Main topics of lecture:
I: Species concepts:
1.- What is a species?
2.- Biological species concept
3.- The phylogenetic species concept
4.- The morphospecies concept
5.- Applying species concept

II: Mechanisms of isolation:
6.- Physical isolation
7.- Changes in chromosomes as a barrier to gene flow

III: Mechanisms of divergence:
8.- Genetic drift
9.- Natural selection
10.- Sexual selection

IV: Secondary contact:
11.- Reinforcement of parental forms
12.- Hybridization, creation of new species, and
hybrid zones

V: The genetics of differentiation and isolation:
13.- Classical genetics
14.- Analyzing quantitative trait loci

1.2.- All human cultures recognize different types of organisms in nature and name
them. People intuitively group like with like. The challenge to biologists has
been to move beyond these informal judgments to a definition of species that
is mechanistic and testable, and to a classification system that accurately reflects
the evolutionary history of organisms

1.3.- This has been difficult to do. In the past 30 years alone there have been at least
half a dozen species concepts proposed. There has been even philosophical debates
about whether the unit we call species actually exists in nature or whether it
is merely a linguistic and cultural construct

1.4.- However, all these definitions agree that species share a distinguishing
characteristics, which is EVOLUTIONARY INDEPENDENCE.
Evolutionary Independence occurs when mutation, selection, migration, and
drift operate on each species separately. This means that species form a
boundary for the spread of alleles. Consequently, different species follow
independent evolutionary trajectories. The differences among species concepts
center on the problem of establishing practical criteria for identifying evolutionary
independence

2.2.- The BSC is compelling in concept and useful in some situations, but it is
often difficult to apply. For example, if nearby populations do not actually overlap,
we have no way of knowing whether they are reproductively isolated. Instead,
biologists have to make subjective judgments to the effect that "if these populations
were to meet in the future, we believe that they are divergent enough that they
would not interbreed, so we will name them different species". The BSC can never
be tested in fossil forms and it is difficult to apply in the many plant groups where
hybridization between strongly divergent populations is routine.

Figure 12.2.: Monophyletic groups. The groups that are
circled on this phylogenetic tree are all monophyletic
The taxa A-J are the smallest monophyletic groups on the
tree

3.2.- Under the phylogenetic species concept (PSC), species are identified
by estimating the phylogeny of closely related populations and finding
the smallest monophyletic groups. The taxa labeled A-J (Fig. 12.2) are
the smallest monophyletic groups on the tree and represent distinct species.
Therefore to be called separate species under the PSC, populations must
have been evolutionary independent enough for DIAGNOSTIC traits
to emerge. Therefore species are named on the basis of statistically
significant differences in the traits used to estimated the phylogeny

3.3.- Putting this criterion into practice is not easy. Many biologists
object to the idea that a "species-specific trait" may be anything that
distinguishes populations in a phylogenetic context. Such traits can be
as trivial as a single substitution in DNA that is fixed in one population
but not in another, or a slight but measurable and statistically increase
in hairiness on the underside of leaves.

3.4.- It is estimated that instituting the phylogenetic species concept
could easily double the number of named species. Proponents of the
PSC are not bothered by that prospect. They respond by saying that
if this increase did occur, it would merely reflect biological reality

5.2.- The red wolf is native to southeastern US, its population had dwindled to
a mere handful of individuals by the early 1970s. Many of the animals that were
left showed characteristics typical of coyotes, which started to become abundant
in the wolf's range in the 1930s. This morphological similarity suggested that
wolves and coyotes were hybridizing extensively.

5.3.- Biologists were able to capture 14 red wolves that apparently did not
have coyote traits. These animals bred readily in captivity and were ready to be
reintroduced in protected habitats.

5.4.- However, under federal conservation biology regulations
(the federal government uses the Biological Species Concept), the extensive
hybridization with coyotes made the species status of the red wolf
questionable. Federal conservation agencies use the Biological Species Concept.
It was suggested that the red coyote is actually a population
of hybrids between the gray wolf and coyote. Morphological studies of
red wolves collected before 1930 indicated that these individuals formed a
morphospecies with intermediate traits between gray wolves and coyotes.
However these data showed that animals collected after 1930 were much
similar to coyotes and therefore it was suggested that hybridization was
a recent phenomenon. This study concluded that red wolves qualify
as a species

5.5.- Genetic studies using DNA collected from specimens captured before 1930
told a different study, as no diagnostic differences were found between
red wolves and coyotes. Instead, the genetic data supported the hypothesis
that red wolves are a hybrid between gray wolves and coyotes, and have no
unique genetic characteristics of their own. This data suggest that
the red wolf's intermediate characteristics are the result of hybridization
and not independent evolution.

The key issue with speciation is that once gene flow is dramatically reduced or
ceases, evolutionary isolation occurs and speciation is initiated

6.2.- The classical theory for how speciation begins is called the allopatric
model. This theory was developed by Erns Mayr who proposed that speciation
is especially likely to occur in small populations that become isolated on the
periphery of a species' range. Population genetic models have shown that
speciation in peripheral populations can occur rapidly when selection for
divergence is strong and gene flow is low.

6.3.- Physical isolation is obviously an effective barrier to gene flow, and
undoubtedly has been an important trigger for the second stage in the speciation
process: Genetic and ecological divergence. Geographic speciation can come
about through dispersal and colonization of new habitats or through vicariance
events, where an existing range is split by a new physical barrier (Fig. 12.5)

Figure 12.5.: Isolation by dispersal and vicariance

GEOGRAPHICAL ISOLATION THROUGH DISPERSAL
AND COLONIZATION

6.4.- A good example of this process can be found with the Drosophila species
of Hawaii. Approximately 500 named species are found in two genera and it
is estimated that 350 additional species still need to be formally described and
named. The ecological diversification of this group is unprecedented. There are
species in all the habitats of the islands and food sources vary widely among
species. They also display a great deal of morphological variation (Fig. 12.6)

Figure 12.6.: Hawaiian Drosophila. This photograph shows
three of the species, there are remarkable differences
in body size, wing coloration, and other traits

6.5.- How did this enormous diversity come to be? The leading explanation is
called founder hypothesis. This hypothesis maintains that this endemism results
when small populations of flies disperse to new habitats or islands. As result, the
colonists found new populations that are cut off from the ancestral species.
Divergence begins after the founding event, resulting from drift and selection
on the genes involved in courtship displays and habitat use. Phylogenetic studies
support this has been the case for these insects (pages 411- 412 of textbook)

GEOGRAPHICAL ISOLATION THROUGH VICARIANCE

6.6.- Vicariance events split a species' distribution into two or more isolated
ranges and discourage or prevent gene flow between them. Researchers have
used phylogenies to study a classical vicariance event: The recent separation
of of marine organisms on either side of Central America. The Isthmus of
Panama closed about 3 million years ago.

6.7.- Phylogenies based on DNA data of populations of snapping shrimp from
either side of the isthmus provide interesting insights on the effect of vicariance
on speciation (Fig. 12.8). Species which are sister in the cladogram are found
on opposite sides of the isthmus!!!
 

Figure 12.8.: Phylogenetic tree of snapping shrimp species
found on both sides of the Isthmus of Panama. The P or C
designations with each number refer to whether the
species is found in the Pacific or Caribbean

6.7.- In an additional experiment males and females of various species pairs
were put together in aquaria and watched for aggressive or pairing interactions,
the researchers found a strong correlation between the degree of genetic
distance between species pairs and how interested the shrimp were in
courting.

6.8.- None of the pairs that formed during the courtship experiments produced
fertile clutches. This last observation confirms that the Pacific and Caribbean
populations are indeed separate species under all three of the species concepts
we have reviewed

6.9.- We would expect that genetic distances and degrees of reproductive isolation
would be identical in all seven species traits. This is not the case. For example,
DNA sequence divergence between species pairs varied from about 6.5% to
over 19%. Why is that?? Because it is unlikely that the land bridge popped up
all at once. Instead the land rose and the ocean gradually split. Therefore
different shrimp populations became also isolated gradually

7.2.- There are two types of polyploidy. Polyploids whose chromosomes
originate from the same ancestral species are called autopolyploids.
Polyploids that result from hybridization events between different species
are called allopolyploids. It is thought that allopolyploids are significantly
more common than autopolyploids.

7.3.- Purely on the basis of its distribution in plants, polyploidization has
to be considered an important mechanism for initiating speciation. Under
the assumption that any plant with a haploid number of chromosomes
greater than 14 is polyploid, Verne Grant estimated that 58% of monocots and
43% of dicots are the descendants of ancestors that underwent polyploidy.
Grant goes so far as to say that polyploidy "is a characteristic of the plant
kingdom"

8.2.- Because genetic drift is a sampling process, its effects are most
pronounced in small populations. This is important because most species
are thought to have originated with low population sizes. Normally,
only tiny numbers of individuals are involved in colonization events,
and peripheral populations tend to be small. As a consequence,
genetic drift has long been hypothesized as the key to speciation's
second stage (See point 8.4 for further discussion on this topic).

8.3.- Genetic drift also will play an important role when populations go through
bottlenecks. Bottleneck effect refers to the occurrence of genetic drift in populations
reduced in size through fluctuations in abundance. Therefore we will expect genetic
drift to play major roles when new populations are formed (founder effect) or when
populations pass for severe reduction in their size (bottleneck effect).

8.4.- In general founder events and population bottlenecks are unlikely to reduce
genetic variation unless the number of founders is tiny and/or the alleles involved
are extremely rare (Fig. 12.A). However (as it was shown in Lectures 8 and 9),
founder events can have other consequences. Although the sample of individuals
is likely to include nearly all of the ancestral population's genes, the frequencies
of the genes may differ from the parental population. Isolated populations can
have exceptionally high frequencies of otherwise rare alleles.

Figure 12.A.: The chance that a founder population will be
not formed only by individuals which are homozygote for one
allele (no reduction of genetic diversity) depends on the
number of founders and the gene frequencies. Here
the chance of having a polymorphic population is shown
for three different frequencies at a two-allele locus. The graph
illustrates the chance that the population has both alleles
after the founder event or population bottleneck

8.4.- The role of drift in speciation events is controversial. Peter Grant and Rosemary
Grant (1996) point out that hundreds of small populations have been introduced to
new habitats around the world in the last 150 years due to the action of humans, but
that few, if any, dramatic changes in genotypes have resulted because of genetic drift.
Although genetic drift once dominated discussions of speciation mechanisms, most
evolutionary biologists now take a much more balanced view. Natural selection has
also been shown to be an important force promoting divergence of isolated populations

9.2.- A good example of selection without geographical isolation is provided by the
apple maggot fly. This species is a major agriculture pest, causing millions of dollars
of damage to apple crops each year. The flies also parasitize the fruits of trees in the
hawthorn groups (species of Crataegus) which are closely related to apples (both are
members of the plant family Rosaceae).

9.3.- Courtship and mating occur on or near the fruits. Females lay eggs in the fruit
while it is still on the tree. Apple trees clearly represent a novel food source for
this fly. Hawthorn trees and the fly are native to North America, but apple trees
were introduced from Europe less than 300 years ago. The main evolutionary
question that we have is: Are the flies that parasitize the fruits of apple and hawthorn
trees forming two distinct sets of populations?

9.4.- This hypothesis implies that natural selection, based on a preference for
different food sources, has created two distinct races of flies that might represent the
starting point for two new species.

9.5.- Because the two host trees and fly populations occur together throughout their
entire ranges it is possible that flies from the same population simply switch from
apple to hawthorn trees and back based on fruit availability. However studies based on
molecular markers show that flies collected from hawthorns versus apples have
statistically significant differences in the frequencies of alleles for six different
enzymes. This is strong support for the hypothesis that hawthorn and apple flies
have diverged and now form distinct populations. Even though the two races
look indistinguishable, they are easily differentiated on the basis of their
genotypes

9.6.- Additional studies support that there are two distinct races. In experiments
where individuals are given a choice of host plants, apple and hawthorn flies
show a strong preference for their own fruit type. Because mating takes
place on the fruit, this habitat preference should result in strong nonrandom
mating. This was confirmed by field studies that found that matings
between hawthorn and apple flies accounted for just 6% of the total observed
matings

9.7.- This example shows that physical isolation is not an absolute requirement
for populations to diverge and also that natural selection for divergence can
overwhelm gene flow and trigger speciation. Populations can diverge even
with low to moderate degrees of gene flow if two important conditions are
met:

        i.- Selection for divergence must be strong (different host preferences
                for both flies)
      ii.- Mate choice must be correlated with the factor that is promoting
                divergence (mating of flies takes place on different host plants)

This kind of speciation that does not require physical isolation is known as
sympatric speciation or the sympatric model of speciation

10.2.- In the Hawaiian Drosophila, for example, sexual selection is thought to have
been a key factor in promoting divergence among isolated populations. For example
males of Drosophila heteroneura have wide, hammer-shaped head (Fig. 12.12a).
Because males butt heads when fighting (Fig. 12.12.b).

Figure 12.12.a.: Male of Drosophila heteroneura have wide,
hammer-shaped head

Figure 12.12.b.: Males of Drosophila heteroneura butt heads to
establish display territories

10.3.- In contrast, males of D. silvestris (a close relative of Drosophila heteroneura)
have heads that are similar in size and shape to female Drosophila heteroneura
(Fig. 12.12c). Instead of head-butting, D. silvestris males fight on the lek by rearing
up and grappling with one another (Fig. 12.12d).

Figure 12.12.c.: Male of Drosophila silvestris have normally
shaped heads

Figure 12.12.d.: Males of Drosophila silvestris fight over display
territories by rearing up and grappling with one another

10.4.- These facts are consistent with the following scenario:

    i.- In the ancestor of silvestris and heteroneura males had normal heads. Females
chose the males who were most successful in combat
    ii.- A mutation occurred in an isolated subpopulation, which lead to males with a
new fighting behavior: head-butting
    iii.- The mutant males were more efficient in combat and females still preferred to
mate with those males who won most contests
    iv.- The mutation increased to fixation. As a result, strong divergence among the
populations occurred due to sexual selection

11.2.- The fate of these hybrids offspring determines the outcome of the speciation event:

        i.- Will the hybrids thrive, interbreed with each of the parental populations, and
                eventually erase the divergence between them?
       ii.- Will hybrids have new characteristics and create a distinct population of their own?
      iii.- What happens if hybrid offspring have reduced fitness relative to the parental
                populations?

11.3.- Theodosius Dobzhanzky reasoned that if the two parent populations have diverged
sufficiently in allopatry, their hybrid offspring should have markedly reduced fitness
relative to the individuals in the parental population. Therefore parents that produce a
hybrid offspring will have a reduction of their fitness, and therefore there should be a
strong selection favoring assortative mating (i.e., mating among individuals from the same
population). That is selection will increase reproductive isolation between parents from
different populations and should favor individuals that choose mates only from the same
population. Selection that reduces the frequency of hybrids in this way is called
reinforcement of parental forms or just reinforcement. If reinforcement occurs, it
would finalize the speciation process by producing complete reproductive isolation

11.4.- Reinforcement will favor assortative mating, therefore it is expected that it will
select for mechanisms of pre-mating isolation.   Selection might favor mechanisms
that alter the aspects of mate choice, genetic compatibility or life history (such as
the timing of breeding). Therefore, reinforcement will lead to divergence in traits that
prevent fertilization to occur and usually results in prezygotic isolation of the two
species.

11.5.- In other situations parent populations could be genetically isolated in the
absence of reinforcement if hybrid offspring are sterile or infertile. This second situation
is known as postzygotic isolation

12.2.- A hybrid zone is a region where interbreeding between diverged populations
occurs and hybrid offspring are frequent. Hybrid zones can be produced after secondary
contact between species that have diverged in allopatry.

12.3.- We have seen that it is possible for hybrid offspring to have lower or higher
fitness than pure-bred offspring, with very different consequences (reinforcement of
parental forms or the formation of a new species). Research on hybrid zones has
confirmed that a third outcome is also possible. Frequently no measurable differences
can be found between the fitnesses of hybrid and pure offspring. The following
three possibilities dictate the size, shape, and longevity of hybrid zones:

            i.- When hybrid and parental forms are equally fit, the hybrid zone is wide.
                Individuals with hybrid traits are found at high frequency at the center of
                the zone and progressively lower frequencies with increasing distance.
                In this type of hybrid zone, the dynamics of gene-frequency change are
                dominated by drift. The width of the zone is a function of two factors:

                    i.a.- How far individuals from each population disperse each generation
                    i.b.- How long the zone has existed

           ii.- When hybrids are less fit than purebreed individuals, the fate of the hybrid
                zone depends on the strength of selection against them. If selection is very
                strong and reinforcement occurs, then the hybrid zone is narrow and short
                lived

         iii.- When hybrids are more fit than purebreds, the fate of the hybrid zone depends
                on the extent of environments in which hybrids have an advantage. If hybrids
                achieve higher fitness in environments outside the ranges of the parental species,
                then a new species may form. If hybrids have an advantage at the boundary
                of each parental population's range, then a stable hybrid zone may form. For
                example many hybrid zones are found in regions called ecotones, where markedly
                different plant and animal communities meet

13.2.- The questions that motivate current research in the genetics of speciation
are focused on the number, and nature of genes that distinguish closely
related species

13.3.- Postzygotic isolation confirms that the population is reproductively
isolated and that the speciation process is complete. Identifying the genes
responsible for postzygotic isolation might tell us something about this
important aspect of isolation

13.4.- One of the best clues in this gene hunt comes from a striking observation
made in experimental crosses between recently diverged taxa. J.B.S. Haldane
discovered  a genetic phenomenon that bears his name as the Haldane's rule.

13.5.- The Haldane's rule establishes that in crosses in which only one sex of
offspring is inviable or infertile while the other sex develops normal. The
heterogametic sex (XY for males in insects and mammals) usually has
reduced viability or fertility. The genetic mechanism of Haldane's rule is the
topic of active modern research because it is likely to reflect some of the key
genetic changes that lead to speciation. In insects and mammals the
homogametic sex is XX (for females)

13.6.- What is about having one of each sex chromosome that contributes to
sterility or inviability when hybrids are formed between divergent populations.
Recently a consensus has emerged that a hypothesis put forth by H.J. Muller in
the early 1940s is probably correct. Muller started out by considering
an autosomal locus (locus no situated on a sex chromosome), "A", and a
X-linked locus B

13.7.- Then he supposed that individuals from one species are fixed for alleles
A₁ and B₁, while individuals from a sister species are fixed for alleles
A₂ and B₂. Further, he supposed that alleles A₁ and B₂ interact to cause inviability

13.8.- Muller pointed out that if females from the first species
(they will have genotype A₁A₁X-B₁X-B₁) mate with males of the second
species (they will have genotype A₂A₂X-B₂Y) then:

            i.- The female offspring will be A₁A₂X-B₁X-B₂. They will be
                    viable because they have copies of A₂ and B₁ which
                    interact to give a normal phenotype
           ii.- The male offspring will be A₁A₂X-B₂Y. They will be
                    inviable because they have alleles A₁ and B₂ which interact to
                    cause infertility or inviability

14.2.- In QTL mapping, researchers obtain F1 and F2 generations and study
the segregation of morphological traits and of several molecular markers. They
try to find correlations between these molecular markers and the quantitative
traits. If a statistically significant associations is found between the trait and the
marker that has been mapped, it implies that a QTL for this trait near the
marker contributes to that trait. We can then estimate how much of the observed
variation is due to the alleles found in each particular QTL (Fig. 12B).
 

Figure 12.A.: Conceptual mapping in F2 populations. DNA
markers throughout the entire genome are tested for the
likelihood they are associated with a QTL. In this example, a
single linkage group with four restriction fragment length
polymorphisms (RFLP) marker loci is shown on the left.
Individuals in the mapping population are analyzed in terms of
marker genotype at each locus by gel electrophoresis,
which produces a banding pattern indicating the parental
allele(s) inherited by each individual (center). For each
marker locus, the indoviduals are divided into classes
according to their marker genotype (two homozygous
parental classes and one heterozygous class). A significant
difference in phenotypes between classes at a particular
locus indicates a QTL is probably nearby. In this example,
there is little difference in mean phenotype between the
three possible genotypic classes at RFLP-1 or RFLP-2,
indicating that they are probably not tightly linked to
the QTL of interest. By contrast, significant differences
in mean phenotype among genotypic classes are observed at
RFLP-3 and RFLP-4, indicating that a QTL is probably located
close to these two markers