Critical Reviews in Plant Sciences, 29:148–161, 2010
Copyright © Taylor & Francis Group, LLC
ISSN: 0735-2689 print / 1549-7836 online
DOI: 10.1080/07352689.2010.481167
Plant Parentage, Pollination, and Dispersal: How DNA
Microsatellites Have Altered the Landscape
Mary V. Ashley
Department of Biological Sciences, University of Illinois at Chicago, Chicago, IL, USA
Downloaded By: [Ashley, Mary V.] At: 01:45 18 May 2010
Table of Contents
I.
INTRODUCTION ............................................................................................................................................. 148
II.
MICROSATELLITES AND PARENTAGE ASSIGNMENT .............................................................................. 149
III.
LIMITATIONS AND CONSTRAINTS .............................................................................................................. 150
IV.
OVERVIEW OF MICROSATELLITE PARENTAGE STUDIES ....................................................................... 151
V.
POLLEN DISPERSAL IN WIND-POLLINATED SPECIES ............................................................................. 154
VI.
POLLEN DISPERSAL IN ANIMAL-POLLINATED SPECIES ........................................................................ 156
VII. PARENTAGE ASSIGNMENT AND SEED DISPERSAL ................................................................................... 157
VIII. CONCLUSIONS ............................................................................................................................................... 158
ACKNOWLEDGMENTS ........................................................................................................................................... 158
REFERENCES .......................................................................................................................................................... 158
DNA microsatellites provide plant ecologists with molecular
markers precise enough to assign parentage to seeds and seedlings.
This allows the exact distance and trajectory of successful pollen
to be traced to characterize pollination patterns. Parentage assignment of established seedlings also allows researchers to accurately
determine how far new recruits have traveled from their seed parent. This paper reviews the history and development of molecular
parentage assignment in studies of native plants, as well as the limitations and constraints to this approach. This paper also reviews
53 articles published in the past 15 years that use parentage assignment to study pollination and seed dispersal in native plants. These
parentage studies have overturned many common assumptions regarding pollen and seed dispersal patterns. They show that longdistance dispersal of pollen is common in both wind and animal
dispersed systems, with average pollination distances commonly
being hundreds of meters. The pollination neighborhood is often
extremely large, and simple dispersal functions based on distance
alone fail to make accurate predictions of pollination. Rather than
hindering gene flow, fragmentation and isolation sometimes, and
perhaps even commonly, results in increased pollination distances.
Studies of seed dispersal using parentage assignment have also
Address correspondence to Mary V. Ashley, Department of Biological Sciences, M/C 066, University of Illinois at Chicago, 845 West
Taylor Street, Chicago, IL 60607, USA. E-mail: [email protected]
yielded some surprises. We now know that it may be erroneous to
assume that seeds growing under the crown of a conspecific adult
are growing beneath their mother, or that seed dispersal distances
are more limited than pollen dispersal distances. Taken together,
the studies to date demonstrate that both seed and pollen dispersal are quite complex phenomena influenced by many ecological
processes.
Keywords
microsatellites, paternity assignment, pollen dispersal,
seed dispersal, DNA markers
I. INTRODUCTION
Because plants are stationary as adults, dispersal of gametes
and offspring nearly always occurs through the movement of
pollen or seeds. Characterizing the distance and direction of
pollen and seed movement is therefore critical importance for
many areas of plant science. Pollination patterns will dictate
the reproductive neighborhood size for a plant, the connectivity
of populations and the impacts of habitat fragmentation. Pollen
movement will also determine the level of contamination of
seed orchards and the risk of gene flow from genetically modified crops to wild species. Seed dispersal, mediated by gravity,
148
Downloaded By: [Ashley, Mary V.] At: 01:45 18 May 2010
PLANT PARENTAGE, POLLINATION, AND DISPERSAL
wind, water, or animals, shapes not only gene flow patterns
but recruitment and demography of plant populations. The seed
dispersal kernel, that is the frequency distribution of the dispersal distances, influences many key demographic and ecological
processes, including colonization, population persistence and
plant community structure. Seed dispersal patterns will determine how rapidly invasive species may spread and how quickly
plants may shift their ranges in response to climate change, both
natural and human-induced.
Despite the importance of seed and pollen dispersal in plant
ecological and community dynamics, the study of these processes has always been challenging. There have been many attempts to study pollination by tracking the physical movement of
pollen using traps (Caron and Leblanc, 1992; Greenwood, 1986)
or dyes (Linhart et al., 1987; Waser and Price, 1982). Other
studies have followed the movements of pollinators (Levin and
Kerster, 1974; Parra et al., 1993; Walther-Hellwig and Frankl,
2000). Problems with these indirect approaches are both logistical and substantive. Logistically, pollen traps will collect whatever pollen the trap intercepts and in nearly all cases will be from
multiple individuals. Substantively, for most studies, it is the actual patterns of fertilization and gene flow rather than pollen
movement and deposition that are of interest. Identifying sires
through paternity assignment is the only direct way to retrace the
path of a pollen grain that has resulted in successful fertilization
of a seed, that is, a dispersal event that is genetically relevant.
Tracking the movement and fates of seeds has proven to be
at least as daunting as tracking pollen movement (Forget and
Wenny, 2005; Xiao et al., 2006). As with pollen, both the origin and destination of seeds must be identified to track seed
dispersal, and few techniques are efficient at both ends. Seed
traps, for example, document only the endpoint of dispersal, not
the origin. Many studies have tagged seeds to track their path
from source to destination, and ecologists have been ingenious
in devising tagging methods, attaching threads (Forget, 1992;
Jansen et al., 2004; Wenny, 2000), wire tin tags (Li and Zhang,
2003, 2007), or magnets (Alverson and Diaz, 1989; Sork, 1984)
to seeds, even spraying seeds with fluorescent microspheres
that can be recovered from fecal samples of birds (Levey and
Sargent, 2000). Seeds have also been labeled with radioactive
(Jensen and Nielsen, 1986; Vander Wall and Joyner, 1998) and
stable isotopes (Carlo et al., 2009). In such seed tagging studies, predation is often so intense that few if any viable seeds
remain at seed stations. The number of dispersed seeds that
survive is typically extremely low and the recovery of these
few of perhaps thousands of marked seeds is often challenging (Jensen and Nielsen, 1986; Jorge and Howe, 2009; Li and
Zhang, 2007; Sork, 1983, 1984; Wenny, 2000). Tagging itself
can influence seed removal rates (Xiao et al., 2006). Aside from
logistical difficulties, the quandary with seed tagging studies
is the complexity of the seed dispersal cycle itself (Wang and
Smith, 2002), with processes such as secondary dispersal, seed
and seedling predation, and seedling competition jumbling the
match between initial seed deposition and the resulting distri-
149
bution of seedling recruits. Using genetic markers to establish
the parentage of established seedlings allows the researcher to
assess the end product of the seed dispersal cycle, and thus evaluate how the cumulative effects of these ecological processes
shape plant demography and vegetation structure. Given this,
advances in genetic marker technology that allow parentage assignment were needed before significant advances in pollen and
seed dispersal biology could occur.
II. MICROSATELLITES AND PARENTAGE ASSIGNMENT
It has been about a quarter century since the discovery
that short, tandem repeats were a common feature of eukaryotic genomes (Hamada et al., 1982; Tautz and Renz, 1984).
These DNA repeats typically have a two to six base pairs sequence repeated a dozen or more times and were dubbed microsatellites, but are also known as simple sequence repeats
(SSRs, Jacob et al., 1991) or short tandem repeats (STRs, Craig
et al., 1988). While first characterized in mammalian genomes
(Stallings et al., 1991; Weber and May, 1989), microsatellites
were subsequently found to be ubiquitous and widely dispersed
in plant genomes (Akkaya et al., 1992; Condit and Hubbell,
1991; Morgante and Olivieri, 1993). As sequence information
from the noncoding regions of genomes accumulated, an important characteristic of microsatellites emerged; microsatellites are
highly variable due to mutations involving the number of repeating units (Akkaya et al., 1992; Dow et al., 1995; Tautz, 1989;
Weber and May, 1989). This hypervariability, in conjunction
with the codominant inheritance of microsatellite alleles, provides a means of distinguishing individuals, and hence a rich
source of neutral genetic markers for mating systems studies,
including inference of parentage (Ashley and Dow, 1994). For
plant ecologists, this meant that parentage of seeds could be established by exclusion, allowing identification of pollen donors,
and in some cases, seed parents. While allozymes have also been
applied to parentage studies in plants (Ellstrand, 1984; Nason
and Hamrick, 1997; Schuster and Mitton, 2000), they generally
lack the levels of variability required for categorical assignment
of individual parents for most of the seeds sampled in a study
(Chakraborty et al., 1988).
Consider a typical microsatellite pollination study in a temperate tree. Most temperate trees are diploid, monoecious, and
wind-pollinated, and occur in low diversity forests. A study
site would be chosen where the density of the study species
would be low or moderate or perhaps a forest fragment would
be chosen that had a manageable number of reproductive adult
trees (<200). Adults, either all of those in a forest fragment,
or within a circumscribed area in the forest, would be sampled
(leaves or cambium tissue) and genotyped at a handful (5–10) of
microsatellite loci. Seeds would then be collected from a set of
‘focal maternal trees’ and also genotyped at these loci. At every
microsatellite locus, a seed would have two alleles, one inherited
from the seed parent and one from the pollen donor. Because
the seed parent has been genotyped, the maternal allele at each
locus is identified, and the genotypes of all the adult trees in the
Downloaded By: [Ashley, Mary V.] At: 01:45 18 May 2010
150
M. V. ASHLEY
stand are examined to find one (and hopefully only one) that
could have contributed all the paternal alleles. Variability across
these loci would likely provide sufficient exclusion probabilities (97–99%) to resolve paternity of seeds with a fair degree
of confidence. In this situation, the exclusion probability is the
probability of excluding an unrelated male as the father of an offspring, given that the mother is known. If none of the candidate
pollen donors provides a match, the researcher can conclude
the pollen source lies somewhere outside the fragment or study
site. By genotyping several hundred seeds, a detailed view of
the distribution of pollination distances, directions, and paternal reproductive success emerges. This scenario matches quite
well some of the earliest application of microsatellite parentage
analysis, and involved studies of American and European oaks
(Dow and Ashley, 1998a, b; Streiff et al., 1999). Modifications
of this study design also was used in an early microsatellite
study of insect-mediated pollination in a tropical tree (Chase
et al., 1996), with the major difference being the geographical
scale of sampling due to the low density of conspecific trees in
the tropics.
Seed dispersal presents some additional challenges for
parentage assignment studies. In the case of dispersed seeds or
seedlings, neither parent is known. Among the sampled adults
at a study site, a seed or seedling may have one parent in the
stand (pollen or seed parent), two parents (both pollen and seed
parent), or no parents (immigrant seed). Most plant species are
cosexual, either producing hermaphroditic flowers with both
male and female parts, or producing separate, monoecious male
and female flowers. For cosexual plants, it is generally impossible to determine whether an assigned parent is the seed
parent or the pollen parent when using nuclear microsatellites.
This problem is circumvented in dioecious species, because the
search will be for a compatible male/female pair (Bittencourt
and Sebbenn, 2007; Hardesty et al., 2005; Slavov et al., 2009).
Maternal and paternal origins could potentially be distinguished
based on organelle DNA (mitochondria or chloroplast) that is
uniparentally inherited (Lian et al., 2008; Ouborg et al., 1999),
but generally mtDNA and cpDNA are not variable enough for individual identification. A recent and more productive approach
has been the use of maternally inherited seed tissue, including
endocarp (Garcia et al., 2007; Godoy and Jordano, 2001; Isagi
et al., 2007; Jordano et al., 2007; Terakawa et al., 2009) pericarp
(Grivet et al., 2005), and seed wing (Jones et al., 2005) tissue,
or megagametophyte tissue of conifers (Iwaizumi et al., 2009;
Iwaizumi et al., 2007) to identify seed parents. This approach is
generally limited to intact seeds; maternal tissues are generally
not present in established seedlings.
III. LIMITATIONS AND CONSTRAINTS
It is important to note that microsatellites are not a panacea
for plant dispersal studies. There are several limitations of this
approach, and while some of these limitations will be overcome
with continued improvement in marker technology, others will
remain for the foreseeable future. The major limitations are the
following: 1) development of microsatellite markers, 2) accurate
genotyping, 3) paternity assignment, and 4) species biology.
The first limitation, marker development, has eased in recent years. This is usually the first step in any microsatellite
study, and involves constructing and screening a genomic library, and from the library, developing and optimizing a useful suite of microsatellite primer pairs. In the 1990s this first
step could be a daunting obstacle, as ecologists and population biologists often struggled to master techniques of cloning,
screening libraries, sequencing and genotyping. As improved
protocols emerged, notably the use of enriched libraries where
microsatellite-containing DNA fragments are selectively cloned
(Glenn and Schable, 2005; Zane et al., 2002), the development
step has been reduced from months to weeks. Researchers with
sufficient funding can outsource this step; there are currently
several companies that will produce microsatellites for a given
species for a few thousand dollars. Also, microsatellites are increasingly available for diverse species; recent issues of Molecular Ecology Resources typically will report new primers for 20
to 30 species, including many plant species. Since microsatellite loci often are transferable to closely related species (those
within the same genus and sometimes families), it is increasingly likely that microsatellite loci are already available for a
plant species of interest. Transfer of published microsatellite
loci to a new species, even if closely related, usually involves
some optimization (adjusted PCR conditions), but by-passes the
entire genomic library screening step.
Technical advances in genotyping (actual scoring allele sizes
of individuals) have been even more dramatic over the past
20 years. The first plant microsatellite studies used nucleotides
labeled with 35 S or other radionucleotides (Dow and Ashley,
1996). This method involved pouring and manipulating large
polyacrylamide gel, autoradiography, and handling and disposal
of radioactive waste. Silver staining (Streiff et al., 1998) was
an alternative but no less laborious option. Scoring of allele
sizes was done by eye, by comparing band sizes to the known
sizes of a labeled DNA ladder run on each gel. Fluorescent
labeling techniques combined with high-volume capillary DNA
sequencing machines has decreased time, improved throughput,
and improved accuracy of genotyping, although the expense of
genotyping is still relatively high, in some cases prohibitively
so. New software, usually developed by the manufacturers of
the DNA sequencing hardware, aids in ‘allele calling’ (sizing)
and data management, although in our experience this step is
not yet completely automated.
On first principles, parentage assignment is quite straightforward; in practice it can be quite thorny. A detailed review
of methods of parentage analysis is beyond the scope of this
paper, and I refer readers to several recent reviews on the subject (Blouin, 2003; Jones and Ardren, 2003; Jones et al., 2010).
Some considerations are of particular relevance to pollination
and seed dispersal studies. In pollination studies, if seeds are
collected directly from a maternal tree, the maternal genotype
Downloaded By: [Ashley, Mary V.] At: 01:45 18 May 2010
PLANT PARENTAGE, POLLINATION, AND DISPERSAL
can be evaluated and the paternal parent (with a matching paternal allele at each locus) can be identified among the pool
of candidate plants. This approach, called strict exclusion, is
often used in pollination studies. However genotyping errors,
null alleles, and (rarely) mutations can lead to false exclusions
(Slavov et al., 2005). Likelihood-based parentage techniques
have become even more popular than exclusion, with the software CERVUS (Kalinowski et al., 2007; Marshall et al., 1998)
the most widely used. These approaches generally accommodate
some degree of mistyping. Programs like CERVUS calculate a
logarithm of the likelihood ratio (LOD score) and assign parentage to the adult with the highest LOD score. Confidence of assignments is estimated by simulation. While popular, CERVUS
can be problematic for plant parentage studies. For example, it
requires the user to input the estimated proportion of candidate
fathers, a value that is often unknown (and of primary interest) in studies of pollen and seed dispersal. Fractional paternity
assignment methods are also available. These approaches split
an offspring among all compatible mates (Devlin et al., 1988;
Neff et al., 2001; Nielsen et al., 2001), and can be used when
exclusion power is too low for categorical assignments. While
fractional techniques have some statistical advantages over categorical assignment for estimating population parameters, I do
not include them in this review because my focus is on precise
tracking of pollen and seed dispersal events.
The larger the pool of candidate parents, the higher the likelihood of Type I and Type II errors in parentage assignment. Type I
errors, also referred to as cryptic gene flow, occurs when a seed
matches a candidate parent by chance, when the true parent is
a plant that was not sampled (false-positive). The likelihood of
Type I errors can be estimated, and rates of pollen immigration
can be corrected (Dow and Ashley, 1996; Slavov et al., 2005).
Type II errors are trickier to accommodate. They occur when
a true parent is excluded due to mistyping at one or more loci,
or even due to mutation. Microsatellite genotyping errors will
occur, either in the laboratory processing, scoring, or data entry.
Technical problems such as null alleles and allelic drop-out will
occur in many microsatellites studies. Fortunately, simulation
studies have shown that parentage assignment approaches such
as CERVUS are quite robust to Type II errors (Oddou-Muratorio
et al., 2003). While rates of genotyping errors can be estimated
(for example, by the frequency of mismatches between a known
seed parent and her seeds), larger numbers of candidate parents will require more microsatellite loci to be used to provide
the needed resolution, which in turn increases the likelihood
of ‘false mismatches.’ Likelihood methods of parentage assignment are better than strict exclusion at accommodating Type II
errors.
Paternity assignment is most straightforward on diploid
species. An offspring (seed) inherits a single paternal allele
from the father, and thus assignment or exclusion of a putative
father is uncomplicated. Another advantage of diploid plants
is that simplicity of scoring genotypes; each individual should
have only one (if homozygote) or two (if heterozygote) alleles
151
at each microsatellite locus. Extra ‘bands’ or ‘peaks’ are suspect and signal technical or scoring issues that can be addressed
through further optimization. Because an estimated 30–50%
of plant species are polyploid (Grant, 1971; Stebbins, 1971),
limiting studies to diploids represents a serious drawback. Unfortunately, inheritance of microsatellites in polyploidy plants
is far from straightforward. Some polyploidy plant genomes are
described as ‘highly diploidized’ because the polyploidy events
are ancient, and chromosomes now form bivalents at meiosis. In
such species, microsatellites might exhibit diploid inheritance,
as in the case of the octaploid strawberry, Fragaria virginiana
(Ashley et al., 2003). In many polyploidy species, scoring an unknown number of alleles and inferring polysomic transmission
patterns will greatly complicate parentage assignment (Hanson
et al., 2007; 2008).
Microsatellite parentage assignment studies are also limited
by the abundance and density of candidate parents. One needs
to sample and genotype most if not all of the candidate parents
in an area relevant to pollen or seed dispersal distances, which
will simply not be possible for plants that occur in densities
of tens to hundreds per square meter or thousands per hectare.
The costs and time involved in genotyping are prohibitive and
even microsatellite markers do not provide enough resolution to
distinguish among thousands of candidate parents. Therefore,
most studies conducted to date have involved forest trees or
other larger plants that occur at low densities (Tables 1 and 2).
IV.
OVERVIEW OF MICROSATELLITE PARENTAGE
STUDIES
To evaluate what conclusions could be drawn from microsatellite parentage studies to date, I conducted a literature
search. I compiled studies that combine microsatellite genotyping and parentage analysis to directly track either pollen or
seed dispersal events, or both (Tables 1 and 2) in native plants.
I searched using Google scholarTM and ISI Web of ScienceTM
to search for articles and tracked references within those articles. I only included studies that were able to assign parentage
to a sizeable sample of genotyped offspring. I did not include
studies that use microsatellite genotyping with indirect methods
such as mating models, spatial genetic structure, or pollen pool
heterogeneity (Smouse et al., 2001) unless they also included
parentage assignment. I also did not include studies involving
cultivated trees or crops. This literature review is not exhaustive,
but I am confident that it is broadly inclusive and representative.
There are a total of 53 articles listed between both Table 1 and
Table 2. The first papers were published in 1996, but only four
papers appeared in the 1990s. Fifteen papers were published in
the period from 2000-2005 and 33 from 2006-2009, showing an
acceleration in publication rate using these approaches. These
studies employed an average of seven microsatellite loci (range
3-11); I did not include an early study (Dawson et al., 1997) that
used a single locus because paternities could not be assigned. I
found 41 papers that measured pollen dispersal by microsatellite
152
M. V. ASHLEY
TABLE 1
Studies that use microsatellites and parentage assignment to track effective pollen movement
Species
Distribution//Study site
Downloaded By: [Ashley, Mary V.] At: 01:45 18 May 2010
Wind pollinated trees
Fraxinus
Temperate/deforested
excelsior
landscape in Scotland
Juglans
Temperate /northern China
mandshurica
Pollen
Breeding System Immigration
mixed
43–68%
monoecious,
28.5%
heterodichogamous,
self-compatible
dioecious
10–47%
Araucaria
angustifolia
Tropical conifer/Brazilian
forest fragment
Araucaria
angustifolia
Tropical conifer/Brazilian
continuous forest
dioecious
54–91%
Quercus
macrocarpa
Temperate/Midwestern US
fragments
monoecious
47–58%
Quercus
macrocarpa
Quercus
macrocarpa
Quercus petraea
Temperate/Midwestern US
fragments
Temperate/Midwestern US
fragments
Temperate/Mixed stand in
continuous forest, France
monoecious
monoecious
71% from
>150 m
57% from
>150 m
69%
Quercus petraea
Temperate/Mixed stand in
central Spain
monoecious
38%
Quercus
pyrenaica
Temperate/Mixed stand in
central Spain
monoecous
34%
Quercus robur
Temperate/Mixed stand in
continuous forest, France
monoecious
65%
Quercus salicina
monoecious
52.2%
monoecious
52.1%
Quercus lobata
Temperate/southern Japan
reserve
Temperate/southern Japan
reserve
Temperate/California
monoecious
∼20%
Quercus
semiserrata
Temperate/Forest fragment in monoecious
Thailand
∼30%
Fagus crenata
Temperate/northern Japan
fragment
monoecious
8%
Pinus densiflora
Temperate/northern Japan
monoecious
67% from
>100 m
Quercus salicina
monoecious
Distance
Reference
80% <100 m, up to
Bacles and
2.9 km
Ennos, 2008
Median 11–14.9 m, up Bai et al., 2007
to 140 m within stand
Mean 83 m within
fragment, up to 2 km
from outside
Mean 102 m within
transect, up to 350 m
within transect
Mean 42 to 70 m within
stand, 100 m from
outside
Mean 77 m within stand
Bittencourt and
Sebbenn,
2007
Bittencourt and
Sebbenn,
2008
Craft and
Ashley,
submitted
Dow and
Ashley, 1996
Mean 76 m within stand Dow and
Ashley, 1998a
Means 22–58 m within Streiff et al.,
stand for individual
1999
trees
Mean 92 m within stand ValbuenaCarabaña
et al., 2005
Mean 270 m within
Valbuenastand
Carabaña
et al., 2005
Means 18–64 m within Streiff et al.,
stand for individual
1999
trees
Mean 66.7 m within
Nakanishi et al.,
stand
2004
Mean 69.2 m within
Nakanishi et al.,
stand
2009
Mean 114 m within
Pluess et al.,
study site
2009
Mean 52 within stand, Pakkad et al.,
up to 570 within
2008
stand
Mean 37 m within
Hanaoka et al.,
stand, up to 700 m
2007
within stand
Not reported
Iwaizumi et al.,
2009
(Continued on next page)
153
PLANT PARENTAGE, POLLINATION, AND DISPERSAL
TABLE 1
Studies that use microsatellites and parentage assignment to track effective pollen movement (Continued)
Species
Distribution//Study site
Distance
Reference
Pinus sylvestris
Temperate conifer/central
Spain isolated relict
monoecious
4.3% from
>30 km
Cercidiphyllum
japonicum
Populus
trichocarpa
Populus nigra
Temperate/riparian forest in
Japan
Temperate/Two contrasting
sites in Oregon, US
Temperate/Czech Republic
dioecious
30%
dioecious
32–54%
Mean 47 m within
stand, up to 629 m
within stand
Mean 129 m within
stand, up to 666 m
Mean 7.6 km
dioecious
23%
10–230 m
monoecious
50%
Mean 88.6 km, up to
160 km
Mean 30-50 m within
stands
Mean 142 m within
study site, up to
350 m within study
site
Mean 1288 m for
pasture trees, 417 m
for fragment trees, up
to 3.2 km within
study site
Mean ∼100 m for
outcross pollen
within study site
Means 240–557 m per
habitat, up to 1 km
within study site
Mean 345 m, up to
740 m within study
site
Mean 21 m within site,
up to 100 m within
study site
Means 134 and 59 m,
up to 528 m within
site
Mean 180 m within
site, up to 700 m
within site
Mean 131 m within
site, up to 121 m
Mean 223 within site
Insect pollinated Trees
Ficus sycomorus Arid/ Ugab River, Namibia
Downloaded By: [Ashley, Mary V.] At: 01:45 18 May 2010
Pollen
Breeding System Immigration
Eucalyptus
wandoo
Pithecellobium
elegans
Arid/Fragmented landscape,
Western Australia
Tropical/Costa Rican forest
fragment
hermaphroditic,
55.3% from
self-compatible
>350 m
hermaphroditic
28–41%
Dinizia excelsa
Tropical/Amazonian forest
fragments and pasture
hermaphroditic
68%
Prunus mahaleb
Arid/Southeastern Spain
gynodioecious
9.5% from
>1.5 km
Tropical/Costa Rican intact
forest, fragments and
pasture
Simarouba amara Tropical/Barro Colorado
Island, Panama
monoecious
∼59%
dioecious
na
Centaurea
corymbosa
monoecious, self
incompatible,
monocarpic
hermaphroditic
17%
Dipteryx
panamensis
Arid/ French Mediterranean
Sorbus torminalis Temperate/two fragmented
European forest sites
48% and
4%
Aesculus
turbinata
Temperate/Japan, riparian
forest
hermaphroditic
49%
Magnolia
obovata
Dipterocarpus
tempehes
Temperate/Japan
hermaphroditic,
protogynous
hermahroditic
40–57%
Tropical, Southeast Asia
∼11%
RobledoArnuncio and
Gil, 2005
Sato et al., 2005
Slavov et al.,
2009
Pospı́šková and
Šálková, 2006
Ahmed et al.,
2009
Byrne et al.,
2008
Chase et al.,
1996
Dick, 2001
Garcia et al.,
2005
Hanson et al.,
2008
Hardesty et al.,
2006
Hardy et al.,
2004
Hoebee et al.,
2007
Isagi et al., 2007
Isagi et al., 2000
Kenta et al.,
2004
(Continued on next page)
154
M. V. ASHLEY
TABLE 1
Studies that use microsatellites and parentage assignment to track effective pollen movement (Continued)
Species
Distribution//Study site
Temperate/Denmark
fragmented forest
Entandrophragma Tropical/Cameroon logged
cylindricum
forests
Swietenia humilis Tropical/Honduran
fragmented forest
Downloaded By: [Ashley, Mary V.] At: 01:45 18 May 2010
Malus sylvestris
Others
Calothamnus
quadrifidus
Symponia
globulifera
Hymenaea
courbaril
Rhododendron
metternichii
Rhododendron
metternichii
Polaskia chichipe
Bird pollinated Australian
shrub
Tropical/bird pollinated
Tropical/Bat pollinated
Amazonian tree
Temperate/Bee pollinated
shrub, Japan
Temperate/Bee pollinated
shrub, Japan
Insect and bird pollinated
columnar cactus, Mexico
Pollen
Breeding System Immigration
monoecious
monoecious
74% not
assigned
∼70%
monoecious
36–100%
hermaphroditic
10–33%
hermaphroditic
35%
hermaphroditic
55%
hermaphroditic
20–30%
hermaphroditic
27–53%
hermaphroditic
27%
paternity assignment in a total of 36 different species (Table 1).
Of the 36 species studied, 17 were wind-pollinated trees. Among
these, oaks (genus Quercus) were well represented with ten papers on seven oak species. There were 15 studies on 15 different
species of insect-pollinated trees. The remainder included a batpollinated tree, a bird-pollinated tree, a bird-pollinated shrub,
an insect-pollinated shrub, and a columnar cactus pollinated
by insects and birds. The list includes only two conifers and
studies of temperate trees outnumbered studies of tropical trees.
Studies of seed dispersal were fewer; I identified 15 papers that
examined seed dispersal in 13 species (Table 2). All were trees,
including one conifer. They included species with a range of
seed dispersal agents, including wind, autochory, and different
frugivores. A few studies are found in both Table 1 and Table 2
because they characterized both pollen and seed dispersal as
part of the same study.
V.
POLLEN DISPERSAL IN WIND-POLLINATED
SPECIES
As mentioned above, the first studies that used microsatellites and paternity assignment to study pollination were winddispersed temperate oaks. Results from these early studies were
surprising, and provided the first indication that much of what
we thought we knew about pollen dispersal, we didn’t know. The
classic view of wind-dispersed pollen envisioned a steep, lep-
Distance
Mean 60 m, up to
300 m within site
mean ∼330 m within
site, up to 2095 m
Mean 1.6 to >4.5 km
Reference
Larsen and Kjær
Lourmas et al.,
2007
White et al.,
2002
Up to 5km between
Byrne et al.,
sites
2007
Mean ∼900 m within
Carneiro et al.,
site
2009
Mean 827 m within site Lacerda et al.,
2008
Not reported
Kameyama
et al., 2000
Not reported
Kameyama
et al., 2001
Overall mean 113 m,
Otero-Arnaiz
for 3 mothers
et al., 2005
>1000 m
tokurtic distribution (Levin and Kerster, 1974). Because pollen
was thought to dissipate quickly in the air column, most pollinations were expected to be between neighboring individuals, with
the implication that most plants would be completely pollinated
with pollen from nearby sources (Ehrlich and Raven, 1969). The
dispersal kernel was thought to have a thin tail, so pollen immigration rates would be relatively low. For this reason, the results
of the first studies of wind pollination, conducted in bur oaks,
Quercus macrocarpa (Dow and Ashley, 1996, 1998a, b; Streiff
et al., 1999) came as quite a surprise. In a relatively isolated
remnant stand of bur oaks, over half the pollinations came from
trees outside the stand, at distances of greater than 150 meters
(Dow and Ashley, 1996, 1998b). Within the stand, pollinations
occurred nearly at random, with no or only a slight advantage
for nearest neighbors (Dow and Ashley, 1998b). Similar results
were soon reported in the European species Q. robur and Q.
petraea, with again over half the pollen immigrating from outside the stand, and mean pollination distances well above mean
nearest neighbor distances (Streiff et al., 1999). These early results in oaks have generally been confirmed as representative of
pollination patterns in many other tree species.
As mentioned above, parentage studies can only document dispersal within a circumscribed study site. Table 2 lists
pollen immigration rates as a percentage of the sampled seeds,
seedlings or juveniles that had all sampled adults excluded as
the parent, or for dioecious species, that had all males excluded.
155
PLANT PARENTAGE, POLLINATION, AND DISPERSAL
TABLE 2
Studies that use microsatellites and parentage assignment to track seed dispersal
Species
Distribution/Study Site
Fraxinus
excelsior
Temperate/deforested
landscape in Scotland
wind
Araucaria
angustifolia
Tropical
conifer/Brazilian
forest fragment
Temperate/Midwestern
US fragments
autochory
Prunus mahaleb
Arid/southeastern
Spain
Prunus mahaleb
Arid/southeastern
Spain
Prunus mahaleb
Arid/southeastern
Spain
Dipteryx
panamensis
Tropical/Costa Rican
intact forest and
fragments
Tropical/Barro
Colorado Island,
Panama
Temperate/Japan,
riparian forest
Temperate/northern
Japan
frugivorous birds
and other
vertebrates
frugivorous birds
and other
vertebrates
frugivorous birds
and other
vertebrates
frugivorous birds
and mammals
Downloaded By: [Ashley, Mary V.] At: 01:45 18 May 2010
Quercus
macrocarpa
Simarouba amara
Aesculus
turbinata
Pinus densiflora
Jacaranda copaia
Abies
sachalinensis
Prunus nigra
Cercidiphyllum
japonicum
Myrica rubra
Dispersal Agent
gravity and
vertebrates
frugivorous birds
and mammals
small mammals
wind
Tropical/Barro
Colorado Island,
Panama
Subboreal
conifer/Japan
wind
Temperate/Czech
Republic
Temperate/riparian
forest in Japan
Temperate/Japan
wind and water
wind
wind
frugivorous
vertebrate
Most of these studies sampled all adults within a fragment or
isolated remnant, or within a circumscribed area. For windpollinated species, pollen immigration is generally high, with
18 of 20 studies finding values of 20% or more, with an overall
average of about 44% across studies. These levels of gene flow
suggest that populations of wind-pollinated trees will be effec-
Sample, dispersal
results and distances
Established seedlings, ∼50% from
outside at >900 ha, seed mediated
gene flow exceeded pollen gene flow
Established seedlings/juveniles, no
seed immigration observed, mean
dispersal 83 m, up to 291 m
Saplings, 6–14% from outside at
>150 m, mean dispersal within
stand 22.8 m, up to 160m
Seed traps, 18% seed immigration,
mean dispersal within stand 6.1 m
up to 320 m
Seed traps, 20% seed immigration,
median dispersal within stand 14 m,
up to 990 m
Seed traps and fecal samples, up to
990 m, show differential dispersal
patterns by birds and mammals
Seed transects, only 7% genotyping
success, 14 seeds assigned that were
dispersed 63-853 m
Established seedlings, mean dispersal
distance 392 m, up to 1 km
Established seedlings, mean dispersal
distance 2.3 m, up to 218 m
Seed traps, 19% seed immigration
from >100 m, Median ∼15 m
within stand
Seed traps, 16% seed immigration,
mean dispersal distance 40-59 m, up
to 711 within site
Recrutiment on fallen logs, 19% seed
immigration, mean 24 m within
stand, up to 237 m
Established seedlings, dispersed up to
370 m
7% seed immigration, up to 302 m
within stand
Feces of macaques (Macaca fuscata),
mean 270 m, up to 634 m within site
Reference
Bacles et al.,
2006
Bittencourt and
Sebbenn, 2007
Dow and Ashley,
1996
Godoy and
Jordano, 2001
Garcia et al.,
2007
Jordano et al.,
2007
Hanson et al.,
2007
Hardesty et al.,
2006
Isagi et al., 2007
Iwaizumi et al.,
2009
Jones et al., 2005
Lian et al., 2008
Pospı́šková and
Šálková, 2006
Sato et al., 2005
Terakawa et al.,
2009
tively panmictic over large spatial scales, and that even small,
isolated fragments or remnants may not be reproductively isolated. There are two notable exceptions. A study of an extreme
isolated stand of Pinus sylvestris in central Spain (RobledoArnuncio and Gil, 2005) and a study of Fagus crenata in northern Japan (Hanaoka et al., 2007) found 4.3% and 8% immigrant
Downloaded By: [Ashley, Mary V.] At: 01:45 18 May 2010
156
M. V. ASHLEY
pollen, respectively. The Pinus sylvestris population was tens
of kilometers from the nearest populations, which makes any
level of pollen immigration impressive. Hanaoka et al. (2007)
suggest that the large size of F. crenata pollen may hinder longdistance dispersal; pollen size was not mentioned as a factor in
the other studies. Interestingly, there has been little or no evidence for directionality in pollen dispersed by wind; prevailing
wind direction does not seem to influence mating patterns in
these species (Dow and Ashley, 1998a; b; Pluess et al., 2009).
This counterintuitive result indicates that plausible assumptions
should not be taken at face value, and need to be empirically
tested.
What is the record for the longest pollination distance documented in a wind-pollinated plant? The answer depends on
whether one requires that the true father be identified, or simply inferred. If identification is required, several studies have
recorded mating events between trees separated by 600–700m
(Hanaoka et al., 2007; Pakkad et al., 2008; Robledo-Arnuncio
and Gil, 2005). Indeed, in most studies, the maximum withinstand pollination distance simply reflect the maximum potential
distances between trees within the study site. Black cottonwood,
Populus trichocarpa, would hold this particular record (Slavov
et al., 2009). The Vinson study site studied in western Oregon
covered an area of >300 km2 , made possible because of the low
density of trees in an arid landscape. The mean pollination distance at Vinson was 7.6 km, with many recorded matings over
10 km. If, on the other hand, the father need not be identified, the
Vinson population of P. trichocarpa is even more remarkable,
because approximately one-third of the pollinations at this site
were from immigrant pollen traveling more than 16 km. By this
relaxed criteria, however, the record may go to the extremely
isolated remnant stand of Pinus sylvestris in central Spain mentioned above (Robledo-Arnuncio and Gil, 2005). Although this
population only received 4.3% immigrant pollen, that pollen
necessarily traveled at least 30 km from the nearest conspecific
population.
In the case of pollen immigration from outside the study site,
only the minimum distances for immigrant pollen can be inferred. Actual distances can only be measured for pollinations
that occurred within the study site. Because such a large fraction
of the pollinations are from unknown distances, microsatellite
paternity studies have generally not been all that useful for describing the tail of the pollen dispersal kernel, except for the
obvious implication that the dispersal kernel is ‘fat-tailed.’ The
P. trichocarpa study (Slavov et al., 2009) is an interesting exception, because sampling at the Vinson study site covered such
a large geographic area that the long-distance component of the
dispersal curve could be characterized. Results indicated that exponential, exponential power, and Weibull functions (commonly
used to model pollen dispersal) are poor predictors of observed
pollinations, especially the long-distance component. Instead,
a model with a two-component process, with local dispersal
described by an exponential distribution and long-distance dispersal described by a uniform distribution, provided a much
better fit to the paternity data. A biological interpretation of this
may be that there exists a portion of wind-dispersed pollen that
follows traditional predictions of exponential declines from the
source and accounts for local pollinations, but a second, often
substantial component may be caught in updrafts and remain
airborne for some time, producing a regional pollen cloud that
effects pollinations over large distances. Paternity studies have
clearly shown that the ‘tail’ of the pollen dispersal curve is a
critical component that is difficult to measure and challenging
to model. Additional studies over large spatial scales are needed
to determine if pollen dispersal should indeed be modeled with
mixed dispersal functions; simple one- and two-parameter functions are likely too simplistic for what now appears to be a rather
complex process.
VI. POLLEN DISPERSAL IN ANIMAL-POLLINATED
SPECIES
Like the first studies in wind-pollinated trees, the first for
an insect-pollinated tree revealed surprisingly high pollination
distances. This was work on Pithecellobium elegans, a large
neotropical canopy tree pollinated by strong-flying hawkmoths.
At their study site in Costa Rica, Chase et al. (1996) found
that the average pollination distance was 142 m, whereas the
average distance of nearest neighbors was only 27 m, and “a
typical lepotokurtic distribution of pollination events was not
evident in this population.”
Using microsatellites and parentage assignment, several studies in tropical trees have tested and rejected Janzen’s ‘living
dead’ hypothesis (Janzen, 1986) that posits that remnant trees
left standing in pastures have no reproductive potential and are
of little conservation value. An important study by Dick (2001)
looked at both seed production and pollen donor diversity in a
prominent Amazonian tree, Dinizia excelsa, in remnant forest,
forest fragments, and isolated pasture trees. Contrary to the ‘living dead’ hypothesis, trees in pasture and forest fragments produced more than three times as many seeds as trees in continuous
forest. Even isolated pasture trees received pollen from multiple
pollen donors at distances of hundreds of meters, thanks to exotic African honeybees that were more common pollinators than
native bees in disturbed habitats. A similar finding was reported
for Swietenia humilis in disturbed habitats in Honduras (White
et al., 2002). Reduction in fragment size resulted in increased
proportions of long-distance pollinations, with one isolated tree
having most of its pollen delivered from >4.5 km. Pollen dispersal in Dipteryx panamensis in fragmented Costa Rican forests
also followed this pattern, with increased dispersal distances in
fragments and pastures compared to continuous forest (Hanson
et al., 2008). The conclusion from these studies is that physically
isolated trees are usually not reproductively isolated. This conclusion holds in quite different landscapes and for species with
a variety of pollinator syndromes. In the vast fragmented agricultural landscapes of southern Australia, Byrne et al. (2007)
report that the bird-pollinated shrub, Calothamnus quadrifidus,
Downloaded By: [Ashley, Mary V.] At: 01:45 18 May 2010
PLANT PARENTAGE, POLLINATION, AND DISPERSAL
received up to 43% of its pollen from other populations up to
5 km away. Extremely isolated stands may sometimes experience low levels of pollen immigration, as shown for Prunus mahaleb (Garcia et al., 2005; but see Herrera, 2009; Hoebee et al.,
2007) and Sorbus torminalis (Hoebee et al., 2007). However,
these cases seem to be an exception, with the general pattern being that physically isolated trees are not reproductively isolated
from conspecifics.
There have been a number of striking examples of longdistance pollen movement for insect-pollinated tropical trees,
including species with pollinators not expected to be strong
fliers. Hardesty et al. (2006) studied the dioecious Simarouba
amara on Barro Colorado Island, Panama, a species pollinated
by small generalist insects. They found an average pollination
distance of 334 m, more than six-fold greater than the mean
distances between nearest male-female pairs (54 m). Notable
mean pollination distances of hundreds of meters have been
reported for both temperate (Hoebee et al., 2007; Isagi et al.,
2000; Isagi et al., 2007) and tropical insect-pollinated trees
(Chase et al., 1996; Hanson et al., 2008; Hardesty et al., 2006).
Dinizia excelsa briefly held the record for the most distant documented insect pollination event, 3.2 km (Dick, 2001), until
4.5 km was reported for Swietenia humilis study the following
year (White et al., 2002). All records were shattered, however,
in a remarkable recent study of the African fig tree Ficus sycomorus that was sampled along 253 km of the lower Ugab River
in Namibia (Ahmed et al., 2009). Paternity assignment of 40
seeds revealed a mean pollination distance of 88 km, with a new
record of 160 km. Remarkably, F. sycomorus is pollinated by a
small host-specific fig wasp, Ceratosolen arabicus, which lives
only 48 hours and must be carried by wind. Unlike the findings for wind-pollinated trees, F. sycomorus shows a marked
east-to-west directionality of pollination flow matching the predominantly easterly winds.
Taken together, the results of paternity studies of pollination
offer solutions to at least two biological paradoxes. The first,
dubbed “Slatkin’s paradox,” is the contrast between direct and
indirect estimates of gene flow, where direct estimates suggest
very low levels of migration and gene flow but indirect measures
such as Wright’s FST indicate high levels of gene flow and little population subdivision (Ahmed et al., 2009; Slatkin, 1987).
The studies reviewed here suggest that solution to the paradox lies in underestimated dispersal by “direct methods,” which
until parentage assignment, including what really amounts to
indirect approaches, those that track pollen or pollinator movements by behavioral observations or the use of pollen dyes of
pollen traps. These approaches fail to adequately characterize
the extent of long distance dispersal. The second is the paradox of forest fragmentation genetics (Kramer et al., 2008). This
paradox involves the contradiction between population genetic
theory, which predicts genetic declines due to drift and inbreeding in forest fragments, and the lack of empirical evidence supporting these predictions. Again, the paradox is solved by the
frequency of long distance pollination revealed by parentage
157
studies. Gene flow occurs across large spatial scales and reproductive neighborhoods will generally extend beyond the boundaries of fragments. Furthermore, a large number of the studies
cited in Table 1 involve trees in fragmented and disturbed habitats, and in most cases the results indicate that pollen-mediated
gene flow is enhanced, not inhibited, by fragmentation.
VII. PARENTAGE ASSIGNMENT AND SEED DISPERSAL
There are fewer microsatellite parentage studies of seed dispersal than pollination, largely because of the challenges outlined above for identifying seed parents, and they vary more
broadly in their design and the type of data they have produced.
Therefore, generalizations from the studies listed in Table 2 are
a bit murkier than those of pollination studies, but similarly
suggest that seed dispersal is a complicated and multifaceted
process. The first study using microsatellites to study seed dispersal involved genotypes of established saplings of Quercus
macrocarpa (Dow and Ashley, 1996). Seed dispersal into the
study stand was 6%, far lower than pollen-mediated gene flow.
Within the stand, approximately half of the saplings had been
dispersed beyond the crown of the maternal tree, the rest were
growing under or near their seed parent. Several subsequent
studies listed in Table 2 have also found fairly low levels of
seed immigration (Bittencourt and Sebbenn, 2007; Jones et al.,
2005; Sato et al., 2005). In marked contrast, a study of Fraxinus
excelsior in Scotland reported very high seed mediated gene
flow (about 50% seed immigration), levels that exceeded gene
flow by pollen, for these wind-dispersed seeds (Bacles et al.,
2006). Impressive seed dispersal distances (hundreds of meters)
are reported in most studies, usually spanning the length of the
study site, but means were generally fairly low (tens of meters,
Table 2).
A study of Simarouba amara on Barro Colorado Island,
Panama, has important implications for tropical ecologists
(Hardesty et al., 2006). These authors found that germinated
seedlings were seldom the offspring of the nearest or nearby reproductive adults. They also showed that modeling approaches
based on seed arrival data into seed traps underestimated
seedling establishment distances by more than 10-fold. In their
study, seed dispersal distances were similar to pollen dispersal
distances. If these results are typical, they suggest that ecological
studies of seed establishment in the absence of genetic parentage studies may result in quite inaccurate estimates of fecundity,
dispersal, and seedling competition.
One interesting study compared both pollen and seed movement across two life stages, seeds and saplings, in a deciduous
Japanese tree Aesculus turbinate (Isagi et al., 2007). Selection was observed against seedlings that had resulted from selfpollinated and crosses between related trees (that were physically closer than unrelated trees). As a result, the effective pollen
movement increased from about 180 to 290 meters between
these two stages. This finding suggests that reproductive performance and effective pollen and seed dispersal should be assessed
158
M. V. ASHLEY
Downloaded By: [Ashley, Mary V.] At: 01:45 18 May 2010
at later growth stages. Another innovative study examined seed
dispersal in Prunus mahaleb by different classes of animals. This
was done by collecting seeds from feces, regurgitation pellets,
and seed traps in different microhabitats (Jordano et al., 2007).
The P. mahaleb seed dispersal kernel is comprised of at least
two components: short-distance dispersal by a diverse array of
small frugivores, and long-distance dispersal by medium-sized
birds and carnivorous mammals. Thus the multimodal nature of
dispersal kernels may hold for both pollen and seed.
VIII. CONCLUSIONS
The major surprise coming from of microsatellite paternity
studies that track pollination events is not that distance is associated with siring success; it usually is. As we have always
assumed, plants that are closer to each other are more likely to
be mating partners that plants than are more distant (although at
local scales even this generalization doesn’t always hold (Chase
et al., 1996; Dow and Ashley, 1996, 1998b). The major surprise is that distance explains only a portion of the variation
in mating patterns, often a relatively modest portion, and this
holds for both wind- and animal-pollinated plants. The traditional view of pollination biology, and the assumption of all
plant dispersal models, is that distance is paramount; it is usually the only parameter considered. The findings reviewed here,
however, suggest that models of seed and pollen dispersal based
simply on distance will provide poor predictions of plant gene
flow and dispersal patterns for many plants. While parentage
studies have certainly made the dispersal scene more complex,
the emerging paradigm of dispersal is much richer and multifaceted. There are intriguing and underappreciated ecological
processes influencing dispersal patterns that we can now begin
to explore. We should begin to look more carefully at processes
such as flowering phenology, wind column dynamics, pollen
properties, pollinator behaviors, and even mate choice to help
explain mating patterns in plants. Perhaps pollen has a maturation process, providing a fertility advantage for pollen that
has traveled greater distances. The distribution and density of
plants across the landscape certainly needs to be incorporated
into future models of pollen dispersal because several studies
reviewed here suggest that the pollination landscape changes
along with plant distributions. The ecology of seed dispersal,
including behaviors of multiple dispersers, secondary dispersal,
and density dependent effects has been considered and investigated previously, but more parentage studies of seedlings will
improve our understanding of how ecological processes contribute to resulting recruitment patterns. More studies in the
future should combine molecular parentage assignment with
the study of ecological processes in the field.
ACKNOWLEDGMENTS
I thank Henry F. Howe and Jeremie B. Fant as well as
my graduate students for their comments that improved the
manuscript.
REFERENCES
Ahmed, S., Compton, S. G., Butlin, R. K., and Gilmartin, P. M. 2009. Windborne insects mediate directional pollen transfer between desert fig trees 160
kilometers apart. Proc. Natl. Acad. Sci. (USA) 106: 20342–20347.
Akkaya, M. S., Bhagwat, A. A., and Cregan, P. B. 1992. Length polymorphisms
of simple sequence repeat DNA in soybean. Genetics 132: 1131–1139.
Alverson, W. S. and Diaz, A. G. 1989. Measurement of the dispersal of large
seeds and fruits with a magnetic locator. Biotropica 21: 61–63.
Ashley, M. V. and Dow, B. D. 1994. The use of microsatellite analysis in
population biology: Background, methods and potential applications. In:
Molecular Ecology and Evolution: Approaches and Applications. pp. 185–
201. Schierwater, B.S.B., Wagner, G. P., and DeSalle, R. Eds., Birkhauser
Verlag, Basel, Switzerland.
Ashley, M. V., Wilk, J. A., Styan, S.M.N., Craft, K. J., Jones, K. L., Feldheim,
K. A., Lewers, K. S., and Ashman, T. L. 2003. High variability and disomic segregation of microsatellites in the octoploid Fragaria virginiana
Mill. (Rosaceae). Theor. Appl. Genet. 107: 1201–1207.
Bacles, C.F.E. and Ennos, R. A. 2008. Paternity analysis of pollen-mediated
gene flow for Fraxinus excelsior L. in a chronically fragmented landscape.
Heredity 101: 368–380.
Bacles, C.F.E., Lowe, A. J., and Ennos, R. A. 2006. Effective seed dispersal
across a fragmented landscape. Science 311: 628–628.
Bai, W. N., Zeng, Y. F., and Zhang, D. Y. 2007. Mating patterns and pollen
dispersal in a heterodichogamous tree, Juglans mandshurica (Juglandaceae).
New Phytol. 176: 699–707.
Bittencourt, J.V.M. and Sebbenn, A. M. 2007. Patterns of pollen and seed dispersal in a small, fragmented population of the wind-pollinated tree Araucaria
angustifolia in southern Brazil. Heredity 99: 580–591.
Bittencourt, J.V.M. and Sebbenn, A. M. 2008. Pollen movement within a continuous forest of wind-pollinated Araucaria angustifolia, inferred from paternity
and TwoGENER analysis. Conserv. Genet. 9: 855–868.
Blouin, M. S. 2003. DNA-based methods for pedigree reconstruction and kinship analysis in natural populations. Trends Ecol. Evol. 18: 503–511.
Byrne, M., Elliott, C. P., Yates, C., and Coates, D. J. 2007. Extensive pollen
dispersal in a bird-pollinated shrub, Calothamnus quadrifidus, in a fragmented
landscape. Mol. Ecol. 16: 1303–1314.
Byrne, M., Elliott, C. P., Yates, C. J., and Coates, D. J. 2008. Maintenance of
high pollen dispersal in Eucalyptus wandoo, a dominant tree of the fragmented agricultural region in Western Australia. Conserv. Genet. 9: 97–
105.
Carlo, T. A., Tewksbury, J. J., and del Rio, C. M. 2009. A new method to track
seed dispersal and recruitment using N-15 isotope enrichment. Ecology 90:
3516–3525.
Carneiro, F. S., Degen, B., Kanashiro, M., de Lacerda, A.E.B., and Sebbenn,
A. M. 2009. High levels of pollen dispersal detected through paternity analysis from a continuous Symphonia globulifera population in the Brazilian
Amazon. For. Ecol. Manag. 258: 1260–1266.
Caron, G. E. and Leblanc, R. 1992. Pollen contamination in a small black
spruce seedling seed orchard for three consecutive years. For. Ecol. Manag.
53: 245–261.
Chakraborty, R., Meagher, T. R., and Smouse, P. E. 1988. Parentage analysis
with genetic markers in natural populations. I. The expected proportion of
offspring with unambiguous paternity. Genetics 118: 527.
Chase, M. R., Moller, C., Kesseli, R., and Bawa, K. S. 1996. Distant gene flow
in tropical trees. Nature 383: 398–399.
Condit, R. and Hubbell, S. P. 1991. Abundance and DNA sequence of two-base
repeat regions in tropical tree genomes. Genome 34: 66–71.
Craig, J., Fowler, S., Burgoyne, L. A., Scott, A. C., and Harding, H.W.J.
1988. Repetitive deoxyribonucleic acid (DNA) and human genome variation: A concise review relevant to forensic biology. J. Forensic Sci. 33: 1111–
1126.
Dawson, I. K., Waugh, R., Simons, A. J., and Powell, W. 1997. Simple sequence
repeats provide a direct estimate of pollen-mediated gene dispersal in the
tropical tree Gliricidia sepium. Mol. Ecol. 6: 179–183.
Downloaded By: [Ashley, Mary V.] At: 01:45 18 May 2010
PLANT PARENTAGE, POLLINATION, AND DISPERSAL
Devlin, B., Roeder, K., and Ellstrand, N. C. 1988. Fractional paternity assignment: theoretical development and comparison to other methods. Theor. Appl.
Genet. 76: 369–380.
Dick, C. W. 2001. Genetic rescue of remnant tropical trees by an alien pollinator.
Proc. R. Soc. Lond. B Biol. Sci. 268: 2391–2396.
Dow, B. D. and Ashley, M. V. 1996. Microsatellite analysis of seed dispersal
and parentage of saplings in bur oak, Quercus macrocarpa. Mol. Ecol. 5:
615–627.
Dow, B. D. and Ashley, M. V. 1998a. Factors influencing male mating success
in bur oak, Quercus macrocarpa. New For. 15: 161–180.
Dow, B. D. and Ashley, M. V. 1998b. High levels of gene flow in bur oak
revealed by paternity analysis using microsatellites. J. Hered. 89: 62–70.
Dow, B. D., Ashley, M. V., and Howe, H. F. 1995. Characterization of highly
variable (GA/CT)(N) microsatellites in the bur oak, Quercus macrocarpa.
Theor. Appl. Genet. 91: 137–141.
Ehrlich, P. R. and Raven, P. H. 1969. Differentiation of populations. Science
165: 12287–11232.
Ellstrand, N. C. 1984. Multiple paternity within the fruits of the wild radish,
Raphanus sativus. Am. Nat. 123: 819–828.
Forget, P. M. 1992. Seed removal and seed fate in Gustavia superba (Lecythidaceae). Biotropica 24: 408–414.
Forget, P. M. and Wenny, D. 2005. How to elucidate seed fate? A review of
methods used to study seed removal and secondary seed dispersal. In: Seed
Fate: Seed Predation, Seed Dispersal and Seedling Establishment. pp. 379–
393. Forget, P. M., Lambert, J., Hulme, P. E., and Vander Wall, S. B. Eds.,
CABI Publishing, Wallingford.
Garcia, C., Arroyo, J. M., Godoy, J. A., and Jordano, P. 2005. Mating patterns,
pollen dispersal, and the ecological maternal neighbourhood in a Prunus
mahaleb L. population. Mol. Ecol. 14: 1821–1830.
Garcia, C., Jordano, P., and Godoy, J. A. 2007. Contemporary pollen and seed
dispersal in a Prunus mahaleb population: patterns in distance and direction.
Mol. Ecol. 16: 1947–1955.
Glenn, T. C. and Schable, N. A. 2005. Isolating microsatellite DNA loci. Methods Enzymol. 395: 202–222.
Godoy, J. A. and Jordano, P. 2001. Seed dispersal by animals: exact identification
of source trees with endocarp DNA microsatellites. Mol. Ecol. 10: 2275–
2283.
Grant, V. 1971. Plant Speciation. Columbia University Press, New York,
London.
Greenwood, M. S. 1986. Gene exchange in loblolly pine: the relation between
pollination mechanism, female receptivity and pollen availability. Am. J. Bot.
73: 1443–1451.
Grivet, D., Smouse, P. E., and Sork, V. L. 2005. A novel approach to an old
problem: tracking dispersed seeds. Mol. Ecol. 14: 3585–3595.
Hamada, H., Petrino, M. G., and Kakunaga, T. 1982. A novel repeated element with Z-DNA-forming potential is widely found in evolutionarly diverse
eukaryotic genomes. Proc. Natl. Acad. Sci. (USA) 79: 6465–6469.
Hanaoka, S., Yuzurihara, J., Asuka, Y., Tomaru, N., Tsumura, Y., Kakubari,
Y., and Mukai, Y. 2007. Pollen-mediated gene flow in a small, fragmented
natural population of Fagus crenata. Can. J. Bot. 85: 404–413.
Hanson, T., Brunsfeld, S., Finegan, B., and Waits, L. 2007. Conventional and
genetic measures of seed dispersal for Dipteryx panamensis (Fabaceae) in
continuous and fragmented Costa Rican rain forest. J. Trop. Ecol. 23: 635–
642.
Hanson, T. R., Brunsfeld, S. J., Finegan, B., and Waits, L. P. 2008. Pollen
dispersal and genetic structure of the tropical tree Dipteryx panamensis in a
fragmented Costa Rican landscape. Mol. Ecol. 17: 2060–2073.
Hardesty, B. D., Dick, C. W., Kremer, A., Hubbell, S., and Bermingham, E.
2005. Spatial genetic structure of Simarouba amara Aubl. (Simaroubaceae),
a dioecious, animal-dispersed neotropical tree, on Barro Colorado Island,
Panama. Heredity 95: 290–297.
Hardesty, B. D., Hubbell, S. P., and Bermingham, E. 2006. Genetic evidence of
frequent long-distance recruitment in a vertebrate-dispersed tree. Ecol. Lett.
9: 516–525.
159
Hardy, O. J., Gonzalez-Martinez, S. C., Freville, H., Boquien, G., Mignot, A.,
Colas, B., and Olivieri, I. 2004. Fine-scale genetic structure and gene dispersal
in Centaurea corymbosa (Asteraceae) I. Pattern of pollen dispersal. J. Evol.
Biol. 17: 795–806.
Herrera, C. M. 2009. A comment on Garcia et al. (2005,2007) and related
papers on mating patterns and gene dispersal in Prunus mahaleb. Mol. Ecol.
18: 4533–4535.
Hoebee, S. E., Arnold, U., Duggelin, C., Gugerli, F., Brodbeck, S., Rotach, P.,
and Holderegger, R. 2007. Mating patterns and contemporary gene flow by
pollen in a large continuous and a small isolated population of the scattered
forest tree Sorbus torminalis. Heredity 99: 47–55.
Isagi, Y., Kanazashi, T., Suzuki, W., Tanaka, H., and Abe, T. 2000. Microsatellite
analysis of the regeneration process of Magnolia obovata Thunb. Heredity
84: 143–151.
Isagi, Y., Saito, D., Kawaguchi, H., Tateno, R., and Watanabe, S. 2007. Effective
pollen dispersal is enhanced by the genetic structure of an Aesculus turbinata
population. J. Ecol. 95: 983–990.
Iwaizumi, M. G., Takahashi, M., Watanabe, A., and Ubukata, M. 2009. Simultaneous evaluation of paternal and maternal immigrant gene flow and the
implications for the overall genetic composition of Pinus densiflora dispersed
seeds. J. Hered. 101: 144–153.
Iwaizumi, M. G., Watanabe, A., and Ubukata, M. 2007. Use of different seed
tissues for separate biparentage identification of dispersed seeds in conifers:
confirmations and practices for gene flow in Pinus densiflora. Can. J. For.
Res. 37: 2022–2030.
Jacob, H. J., Lindpaintner, K., Lincoln, S. E., Kusumi, K., Bunker, R. K., Mao,
Y. P., Ganten, D., Dzau, V. J., and Lander, E. S. 1991. Genetic mapping of
a gene causing hypertension in the stroke-prone spontaneously hypertensive
rat. Cell 67: 213–224.
Jansen, P. A., Bongers, F., and Hemerik, L. 2004. Seed mass and mast seeding
enhance dispersal by a neotropical scatter-hoarding rodent. Ecol. Monogr. 74:
569–589.
Janzen, D. H. 1986. The future of tropical ecology. Annu. Rev. Ecol. Syst. 17:
305–324.
Jensen, T. S. and Nielsen, O. F. 1986. Rodents as seed dispersers in a heath oak
wood succession. Oecologia 70: 214–221.
Jones, A. G. and Ardren, W. R. 2003. Methods of parentage analysis in natural
populations. Mol. Ecol. 12: 2511–2523.
Jones, A. G., Small, C. M., Paczolt, K. A., and Ratterman, N. L. 2010. A practical
guide to methods of parentage analysis. Mol. Ecol. Res. 10: 6–30.
Jones, F. A., Chen, J., Weng, G. J., and Hubbell, S. P. 2005. A genetic evaluation
of seed dispersal in the neotropical tree Jacaranda copaia (Bignoniaceae).
Am. Nat. 166: 543–555.
Jordano, P., Garcia, C., Godoy, J. A., and Garcia-Castano, J. L. 2007. Differential
contribution of frugivores to complex seed dispersal patterns. Proc. Natl.
Acad. Sci. (U S A) 104: 3278–3282.
Jorge, M. and Howe, H. F. 2009. Can forest fragmentation disrupt a conditional mutualism? A case from central Amazon. Oecologia 161: 709–
718.
Kalinowski, S. T., Taper, M. L., and Marshall, T. C. 2007. Revising how the computer program CERVUS accommodates genotyping error increases success
in paternity assignment. Mol. Ecol. 16: 1099–1106.
Kameyama, Y., Isagi, Y., Naito, K., and Nakagoshi, N. 2000. Microsatellite
analysis of pollen flow in Rhododendron metternichii var. hondoense. Ecol.
Res. 15: 263–269.
Kameyama, Y., Isagi, Y., and Nakagoshi, N. 2001. Patterns and levels of gene
flow in Rhododendron metternichii var. hondoense revealed by microsatellite
analysis. Mol. Ecol. 10: 205–216.
Kenta, T., Isagi, Y., Nakagawa, M., Yamashita, M., and Nakashizuka, T.
2004. Variation in pollen dispersal between years with different pollination conditions in a tropical emergent tree. Mol. Ecol. 13: 3575–
3584.
Kramer, A. T., Ison, J. L., Ashley, M. V., and Howe, H. F. 2008. The paradox of
forest fragmentation genetics. Conserv. Biol. 22: 878–885.
Downloaded By: [Ashley, Mary V.] At: 01:45 18 May 2010
160
M. V. ASHLEY
Lacerda, A.E.B., Kanashiro, M., and Sebbenn, A. M. 2008. Long-pollen movement and deviation of random mating in a low-density continuous population
of a tropical tree Hymenaea courbaril in the Brazilian Amazon. Biotropica
40: 462–470.
Larsen, A. S. and Kjær, E. D. 2009. Pollen mediated gene flow in a native
population of Malus sylvestris and its implications for contemporary gene
conservation management. Conserv. Genet.: 1–10.
Levey, D. J. and Sargent, S. 2000. A simple method for tracking vertebratedispersed seeds. Ecology 81: 267–274.
Levin, D. A. and Kerster, H. W. 1974. Gene flow in seed plants. Evol. Biol. 7:
139–220.
Li, H. J. and Zhang, Z. B. 2003. Effect of rodents on acorn dispersal and survival
of the Liaodong oak (Quercus liaotungensis Koidz.). For. Ecol. Manag. 176:
387–396.
Li, H. J. and Zhang, Z. B. 2007. Effects of mast seeding and rodent abundance
on seed predation and dispersal by rodents in Prunus armeniaca (Rosaceae).
For. Ecol. Manag. 242: 511–517.
Lian, C., Goto, S., Kubo, T., Takahashi, Y., Nakagawa, M., and Hogetsu, T.
2008. Nuclear and chloroplast microsatellite analysis of Abies sachalinensis
regeneration on fallen logs in a subboreal forest in Hokkaido, Japan. Mol.
Ecol. 17: 2948–2962.
Linhart, Y. B., Busby, W. H., Beach, J. H., and Feinsinger, P. 1987. Forager
behavior, pollen dispersal, and inbreeding in two species of hummingbirdpollinated plants. Evolution 41: 679–682.
Lourmas, M., Kjellberg, F., Dessard, H., Joly, H. I., and Chevallier, M. H. 2007.
Reduced density due to logging and its consequences on mating system and
pollen flow in the African mahogany Entandrophragma cylindricum. Heredity
99: 151–160.
Marshall, T. C., Slate, J., Kruuk, L.E.B., and Pemberton, J. M. 1998. Statistical
confidence for likelihood-based paternity inference in natural populations.
Mol. Ecol. 7: 639–655.
Morgante, M. and Olivieri, A. M. 1993. PCR-amplified microsatellites as markers in plant genetics. Plant J. 3: 175–182.
Nakanishi, A., Tomaru, N., Yoshimaru, H., Kawahara, T., Manabe, T., and
Yamamoto, S. 2004. Patterns of pollen flow and genetic differentiation among
pollen pools in Quercus salicina in a warm temperate old-growth evergreen
broad-leaved forest. Silvae Genet. 53: 258–264.
Nakanishi, A., Tomaru, N., Yoshimaru, H., Manabe, T., and Yamamoto,
S. 2009. Effects of seed- and pollen-mediated gene dispersal on genetic structure among Quercus salicina saplings. Heredity 102: 182–
189.
Nason, J. D. and Hamrick, J. L. 1997. Reproductive and genetic consequences of
forest fragmentation: two case studies of neotropical canopy trees. J. Hered.
88: 264–276.
Neff, B. D., Repka, J., and Gross, M. R. 2001. A Bayesian framework for
parentage analysis: the value of genetic and other biological data. Theor.
Popul. Biol. 59: 315–331.
Nielsen, R., Mattila, D. K., Clapham, P. J., and Palsboll, P. J. 2001. Statistical
approaches to paternity analysis in natural populations and applications to
the North Atlantic humpback whale. Genetics 157: 1673.
Oddou-Muratorio, S., Houot, M. L., Demesure-Musch, B., and Austerlitz, F.
2003. Pollen flow in the wildservice tree, Sorbus torminalis (L.) Crantz. I.
Evaluating the paternity analysis procedure in continuous populations. Mol.
Ecol. 12: 3427–3439.
Otero-Arnaiz, A., Casas, A., and Hamrick, J. L. 2005. Direct and indirect
estimates of gene flow among wild and managed populations of Polaskia
chichipe, an endemic columnar cactus in Central Mexico. Mol. Ecol. 14:
4313–4322.
Ouborg, N. J., Piquot, Y., and Van Groenendael, J. M. 1999. Population genetics,
molecular markers and the study of dispersal in plants. J. Ecol. 87: 551–
568.
Pakkad, G., Ueno, S., and Yoshimaru, H. 2008. Gene flow pattern and mating
system in a small population of Quercus semiserrata Roxb. (Fagaceae). For.
Ecol. Manag. 255: 3819–3826.
Parra, V., Vargas, C. F., and Eguiarte, L. E. 1993. Reproductive biology, pollen
and seed dispersal, and neighborhood size in the hummingbird-pollinated
Echeveria gibbiflora (Crassulaceae). Am. J. Bot. 80: 153–159.
Pluess, A. R., Sork, V. L., Dolan, B., Davis, F. W., Grivet, D., Merg, K., Papp, J.,
and Smouse, P. E. 2009. Short distance pollen movement in a wind-pollinated
tree, Quercus lobata (Fagaceae). For. Ecol. Manag. 258: 735–744.
Pospı́šková, M. and Šálková, I. 2006. Population structure and parentage analysis of black poplar along the Morava River. Can. J. For. Res. 36: 1067–
1076.
Robledo-Arnuncio, J. J. and Gil, L. 2005. Patterns of pollen dispersal in a small
population of Pinus sylvestris L. revealed by totalexclusion paternity analysis.
Heredity 94: 13–22.
Sato, T., Isagi, Y., Sakio, H., Osumi, K., and Goto, S. 2005. Effect of gene
flow on spatial genetic structure in the riparian canopy tree Cercidiphyllum
japonicum revealed by microsatellite analysis. Heredity 96: 79–84.
Schuster, W.S.F. and Mitton, J. B. 2000. Paternity and gene dispersal in limber
pine (Pinus flexilis James). Heredity 84: 348–361.
Slatkin, M. 1987. Gene flow and the geographic structure of natural populations.
Science 236: 787.
Slavov, G.T., Howe, G. T., Gyaourova, A. V., Birkes, D. S., and Adams, W. T.
2005. Estimating pollen flow using SSR markers and paternity exclusion:
accounting for mistyping. Mol. Ecol. 14: 3109–3121.
Slavov, G. T., Leonardi, S., Burczyk, J., Adams, W. T., Strauss, S. H., and
Difazio, S. P. 2009. Extensive pollen flow in two ecologically contrasting
populations of Populus trichocarpa. Mol. Ecol. 18: 357–373.
Smouse, P. E., Dyer, R. J., Westfall, R. D., and Sork, V. L. 2001. Two-generation
analysis of pollen flow across a landscape. I. Male gamete heterogeneity
among females. Evolution 55: 260–271.
Sork, V. L. 1983. Mammalian seed dispersal of picgnut hickory during three
fruiting seasons. Ecology 64: 1049–1056.
Sork, V. L. 1984. Examination of seed dispersal and survival in red oak, Quercus
rubra (Fagaceae), using metal-tagged acorns. Ecology 65: 1020–1022.
Stallings, R. L., Ford, A. F., Nelson, D., Torney, D. C., Hildebrand, C. E., and
Moyzis, R. K. 1991. Evolution and distribution of (GT)n repetitive sequences
in mammalian genomes. Genomics 10: 807–815.
Stebbins, G. L. 1971. Chromsomal Evolution in Higher Plants. Addison-Wesley,
Reading, MA.
Streiff, R., Ducousso, A., Lexer, C., Steinkellner, H., Gloessl, J., and Kremer,
A. 1999. Pollen dispersal inferred from paternity analysis in a mixed oak
stand of Quercus robur L. and Q. petraea (Matt.) Liebl. Mol. Ecol. 8: 831–
841.
Streiff, R., Labbe, T., Bacilieri, R., Steinkellner, H., Glössl, J., and Kremer, A.
1998. Within-population genetic structure in Quercus robur L. and Quercus
petraea (Matt.) Liebl. assessed with isozymes and microsatellites. Mol. Ecol.
7: 317–328.
Tautz, D. 1989. Hypervariability of simple sequences as a general source for
polymorphic DNA markers. Nucleic Acids Res. 17: 6463–6471.
Tautz, D. and Renz, M. 1984. Simple sequences are ubiquitous repetitive components of eukaryotic genomes. Nucleic Acids Res.12: 4127–4138.
Terakawa, M., Isagi, Y., Matsui, K., and Yumoto, T. 2009. Microsatellite analysis of the maternal origin of Myrica rubra seeds in the feces of Japanese
macaques. Ecol. Res. 24: 663–670.
Valbuena-Carabaña, M., González-Martı́nez, S. C., Sork, V. L., Collada, C.,
Soto, A., Goicoechea, P. G., and Gil, L. 2005. Gene flow and hybridisation in
a mixed oak forest (Quercus pyrenaica Willd. and Quercus petraea (Matts.)
Liebl.) in central Spain. Heredity 95: 457–465.
Vander Wall, S. B. and Joyner, J. W. 1998. Recaching of Jeffrey pine (Pinus
jeffreyi) seeds by yellow pine chipmunks (Tamias amoenus): potential effects
on plant reproductive success. Can. J. Zool. 76: 154–162.
Walther-Hellwig, K. and Frankl, R. 2000. Foraging distances of Bombus muscorum, Bombus lapidarius, and Bombus terrestris (Hymenoptera, Apidae). J.
Insect Behav. 13: 239–246.
Wang, B. C. and Smith, T. B. 2002. Closing the seed dispersal loop. Trends
Ecol. Evol. 17: 379–385.
PLANT PARENTAGE, POLLINATION, AND DISPERSAL
Downloaded By: [Ashley, Mary V.] At: 01:45 18 May 2010
Waser, N. M. and Price, M. V. 1982. A comparison of pollen and fluorescent dye
carry-over by Natural pollinators of Ipomopsis aggregata (Polemoniaceae).
Ecology 63: 1168–1172.
Weber, J. L. and May, P. L. 1989. Abundant class of human DNA polymorphisms
which can be typed using the polymerase chain reaction. Am. J. Hum. Genet.
44: 388–396.
Wenny, D. G. 2000. Seed dispersal, seed predation, and seedling recruitment of
a neotropical montane tree. Ecol. Monogr. 70: 331–351.
161
White, G. M., Boshier, D. H., and Powell, W. 2002. Increased pollen flow
counteracts fragmentation in a tropical dry forest: An example from Swietenia
humilis Zuccarini. Proc. Natl. Acad. Sci. U. S. A. 99: 2038–2042.
Xiao, Z. S., Jansen, P. A., and Zhang, Z. B. 2006. Using seed-tagging methods
for assessing post-dispersal seed fate in rodent–dispersed trees. For. Ecol.
Manag. 223: 18–23.
Zane, L., Bargelloni, L., and Patarnello, T. 2002. Strategies for microsatellite
isolation: a review. Mol. Ecol. 11: 1–16.