Professional Documents
Culture Documents
Mechanisms
Bundlethe more
reproduction of those variable individuals are non-random. Organ-
books you buy,
isms respond evolutionarily to changes in their environment and
the greater your
of Evolution
other selection pressures, including global climate change. The
discount! importance of spatial structure of a population in relation to how
it affects the strength of gene flow and/or genetic drift, as well as
THE CONTENT the genetic variation and evolution of populations, is shown. Gene
flow tends to reduce variation between populations and increase it
Energy Physics
within populations, whereas genetic drift tends to reduce genetic
Engineering variation, especially in small, isolated populations. The mecha-
Biotechnology nisms of evolution can lead to speciation, which requires both time
Biology and genetic isolation of populations, in addition to natural selec-
Mathematics tion or genetic drift.
Chemistry
Christopher J. Paradiseis professor of biology and environ-
mental studies at Davidson College. He teaches introductory
THE TERMS biology, ecology, entomology, and topical seminars on ecotoxi-
Perpetual access cology and renewable natural resources. He also occasionally
leads a study abroad program in India. His research evaluates
for a one time fee
anthropogenic factors that influence insect biodiversity at a
No subscriptions or
variety of scales. His current research interests include effects of
access fees
land use patterns on pollinator communities in parks.
Unlimited
concurrent usage A. Malcolm Campbellteaches biology at Davidson College,
Downloadable PDFs NC. He received national and international education awards:
Genetics Society of America (2013); American Association for the
Free MARC records
Advancement of Science (2012); and American Society for Cell
Biology (2006). He was the founding co-editor in chief of CBE Life
For further information,
Sciences Education; founding director of Genome Consortium
a free trial, or to order,
contact: for Active Teaching (GCAT); and member of the American Soci- Christopher J. Paradise
sales@momentumpress.net ety for Cell Biology governing council (20122014).
A. Malcolm Campbell
Mechanisms of Evolution
Mechanisms of Evolution
10 9 8 7 6 5 4 3 2 1
Keywords
evolution, population, gene flow, genetic drift, adaptation, natural
selection, phenotypes, behavior, predation, selective agent, heterozygous,
homozygous, heritability, dispersal, genotype, gene flow, genetic isola-
tion, genetic distance, non-adaptive evolution, population bottleneck,
inbreeding
Contents
Preface...................................................................................................ix
Acknowledgments....................................................................................xi
Introduction.........................................................................................xiii
Chapter 1 Selection Acts on Individuals with Variable
Characteristics....................................................................1
Selection can Act on Behaviors...........................................1
Natural Selection on a Discrete Trait..................................6
Chapter 2 Species May Evolve in Response to Climate Change........11
Range Expansion.............................................................12
Evolutionary Response to Changing Rainfall...................14
Ethical, Legal, Social Implications: Data are
Needed to Formulate Policy, but Science is
Often Misused in the Process........................................17
Chapter 3 Two Seemingly Isolated Populations may not
Actually be Isolated..........................................................23
Some Populations Have Limited Dispersal.......................23
Dispersal Links Geographic and Genetic Distance...........27
Chapter 4 Populations can Evolve in the Absence of
Natural Selection..............................................................33
Population Isolation Affects Genetic Diversity.................33
A Population Bottleneck Reduces Genetic Diversity.........37
Conclusion............................................................................................41
Glossary................................................................................................43
Index....................................................................................................45
Preface
This book about mechanisms of evolution is part of a thirty book se-
ries that collectively surveys all of the major themes in biology. Rather
than just present information as a collection of facts, the reader is treated
more like a scientist, which means the data behind the major themes are
presented. Reading any of the thirty books by Paradise and Campbell
provides readers with biological context and comprehensive perspective
so that readers can learn important information from a single book with
the potential to see how the major themes span all size scales: molecular,
cellular, organismal, population and ecologic systems. The major themes
of biology encapsulate the entire discipline: information, evolution, cells,
homeostasis and emergent properties.
In the twentieth century, biology was taught with a heavy emphasis
on long lists of terms and many specific details. All of these details were
presented in a way that obscured a more comprehensive understanding.
In this book, readers will learn about mechanisms of evolution and some
of the supporting evidence behind our understanding. The historic and
more recent experiments and data will be explored. Instead of believing
or simply accepting information, readers of this book will learn about the
science behind the mechanisms of evolution the way professional scien-
tists dowith experimentation and data analysis. In short, data are put
back into the teaching of biological sciences.
Readers of this book who wish to see the textbook version of this
content can go to www.bio.davidson.edu/icb where they will find
pedagogically-designed and interactive Integrating Concepts in Biology
for introductory biology college courses or a high school AP Biology
course.
Acknowledgments
Publishing this book would not have been possible without the generous
gift of Dr. David Botstein who shared some of his Breakthrough Prize
with co-author AMC. Davids gift allowed us to hire talented artists (Tom
Webster and his staff at Lineworks, Inc.) and copyeditor Laura Loveall.
Thanks go to Kristen Mandava of Mandava Editorial Services for project
management and guidance. In particular, we are indebted to Katie Noble
and Melissa Hayban for their many hours and attention to detail.
Kristen Eshleman, Paul Brantley, Bill Hatfield and Olivia Booker
helped us with technology at Davidson College. We are grateful to ad-
ministrators Tom Ross, Clark Ross, Carol Quillen, Wendy Raymond,
Verna Case, and Barbara Lom who had confidence in us and encouraged
us to persist despite setbacks along the way.
Thanks to my wife Amy Brooks for her constant support during the
development of this textbook, and my daughter Evelyn for her endless
energy. Thanks to Malcolm Campbell for his steadfast resolve and opti-
mism. Without him, this book would not exist. Thanks to collaborator
Laurie Heyer for taking my sometimes half-baked math ideas and turn-
ing them into powerful and elegant Bio-Math Explorations. I learned a
lot from both of them. While the math is largely absent from this
book, our collaboration with her made this a better book. Nancy Stamp
at Binghamton University, and Bill Dunson and Richard Cyr at The
Pennsylvania State University influenced me greatly in how I think as
a scientist and approach my teaching. Finally, I thank my students in
Integrated Concepts in Biology II, who enthusiastically participated in
our experiment to redesign introductory biology, starting with the text
and ending with a new approach to teaching biology.
Introduction
Although the unit of study in evolution is most often the population,
evolution can be and is studied at the levels of the molecule, cell, and
organism. In this book, evolution will be studied at the level of the in-
dividual organism and the population, that is all the individuals of the
same species living in the same place at the same time, by examining how
natural selection acts on individuals that possess certain traits, the evolu-
tionary consequences of rapid environmental changes, and the evolution-
ary consequences of gene flow and genetic drift, which are all manifested
at the population level. The main themes of the evolution will be evident
throughout the book. For instance, natural selection is the mechanism
that accounts for adaptation to the selective pressures of interacting spe-
cies. Life continues to evolve as the environment changes, and humans
are a major contributor to changes in our 21st century environment.
CHAPTER 1
Selection Acts on
Individuals with Variable
Characteristics
80
= 36 hour survival
60 = 60 hour survival
percent survival
40
20
0
timid ordinary bold
type of guppy
Figure 1 Total survival (pooled for all ten trials) of guppies with
different behavioral tendencies to inspect potential predators. Survival
was measured at two time points during the experiment.
Source: Data from Dugatkin 1992 in text.
Selection Acts on Individuals with Variable Characteristics 3
flee from predators before the predator gets too close. This may enhance
their survival, and may be something they were unable to do in the first
experiment, where survival of bold guppies was zero. The variation in be-
havioral tendencies is associated with variation in appearance, as colorful
males were generally bolder, especially in the presence of females. Drab
males spent less time inspecting predators, because they spent more time
near females when females were present. However, in choice tests, fe-
males preferred to mate with bold males when a predator was nearby and
bright males when predators were absent. Inspection behavior was not an
issue when there was no predator nearby, but because of the correlation
between brightness and boldness, females may be able to assess male bold-
ness by examination of their color alone.
In Dugatkins first experiment, bold males were selected againstthat
is, their survival was very low because of their tendency to inspect preda-
tors. A scientist may wonder then how individuals with that behavior
remained in the population, or why the population did not consist of all
timid males. Because boldness is preferred by females that counters preda-
tion selection against the boldness.
In addition, variation in color pattern, which is associated with be-
havior, is used by females as a signal for potential boldness when they
cannot actually observe how a male would behave if there were a predator
present. Bold males inspect more often, which may signal to the predator
that it has been spotted, and this behavior allows bold individuals to flee
an approaching predator earlier than a timid individual. Females may
prefer these types of males because bright color indicates their boldness,
and boldness is related to their ability to detect predators. Bold behavior
may be associated with other aspects of success, such as higher rates of
feeding. The value of this to the female is that bold, healthy males may
contribute more advantageous genes to the females offspring. Phenotypes
that remain in a population are maintained by providing an advantage to
the possessor, whereas phenotypes selected against reduce the ability to
survive or reproduce.
If boldness and brightness are then favored in the environment, why
are there any timid, drab males left in the population? Even though fe-
males prefer bold, bright males, timid males are still able to mate, which
helps explain why this characteristic remains. Dugatkin and Godin found
6 MECHANISMS OF EVOLUTION
that the correlation between brightness and boldness was not perfect; not
all bold males were brightly colored. Multiple genes may code for proteins
involved in those characteristics, and the environment in which a guppy
develops may also contribute to that variation. Natural selection is not
perfect, and there are many factors that affect the success or failure of any
one individual. Timid males may have higher survival than bold males
under some conditions. If all timid, drab males were eliminated from the
guppy population, the remaining population of bright, bold males will be
less variable, and this may have negative consequences in an environment
with a high abundance of predators. Such a population of bold and bright
guppies could be selected against and face extinction.
many annual censuses, and what they found was that the proportion of
white flowers was consistently at or close to 0 on the west side of the
ravine and consistently at or close to 1.0 on the east side of the ravine.
The researchers hypothesized that there was intense local selection for
flower color, such that blue flowers were favored on the west side of the
ravine and white on the east. To support this hypothesis, the researchers
tested four other genes, predicting that other genetic loci would show
no pattern across the ravine if selection were only on flower color. To
determine the frequencies of alleles of the four genetic loci, the scientists
collected individual desert snow plants across the ravine. They extracted
enzymes and separated allozymes, variants of an enzyme, for the four
genes using electrophoresis. Electrophoresis is the process in which large
molecules can be separated according to size and electrical charge by ap-
plying an electric current to them in a gel. The scientists found exactly
what they predicted; the other genes that do not influence flower color
were not selected for or against.
The scientists also planted both white and blue flowered plants in
plots on both sides of the ravine to determine their seed production suc-
cess in the two habitats. Note that 1995 was wetter than average, and
1996 was drier than average. In 1995, there was no difference in seed
production between blue and white plants on the west side but on the
east side, where white flowers historically dominated, seed production in
white flowers was significantly greater, by about 30%, than seed produc-
tion of blue plants. The trend was reversed in 1996, the drier than aver-
age year, where blue plants produced significantly more seeds than white
plants on the west side, but both white and blue plants produced equal
numbers of seeds, on average, on the east side.
Finally, Schemske and Bierzychudek collected data on environmental
factors on the two sides of the ravine to determine what selective factors
there might be in the two habitats. They looked at the other plants in the
community, which are potential competitors, as well as soil properties.
Nine out of ten plant species had differences in area covered on the west
side versus the east side, with six covering more area on the west side than
the east, and three covering more area on the east side than the west. The
researchers also found significant differences in soil chemistry, with five
out of ten variables being different on one side than the other.
8 MECHANISMS OF EVOLUTION
unmeasured factor could have then given rise to variation in plant com-
munity composition. Either of these factors, soil or the other species of
plants present on either side of the ravine, could be the source of selection
for flower color on the two sides of the ravine, although Schemske and
Bierzychudek did not test individual factors in the soil or in the plant
community. However, Schemske and Bierzychudek showed that ecologi-
cal factors can and do vary, and this variation leads to natural selection
on a local scale over short periods of time. Natural selection can eliminate
certain characteristics from a population, thereby reducing variation. But
natural selection can also maintain variable characteristics by favoring
certain types in different local habitats. Variation among individuals in a
population can also lead to descent with modification over much longer
periods of time.
Bibliography
Dugatkin LA: Tendency to inspect predators predicts mortality risk in the
guppy (Poecilia reticulata), Behav Ecol 3:124127, 1992.
Godin J-GJ, Davis SA: Who dares, benefits: predator approach behaviour
in the guppy (Poecilia reticulata) deters predator pursuit, P Roy Soc
Lond B Bio 259(1355): 193200 1995.
Godin J-GJ, Dugatkin LA: Female mating preference for bold males
in the guppy, Poecilia reticulata. Proc Natl Acad Sci USA 93(19):
1026210267, 1996. Available online: http://www.jstor.org/stable/
40373.
Schemske DW, Bierzychudek P: Spatial differentiation for flower color
in the desert annual Linanthus parryae: was Wright right? Evolution
61(11):25282543, 2007.
CHAPTER 2
Europe
0.0
Australia
Although climate has changed often during the history of the planet, the
changes occurring now are rapid and global, and they are outside the range
experienced by humans since before the development of agriculture. Ecolog-
ical systems have begun to change during the past few decades, and studies
have documented changes in species interactions, seasonal activity patterns,
and expansion of geographic ranges. In this chapter, some of the data on
evolutionary responses of species to these climate changes will be explored.
Range Expansion
Thomas and colleagues studied several species of insects in England. Two of
the species they analyzed were bush crickets, which is a type of grasshopper.
Species May Evolve in Response to Climate Change 13
The scientists tracked the year each species and population was first re-
corded throughout its current range, using published records of the spe-
cies. They found that the range of each species has changed over time,
becoming larger for each species. Both species of bush cricket are spread-
ing northwards and inland from southern coastal regions. The expansion
of geographic range, noted for many species of insects, may be related to
evolutionary changes in species. Not all species will respond in the same
way; in fact, some species may become extinct as climate changes. In this
section, potential evolutionary responses to climate change are considered.
In the case of the bush crickets, both species have a long-winged
form that had been relatively rare in sampled populations in the past,
and they determined the proportion of long-winged individuals in the
populations, and analyzed that proportion according to the date a popu-
lation was first discovered. Populations that had been known for longer
periods of time tended to have lower proportions of long-winged forms.
In addition, the frequency of long-winged individuals has changed over
time, and some recently discovered populations have the highest frequen-
cies. Those populations are also at the edges of the expanding geographic
range. The increased frequencies of long-winged individuals in recently
established populations could mean that environmental variables have
changed and are affecting the development of immature crickets, or there
could be genetic differences that affect the temperature a cricket must
experience when developing before it will become a long-winged adult.
Thomas and colleagues concluded that, even in the face of climate
change, there are likely to be genetic, evolutionary changes in the popula-
tions because high population density plays a strong role in development
of the long-winged form, and newly founded populations typically have
both low population densities and high proportions of long-winged in-
dividuals. In addition, if the crickets expand their ranges as temperatures
rise, scientists would expect to observe similar proportions of long-winged
individuals in populations that are in the north now as would have been
observed in the south decades earlier. That is not the case. Thomas and
colleagues concluded that there were evolutionary changes in these two
species that facilitated the expansion of their geographic ranges. They fur-
ther concluded that dispersal of long-winged individuals can speed up
range expansion disproportionately as they increase in the population.
14 MECHANISMS OF EVOLUTION
were planted in pots in a greenhouse. Each of 300 pots received seeds from
each site and each time: two from the ancestor (1997) cross (one from wet
soil and one from dry) and two from the descendant (2004) cross (one
from wet soil and one from dry), resulting in four plants per pot.
For the first 33 days pots were watered daily. Researchers stopped
watering one-third of the pots at that point, representing a short rainy
season. They then stopped watering the next third of the pots after
day 51, representing a medium season; and then they stopped watering
the last third after day 81, representing a long season. They recorded the
day that each seed germinated and each plant flowered.
The trend in flowering times for all growing season lengths was similar.
In particular, the distribution of flowering times was similar and highly
overlapping for ancestral and descendant plants from the sandy dry soil
site. However, for plants from the wet soil site, the distribution of flower-
ing times in descendant plants was shifted to earlier flowering times, with
the median flowering time being about two weeks earlier in descendant
than in ancestral plants.
Thus the median flowering time for plants from the dry soil site
were all earlier than plants from the wet soil site, regardless of whether
the plants were ancestral (1997) or descendant (2004). The researchers
found large differences in median flowering times for plants from the
wet soil site, depending upon whether the plants were offspring of an-
cestral, descendant, or hybrid parents and depending upon their site of
origin. Descendant plants that originated from the post-drought time
(2004) flowered earlier than plants from the pre-drought time (1997).
The major environmental change between 1997 and 2004 was a change
to much less than average precipitation, leading to shorter rainy sea-
sons. This led to a change in the population from the wet soil site that
made it look more similar to the plants from the dry soil site. Plants
from the dry soil site had evolved to flower earlier because the ances-
tral plants had already experienced dry conditions; thus, they did not
respond to the dry period between 2000 and 2004 as the plants from
the wet soil site did.
Plant survival was also monitored. No matter where or when plants
were from, plant survival was near 100% when exposed to a long growing
season. In addition, both ancestral and descendant plants from the sandy
16 MECHANISMS OF EVOLUTION
dry soil site had survival over 90% no matter the length of the growing
season. Ancestral plants from the wet soil site when exposed to short and
medium growing seasons had only 50 and 44% survival, respectively, but
descendant plants had 75 and 72% survival when exposed to short and
medium growing season lengths, respectively.
Mortality differences among the treatments were thus profound.
When seasons are short, plants that delay flowering are unable to pro-
duce seeds before dry conditions kill them. The higher survival of 2004
offspring indicates that this population of plants had evolved to the drier
conditions in as short a time as 7 years. Survival of ancestral and de-
scendant plants from the dry soil type was always high, regardless of the
length of the wet season. This suggests local adaptation, that is, there were
genetic differences between the two populations. One population had
already evolved to dry conditions, and the other one evolved during the
period between 1997 and 2004.
In a second experiment designed to assess heritability in flowering
times, Franks and his colleagues planted seeds from the ancestral (1997)
parents in the same pots. Heritability is the proportion of observed varia-
tion attributable to variation in genes or the extent to which individual ge-
netic differences contribute to individual differences in observed traits. As
in the drought tolerance experiment, they measured the time to flowering.
They estimated heritability as the slope of the line that best fits the graph
of mean flowering time of a set of offspring versus the mean flowering time
of the parents (Table 1). If all variation in flowering time was due to dif-
ferences in environmental factors, the slope of the line would be zero, and
there would be no correlation between mean offspring flowering time and
mean parent flowering time. The higher the heritability, the more variation
in offspring flowering times is attributable to variation in parent flowering
times, and thus offspring inherited their flowering time from their parents.
Genetic differences over time and between the two populations are
also indicated by the heritability estimates. Estimates greater than zero in-
dicate that there is variation in flowering times caused by genetic changes.
Heritability in these wild mustard populations was high, that is, up to
almost half of the observed variation in flowering times was due to varia-
tion in parent flowering times. Wild mustard exposed to drier conditions
appear to have evolved and be better adapted to short wet seasons. Rapid
Species May Evolve in Response to Climate Change 17
evolution by natural selection can occur if the selective force is strong and
the heritability is high.
The two examples presented here demonstrate that evolution can
occur quite rapidly as climate changes. Climate change involves more
than just warming, and organisms can respond evolutionarily to a vari-
ety of climate factors. Species can respond both evolutionarily and eco-
logically to more than one factor simultaneously. For the response to be
evolutionary, there must be heritable genetic variation for the trait in
question. The changes in species affect entire ecological systems, because
evolutionary changes may allow for range expansion or changes in inter-
actions among species. Species that do not respond evolutionarily may go
extinct, which also will affect ecological systems.
activity to establish that the activity will not or is very unlikely to result
in significant harm. If the precautionary principle had been applied when
we began using fossil fuels, it would have been up to the proponents of
this new fuel to establish that its use will result in no adverse effects. Now
that science knows there are adverse effects, one could argue that the pro-
ponents of doing nothing to curb greenhouse gas emissions must estab-
lish that continued activity will not result in further environmental harm.
Conversely, the proponents of changing our economy to one based
on renewable energy and efficiency must establish that any new activities
will not result in economic or environmental harm. Those proponents
argue that there is an obligation that because the predicted level of harm
of not changing course is high, we must take action to prevent or mini-
mize such harm even when the absence of scientific certainty makes it
difficult to predict the likelihood of harm occurring, or the level of harm,
should it occur. While preventative measures may have a short-term dev-
astating effect on the economy, applying the precautionary principle to
act now to prevent future catastrophe might still be better economically
and ecologically in the long-term.
Scientific uncertainty should not preclude preventative measures
being taken to protect the environment. But, if we cannot reject the nega-
tive (that climate change is not caused by human actions) using the scien-
tific method and a controlled experiment, humans can still use science to
support the affirmative that climate change is caused, in part, by human
actions. A weight-of-evidence approach can be used, where data from
different scientific disciplines, including biology, meteorology, climate
science, chemistry, physics, and geology all point to rapid climate change
that has been caused at least in part by humans and will likely move
Earths recent climate outside of the range humans have experienced since
the dawn of the agricultural revolution about 10,000 years ago.
Bibliography
Franks SJ, Sim S, Weis AE: Rapid evolution of flowering time by an an-
nual plant in response to a climate fluctuation, Proc Natl Acad Sci
USA 104(4):12781282, 2007.
Species May Evolve in Response to Climate Change 21
Population structures, that is, how the density and location of individu-
als in a population vary, are often complex in time and space. That is,
populations of the same species may be geographically separated from one
another, or they may become separated over time due to geologic events,
such as formation of mountains or erosion of canyons. In this chapter, sev-
eral examples of spatial structure and dispersal will be explored to better
understand how dispersal can maintain genetic similarity of populations.
Consider a species of squirrels that lives on either side of the Colorado
River. Although an individual squirrel may occasionally cross the river,
once the Grand Canyon was hollowed out by the river squirrels were
much less likely to cross that divide, and they became two separate popu-
lations. Over time, those separated populations evolved differences in the
patterns of their fur color and are recognized by experts as two subspe-
cies of the same species. Might they become two separate species, given
enough time? Possibly, but time is only one factor in a speciation event.
If two populations diverge enough, a speciation event may occur. In this
chapter, a mechanism of evolution that can counter the divergence and
spatial separation of two populations will be explored.
Kay and Rytas Vilgalys studied the population of a fungus, the oyster
mushroom, which feeds on and decomposes logs in moist forested areas.
Fungi are multicellular heterotrophic organisms with chitinous cell walls.
Fungi typically grow as hyphae, filaments composing the vegetative part
of a fungus, often forming a dense mat on or in solid substrates, and are
not observed by most people. Many people notice the mushroom, which
is the fleshy, spore-bearing fruiting body of a fungus, typically produced
aboveground or on a food source, such as a decomposing log. Mushrooms
contain spores that, when released, lead to dispersal, or spreading, of the
fungus.
Kay and Vilgalys mapped the location of each log in a forest that had
sprouting mushrooms, recorded the position of clusters of mushrooms
on each log, and collected samples of mushrooms. To determine how
many distinct individuals had colonized each log, they performed genetic
testing. Counting individual mushrooms is not a good indicator of how
many individuals live on a log, because one individual fungus in a log can
grow multiple mushrooms.
Kay and Vilgalys found between 1 and 16 distinct individuals on each
of 21 logs, for a total of 53 individual fungi. The researchers mapped the
location of rotting logs and noted the number of individual fungi on each
log. Many logs contained only one individual, several contained 24,
several more contained 6 or 7, and one contained 16 distinct individuals.
These logs were spread throughout the forest and there was no apparent
spatial pattern in this distribution of individuals.
The scientists compared the genetic similarity of individuals and their
distribution relative to related individuals. They did this for the forest, but
also for each individual log. One individual fungus that colonizes a log
through dispersal might grow to take over that log through asexual repro-
duction, or other individuals may arrive through dispersal. The fungi in
this forest can be examined at the spatial scale of the log (one or several
individuals), or at the scale of the whole forest (the population).
The movement and genetic relatedness of individuals within the popu-
lation are important factors in determining the population structure and the
evolution of the population. If logs were farther apart than spores generally
travel, new populations could form, having become isolated by distance.
This could affect the genetic relatedness and the evolution of the species.
TWO ISOLATED POPULATIONS MAY NOT BE ISOLATED 25
3
distance (m)
mean flight
0
A bumble bees butterflies
2.0
flowers visited per plant
mean number of
1.5
1.0
0.5
0.0
bumble bees butterflies
B polinator type
Figure 3 Mean flight distance. (A) and number of flowers visited per
plant (B) for bumblebees and butterflies.
Source: From Schmitt, 1980, Tables 1 and 2.
TWO ISOLATED POPULATIONS MAY NOT BE ISOLATED 27
0.5 m. About 92% of the bumblebee flights were less than 1 m in length.
Schmitt found that butterflies are much more likely than bumblebees to
travel distances longer than 1 m, and that butterflies occasionally fly very
long distances, while bumblebees never do, at least not while they are
foraging.
If populations are partitioned into small local clusters that are not
genetically connected and natural selection acts differentially on those
clusters, they may diverge evolutionarily. That is, genetic, morphological,
behavioral, or physiological differences could arise such that individuals
in one population no longer recognize individuals from another popu-
lation as belonging to the same species. A species dispersal ability will
partly determine whether populations diverge genetically.
In the case of the sunflower-like plants, dispersal of pollen is deter-
mined by the pollinator. This led to mean flight distances between flowers
being much lower for bumblebees than butterflies (Figure 3A). Butterflies
sometimes visited flowers close together, but they often flew a significant
distance before stopping to collect nectar. Foraging bees visited a slightly
higher number of flowers on a single plant than butterflies (Figure 3B).
These observations make sense in light of what is known about foraging
behavior of bees and butterflies. Recall that bees require a lot of energy,
and they obtain this energy by visiting many flowers in rapid succession.
Butterflies, on the other hand, are feeding and looking for mates, so they
often fly right by available flowers.
In populations where bumblebees are the primary pollinator, Schmitt
concluded that population clusters would be small, both in numbers of
individuals and area occupied. When butterflies are the pollinator, popu-
lations are predicted to be more spread out and to be connected geneti-
cally, because gene flow and mating between widely separated individuals
would be much more substantial.
their genomes into the other population. This makes the two populations
more similar genetically and consequently decreases the genetic distance
between the populations.
If the two populations are very different, the genetic distance is
near 1. Isolated populations that have no gene flow or dispersal of
individuals into their population may evolve over time, either through
natural selection or non-adaptive evolution. Those genetic changes will
make that isolated population different than other populations of the
same species, increasing the genetic distance between that population and
any other population to which it is compared. With genetic distance,
scientists can better understand the influence of pollination and other
factors in the genetic diversity of bladder campion and other species.
Olson and McCauley examined the effects of geographic distance
between populations on the genetic distance and population structure.
Geographic distance is easy to determine from a map, and there are also
methods to measure genetic distance.
To quantify the relationship between genetic distance and geographic
distance, Olson and McCauley found the genetic distance and geographic
distance between every possible pair of populations, and found the cor-
relation coefficient between these two measurements. They found that as
geographic distance between pairs increased, genetic distance decreased,
meaning the populations farther apart geographically were more similar
genetically (correlation coefficient = 0.246). This decrease was statisti-
cally significant, with a p-value of 0.002. This result was unexpected by
the researchers; they predicted that populations farther apart would have
higher genetic distances.
The scientists hypothesized that there would be a positive relation-
ship because they suspected that dispersal between closely spaced pop-
ulations would be common, decreasing the genetic distance between
those populations. It was expected that dispersal to far populations
would be less common, leading to larger genetic distances between
populations spaced farther apart. It is possible that long distance dis-
persal in this species is more common than believed, leading to the
observed negative relationship. Populations founded by such long dis-
tance dispersal might retain their genetic closeness to the parent popu-
lation for some time.
30 MECHANISMS OF EVOLUTION
those five genes. Only one out of 22 genes had multiple alleles that
were found in different frequencies among starling populations in the
four different states, and the estimate of variation among populations
was low.
As with oyster fungi, most starlings disperse only a short distance
away from their nest, relative to the maximum dispersal distance of this
very mobile species. But a small percentage of them fly very far distances
from where they were born. While dispersal abilities and distances are
related to the mobility of species, biologists know that species can disperse
far enough to spread genes from one area to another but that each had
a very different population structure. Even in populations separated by
many kilometers, genetic similarities may exist. This indicates very little
genetic isolation. A species may have populations separated spatially but
connected genetically, and this helps maintain the species as a distinct
entity. Clearly this is true for starlings. A female that flies 2,000 km and
mates with a male in its new population mixes her genes with the popu-
lations genes, and this decreases the genetic distance between the two
populations.
Gene flow between populations is a mechanism of evolution that
can prevent divergence of populations and speciation events. The flow of
genetic material between populations helps maintain genetic similarity
among individuals living in different areas. A species may be structured
as a set of discrete, but genetically connected, populations linked through
dispersal of individuals, as a set of discrete but disconnected populations,
or as one large population. The way that a population is structured spa-
tially affects gene flow and may increase or decrease it depending upon
proximity of populations and dispersal ability of the species. Mutations
or adaptations that arise in one population can spread among populations
through the process of gene flow. Gene flow can thus prevent speciation
events from occurring, and it affects genetic diversity. Genetic diversity
between populations decreases with increasing gene flow. Genetic diver-
sity within populations may increase with increasing gene flow as new
alleles may enter the population with migration. In the next chapter, how
isolated populations may evolve in the absence of natural selection and
gene flow will be examined.
32 MECHANISMS OF EVOLUTION
Bibliography
Cabe PR: Dispersal and population structure in the European starling,
The Condor 101: 451454, 1999.
Campbell DR, Waser NM: Variation in pollen flow within and among
populations of Ipomopsis aggregata, Evolution 43(7):14441455,
1989.
Kay E, Vilgalys R: Spatial distribution and genetic relationships among
individuals in a natural population of the oyster mushroom Pleurotus
ostreatus, Mycologia 84:173182, 1992.
Olson MS, McCauley DE: Mitochondrial DNA diversity, population
structure, and gender association in the gynodioecious plant Silene
vulgaris, Evolution 56: 253262, 2002.
Schmitt Je: Pollinator foraging behavior and gene dispersal in Senecio
(Compositae), Evolution 34:934943, 1980.
CHAPTER 4
It is known that gene flow can reduce the genetic distance between popu-
lations or even disrupt divergence of populations into two species. Di-
vergence of two isolated populations can occur because of variation in
environmental selection pressures in two populations. However, consider
two populations of the same species that are isolated from each other
and live in environments with similar selective pressures. In the absence
of natural selection, or adaptive evolution, can populations still diverge?
The conditions under which such non-adaptive evolution occurs may
relate to the structure of the population, which can be affected by gene
flow, the latter of which connects populations genetically. Non-adaptive
evolution is evolution of a population caused by random changes in the
genome that are neither beneficial nor harmful to individuals. Popula-
tions separated in space may exchange very little, if any, genetic material.
It is important to examine the population structure of a species when at-
tempting to determine the evolutionary mechanisms at play, as described
in previous chapters. Can divergence of populations or even speciation
occur in the absence of adaptive evolution and gene flow?
studied three species of plants that live in the Swiss Alps: alpine willow-
herb (Epilobium fleischeri), a species of rose (Geum reptans), and yellow
bellflower (Campanula thyrsoides). Populations of these plants are sepa-
rated by anywhere from 5 to 30 kilometers. These plants are all insect
pollinated and seeds are dispersed by the wind.
Kuss and his colleagues studied these populations of these species
within the same geographic area using the same genetic analyses. They
predicted that they would find a genetic structure of each species that cor-
related with the fragmented and isolated population structure, but that
the species with the best long-distance dispersal adaptations would be
less affected by isolation. Thus, populations of species with poor dispersal
abilities that are widely separated would be less similar genetically. The
scientists also predicted that small populations of each species would have
a lower degree of genetic diversity.
Kuss and his colleagues extracted DNA and separated randomly se-
lected segments of DNA using Random Amplification of Polymorphic
DNA (RAPD). Multiple fragments of different sized DNA sequences are
produced in this process, and these can be separated by size on a gel. A
particular pattern of bands of different sizes will be produced for each
individual plant. Two plants that have the same exact genome will have
the same pattern. Individuals who differ genetically will have different
banding patterns, because DNA sequences would be of different size and
would end up in different positions on the gels. In this way, RAPD can
be used to analyze the genetic diversity of individuals and populations.
Genetic variation was measured as the number of unique band pat-
terns and their frequencies within a population. In theory, a population
that had zero genetic variation contained individuals that all had the exact
same banding pattern. A population with more than one banding pat-
tern has segments of DNA that have more than one allele. Kuss and his
colleagues determined that there were 47 to 89 segments that had more
than one allele, depending upon the species and they then determined the
percentage of genes that had multiple alleles (Figure 4A).
From the frequency of bands of particular sizes, the scientists could
also estimate which individuals were homozygous for particular seg-
ments and which individuals were heterozygous (that is, they had two
alleles for the same DNA region). Heterozygosity, the percentage of
Populations can Evolve in the Absence of Natural Selection 35
% heterozygosity ( 1 SE)
80 20
60 15
40 10
20 5
0 0
willowherb rose bellflower willowherb rose bellflower
A B
by geographic distance than the other species. Even though there are
trends with geographic distance, it turns out that even the highest genetic
distances were all below 0.5, suggesting that populations are still pretty
similar genetically, even when far away from each other geographically.
Evolution could be occurring in situations with such population struc-
tures, and divergence could occur in the isolated populations.
The populations of these three species are fragmented, and this has
led to population isolation due to distance between populations. The
farther apart two populations are, the less likely they are to exchange
genetic material by gene flow; and in this study, even populations close
to each other were genetically dissimilarto a degree. This effect was
most pronounced for the rose (G. reptans). The mechanisms for dispersal,
insects for pollen and wind for seeds, are unlikely to lead to long distance
dispersal for this species. Wind may seem like it could carry seeds a long
distance, but seeds may not be carried far in mountains with complex
topography.
There was a lot of variation in the data and a lot of scatter in the plot
of genetic distance versus geographic distance. Some pairs of populations
that were geographically close had a high level of genetic distance, and
others at the same distance had a low level, indicating perhaps that there
was some randomness in the flow of genetic materialjust by chance,
pollen or seeds from one population may make it to another population
regardless of distance. The average genetic diversity of these three moun-
tain species was similar, and each had a low percentage of heterozygosity,
which is predicted for isolated populations. Figure 4A indicates that up to
80% of genetic loci were estimated to be homozygous.
However, whether heterozygous or homozygous, there was a high per-
centage of DNA segments with multiple alleles within populations. This
was contrary to the prediction of Kuss and his colleagues. If the popula-
tions are isolated, as they appear to be, then genetic diversity should de-
cline over time as alleles are eliminated by chance events. Just by chance,
individuals possessing a certain allele may not produce surviving off-
spring. Those chance events are part of the nature of genetic drift, which
is a form of non-adaptive evolution. Think of genetic drift as the accumu-
lation of changes in the genetic makeup of a population over time due to
random events. The fact that many alleles were not eliminated indicates
Populations can Evolve in the Absence of Natural Selection 37
that other evolutionary mechanisms may have been acting on the popula-
tions, or that not much time had passed since isolation occurred.
It was also surprising to the researchers that they did not find a corre-
lation between measures of genetic diversity and population size. Genetic
drift is often strongly affected by population size, because small popula-
tions are more likely, just by chance, to lose alleles that have low frequency
in the population. In very small populations, or in species that have low
dispersal abilities, loss of genetic diversity may be more extreme. For in-
stance, orchids, plants that have been studied due to their extremely high
diversity, often exist in small, isolated populations. If natural selection
acted differentially on these isolated populations, then speciation events
could occur, given enough time. But if natural selection is not responsible
for divergence of populations, it might be concluded that non-adaptive
evolution played a role, as Kuss and colleagues did.
0.8
heterozygosity ( 1 SD)
0.6
0.4
0.2
0.0
A Netherlands museum Norway Austria
number of alleles ( 1 SD)
0.0
B Netherlands museum Norway Austria
in Norway and Austria. Each was genetically less similar to the pres-
ent Dutch population. This suggests that the Norwegian and Austrian
populations have not changed genetically in the last 50 to 60 years,
but that the Dutch population has. The three closely-related popula-
tions were large and genetically connected to other populations, which
helped maintain genetic diversity.
Genetic drift, whether caused by isolation of small, fragmented popu-
lations or a population bottleneck, is an important mechanism of evolu-
tion and increases genetic distance between isolated populations of the
same species. Drift, like natural selection, causes loss of genetic diversity
over time. The mechanisms for loss of diversity are different from natural
selection as genetic drift leads to random, non-adaptive elimination of al-
leles from small, isolated populations. Given enough time and isolation,
genetic drift and natural selection are powerful evolutionary forces that
may result in speciation events. Time and isolation are important fac-
tors in both natural selection and genetic drift, whereas lack of isolation
caused by gene flow may prevent speciation events.
Bibliography
Kuss P, Pluess AR, gisdttir HH, et al.: Spatial isolation and genetic
differentiation in naturally fragmented plant populations of the Swiss
Alps, J Plant Ecol 1:149159, 2008.
Larsson JK, Jansman HAH, Segelbacher GS, et al.: Genetic impoverish-
ment of the last black grouse (Tetrao tetrix) population in the Neth-
erlands: detectable only with a reference from the past, Mol Ecol
17:18971904, 2008.
Conclusion
The three mechanisms of evolution that explored in this chaptergenetic
drift, gene flow, and natural selectionhave variable effects on popula-
tions depending on the environmental conditions and the structure of a
population. Even though these mechanisms affect populations, they act
on individuals, which then leads to changes in populations. Individuals
do not evolve, populations do; and they can do so in isolation from other
populations of the same species, leading to speciation, or populations of
the same species can be connected through gene flow, which may pre-
vent speciation events from occurring. Evolution can be adaptive or non-
adaptive, depending upon the environmental context, the population
structure, and the presence or absence of various selective factors. Adaptive
evolution occurs through natural selection, whereas non-adaptive evolu-
tion occurs primarily through genetic drift. Further exploration of the
mechanisms of evolution will take place in other books in this series, in-
cluding Evolutionary History and Evolution of Interactions in Communities.
Glossary
adaptation. A change or the process of change by which a species becomes better
suited to its environment via the mechanism of natural selection.
allozymes. A variant of an enzyme
behavior. The way in which an animal acts in response to an internal or external
stimulus.
dimorphism. Refers to a population that contains individuals that are of one of
two phenotypes.
dispersal. Dispersal means to spread from the place of birth.
DNA.Deoxyribonucleic acid, the carrier of genetic information, is a mac-
romolecule present in nearly all living organisms as the main constituent of
chromosomes.
dominant allele. An allele that expresses its phenotypic effect even when hetero-
zygous with a recessive allele.
electrophoresis. The process in which large molecules can be separated according
to size and electrical charge by applying an electric current to them in a gel.
evolution. The scientific explanation for the origin of life and its continual change
over time.
fungus. Fungi are multicellular heterotrophic organisms with chitinous cell walls.
gene flow. Gene flow is the movement and incorporation of alleles from one
population to another.
genetic distance. Genetic distance measures the genetic similarity between two
populations by comparing their levels of genetic diversity.
genetic drift. The loss of alleles by random causes and not natural selection.
genetic isolation. Genetic isolation occurs when two populations of the same
species have separate and distinct genomes.
genotype. Genotype is the genetic composition that determines an organisms
appearance or behavior.
heritability. Heritability is the proportion of observed variation attributable to
variation in genes or the extent to which individual genetic differences contribute
to individual differences in observed traits.
heterozygous. Heterozygous individuals have two different versions, or alleles,
of a gene.
homozygous. Homozygous individuals have two copies of the same version, or
allele, of a gene.
hyphae. Hyphae are the filaments composing the vegetative part of a fungus,
often forming a dense mat.
inbreeding. Inbreeding is reproduction between close relatives.
44 GLOSSARY
Bees, 25 Electrophoresis, 7
median flight distance, 2627 Environmental Protection
Behavior, selection acting on, 16 Agency, 18
Bierzychudek, Paulette, 69 Epilobium fleischeri. See Alpine
Black grouse, 3740 willowherb
Bladder campion European starlings, 3031
allele combinations of, 28 Evolution
genetic distance and geographic non-adaptive, 33
distance, 2830 of population, 3340
widespread dispersal of, 30 range expansion, 1213
Brassica rapa. See Wild mustard response to rainfall, 1417
Bush cricket, 1213 of species. See Species evolution,
Business-as-usual policy, 19 in response to climate
Butterflies, 25 change
median flight distance, 2627 Executive Orders, 18
The Momentum Press digital library is very affordable, with no obligation to buy in future
years.
For more information, please visit www.momentumpress.net/library or to set up a trial in the
US, please contact mpsales@globalepress.com.
EBOOKS Mechanisms of Evolution
FOR THE Christopher J. Paradise A. Malcolm Campbell
APPLIED BIOLOGY COLLECTION
Three of the four major mechanisms of evolution, natural selec-
SCIENCES
tion, genetic drift, and gene flow are examined. There are 5tenets
LIBRARY of natural selection that influence individual organisms: Individuals
Create your own within populations are variable, that variation is heritable, organ-
Customized Content isms differ in their ability to survive and reproduce, more individu-
als are produced in a generation than can survive, and survival&
Mechanisms
Bundlethe more
reproduction of those variable individuals are non-random. Organ-
books you buy,
isms respond evolutionarily to changes in their environment and
the greater your
of Evolution
other selection pressures, including global climate change. The
discount! importance of spatial structure of a population in relation to how
it affects the strength of gene flow and/or genetic drift, as well as
THE CONTENT the genetic variation and evolution of populations, is shown. Gene
flow tends to reduce variation between populations and increase it
Energy Physics
within populations, whereas genetic drift tends to reduce genetic
Engineering variation, especially in small, isolated populations. The mecha-
Biotechnology nisms of evolution can lead to speciation, which requires both time
Biology and genetic isolation of populations, in addition to natural selec-
Mathematics tion or genetic drift.
Chemistry
Christopher J. Paradiseis professor of biology and environ-
mental studies at Davidson College. He teaches introductory
THE TERMS biology, ecology, entomology, and topical seminars on ecotoxi-
Perpetual access cology and renewable natural resources. He also occasionally
leads a study abroad program in India. His research evaluates
for a one time fee
anthropogenic factors that influence insect biodiversity at a
No subscriptions or
variety of scales. His current research interests include effects of
access fees
land use patterns on pollinator communities in parks.
Unlimited
concurrent usage A. Malcolm Campbellteaches biology at Davidson College,
Downloadable PDFs NC. He received national and international education awards:
Genetics Society of America (2013); American Association for the
Free MARC records
Advancement of Science (2012); and American Society for Cell
Biology (2006). He was the founding co-editor in chief of CBE Life
For further information,
Sciences Education; founding director of Genome Consortium
a free trial, or to order,
contact: for Active Teaching (GCAT); and member of the American Soci- Christopher J. Paradise
sales@momentumpress.net ety for Cell Biology governing council (20122014).
A. Malcolm Campbell