ArticlePDF Available

Intraspecific Phylogeography: The Mitochondrial DNA Bridge Between Population Genetics and Systematics

November 1987
Annual Review of Ecology and Systematics 18(1):489-522

DOI:10.1146/annurev.es.18.110187.002421

Authors:

Jonathan Arnold

University of Georgia

Eldredge Bermingham

Smithsonian Institution

Show all 8 authorsHide

Content uploaded by Jonathan Arnold

Content may be subject to copyright.

Intraspecific Phylogeography: The Mitochondrial DNA Bridge Between Population Genetics

and Systematics

Author(s): John C. Avise, Jonathan Arnold, R. Martin Ball, Eldredge Bermingham, Trip

Lamb, Joseph E. Neigel, Carol A. Reeb, Nancy C. Saunders

Source:

Annual Review of Ecology and Systematics,

Vol. 18 (1987), pp. 489-522

Published by: Annual Reviews

Stable URL: http://www.jstor.org/stable/2097141

Accessed: 25/02/2009 23:39

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at

http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless

you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you

may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at

http://www.jstor.org/action/showPublisher?publisherCode=annrevs.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed

page of such transmission.

JSTOR is a not-for-profit organization founded in 1995 to build trusted digital archives for scholarship. We work with the

scholarly community to preserve their work and the materials they rely upon, and to build a common research platform that

promotes the discovery and use of these resources. For more information about JSTOR, please contact support@jstor.org.

Annual Reviews is collaborating with JSTOR to digitize, preserve and extend access to Annual Review of

Ecology and Systematics.

http://www.jstor.org

Ann. Rev. Ecol. Syst. 1987. 18:489-522

INTRASPECIFIC

PHYLOGEOGRAPHY:

The

Mitochondrial

DNA Bridge Between

Population

Genetics and Systematics

John C. Avisel, Jonathan Arnold', R. Martin Ball', Eldredge

Bermingham 2, Trip Lamb"3, Joseph E. Neigell 4, Carol A.

Reebl, and Nancy C. Saunders"

'Department

of Genetics,

University

of Georgia,

Athens,

Georgia

30602;

2NMFS/

CZES, Genetics, 2725 Montlake

Boulevard

East, Seattle, Washington

98112;

3Savannah

River Ecology

Laboratory,

Drawer

E, Aiken, South

Carolina

29801;

4Department

of Microbiology

and

Immunology,

School

of Medicine,

University of

California,

Los

Angeles,

California

90024;

5School

of Veterinary

Medicine,

Virginia

Tech University,

Blacksburg,

Virginia

24046

INTRODUCTION

A recurring debate in evolutionary biology is over the extent to which

microevolutionary

processes operating

within species can be extrapolated

explain macroevolutionary

differences among species and higher taxa (36,

38, 45, 46, 53, 67, 68, 80). As discussed by Stebbins & Ayala (83), several

issues involved must be carefully distinguished, such as (a) whether

micro-

evolutionary processes (e.g. mutation, chromosomal

change, genetic drift,

natural

selection) have operated

throughout

the history of life (presumably

they have); (b) whether

such known processes can by themselves account for

macroevolutionary

phenomena;

and (c) whether these processes can predict

macroevolutionary

trends and patterns. In another, phylogenetic sense,

489

0066-4162/87/11 20-0489$02.00

490 AVISE

ET AL

macroevolution

is ineluctably

an extrapolation

of microevolution:

Organisms

have parents,

who in turn had

parents,

and so on back

through time. Thus, the

branches in macroevolutionary

trees have a substructure that consists of

smaller branches

and twigs, ultimately

resolved as generation-to-generation

pedigrees (Figure 1). It is through these pedigrees that genes have been

transmitted,

tracing the stream of heredity

that is phylogeny.

It would seem that

considerations of phylogeny

and

heredity

should

provide

a logical starting point for attempts to understand any connections of

macroevolution

to microevolution.

Yet amazingly,

the discipline traditionally

associated

with heredity

and microevolutionary process (population

genetics)

developed and has remained

largely

separate

from those fields associated with

phylogeny and macroevolution

(systematics

and

paleontology). Thus, several

classic textbooks in population

genetics (35, 39, 64) do not so much as index

"phylogeny," "s

ystematics," or "speciation," while the equally important

textbooks in systematics (55, 81, 96) can be read and understood

with only

the most rudimentary

knowledge of Mendelian and population genetics.

Notwithstanding

some evidence for recent increased communication between

these disciplines (40, 71), too many systematists

and population

geneticists

continue to operate

in largely separate

realms, employing

different

languages

and concepts to address issues that should be of importance

to all.

X n X tn u)

sV) *0 G D E pedigrees

E E

C O A R S E F I N E

R E S O L U T I O N

Figure I At closer levels of examination,

macroevolutionary

trees (such as the one on the left

summarizing relationships among some of the vertebrate

classes) must in principle have a

substructure

consisting

of smaller

and smaller

branches,

ultimately

resolvable as family pedigrees

through

which genes have been transmitted. Some branches

in the pedigree

on the right

have been

darkened to indicate the transmission

path of mtDNA from the earliest pictured

female.

PRINCIPLES OF mtDNA PHYLOGEOGRAPHY 491

It might also be supposed

that the newer field of molecular

evolution, with

its obvious grounding

in genetics and yet a concern with phylogeny, would

have facilitated a firmer linkage between micro- and macroevolutionary

study. And to some extent it has, by allowing evaluations of large-scale

evolutionary trees in terms of DNA and protein characters with a known

genetic basis. But movement in the opposite direction-extending phyloge-

netic principles and reasoning to the microevolutionary

level-has been

negligible. Thus, when molecular

evolutionists work at the intraspecific level,

they tend to adopt the terms and concepts of population genetics, such as

"variances in allele frequencies," "genetic drift," "mutation-selection bal-

ance," "fitness,"

and so forth, but not the terms and concepts of systematics,

such as "monophyletic groups," "parsimony networks," "clades," or "syn-

apomorphic character states." Conversely, when systematists work at the

intraspecific level, for example to describe subspecies, it is usually with

morphological

or behavioral traits whose genetic basis (or even control) is

poorly known. Worse yet, geographic

locale per se is too often the primary

basis for assigning newly collected specimens to "subspecies," so that at-

tempts to understand

any relationship

between phylogenetic differentiation

and spatial separation

flirt with circular

reasoning.

The reasons

that the field of molecular evolution has not contributed

greatly

to the incorporation

of phylogenetic principles

into population genetics are,

we suspect, primarily

twofold. First, the inception

and early development

molecular evolution (reviewed in 18, 63, 70) largely coincided with (and

stimulated) the rise of the neutralist school of thought (58, 59), which

challenged a common view that genetic variability

was molded primarily by

natural selection. Thus, molecular evolutionists were (justifiably) preoccu-

pied with the selectionist-neutralist

debate and never

gained very close contact

with various

schools of systematic thought

that were also growing actively at

that time (55, 81). Second, in the early years protein electrophoresis

was the

only molecular

technique readily applicable

to comparisons

at the intraspecif-

ic level. Yet allozymes of a particular

locus are qualitative,

multistate traits

the phylogenetic

order of which cannot be safely inferred from the observable

property, electrophoretic mobility. Furthermore, allozymes are encoded by

nuclear

genes that segregate

and recombine

during

each generation

of sexual

reproduction.

These attributes of allozymes understandably

directed methods

of data

analysis

toward concerns with allele frequencies

and heterozygosities,

which in turn channeled thinking back into the traditional

framework and

language of population genetics and away from phylogeny.

The purpose

of this report

is to make a case that animal

mitochondrial DNA

(mtDNA) (by virtue of its maternal, nonrecombining

mode of inheritance,

rapid pace of evolution, and extensive intraspecific polymorphism) permits

and even demands an extension of phylogenetic thinking

to the microevolu-

492 AVISE

ET AL

tionary level. As such, data from mtDNA can provide a liaison service for

expanding communication

between systematists and population

geneticists.

With empirical

and conceptual

channels opened, it might then be possible to

reconsider various connections between micro- and macroevolutionary

change as interpreted

against a continuous genealogical backdrop. This re-

view will be a success if it stimulates further

dialogue in these areas.

MTDNA-NOT "JUST ANOTHER" MOLECULAR

MARKER

If one were to specify the properties

desired of an ideal molecular system for

phylogenetic analysis, the wish list might include the following. The mole-

cule should: (a) be distinctive, yet ubiquitously

distributed,

so that secure

homologous comparions

could be made among a wide variety of organisms;

(b) be easy to isolate and assay; (c) have a simple genetic structure lacking

complicating features such as repetitive DNA, transposable elements,

pseudogenes, and introns; (d) exhibit a straightforward

mode of genetic

transmission, without recombination or other genetic rearrangements; (e)

provide suites of qualitative character states whose phylogenetic in-

terrelationships

could be inferred

by reasonable

parsimony

criteria; and, for

purposes of microevolutionary

analysis, (t) evolve at a rapid

pace such that

new character

states commonly arise within the lifespan of a species. To a

remarkable

degree, the mitochondrial

DNA of higher

animals

meets all of the

above criteria.

Molecular properties

of animal mtDNA have been reviewed previously

(10, 25, 26), so only a brief synopsis sufficient for current discussion is given

here. The reader is directed

to the earlier

papers

for details and

qualifications.

In higher animals, mtDNA is a small, covalently closed circular molecule,

about 16-20 kilobases long. It is tightly packed

with genes for 13 messenger

RNA's, 2 ribosomal

RNA's, and 22 transfer RNA's. In addition to these 37

genes, an area known as the "D-loop" (in vertebrates and echinoderms)

or "A

+ T-rich" region (in Drosophila), roughly 0.8 kilobases long, appears to

exercise control over mtDNA replication and RNA transcription.

Introns,

repetitive DNA, pseudogenes, and even sizeable spacer sequences between

genes, are all absent. Gene arrangement

appears

very stable, at least within a

taxonomic class or phylum. For example, gene order is identical in assayed

mammals and frogs but differs from that in Drosophila. Nonetheless, evolu-

tion at the nucleotide sequence level is rapid, perhaps

1-10 times faster than

typical single-copy nuclear DNA (28, 92). Most of the genetic changes are

simple base substitutions;

some are small addition/deletions

(one or a few

nucleotides); and fewer still involve large length differences (up to several

hundred nucleotides). The size differences are usually (though not ex-

PRINCIPLES

OF mtDNA PHYLOGEOGRAPHY 493

clusively; 69) confined to the control region

of the molecule, which in general

is evolving especially rapidly. The final and perhaps

most important

point is

that, to the best of current

knowledge (50, 60), inheritance

of animal mtDNA

is strictly

maternal.

Thus, unlike the situation

for nuclear

DNA, the mtDNA

mutations

arising in different individuals

are not recombined

during sexual

reproduction.

No molecular system is likely to be perfect for phylogenetic analysis, and

mtDNA does have some potential

and real limitations that need to be recog-

nized:

Heteroplasmy

Most somatic cells (and mature

oocytes) contain hundreds or thousands of

mtDNA molecules, so that

at its inception

a new mutation

will either

generate

or add to a heteroplasmic

condition in which two or more genotypes coexist

within an individual. On theoretical grounds, it was originally feared that

heteroplasmy

might be extensive and hopelessly complicate mtDNA study,

but empirical

experience

proved

this worry

to be unjustified.

Cases of hetero-

plasmy have been discovered (20 and references

therein)

but are unusual

and

therefore of little impact in routine surveys of animal mtDNA. Current

thinking is that mutations

within a cell line (as opposed to paternal

leakage

of mtDNA via sperm) generate most instances of heteroplasmy, and that

the heteroplasmic

state is quite transitory,

due to rapid sorting of mtDNA

molecules in germcell lineages (34, 52, 76, 86). Thus, as phrased by

Wilson et al (97), "The vast majority of individuals tested seem effec-

tively haploid as regards the number of types of mtDNA transmitted

the next generation (although polyploid as regards the number

of mtDNA

copies per cell)."

Homoplasy

An ideal phylogenetic

marker

would be free from reversals

as well as parallel

or convergent

evolutionary

change (homoplasy).

In one respect, mtDNA falls

short of this standard-many restriction sites have been observed to "blink"

on and off repeatedly

during evolution (e.g. 43, 61). This phenomenon

presumably

most often attributable

to recurrent transitional base substitutions

(3) at some nucleotide sites. If particular

positions in the mtDNA genome are

considered "characters,"

and if evolutionary change at these positions has

been especially rapid with respect to the time since separation

of assayed

lineages, then the small number of alternative character states assumable

insures that some homoplasious changes will have occurred. Nonetheless,

because mtDNA genomes are nonrecombining,

the entire molecule can jus-

tifiably be considered

the "character,"

in which case the number

of possible

character

states becomes astronomical.

494 AVISE

ET AL

Typical empirical surveys of mtDNA (see beyond) effectively involve

assay of at least several hundred

base-pairs

of information per individual.

When viewed this way, any widespread

and intricate

similarities

present in

mtDNA are most unlikely

to have arisen by convergent

evolution, and so they

must primarily

reflect phylogenetic

descent (or, conversely stated, any wide-

spread and intricate

differences observed

among mtDNA molecules could not

be overcome by wholesale convergent mutation).

The effects of homoplasious

change in mtDNA are thus probably

limited to introduction of circumscribed

ambiguity in tree or network

placements

of mtDNA genotypes. Furthermore,

approaches for recognizing and treating homoplasy in mtDNA have been

suggested (10, 89, 90).

Scale

Some nucleotide positions in mtDNA are far more labile evolutionarily

than

are others, presumably

due to relaxed selective constraints

(4). The initial

rapid pace of mtDNA differentiation

(estimated at about 2% sequence di-

vergence per million years in mammals; 28) is attributable primarily to

changes at these sites, after which further mtDNA differences accumulate

much more slowly. The overall effect is that beyond perhaps about 8-10

million yr, a plot of mtDNA nucleotide

sequence

divergence (p) against

time

(t) becomes curvilinear, eventually reaching

a plateau

where estimation

of t

from p is pointless (28). For this reason, unless special precautions

are taken

to work

only with more slowly evolving portions

of the molecule, meaningful

phylogenetic comparisons

from conventional mtDNA surveys will normally

be confined to conspecific populations and closely related species whose

separations

date to within the last few million years.

At the other end of the scale, for very recently disjoined populations

species, it is likely that a substantial fraction of observed mtDNA sequence

differences arose prior

to population separation (i.e. they represent

retention

of polymorphisms originally

present

in ancestral

parental

stock). There are at

least two ways to deal with this potential complication.

First, from a popula-

tion genetic perspective, statistical corrections

can be applied (72, 97). For

example, let Ax, Sy, 8A, and 8xy represent

the mean pairwise mtDNA di-

vergence values between individuals

of population

X, of population Y, of the

ancestral population, and between individuals in population X versus Y,

respectively. Although 8A cannot be observed

directly, it can be estimated

assuming that

SA = 0.5 (&x + Sy)-

Then the corrected

distance estimate between populations

X and Y becomes

8 = 8XY -

PRINCIPLES OF mtDNA PHYLOGEOGRAPHY 495

A second way to deal with the predicament

involves a shift to a phylogenetic

perspective. Since mtDNA genotypes in different lines do not recombine,

individual organisms (rather

than populations

or species) can justifiably be

considered as the basic operational

taxonomic

units

(OTU's) in a phylogenetic

reconstruction (62). This straightforward approach,

in which individuals

(or,

more precisely, their mtDNA genotypes) constitute the tips of hypothesized

evolutionary trees (or nodes of evolutionary networks), can be especially

informative.

For example, with respect

to matriarchal

phylogeny, it is biolog-

ically quite plausible

that some individuals

may truly

be more closely related

to members

of another

species than

they are to conspecifics, owing solely to

particular patterns

of maternal

lineage survival and extinction accompanying

the speciation process (73, 85; also see below).

Selection Versus

Neutrality

The longstanding

debate about whether the dynamics

of genetic variation are

governed primarily

by natural selection or by genetic drift of neutral

muta-

tions, can also be extended

to mtDNA. In our

view, the phylogenetic

value of

mtDNA does not, however, completely hinge on the outcome. Thus, even if

mtDNA genotypes prove commonly to differ with respect

to fitness, properly

identified synapomorphic

(shared-derived)

character states should still permit

recognition

of monophyletic assemblages (clades)

of molecules. Nonetheless,

in some kinds of data analyses involving genetic distance estimates and

molecular clock concepts to date separation

events, it would be especially

important

to know whether mtDNA variability

is neutral

(although, particu-

larly when longer spans

of time are involved and much genetic information is

assayed, the magnitude of genetic differentiation

under some models of

natural selection should also be well-correlated

with time; 9, 44).

Two senses in which mtDNA variability

might be deemed neutral need to

be carefully distinguished.

First, in a mechanistic

sense, we already

"know"

that most of the particular

mtDNA genotypic variants segregating in pop-

ulations probably

have, by themselves, absolutely no differential effect on

organismal fitness. These include, for example, base substitutions in silent

positions of protein-coding

genes, and some substitutions and small addition/

deletions in the nontranscribed

D-loop region. These changes are dis-

proportionately

common in mtDNA (25) and

are

ones for which only the most

ardent selectionist would argue a direct link to organismal

fitness. On the

other hand, mtDNA contains genes whose products

(usually in collaboration

with those of the nuclear

genome) are crucial to production

of energy neces-

sary for animal survival

and reproduction (49, 74). Some mtDNA mutations

must, then, be highly visible to selection. When they arise, each such

mutation will by chance be associated with a particular array

of mechanistical-

ly neutral

variants elsewhere in the molecule. Since mtDNA is maternally

496 AVISE ET AL

inherited, these associations will not be dissoluted by recombination (95). In

this second, dynamic sense, mechanistically neutral

mtDNA variants

may,

through linkage to selected mtDNA mutations, have evolutionary

histories

that are at times influenced or even dominated by effects of natural

selection.

A deeper

understanding

of such possibilities

poses a stiff challenge for future

study. Not only will knowledge

be required

of the periodicity

and intensity of

selection on fitness-related mtDNA mutations, but historical accidents of

association with neutral markers will have to be taken into account. Further-

more, all this action must be understood within the context of ever-changing

nuclear

gene backgrounds

whose epistatic interactions with mtDNA are

likely

to be of great importance

(5, 25, 49, 78).

Lineage Sampling

Bias

The phylogenies inferred from mtDNA comparisons

represent

the presumed

historical sequences of mutational events accompanying the differentiation of

maternal lines. An mtDNA phylogeny is thus an example of a molecular

genealogy-a record

of evolutionary

changes in a piece of DNA, in this case

one that has a history of maternal

transmission. In general, any organismal

phylogeny must in some sense represent a composite attribute of many

molecular

genealogies, including

those for all nuclear

genes, each of which in

any generation

could have been transmitted

through

male or female parents.

As phrased by Wainscoat

(93), "We inherit our mitochondrial DNA from

just

one of our sixteen great-great

grandparents, yet this maternal ancestor has

only contributed one-sixteenth

of our nuclear DNA." The asexual, maternal

transmission

of mtDNA is thus a double-edged

sword. Although

the informa-

tion recorded

in mtDNA represents

only one of many molecular

tracings in

the evolutionary

histories of organisms,

it is nonetheless a specified genealog-

ical history (female -> female -> female), and one whose molecular record

has not been complicated by the effects of recombination.

INTRASPECIFIC PHYLOGENY

AND GEOGRAPHIC

POPULATION

STRUCTURE

Most mtDNA surveys of natural

populations

have involved the technically

expedient restriction

enzyme approach.

MtDNA is isolated from individual

animals, digested with particular

endonucleases, and the resulting digestion

products separated

by molecular weight through gels. The "raw" data then

consist of restriction fragment digestion profiles on gels, or with some

additional

effort, restriction site maps. The evolutionary

changes

in restriction

sites underlying the differing digestion profiles or site maps can often be

inferred

simply, and a parsimony

network

summarizing

the presumed

history

of genotypic interconversions can be generated. A straightforward

example

PRINCIPLES OF mtDNA

PHYLOGEOGRAPHY 497

involving BstEII

sites observed

in the mtDNA from

Peromyscus

maniculatus

(61) is presented

in Figure 2. A typical survey

now often includes data from

ten or more enzymes and involves, on average, 40-100 or more restriction

sites per individual. The recognition sequence

of each employed endonucle-

ase is either four, five, or six base-pairs

in length, so a routine survey of

mtDNA would effectively screen

individuals for genetic

differences at several

hundred

nucleotide positions.

To exemplify more fully the kinds of phylogenetic implications

inherent

such data, we briefly summarize

results from a typical natural

population

survey. Bermingham

& Avise (19) used 13 restriction endonucleases

to score

an average of 54 sites per individual

in the mtDNA of 75 bowfin fish (Amia

calva) collected from river drainages

from South Carolina

to Mississippi. A

total of 13 distinct mtDNA genotypes (which for simplicity can be called

"clones")

were observed. Figure

3A shows a hand-drawn

parsimony

network

(constructed by an extension of the approach exemplified in Figure 2) in-

terconnecting these clonal genotypes, and in Figure 3B this network is

superimposed over the geographic sources of the collections. Two major

genetic (and geographic) assemblages of mtDNA clones are apparent-an

eastern assemblage of nine related clones observed in bowfin from South

Carolina, Georgia, and Florida; and a western assemblage of four related

A B C E F

i0.2_

5 = __ -

5.1 -_

4.1 -

3.0-

2.0 -

1.6-

1.0- _

0.5 -

Figure 2 (left) Diagrammatic representation

of the five BstElI digestion profiles observed

among mtDNA's isolated from samples of Peromyscus maniculatus (61). The leftmost lane

shows selected sizes (in kilobases) of fragments

in a molecular

weight standard.

(right)

Restric-

tion site maps (obtained from double-digestion procedures) corresponding

to the fragment

profiles

on the left. These site maps

have been interconnected

into a parsimony

network

reflecting

probable evolutionary relationships

among the BstEII

patterns.

Arrows indicate direction

of site

loss and not necessarily direction

of evolution.

498 AVISE

ET AL

clones in bowfin from Alabama and

Mississippi.

At least four assayed restric-

tion site changes distinguish any eastern

from any western genotype.

As drawn, these parsimony

networks are unrooted,

but additional

hypoth-

eses about phylogenetic orientation

can be advanced. By several criteria,

mtDNA clone 1 is a likely candidate

for the ancestral genotype within the

eastern assemblage

of Amia calva: (a) It is by far the most common eastern

genotype, occurring

in 30 of 59 assayed specimens; (b) it is geographically

the most widespread, observed in nine of the ten eastern river drainages

surveyed; and (c) in the parsimony

analysis, it forms the hub of a network

whose spokes connect separately

to seven other mtDNA genotypes (Figure

3A). Clone 1 is also at least one mutation step closer to the western mtDNA

Figure 3 Phylogenetic networks and phenograms summarizing evolutionary relationships

among 13 mtDNA genotypes observed in a sample of 75 bowfin fish, Amia calva (19). (A)

Hand-drawn

parsimony

network.

Slashes

crossing

branches indicate restriction site changes

along

a path;

heavier lines encompass

2 major arrays

of mtDNA genotypes

distinguishable by at least 4

restriction site changes. (B) The parsimony

network in A superimposed

over the geographic

sources of collections. (C) Wagner parsimony

network computer generated

from a presence-

absence site matrix. Inferred restriction site changes are indicated,

and numbers

in the network

represent

levels of statistical

support

(by bootstrapping)

for various clades. (D) UPGMA pheno-

gram, where p is estimated nucleotide sequence divergence.

PRINCIPLES

OF mtDNA PHYLOGEOGRAPHY 499

CLONE LOCALE

4 j COOPER

APALACHICOLA

\ ~ ~~~~ 6

se 9

< g | ] ESCAMBIA

40 12 MISSISSIPPI

CLONE LOCALE

4 COOPER

5 to

3 APALACHICOLA

ESCAMBIA

12 MISSISSIPPI

0.010 0.008 0.006 0.004 0.002 0.000

clade than are any other mtDNA genotypes in the east. In the western

genotypic array, clones 10 and 12, which occur in the drainages most

proximate

to those in the east (Figure

3B), are genetically closest to clone 1

(each differs by four assayed mutation

steps); and fish in the most westerly

drainage show further

distinction from these clones (Figures 3A and 3B).

Some data sets are far too large for such easy analysis by hand, and

computer

assistance is required.

Several tree-building

software packages are

500 AVISE

ET AL

available (42); we routinely employ various

algorithms

in the PHYLIP pack-

age distributed

by Joe Felsenstein. For example, Figure 3C shows a Wagner

parsimony network (from the METRO annealing algorithm

in PHYLIP) of

mtDNA genotypes in bowfin fish generated from a matrix consisting of

presence-absence information for each restriction site in each mtDNA clone.

Particular site changes along various branches of the network

are shown, and

numbers indicate the levels of statistical support (the proportion of times that a

group was distinguished

in a bootstrap analysis; 41) for a given hypothesized

mtDNA clade.

It is also possible to convert mtDNA fragment

or site data into estimates of

nucleotide sequence divergence (p) between genotypes (e.g. 72, 91), and the

resulting distance matrixes can provide the basis for tree or phenogram

construction. Figure 3D shows a UPGMA phenogram (81) for the mtDNA

clones in Amia calva. The eastern versus western clonal assemblages are

again apparent

and differ in nucleotide

sequence by an average

of about 1%.

In general then, many qualitative

and quantitative

methods of tree construc-

tion can be applied to mtDNA data. It is beyond the scope of this review to

address

the ongoing debate about "best" methods for phylogeny reconstruc-

tion (and Avise's views have been presented elsewhere; 6). Suffice it to say

that in our experience, tree-constructing algorithms involving philosophically

distinct methodologies usually produce very similar outcomes when applied

to a given set of mtDNA data. The pictured

networks

and phenograms

for

Amia calva (Figure 3) are merely a case in point.

In our laboratory,

similar surveys of geographic

variation

in mtDNA have

now been completed or are in progress for about 20 species, including

mammals, birds, reptiles, amphibians,

marine and freshwater

fishes, and an

invertebrate (the horseshoe crab). The remainder

of this section summarizes

major

features

of these data (Table 1) in the context of qualitative patterns

geographic population

structure. The original papers

should be consulted for

details.

In principle, intraspecific phylogenies overlaid

on geographic maps could

yield many

kinds of outcomes. Five major categories

of possibilities and their

provisional interpretations

are summarized in Figure 4. For example, an

mtDNA phylogeny itself could show discontinuities

(or genetic "breaks")

which arrays of related genotypes differ from other such arrays by many

mutational

steps. Such genetically

distinct mtDNA assemblages might occupy

separate geographic regions within the range

of a species (category I, Figure

4), or they could co-occur geographically (category II). Alternatively,

mtDNA phylogenies themselves might be more or less continuous

genetical-

ly, and spatially either disjunct (category III), totally overlapping (category

IV), or nested (category V). We have empirical examples approximating

almost all of these theoretical

outcomes.

PRINCIPLES

OF mtDNA PHYLOGEOGRAPHY 501

ZW~~~~

LU0 U In O E > o S W Z

2 2 ~ 2

ZLJ N , CL

z +

o >-

z ~~~~~~~~~~~~~

+~~z fa

x U Z N

N z I-~~~~~~~~~~~~~~~w

z )C

5 Z + I~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~Li

o ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~C

1cr -i 4-

LL ?

L'i U) U) - U) -i~U A) U)

0' U 0

HJ 0 0 c

uL' C

C.) C.) I '

U) U)z 0 0

z~~~~~~~~~~~~~~~~~~~~~~'

0 >

(9 CD

LU x7:

1-r

o~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~i

502 AVISE ET AL

Category I-Phylogenetic Discontinuities, Spatial Separation

In our experience, this is the most commonly

encountered

situation. It applies

for example, to the Amia calva case history

already

detailed

above, in which

arrays

of related mtDNA genotypes occurred in eastern

versus western river

drainages in the southeastern United States, and the two arrays were dif-

ferentiable by at least four assayed mutation steps (and p -0.01). The

magnitudes of genetic breaks

distinguishing

populations from different

geo-

graphic regions are in fact often considerably greater

than that observed in

Amia calva. For example, in the redear sunfish

Lepomis

microlophus, which

was sampled from the same river drainages

as Amia, eastern

versus western

arrays

of mtDNA genotypes differed

by 17 or more

assayed

mutation

steps (p

=0.09), yet maximum differentiation within either the eastern or western

mtDNA assemblages was always less than

p = 0.007 (19). Other species in

which we have observed discontinuous intraspecific

mtDNA phylogenetic

networks, with a strong geographic orientation, include the pocket gopher

(Geomys pinetis), deer mouse (Peromyscus maniculatus), bluegill sunfish

(Lepomis

macrochirus), spotted

sunfish

(L. punctatus),

warmouth

sunfish

(L.

gulosus), mudpuppy

salamanders

(Necturus

alabamensis and relatives), des-

ert tortoise (Scaptochelys agassizii), and horseshoe crab (Limulus

polyphe-

mus). References and relevant data from these studies are summarized

Table 1.

The most likely explanation

for major genetic discontinuities that display

geographic orientation involves long-term, extrinsic (i.e. zoogeographic)

barriers to gene flow, such that conspecific populations occupy easily

recognizable branches on an intraspecific

evolutionary

tree. Another

possibility, not mutually

exclusive, is extinction of intermediate

genotypes in

widely distributed

species with limited dispersal

and gene flow capabilities.

Apart

from the mtDNA

phylogeographic patterns

per se, is there

additional

support

for the significance of historical

zoogeography

in shaping intraspecif-

ic genetic architectures? At least two empirical lines of evidence can be

advanced.

First, populations

separated

for long times by zoogeographic

barri-

ers should also accumulate differences in the nuclear

genome. Few studies

have assayed nuclear

genes (or their products)

in concert with mtDNA, but

two that have done so-involving the pocket

gopher, Geomys

pinetis (13) and

the bluegill sunfish, Lepomis macrochirus (12)-found dramatic

allozyme

frequency distinctions between major mtDNA phylogenetic groups (Figure

5). Second, strong

biogeographic

barriers

should mould the genetic structures

of independently evolving species in concordant fashion. Five species of

freshwater fishes have been assayed for mtDNA differentiation across river

drainages in the southeastern United States-and, remarkably,

all showed

strong patterns of congruence in the geographic placements of the major

mtDNA phylogenetic breaks (Figures 3, 5, and 6). To account for these

PRINCIPLES

OF mtDNA PHYLOGEOGRAPHY 503

Z0)

C)0

0 Q N 0 00

00t

o oo. 00~~~~~~~~~~~~~~~~~~~~~~oc

CO~~~~~~~~~~~~~~~~~~~~~~~~~~~C

.0~~~~~~~~~~~~~~~~~~

CO~~~~~~~~~~~~~'

o~~~~

0 (L)~~~~~~~~~~~0

00 0

0)~~~~

.U 0) 0

0 0

00 Z~~~~~~~~~0

0)~~~~~~~~~~~~~~~~~0

0~~~~~~~~~~~~~~~ 0

En r.0)0 ,

0 0~~~~~~~~0C

- _ 0)~~ ~

504 AVISE ET AL

124 E 72

0 42

00 cis

4.) cis

O O

0 O O 0 'IC -

0 O O O O O O O

0 CZ -

0 0

Cld

s-4 11

cis

0 Z -cl O 00

u cis cis

r. cis cis

cis m O m

z 0

cis cis Ic; - to

7:1 cis

cis s.. = 0 cis

0 cis 9

U O u

to 0 0 cis LL,

0 0 L14 0 0 w

4.) -a w 0

4- cis

0 O 0 0 0 0

Z Z

tn 00 C14 tn

z A

to 4.4

0 0

E 0

40-4

14)

-O 0 E qj--

0 0 Z +-, = Ci 0

t4... 0

PRINCIPLES OF mtDNA PHYLOGEOGRAPHY 505

0 co

0 co co I" r- r-

0 ci 6 ci

> >

C14

.14

O 0

ed u 0

Au 0

C5 C5

co u

'4-J

z z z

.14

0 0

U r-

en 00 en

0 0

.,CJ

Ln p.,

5! e

OZ E = E Cb Ow

506 AVISE ET AL

results, we advanced

a detailed biogeographic

reconstruction-one that im-

plicates historical patterns of river drainage isolation and coalescence associ-

ated with Pliocene and Pleistocene changes in sea level (19).

Preliminary evidence suggests that intraspecific phylogeographic dis-

continuities in mtDNA may commonly align with the boundaries of

zoogeographic provinces as identified by more conventional

biogeographic

data. For example, from lists of distributional limits of freshwater fish

species, Swift et al (84) identified two major zoogeographic provinces (east

versus west of the Apalachicola River), plus additional

subprovinces,

in the

southeastern United States. Boundary

zones between these regions agreed

quite well with the concentrations

of intraspecific phylogenetic breaks in

Geomys pinetis

mtDNA

Lepomi s m

ac roc hi rus

0 00 00 ;

Got-2 0t

DNA

Figure 5 Empirical examples in which highly divergent

mtDNA phylogeographic groupings

also proved distinct in allozyme frequencies. Above: Data for southeastern

pocket gopher,

Geomys pinetis (13). On the left are pie diagrams summarizing geographic

distributions

(in three

southern

states) of the two electromorphs (labeled

"95" and "100")

of the albumin locus. On the

right is an mtDNA phylogenetic network,

the most dramatic feature of which is the large genetic

gap (p 0.034) distinguishing

the same arrays

of eastern versus western

samples. Below: Data for

the bluegill sunfish, Lepomis

macrochirus. On the left are

pie diagrams summarizing geographic

distributions

of electromorphs (labeled

"100"

and

"58")

of the Got-2 nuclear locus (from 11). On

the right are pie diagrams

of frequencies

of two highly distinct (p = 0.085) mtDNA genotypes

(from 12).

PRINCIPLES OF mtDNA PHYLOGEOGRAPHY 507

Figure 6 Geographic distributions of major mtDNA clades in three additional species of

sunfish, Lepomis (from 19). Within each species, major

mtDNA

phylogenetic

breaks

(,a

= 0.062,

0.063, and 0.087 for L. punctatus,

L. gulosus, and

L. microlophus, respectively) distinguish

fish

in eastern rivers from those in drainages

further

to the west.

mtDNA for the five widely distributed

fish species assayed by Bermingham

Avise (19). In another example, a genetic discontinuity between two

phylogenetic assemblages of mtDNA's in the coastal horseshoe crab (77)

occurred near Cape Canaveral, Florida, a region long-recognized as tran-

sitional between warm-temperate

and tropical marine faunas (1, 22).

Recognizable biogeographic provinces presumably

exist because of environ-

mental

impediments (ecological and/or

physical;

historical as well as contem-

508 AVISE

ET AL

porary) to dispersal and gene flow. These impediments

are conventionally

recognized as reflected in concentrations of distributional

limits for many

species; perhaps

they may also be reflected

in concentrations of intraspecific

phylogenetic discontinuities

within species that

have geographic

distributions

extending across zoogeographic provinces.

Category Il-Phylogenetic Discontinuities, Lack of Spatial

Separation

In surveys from our laboratory,

we have no good empirical

examples of the

situation diagrammed

in category II, Figure 4-mtDNA phylogenetic dis-

continuities not associated with spatial separation.

Indeed, it has even been

rare to observe large mtDNA differences (i.e. greater than about 1-2%

nucleotide

sequence divergence)

between conspecific individuals

collected at

any given geographic

site. One example involved

the deer

mouse

Peromyscus

maniculatus (61), in which, for unknown

reasons, two or more moderately

divergent

mtDNA clones were occasionally found within particular localities

in the eastern United States. For example, two mtDNA clones collected in

Giles County, Virginia, differed by five restriction

sites changes (in assays

with eight endonucleases) and an estimated sequence divergence of p-

0.013. Even in P. maniculatus, however, the largest genetic differences in

mtDNA (p values greater

than about

0.03) were invariably between

mice from

different regions of North America (61), so that the overall pattern is more

consistent with category I in Figure 4.

In a large sample

of bluegill sunfish

(Lepomis

macrochirus) collected from

Lake Oglethorpe

in north

Georgia,

two grossly different mtDNA genotypes (p

= 0.085) did co-occur in roughly equal frequency

(12). Further

analysis of

this situation, involving more extensive geographic sampling as well as

comparisons

with nuclear

genotypes (1 1), revealed that the Lake Oglethorpe

population

is probably

a random-mating, hybrid

swarm

arising

from second-

ary

contact

between

allopatrically

evolved races

of bluegill. In such secondary

admixture zones (as well as in cases where reproductively

isolated sibling

species are inadvertently assayed as if belonging

to a single species), mtDNA

phylogenetic

discontinuities

in the absence

of current

spatial separation

are

course to be expected.

Category II-Phylogenetic Continuity,

Spatial Separation

Not all assayed species have exhibited the large mtDNA phylogeographic

"breaks" characteristic

of category I. Another

commonly encountered situa-

tion is one in which mtDNA parsimony

networks

are more

or less continuous,

with consistently small numbers

of mutational

steps (and fairly low p values)

between phylogenetically adjacent

clones, each of which is nonetheless

con-

fined to a subset of the geographic

range

of the species (III, Figure 4). Such a

situation is approximated

in the marine oyster toadfish Opsanus tau (16).

PRINCIPLES OF mtDNA PHYLOGEOGRAPHY 509

Among 43 individuals

sampled from Massachusetts

to Georgia, five closely

related but geographically localized mtDNA genotypes were observed, each

differing from its apparent closest relative by only one or two assayed

restriction sites (and associated p < 0.008). The two most common

genotypes

were respectively confined to collections north versus south of the Cape

Hatteras area in North Carolina (another boundary region between

zoogeographic provinces (22). Yet mean sequence divergence between the

five mtDNA genotypes in 0. tau was only p -0.005. Other

species in which

we have observed limited differentiation yet geographic localization of

mtDNA clones include the gulf toadfish (Opsanus beta), diamondback ter-

rapin (Malaclemys terrapin), and old-field mouse (Peromyscus

polionotus)

(Table 1). This phylogeographic pattern

is also characteristic

of the differenti-

ation observed within particular regions for most of the category-I species

previously listed (in other words, category III is similar

to the within-region

pattern

of category I-Figure 4).

The most likely explanation

for geographic

localization

of mtDNA clones

and clades, in the absence of major phylogenetic breaks,

involves historically

limited gene flow between populations in species not subdivided by firm

long-term zoogeographic barriers

to dispersal. Thus, recently arisen muta-

tions are confined to subsets

of the species' range, and the overall population

structure may conform more or less to either

the "island" or "stepping

stone"

models in traditional

population genetics (54).

Category IV-Phylogenetic Continuity,

Lack of Spatial

Separation

Within a few species, closely related mtDNA genotypes appear not to be

geographically localized. Perhaps

the best example involves the American

eel, Anguilla rostrata (14). In 109 eels taken from seven locales between

Maine and

Louisiana,

numerous related

mtDNA genotypes

were detected, yet

each (when present in two or more individuals) was geographically wide-

spread. Similarly, in the hardhead sea catfish (Ariusfelis), two related

clades

of mtDNA genotypes ( -0.006) were both widely distributed

along the

South Atlantic and Gulf of Mexico coastlines (16). Other assayed species

exhibiting limited mtDNA phylogenetic diversity and relatively little geo-

graphic structure include the marine

gafftopsail

catfish (Bagre marinus), the

red-winged blackbird (Agelaius phoeniceus) (Table 1), and, to an argued

extent, humans (24, 94).

We propose

that

geographic populations

of species exhibiting

this category

of intraspecific phylogeography have had relatively extensive and recent

historical

interconnections

through gene flow. This would require the absence

of firm and longstanding zoogeographic

barriers

to movement, as well as life

histories conducive to dispersal. All of the above examples can at least

provisionally be understood in these terms. American eels have a

510 AVISE ET AL

catadromous life history-mass spawning takes place in the western tropical

mid-Atlantic

Ocean, and larvae

are

transported (perhaps

passively) to coastal

regions by ocean currents. Young eels mature

in freshwater

before completing

the life cycle by migrating

back to the mid-Atlantic for spawning. Thus, any

freshwater

population

in the Americas may contain a nearly

random draw of

genotypes from what is effectively a single mating pool. The eel life-history

pattern

is highly unusual, but other marine

fishes, such as the marine

catfish

which are active swimmers

as adults, may also prove

to exhibit the "category-

IV"

phylogeographic pattern (8, 16, 48). Marine fish occupy a realm relative-

ly free of solid geographic

barriers

to dispersal (at least over major portions

their ranges and in comparison

to those of their freshwater

counterparts that

are necessarily confined to specific drainages

for moderate

lengths of evolu-

tionary time). Many marine fishes also possess great dispersal capabilities,

either as pelagic larvae, juveniles, and/or adults.

Birds constitute another group of potentially highly mobile animals for

which the "category-IV"

phylogeographic pattern may prove to be common.

For example, populations

of the red-winged blackbird,

Agelaius phoeniceus,

(the only avian species extensively assayed at the time of this writing)

exhibited very limited mtDNA phylogeographic

structure across all of North

America (Table 1). Red-winged blackbirds are known to be moderately

nest-site philopatric (average distances between banding and recovery at

nesting sites in successive years are generally less than 50 km; 37), so the

documentation

of a "category-IV"

pattern clearly cannot be taken to imply

panmixia or even long-distance gene flow on a generation-to-generation

scale. Rather, we suspect that these blackbirds (and other species in

phylogeographic category IV) have had a relative fluidity of movement (in

birds, obviously facilitated

by the capacity

for flight) over a recent

evolution-

ary

time scale such that

populations

have been in solid genetic contact

within,

perhaps, the last few tens of thousands of generations (see next section).

For humans, assays of mtDNA from individuals of diverse racial and

geographic

origin revealed

only a weak tendency

for phylogenetic

structuring

of groups, according

to Cann

et al (33; see also 24, 27, 30, 31, 32). Based on

a conventional mtDNA clock calibration of 2% nucleotide sequence di-

vergence per million years, Cann et al (31) proposed a mean interracial

divergence

time in humans

of about

50,000 years, and Brown (24) and Cann

et al (33) hypothesized a common (female) ancestor for all humans about

200,000 years ago. Cann et al -(33)

also argue

from the mtDNA data that

this

female ancestor lived in Africa. For any species whose numbers and ranges

have expanded dramatically

from a single refugium or place of origin in

recent evolutionary times, mtDNA phylogeographic differentiation should

similarly be quite limited.

Using independently

obtained data, Johnson et al (57) report a greater

degree of geographic

and racial clustering

of human

mtDNA genotypes than

PRINCIPLES

OF mtDNA

PHYLOGEOGRAPHY 511

did the Cann et al (31, 32, 33) research group. MtDNA genotypes

thought to

be ancestral for humans were geographically

and racially widespread, but

genotypes presumed

derived

were often race specific. Thus, according

to the

Johnson et al data (57; see also 94), humans are better characterized

exhibiting the category-V phylogeographic

pattern (see below).

Category V-Phylogenetic Continuity,

Partial Spatial

Separation

The four phylogeographic

categories listed above are of course somewhat

arbitrarily

selected though distinct points from a wide field of possibilities.

We include

category

V here and

in Figure

4 only to provide

example

of one

type of intermediate situation. In this category, some mtDNA genotypes are

geographically

widespread,

while allied genotypes

are

localized, such that the

overall pattern is one of a nested series of phylogeographic

relationships.

Besides humans, we have already mentioned one other example. In the

eastern

mtDNA clonal assemblage

of the bowfin fish Amia calva (Figure 3),

mtDNA genotype 1 was present

in nine of ten surveyed

river

drainages, while

each of eight other genotypes was apparently

confined to one or a few

adjacent

drainages

within the range of genotype 1. A reasonable

hypothesis

for this eastern assemblage is that genotype 1 is plesiomorphic

(ancestral),

while the other genotypes are apomorphic

(derived). Individual

fish sharing

the derived states

(i.e. possessing synapomorphic

mtDNA traits)

form

various

monophyletic

groups (with respect

to maternal

ancestry). But because of the

possibility of joint retention of the ancestral

condition, individuals sharing

genotype 1 do not necessarily

form a clade within the eastern

assemblage

bowfin (although

compared

to bowfin in the western

drainages,

they may still

form a broader

clade including all eastern genotypes).

Phylogeographic

category V might be anticipated

in species or subsets of

species with historically

intermediate levels of gene flow between geographic

populations. Thus, unlike category III, presumed

ancestral

genotypes occur

over a broad

area;

while unlike category

IV, newly arisen mutations have not

yet spread throughout

the range of a species. Nonetheless, in practice this

intermediate

situation may normally

be difficult to distinguish clearly from

categories III or IV, respectively (Table 1).

MTDNA EVOLUTIONARY

TREES ARE SELF-PRUNING

mtDNA transmission

is the female analogue

of "male surname transmission"

in many human societies: Progeny of both sexes inherit mitochondria

from

their mothers, but only daughters

subsequently

transmit

mtDNA to future

generations.

Thus, mtDNA (and surnames)

are examples of asexually trans-

mitted traits

within

otherwise

sexually reproducing

species. Realistic statistic-

al models of mtDNA evolution must accommodate this mode of inheritance;

512 AVISE ET AL

they must also somehow account for the empirical

rarity

of major mtDNA

phylogenetic gaps within local populations

(category II, Figure 4), and the

common occurrence

of monophyletic

groupings among

allopatric

populations

(categories I and III, Figure 4). The method of applying

generating

functions

to the distributions

of family size in a branching process (51) is a relevant

probabilistic approach that has been used to study the dynamics of surnames

(65, 66, 82) as well as mtDNA lineages (15, 73).

Assume, for example, that adult females within a population produce

daughters

according

to a Poisson distribution

with mean

,. The probability of

loss of a given female lineage after one generation

(or the proportion

of such

lineages lost from the population)

is then e- , and

the probability of loss after

G generations is given by the generating function PG = e4X- 1), where x

equals the probability

of loss in the previous generation

(82). (Generating

functions are also available for other parametric

family size distributions,

such as the binomial.) In the Poisson case, if mothers

leave on average one

surviving daughter, about 37% of the maternal

lineages will by chance go

extinct in the first generation, and less than 2% of the original mothers will

likely have successfully contributed mtDNA molecules to the population 100

generations later.

Avise et al (15) used an extension

of this approach

to estimate

probabilities

(iT) of survival

of two or more independent

mtDNA lineages through

time. In

the Poisson situation, with ,u (and hence also the variance, v) equal to 1.0,

within about 4n generations

all individuals within a stable-sized population

begun with n females will with high probability

trace maternal

ancestries to a

single foundress

(Figure 7); and

the times in generations

to intermediate levels

of iT are roughly n to 2n. That is, all mtDNA sequence differences would

almost certainly

have arisen less than 4n generations earlier

and more prob-

ably within n to 2n generations.

Lineage sorting

can be much

rapid

than

this when the variance in progeny numbers across females is greater. For

example, in computer

simulations

where females produced

daughters

accord-

ing to a negative binomial distribution with , = 1.0 and v = 5.0, individuals

invariably stemmed from a single female ancestor

less than 2n generations

earlier (Figure 7). The variances in progeny survival between families are

probably large in many species.

In general

then, stochastic mtDNA lineage extinction

within a population

expected to occur at a (counterintuitively)

rapid pace, with the net effect of

continually

truncating

the frequency spectrum

of times to common mtDNA

ancestry. In other words, due simply to the stochastic lineage turnover

associated with the vagaries of reproduction,

mtDNA evolutionary

trees are

continually "self-pruning."

This line of reasoning may largely account for

limited mtDNA sequence divergence values (e.g. p usually much less then

about 0.01) observed within local populations,

or within entire species char-

PRINCIPLES OF mtDNA PHYLOGEOGRAPHY 513

1.0 .0 ;

n1 POISSON

1, l

0.9 \,=20~ LI9O

nz2 ~~~~~~~~~~~~~NEGATIVE

0.8 \ ~~~\08 BINOMIAL

0.7 0d n = 65

' d n n

0.6 n=1 0 w m 0.6 1 t 0 1 5.

05 \n=6 \\n n,0 0 -0r005

0.4\ 0.4-

0.3 03h vo20.0

0.2 \ \ \\ \ \ " 0.2

gaps

wihn sol mayseis(al loipisteoeaino rcse

I 10 lo 102 30 104 105 10 102 103

G G

Figure 7 left: Solid lines are theoretical

probabilities

(,Tr)

of survival

of two or more mtDNA

lineages through

G generations

within populations

founded by n females producing daughters

according

to a Poisson distribution

with mean 1.0. Dashed

lines are conditional

probabilities

(7re)

that two or more lineages survive, given that the population remains

extant. Right:

ie values

within populations

founded by n = 65 females producing daughters according to a negative

binomial distribution

with mean 1.0 and variances (v) ranging from 1.1 to 20.0 (from 15).

acterized by historically high levels of gene flow and/or recent expansion

from a single refugium.

On the other

hand, the empirical

existence of major

mtDNA phylogenetic

gaps within so many species (Table 1) also implies the operation

of processes

acting dramatically

to inhibit extinction of some mtDNA clades, such that

much larger genetic differences than those normally observed within pop-

ulations have had time to accumulate. Since such phylogenetic gaps are

almost invariably

observed between allopatric

populations,

it seems reason-

able that long-term

population

isolation is responsible. Suppose a particular

species has been subdivided

historically

into two or more spatially isolated

populations. Although the genetic distances between lineages within each

population will be limited by the balance between the rate of novel mtDNA

mutations and female lineage extinction, at least one mtDNA lineage per

extant population will be retained

indefinitely, and the distinction

between

these mtDNA lineages could be no less than that which

had

accumulated

since

the time of the original population

separation.

Neigel & Avise (73) have, by computer

simulation

of branching processes,

formalized

these latter ideas and couched them

in the language

of systematics.

(Our models were developed for "species differences"

in mtDNA, but they

apply equally well to expected relationships

between spatially isolated con-

specific populations).

Suppose

that

from

ancestral

stock, populations

A and

have separated

recently (less than about n generations

earlier,

where n is the

514 AVISE

ET AL

carrying capacity of females in each daughter population).

Then because of

stochastic

mtDNA lineage sorting at and subsequent

to population

separation,

it is likely that some individuals within A are in reality more closely related

(i.e. have shared

a female ancestor more recently) to some individuals of B

than they are to other

members

of A; and

conversely, some members of B are

phylogenetically

closer to some A's than to some B's. Populations A and B

could thus be said to be polyphyletic in matriarchal ancestry. However,

through time mtDNA lineage extinction continues inexorably within pop-

ulations, such that after about 3n to 4n generations, populations

A and B

would with high probability

appear monophyletic

with respect

to one another.

At intermediate times of separation

from

ancestral stock, populations

A and

could reasonably

be expected to exhibit a paraphyletic

relationship. Figure 8

plots results of a typical computer simulation summarizing

probabilities

poly-, para-, and monophyly of isolated populations as a function of time

since population separation.

These times were observed under the Poisson

distribution

of family size. For larger

variances

in progeny

numbers (or when

founder effect is severe), expected times to reciprocal

monophyly of pop-

ulations would be even lower. Thus, particularly

for species composed of

many small demes (and limited gene flow between them), allopatric pop-

ulations should

often be monophyletic

in matriarchal

ancestry. Whether

or not

they would appear

to be so in an mtDNA survey might well depend on the

level of laboratory effort expended (i.e. the number of restriction sites

assayed) in the search for synapomorphic character states. The larger

phylogenetic gaps (category I, Figure 4) should occur between demic arrays

1.0 - II_.

p 0.5\ /

N 2N 3N 4N

GEN.

Figure 8 Example of results from computer simulations monitoring probabilities (P) of

monophyly (curve I), polyphyly (II) and paraphyly (III a,b) of two isolated populations G

generations after their separation from ancestral stock (from 73). In each of 400 replicate

simulations, the daughter populations

were founded by 300 and 200 individuals, respectively,

drawn at random from an ancestral stock. The daughter

populations

were subsequently

density

regulated

at carrying capacities

k = 300 and

k = 200 by constraining

the mean number of female

progeny per mother as follows: ,t = e(k-1)/k. N is equal to 300.

PRINCIPLES

OF mtDNA

PHYLOGEOGRAPHY 515

isolated for especially long periods

of time and should be easiest to detect (as

well as most strongly

supported

statistically

by procedures

such as bootstrap-

ping; Figure 3C).

Can these models be related to observed rates of mtDNA evolution and

census population

sizes in more concrete fashion?

From

empirical

experience,

individuals in localized geographic areas, and in entire species in

phylogeographic

category IV, usually show estimates of mtDNA nucleotide

sequence divergence much less than about

p = 0.008. Using Brown et al's

(28) conventional

mtDNA clock calibration,

this implies an upper bound on

times since common female ancestry

of about

400,000 years. This would be

roughly compatible

with expectations

for a population

(or set of populations

well inteconnected

by gene flow) of perhaps

n 400,000 females, provided

the population

has a generation

length of 1 year, has been fairly stable

in size,

and has ,u and v near 1.0. But because branching process theory yields only

probabilistic

outcomes, and because mtDNA lineage survivorship

is likely to

have a large stochastic component, it would not at all be surprising

for that

same population

to have been of size anywhere

from, say, n = 200,000 to

1,000,000 or more (Figure

7). And some very different

demographic

scenar-

ios would not be ruled out. For example, absolute population

size could have

been vastly larger throughout

much of the evolution of the species, but by

chance, two mtDNA lineages dating to 400,000 years ago happened to

squeeze through

more recent

bottlenecks

in population

size. Thus, evolution-

ary reconstructions regarding

population

size and times of ancestry

should be

presented

with due caution; and for most populations, we may never have

direct and detailed knowledge of historical demography

against which to

evaluate possible inferences from present-day

mtDNA diversities.

Nonetheless, unless the rate of mtDNA evolution is anomalously

high in

particular

populations, the major phylogeographic

gaps observed in many

species strongly suggest long times since common female ancestry

for some

conspecifics-much longer than is observed empirically within local pop-

ulations. For example, from the mtDNA data for the eastern

versus western

monophyletic groupings of the redear

sunfish Lepomis microlophus

(Figure

6), mean population

separation

(corrected

for within-region

divergence) oc-

curred about 4 million years ago, and some mtDNA lineages within the

species may date to as much as 5 million years B P (20). Similar values apply

to allopatric

clades within several other species in Table 1.

ECOGEOGRAPHY

AND PHYLOGEOGRAPHY

Data on within-species variability in mtDNA thus lend themselves to ex-

amination

from two vantages: (a) phylogenetic interrelationships

among the

mtDNA molecules themselves and (b) geographic distributions of the

516 AVISE ET AL

phylogenetic groupings. Jointly, these elements constitute concerns of a

discipline that might be termed intraspecific phylogeography.

Notwithstanding

occasional examples of concern with the influence of

historical population subdivision in shaping genetic architecture at the in-

traspecific level (e.g. 2, 75), attention seems more conventionally to have

been focused on possible adaptive

explanations (the "adaptationist paradigm")

for geographic

differences

in attributes such as morphology

or behavior

(47).

One line of evidence for this preoccupation

has been the formulation

several "ecogeographic

rules"

summarizing recognizable

trends

in presumed

adaptive responses to geographically

varying

environmental conditions (23).

For example, Bergmann's

rule notes a tendency in homeotherms for larger

body sizes at higher latitudes (presumably

a surface/volume

adaptation

for

heat conservation

in colder climates);

Allen's rule notes a latitudinal trend

lengths of limbs (shorter extremities may similarly conserve heat in cold

climates); and Gloger's rule notes a tendency

for populations

in humid areas

to be more heavily pigmented (probably a manifestation of selection for

background-matching

to predation

and competition).

While these and

other ecogeographic rules at best represent general trends with many ex-

ceptions (98), they have been provocative

and informative

constructs.

In this

same spirit, we want to suggest several

phylogeographic

hypotheses

that may

serve as a stimulus for further considerations

of geographic trends in in-

traspecific phylogeny.

We take it as axiomatic that the extended pedigree within any species

constitutes

its intraspecific phylogeny and that

genes transmitted

through

this

pedigree can in principle provide genealogical tracings

of hereditary

history.

As emphasized, data from mtDNA allow estimation of one specified com-

ponent of the pedigree-the matriarchal

phylogeny. Thus, the historical

picture recorded in mtDNA is far from a complete characterization of in-

traspecific phylogeny, and that picture may be especially distorted

if males

and females differ in phylogeographically

relevant characteristics,

such as

variances

in progeny numbers

or levels of dispersal (87, 88). Yet techniques

of mtDNA assay have provided

the first extensive and readily

accessible data

in the form of "gene genealogies" at the intraspecific

level. The following

phylogeographic

hypotheses are motivated by the mtDNA data and theory

currently

available and are offered within that context.

Phylogeographic Hypotheses

(a) Most species are composed of geographic populations whose members

occupy different

branches of an intraspecific, phylogenetic tree. Such geo-

graphic partitioning

of phylogenetic

branches can be termed

phylogeographic

population

structure.

The magnitude

of genetic distance between branches

can

range from small to great, but not uncommonly, geographic clades are

distinguished

by large phylogenetic gaps or breaks.

PRINCIPLES OF mtDNA PHYLOGEOGRAPHY 517

(b) Species with limited phylogeographic

population structure have life

histories conducive to dispersal and have occupied ranges free of firm

impediments to gene flow. Such species have had a relative fluidity of

geographic movement over recent evolutionary

time and may be especially

common in certain groups such as flying insects, birds, and marine fishes, or

in species such as the human that have expanded recently from a single

refugium. Genealogical

distances

within such species are constrained because

of the inevitable extinction

of lineages expected within populations behaving

as a single demographic

unit in evolution.

by large phylogenetic gaps usually

arise from long-term

extrinsic (zoogeographic) barriers to gene flow. Since

reproduction leads to a continual turnover

of lineages, isolated populations

should evolve through

time to a condition of reciprocal monophyly, and the

time of isolation

should be positively correlated

(all else being equal) with the

magnitude of genealogical differentiation.

This hypothesis has a series of

corollaries that also serve as predictions for further

empirical tests of the

expectation:

(i) As time since isolation increases, the degree of phYlogeographic

concordance across

separate gene genealogies increases. That is, phylogenetic differentiation between long-

isolated

populations (either

in refugia

or in situ) should be reflected in appropriate assays of

numerous nuclear as well as cytoplasmic genes.

(ii) The geographic placements of phvlogenetic

gaps are concordant across species. That

is, long-term barriers to gene flow should tend to mold the intraspecific genetic

architectures

of species with similar life histories in geographically

concordant fashion.

(iii) Phvlogenetic gaps within species are geographically concordant with boundaries

between traditionally recognized zoogeographic provinces. That is, to the extent that

biogeographic provinces

reflected

in species' distributional limits exist because of environ-

mental barriers to gene flow, such barriers

may also tend to result in geographic con-

centrations

of boundaries

between well-differentiated clades within species.

Whether or not these hypotheses are confirmed with additional

data, we

feel that concern

with intraspecific phylogeography

should assume a place in

evolutionary study at least commensurate with ecogeography. Indeed,

ecogeography will also benefit from this new enterprise. Let us give two

empirical examples. In the deer mouse Peromyscus maniculatus,

mammalo-

gists have recognized two distinct morphotypes-a long-tailed, long-eared

form typically associated

with forest environments,

and a short-tailed,

short-

eared form more characteristic of grasslands

(21). Data from mtDNA clearly

indicate that at least with respect to matriarchal

ancestry, these morphotypes

do not constitute separate

evolutionary

clades (61). The extensive mtDNA

phylogenetic structure

in P. maniculatus across North America is strongly

oriented to geography

and bears no consistent relationship

to these morpho-

logical distinctions. Such findings add support

to earlier

suggestions that the

518 AVISE ET AL

ear and tail length differences

represent

selection-driven

responses to ecologi-

cal challenges posed by forests and grassland

and have arisen more than

once

in separate evolutionary

lines. For a counterexample,

in the bluegill sunfish

Lepomis macrochirus, ichthyologists

have also recognized

two distinct mor-

phological and physiological forms (56). In this case the morphological

"races" proved to belong to highly divergent branches in an intraspecific

evolutionary

tree (Figure 5; 11, 12). This of course

in no way excludes natural

selection as a possible factor influencing

the evolution of these racial differ-

ences.

In a recent review of geographic variation in allozymes, Selander &

Whittam (79) concluded: "studies of protein polymorphisms

indicate that a

great

variety of organisms,

ranging

from bacteria to humans .. ., are strongly

structured

genetically and that their evolution cannot be understood

without

reference

to this structure."

Data from mtDNA have revealed

an even greater

degree of population

structure

for many species. But more importantly,

the

nature

of assayable mtDNA differences has allowed relatively unambiguous

documentation

of a strong

phylogenetic

component

to geographic

differentia-

tion. Most species have a rich phylogeographic

diversity characterized by

localized clades and, not infrequently,

important

phylogenetic gaps between

allopatric

populations.

Many mtDNA lineages within

species date to common

ancestors several million years BP. Thus, no longer will it be defensible to

consider

species as phylogenetically

monolithic entities

in scenarios

of specia-

tion or macroevolution. Phylogenetic differences within species are quali-

tatively of the same kind as, though often smaller in magnitude

than, those

normally pictured

in higher-order

phylogeny reconstructions.

To paraphrase

and update the statement

by Selander & Whittam

quoted above: Studies of

mtDNA polymorphisms

indicate that

a great

many species are strongly

struc-

tured phylogenetically and that their evolution cannot be fully understood

without references to this intraspecific

phylogeographic

structure.

SUMMARY

Mitochondrial

DNA has provided the first extensive and readily accessible

data available to evolutionists in a form suitable for strong genealogical

inference at the intraspecific level. The rapid pace of mtDNA nucleotide

substitution, coupled with the special mode of maternal nonrecombining

mtDNA inheritance,

offers advantages

for phylogenetic

analysis

at the micro-

evolutionary

level that

will not be matched

easily by any nuclear

gene system.

These peculiarities of mtDNA data have literally forced the addition of a

phylogenetic perspective

to studies of intraspecific

evolutionary process and

as such have provided an empirical and conceptual bridge between the

nominally rather

separate

disciplines of systematics

and population

genetics.

PRINCIPLES

OF mtDNA PHYLOGEOGRAPHY 519

MtDNA has also served to clarify thinking

about the distinction

between (yet

relevance of) gene genealogies to organismal

phylogeny.

Many species have proved

to exhibit a deep and geographically

structured

mtDNA phylogenetic history. Study of the relationship between genealogy

and geography constitutes a discipline that can be termed intraspecific

phylogeography. We present several phylogeographic

hypotheses that were

motivated

by available data and that

represent

possible trends

whose broader

generality

remains

to be tested. Study

of intraspecific

phylogeography

should

assume a place in evolutionary biology at least commensurate

with that of

ecogeography, with mutual benefit resulting

to both disciplines. Theories of

speciation and macroevolution

must now recognize and accommodate the

reality of phylogeographic

differentiation

at the intraspecific

level.

ACKNOWLEDGMENTS

We wish to thank Dr. Bob Lansman for introducing us to mitochondrial

DNA. John C. Avise's laboratory

has been supported

by grants

from NSF.

Publication costs were funded by contract DE-AC09-76SR00-819 between

the US Department

of Energy and the University of Georgia Institute of

Ecology.

Literature

Cited

1. Abbott, R. 1957. The tropical western

Atlantic province. Proc. Phila. Shell

Club 1:7-11

2. Ammerman, A. J., Cavalli-Sforza, L.

L. 1984. The Neolithic Transition and

the Genetics of Populations in Europe.

Princeton, NJ: Princeton

Univ. Press

3. Aquadro,

C. F., Greenberg,

B. D. 1983.

Human mitochondrial DNA variation

and evolution: Analysis of nucleotide

sequences from seven individuals. Ge-

nietics 103:287-312

4. Aquadro, C. F., Kaplan, N., Risko, K.

J. 1984. An analysis of the dynamics

mammalian mitochondrial DNA se-

quence evolution. Molec. Biol. Evol.

1:423-34

5. Asmussen, M. A., Arnold, J., Avise, J.

C. 1987. Definition and properties of

disequilibrium

statistics for associations

between nuclear and cytoplasmic

genotypes. Genetics. 115:755-68

6. Avise, J. C. 1983. Protein variation

and

phylogenetic reconstruction.

In Protein

Variation:

Adaptive

and Taxonomic

Sig-

nificance, ed. G. Oxford, D. Rollinson,

pp. 103-30. London:

Syst. Assoc. Publ.

Br. Mus. Nat. History

7. Avise, J. C. 1986. Mitochondrial

DNA

and the evolutionary genetics of higher

animals. Phil. Trans. R. Soc. Lond.

B312:325-42

8. Avise, J. C. 1987. Identification

and in-

terpretation of mitochondrial DNA

stocks in marine

species. In Proc. Stock

Identification Workshop,

ed. H. Kumpf,

E. L. Nakamura,

pp. Panama

City, FL:

Publ. Natl. Oceanogr. Atmos. Admin.

In press

9. Avise, J. C., Ayala, F. J. 1975. Genetic

change and

rates of cladogenesis. Genet-

ics 81:757-73

10. Avise, J. C., Lansman, R. A. 1983.

Polymorphism

of mitochondrial

DNA in

populations

of higher animals. In Evolu-

tion of Genes and Proteins, ed. M. Nei,

R. K. Koehn, pp. 147-64. Sunderland,

Mass: Sinauer

11. Avise, J. C., Smith, M. H. 1974.

Biochemical

genetics of sunfish. I. Geo-

graphic variation and subspecific in-

tergradation in the bluegill, Lepomis

macrochirus. Evolution 28:42-56

12. Avise, J. C., Bermingham, E., Kessler,

L. G., Saunders,

N. C. 1984. Character-

ization of mitochondrial

DNA variability

in a hybrid

swarm between subspecies of

bluegill sunfish (Lepomis

macrochirus).

Evolution 38:931-41

13. Avise, J. C., Giblin-Davidson, G.,

520 AVISE FT Al.

Laerm, J., Patton, J. C., Lansman, R.

A. 1979. Mitochondrial DNA clones

and matriarchal

phylogeny within and

among geographic populations of the

pocket gopher, Geomvs pinetis. Proc.

Natl. Acad. Aci. USA 76:6694-98

14. Avise, J. C., Helfman, G. S.. Saunders,

N. C., Hales, L. S. 1986. Mitochondrial

DNA differentiation in North Atlantic

eels: Population genetic consequences

an unusual life history pattern. Proc.

Natl. Acad. Sci. USA 83:4350-54

15. Avise, J. C., Neigel, J. E., Arnold, J.

1984. Demographic

influences on mito-

chondrial DNA lineage survivorship

animal populations. J. Mol. Evol.

20:99-105

16. Avise, J. C., Reeb, C. A., Saunders,

C. 1987. Geographic population struc-

ture and

species differences in mitochon-

drial DNA of mouthbrooding marine

catfishes (Ariidae)

and demersal

spawn-

ing toadfishes (Batrachoididae).

Evolu-

lion. In press

17. Avise, J. C., Shapira,

J. F., Daniel, S.

W., Aquadro, C. F., Lansman, R. A.

1983. Mitochondrial DNA differentia-

tion during the speciation process in

Peromvscus. Mol. Biol. Evol. 1:38-

18. Ayala, F. J., ed. 1976. Molecular

Evolution. Sunderland,

Mass: Sinauer

19. Bermingham, E., Avise, J. C. 1986.

Molecular zoogeography of freshwater

fishes in the southeastern United States.

Genetics 113:939-65

20. Bermingham, E., Lamb, T., Avise, J.

C. 1986. Size polymorphism

and hetero-

plasmy in the mitochondrial DNA of

lower vertebrates.

J. Hereditv 77:249-

21. Blair, W. F. 1950. Ecological factors in

speciation of Peromyscus. Evolution

4:253-75

22. Briggs, J. C. 1974. Marine Zoogeogra-

phv. New York: McGraw-Hill

23. Brown, J. H., Gibson, A. C. 1983.

Biogeographv. St. Louis, Mo: C. V.

Mosby

24. Brown, W. M. 1980. Polymorphism

mitochondrial DNA of humans as re-

vealed by restriction endonuclease anal-

ysis. Proc. Natl. Acad. Sci. USA

77:3605-9

25. Brown, W. M. 1983. Evolution of an-

imal mitochondrial

DNA. In Evolution

of Genes and Proteins, ed. M. Nei, R.

K. Koehn, pp. 62-88. Sunderland,

Mass: Sinauer

26. Brown, W. M. 1985. The mitochondrial

genome of animals. In Molecular Evolu-

tionarv Genetics, ed. R. J. Maclntyre

pp. 95-130. New York: Plenum

27. Brown, W. M., Goodman,

H. M. 1979.

Quantitation of intrapopulation

variation

by restriction

analysis of human mito-

chondrial DNA. In Extrachromosomal

DNA, ed. D. J. Cummings, P. Borst, I.

B. Dawid, S. M. Weissman, C. F. Fox,

pp. 485-500. New York: Academic

28. Brown, W. M., George, M. Jr., Wilson,

A. C. 1979. Rapid evolution of animal

mitochondrial DNA. Proc. Natl. Acad.

Sci. USA 76:1967-71

29. Brown, W. M., Prager, E. M., Wang,

A., Wilson, A. C. 1982. Mitochondrial

DNA sequences of primates:

Tempo and

mode of evolution. J. Mol. Evol.

18:225-39

30. Cann. R. L., Wilson, A. C. 1983.

Length mutations in human mitochon-

drial DNA. Genetics 104:699-711

31. Cann, R. L., Brown, W. M., Wilson,

A. C. 1982. Evolution of human

mitochondrial DNA: A preliminary

re-

port. In Human Genetics, Part A, The

unfolding genome, ed. B. Bonne-Tamir,

P. Cohen, and R. N. Goodman, pp.

157-65. New York: Liss

32. Cann, R. L., Brown, W. M., Wilson,

A. C. 1984. Polymorphic sites and the

mechanism of evolution of human

mitochondrial DNA. Genetics 106:479-

33. Cann. R. L., Stoneking, M., Wilson, A.

C. 1987. Mitochondrial DNA and hu-

man evolution. Nature 325:31-36

34. Chapman, R. W., Stephens, J. C.,

Lansman, R. A., Avise, J. C. 1982.

Models of mitochondrial DNA transmis-

sion genetics and

evolution in higher

eu-

caryotes. Genet. Res. 40:41-57

35. Crow, J. F., Kimura, M. 1970. An In-

troduction to Population Genetic

Theorv. New York: Harper & Row

36. Dobzhansky,

T. 1937. Genetics and the

Origin of Species. New York:

Columbia

Univ. Press

37. Dolbeer, R. A. 1978. Movement and

migration

patterns

of red-winged

black-

birds: A continental overview. Bird

Banding 49:17-34

38. Eldrege, N., Gould, S. J. 1972. Punctu-

ated

equilibria:

An alternative to phyletic

gradualism.

In Models in Paleobiology,

ed. T. J. M. Schopf, pp. 82-115. San

Francisco, Calif: Freeman, Cooper

39. Falconer, D. S. 1981. Introduction to

Quantitative

Genetics New York:

Long-

man. 2nd ed.

40. Felsenstein, J. 1982. Numerical

methods

for inferring

evolutionary

trees.

Q. Rev. Biol. 57:379-404

41. Felsenstein, J. 1985. Confidence limits

on phylogenies: An approach utilizing

the bootstrap.

Evolution 39:783-91

PRINCIPLES OF mtDNA PHYLOGEOGRAPHY 521

42. Fink, W. L. 1986. Microcomputers and

phylogenetic analysis. Science 234:

1135-39

43. George, M. Jr, Ryder, 0. A. 1986.

Mitochondrial DNA evolution in the

genus Equus. Mol. Biol. Evol. 3:535-46

44. Gillespie, J. H. 1986. Rates of molecu-

lar evolution. Ann. Rev. Ecol. Syst.

17:637-65

45. Goldschmidt, R. 1940. The Material

Basis of Evolution. New Haven, Conn:

Yale Univ. Press

46. Gould, S. J. 1980. Is a new and general

theory of evolution emerging? Paleobi-

ology 6:119-30

47. Gould, S. J., Lewontin, R. C. 1979.

The spandrels

of San Marco

and the Pan-

glossian paradigm: A critique of the

adaptationist programme. Proc. Royal

Soc. London Ser. B 205:581-98

48. Graves, J. E., Ferris, S. D., Dizon, A.

E. 1984. Close genetic similarity of

Atlantic and Pacific skipjack tuna (Kat-

suwonus pelamis) demonstrated with

restriction endonuclease analysis of

mitochondrial DNA. Mar. Biol. 79:315-

49. Grivell, L. A. 1983. Mitochondrial

DNA. Sci. Am. 248:78-89

50. Gyllensten, U., Wharton, D., Wilson,

A. C. 1985. Material inheritance of

mitochondrial

DNA during backcrossing

of two species of mice. J. Hered.

76:321-24

51. Harris, T. 1963. The Theory of Branch-

ing Processes. Berlin: Springer-Verlag

52. Hauswirth,

W. W., Laipis, P. J. 1985.

Transmission genetics of mammalian

mitochondria:

A molecular model and

experimental

evidence. In Achievements

and Perspectives of Mitochondrial Re-

search, Vol. II: Biogenesis, ed. E.

Quagliariello, E. C. Slater, F. Palmieri,

C. Saccone, A. M. Kroon, pp. 49-60.

New York: Elsevier

53. Hecht, M. K., Hoffman, A. 1986. Why

not neo-Darwinism? A critique of

paleobiological challenges. Oxford

Surv. Evol. Biol. 3:1-47

54. Hedrick, P. W. 1983. Genetics of Pop-

ulations. New York: Van Nostrand

Reinhold

55. Hennig, W. 1966. Phylogenetic Sys-

tematics. Urbana:

Univ. Ill. Press

56. Hubbs, C. L., Allen, E. R. 1944. Fishes

of Silver Springs, Florida. Proc. Fla.

Acad. Sci. 6:110-30

57. Johnson, M. J., Wallace, D. C., Ferris,

S. D., Rattazzi, M. C., Cavalli-Sforza,

L. L. 1983. Radiation of human

mitochondrial

DNA types analyzed by

restriction endonuclease cleavage pat-

terns. J. Mol. Evol. 19:255-71

58. Kimura, M., Ohta,

T. 1971. Theoretical

Aspects of Population Genetics. Prince-

ton, NJ: Princeton

Univ. Press

59. King, J. L., Jukes, T. H. 1969. Non-

Darwinian evolution: Random fixation

of selectively neutral mutations.

Science

164:788-98

60. Lansman,

R. A., Avise, J. C., Huettel,

M. D. 1983. Critical experimental

test

of the possibility of "patemal

leakage"

of mitochondrial DNA. Proc. Natl.

Acad. Sci. USA 80:1969-71

61. Lansman,

R. A., Avise, J. C., Aquadro,

C. F., Shapira, J. F., Daniel, S. W.

1983. Extensive genetic variation in

mitochondrial

DNAs among geographic

populations of the deer mouse, Per-

omyscus maniculatus. Evolution 37:1-

62. Lansman,

R. A., Shade, R. O., Shapira,

J. F., Avise, J. C. 1981. The use of

restriction endonucleases to measure

mitochondrial DNA sequence related-

ness in natural populations. III. Tech-

niques and potential applications. J.

Mol. Evol. 17:214-26

63. Lewontin, R. C. 1974. The Genetic

Basis of Evolutionary

Change. Colum-

bia Univ. Press, NY.

64. Li, C. C. 1955. Population Genetics.

Chicago: Univ. Chicago Press

65. Lotka, A. J. 1931. Population analy-

sis-the extinction of families. I. J.

Wash. Acad. Sci. 21:453-59

66. Lotka, A. J. 1931. Population analy-

sis-the extinction of families. II. J.

Wash. Acad. Sci. 21:453-59

67. Mayr, E. 1963. Animal Species and

Evolution. Cambridge, Mass: Belknap

Press, Harvard

68. Mayr, E. 1982. Speciation and macro-

evolution. Evolution

36:1119-32

69. Moritz, C., Brown, W. M. 1986. Tan-

dem duplication

of D-loop and ribosom-

al RNA sequences

in lizard mitochondri-

al DNA. Science 233:1425-27

70. Nei, M. 1975. Molecular Population

Genetics and Evolution. New York:

Elsevier

71. Nei, M. 1987. Molecular Evolutionary

Genetics. New York: Columbia Univ.

Press

72. Nei, M., Li, W.-H. 1979. Mathematical

model for studying genetic variation

terms of restriction endonucleases.

Proc. Natl. Acad. Sci. USA

76:5269-73

73. Neigel, J. E., Avise, J. C. 1986.

Phylogenetic relationships

of mitochon-

drial DNA under various demographic

models of speciation. In Evolutionary

Processes and Theory, ed. E. Nevo, S.

Karlin, pp. 515-34. New York: Aca-

demic

522 AVISE ET AL

74. O'Brien, T. W., Denslow, N. D., Har-

ville, T. O., Hessler, R. A., Matthews,

D. E. 1980. Functional and structural

roles of proteins in mammalian

mitochondrial

ribosomes. In The Orga-

nization and Expression of the Mito-

chondrial

Genome, ed. A. M. Kroon, C.

Saccone, pp. 301-305. New York: Else-

vier

75. Ochman, H., Jones, J. S., Selander, R.

K. 1983. Molecular area effects in

Cepaea. Proc. Natl. Acad. Sci. USA

80:4189-93

76. Rand, D. M., Harrison, R. G. 1986.

Mitochondrial DNA transmission genetL

ics in crickets. Genetics 114:955-70.

77. Saunders,

N. C., Kessler, L. G., Avise,

J. C. 1986. Genetic variation and geo-

graphic differentiation

in mitochondrial

DNA of the horeshoe crab, Limulus

polyphemus. Genetics 112:613-27

78. Schweyen, R. J., Wolf, K., Kaudewitz,

F. eds. 1983. Mitochondria 1983, Nuc-

leo-Mitochondrial Interactions. New

York: de Gruyter Publ.

79. Selander, R. K., Whittam, T. S. 1983.

Protein polymorphism and the genetic

structure of populations. In Evolution of

Genes and Proteins, ed. M. Nei, R. K.

Koehn, pp. 89-114. Sunderland,

Mass:

Sinauer

80. Simpson, G. G. 1944. Tempo

and Mode

in Evolution. New York: Columbia

Univ. Press

81. Sneathz P. H. A., Sokal, R. R. 1973.

Numerical Taxonomy. San Francisco:

W. H. Freeman

82. Spiess, E. B. 1977. Genes in Pop-

ulations. New York: Wiley

83. Stebbins, G. L., Ayala, F. J. 1981. Is a

new evolutionary synthesis necessary?

Science 213:967-71

84. Swift, C. C., Gilbert, C. R., Bortone, S.

A., Burgess, G. H., Yerger, R. W.

1986. Zoogeography of the freshwater

fishes of the southeastern

United States:

Savannah River to Lake Ponchartrain.

Zoogeography of North American

Freshwater

Fishes, ed. C. H. Hocutt,

0. Wiley, pp. 213-65. New York:

Wiley-

85. Tajima, F. 1983. Evolutionary

relation-

ship of DNA sequences in finite pop-

ulations. Genetics 105:437-60

86. Takahata,

N. 1985. Population genetics

of extranuclear

genomes:-

A model and

review. In Population Genetics and

Molecular Evolution, ed. T. Ohta, K.

Aoki, pp. 195-212. Berlin: Springer-

Verlag

87. Takahata, N., Palumbi, S. R. 1985. Ex-

tranuclear differentiation and gene flow

in the finite island model. Genetics

109:441-57

88. Takahata, N., Slatkin, M. 1984.

Mitochondrial gene flow. Proc. Natl.

Acad. Sci. USA 81:1764-67

89. Templeton, A. R. 1983. Phylogenetic

inference from restriction endonuclease

cleavage site maps with particular

refer-

ence to the evolution of man and the

apes. Evolution 37:221-44

90. Templeton, A. R. 1983. Convergent

evolution and nonparametric

inferences

from restriction data and DNA se-

quences. In Statistical Analysis of DNA

Sequence Data, ed. B. S. Weir, pp.

151-79. New York: Marcel Dekker

91. Uphold, W. B. 1977. Estimation of

DNA sequence divergence from com-

parison of restriction endonuclease di-

gests. Nucleic Acids Res. 4:1257-65

92. Vawter, L., Brown, W. M. 1986. Nu-

clear and mitochondrial DNA com-

parisons

reveal extreme rate variation in

the molecular clock. Science 234:194-

93. Wainscoat,

J. 1987. Out of the garden

Eden. Nature 325:13

94. Wallace, D. G. 1983. Structure and

evolution of organelle DNAs. In

Endocytobiology,

Vol. II, ed. H. E. A.

Schenk, W. Schwemmler, pp. 87-100.

New York: de Gruyter

95. Whittam, T. S., Clark, A. G., Stone-

king, M., Cann, R. L., Wilson, A. C.

1986. Allelic variation in human

mitochondrial genes based

on patterns

restriction site polymorphism. Proc.

Natl. Acad. Sci. USA 83:9611-15

96. Wiley, E. 0. 1'981.

Phylogenetics. New

York: Wiley

97. Wilson, A. C., Cann, R. L., Carr, S.

M., George, M. Jr., Gyllensten, U. B.,

Helm-Bychowski, K. M., Higuchi, R.

G., Palumbi, S. R., Prager, E. M.,

Sage, R. D., Stoneking, M. 1985.

Mitochondrial DNA and two per-

spectives on evolutionary genetics. Biol.

J. Linn. Soc. 26:375-400

98. Zink, R. M., Remsen, J. V. Jr. 1986.

Evolutionary processes and patterns

geographic

variation in birds.

In Current

Ornithology,

Vol. 4, ed. R. F. Johnston,

pp. 1-69. New York: Plenum

Analysis of Mitochondrial COI Gene for Species/ Breed Level Identification of the Genus Capra (Linnaeus, 1758)

Article

Full-text available

Feb 2025
PAK J ZOOL

The current study mainly aimed to identify the goat breed at species level through COI gene sequencing. In this study the effectiveness of COI gene was examined in distinguishing the goat breeds which has not been inclusively investigated in mammals. To fill this gap, we determined the COI barcodes for six different breeds of goat and found them all as domestic goats (Capra hircus Thomas, 1911). Mean intraspecific genetic distance was 0.0016%, obtained from 528 COI positions, lower than the average interspecific genetic distance 0.03%, indicating the discrimination efficacy of COI gene. Using the COI gene sequences the phylogenetic analysis confirmed that the domestic goat breed has a very close kinship with Siberian goat breeds, Capra sibirica (Pallas, 1776), a wild goat species. It is concluded that molecular based identifications of animal species are more accurate, compared to morphological approaches for identification of the indigenous breeds. Furthermore, for future study, the COI gene inclusive exploration is suggested to find the genetic divergence in goat breeds of the country.

Unraveling the Genetic Diversity and Phylogeography of the “King of Vitamin C” Fruit (Rosa roxburghii Trattinnick) in Chinese Southwest

Article

Full-text available

May 2025

The southwestern region of China is one of the last repositories of biodiversity. Exploring the genetic diversity and phylogeographical structure of representative species in this area can not only provide insights into their evolutionary history but also help us specify protection plans. We conducted extensive sampling and combined ISSR, cpDNA (psbA‐trnH, atpF‐trnH, trnL‐trnF, accD‐psaI, trnG‐trnS), and single‐copy nuclear gene (ncpGS and GAPDH) data with the MaxEnt model to investigate the genetic diversity, phylogeographic pattern and potential distribution of Rosa roxburghii Trattinnick, the “King of Vitamin C” fruit, in this region. The study revealed high genetic diversity within R. roxburghii populations, with Guizhou Province identified as the center of origin. At least two glacial refugia were presented during the Quaternary ice age. The research emphasizes the impact of climate change and geographic isolation on genetic differentiation. This study contributes to the understanding of biodiversity in the southwestern region of China and offers a new perspective on the conservation of wild Rosa roxburghii and other plant resources.

Genetic Diversity Analysis of Sambar Deer (Cervus unicolor) Based on Mitochondrial DNA D-loop Sequence

Article

Full-text available

Apr 2025

This study aimed to analyze the genetic diversity of the sambar deer (Cervus unicolor) population through mitochondrial D-loop DNA sequencing. The findings provide valuable insights into the conservation status of this vulnerable species in the region and inform future management strategies for Cervus unicolor. Blood samples from 10 individual sambar deer, consisting of 8 females and two male deer, were amplified using PCR and then sequenced. Data analysis of the genetic diversity of sambar deer was carried out using the genome sequencing method from NCBI, Bioedit 7.7.1 software, DNAsp 5.1, and MEGA 11 Software. The results of this study were that the DNA concentration test in sambar deer had an average of 12.375 ng/uL, with an average DNA purity test = 1.34, with 10 samples divided into 3 haplotypes. The level of genetic diversity of sambar deer from all samples is π = 0.01745 ± 0.00380, and haplotype diversity of Hd = 1,000 ± 0.045. Based on the phylogenetic tree, there are two parts: the Asian and Kalimantan regions. The conclusion of the current study showed that sequencing analysis of Sambar deer shows relatively high diversity and is a reasonable basis for performance selection and development of modern Sambar deer breeding.

Mechanisms Maintaining Mitochondrial DNA Polymorphisms: The Role of Mito-Nuclear Interactions, Sex-Specific Selection, and Genotype-by-Environment Interactions in Drosophila subobscura

Article

Full-text available

Apr 2025

Experimental mito-nuclear introgression lines (MNILs) were established by backcrossing isofemale lines of D. subobscura originating from the same populations. MNILs were subjected to a series of life-history experiments designed to test the fitness of the bearers of different combinations of two main mtDNA haplotypes on their own nuclear background, as well as on the background of the opposite haplotype. By having 11 replicas of the four mito-nuclear combinations, we could test not only the adaptive significance of the differences between the two main haplotypes but also the influence of additional variation present within each of the 11 combinations on fitness. Testing the fitness of individuals of both sexes enabled us to examine if sex-specific selection has a role in maintaining the frequencies of the two mtDNA haplotypes in nature. Conducting the fitness assays on two different temperatures enabled us to test whether different temperatures favor specific mtDNA haplotypes or mito-nuclear genotypes and consequently promote stable sympatric mtDNA variation. The results show weak signature of genotype-by-environment interactions, and no sex-specific selection regarding differences between the two main haplotypes. However, individual models across different life-history components showed these two mechanisms at play in promoting mtDNA variability present in specific mito-nuclear crosses. Our models show that mito-nuclear interactions are, in fact, more important as units of selection.

Caracterización molecular de Panstrongylus chinai del norte del Perú y su relación filogenética con poblaciones ecuatorianas empleando el gen COIMolecular characterization of Panstrongylus chinai from northern Peru and its phylogenetic relationship to Ecuadorian populations using the COI gene

Article

Full-text available

Mar 2025

Objective. To determine the molecular characterization of Panstrongylus chinai from northern Peru and its phylogenetic relationship with Ecuadorian populations using the Cytochrome C Oxidase Subunit I (COI) gene. Materials and methods. We analyzed three adult female P. chinai specimens from populations reared under laboratory conditions, from rural localities in the department of Piura. The legs of each specimen were dissected from the coxa to the tibia, discarding the tarsi and nails, then the DNA was extracted, and a polymerase chain reaction (PCR) of the COI gene was carried out. The PCR products were sequenced by Sanger and analyzed with DNA sequences of the COI gene of P. chinai from Ecuador, obtained from the NCBI portal, Genbank. The DNA sequences of the study, together with similar sequences found in the NCBI database, were inserted into the MEGA v.11 software to construct a phylogenetic tree. They were then transferred to the DnaSP v.5 software for molecular characterization by haplotypes. Results. Molecular characterization revealed the presence of three haplotypes circulating in the department of Piura, different from the haplotype previously reported in Ecuador. Likewise, phylogenetic analysis suggests the emergence of the evolutionary process of cladogenesis, in which the Ecuadorian variant may have originated from populations of P. chinai from northern Peru. Conclusions. P. chinai from Ecuador and northern Peru have different molecular characteristics and a descending phylogeny, which infer distribution from Peru to Ecuador.

Hybridization ddRAD‐sequencing and phenotypic analysis clarify the phylogeographic structure and evolution of an alpine Chrysanthemum species with a sky island distribution

Article

Full-text available

Apr 2025

The phylogeographic structure of a species is the result of historical intraspecific differentiation and influences the pace and trajectory of speciation. Therefore, study of the phylogeographic structure of species and the mechanisms underlying its formation can shed light on the evolutionary history and speciation of species, as well as enhance our understanding of the generation and maintenance of species diversity. Chrysanthemum hypargyrum is an alpine species endemic to central China. It is restrictively distributed to three isolated mountain ranges, and its populations exhibit a sky island distribution and some morphological variation to different environments. In this study, we investigated the morphogenetic divergence, phylogeographic structure, and evolutionary history of this species through hybridization ddRAD-sequencing, phenotypic analysis, and species distribution modeling. Our results indicate that C. hypargyrum originated in the Daba Mountains and has since diverged into three lineages. The phylogeographic structure and distribution of this species are mainly attributed to geographic isolation, the founder effect and Quaternary climate oscillations as its range expanded. The divergence of its three major lineages coincided with Pliocene mountain uplifts and Pleistocene climatic fluctuations. The current sky island distribution has also promoted the diversification and phylogeographic structure of C. hypargyrum.

Genetic diversity of the zigzag ladybird beetle, Cheilomenes sexmaculata (F.) (Coleoptera: Coccinellidae) with its distribution in India and implications for biological control

Article

Full-text available

May 2025

Biological control is considered the backbone of any integrated pest management programme. Exploring native ranges to identify new biological control agents is technically challenging and time-consuming; however, incorporating biogeographical insights into the selection process enhances the likelihood of success in biological control efforts. The zigzag beetle, Cheilomenes sexmaculata (F.), the most important and abundant of all ladybird beetles, feeds on a diverse range of prey. To study the population structure of C. sexmaculata, individuals were collected from five different zones, consisting of 25 subpopulations (subsets from their respective main zones) that uniformly represented India. We identified 109,585 contigs, 304,784 SNPs, and 46,583 INDELs, among which 5 INDELs and 9 SSRs were polymorphic. We observed high genetic diversity parameters for populations collected from the West Zone in all SNPs, SSRs, and INDELs markers. However, gene flow (Nm) was restricted, and FST was high in the South Zone and East Population (FST=0.45) populations, while the populations from the Central, North and West zones recorded zero FST. The structure and dendrogram construction analysis revealed individuals from Coorg and Mudigere of the South Zone and Barapani of the East Zone populations formed separate clusters. Mantel correlation revealed genetic distances decreased with the increasing geographical distance SNPs (Rxy= − 0.184 and P = 0.07); INDELs (Rxy= − 0.204 and P = 0.05) and SSRs (Rxy= − 0.339 and P = 0.06) indicated moderate genetic differentiation in the South and East zones populations due to geographic barriers, genetic drift and local environmental conditions. Populations in the North, West, and Central zones showed high gene flow, likely due to modern domestication and human activities. Additionally, the findings indicated that the populations from the West, North, and Central zones are better strains due to their high genetic fitness, so these populations can be used for classical and augmentative biological control.

Phylogenetic Analyses of Bostrichiformia and Characterization of the Mitogenome of Gibbium aequinoctiale (Bostrichiformia Ptinidae)

Article

Apr 2025

Background: Ptinidae, within the infraorder Bostrichiformia, are a cosmopolitan, ecologically diverse but poorly known group. The phylogeny within Bostrichiformia and the monophyly of Ptinidae and its phylogenetic placement in Bostrichiformia remain contentious. Methods: In this research, we determined the entire mitochondrial genome (mitogenome) of Gibbium aequinoctiale, the first representative mitogenome of the subfamily Ptininae, and reconstructed the phylogenetic relationships for Bostrichiformia based on four mitochondrial datasets using maximum likelihood (ML) and Bayesian inference (BI) methods. Results: The mitogenome of G. aequinoctiale is a circular molecule spanning 17,020 bp and harbors 37 mitochondrial genes and a presumed control region (CR). The mitogenome exhibited a marked preference for the utilization of A and T bases, which was also observed in three kinds of genes and CR. AAT was inferred as the putative candidate initiation codon for cytochrome oxidase subunits 1 (COI). The control region contains three tandem repeats (TDRs) and one poly-thymine stretch (Poly-T) in both coding strands. The phylogenetic results appeared to support the monophyly of four families, Nosodendridae, Derodontidae, Dermestidae, and Bostrichidae, and the basal position of the latter two families within Bostrichiformia. However, the family Ptinidae was not verified as monophyly because of one species diverging from the main lineage. Three families, Dermestidae, Bostrichidae, and Ptinidae, clustered as the major clade in Bostrichiformia, among which Bostrichidae and Ptinidae grouped together as sister groups. Conclusions: The present study provides valuable mitochondrial information for Ptinidae and provides novel perspectives on the inner phylogeny within the infraorder Bostrichiformia.

Leveraging Whole Genomes, Mitochondrial DNA and Haploblocks to Decipher Complex Demographic Histories: An Example From a Broadly Admixed Arctic Fish

Article

Apr 2025
MOL ECOL

The study of phylogeography has transitioned from mitochondrial haplotypes to genome‐wide analyses, borrowing from population genomics methods along the way. Whole‐genome sequencing allows the study of both mitochondrial and nuclear DNA and provides the density of markers to investigate recombination along the genome. This level of resolution could unravel complex histories of admixture between lineages, which are commonly observed in species evolving in recently deglaciated habitats. In this study, we sequenced 1120 Arctic Char genomes from 33 populations across Canada and Greenland to characterise patterns of genetic variation and diversity, and how they are shaped by hybridisation between the Arctic and Atlantic glacial lineages. Mitochondrial genomes across the study area were predominantly of Arctic origin, except in Greenland, where we observed some Atlantic descent. Through admixture analyses and demographic inferences on nuclear markers, we identified that all Canadian populations under the 66th parallel showed introgression from the Atlantic lineage, leading to higher genetic diversity. By scanning the genome using local principal component analyses, we identified putative large low‐recombining haploblocks as local ancestry tracts from either lineage. Since haplotypes might retain different signatures of postglacial histories by sheltering sequences from recombination, we attempted to infer origins of recolonisation using whole genomes vs. ancestry tracts for the Arctic lineage. Despite limitations, we unveiled clues suggesting a complex postglacial history in Arctic Char. Overall, our study demonstrates that, even at low depth, making the most of whole‐genome sequencing by analysing several genomic compartments provides a versatile and powerful way to address phylogeographic dynamics.

What Does Phylogeography Teach Us About Population Genetics in the Neotropics?

Chapter

Apr 2025

The neotropical region is a highly significant and biodiverse area, spanning through a large portion of Central and South America and encompassing diverse landscapes. The region holds immense global importance for biodiversity, as it offers essential ecological services crucial to the planet’s health. Ensuring the protection of the region and its unique biodiversity is of utmost importance for the well-being of current and future generations. To this end, phylogeography, a relatively recent field, provides a historical and evolutionary context for biodiversity studies since it investigates the historical processes that have influenced genetic diversity. Phylogeography enables the formulation of biogeographic hypotheses, the description of the evolution of reproductively isolated population units, the inference of processes relating to the origin, distribution, and maintenance of biodiversity, and the creation of hypotheses concerning the influence of past environmental changes, physical and biotic characteristics of populations, population structure, and genetic diversity. This chapter presents an overview of how phylogeography can be instrumental in identifying divergent lineages, understanding species’ responses to climatic changes, identifying refuge areas, dating geographical events, and maintaining the neotropical biota. Phylogeography has shed light on high levels of genetic diversity, genetic structure, hybridization, and various historical events, thereby serving as a pillar in population genetics research. Furthermore, advancements in molecular techniques and more efficient machine learning have facilitated the availability of larger datasets for the region, leading to fundamental insights into the evolutionary processes that have given rise to the extensive biodiversity we observe in the Neotropics today.

Evolutionary processes and patterns of geographic variation in birds

Chapter

Full-text available

Jan 1986

Movement and Migration Patterns of Red-Winged Blackbirds: A Continental Overview

Article

Full-text available

Jan 1978

Richard Dolbeer

From 1924 through 1974, about 11,000 recoveries accumulated from the banding of over 700,000 Red-winged Blackbirds(Agelaius phoeniceus) in North America. A few studies have examined some of these data for specific localities during certain times of the year; however, no attempt has been made to examine the total recovery data to compile a general picture of continental movement and migration patterns. Increasing attention is being given to blackbird (Icteridae) and Starling (Sturnus vulgaris) populations in North America because of their reported crop depredations, health hazards, and nuisance aspects, especially when congregating in large roosts (Meanley,1975;Graham, 1976). Solutions to these problems are difficult to find because the problems are widespread and the populations are highly mobile. Management measures necessary in one area may have effects in areas far removed. Thus, effective management requires a thorough knowledge of the overall movement and migration pattern of each species. The objectives of this study are to: (1) determine the general continental movement and migration patterns of the Red-wing during the annual cycle, and (2) examine banding and recovery numbers in relation to population numbers on a regional basis to pinpoint areas where additional banding and/or recovery effort is needed.

Zoogeography of the freshwater fishes of the southeastern United States: Savannah River to Lake Pontchartrain.

Chapter

Full-text available

Jan 1986

Why not neo-Darwinism? A critique of paleobiological challenges

Article

Jan 1986

Nucleo-mitochondrial Interactions

Article

Jan 1983

Characterization of Mitochondrial DNA Variability in a Hybrid Swarm Between Subspecies of Bluegill Sunfish (Lepomis macrochirus)

Article

Sep 1984

We begin a characterization of the evolutionary dynamics of mitochondrial DNA (mtDNA) in fishes by examining restriction site variability in 189 bluegill sunfish (Lepomis macrochirus). A total of 15 endonucleases was employed to map 37 restriction sites in mtDNA from selected individuals. Genome size was approximately 16.2 kilobases. All differences between genotypes could be accounted for by gains or losses of individual sites, without additions, deletions, or rearrangements affecting more than about 50-250 base-pairs. Two highly distinct mtDNA genomes, differing by 20 assayed mutation steps and an estimated 8.5% sequence divergence, were discovered. Both genomes were observed in high frequency in a sample of 151 bluegill from a north-Georgia population, which on the basis of allozyme genotype appears to represent a freely-interbreeding hybrid swarm between two bluegill subspecies. Within the hybrid population, mtDNA and allozyme genotypes were associated approximately at random. The distinct genetic markers provided by the mtDNA genomes provided an improved test of the possibility of within-individual mtDNA polymorphism because true heteroplasmicity could readily be distinguished from results of incomplete digests. Nonetheless, in somatic or germ cells, no individual bluegill exhibited mtDNA heteroplasmicity (at 5% or greater level) that could be explained by paternal mtDNA transmission. Additional samples from Louisiana to Florida tentatively confirm that the geographic distributions of the two distinct mtDNA genomes are highly concordant with the previously described ranges of L. m. macrochirus and L. m. purpurescens defined by morphology and allozymes. Overall, the results on mtDNA variability in bluegill conform to and strengthen some of the more straightforward expectations about the pattern of evolution of uniparentally-transmitted genomes in sexually reproducing populations.

Non-Darwinian Evolution

Article

May 1969

Most evolutionary change in proteins may be due to neutral mutations and genetic drift.

Phylogenetic Inference From Restriction Endonuclease Cleavage Site Maps with Particular Reference to the Evolution of Humans and the Apes

Article

Mar 1983

Alan Templeton

The estimation procedure utilizes a compatibility analysis between enzyme data sets of the most parsimonious trees constructed from the restriction enzyme. Next, a non-parametric test is given for comparing alternative phylogenies. A 2nd non-parametric test is developed for testing the molecular clock hypothesis. To illustrate the power of these procedures, data derived from the mitochondrial DNA and globin DNA of man and the apes are analyzed. Although previous analyses of these data led to the speculation that 10 times more information would be required to resolve the evolutionary relationships between man with chimps and gorillas, this algorithm resolved these relationships at the 5% level of significance. The molecular clock hypothesis was rejected at the 1% level. The implications of this phylogenetic inference when coupled with other types of data lead to the conclusion that knuckle-walking - not bipedalism - is the evolutionary novelty in mode of locomotion in the primates and that many other hominid features are primitive whereas their African ape counterparts are derived.-from Author

Ecological Factors in Speciation of Peromyscus

Article

Sep 1950

W. Frank Blair

Geographic variation in mice of the genus Peromyscus is attributable to mutations, selection by the environment, and reduced gene flow through partial isolation of large or small segments of the species population. Wild genotypes are difficult of analysis in this group, but some mendelian alleles having important effects on wild populations have been described. Selection by the physical and biological environment is probably the most important agency determining the characteristics of sub-populations within the species range. There is evidence that simple predation is a selective agency of major importance in this group. Partial isolation of sub-populations because of ecological barriers and because of isolation by distance is another important factor in geographic variation. The home-range habit tends to restrict gene flow; dispersal of young on reaching sexual maturity has the opposite effect. The species is always unevenly distributed in its range due to preferences for certain environments, tolerance of others and avoidance of still others. The pattern of distribution varies within the species range, and in some parts of the range it may be more favorable to local differentiation due to isolation than it is in others. Geographic races are ecological phenomena. There is no difference between ecological races, geographic races and subspecies. Species with the greatest geographic ranges, occupying many environments, have the greatest number of geographic races, and the number of races is roughly proportional to the area occupied by the species. There is a correlation, in some species, between certain body dimensions and the vegetational environments occupied. There is a general correlation between pelage color of Peromyscus and the color of the soils on which the mice live. The width of zones of intergradation between geographic races is related to the rate of geographic change in environment. The present distribution and biological relations in four cenospecies of Peromyscus indicate that speciation is commonly initiated in this group by the simple geographic isolation of a segment of the population. There is evidence that new species have originated at the periphery of the range, on coastal islands, and probably on isolated mountain ranges. No marked change in habitat preference has accompanied speciation in Peromyscus, and it is unlikely that invasion of a new environment would have sufficient isolating effect to permit speciation. Opposite ends of chains of races have differentiated morphologically and in habitat preference, and intrinsic isolating mechanisms have possibly developed between some of these end populations. Extinction of intermediate populations would complete speciation in this case. There is no evidence that speciation would go to completion (i.e., breeding discontinuity) without this outside intervention. Geographic variation is favorable to the survival and evolution of the species, without the splitting off of new species. Wide distribution of the species population is favorable for speciation, because the most widely distributed population will have the most opportunities to be split by external factors into separate-breeding populations.

Speciation and Macroevolution

Article

Nov 1982

Ernst Mayr

Intraspecific Phylogeography: The Mitochondrial DNA Bridge Between Population Genetics and Systematics

Recommended publications

Phylogeography of the bullhead Cottus gobio (Pisces: Teleostei: Cottidae) suggests a pre-Pleistocene...

Looking forwards or looking backwards in avian phylogeography? A comment on Zink and Barrowclough 20...

Population structure and genetic diversity of Southeast Queensland populations of the Wallum Froglet...

The usefulness of maternally inherited genetic markers for phylogeographic studies in village chicke...