Evolution of Quantitative Characters
How does selection act on quantitative traits?
Three modes of selection
Directional Selection
Fitness profile – one tip of the character distribution is at an
advantage
- Fitness profile – Fitness (y) against character value (x)
Consequences for trait’s frequency distribution = shift in mean
value
Imagine a population undergoes a directional selection for high
trait values
- Blue spots show chosen parents and their progeny
1. Slope of line estimates heritability
2. Difference between mean of selected
parents and population mean = S, selection
differential
3. Difference between mean of progeny and
population mean = R, response to selection
Heritability and Regression
If we regress mean offspring on mid-parent values –
why is the gradient an estimate of heritability of
the trait?
Statistics: Regression of a variable y on variable x = covariance x and y/ variance of x
Quantitative genetics: covariance of offspring and mid-parent value = ½ V A, variance of
mid-parent values = ½ VP
1. S is same value as previously
2. Slope of regression is shallower
3. Response to selection is smaller
Breeder’s equation: R = h2S
Can estimate h2 from selection experiment: impose directional selection of known
strength (so S value known), measure the response to selection (R value estimated)
then h2 = R/S
Artificial Selection (form of directional selection)
, Selecting for specific phenotypes can lead to large phenotypic differences from
wild-type
Eg. domesticated breeds of pigeon all stem from wild rock pigeon by selective
breeding – fantail: 32 tail feathers vs, 12 of mere ancestor
Direct and correlated responses
- Direct response: change in the trait under selection
- Indirect or correlated responses, changes in other traits
Eg. sexual selection
Context is female preference for male ornamental traits – AS can reveal
correlations between genes for female preference and genes for male
ornaments
Does artificial selection on male ornament cause a correlated response in female
preference?
3 selection regimes maintained >13 generations, 2 replicates per selection regime
AS on allometric shape of male flies
Long eyespan relative to body size – measure 50 males, choose 10 with longest ES to
be parents
Short eyespan relative to body size – measure 50 males choose 10 with shortest ES
to be parents
Control (unselected – 10 males chosen at
random from 50)
There was a direct response to selection:
Relative ES increased in the long lines and
decreased in the short lines
Is there a correlated response to selection, ie.
is there a change to female preference when
male traits are selected?
- Used choice chambers
- There is a correlated response in preference – unselected
control females showed strong preference for L males
- Females from L lines also preferred L males
- C and L females had similar preference level
- BUT S line females perferred S males – complete reversal
of the usual preference observed in field and laboratory
populations
Directional selection in the field
May be quite common especially in changeable environments
Eg. Selection on bill size in Darwin finch
, In past 3 000 000 years, adaptive radiation of feeding specialisation – cactus
flowers, use twig tools for insects from bark, eggs, leaves, blood, ticks
Key studies on Medium Ground Finch, Geospiza fortis
- Simple island ecology: identify all seed types in finch diet – use bill to crack open
seeds
- Variation in bill thickness some individuals more efficient at digesting hard
seeds
1977 severe drought – drought resistant plants have larger seeds = imposed
directional selection on beak traits in finches
Birds with deeper and stronger beaks could crack
seeds better, smaller died of starvation
Beak traits have heritability of ~0.7
Net effect of selection was an increase in average
beak depth of survivors relative to original
population
- Pattern of directional selection changes
depending on climate changes
- Change due to selection, the S values for
several traits in 1977 + 1982 (drought), 1985
(very wet)
- Direction of S can be reversed
- Suite of traits changed synchronously, all positive or all negative
- Traits move in concert – directional selection can retain genetic variation
(though the AVERAGE bill depth of survivors relative to original population
changed)
Directional selection (naturally occurring) – is often not long sustained in one direction,
with reversals in sign from time to time
- In the long term, fluctuating directional selection may maintain genetic
variability in traits since different character optima exist at different times
- Alleles that increase and alleles that decrease a trait value will have bouts of
favourable selection and can persist
Stabilising Selection
Fitness profile – fitness peak at intermediate values of trait with selection against
the extremes
How does selection act on quantitative traits?
Three modes of selection
Directional Selection
Fitness profile – one tip of the character distribution is at an
advantage
- Fitness profile – Fitness (y) against character value (x)
Consequences for trait’s frequency distribution = shift in mean
value
Imagine a population undergoes a directional selection for high
trait values
- Blue spots show chosen parents and their progeny
1. Slope of line estimates heritability
2. Difference between mean of selected
parents and population mean = S, selection
differential
3. Difference between mean of progeny and
population mean = R, response to selection
Heritability and Regression
If we regress mean offspring on mid-parent values –
why is the gradient an estimate of heritability of
the trait?
Statistics: Regression of a variable y on variable x = covariance x and y/ variance of x
Quantitative genetics: covariance of offspring and mid-parent value = ½ V A, variance of
mid-parent values = ½ VP
1. S is same value as previously
2. Slope of regression is shallower
3. Response to selection is smaller
Breeder’s equation: R = h2S
Can estimate h2 from selection experiment: impose directional selection of known
strength (so S value known), measure the response to selection (R value estimated)
then h2 = R/S
Artificial Selection (form of directional selection)
, Selecting for specific phenotypes can lead to large phenotypic differences from
wild-type
Eg. domesticated breeds of pigeon all stem from wild rock pigeon by selective
breeding – fantail: 32 tail feathers vs, 12 of mere ancestor
Direct and correlated responses
- Direct response: change in the trait under selection
- Indirect or correlated responses, changes in other traits
Eg. sexual selection
Context is female preference for male ornamental traits – AS can reveal
correlations between genes for female preference and genes for male
ornaments
Does artificial selection on male ornament cause a correlated response in female
preference?
3 selection regimes maintained >13 generations, 2 replicates per selection regime
AS on allometric shape of male flies
Long eyespan relative to body size – measure 50 males, choose 10 with longest ES to
be parents
Short eyespan relative to body size – measure 50 males choose 10 with shortest ES
to be parents
Control (unselected – 10 males chosen at
random from 50)
There was a direct response to selection:
Relative ES increased in the long lines and
decreased in the short lines
Is there a correlated response to selection, ie.
is there a change to female preference when
male traits are selected?
- Used choice chambers
- There is a correlated response in preference – unselected
control females showed strong preference for L males
- Females from L lines also preferred L males
- C and L females had similar preference level
- BUT S line females perferred S males – complete reversal
of the usual preference observed in field and laboratory
populations
Directional selection in the field
May be quite common especially in changeable environments
Eg. Selection on bill size in Darwin finch
, In past 3 000 000 years, adaptive radiation of feeding specialisation – cactus
flowers, use twig tools for insects from bark, eggs, leaves, blood, ticks
Key studies on Medium Ground Finch, Geospiza fortis
- Simple island ecology: identify all seed types in finch diet – use bill to crack open
seeds
- Variation in bill thickness some individuals more efficient at digesting hard
seeds
1977 severe drought – drought resistant plants have larger seeds = imposed
directional selection on beak traits in finches
Birds with deeper and stronger beaks could crack
seeds better, smaller died of starvation
Beak traits have heritability of ~0.7
Net effect of selection was an increase in average
beak depth of survivors relative to original
population
- Pattern of directional selection changes
depending on climate changes
- Change due to selection, the S values for
several traits in 1977 + 1982 (drought), 1985
(very wet)
- Direction of S can be reversed
- Suite of traits changed synchronously, all positive or all negative
- Traits move in concert – directional selection can retain genetic variation
(though the AVERAGE bill depth of survivors relative to original population
changed)
Directional selection (naturally occurring) – is often not long sustained in one direction,
with reversals in sign from time to time
- In the long term, fluctuating directional selection may maintain genetic
variability in traits since different character optima exist at different times
- Alleles that increase and alleles that decrease a trait value will have bouts of
favourable selection and can persist
Stabilising Selection
Fitness profile – fitness peak at intermediate values of trait with selection against
the extremes
