Behavioral Ecology,
Год журнала:
2021,
Номер
32(6), С. 1321 - 1329
Опубликована: Авг. 5, 2021
Abstract
Hemispheric
brain
lateralization
can
drive
the
expression
of
behavioral
asymmetry,
or
laterality,
which
varies
notably
both
within
and
among
species.
To
explain
these
left–right
asymmetries
in
animals,
predator-mediated
selection
is
often
invoked.
Recent
studies
have
revealed
that
a
relatively
high
degree
correlates
positively
with
traits
known
to
confer
survival
benefits
against
predators,
including
escape
performance,
multitasking
abilities,
group
coordination.
Yet,
we
still
know
comparatively
little
about
1)
how
consistently
predators
shape
lateralization,
2)
importance
sex-specific
variation,
3)
heritable.
Here,
take
advantage
model
system
radiation
Bahamas
mosquitofish
(Gambusia
hubbsi)
measure
hundreds
wild
fish
originating
from
multiple
blue
holes
differ
natural
predation
pressure.
Moreover,
estimated
heritability
this
trait
using
laboratory-born
one
focal
population.
We
found
but
not
particular
direction
(left
right)
differed
significantly
across
low
risk
environments.
Fish
high-predation
environments
were
more
strongly
lateralized,
especially
females.
further
confirmed
genetic
basis
species,
significant
additive
variation
population
examined.
Our
results
reveal
represents
key
ecological
factor
has
likely
shaped
origin
maintenance
widespread
phenomenon,
even
potentially
explaining
some
patterns
laterality
recently
described
animals.
Annual Review of Ecology Evolution and Systematics,
Год журнала:
2018,
Номер
49(1), С. 303 - 330
Опубликована: Авг. 15, 2018
Parallel
evolution
across
replicate
populations
has
provided
evolutionary
biologists
with
iconic
examples
of
adaptation.
When
multiple
colonize
seemingly
similar
habitats,
they
may
evolve
genes,
traits,
or
functions.
Yet,
replicated
in
nature
the
laboratory
often
yields
inconsistent
outcomes:
Some
along
highly
trajectories,
whereas
other
to
different
extents
distinct
directions.
To
understand
these
heterogeneous
outcomes,
are
increasingly
treating
parallel
not
as
a
binary
phenomenon
but
rather
quantitative
continuum
ranging
from
nonparallel.
By
measuring
populations’
positions
this
(non)parallel
continuum,
we
can
test
hypotheses
about
and
ecological
factors
that
influence
extent
repeatable
evolution.
We
review
evidence
regarding
manifestation
laboratory,
natural
populations,
applied
contexts
such
cancer.
enumerate
many
genetic,
ecological,
processes
contribute
variation
Proceedings of the Royal Society B Biological Sciences,
Год журнала:
2018,
Номер
285(1882), С. 20180036 - 20180036
Опубликована: Июль 4, 2018
The
question
of
parallel
evolution-what
causes
it,
and
how
common
it
is-has
long
captured
the
interest
evolutionary
biologists.
Widespread
urban
development
over
last
century
has
driven
rapid
responses
on
contemporary
time
scales,
presenting
a
unique
opportunity
to
test
predictability
parallelism
change.
Here
we
examine
evolution
in
an
acorn-dwelling
ant
species,
focusing
heat
island
signal
ant's
tolerance
these
altered
temperature
regimes.
Using
common-garden
experimental
design
with
acorn
colonies
collected
from
rural
populations
three
cities
reared
under
five
treatments
laboratory,
assessed
plastic
shifts
cold
F1
offspring
worker
ants.
In
two
cities,
found
evolved
losses
tolerance,
compression
thermal
breadth.
Results
for
were
more
complex:
one
city,
evidence
simple
populations,
though
another,
difference
population
depended
laboratory
rearing
temperature,
only
became
weakly
apparent
at
warmest
temperatures.
appeared
be
adaptive,
as
our
analysis
fitness
consequences
warming
revealed
that
while
produced
sexual
reproductives
warmer
temperatures,
fewer.
Patterns
natural
selection
tolerances
supported
findings
trade-offs
local
adaptation
across
acted
opposite
directions
between
coldest
Our
study
provides
mixed
support
rise,
and,
importantly,
suggests
promising
use
non-parallel
scales.
The
study
of
parallel
ecological
divergence
provides
important
clues
to
the
operation
natural
selection.
Parallel
often
occurs
in
heterogeneous
environments
with
different
kinds
environmental
gradients
locations,
but
genomic
basis
underlying
this
process
is
unknown.
We
investigated
genomics
rapid
adaptation
marine
snail
Littorina
saxatilis
response
two
independent
axes
(crab-predation
versus
wave-action
and
low-shore
high-shore).
Using
pooled
whole-genome
resequencing,
we
show
that
sharing
regions
high
differentiation
between
generally
low
increases
at
smaller
spatial
scales.
identify
shared
for
each
axis
most
these
overlap
candidate
chromosomal
inversions.
Several
inversion
are
divergent
polymorphic
across
many
localities.
argue
inversions
could
store
variation
fuels
environments,
possibly
as
balanced
polymorphism
by
adaptive
gene
flow.
PLoS Genetics,
Год журнала:
2020,
Номер
16(4), С. e1008658 - e1008658
Опубликована: Апрель 17, 2020
Understanding
the
extent
to
which
ecological
divergence
is
repeatable
essential
for
predicting
responses
of
biodiversity
environmental
change.
Here
we
test
predictability
evolution,
from
genotype
phenotype,
by
studying
parallel
evolution
in
a
salmonid
fish,
Arctic
charr
(Salvelinus
alpinus),
across
eleven
replicate
sympatric
ecotype
pairs
(benthivorous-planktivorous
and
planktivorous-piscivorous)
two
evolutionary
lineages.
We
found
considerable
variability
eco-morphological
divergence,
with
several
traits
related
foraging
(eye
diameter,
pectoral
fin
length)
being
highly
even
This
suggests
repeated
predictable
adaptation
environment.
Consistent
ancestral
genetic
variation,
hundreds
loci
were
associated
within
lineages
eight
shared
variation
was
maintained
despite
histories,
ranging
postglacial
sympatry
(ca.
10-15kya)
pre-glacial
20-40kya)
secondary
contact.
Transcriptome-wide
gene
expression
(44,102
genes)
replicates,
involved
biological
processes
characteristic
morphology
physiology,
revealed
parallelism
at
level
regulatory
networks.
not
only
plastic
but
part
genetically
controlled
cis-eQTL.
Lastly,
that
magnitude
phenotypic
largely
correlated
differentiation
divergence.
In
contrast,
direction
change
mostly
determined
interplay
adaptive
expression,
ecosystem
size.
Ecosystem
size
further
explained
putatively
adaptive,
ecotype-associated
genomic
patterns
lineages,
highlighting
role
stochasticity
evolution.
Together,
our
findings
demonstrate
eco-morphology
contingencies,
overcoming
variable
histories
backgrounds.
Trends in Ecology & Evolution,
Год журнала:
2019,
Номер
35(1), С. 10 - 21
Опубликована: Сен. 12, 2019
Experimental
speciation
is
an
excellent
complement
to
snapshot
studies
of
natural
populations
because
it
can
disentangle
recurring
problems
that
confound
populations.Experimental
made
early
significant
contributions
understanding
evolutionary
processes
mediating
the
evolution
reproductive
isolation.Over
past
decade,
genomics
has
provided
better
predictions
on
how
barrier
loci
spread
in
genome
and
speciation-with-gene-flow
occur.These
developments
remain
difficult
test
have
not
been
widely
adopted
experimental
research.Future
integration
genomic
tools
framework
will
provide
a
step-change
these
outstanding
questions.
Speciation
result
generate
barriers
gene
flow
between
populations,
facilitating
isolation.
typically
studied
via
theoretical
models
tests
populations.
enables
real-time
direct
theory
long
touted
as
critical
other
approaches.
We
argue
that,
despite
its
promise
elucidate
isolation,
underutilised
lags
behind
research.
review
recent
experiments
outline
for
be
implemented
address
current
questions
are
otherwise
challenging
answer.
Greater
uptake
this
approach
necessary
rapidly
advance
speciation.
The
progression
outcome
(see
Glossary)
depend
interactions
forces
[1Ravinet
M.
et
al.Interpreting
landscape
speciation:
road
map
finding
flow.J.
Evol.
Biol.
2017;
30:
1450-1477Crossref
PubMed
Scopus
(256)
Google
Scholar]
act
with
varying
importance
over
space
time
either
facilitate
or
impede
isolation
(RI)
[2Abbott
R.
al.Hybridization
speciation.J.
2013;
26:
229-246Crossref
(1312)
Scholar].
RI
may
arise
through
action
genetic
drift
and/or
divergent
selection,
continuous
migration
secondary
contact,
impacted
by
population
size
structure,
influenced
properties
such
mutation
recombination
rates
[3Ortiz-Barrientos
D.
James
M.E.
Evolution
1519-1521Crossref
(6)
Understanding
relative
focus
A
classic
highly
successful
studying
involves
identifying
phenotypically
trait
testing
association
level
extant
[4McKinnon
J.S.
al.Evidence
ecology's
role
speciation.Nature.
2004;
429:
294-298Crossref
(381)
Scholar,
5Le
Pennec
G.
al.Adaptation
dislodgement
risk
wave-swept
rocky
shores
snail
Littorina
saxatilis.PLoS
One.
12e0186901Crossref
(25)
6Huber
S.K.
al.Reproductive
sympatric
morphs
Darwin's
finches.Proc.
Soc.
B
Sci.
2007;
274:
1709-1714Crossref
(116)
increasing
application
high-throughput
data
(Box
1)
used
reconstruct
history
(e.g.,
demography)
infer
leading
speciation,
often
timescale
3Ortiz-Barrientos
7Wolf
J.B.W.
Ellegren
H.
Making
sense
islands
differentiation
light
speciation.Nat.
Rev.
Genet.
18:
87-100Crossref
(241)
This
analogous
use
forward
genetics
study
function
gene,
but
applied
RI.
Here,
begins
phenotype
proceeds
identify
potential
caused
build
up
diverged
Many
support
success
4McKinnon
However,
method
actually
backward
looking,
reflecting
static
contributed
divergence.
Realistically,
signals
likely
erased
overwritten
progresses.
Thus,
challenged
deduce
multiple
impacting
phenotypic
factors
influence
sequentially
simultaneously,
same
different
directions,
inferred
histories.Box
1Speciation
GenomicsThe
reduced
cost
expanded
ability
25Nosil
P.
al.Ecological
explanations
(incomplete)
speciation.Trends
Ecol.
2009;
24:
145-156Abstract
Full
Text
PDF
(513)
Of
interest
distributed
across
genomes
they
evolve
during
Predicted
patterns
based
whether
geographically
separated
without
flow,
occurring
initial
divergence
following
contact.
In
allopatry,
substantially
constrained
extent
linkage
strength
selection
producing
contrast,
speciation-with-gene-flow,
opposed
both
erode
associations
genes
under
[82Nosil
Feder
J.L.
Genomic
causes
consequences.Philos.
Trans.
2012;
367:
332-342Crossref
(255)
genic
view
posits
initiated
acting
against
at
specific
targets
interested
initiate
(through
build-up
disequilibrium)
RI,
including
subsequent
dependent
architecture
[25Nosil
26Nosil
Harmon
L.
Niche
dimensionality
ecological
speciation.in:
Butlin
R.K.
Patterns
Diversity.
Cambridge
University
Press,
2009:
127-154Crossref
83Wu
C.-I.
process
2001;
14:
851-865Crossref
(910)
84Turner
T.L.
al.Genomic
Anopheles
gambiae.PLoS
2005;
3e285Crossref
(550)
predicted
depending
primary
[85Richards
E.
al.Searching
era.bioRxiv.
2019;
(Published
online
January
28,
2019.
https://doi.org/10.1101/367623)Google
Scholar].Speciation
begun
issues
evolving
response
drift,
their
effect
sizes,
distribution,
associations,
builds
increases,
along
inferring
demographic
[86Ellegren
al.The
species
Ficedula
flycatchers.Nature.
491:
756Crossref
(421)
87Nadeau
N.J.
hybridizing
Heliconius
butterflies
identified
large-scale
targeted
sequencing.Phil.
B.
343-353Crossref
(235)
88Marques
D.A.
nuptial
colour.PLoS
2016;
12e1005887Crossref
(153)
89Smadja
C.M.
al.Large-scale
candidate
scan
reveals
chemoreceptor
host
plant
specialization
pea
aphid.Evolution
(N.
Y).
66:
2723-2738Google
there
well-reviewed
confounding
influencing
heterogeneity
unrelated
history,
time,
variation
strength,
timing
[7Wolf
9Noor
M.A.F.
Bennett
S.M.
Islands
mirages
desert?
Examining
restricted
maintaining
species.Heredity
(Edinb).
103:
439-444Crossref
(266)
10Cruickshank
T.E.
Hahn
M.W.
Reanalysis
suggests
due
diversity,
flow.Mol.
2014;
23:
3133-3157Crossref
(619)
Scholar]),
disentangling
remains
Models
rate,
direction
magnitude
tend
rely
summary
measures
comparing
limited
sets
hypothesised
scenarios.
Additionally,
impact
sometimes
clearly
[5Le
90Jiggins
C.D.
Ecological
mimetic
butterflies.Bioscience.
2008;
58:
541-548Crossref
(100)
91Le
Rouzic
A.
al.Strong
consistent
associated
armour
reduction
sticklebacks.Mol.
2011;
20:
2483-2493Crossref
(56)
92Forbes
A.A.
al.Revisiting
particular
shifts
initiating
insect
speciation.Evolution
71:
1126-1137Google
Scholar],
frequently
characterise
pressures
–
increasingly
so
further
traced
back
history.
differentiation,
[93Langerhans
R.B.
Predictability
parallelism
multitrait
adaptation.J.
Hered.
109:
59-70Crossref
(34)
Alongside
development
which
coestimate
demography
directly
observe
designed
track
powerful
apply
systems
where
observation
unavailable.
Laboratory
(EE)
challenges
manipulating
thought
many
generations
then
(ES)
reverse
function.
It
putative
conditions
therefore
identifies
circumstances
ES
complements
(Table
also
stand-alone
real
time.
several
decades
when
influential
contribution
was
last
reviewed,
10
years
ago
Fry
[8Fry
J.D.
Garland
T.J.
Rose
Michael
Evolution.
California
631-656Crossref
technique
seemed
poised
exponentially
accelerate
outlined
neglected
well
suited
Since
Fry's
review,
advanced
incorporate
more
sophisticated
ideas
constraints
Snapshot
signatures
conventional
vexed
inference
problems,
limiting
[9Noor
11Westram
A.M.
Ravinet
Land
ahoy?
Navigating
while
avoiding
shipwreck.J.
1522-1525Crossref
(3)
provides
potent
controlling
environmental
approaches
cannot
'A
Selection
New
Challenges
That
Can
Address').Table
1ES
Studies
Using
Natural
Populations
Are
Highly
ComplementaryaSeveral
advantages
(bold)
limitations
each
matched
illustrate
complementary
nature.Laboratory-based
ESComparative
methods
populationsRare
(but
important)
serendipitous
events
missed
unless
experiment
largeBetter
represents
given
process,
rather
than
just
occurrenceStarting
characteristics
defined
quantified
priori
researcherIn
most
cases,
ancestral
genomesTypically
reliant
upon
standing
aloneGreater
de
novo
introgression
from
play
roleEnvironment
controlled
kept
constant
manipulated
manner,
throughoutOften
determine
environment
required
delineate
geography
restricting
flowMany
effects
laboratory
adaptation.
If
adaptation
occurred
pre-EE,
diversity
lowerPopulations
close
equilibrium
wildEvolutionary
responses
replicated
series
lines
robustly
link
responsesNo
true
replication.
Lack
create
uncertainty
change
variableEvolution
traits
what
performed
culture
conditions.
Low
niche
means
only
simple
contrasts
madeA
much
wider
range
selected
ariseGene
accurately
reliably
determined
levels,
local
RIDifficult
ongoing
flowLimited
subset
organisms
suitable
EECan
any
diverging
populationsEasy
separate
intrinsic
extrinsic
forms
RILaboratory
settings
exclude
aspects
speciesCan
assess
full
isolating
mechanisms
found
wildPhenotypic
collected
high
temporal
resolution
providing
estimates
hindsight
underlying
changesEven
if
reconstructed,
single
snapshot,
dataExperiments
cover
short
timescales
subsets
processLong
(although
histories
must
inferred)a
Several
nature.
Open
table
new
tab
we
decade
examine
progress
original
areas
research
progressed
since
applied.
using
combined
enable
rapid
advances
suggested
could
address:
efficacy
generating
RI;
types
barriers;
feasibility
parapatric
speciation;
reinforcement
summarise
topics
used,
two
fundamental
ignored
framework.
To
maintain
differences
emerge
Barriers
prezygotic
(premating
postmating,
prezygotic)
postzygotic
stage,
Initial
relatively
strong
even
arbitrary
no
clear
mechanism
conditions,
disruptive
did
generally
lead
divergently
had
little
relevance
fitness
few
altered
tested
Most
sexual
conflict
[12Parker
G.A.
Partridge
Sexual
speciation.Philos.
1998;
353:
261-274Crossref
(364)
13Gavrilets
S.
Is
"engine
speciation"?.Cold
Spring
Harb.
Perspect.
6:
a017723Crossref
(46)
equivocal
Subsequent
work
continues
fail
find
[14Gay
al.Does
faster
larger
sexually
antagonistic
coevolution?.Biol.
Lett.
5:
693-696Crossref
15Plesnar-Bielak
al.No
evidence
bulb
mite
Rhizoglyphus
robini.PLoS
8:
1-8Crossref
(7)
16Debelle
al.Sexual
assortative
mating:
test.J.
29:
1307-1316Crossref
(11)
17Reding
L.P.
hinders
yeast.Biol.
9:
20121202Crossref
(14)
increase
likelihood
assessing
[15Plesnar-Bielak
One
species,
Drosophila
melanogaster,
independently
laboratories
one
[18Syed
Z.A.
manipulation
coevolution
melanogaster.Sci.
Rep.
7:
3330Crossref
(
Global Change Biology,
Год журнала:
2018,
Номер
24(8), С. 3791 - 3803
Опубликована: Апрель 27, 2018
Human
activities
reduce
biodiversity
but
may
also
drive
diversification
by
modifying
selection.
Urbanization
alters
stream
hydrology
increasing
peak
water
velocities,
which
should
in
turn
alter
selection
on
the
body
morphology
of
aquatic
species.
Here,
we
show
how
urbanization
can
generate
evolutionary
divergence
two
species
fish,
western
blacknose
dace
(Rhinichthys
obtusus)
and
creek
chub
(Semotilus
atromaculatus).
We
predicted
that
fish
evolve
more
streamlined
shapes
within
urbanized
streams.
found
urban
streams,
consistently
exhibited
bodies
while
showed
deeper
bodies.
Comparing
modern
populations
with
historical
museum
collections
spanning
50
years,
(1)
rapidly
became
bodied
streams
experienced
over
time,
(2)
had
already
achieved
deepened
years
ago
were
then
(and
no
additional
deepening
time),
(3)
remained
relatively
shallow
stayed
rural
time.
By
raising
from
five
under
common
conditions
laboratory,
morphological
differences
largely
reflected
genetically
based
differences,
not
velocity-induced
phenotypic
plasticity.
suggest
rapid,
adaptive
responses
to
disturbance,
these
vary
unpredictably
different
Integrative and Comparative Biology,
Год журнала:
2019,
Номер
59(3), С. 705 - 715
Опубликована: Май 23, 2019
Animals
use
a
diverse
array
of
motion
to
feed,
escape
predators,
and
reproduce.
Linking
morphology,
performance,
fitness
is
foundational
paradigm
in
organismal
biology
evolution.
Yet,
the
influence
mechanical
relationships
on
evolutionary
diversity
remains
unresolved.
Here,
I
focus
many-to-one
mapping
form
function,
widespread,
emergent
property
many
systems
nature,
discuss
how
redundancy
influences
tempo
mode
phenotypic
By
supplying
possible
morphological
pathways
for
functional
adaptation,
can
release
morphology
from
selection
performance.
Consequently,
decouples
diversification.
In
fish,
example,
parallel
evolution
weaker
traits
that
contribute
mechanically
redundant
motions,
like
suction
feeding
than
with
one-to-one
form-function
relationships,
lower
jaw
lever
ratios.
As
complexity
increases,
historical
factors
play
stronger
role
shaping
trajectories.
Many-to-one
mapping,
however,
does
not
always
result
equal
freedom
The
kinematics
complex
often
be
reduced
variation
few
high
effect.
various
different
four-bar
linkage
systems,
output
(kinematic
transmission)
highly
sensitive
size
one
or
two
links,
insensitive
others.
faster
rates
are
biased
Mechanical
sensitivity
also
results
evolution-evolutionary
transitions
coupled
transition
linkages
other
words,
dynamics
actually
approximate
simpler,
when
strong.
When
examined
macroevolutionary
framework,
same
system
may
experience
distinct
selective
pressures
groups
organisms.
For
performance
tradeoffs
organisms
structure
more
functions.
general,
less
and,
sometimes,
slower
rate
These
trends
unevenness
lineage
across
tree
life.
Finally,
informs
our
understanding
relative
roles
determinism
contingency
Frontiers in Plant Science,
Год журнала:
2020,
Номер
11
Опубликована: Дек. 8, 2020
Parallel
evolution
provides
powerful
natural
experiments
for
studying
repeatability
of
and
genomic
basis
adaptation.
Well-documented
examples
from
plants
are,
however,
still
rare,
as
are
inquiries
mechanisms
driving
convergence
in
some
traits
while
divergence
others.
Arabidopsis
arenosa
,
a
predominantly
foothill
species
with
scattered
morphologically
distinct
alpine
occurrences
is
promising
candidate.
Yet,
the
hypothesis
parallelism
remained
untested.
We
sampled
populations
all
regions
known
to
harbor
ecotype
used
SNP
genotyping
test
repeated
colonization.
Then,
we
combined
field
surveys
common
garden
experiment
quantify
phenotypic
parallelism.
Genetic
clustering
by
region
but
not
elevation
coalescent
simulations
demonstrated
parallel
origin
four
mountain
regions.
Alpine
exhibited
height
floral
which
persisted
after
two
generations
cultivation.
In
contrast,
leaf
were
distinctive
only
certain
region(s),
reflecting
mixture
plasticity
genetically
determined
non-parallelism.
demonstrate
varying
degrees
causes
non-parallelism
across
within
plant
species.
along
sharp
gradient
makes
A.
candidate
Evolution,
Год журнала:
2020,
Номер
74(7), С. 1466 - 1481
Опубликована: Июнь 9, 2020
A
growing
number
of
empirical
studies
have
quantified
the
degree
to
which
evolution
is
geometrically
parallel
by
estimating
and
interpreting
pairwise
angles
between
multiple
replicate
lineages'
evolutionary
change
vectors
in
multivariate
trait
space.
Similar
comparisons,
distance
space,
are
used
assess
convergence.
These
approaches
amount
element-by-element
interpretation
matrices,
typically
testing
for
differences
among
vectors,
compared
a
null
hypothesis
perfect
parallelism.
We
suggest
complimentary
set
approaches,
co-opted
from
quantitative
genetics,
involving
eigen
analysis
comparison
among-lineage
covariance
matrices.
Such
allow
one
identify
major
axes
(e.g.,
alternative
adaptive
solutions),
also
definition
biologically
tenable
hypotheses,
such
as
drift,
against
patterns
can
be
tested.
Reanalysis
dataset
across
replicated
lake/stream
gradient
threespine
stickleback
reveals
that
most
variation
direction
captured
just
few
dimensions,
indicating
greater
extent
parallelism
than
previously
appreciated.
applying
may
often
necessary
fully
understand
form
convergent
evolution.
Journal of Evolutionary Biology,
Год журнала:
2022,
Номер
35(11), С. 1414 - 1431
Опубликована: Сен. 13, 2022
Abstract
Examples
of
parallel
evolution
have
been
crucial
for
our
understanding
adaptation
via
natural
selection.
However,
strong
parallelism
is
not
always
observed
even
in
seemingly
similar
environments
where
selection
expected
to
favour
phenotypes.
Leveraging
this
variation
within
well‐researched
study
systems
can
provide
insight
into
the
factors
that
contribute
adaptive
responses.
Here
we
analyse
results
36
studies
reporting
446
average
trait
values
Trinidadian
guppies,
Poecilia
reticulata
,
from
different
predation
regimes.
We
examine
how
extent
predator‐driven
phenotypic
influenced
by
six
factors:
sex,
type,
rearing
environment,
ecological
complexity,
evolutionary
history,
and
time
since
colonization.
Analyses
show
guppies
highly
variable
weak
on
average,
with
only
24.7%
among
populations
being
explained
regime.
Levels
appeared
be
especially
colour
traits,
decreased
increasing
complexity
history
(i.e.,
when
estimates
a
single
drainage
were
compared
pooled
between
two
major
drainages).
Suggestive
–
but
significant
trends
warrant
further
research
include
interactions
sexes
categories.
Quantifying
accounting
these
other
sources
‘replicates’
leveraged
better
understand
which
drive
nonparallel
aspects
divergence.