuring the numbers of alleles in a sample of individuals from within a population.

Such measurements can be seen in terms of probability, which can range from 0 to 1, wherein the sum of all allele frequencies at any given locus = 1.

The "Allelic frequency" (p) for any given trait can be calculated by dividing the number of copies of that particular allele in a population by the sum of alleles in the population.

According to Mendelian genetics; If only two alleles (A and a) are present for a given locus, and are found among the members of a diploid population, they may combine to form three different genotypes: AA, Aa, and aa.


The Allelic frequencies can be calculated using simple mathematics with the following variables.

  • n AA = the number of individuals that are homozygous for the A allele (AA).
  • n Aa = the number of individuals that are heterozygous (Aa).
  • n aa = the number of individuals that are homozygous for the a allele (aa).
  • Note that n AA + n Aa + n aa must always = N, the total number of individuals within the population.

    Now let's look at it from the perspective of each alle at a given locus,

    For the sake of discussion, let

    p = the frequency of allele A.
    q = the frequency of allele a.


    It is possible that the numbers of homozygous dominant and heterozygotes and homozygous recessives can change without changing the probability of finding individual alleles (p's and q's).

    What about an example of how such equations can be used to calculate the frequencies of allele "A" and "a" within a population.

    In essence, a population that is not changing genetically is said to be at Hardy–Weinberg equilibrium; in that the allelic and genotypic frequencies within a population that has reached this state do not change from generation to generation.

    To appreciate the importance of the HW equilibrium, Five essential assumptions about the population must be made.

  • Mating is random.
  • Population size is very large.
  • There is no migration between populations.
  • Mutation can be ignored.
  • Natural selection does not affect the alleles under consideration.
  • IF the above conditions of the Hardy–Weinberg equilibrium are met, two consequences

    must follow.

    (a) The frequencies of alleles at a given locus will remain constant from generation to generation.

    (b) After one generation of random mating, the genotypic frequencies will not change.

    Restating the second result in the form of an equation that takes into account the allelic frequencies, produces the Hardy–Weinberg equation:

    p2 + 2pq + q2 = 1.

    An example of the Hardy–Weinberg equation in use, and how it is derived from Mendelian first principles.

    Note that this example also shows that there are two ways to produce a heterozygote, hence the overall probability for obtaining a heterozygote is "2pq", not just "pq".

    So, why is the Hardy–Weinberg equilibrium important?


    Because it suggests that allelic frequencies remain the same from generation to generation unless some agent acts to change them.


    Thus, dominant alleles would not necessarily "dominate" the presence of the recessive allele unless either one had an effect upon any of the five criterea necessary to appreciate the HW equilibrium.

    It also illustrates the distribution of genotypes that can be expected (for a given population) at "genetic equilibrium" for any value p or q (which can be determined empirically).

    But the conditions for any maintenance of the equilibrium are far too stringent for any given natural population, so...

    is it relevant to the real world?

    Yes. It is the equivalent of a "null hypothesis" to the scientific method. ie it is the condition where "nothing happens".

    As such, it allows scientists to determine (a) whether evolutionary agents are operating and (b) the identity of the agents that might be operating to change the pattern away from the equilibrium.... Erin Brokovitch story

    Microevolution: Changes in the Genetic Structure of Populations

    "Evolutionary agents" cause changes in the allelic and genotypic frequencies within a population.
    Since the changes in the gene pool of a population constitute small-scale evolutionary changes, they are referred to as microevolution.

    We have discussed -to varying degrees- that some of the known evolutionary agents are mutation, gene flow, random genetic drift, non-random mating, and natural selection.

    To a limited extent, we have also discussed, mutations, which are changes in the genetic material that can either be deleterious or beneficial to the function of an allele/gene.

    While most of these mutations appear to be random (e.g. copying errors, as the DNA is synthesized), and are normally either harmful or "neutral" (i.e. they do not affect their bearers ability to survive, and or procreate), some mutations are actually beneficial. Whatever the direct consequence they provide for a heterogenous population.

    Indeed, the origin of all genetic variation is heterogeneity in the germ-line cells (why do we not care about somatic cell variations?)

    Even though mutations are sufficient to create considerable genetic variation mutation rates, are relatively low; approximately one mutation per million loci is a typical frequency.

    Thus, even though the very presence of mutations within a population means that one of the principle conditions that are necessary for the Hardy–Weinberg equlibrium to exist can never be met, the rate at which mutations arise at any particular locus is so low that the consequences of any neutral mutations would result in only very small deviations from Hardy–Weinberg expectations.

    If large deviations are found within a population, then either the mutation is selective, or it would be appropriate to look for reasons why the mutation rate is so high -i.e. look for evidence of and additional agent that is causing the problem.

    Such analyses have added important insights into how we can view evolutionary changes that have occurred over the ages, and have allowed us to appreciate that evolutionary rates can vary in two ways: slowly through "neutral" mutational changes to a gene pool and/or quite dramatically by some other changes in the assumptions that would normally hold a gene pool in some form of equilibrium.

    Hopefully helping to understand some of the different variables that have helped create evolutionary changes over evolutionary time and how the rates of change can differ for different types of living organisms. For example, it can detect potential "bottlenecks" in population development, as the ferquency of alleles, under this circumstance would be severely reduced.






    Copyright © Department of Biology, Georgia State UniversityView Legal Statement Contact Us


    Quick Facts
    Governance and Strategy
    University Policies
    Contact Georgia State


    Colleges & Schools
    Degrees & Majors
    Academic Guides


    College of Law
    Financial Aid


    Commercial Development


    University Library
    College of Law Library

    Campus Life

    Career Services