What is Science?
- A community of scholars who study the nature of the physical world
- A set of methods used to investigate the nature of the physical world
- The sum total of human knowledge about the physical world
- A series of explanations developed to account for why the world is the way it is
- Deduction: inference in which the conclusion necessarily follows from the premises
- Induction: inference about general conclusions
- The hypothetico-deductive method: an approach involving the generation of explicit
predictions that can be tested by making new observations
Hypothesis is the general assumption, prediction is what you specifically think will
happen in the experiment that will lead to the general assumption
- Prediction is used to test hypothesis!
- Fact: accepted as “true” & has been repeatedly confirmed
- Theory: well-substantiated explanation of some aspect of the natural world that can
incorporate facts, laws, inferences & tested hypotheses (explain facts)
- Law: a descriptive generalization about how some aspect of the natural world behaves
under stated circumstances.
- Science is repeatable, testable, based on observation & inference, published primarily in
peer-reviewed journals, & always subject to modification.
- Neo-Lamarckism: evolution is the coordinated addition of developmental stages in a
species, not natural selection
- Some new additions caused certain organs to be used or disused, which
instigated a Lamarckian mechanism of change.
- Mutationism: new species are formed all at once with no intermediates
- Orthogenesis: internal forces push species along evolutionary trajectories
- Saltationisn: natural selection is important only in producing minor, within-species
- Evolution accomplished by macromutations (rearrangement of chromosomes)
- “The Modern Synthesis”: Darwinians attempted to synthesize Darwinian natural
selection & Mendelian genetics.
- Hardy-Weinburg Equillibrium Principle: what happens to allele frequencies under
idealized conditions? p + 2pq + q = 1
The Basis of Evolutionary Thought
- The scientific community overwhelmingly agrees that evolution has occurred (it is a
- Scientists are not part of a vast global conspiracy with a particular social agenda.
Individual scientists may have all manner of different political, religious & social views. - Scientists are not stupid
Therefore the evidence in support of evolution as a historical reality must be strongly
convincing to scientists for a good reason.
- FACT: Species are related by descent/have common ancestry. Fossil record,
biogeography, morphology, DNA, etc. (this change exists)
- THEORY: The mechanisms that explain how this change occurs. EX. Darwinian
natural selection (how change occurs)
- PATH: The actual historical pathway followed during the course of evolution on Earth
(the actual path historically)
- Good point of evidence for today, less for Darwin
- Darwin: mainly evidence for extinction & the vertical order & geographic distribution
More extensive than Darwin thought
- Ex. fossil of bird with teeth and feathers
- OR an intermediate feature
- Fossil species that exhibit some traits found only in the ancestor and some only in the
descendant, or possessing a feature that is intermediate in form between its ancestors &
- It does not need to be a direct ancestor to be informative!
Imagine digging up a cemetery in England from the 14 century: very unlikely
to find one of your direct ancestors, but this would still tell you a lot about the
people from that time.
- Fossil Gap: find rocks of the correct age & proper environment; then look for fossils
- Ex. the evolution of tetrapod limbs (started as fins, probably used to graze ocean floor)
- Species are distributed in clear patterns geographically & not just by habitat.
- Major patterns are known to correspond to Earth history
- Distribution of life reflects geographic history (continental drift)
- Human, cat & bat all derived from same limb structure
- Avatisms: the reappearance of ancestral features in modern individuals
- Ex. embryos of chicken mutants have been found to grow crocodile-like teeth – thus
birds maintain the genes for making reptilian teeth (i.e. descended from an animal with
- Vestigial structures: no longer necessary – does not mean dysfunctional
- Tail bone, appendix, ear muscles, goose bumps, wisdom teeth - Suboptimality due to history: backpain due to vertical spine, retina inverted – has blind
spot & can detach, crossover of trachea & esophagus – choking, difficult childbirth due to
large heads & upright hips
- Ex. whales have vestigial structures – hip bones, non-functional genes for smell,
remnants of muscles to move non-existent ears
- Pakicetus is oldest known cetacean: mostly terrestrial, linked to whales due to
- Ambulocetus (walking whale): amphibious, hind legs better suited to swimming
All genetic analyses agree that all modern whales are most closely related to a
certain group of terrestrial mammals, rather than other aquatic vertebrates.
Closest living relative is the hippo!
- Most closely related living group (sister taxon) does not equal ancestor!
- Debates about the specific path or mechanism of evolution do not change
the fact of evolution.
EVIDENCE FOR EVOLUTION
- Comes from many independent sources
- Has met countless hypothetico-deductive predictions
- Is abundant, even for supposedly “difficult” groups
- Is so extensive that scientists don’t spend time trying to establish the fact anymore,
instead they focus on explaining the mechanisms by which it ocurs (theory) & on
reconstructing its historical details (path)
- Phylogeny: an evolutionary tree (phylogenetics)
- Shows the path of evolution
- Polytomy: multibranching/unresolved node
- Topology: branching pattern
- Cladogram: topology only, branch lengths have no meaning
- Phylogram: branch length signify amount of divergence or time
There may be a “true” phylogeny – ie. An actual historical set of relationships – but
any reconstructed phylogenetic tree is a hypothesis about relationships & patterns of
Cladistics & “Natural” Classification
- A “natural” classification system is not based on superficial similarity, but on
- According to cladistics, only monophyletic groups (“clades”) should receive taxonomic
- Monophyletic group: A group composed of a collection of organisms, including the
most recent common ancestor of all those organisms & all the descendants of the
The basis of a natural classification system - Paraphyletic group: A group composed of a collection of organisms, including the most
recent common ancestor of all those organisms, but omits some of the descendants of the
most recent common ancestor.
Results when some descendants appear very different.
- Polyphyletic group: A group composed of a collection of organisms in which the most
recent ancestor of all the organisms is not included, usually because the common ancestor
lacks the characteristics of the group.
Puts distant relatives together to the exclusion of closer relatives. Can result
from convergent evolution.
- The lineages of all living species have been evolving for exactly the same amount of
time because they all go back to a shared ancestor.
- Terminal nodes usually depict MODERN taxa; none are “ancestral” to others
- A long branch does not imply that no change has occurred: BEWARE INCOMPLETE
- “Ancestral” & “derived” refer to TRAITS, not species. Every species is a mixture of
ancestral & derived traits.
Intro to Microevolution & Population Genetics
- Microevolution: small-scale processes operating within populations to change allele
- Macroevolution: large-scale patterns of change above the species level, including the
origin of new species.
- Extrapolationists believe that macroevolution is just microevolution extrapolated over
long periods of time.
- It is the prevalent view since microsynthesis
- Usually those who study population-level processes
- Often define “evolution” as nothing more than “a change in allele frequencies”
- Macroevolutionists believe there is more to macroevolution that just microevolutionary
- Those who study the big patterns (e.g. paleotologists, genome biologists)
- “Multi-level” or “hierarchical selection”: natural selection operates at multiple
levels, including among organisms in populations, but also within genomes,
among groups & even among species.
Mendel’s Laws of Inheritance
- First law – segregation: (1) Alternative versions of genes (alleles) account for variations
in inherited characters. (2) A diploid organism inherits two alleles, one from each parent.
(3) “Dominant” alleles are expressed while “recessive” alleles remain unexpressed. (4)
The two alleles for each character segregate during gamete production.
A |1 = 20% A + 50%1A 2
- Second law – independent assortment: During gamete formation the segregation of the
alleles at one locus is independent of the segregation of the alleles of another locus
A |1 + 2 |B 1 22% A |B + 251 A1|B + 25% A2|B1+ 25% A |B1 2 2 2
- NOTE: as laws, these apply to specific conditions (i.e. diploids, genes not linked, no
partial inheritance, etc.) Some Definitions
- Mendelian genetics: allows predictions about what happens to alleles within families or
in a single cross
Add natural selection & genetic drift (modern synthesis) = population genetics:
allows predictions about what happens to alleles in entire populations across
- Population: for sexual species, a group of interbreeding individuals & their offspring
- Locus: the physical location of a gene on a chromosome
- Alleles: alternate (i.e. different & mutually exclusive) forms of a gene
- Gene pool: the set of all copies of all alleles in a population that could potentially be
contributed by the members of one generation to the members of the next generation.
- Frequency: the proportional representation of a phenotype, genotype, gamete, or allele
in a population (e.g. 6 out of 10 have blue eyes = 60% = frequency of 0.6)
- Genotype: the set of genes possessed by an organism; the combination of alleles at a
- Phenotype: the physical expression of the genotype (in combination with environment).
- Homozygote: individual with the same 2 alleles at a given locus
- Heterozygote: individual with 2 different alleles at a given locus
- Mendel’s work was rediscovered in 1900 & it was suggested that “genes” were discrete
& particulate, meaning they did not vary smoothly & they did not blend
- Some Mendelians used this to argue against gradual natural selection &
- However those who measured variation in physical traits in populations argued
that phenotypic variation is continuous.
- Yule opposed Mendelism
- Punnett was a hardcore Mendelian & knew Yule was wrong but couldn’t pinpoint it
The Hardy-Weinberg Equillibrium Principle
- Is based on one locus with two alleles
- Assumes an “idealized” population that:
- Is infinite (or at least very large)
- Is diploid
- Reproduces sexually
- Has non-overlapping generations
- Mates randomly
- Experiences NO mutation, selection or migration
(These are intentional oversimplifications)
In other words, the issue is what happens to existing alleles when: no new alleles arise
by mutation, neither allele is selectively advantageous, no new alleles are migrating in or
out, no alleles 2re lost by 2hance, & alleles are being mixed randomly.
As long as p + 2pq + q = 1, any p & q starting frequencies will stay in equilibrium
- CONCLUSION #1
a) If the idealized conditions are met, then the allele frequencies will not
change from generation to generation. b) Equilibrium will be reached after just one generation of random mating
under the idealized conditions, regardless of starting frequencies.
i.e. initial frequencies & dominance don’t matter – allele frequencies
won’t change unless something changes them.
- CONCLUSION #2
- Under the idealized conditions, if allele frequencies are given by p & q,
then genotype frequencies can be calculated simply by multiplying allele
- Alleles (p : q) Genotypes (p : 2pq : q )
- SO… this allows you to calculate the frequency of alleles (including recessive ones) if
the population is in equillibrium & genotype frequencies are known
(Like a punnett square for the entire population)
- Can also be used to find a deviation from the equillibrium, showing that something
interesting is happening (i.e. microevolution!)
Mutation: the source of genetic variation
Pre-molecular Views of Heredity
Lamarck: Organisms Weismann: Germline &
change in response to somatic lines are separate.
environmental needs; use of VS. Only changes to copies of
an organ accentuates it & genes found in cells of the
the change is passed on germline are passed on.
Darwin: “Pangenesis”: Mendel: Genes are
blending of “gemmules” in “particulate” & are inherited
the blood, offspring inherit a VS. in discrete units. No
mix of traits blending. - In early 1900’s gene was a vague term
- Miesher stumbled across a phosphorous-rich substance in nuclei which “cannot belong
among any of the protein substances known hitherto”.
- Named it “nuclein”
- Thought it was a storehouse for phosphorus
- Termed chromatin (DNA + histone proteins) after various chemical stains
- Termed chromosome shortly after to describe stainable threads in the nucleus
- Renamed nucleic acid (from nuclein) after protein-free samples were found
- Chemical components were discovered – renamed desoxyribose nucleic acid (& later
deoxyribonucleic acid) in reference to sugar component.
- 40s-50s were characterized by research in role of DNA & structure
- It required demonstration with several independent lines of evidence before scientists
were fully convinced that DNA, & not protein, was the molecule of heredity.
- DNA but not protein is involved in the “transformation” of bacteria from non-
virulent into virulent strains
- DNA but not protein amount is constant within species
- DNA but not protein is used by viruses to alter host cells
- Hershey-Chase blender experiment
- 1) DNA contains a 1:1 ratio of purine (A,G) & pyrimadine (T,C) bases
- 2) Base pair composition varies among species
- IN CONLCUSION… it took decades of research to infer the structure & function of
DNA & overall much of our current knowledge has come fairly recently (1940s+)
DNA RNA protein DNA (repeat)
- But proteins also affect the expression of genes & new genes can arise from
RNA (e.g. retroviruses, transposable elements)
- Errors in the genetic system
- Occur because the genetic system & its quality control & correction mechanisms are not
- Mutations are the source of new genetic variation, and without them evolution would be
- HOWEVER, mutations occur without regard for any consequences; good or bad
- Point mutations (new alleles)
- Transition: purine to purine/pyrimadine to pyrimadine – more common
- Transversion: purine to pyrimadine or vice versa
- Synonymous substitutions (silent): no change in the amino acid
- Non-synonymous substitutions (replacement): changes the amino acid
- Frame shift: changes all consecutive amino acids
- Mutation to stop codon ends protein sequence early
- Gene duplications (new genes) - Occur by unequal crossing over (can result in deletion)
- Can lead to NEW GENES
- Chromosomal mutations (new gene order)
- Mostly change the order of genes & their relative location with respect to
- Inversions: part of chromosome breaks off & inverts (flips)
- Translocations: a piece of one chromosome breaks off & joins another (can be
reciprocal or non-reciprocal)
- Inversions, breakages & fusions change gene order rather than generating new
alleles or genes.
- Chromosome number is a very flexible character, with lots of breakages, fusions
& even duplications common in evolution.
- Related species can have quite different chromosome numbers. However, they
are likely to show evidence of relatedness in terms of banding patterns.
- Polyploidy (genome duplications)
- Can occur by hybridiation or errors of meiosis/germline mitosis
- Adds a second copy of the entire genome & thus every gene
- Very common in plants
- Can instantly create a new species
- The effects of mutated genes may occur in any part of the body, but a mutation is only
relevant in evolution if it occurs in a copy of the gene found in the germline
- Mutation is NOT a very powerful force of evolution by itself
p n p e 0
u = mutation rate (mutations per genome per generation)
n = population size?
P n frequency of p at population size n
P = frequency of p at population of start
0 - Variation in mutation rates:
- Sexual vs. asexual reproduction
- Short vs. long generation time
- Exposure to mutagens
- Differences in repair efficiency
- Other properties of gene/genome
- Males contribute more muations to the human gene pool than females because sperm go
through >400 divisions compared to approx. 24 divisions in female eggs.
Therefore there are many more opportunities for errors in replication
- Small random genetic changes may or may not be beneficial to organisms but large
ones are almost always disasterous
However mutations with large effects can & do occur & may have significant
consequences for macroevolution
- Most mutations are not harmful – most are neutral (most occur in non-coding DNA).
However, of those that affect fitness, most are deleterious
- Even a veryyy low rate of beneficial mutation is sufficient – selection is the important
- Whether a mutation is beneficial or harmful depends on the environment.
Example of beneficial mutations: a single amino acid substitution confers
pesticide resistance in blowflies.
- Artificial selection is performed by humans, whether consciously or unconsciously.
1) Individuals within populations are variable
2) This variability is at least partly heritable
3) Not everyone survives & reproduces, & some individuals are more successful than
4) The differential survival & reproduction of individuals is associated with the heritable
variation among individuals (i.e. it is not random) CONSEQUENCES OF POSTULATES…
- Individuals with traits that promote improved survival and/or reproduction relative to
conspecifics will leave more offspring. & thus, the alleles that enhance the survival &
reproduction of these individuals will be passed on more than alleles that do not.
- Individuals who survive & reproduce most successfully have the highest “fitness”
- If the differences that contribute to higher reproductive success are not heritable, they
are irrelevant to evolution (e.g. if you work out you may be healthier, but this has nothing
to do with evolution)
- Fitness is not circular or ad hoc
- Fitness can be predicted based on traits that would be expected a priori to improve
survival & reproduction in a particular environment
- Fitness can be measured (i.e. as the number of offspring) to test these predictions
Organisms do not evolve, populations do!! (watch simulations)
- In real populations, both survival & reproduction usually play a role in natural selection,
sometimes in the same direction, sometimes in opposite directions (e.g. an allele
increases reproduction but shortens lifespan).
- What matters is net reproductive success from one generation to the next. ANTIBIOTIC RESISTANCE
When treated with antibiotics, most of a population is killed. Those who survive happen
to contain genes that confer resistance, & will reproduce. Over time, these resistant
alleles will become more common in the population & if they become the only remaining
allele, they are said to be resistant to the antibiotic.
Consequences of Selection
- Natural selection can remove unfit alleles or increase the frequency (or even “fix”) new
mutations that increase fitness.
Selection & Hardy-Weinberg
- Natural selection violates one of the conditions for Hardy-Weinberg equillibrium, so…
- Natural selection can make it impossible to calculate genotype frequencies just
by multiplying allele frequencies.
- Natural selection can drive genotype frequencies away from the values predicted
under Hardy-Weinberg (BUT one generation of random mating without selection
will restore H-W equillibrium)
- Unfit alleles that are selected against can still persist in the population if they are
generated often enough by mutation.
- The frequency of a deleterious recessive allele at equillibrium is given by:
[ Square root of mutation rate over selection coefficient (root of u / s) ]
Types of Natural Selection
- Different types of selection can be defined based on which phenotypes (& thus
genotypes) are selected in the population. This determines the long-term outcome.
- Selection against both extremes (or for intermediate phenotypes)
- Individuals with intermediate traits have the highest fitness for the environment
- Does not alter the population average (only the variance)
- Selection for both extremes (or against intermediate phenotypes)
- Generates a bimodal distribution
- May lead to 2 divergent populations
- Selection against one extreme or the other
- Moves population average in one direction. If new phenotypes arise by mutation or
recombination, the population may continue to be pushed in that direction.
- If removing deleterious allele = “purifying selection”
- Favours the maintenance of more than one allele in the population (polymorphism) - Can occur by heterozygote advantage (overdominance) or frequency-dependent
- Occurs when the fitness of an allele depends on its abundance
- Can be positive (fitness increases with abundance) or negative (fitness decreases with
- Individual organisms under selection do not change. They either live or die & reproduce
or fail to reproduce.
- Natural selection acts on organisms, but the consequences occur in populations.
- Natural selection is based on phenotypes, but what matters is the resulting change in
allele frequencies in populations.
- Not all or nothing. Even a slight reproductive advantage can have major effects on allele
frequencies over many generations.
- The effects of natural selection accumulate gradually from one generation to the next
- Natural selection does not create new variation (mutation does) – most forms of
selection deplete variation
- Populations adapt to the conditions of the past, not the future.
- No individuals change when a population evolves, only the proportions of different
alleles or traits do.
- Difference in reproductive output alone can lead to natural selection. Genetic Drift
- For alleles that are neutral & not subject to selection, there can be microevolution by
chance via genetic drift.
i.e. random changes in allele frequencies occur by genetic drift
- Genetic drift represents sampling error, in which only an unrepresentative subset of the
population passes on alleles
- This subsampling is unrelated to fitness, and so this is due to chance & NOT
- Unlike natural selection, it is random.
- Direction of drift each generation is unpredictable
- Any previous changes in allele frequency are irrelevant to predicting future
changes by drift (like a coin toss)
- Random sampling error is caused by genetic drift. Sampling error can be caused by:
1) Population bottlenecks
- A population experiences a drastic reduction in size
- Can be due simply to dumb luck
- EX. Northern elephant seals hunted down to about 20 individuals; today
there are 30000 all descended from those lucky 20 – therefore NO
2) Founder effects
- A subset of the original population founds a new population
3) Gamete sampling error
- An unrepresentative subsample of gametes mixes to form zygotes
- The more zygotes that are picked, the closer the allele frequencies are to
the expected frequency.
- Fewer samples means more sampling error!
- Drift can change allele frequencies in populations (i.e. cause microevolution), but it
does NOT lead to adaptation.
How much change can result purely by chance?
- Drift can cause allele frequencies to fluctuate in populations, HOWEVER, if no other
forces are acting, eventually a neutral allele will either become fixed (frequency of 1.0) or
will be lost (frequency of 0) purely by chance.
- Even deleterious alleles can become fixed if drift is strong enough
- “Drunkard’s walk”: stumbling next to train tracks – will eventually either
stumble into safety (fixed) or onto the tracks (loss).
- At any given time, the probability of a neutral allele becoming fixed by chance equals
its current frequency
- So if p = 0.6 right now, then there is currently a 60% chance that it (vs.
alternative) is the one that will become fixed.
- If an allele has just arisen by mutation, its frequency (& likelihood of reaching fixation)
will depend on population size (N): 1/(2N)
- For an allele that is already present in the population, simply multiply this by the
number of copies (x) of the allele: x/(2N) - The average time to fixation of a newly arisen neutral allele that does become fixed is
So drift is much more influential in small populations whereas selection is stronger in
Effective popuation size
- What if everyone in the population doesn’t mate? Then the effective population size
will not be the same as the absolute population size.
- EPS is the number of individuals who are actually contributing alleles to the next
N =e(4N N)m/ fN + Nm f
- There are 246 people, 2/3 are females
- Therefore there are 82 males, so there would be 82 matings assuming each individual
mates only once. Therefore 82 females won’t get to mate.
The effective population size would be 164
- If all females mate once & each male mates twice = 219
- If 164 (all) females mated with 1 male, Newould be 3.975
- 1 male & 100 females = 3.960
- 1 male & 10,000 females = 3.9996
- 1 male & 3,500,000,000 females = 3.99999999
Drift among populations
- Drift reduces heterozygosity within populations, but increases variation among
- The proportion of populations expected to become expected to become fixed for a given
allele by drift is equal to the initial frequency of that allele (e.g. 0.75 initial frequency,
then 75% of the populations should become fixed for that allele)
- The overall average frequency of the allele among all populations does not change, but
the frequency of heterozygotes goes to zero as different populations all become
homozygous for one or the other allele.
Is drift im