Why should Ecologists care about evolution?

Evolution works on a much more rapid time scale than previously thought
Rapid Evolution of Resistance in a Crop Pest
Pink bollworms (Pectinophora gossypiella) are a major pest of cotton. Pink bollworm
caterpillars eat the flowers and seeds of cotton plants, reducing their fiber
production.1 Adding insult to injury, the caterpillars stain any fibers infested plants do
manage to make, diminishing the crop's commercial value.
A single female can lay between 100 and 200 eggs.3 A pink bollworm population can go
through as many as five generations during a single growing season.
Pink bollworm populations exposed to chemical insecticides evolve resistance.
The populations evolved across generations, quickly becoming resistant to the
insecticides' effects.
at least under the right circumstances, populations can evolve quickly. In one generation,
Tabashnik's pink bollworm population (as well as yours) went from about 20% resistant
to 100% resistant. If rapid evolution is common in nature, then it is something ecologists
need to be aware of at all times

More armor plating gives better protection against cutthroat trout,23 but this protection
comes at a cost. Compared to lightly armored sticklebacks, completely armored
individuals grow more slowly, survive the winter at lower rates, and breed later.4 In
addition, because heavily armored sticklebacks grow more slowly, they are more
vulnerable to predatory insects that eat stickleback fry.5
Sticklebacks tend to be more heavily armored in lakes and streams where they are more
vulnerable to attack by predatory fish, and less heavily armored where they are less
vulnerable
During this brief time, the stickleback population shifted from mostly lightly armored
fish to mostly intermediate and heavily armored fish. This is a rapid evolutionary change.
Populations in nature can evolve rapidly enough to change the ecological interactions
between species during the course of a typical ecological study. Ecologists must be aware
of evolution.

Assertion 1: Individuals Vary

The theory of evolution by natural selection's first assertion is that everybody is
different.
Assertion 2: Offspring Resemble Their Parents
The theory of evolution by natural selection's second assertion is that at least some of the
differences among individuals are passed genetically from parents to offspring. That is,
offspring resemble their parents.

Assertion 3: Survival and Reproduction are Selective

The theory of evolution by natural selection's third assertion is that individuals with
certain traits survive and reproduce at higher rates than those with other traits.

Any population that satisfies the three assertions of the theory of evolution by natural
selection will evolve.
Daphne Major lacks a permanent source of fresh water. The scientists bring their own; the
plants and animals that live there rely on the short spring rainy season to sustain them
through the rest of the year.
Medium ground finches harvest the rainy season's crop of seeds, which dwindles as the
arid months drag on. The finches crack the seeds open with their beaks. It is the birds'
beaks that are the focus of our story.
The beaks of Daphne Major's medium ground finches vary considerably in depth.
Much of the finch population's variation in beak depth is transmitted from parents to
offspring. 65% of the variation in beak depth among Daphne Major's medium ground
finches is due to differences in genes.

e finches scratched for survival during the drought, they ate the small, soft seeds first. As
these dwindled, the birds who found the most to eat were the ones that could open large,
hard seeds. Under these circumstances, small differences in beak size became matters of
life and death.
The graph at lower right compares the distribution of beak sizes among the medium
ground finches living on Daphne the year before the drought with the distribution among
the survivors.2 By a small but statistically detectable margin of a bit more than 0.5 mm,
the survivors had larger beaks, on average, than the birds on the island before the
drought. In other words, survival was non-random.

Fitness refers to how evolutionarily successful an individual is within a certain

environment, relative to other individuals. In practice, this is often measured as the
number of offspring that are produced that themselves grow up to produce offspring. But
the theoretical definition includes all future generations that come from the individual, or
more generally, the genetic contribution of this individual to future generations.

An adaptation is a genetically-determined trait that increases an individual's fitnessthat

is, its genetic contribution to future generations. Any trait that improves an individual's
chances of surviving and reproducing, compared to chances of survival/reproduction for
individuals lacking the trait, is an adaptation.
The possession of an adaptation is said to be adaptive.

Natural selection is a systematic difference in the survival and reproductive success of

individuals with different genotypes or phenotypes.
If, for example, snails with thin shells are eaten by crabs while snails with thick shells
survive to reproduce, thick shells are said to be naturally selected.
If the traits that influence survival and reproduction are heritable, then natural selection
can lead to evolution.

Assertion 1: Individuals vary in phenotype. That is, they look, function, and act
differently.
Assertion 2: Differences in phenotype are at least partly due to differences in genotype.
That is, some of the variation among individuals in appearance, function, and behavior
arises because they carry different alleles for some of their genes.
Assertion 3: Some phenotypes, and thus some genotypes, survive and reproduce at
higher rates than others.

A quantitative trait is a character with a continuous distribution (as opposed to discrete

categories). Examples in humans include skin color and running speed. The term comes
from the fact that an individual's phenotype can only be assessed by taking
measurements.
Typically, quantitative traits follow normal distributions (bell curves). There are a few
individuals with extreme phenotypes and many individuals with intermediate phenotypes.
Phenotypes for quantitative traits are typically determined by the combined influence of
genotype and environment.
Discrete traits occur in distinct Categories: Trait is there or it is not (examples: albinism,
cystic fibrosis, Huntingtons disease) Mendelian inheritance, single genes, dominance,
recessiveness
Genetic drift is change in the frequencies of alleles from generation to generation, or
between populations, that happens as result of sampling error. Sampling error is the
random deviation between the frequencies we see in a randomly-chosen subset of a
population versus the population as a whole.
For example, if draw 10 marbles from a bag containing 600 green marbles and 400
orange marbles, we will often, just by chance, end up with something other than 6 green
and 4 orange. Likewise, if we draw eggs and sperm at random from the gene pool to
make the next generation's zygotes, we will often, just by chance, end up with allele and
genotype frequencies somewhat different from those in the previous generation. This is
genetic drift.

gene flow (also known as gene migration) is the transfer of alleles orgenes from one
population to another.
Chapter 2

Handling time (h) is the time needed to process a prey item. It includes the time required
to capture or collect the prey; to clean, shell, or otherwise prepare it; and the time needed
to eat it. While a forager may not complete all food-handling tasks in one bout, an
important assumption of many foraging models is that foragers cannot search for
additional prey while they are handling a current prey item. Therefore, handling time
affects a prey item's profitability.
Prey profitability accounts for both handling time and energy. It is the energy gained per
unit of time spent handling that item:
E/H

The transition between specialist and generalist is predicted to be abrupt and to depend
solely on how easily the more profitable prey item is found. If it is easy enough, the
forager should ignore less profitable items; but if it is hard, less profitable items should be
taken too. Charnov's model predicts that a forager should switch from a specialist to a
generalist when the following condition holds:
S1> E1H2/E2 H1
Game theory is the study of strategic decision-making in contestswith two or more
players. Mathematics and logic are used to predict the rational behavior of individuals
playing the game.
Game theory assumes that all players act in their own self-interest to secure the best
possible outcome given the rules of each game and that the payoff for any strategy
depends on other players' strategies.

E = (PwinV) (PloseI) Display I = injury cost

Individuals with higher scores should produce more offspring and, consequently, their
behavioral phenotypeshould be better represented in subsequent generations.
An ESS is a strategy (i.e., trait or behavior) that, when adopted by all members of a
population, cannot be successfully invaded by an individual with a different strategy. The
ESS concept was originally proposed to identify a behavioral strategy that could not be
bettered by an alternate strategy.
Evolutionarily stable strategies can be either pure or mixed.
In game theory, a pure strategy is one that is not composed of other strategies a player
could adopt. It is a set of behaviors an individual always performs in a given type
of contest.
In game theory, a mixed strategy is composed of a mix of otherstrategies, each of which
is played with some probability. For example, a mixed strategy could be to play hawk
50% of the time and dove 50%.
Dove can never be a pure ESS, because a Hawk can always invade a population
comprised entirely of Doves.
ixed ESSs are quite common in nature and can take two forms.

Each individual always plays a particular strategy and, at equilibrium, there is a

certain frequency of each type of individual.

Each individual in the population plays both strategies, each with certain
frequency.

Intrasexual selection results when one sex directly competes with members of the same
sex for mating opportunities. Intrasexual competition drives the evolution of secondary
sexual characteristics that improve fighting ability (e.g., big size, weapons such as horns).
Because such contests are most common between males, intrasexual selection is often
referred to as male-male competition.
Intersexual selection results when one sex chooses a mate from among members of the
opposite sex. Intersexual selection drives the evolution of secondary sexual
characteristics (ornaments) that make members of the chosen sex (typically males) more
attractive to the choosing sex (typically females). Because females are more commonly
the choosers, intrasexual selection is often simply called female choice.

Good Genes. Showy males offer genes that increase the fitness of a female's
offspring by, for instance, ensuring heterozygosity or conferring advantages such
as disease resistance. This hypothesis assumes ornaments are an honest signal that
a male has superior genes. For example, long tails may be a handicap that only
superior males can maintain while avoiding predation.2

Sexy Sons. Female preference creates an evolutionary feedback loop that selects
for more and more extreme male ornaments. Females who prefer males with
extreme traits (e.g., long tails) produce sons that inherit their father's long tail and
daughters that inherit her preference for long tails. Over time, this runaway
selection process can drive the evolution of ever longer tails, stopping only when
the reproductive benefit of the tail is balanced by the accompanying survival cost.

Sexual selection is a form of evolution by natural selection that acts on traits

affecting mating success.
Biologists distinguish between two forms of sexual selection: (1) competition for
access to mates, typically occurring between males and therefore called malemale competition; and (2) mate choice, typically made by females and therefore
called female choice.
Anisogamy is a form of sexual reproduction in which the zygote is formed from
the union of two gametes of unequal size and unlike form.
By convention, the sex producing the larger gamete is female and the smaller
gamete is male.

Monogamy is a mating system in which a male and female have only a single mating
partner per breeding season.
Polygyny is a mating system in which a male mates with more than one female but each
female mates with only one male during a single breeding season.
Polygynandry describes mating systems where both males and females take multiple
partners during a given breeding season. As a social mating system, polygynandry is rarer
than the other three common mating systems (monogamy, polygyny, and polyandry), but
because many individuals mate outside their bond, genetic polygynandry may be more
common.
Polyandry is a mating system in which a female mates with more than one male but each
male mates with only one female during a single breeding system.

Small Atlantic salmon (Salmo salar L.) take this approach. Some male salmon migrate to
the sea, spend up to 3 years feeding and growing, and then return to the river of their birth
as large, fertile adults. Other males, known as precocious parr, remain in the river and
reach sexual maturity after only about a year, never growing large. Instead of fighting,
these small males "sneak" fertilizations by hiding near females, then racing in and
releasing sperm when the large males are spawningwith the females.
enetic analysis indicates that these sneakers sire about 60% of juvenile fish
As defined by Robert Trivers, parental investment is any investment in an individual
offspring that increases the offspring's chance of surviving and reproducing, at the cost of
the parent's ability to invest in other offspring.
Parents should only invest resources in their current offspring if they can't do better by
investing those resources in their own survival or in future offspring
Sunfish appealed to Neff because, like salmonids, males display two life history
strategies called "parental" and "cuckolder". Parentals are large, territorial males that
compete for nesting sites, guard potential mates and, most importantly, provide care for
offspring. Cuckolders do none of these things. When small, they sneak fertilizations by
darting in while parentals are spawning, similar to small Atlantic salmon males. As they
become bigger they look and behave like females. As a result, they no longer need to
sneakinstead, these female mimics achieve fertilizations by duping a parental male into
thinking it is simultaneously spawning with two females, as can be seen in the image to
the right.
Neff and others hypothesized that a parental male's willingness to provide care to
offspring should depend on how certain he is that he is the father.23 After all, a male gains
no fitness benefits from caring for offspring of other males! This hypothesis has proven
difficult to test because paternity certainty is difficult to experimentally manipulate.
First, parental males cannot determine if any eggs in their nest were fertilized by another
male. If no other males are present during spawning, a parental male can be relatively
confident that he fertilized the eggs, but if another male is present, his paternity certainty
is likely to decline. Second, once eggs hatch, parental males can use water-borne
chemicals released by newly hatched fry (young fish) to determine paternity.1 They can
smell which fry are theirs.
Neff quantified parental care at three points during this second experiment, with the
results shown on the right, in the bottom graph.
1. Baseline care levels before the egg manipulation: At this point, both groups
provided similar levels of care (red bars).
2. The day after the egg manipulation: Both groups increased care after the
manipulation, but there was no significant difference between treatment males,
whose paternity had been reduced, and control males (yellow bars).

3. The day after eggs hatched: Treatment males provided significantly less care to
their fry than control males (blue bars).
This second experiment lends further support to the hypothesis that males adjust parental
care in response to paternity certainty.
Life history patterns play an important role. In mammals, for example,
internal gestation and lactation mean that females provide all of the initial care. As a
result, female-only care is the rule in 91% of mammalian species.1 In contrast, male and
female birds are equally capable of incubating eggs and feeding chicks, and two parents
often are more successful than one. This helps explain why biparental care occurs in 81%
of bird species.2
Of course, for many species, neither parent offers any care. For example, the majority of
fishes, amphibians, and reptiles don't provide any care beyond the nutrition and energy in
eggs. The offspring of these species tend to be precocial, meaning they can fend for
themselves as soon as they hatch.
Cooperation occurs when an actor behaves in a manner that both benefits the recipient
and improves the actor's inclusive fitness.
Cooperation includes acts of mutual benefit as well as those that may appear altruistic but
are actually influenced by reciprocity orkin selection.
The Vampire Bat's Dilemma game is a variation on the Prisoner's Dilemma, in which bats
attempt to maximize their fitness. Each player can adopt one of
two strategies: share ("cooperate") orwithhold ("defect").
A bat playing share will share a blood meal with another bat, giving up some of her own
fitness to help a hungry roostmate.
A bat playing withhold does not share a meal with her roostmate.
The game assumes bats venture out to forage each night. Upon returning, hungry bats beg
meals from well fed bats.
Instead, they play the game every night for years, mostly with the same roostmates (bat
colonies are relatively small and stable).1 As a result, a bat's lifetime fitness depends not
on each individual decision to share or withhold, but rather on their overall pattern of
behavior.

it-for-tat is successful for three reasons:

1. It is nice, cooperating on the first move.
2. It is retaliatory, punishing players that defect.
3. It is forgiving, cooperating once a previously defecting player begins to
cooperate.
Tit-for-tat is a model for how cooperation can evolve.2 If individuals repeatedly interact,
tit-for-tat and other such strategies can have higher fitness than simply defecting all the
time.
reciprocity is favored when (1) the benefit for the recipient is greater than the cost to the
actor; (2) there are frequent opportunities for repayment; and (3) individuals can
recognize each other and remember past behavior.1
Bats given a blood meal gain more time until starvation than donors give up. Female bats
form associations that last for years, ensuring frequent opportunities to repay shared
meals. Finally, bats recognize one another and preferentially share food with bats in their
social network, strongly suggesting a tit-for-tat pattern of sharing. In fact, researchers
have found that the best predictor of the amount of food donated was the amount of food
the donor had previously received from the other bat,
Hamilton's rule states that a social behavior will be favored by selection if:
where r is the coefficient of relatedness, B is the fitness benefit to the recipient, and C is
the fitness cost to the actor.
rB-C>0
Inclusive fitness is the sum of an individual's direct and indirectfitness.
Direct fitness is the fitness an individual realizes through its own reproduction. It is
generally measured as the number of offspring that live to reproductive age.
Indirect fitness is the fitness an individual realizes by helping nondescendent kin. For
example, indirect fitness is bolstered by offering assistance that either improves the
survival of these relatives or that helps them to raise their offspring.

Chapter 3
Variation in life history strategies can be attributed to fundamental physical constraints.
The limited resources available must be divided among all of an organism's biological
needs for survival and reproduction (e.g., maintenance, defense, growth). The need to
allocate limited resources generates trade-offs. For example, energy spent on growth
cannot be spent on producing eggs.
Once energy and nutrients are acquired, whether these resources are allocated to growth,
generating gametes, raising offspring or warding off predators depends to an extent on
the organism's environment.
These researchers found that the mean clutch size was 8.5, but the clutch size that
produced the most surviving chicks was 12.1 This is a somewhat counter-intuitive result.
If Lack's hypothesis was correct, the average clutch size should have the highest survival
rate. Why the discrepancy? Boyce and Perrins hypothesized that individuals lay fewer
eggs than the apparently ideal number as a result of year-to-year variation in food
availability. For example, laying 12 eggs in a good year might be great, but in a sparse
year it could threaten survival of the entire clutch.
R = birth deaths
R = rate of growth
When an age pyramid diagram for such a species depicts a large percentage of prereproductive individuals (i.e., has a wide base), the population is likely to be increasing in
size
In contrast, when a high proportion of individuals are in the older, post-reproductive age
classes (i.e., has an age pyramid with a narrow base), the population is likely to be either
stable or declining.
The number of individuals at the beginning of each age class is denoted in the life table
by nx,
bx, records the age-specific birth rate, which, for x = 0, is the number of births that
occurred for this cohort during the first age classi.e., between the ages of 0 and 10.
This is shown in the Number of births box below the life table. Because children do not
have children, the value is 0. Enter 0 under bx in the top row.

Survivorship, lx, is the fraction of the cohort alive at the beginning of an age interval. To
calculate lx, divide the number of individuals alive at the start of an age class, nx, by the
initial cohort sizei.e., the number of babies born at time zero, n0 (here, 1,000).
Fecundity, denoted mx, is the per-female rate of offspring production for age class x. To
calculate mx, divide the number of births occurring for an age class by the number of
survivors at the beginning of that age class.
R0 (net reproductive rate) = summation of lxmx
f R0 is less than 1.0, each individual that started the cohort has, on average, less than one
offspring before dying. This means the population will decline by that proportion each
generation.
ifR0 = 2.0, the next generation will be twice as large as the current.
Generation time, T, is defined as the average age of a mother producing female offspring,
and is calculated according to this formula:
Summation of x(age cohort)lxmx all divided by R0
To examine the potential effect of conservation efforts focused on egg versus non-egg
stages, the researchers then investigated what happens to the population growth rate, r, if
the mortality rate of any given life stage was reduced by 90%. Interestingly, the life table
models suggested that the best way to preserve loggerhead turtles was to protect adults
and juveniles, particularly large juveniles (see figure on right).
R = ln(R0) / T
ecundity increases linearly with body mass. This relationship is seen in many other
species. Sea turtles, discussed previously, show greater fecundity with larger body mass,
as do some mammals such as red squirrels in Britain.2, 3
This trade-off is particularly pronounced in species that produce a large number of
offspring during a single reproductive event in their lives (known as semelparous
organisms). These species, such as mosquitoes, butterflies, and salmon, can benefit by
reproducing lateronce individuals have had a chance to grow bigger, they will have
more energy and somatic resources for offspring production. One obvious cost of
delaying reproduction is that individuals incur a greater probability of dying before
reproducing, so species that evolve delayed reproduction are trading increased chance of
reproduction for increased investment in reproduction, which often results in more
successful offspring.
Any individual organism has a finite amount of energy and biomass, which must be
partitioned and used toward different life functions such as feeding, growth, respiration,

defense and reproduction. The Principle of Allocation expresses this fact: any energy or
biomass used for one function, such as growth, reduces the amount of energy available
for another function, such as reproduction.

What are there three elements of a life history strategy?

Pattern of development and growth (metamorphosis trade-off example)
Lifespan (cellular degeneration vs. reproduction trade-off)
Timing and quantity of reproduction (yuccas vs. agave; antechinus; ducks
abandoning their brood)
semelparity
A life cycle characterized by one bout of reproduction over an individual's lifetime. This
is in contrast to iteroparous species which undergo multiple bouts of reproduction during
their life span. Semelparous is the term applied to a species that exhibits semelparity.
A life cycle characterized by multiple bouts of reproduction during an individual's
lifetime. This is in contrast to semelparous species which undergo a single bout of
reproduction during their life span.Iteroparous is the term applied to a species that
exhibits iteroparity.
Chapter 4
geometric population growth
Geometric population growth is a growth model for populations with non-overlapping
generations and unlimited resources. Successive generations differ in size by a constant
ratio.

Populations change depending on the birth and death rates within that population. If the
birth rate exceeds the death rate, the population size will obviously increase. Vice versa,
if the death rate exceeds the birth rate, the population size will decrease. Even if we do
not know birth and death rates for a population, we can estimate their net effect by
measuring the difference in numbers of individuals from one time period to the next.
A simple measure of population growth is the ratio of the population size at one time
(Nt+1) to the population size in the previous time step (Nt). This is known as the finite rate

of increase, denoted by lambda (), and is the same as the estimated growth rate that you
have been manipulating for waterhemp.
The finite rate of increase for a population with non-overlapping generations growing
geometically is the ratio of the population size at one time step ( Nt+1 ) to the size of the
population at the previous time step ( Nt ).
= Nt+1 / Nt
Usually, the time step used is equivalent to the generation time (e.g., a year for an annual
plant). The geometric growth model assumes the population's growth is not restricted by
limiting resources, predation, etc.
Geometric growth equation
The geometic growth equation is a useful population growth model for organisms with
non-overlapping generations.
To calculate lambda (), the finite rate of increase:
To calculate the population size for geometric growth, if is known:
The generalized form of the geometric growth equation is:
Here,

is the population's finite rate of increase (per time unit).

t is time.

Nt is the population size at time t

Nt+1 is the population size at time t + 1

N0 is the initial population size, at time 0.

The model also assumes that there is only one discrete reproduction event per unit of time
for all individuals. This is often true for annual plants like waterhemp, which only set
seeds once a year.
The geometric growth model applies to organisms that have one distinct bout of
reproduction each year, so individuals are added to the population all at once

Exponential growth equation

Exponential population growth occurs when the increase in a population's size is
proportional to its current size.
The population size ( N ) as a function of time ( t ) is:
Nt= N0*e^rt
Here, r is the per capita growth rate. The relationship betweenr and , the finite rate of
increase, is:
R = ln()
The instantaneous rate of change for the population, dN / dtis: rN

Note that when the population is small, dN / dt will be comparatively small. Conversely,
when the population is large,dN / dt will be comparatively large.
For more continuously-reproducing populations
When r is greater than 0, the population is increasing; when r is less than 0, the
population is decreasing.
intrinsic growth rate, r
The intrinsic population growth rate (or intrinsic rate of increase), symbolized by r, is the
population growth rate, per individual, for a population with unlimited resources.
The intrinsic rate of growth is the maximum possible per capita growth rate in a certain
environment, with unlimited resources.
If the environment is completely ideal for the species (the best levels for temperature,
precipitation, etc.), then a population will grow as fast as it possibly can for that species.
he symbol rmaxrepresents this maximum intrinsic growth rate, because it is the theoretical
maximum value of r. Life history characteristics and survivorship determinermax for a
species.

The actual per capita growth rate, which is also r for exponential growth, will approach
the maximum, rmax, when environmental conditions are optimal and resources are
unlimited. As conditions become suboptimal, r declines. If a species exists outside of its
environmental tolerance range, r will become negative and the population size decreases.
Populations do grow exponentially, but typically only for short periods of time during
which necessary resources remain abundant. For example, when
introduced exotic species first colonize a new habitat, they often find an abundance of
food and other resources available to them. In addition, there may be a lack of predators
that feed upon the invading species and few native species that are able to
successfully compete with the invader for key resources. Under these conditions, the
growth rate can stay relatively constant for a period of time during which the population
can grow exponentially.
carrying capacity, K
A population's carrying capacity is the maximum number of individuals that its
environment can sustainably support. The carrying capacity is usually constrained by
limiting factors such as space, nutrients, food availability, water, and light.

Logistic Growth Equation

Logistic growth is a model of population growth in which population growth rate (dN/dt)
declines as the population size increases and approaches its maximum possible size (its
carrying capacity, K). If the population size is larger than K, growth is negative and the
population shrinks.
The logistic growth equation models population growth in the presence of a limiting
resource. The equation is:
where r is the intrinsic growth rate, K is carrying capacity, andN is population size.
Dn/dt = rN (1-N/K)
When a population grows logistically, the instantaneous population growth rate and the
per capita growth rate changes as the population grows and resources are depleted, even
though the intrinsic growth rate (r in the logistic equation) remains constant.
INSTANTENOUS pop growth highest at the medium pop sizes, per capita growth
highest when smallest population

density dependence
The influence of density-dependent factors on the dynamics of a population changes with
the size of that population. Normally, the larger the population size, the greater the effect
these factors have. Examples include disease, competition for food, parasitism, and
predation.

density independence
The influence of density-independent factors on the dynamics of a population does not
change with the size of that population. These factors have the same relative impact on
the population regardless of how many individuals are present. Events such as storms,
fires, volcanoes, and floods are examples of density-independent factors that lead to the
same proportion of the population being affected whether the population is large or small.

source-sink population dynamics

In some areas, one population persists stably over a long period of time while nearby
populations are not stable and can go extinct.
If emigrants from the stable population (the source) often rescuethe other populations
(the sinks) from extinction, we call the whole system of populations a source-sink system,
and this system has source-sink population dynamics.

metapopulation
A metapopulation is a collection of spatially distinct subpopulations of the same species
that are connected via dispersal. It is sometimes referred to as a population of
populations. In the strictest definition of a metapopulation, the whole system persists over
time, even though individual populations sometimes go extinct.
equilibrium patch occupancy highest when When patches are larger (lower extinction
rate) and closer together (higher colonization rate).
8

environmental stochasticity
Environmental stochasticity refers to seemingly random variability in resource
availability, ecological community composition, predation pressure, weather events, etc.,
which often causes fluctuations in a population's growth rate.

demographic stochasticity
Population growth rate is determined by birth and death rates. By chance, there could be
many births in a row, leading to a higher population growth than you would expect. Or by
chance, there could be a number of deaths in a row, leading to unexpected extinction.
Such a chance sequence of events leading to variation in a population's growth is known
as demographic stochasticity.