{.4
i1 ..
AL“? 1»
*5 u
n z...
E u.

t;

. v V

.. a}?
. GAB“ .
a. 5%
. $.fl.‘

. .r 5

.3:

:33:

Pan.
.5!

 

. P N
5.1.. 55%».

*ij , ”LIBRARY
" ,7 Michigan State

 

 

 

 

” .C) x . .
0" UniverSIty
This is to certify that the
dissertation entitled

Genetic Diversity of White Pine (Pinus strobus)
in the Beaver Island Archipelago

presented by

Eric Russel Myers

has been accepted towards fulfillment
of the requirements for

Doctoral degree in Philosophy

/630%kafig é:;@0éuy»\i
/

Major piloéssor

Date November 7, 2002

MS U is an Affirmative Action/Equal Opportunity Institution 0- 12771

PLACE IN RETURN Box to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJCIRCJDatoDuopes-sz

 

GENETIC DIVERSITY OF WHITE PINE (Pinus strobus) IN THE BEAVER ISLAND
ARCHIPELAGO

By

Eric R. Myers

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
For the degree of

DOCTOR OF PHILOSOPHY

Department of Forestry / Plant Breeding and Genetics

2002

ABSTRACT
GENETIC DIVERSITY OF WHITE PINE (Pinus strobus) IN THE BEAVER ISLAND
ARCHIPELAGO

By

Eric R. Myers

Geographical isolation is thought to permit divergence among populations by
disrupting gene flow. This study examined the long-term effects of fragmentation on
variable sized, insular populations of white pine (Pinus strobus). The four study sites
were connected to mainland Michigan roughly 10,000 y.b.p. and have been isolated from
the mainland and each other since approximately 6,000 y.b.p. Results determined that
white pine harbors high amounts of genetic diversity within populations, and little
diversity among fragmented populations. This study found slight divergence with
allozymes. The confidence interval for F st among islands was 0.00 to 0.03 and among
sample plots was 0.0] to 0.07. Diversity measures based on microsatellite markers also
indicated some divergence. The Rst value among islands was 0.01, and 0.03 among
sample plots. Estimates of inbreeding within islands were significantly different form
zero. The confidence interval for Fis among islands ranged from 0.02 to 0.12, but was
not significantly different from zero among sample plots. In addition, significant positive
spatial autocorrelation was found in distance classes less than 5 km and 8.5 km among
islands. Negative spatial autocorrelation was found at distances greater than 13.5 km
among islands. Within the largest island, Beaver Island, significant positive spatial

autocorrelation was found in distance classes less than 3.6 km, and a negative

autocorrelation was found at distance classes greater than 13.5 km. These results suggest
that the fragmentation and subsequent isolation has resulted in some divergence and
inbreeding within islands, but that a limited amount of migration has prevented more

severe divergence.

Acknowledgements

I wish to thank many people who helped me complete this project. First, my
graduate committee, Dr. B. Epperson, Dr. D. Dickmann, Dr. J. Hancock and Dr. F.
Telewski for not only the valuable help with this project but also for the knowledge
gained from them as instructors. I would also like to thank Dr. Dickmann for flying me
around the islands to survey for white pine. I am especially thankful for my wife, Carrie
Myers for her help as a "volunteer" field assistant and a GPS operator. Much thanks also
to Dr. Rosemarie Walter for teaching me many of the molecular laboratory techniques
used in this work, and all her help with the inevitable problems which arise in a
molecular laboratory. Thanks also to Dr. Myong Chung for running the starch gels and
for comments and suggestions through correspondence. Also, thanks to Shari Tjugum-
Holland for determining the SSR lengths and Robert Goodwin for providing me the GPS
equipment and helping me with data transfer. I thank the Central Michigan University
biological station for providing some transport to other islands, and for allowing me to
store samples at the biostation. I 'd like to thank Carol Graysmith for assistance with
computer sofiware and computer equipment and Dr. Scribner’s lab for allowing me to use
the flurometer. Last but not least, I thank my family and friends for their support and

encouragement.

iv

Table of contents (continued)

Spatial AutocorrelationAnalysis.....................................
Moran’sIStatistic

Chapter 4 Discussion
Discussion...

Write'sFStatistics...
Nei'sGeneticDistanc.e.............................
SpatialAutocorrelatio.n......................................................

Appendix

Allozyme Frequencies Among Plots...... .
...103

107
...111

Haplotype Frequencies Among Plots...
Diameter at Breast Height of Sampled Trees
Workscited”

vi

88
88

.... ..92
...97

98
.98

101

List of Tables

Table 1 Comparative effects on the genetic diversity ofpopulations... ..14
Table 2 Comparative effects on genetic measures ofpopulations... ....14
Table 3 Number ofallozyme alleles found on islands... .35
Table 4 Allele frequencies ofallozymes for islands... .. .36
Table 5 Descriptive statistics ofallozymes among islands... ...37
Table 6 Chi square test for Hardy-Weinberg..... .....38
Table 7 F Statistics ofallozymes among islands... ........39
Table 8 Fis among allozyme loci and islands... .... .40
Table9GeneDiversityofallozymesamongislands .. 41
Table 10 Nei’ 5 genetic distance and genetic identity among islands ......................... 42
Table 11 Descriptive Statistics of allozymes among Plots... .44
Table 12 Standard F Statistics ofallozymes among Plots... .. .45
Table 13 Nei’ 5 Genetic Distance and Genetic Identity among Islands ....46
Table 14 Haplotype frequencies among islands... ......49
Table 15 Descriptive statistics for haplotypes among islands .... ...50
Table 16 F st of haplotypes among islands... .. 51
Table 17 Nei' 5 genetic distance and genetic identity of haplotypes among islands ...52
Table 18 Descriptive statistics for haplotypes among plots... .. .. .. .. .55
Table 19 Fst of haplotypes among plots... ...56
Table 20 Nei' 5 genetic distance and genetic identity of haplotypes among plots ........... 57
Table 21 Descriptive statistics of allozymes within Garden Island... .. .....60
Table 22 Fis measured with allozymes within Garden Island... .... ...61
Table 23 Descriptive statistics of allozymes within Hog Island... ...63
Table 24 F statistics of alleles within Hog Island... .. .64
Table 25 Nei' 5 genetic distance and genetic identity of 5116511312; within hog island ...... 65
Table 26 Descriptive statistics of allozymes within High Island... . .67
Table 27 F statistics of alleles within High Island... ...68
Table 28 Nei' 3 genetic distance and genetic identity of allozymes within High Island... .69
Table 29 Descriptive statistics of allozymes within Beaver Island... . ..71
Table 30 F statistics of alleles within Beaver Island... .. .72
Table 31 Nei' 3 genetic distance and genetic identity of haplotypes within Beaver Island73
Table 32 Descriptive statistics measured with haplotypes within Garden Island... 7.5
Table 33 Descriptive statistics measures with haplotypes measures within Hogm Island...”
Table 34 F statistics measures within hog island... . .. .. .78
Table 35 Nei' 3 genetic distance and identity of haplotypes within hog islands... . .79
Table 36 Descriptive statistics measures with haplotypes measures with High Island .81
Table 37 F statistics measures within High Island” . .. .. .82
Table 38 Nei' 3 genetic distance and identity of haplotypes within High Islands ...83
Table 39 Descriptive statistics measures with haplotypes measures with High Island... .85
Table 40 F statistics measures within High Island... . .. 86
Table 41 Nei' 5 genetic distance and identity of haplotypes within Hrgh Islands ........... 87
Table42Moran'sIstatisticforthearchipelago. 89

Table 43 Moran's I statistic forBeaverIsland... .......91

vii

List of Figures

The Beaver Island Archipelago...

White pine distribution amongthe Beaverlsland Archrpelago...
...58

UPGMA phenogram ofallozymes within beaverisiand..fff.'IffffIffififffIfffIfffff...74

UPGMA phenogram of allozymes among plots...
UPGME phenogrom of haplotypes among plots.

viii

Introduction

Habitat Fragmentation

Habitat fragmentation, resulting in the formation of isolated ecological islands is
one of the most important issues facing wild species today (Saunders et al., 1991). In
almost every nation and continent, the previously continuous landscape is becoming a
fragmented patchwork of wilderness and human development (Chiras et al., 2002).
Habitats are being fragmented by agriculture, housing, commerce and other human
development. Habitat fragmentation can lead to a reduction in biodiversity (the number
of species in a given area) as useful or unique parts of the habitat are lost (Saunders et al.,
1991; Blake 1991). Along with the obvious loss of habitat due to human development,
fragmentation can also lead to a decrease in genetic diversity of the species in the
fragmented habitat. A loss of genetic diversity has been theorized to reduce the fitness of
a population and can lead to species extinction (Ledig, 1992)

Several factors can alter the genetic diversity of individuals in fragmented areas.
When a population’s numbers are small, it can be considered to be in a population
bottleneck. A population bottleneck can be described as a drastic reduction in the size of
the population. The individuals that survive the bottleneck will be the basis of the genetic
diversity for future generations in the population. The smaller the number of individuals
that survive the population bottleneck, the smaller the gene pool will be in future
subsequent generations. With a small gene pool, there will potentially be a loss of
variation in the genetic make up of individuals in the population. A bottleneck usually
results in a loss of heterozygosity and polymorphism as low frequency alleles are lost

from the gene pool. A potential indicator of past population bottlenecks is when the

number of alleles is lower compared to unaffected populations (Comuet and Luikart,
1996).

Fragmented and small populations are influenced by other factors that can affect
their genetic diversity. For instance, factors such as genetic drift and inbreeding, which
reduce genetic diversity in populations, are usually problematic in small populations.
Genetic drift is a random process where sampling between generations can ultimately fix
or lose specific genes in a population. Inbreeding results when individuals have
consanguineous matings. Conversely, other factors can increase genetic diversity in a
population such as mutation and migration. Mutation is a change in the genetic sequence
of an individual’s DNA and is the ultimate cause of genetic differences. Migration in this
text is the movement of seeds and/or pollen between populations. The overall genetic
diversity in a population will depend on the strength and intensity of the varying factors
as well as stochastic events.

The scope of this study will examine factors acting on variable sized island
populations of white pine (Pinus strobus) and the genetic diversity which results. This
study will use allozymes and DNA sequence variations (DNA length polymorphisms) in
the chloroplast genome to estimate genetic diversity among and within insular white pine
populations. The variable sized populations are located in northern Lake Michigan at
different distances from adjacent populations. The group of islands is collectively known
as the Beaver Island Archipelago (BIA). Perhaps the greatest utility of this study site is
that the BIA is a running model. It is a conglomeration of discrete, variable sized
populations that have been naturally fragmented for thousands of years. To explain, the

islands were exposed from the Laurentide ice sheet around 15,000 thousand years ago.

Figure 1 The Beaver Island Archipelago

v

 

ESE
.563—

   

    

.1.o.,,.‘..,., ., ... u. ...-a . o. 3.0.80. “Hus-91. ...... ....0 o o'.
_. 1...... . n. ,. ...-o .l... numwn.$ ‘9... up.
1.00.. . I. . . ..r ‘ “t .
o - 0.0.0,...130‘0Hoyuots... “.5,
. m. ...... . , ... Mr
.00 . . 0 .. us 11“»

3808 83 III...-
38. e ..5 8.» via-m I

   

. C

1..
. «-

     

   
  

l....... “...—9:

Figure 2 White Pine Distribution on the

 

Archi ela o

        

...—«.m—
hegem—

 

mug—mu
wen .

  

mas—mu.
apnoea—«U

“in.

.5582.

use 353 D

E88 2:; BE? Z

Then, around 10,000 years ago, the lake levels were dramatically reduced, (Hansel, et al.,
1985) and the islands would have been connected to each other and also to mainland
Michigan via a land bridge. The islands would have been part of mainland Michigan
intermittently for 4,000 years until around 6,000 years ago when the lake levels reached
a height similar to current lake levels. The Great Lakes reached a peak in lake levels
around 4,000 years ago during the Nipissing period, when the lake levels were roughly
7m above present day levels (Hansel, et al., 1985). The lake levels remained high enough
so that the islands would have been isolated before the Nipissing period. The lake levels
have been at present levels for around 2,500 years. Based on the geological history, it is
hypothesized that there will be a loss of genetic diversity among the archipelago.
Moreover, there should also be a greater loss of diversity within the small islands which
have smaller populations next to the Lake Michigan shoreline. Populations next to the
shoreline would have been severely affected by the fluctuating lake levels.

Time scales are of importance when studying and /or managing a species. Here,
we have a chance to study the effects of long-term fragmentation and possible isolation
under natural conditions with a relatively well known start to ecological succession. It is
uncommon to have a situation where the known ecological events can be compared with
the genetic diversity which results (Comps et al., 2001). This study will use two types of
assays to measure genetic diversity. Allozymes will be used to measure genetic diversity
of BIA, which will add to the vast array of studies that have used allozymes to measure
genetic diversity (See Hamrick and Godt, 1990). In addition, this study will measure
genetic diversity with microsatellites, a relatively new tool, and will be one of the first

studies to use chloroplast markers to examine white pine genetic diversity in fragmented

 

 

populations. This study can also be useful in that it will allow a direct comparison
between the traditional allozyme measures and the more recent microsatellite techniques.
Lastly, Moran’s I statistic will be used to examine spatial autocorrelation trends in
genetic markers in space among the archipelago, as well as within Beaver Island.

Forces Affecting a Population’s Genetic Diversity

Selection: One mechanism hypothesized by Darwin to affect differences among
individuals is termed “Natural Selection”. Selection was one of the first mechanisms
invoked to explain how the vast diversity of life found on this planet evolved. Basically,
natural selection is effective when genetic differences result in different phenotypes.
Different phenotypes will provide different abilities to organisms to survive; this is
termed absolute fitness, whether or not an individual survives. Selection can also act to
increase or decrease the number of surviving offspring. When survivorship is compared
to other members in the population, it is termed relative fitness.

Detection of selection acting specifically on genes or gene products can be
difficult. First, the effect of selection on a specific gene is usually weak so that the
consequences of selection ofien takes many generations to make noticeable changes in a
population. This problem can be compounded by the fact that the strength of selection
can fluctuate from season to season and year to year. Therefore, without long-term
observations it can be difficult to prove that selection is responsible for genetic
differences among individuals or populations. Furthermore, linkage disequilibrium can
obscure the effects of selection. Linkage disequilibrium is defined as two genes that are
not in random association (Hartl and Jones, 1998). If genes that are neutral in benefit or

even mildly deleterious are linked to beneficial genes, all linked genes increase in

frequency. Conversely, even beneficial genes can decrease in frequency if they are
linked to extremely harmful or lethal genes. This is one of the problems with studying
factors that affect individual loci. Selection is not viewed as a major force affecting the
genetic diversity in this study. First, the huge amount of variability in white pine
combined with the sampling scheme will not allow analysis of the effects due to
selection. Second, some of the genetic markers used in this study are from regions of the
genome that are non-coding, and therefore will not be directly affected by selection.
Mutation: The ultimate cause of genetic differences is mutation. Essentially, a
mutation results in a new or unique DNA sequence. Any change that occurs in the DNA
of germ line cells will be passed on to consecutive offspring. Most of the work on
mutations have been done on humans. Most of the germ line mutations in human genes
potentially represent errors of endogenous error-prone processes involving either
physical, chemical, or enzymatic mechanisms (Cooper and Krawczak, 1993 ). There are
many causes of mutations in the genetic material of an organism; ultraviolet light and
cosmic radiation are natural mutagens and there are numerous anthropogenic chemical
and organic compormds that are considered mutagens. Mutations also occur with some
frequency as errors during DNA replication. During replication, the polymerase
mechanism can slip, adding to or subtracting from the template sequence. A loss of
nucleotides from the original sequence is called a deletion, when one or more continuous
nucleotides are excised from the template sequence (Hale and Marghan, 1991). The
addition of one or more nucleotides between two adjacent nucleotides in the existing
DNA sequence is termed an insertion (Hale and Marghan, 1991). Mutations that result

from a single nucleotide change are often called point mutations. The impact of point

mutations may seem trivial, but the addition or subtraction of a single nucleotide in an
active gene can drastically change the structure/function of the resulting gene product.
Fragments of DNA can also be inserted or deleted from areas of an organism’s genome.
These fragments are often called indels, a hybrid term put together to describe sequence
length change when the direction of the change is unspecified. Moreover, it is possible
for DNA sequences to be altered by transposition, where DNA sequences excise and
insert autonomously. These sequences often leave behind DNA fragments in the “host”
genome. Sequences of DNA can also be changed when viruses insert and later excise,
leaving fragments of DNA behind.

Another form of DNA alteration occurs when pieces of a chromosome are moved
or rearranged. Chromosomal rearrangement is not uncommon during meiosis and is
termed chromosomal crossover. In addition, the structure and number of one or more
chromosomes can change due to improper segregation of gametes. With unequal
segregation and/or crossing over, perhaps more important than creating unique DNA
sequences is the possibility of creating multiple copies of genes, freeing the second gene
from the effects of selection. Multiple gene copies allows mutations to alter the second
gene/ gene product while the metabolic need is still met by the unmutated gene.

Mutations constantly create new alleles or DNA sequences, however, the initial
frequency of new mutations (alleles) is low (l/2N with N= the number of the diploids in
the population). Therefore, the effects of mutations on changing genetic diversity are
weak in large populations. In small populations, the effects of mutation can be quite
noticeable and effective in changing the genetic diversity. In the field of population

genetics the cause of DNA alteration is usually not the primary interest, rather,

population genetic studies are concerned with how these genetic differences change in
frequency.

Recombination: The act of recombination often occurs during chromosomal
crossing over. Recombination rarely occurs in the genomes of mitochondria and
chloroplasts and consequently should have no effect on the chloroplast markers used in
this study. The allozyme markers could be subject to recombination and a recombination
event would be detected in the laboratory.

Genetic Drift: A factor that can affect genome-wide genetic diversity in
fragmented population is termed genetic drift. Genetic drift can be the most important
factor controlling genetic diversity in small populations (less than 50 individuals) (Hartl
and Clark, 1997). Genetic drift can occur between seed establishment and reproductive
maturity. If many individuals are unable to reach reproductive maturity, then the
possibility exists for some genes to be lost or fixed from the population. Genetic drifi
also occurs between generations due to random fluctuations in gametic allele frequencies.
In other words, an almost infinite number of gametes are produced each generation, but
only a specific number of organisms result from a sample of the gametes. With fewer
individuals, the possibility of allele frequencies changing between generations increases.
For example, one individual plant can produce millions of pollen spores in a single
season but often produces a smaller number of recipient female gametes. This dissimilar
ratio of gametes results in a greater possibility of a sampling phenomenon dramatically
changing the allele frequencies. With small populations, consisting of a few individuals,

some alleles could be lost or fixed in every member in the population, due to the

sampling of gametes. Genetic drift tends to decrease genetic diversity found within
populations, resulting in a loss of rare alleles and an increase in homozygosity.

Genetic drift changes the genetic diversity within populations but can also alter
the genetic diversity found among populations. Since drift is a random process,
fragmented populations affected by drifi can be fixed for different allele frequencies. The
effect would be for drift to increase the genetic diversity found among populations as
different populations loose or become fixed for different alleles.

Isolation by Distance / In breeding: A phenomenon observed by Alfred Russel
Wallace over a century ago was that over geographical distances differences among
species’ composition tend to occur. This observation helped lead to the theory of natural
selection which is based on the fact that different phenotypes will have greater or lesser
fitness (survival) values in different areas. This phenomenon of detectable differences
occurring over space can be further advanced by a theory in population genetics; with
increased geographic distance there will be an increase in genetic differences between
populations. This theory is based on the fact that due to the limit of migration, individuals
will tend to mate with their neighbors thereby increasing the frequency of local alleles.
The phrase isolation by distance was described by Sewal Wright ( 1943; 1946) to explain
the accumulation of local genetic used under geographically separated areas. The theory
of isolation by distance has been developed by Wright (1946), Malecot (1948) and others.
The most severe result of isolation by distance is inbreeding. When plants mate in close
proximity there is the clmnce inbrwding will occur. Inbreeding is the gauge used to
decide whether individuals are mating with relatives or more distant members of the

population. Inbreeding can be especially dramatic in small populations with a restricted

10

range resulting from habitat fragmentation. With small fragmented populations there is an
increased chance that individuals will have consanguineous matings. Along with
consanguineous matings, selfing can greatly increase the effects of inbreeding.

With inbreeding, the probability that two recessive deleterious alleles will come
together in an individual increases. However, in large populations the effects of the
deleterious alleles will often be masked as heterozygotes. Inbreeding is usually not
detectable or problematic in large populations. The effects of inbreeding can be more
easily seen in small populations because the number of deleterious alleles will be
relatively substantial compared to large populations. Populations with individuals
homozygous for deleterious alleles often show a higher mortality rate and lower
fecundity (Hartl and Clark, 1997; Chiras et al., 2002) .

The effect of inbreeding can be a major problem with fragmentation which can
prevent immigration of individuals from other areas, especially when there is inhospitable
habitat between populations. Limited migration results in mating with neighbors, who
over generations will be more related. Continued inbreeding will result in a reduction of
heterozygosity. This loss of heterozygosity can be measured within a population or
among groups of populations using Wright’s F statistics.

Biparental Inbreeding: In a species like white pine, where reproduction does
not ofien include selfing, inbreeding can still be a problem. Inbreeding can be biparental.
Biparental inbreeding can be important in the case of plants because individuals tend to
grow in close proximity to their parents and siblings. Matings between progeny of the
same two parents are full sibs while mating between individuals with the same maternal

or paternal parents are half siblings. Mating with one’s parents or siblings often results

11

in having two copies of the exact same alleles which would be identical by decent
(autozygous). So, with biparental inbreeding there is an increased chance for individuals
to be homozygous for deleterious alleles.

Any form of inbreeding tends to reduce genetic diversity within populations but it
does not change the gene frequency within a population. In other words, the same
number of alleles will still be found in the population as homozygotes, but there will be
less of a potential for loss of any alleles in the populations as with genetic drift. Unlike
genetic drift, the effects of biparental inbreeding can decrease genetic diversity and
increases spatial structure even in populations that are large in size.

Migration: In the context of this study migration will mean gene migration which
involves the movement of genes via pollen or seed. Migration can be viewed in terms of
the influence of one population from another. The net effect of migration on the
receiving population is determined by several factors: the initial gene frequency in the
receiving population, the rate of immigration and the frequency of the allele in the donor
population. In one very simple model, migration can be expressed mathematically
(Hartle and Clark, 1997)

P‘ =( l -M)P+MP"'
Where P‘ is the gene frequency in the next generation, M is the migration rate, P is the
original gene frequency in the receiving population and P“ is the allele frequency in the
donor population. Pollen tends to travel long distances, especially in wind pollinated
(anemophilous) plants. Even plants that rely on animals to disperse pollen can have
extensive pollen flow between individuals and populations. Pollen can contribute to

changes within a population’s genetic structure (Ledig, 1998). Seeds are usually larger,

12

and therefore usually travel shorter distances than pollen does in anemophilous species.
When there is a high amount of gene flow, the genetic diversity will usually be high
within populations, and there will be a low amount of divergence or diversity among
populations. With low gene flow, individual populations will be subdivided into a
mosaic of gene distributions or deme’s and there will be an increase in the genetic
diversity found among populations. Migration can decrease the genetic diversity found
among p0pulations. This can happen in two ways: first, if there is pollen from one
population entering all populations, (one way migration) recipient populations will tend
to have the same genotype; second, if populations are consistently exchanging pollen and
seeds, there will tend to be a blending of genotypes among all populations.

Fragmented populations can avoid the loss of genetic diversity associated with
isolation/inbreeding, bottlenecks, and genetic drift, if there is a high amount of gene flow.
By introducing new alleles or DNA sequences, migration usually increases the genetic
diversity within populations. Migration will be examined by looking at the occurrence of
private alleles, spatial autocorrelation measures and the levels of heterozygosity.

The study of population genetics involves the genetic differences among
individuals in one or more populations (a population being a group of breeding
individuals in a discrete area). A population's genetic diversity is influenced by several
factors such as population size, mating system, geographical distribution, habitat

fragmentation and other factors such as selection, mutation, migration and genetic drift.

l3

Table 1 Comparative effects on the genetic diversig of populations

 

 

 

 

 

 

 

 

 

 

Diversity within populations [JDiversity among populations |
flratlon Increase Decrease
Genetic drift Decrease increase
Inbreeding Decrease Increase
Fragmentztion Decrease Increase
Mutation increase Increase

 

 

 

 

 

 

 

 

Table 2 Comparative effects on genetic measures of populations

 

 

 

 

 

 

 

 

 

 

 

 

 

Fit Fis [Fst
Migration Decrease Decrease Decrease
Genetic drift Increase Increase increase
Inbreeding increase Increase increase
Fragmentation increase Increase Increase
Mutation Increase Decrease Increase
Levels of Genetic Diversity

Genetic diversity exists on three levels, within individuals, within populations and
within a species. First, there can be genetic diversity within individuals, for example,
diploid and polyploid individuals can have multiple alleles for a particular gene. An
allele is a particular form of a gene; diploid organisms have two copies at every gene.
Individuals with both alleles producing identical gene products or DNA length are
considered homozygous, whereas individuals that have two different alleles are
heterozygous for that locus. Individuals that are heterozygotic for a particular locus have
more genetic diversity than individuals that are homozygotic.

The second level of genetic diversity is at the population level. Genetic

differences at the population level can be measured at two levels, within, and among
populations. Depending on the scientific question being tested, it may be better to focus
on one type of diversity or the other. In populations, DNA markers (discussed below)

can be used to estimate the frequency of alleles among individuals in the population(s).

l4

Measurements of genetic diversity (descriptive statistics) such as the percentage of
polymorphic loci (P), number of alleles at a loci (A), the number of polymorphic alleles
(Ap), genetic diversity (He) and observed heterozygosity (Ho) can be used to estimate if
the current diversity is due to particular forces acting on the population(s) now or in the
past. Spatial autocorrelation of alleles in plant populations can also be informative in
understanding the influences on a p0pulation.

The third level of genetic diversity is the species level. Genetic diversity at this
level must often be addressed using measurements among populations because of the vast
number of individuals in a species. Descriptive statistics can also be used to measure
diversity at the species level. The diversity measures are compared among populations
over the range of a particular species to look for patterns or correlations that could
suggest selection, migration, isolation or recent speciation events.

Genetic Markers Used in This Study

The first markers used to differentiate between organisms and varieties were
observable differences such as color and shape. One of the fast molecular or genetic
markers used to test for diversity was allozymes (For example, Soltis et al., 1983).
Today, allozymes are perhaps one of the least expensive and effective markers used to
estimate population genetic diversity. The use of allozymes is also beneficial because the
results can be compared to many other organisms (summarized in Hamrick and Godt,
1990). However, allozyme loci may not be able to display the true diversity of an
organism’s genome. It is possible for allozymes to be identical by state, meaning that the
allozymes are the same allele (migrate the same on an electrophoretic medium) but they

could have different DNA sequences. The difference in alleles that are identical by state

15

and identical by descent (replica’s of a single gene from a common ancestor) cannot be
fully distinguished using allozymes as genetic markers.

In order to more specifically characterize differences in the genetic makeup of
individuals, assays of the actual DNA of organisms were devised. With the advent of the
Polymerase Chain Reaction (PCR), (Mullis and Faloona, 1987; Saiki et al., 1987) it
became possible to amplify specific areas of an organisms genome. Today, there are
numerous types of genetic markers that use PCR to estimate genetic diversity. One such
type is termed microsatellites or simple sequence repeats (SSR’s). Microsatellites are
areas of an organism's genome, usually in non-coding regions (Hancock, 1985), where
there are repeats of a specific nucleic acids motif. For example, (ACACACACAC)
would be described as a five repeat of an adenosine and cytosine motif. The number of
nucleotides in a repeat can vary from one to many. Microsatellites have been used in
numerous other studies (Hearne, et al., 1992; Morgante and Olivieri, 1993; Queller et al.,
1993. Echt et al., 1996; Rajora et al., 2000; Walter and Epperson, 2001). When the DNA
sequence of flanking regions (both upstream and down stream) of an SSR are determined,
specific primers can be made to amplify the microsatellite sequence. The use of SSR’s in
molecular studies is useful because the specific length of the SSR can be consistently
amplified and there are usually high levels of polymorphisms (Powell, et al., 1996).

Genetic diversity can be measured using nuclear DNA or using the DNA of
organelles such as mitochondria and chloroplasts. The different genomes (organism or
organelle) have their potential usefulness. Nuclear DNA is inherited from both the
mother and father, (bipmentally) while mitochondria and chloroplast DNA can be

inherited (uniparentally) either maternally, as mitochondria are in many organisms, or

16

paternally as is the case of chloroplasts in the genus Pinus. Organelle DNA is obviously
in greater abundance since a single cell has but one pair of genes at a nuclear locus, and
potentially hundreds of copies of organelle's DNA. This fact allows for more consistent
amplification of organelle DNA under in vitro conditions compared to nuclear DNA.

Mutation could be a major cause of variation in SSR diversity examined in this
study. In vertebrates it has been suggested that regions with repeats have a tendency to
increase in the number of repeats over time (Weber and Wong, 1993). A potential
drawback to SSR’s is that a mutation may occur at the primer binding site, which results
in no amplification or a null allele. With nuclear DNA this could result in heterozygotes
being scored as homozygotes. With organelle DNA the result would be an inability to
amplify SSR’s in an individual. If a mutation occurs in the primer binding region it will
in effect result in a underestimate in the genetic diversity.
Natural History of White Pine

Eastern white pine can live up to 450 years and can reach heights of 90-180 feet,
has cones usually around five inches in length (4-8 inches) and needles which persist for
2 years, rarely 3 (Preston, 1989). Mature white pine has a shape that perhaps can best be
described as irregular, with occasional branches protruding out in seemingly random
directions. Mature trees are usually broad crowned with a single straight trunk. White
pine is also one of the tallest trees in northeastern North America, making it conspicuous
as mature trees emerge above the forest forming a super canopy.

The current distribution of white pine is from Minnesota to Maine through the
Great Lakes region and from Newfoundland to Georgia through the Appalachians

(Preston, 1989). There are also isolated populations of white pine south of the Great

17

Lakes region but these are usually in isolated habitats such as lake dunes (Daniel and
Sullivan, 1981). One can also observe white pine on almost every lake shore throughout
BIA. Historical evidence suggests that white pine survived the last glacial age in what is
now the southeastern United States (Davis, 1981; Pieliou, 1991). Pollen records indicate
that white pine has been in the Great Lakes region for several thousand years (Jacobson,
1979; Macdonale et al., 1998)

White pine is found consistently in two forest types in the Great Lakes region,
boreal and northern hardwoods. It is also occasionally found in bog and northern white-
cedar communities as well. White pine’s strategy for surviving in boreal forests in the
Great Lakes region is for individuals to survive near water, or on rock outcrops where fire
intensity is low (Frelich and Reich 1995). During prehistoric times, fires would kill
invading white cedar, spruce and fir trees but would only scar the large thick barked pines
(Frelich and Reich 1995). White pine could then establish in areas where fire destroyed
the existing vegetation, producing even aged stands of white pine. Further inland,
depending on the time of year of the fire, aspen and birch could be the first trees to
succeed in after a burn. Since white pine is moderately shade tolerant it could gradually
stock the under story of adjacent aspen or paper birch stands with seedlings for decades
after the fire. This type of establishment would result in uneven aged stands of white
pine. It was also shown in Wisconsin (Swain 1978) that white pine increased in
frequency during wetter climactic times when the more fire adapted species such as paper
birch, oaks and aspens began to decline. Pollen analysis suggests that white pine was

able to succeed the early successional fire adapted trees. (Swain 1978)

18

Many forests described as northern-hardwoods contain white pine. Some areas
have soil with nutrients and permeability that are unfavorable for northern hardwood
trees but would support white pine. White pine has a generally high tolerance for soil
type and nutrient availability. (Smith and I-Iinkley, 1995; Whitney, 1986). This range of
soil tolerance allows white pine to persist in habitats that are widely distributed across the
landscape of the northern Great Lake states. This would also allow for white pine stands
to act as a seed and pollen source to invade any micro-disturbance such as wind throw
areas. Furthermore, white pine’s soil tolerance would allow for it to persist in other
areas that are not optimum for northern hardwood species such as glacial outwashes
(Daniel and Sullivan, 1981).

Reproductive Mechanisms

The mating system is an important characteristic because it has an efi‘ect on the
distribution of genotypes of populations. White pine, like most conifers has a high
amount of genetic diversity (Hamrick and Godt, 1990). Reproduction in white pine, as in
most pines, is sexual. There is no reported coppice reproduction in white pine, no
layering of branches to form new trees has been reported and no documentation of
apomixis is in the current literature. White pine is an outcrossing, monecious,
anemophilous gymnosperm. In white pine populations, the pollen is released in the
spring from male strobili. The pollen is captured in the female strobili and adheres to a
sticky fluid about the micropyle. (Raven et al., 1992). As the fluid evaporates, the pollen
is drawn into the micropyle and at this time a pollen tube forms and slowly digests its
way through tissues toward the megaspores (Raven et al., 1992). Fertilization occurs the

following year. Then afier several more months, winged seeds form that will be released

19

in autmnn almost 1.5 years after the pollen was released. At maturity cone scales
separate and the winged seeds are released and can be carried considerable distances by
wind.

White pine's reproductive mechanism helps to maintain high levels of genetic
diversity. First, mechanisms that increase genetic diversity can be found at the gametic
interface: female cones secrete a pollination droplet between the cone scales, and pollen
lodges on the droplet and will be drawn in (reabsorbed) and eventually will be deposited
in the pollen chamber. If pollen is plentiful, two to four pollen grains can be lodged in
the pollen chamber (Ledig, 1998). One to many archegonia can be formed in a egg
(Ledig, 1998). When extra pollen adheres to the female cone, this allows sampling
(randomization) so that not all seeds, even in a single cone will be as likely to have the
same father. White pine is a primarily out crossing anemophilous species with pollen
that has the capacity to travel long distances; pollen grains have two small air bladders
that aid in the pollen remaining air borne for long periods of time. An increase in
diversity results from the fact that millions to billions of pollen grains are released in a
white pine population, thereby increasing the sampling of gametes via pollen flow.
White pine has an out crossing rate that approaches 1.0 (Beaulieu and Simon, 1994); with
low amounts of selfing there will be a low amount of inbreeding, which should
increase/maintain genetic diversity.

Beaver Island Archipelago

The study site, BIA provides a model system for studying variable sized

populations of white pine and the genetic diversity that results from isolation and variable

population size. Not only does this system contain variable sized and spaced populations

20

of white pine, but also the relative starting point of ecological succession can be
estimated During the last ice age BIA, as was most of the Great Lakes region, was
covered by the Laurentide ice sheet. As the ice sheet melted, proglacial lakes were
formed. The proglacial Lake Algonquin began to form around 16,000 years ago (Pielou,
1991); today, the remnants of proglacial Lake Algonquin consists of Lake Michigan and
Lake Huron. Eventually, the Algonquin lake levels were dramatically lowered as the
retreating lobes of the glacier revealed new outlets to the ocean for the glacial lakes. It is
at this time of low water, around 10,000 years ago, that the BIA was connected to what is
now the northern lower peninsula of Michigan. It was also at this time that white pine
was migrating through the Great Lakes area (Pielou, 1991; Macdonale, 1998). The BIA
was then connected as a peninsula or a chain of islands to mainland Michigan
intermittently for around 3,500 years. The BIA has been isolated for roughly the last six
thousand years. The lakes reached a maximum height about 4,000 years ago during the
Nipissing period, about seven meters above current levels. During the Nipissing period
the islands would have been much smaller in size. After the Nipissing period, the lake
levels decreased for about two thousand years and have remained relatively stable for the
last 2,500 years.

Native Americans frequented BIA perhaps as soon as it was revealed. The effect
that Native Americans had on the ecosystem is unknown but assumed to be minimal. The
first European's to visit BIA were French trappers in the17th century (1650’s). It has
been suggested (Hatt, 1950) that there was some colonization and farming by the French
in the seventeenth century. There would have also been some Native American's on BIA

at that time. The islands were reinhabited by lurnbermen and some settlers during the

21

1800’s. The BIA proved to be in a logistic place one-half way between Detroit and
Chicago. ‘The old steamers would stop to refuel in the deep harbors of the islands in the
archipelago especially from 1840-1889’(Hatt, 1950). Logging was mostly for cordwood
until fossil fuels replaced wood as an energy source in the late 1800’s. Logging
continued until today, where single tree selection is a common practice on Beaver Island.

Four islands in the archipelago harbor white pine populations: Garden Island, Hog
Island, High Island, and Beaver Island (Figure 1, page 1)

Garden Island: 7.8 square miles, 20.7 miles of shore line. There were Native
American residents on the island until 1839. Garden Island had yearlong European-
Americans inhabitants during the 1800’s. The island has been uninhabited since 1938
(Hatt, 1950). Garden Island had the smallest number of white pine trees of any of the
four islands. White pine is found only in one bay, Northcut bay. This is a sandy bay with
a large area for camping. White pine is also found among otherwise pure stands of
northern white-cedar (Thuja occidentalis) by the lakeshore. White pine is in the inner
dunal area with white-cedar, balsam fir (Abies balsamea) and other inner dunal plants.
White pine is also formd interspersed in the forest in the watershed of Northcut bay with
red pine (Pinus resinosa) and quaking aspen (Populous tremuloides). White pine
regeneration is limited to the dune areas, where there is a slow succession toward the lake
from forest to dunal forbs, to sand, to the lake.

Hog Island: 3.9 square miles, 16.0 miles of shore line. Hog Island probably
never had permanent settlers but was logged most likely during the late 19th century.
White pine is prominent on the lakeshore on the northern half of the east side of Hog

Island The stand of white pines extends back into the island several dozen meters.

22

White pine is found growing in a near boreal forest of balsam fir, white spruce and white
cedar. There are several white pine trees (around 13) on the northwest side of Hog
Island. In both populations, white pine reproduction is limited to the dune areas, where
there is a slow succession toward the lake from forest to dunal forbs, to sand.

High Island: 5.8 square miles, 12.5 miles of shore line. Hog Island was settled
and inhabited by Mormon settlers in the 1800’s. Evidence of their agriculture and
settlement still remains. White pine is formd on the east and west lakeshore on the
northern part of the island. The east side white pine population ranges through two large
bays and rounds the northeast corner until the tree composition changes from a near
boreal forest to a deciduous forest. The white pine stand on the west side is found at the
southern end of the large sand dunes. White pine is intermixed with red pine in a near
boreal community. The population on the west side is much smaller in size. There is
limited white pine reproduction on both populations in the inner dunal areas. White pine
reproduction is limited to the dune areas, where there is a slow succession toward the lake
from forest to drum] forbs, to sand.

Beaver Island: 58.4 square miles, 41.6 miles of shore line. Beaver Island had
Native American residents when French explorers visited the island in the 1600’s. A
French colony did not survive but the island was visited by trappers and explores, and
was recolonized in the 1800’s by Mormons. The population of Mormons was estimated
to be between 1300-2600 people in 1855. The island human population never again grew
large (Hart, 1950) until, today, where there are about 450 year round residents with

thousands of seasonal visitors.

23

Beaver Island had the greatest number of white pine in the archipelago, growing
in many bays at the edge of the Lake Michigan shore, along most inland lake shores, on
old beach bluffs representing higher lake levels, and in almost pure stands in some
meadows and abandoned agricultural areas. It is quite abundant in some old fields, and
now grows on soils unsuitable for agriculture. White pine reproduction can be quite
prolific in some areas of Beaver Island. The northern half of Beaver Island has a large
contingent of white pine interspersed in almost every plant community, from bogs to
fields, to northern white-cedar swamps to northern hardwood forests, but in the southern

part of the island, white pine is limited to lakeshores and clearings.

24

Materials and Methods

Sampling: The BIA was surveyed using an airplane during the first week of April 2000.
At this time no deciduous trees had leaves, and white pine stands were identified by
vegetative patterns and by the emergent silhouette shape of individual white pine. White
pine is formd on four islands: Beaver, High, Hog and Garden Islands. The approximate
size of white pine populations in area, ntunber of trees, as well as location, were mapped.

The sampling procedure was to locate specific areas within the white pine
populations, where 13-20 contiguous trees would be sampled. The sampled areas are
hereafter referred to as plots. The exact area to be sampled was chosen so that roughly 20
trees would be included in the plot. Trees with a diameter less than 12 cm were not
sampled Plots were spaced throughout the archipelago to allow estimates of genetic
diversity within and among islands (white pine populations). Most plots were located in
areas where white pine was a dominant or codominant tree. From May until July 2000,
19 plots were sampled throughout the four islands containing white pine, except Garden
Island. White pine on Garden Island was not a dominant tree, but was intermixed with
red pine, and quaking aspen, in the upland areas and occurred sporadically with northern
white-cedar, near the shoreline. All sampled trees on Garden Island are considered a
single plot. Four hundred twenty eight trees were sampled from the four islands. Each
tree was tagged with sequentially numbered aluminum tags and the plot was marked
using a global positioning system (GPS) (Geotracker II). Needle and bud tissue were
taken with a hand prunners, a pole prunner or with a shotgun depending on the height of
the lowest living needle tissue. Several grams of needle and bud tissue were collected

from each tree. The plant material was divided into roughly two halves with each half

25

being wrapped in a wet paper towel and kept on ice until returning to Michigan State
University. Upon returning to Michigan State University, one sample was frozen, the
other was refiigerated. The refrigerated samples were used for the allozyme analysis, the
frozen samples were used for extracting chloroplast DNA.

The sampling scheme was constructed to cover the entire archipelago to give an
estimate of the genetic diversity of the archipelago as a whole. The sampling scheme
also should provide an estimate of the genetic diversity found within each island. There
were 12 sampling plots on Beaver Island. High Island had 3 sample plots, one on the
west side and two at the north and south ends of the eastern population. Hog Island had
four sample sites. Almost every tree was sampled from the western population (this was
the smallest plot with 13 trees), and three plots were sampled from the eastern population,
one at the northern end of the eastern white pine population, one in the middle, and one
plot from the southern most part of the eastern shore population. The Garden Island
population was sampled along transects throughout the entire population because white
pine was found to be too dispersed to group trees into a plot. It is estimated that 15%-
25% of the entire Garden Island population was sampled.

Allozyme Extraction and Genotyping: The allozyme laboratory work was done in
South Korea in the lab of Dr. Myong G. Chung. Collected needles were kept at 4 °C and
then shipped to his laboratory, and stored at 4 °C until protein extraction. Of the twenty-
one putative loci systems assayed by Dr. Chung, five were polymorphic: Pgm-l, Pgi-2,

Mdh-2, Tpi-2 and Aat-3.

26

The following protocol was reported by Dr. Chung:

Leaf samples were cut and crushed with a mortar and pestle, and phosphate-
polyvinylpyrrolidone extraction buffer was added (Mitten et al., 1979). Enzyme extracts
were absorbed onto 4mm x 6mm wicks cut from Whatman 3MM chromatography paper,
and stored at -—-70 °C until allozyme analysis. Starch gels (11.5 %) were stained for 13
enzyme systems, which produced 21 putative loci. A Poulik buffer system, a
modification (Haufler, 1985) of Soltis et al. (1983) “system 6” was used to resolve
alcohol dehydrogenase (Adh), fluorescencent esterase (Fe), glutamate dehydrogenase
(Gdh), leucine aminopeptidase (Lap-1 , Lap-2), phosphoglucomutase (Pgm-I), and
triosephosphate isomerase (Tpi-I, Tpi-Z). A histidine citrate buffer system, a
modification (Chung and Kang, 1994) of Soltis et al. (1983) “system 11” was used to
resolve isocitrate dehydrogenase (Idh), fonnate dehydrogenase (F dh), malate
dehydrogenase (Mdh-I, Mdh-Z, Mdh-3), 6-phosphogluconate dehydrogenase (Pgd), and
phosphoglucoisomerase (Pgi-l , Pgi-Z). Soltis et al. (1983) “system 7” was used to
resolve diaphorase (Dia-I, Did-2) and aspartate aminotransferase (Aar-I, Ant-2, Aat—3).
Stain recipes were taken from Soltis et al. (1983), except for diaphorase (Cheliak and
Pitel, 1984). The genetic basis of allozyme banding patterns was inferred from
segregation patterns with reference to typical subunit structure (Weeden and Wendel,
1989; Wendel and Weeden, 1989) and conceptual methods described in Gillet (1988).
Putative loci were designated sequentially, with the most anodally migrating isozyme
designated ‘1’, the next ‘2’, and so on. Similarly, alleles were designated sequentially
with the most anodally migrating alleles designated with superscript ‘a’. (M.G. Chung
Pers. Comm)

DNA Extraction and Genotyping
Chloroplast Microsatellite Markers: Six different primer pairs (pt9383 cp2; pt15169
cp3; pt30204 cp5; pt36480 cp6; pt637l8 cpl 1 and pt71936 cp12) designed from known
sequences of black pine, Pinus thunbergir‘ (Vendramin et al. 1996), were used to amplify
specific regions of white pine’s chloroplast genome. The forward strand of each primer
pair was labeled with one of three phosphoramidites: 6-FAM, HEX and NED (AB1) for
detection of alleles. This permitted multiplexing and simultaneous analysis of
polymorphic primer pair loci.

DNA Isolation: Chloroplast DNA isolation was conducted in the forest genetics
lab, Michigan State University. The protocol consisted of first weighing ~120

milligrams of needle tissue for each tree. The needle tissue was then cut finely with a

27

razor, and placed into a 2 ml screw cap centrifuge tube between one ceramic bead at the
bottom of the tube and another bead on top of the needle tissue. Then, 750 ml of CT AB
buffer (Tris/EDTA/NaCl/CTAB) was added to the sample. The amalgamate in the tube
was placed into the FastPrep F P 120 (Savant Bio 101) cell disrupter for 2 cycles of 45
seconds on speed setting 4.5.

Afterwards, the samples were placed into a water bath (GP-200 Neslab) at 65
degrees centigrade for 20-30 minutes, to further aid in cell lyses. The samples were then
taken to the Genomics Technology Support Facility at Michigan State University using
the automatic DNA isolation system AutoGen 850 a(AutoGen, Inc.) for DNA isolation,
applying the plant protocol “Plant tissue DNA (system 4)”. The dry DNA was then
placed into 100 ml of TE buffer solution, and the concentration was determined with a
Hoefer DyNA Quant 200 flurometer. The DNA was then diluted with water to

approximately Sng per 111. The modal yield of DNA was around 500 ng/ul but sample

DNA concentrations had a wide range from less than 10 ng/ul to over 3000 ng/ul.

PCR Amplification: Polymerase chain reaction amplification was conducted in a total
volume of 15 pl containing 30ng of template DNA and 9 pl of PCR buffer. The PCR
buffer was a PFU/SCR buffer consisting of 20mM Tris-HCl, pH 8.75, 10mM (NI-I4)2804,
10mM KC], 2mM MgSO4, 0.1mM Cresol red, 0.1% Triton X-100, 10% bovine serum
albumin, and 6% sucrose. The reaction tube contained 200 pl of each dNTP, 3.5 mM
MgCl, 200 nM of each primer and 0.04 units of Amplitaq Gold (Perkin-Elmer). The
PCR reaction was conducted using a ptc-100 Programmable Thermo Cycler. The
protocol began with two cycles at 94°C for one minute (denaturing) 65°C for one

minute (annealing) and 70°C for 35 seconds (extension) followed by 18 cycles of 93°C

28

for 45 seconds (den. ), 64°C for 45 seconds (ann) with 05°C decrease during the
remaining cycles, and then 70°C for 45 seconds (ext). This was then followed by 20
cycles at 92°C for 30 seconds (den), 55°C for 30 seconds (am), and then 70°C for 20
minutes (ext). At least 7% of all samples were checked for a PCR product using a 2.5%
agarose gel.

Three PCR products with different fluorescent labels were then combined into a
single capillary tube and run on a 3100 Genetic Analyzer (Applied Bio Systems) at the
Genomics Technology Support Facility at Michigan State University.

Sample genotypes were viewed using an OS 9.2 Macintosh computer running
Genescan ver. 2.1. The lab protocol and techniques allowed for one base pair
differences to be resolved. At least 10% of the polymorphic primer pairs were repeated
in order to check for consistency. Sample genotypes were recorded and polymorphic
primer pairs were combined to form a lmplotype for each tree.

Statistical Analysis of Allozymes: For the analysis of allozyme diversity, each
locus was considered polymorphic if two or more alleles were detected. Of the 21
enzyme systems only five were polymorphic (Pgm-l , Pgi-2, Mdh-Z, Tpi-2, Aat-3).
Allozyme analysis was conducted using these five polymorphic systems. The program
F stat (Goudet, 2001) was used to measure allele frequencies among the islands and plots.
Standard genetic measures (descriptive statistics), F-statistics (F, f and 9 as in Weir,
1996), and Nei’s genetic distance and genetic identity (Nei, 1978) were calculated using
Lewis and Zaykin’s (2000) GDA program both among islands and within each island.
Descriptive statistics include the average sample size (n), proportion of loci polymorphic

(P), average number of alleles per locus (A) and polymorphic locus (Ap), expected

29

proportion of heterozygotes (He), observed proportion of heterozygotes (Ho), and
inbreeding coefficient (f). Wright’s standard F statistics estimate the parameters f
(corresponding to Fis), F (Fit), and Theta (F st). For the allozymes analysis, F -statistics
bootstrapping (1000 replicates) was done to obtain 95% bootstrap confidence intervals
for the parameters f (=F is), F (=F it), and Theta (=F st). The allozyme allele frequencies
among islands were compared to Hardy-Weinberg (H-W) expected values using a chi-
square tests in GENEPOP (Yeh and Boyle, 1997) to measure the relative excess of
homozygotes or heterozygotes compared with panmictic expectations

Spatial autocorrelation was conducted for averaged allele frequencies for each
polymorphic loci among plots using SAAP ver. 4.3 (D. Wartenberg, 1989). For diallelic
loci, only one allele was considered because the second allele would contribute identical
information. For multiallelic loci, all alleles at that locus were used for analysis. Spatial
autocorrelation analysis was conducted on the average allele frequencies of the 20 plots
and a separate analysis was done within Beaver Island. The analysis of the Beaver
Islands plots was conducted because the number of joins in the smaller distance classes
would be influenced by the results from within Beaver Island since most plots (12/20)
were found within Beaver Island. The spacing of plots allowed for seven distance classes
throughout the archipelago, and five distance classes within Beaver Island. Moran’s I
statistics (Sokal and Oden, 1978) were calculated for each of the distances classes by

I=N22(WijZiZj)/(22Wij2212)

Where N is the number of plots, Wij is a join in the weighting matrix, where wij is set
equal to 1.0 if the ith and jth plots are in the distance class and zero otherwise. Zi= xi-x,

Zj= xj-x and the variables xi and xj are the genotypic scores for the ith and jth plots

30

respectively and x is the mean score for all plots. Each 1 value was used to test for
significant deviations from the expected values E(I) = -1/(N-1) (Cliff and 0rd, 1981),
under the random distribute null hypothesis. In addition, the I values for each allele in a
distance class were combined and averaged to create a standard normal deviate (SND) to
test for significance within distance classes. A significant positive SND indicates that
plots in the distance class have similar gene frequencies. A significant negative value
indicates that the plots in the distance class have gene frequencies more dissimilar than
expected by chance.

Statistical analysis of Microsatellites: Statistical analyses were conducted using
polymorphic locations of the chloroplast genome. Three primer pairs (pt15169; pt30204;
pt63718) were polymorphic and all analyses were conducted using the polymorphic
primer pairs. For each population and plot, standard diversity measures and F st (Wright,
1965) were computed using the Lewis and Zaykin (2000) GDA program both within and
among islands as was done with the allozymes. Since the microsatellite data was treated
as haplotypes, the following measures were estimated, the proportion of loci polymorphic
(P), and polymorphic locus (Ap), observed proportion of heterozygotes (Ho), and
inbreeding coefficient (f) were not conducted. For haplotype data, the average number of
alleles per locus (A) is the average number of haplotypes and He is gene diversity. The
program F stat (Goudet, 2001) was used to estimate Rst and to give haplotype frequencies

among the islands and plots.

3]

Abbreviations for Islands and Plots

Island abbreviations:

Garden Island ................. Ga
Beaver Islands ................ Be
High Island ..................... Hi
Hog Island ...................... Ho

Plot Abbreviations:

Garden Island ................. GI
Hog West ....................... Ho-W
Hog North East ............. HN-E
Hog Middle East ........... I-lM-E
Hog South East ............. HS-E
High South East............HiS-E
High N. East ................... HiN-E
High West ....................... Hi-W
Powerline ........................ PL
Blue Trail ........................ BT
Foot Bridge ..................... FB
French Bay ..................... FBy
Green’s Bay .................... GB
Wujek ............................. WK
CMU ............................... CMU
DUDA ............................. DA
Font Lake ........................ FL
Barney’s Lake .................. BL
Sloptown .......................... SN
Airport ............................. AT

32

Results

Allozyme genetic diversity measures among islands, Tables 3-10

Genetic diversity measures of allozymes for all islands are found in Tables 3-10. Table 3
shows the nrunber of alleles found on each island. The number of alleles is the same for
all islands with the exception of Pgi-2 and Tpi-2. The locus Pgi-2 had a fourth private
allele on Beaver Island, and Tpi-2 was monomorphic only on Hog Island, and dialleleic
elsewhere. Table 4 shows the allele frequencies among islands. The locus Pgi-2 was the
only one to show substantial variability in allele frequency between islands. The other
loci had rather consistent frequencies among islands. Table 5 contains the results for the
descriptive statistics. Descriptive measures were found to be consistent with the mean
among islands for measures of the proportion of loci polymorphic (P mean = 0.95),
average number of alleles per locus (A mean = 2.4) and polymorphic locus (Ap mean =
2.48), expected proportion of heterozygotes (l-le mean = 0.33) and the observed
proportion of heterozygotes (Ho mean = 0.31). The measure of inbreeding coefficient (f
mean = 0.07) was highest on Hog Island (0.10), which is somewhat higher than expected
for an outcrossing plant. Table 6 demonstrates the results of testing for Hardy-Weinberg
equilibrium. The locus Pgi-2 was found not to be in Hardy-Weinberg equilibrirun on
Beaver, High and Garden Islands, as was Aat-3 on Hog Island and Tpi-2 on Beaver
Island.

Table 7 lists the results for the standard F statistics. Significant deviations from
zero were found for F is, (95% confidence interval 0.02-0. 12) and for Fit (95% confidence
interval 0.04-0.12). The Fis and Fit values for individual loci were very high for Tpi-2

and Aat-3, which may have skewed the overall results among islands. Table 8 lists the

33

values of F is among islands and loci. It is not uncommon for Fis to have a wide range of
values in genetic studies, which is indeed the case here. The F is values ranged from -
0.244 to 0.318. The mean Fis over all loci for Garden was 0.04, Hog, 0.10, High 0.07
and for Beaver Island was 0.06. These values suggest that some inbreeding is occurring
in populations of white pine. Table 9 lists the gene diversity values among islands and
loci. Gene diversity is used as a measure of the evenness of allele frequencies. High
values for gene diversity suggest evenness in the allele frequencies, whereas low values
suggest one or a few alleles are in very high frequency. The gene diversity measures
were relatively high for all loci among islands, except Tpi-2 where a second allele occurs
. on three islands in a very low frequency. Table 10 shows the values for Nei’s genetic
distance and genetic identity. The values for Nei’s genetic distance were quite low,
ranging from 0.00 to 0.03. Values for genetic identity were high, ranging from 0.99 to

1.00.

34

Table 3 Number of alleles for each polymorphic locus found on each island

Garden Hog High Beaver

Pgrn-l 2 2 2 2
Pgi-2 3 3 3 4
Mdh-2 3 3 3 3
Tpi-2 2 1 2 2
Aat-3 2 2 2 2

35

Table 4 Allele frequencies of allozyme loci among islands

Island (sample size)

Allele Garden Hog High Beaver

(57) (53) (34) (196)
Locus: Pgm-l

1 0.798 0.877 0.912 0.839

2 0.202 0.123 0.088 0.161
Locus: Pgi-2

1 0.132 0.236 0.338 0.191

2 0.018 0.009 0.147 0.051

3 0.851 0.755 0.515 0.753

4 0.000 0.000 0.000 0.005
Locus: Mdh-2

1 0.649 0.528 0.618 0.671

2 0.237 0.387 0.294 0.222

3 0.114 0.085 0.088 0.107
Locus: Tpi-2

1 0.026 0.000 0.029 0.023

2 0.974 1.000 0.971 0.977
Locus: Aat-3

1 0.684 0.670 0.632 0.671

2 0.316 0.330 0.368 0.329

36

Table 5 Descriptive statistics of polymorphic allozymes for the islands including
measures of, sample size (n), proportion of loci polymorphic (P), average number of
alleles per locus (A) average number of alleles per polymorphic locus (Ap), expected
proportion of heterozygotes (He), observed proportion of heterozygotes (Ho), and
inbreeding coefficient (1). Values are averaged over loci

 

 

a) Island 11 P A An He Ho f
Garden 57.00 1.00 2.40 2.40 0.32 0.31 0.04
Hog 53.00 0.80 2.20 2.50 0.32 0.29 0.10
High 34.00 1.00 2.40 2.40 0.37 0.34 0.07
Beaver 196.00 1.00 2.60 2.60 0.33 0.31 0.06
Mean 85.00 0.95 2.40 2.48 0.33 0.31 0.07

b) Private alleles among islands

Locus Allele Frequency Found in
Pgi-2 4 0.005 Beaver

37

Table 6 Chi square test for Hardy-Weinberg equilibrium

 

Tiand Locu_s P-value
Garden Pgm-l 0.062
Garden Pgi-2 <0.000*
Garden Mdh-2 0.323 .
Garden Tpi-2 0.868
Garden Aat-3 0. 136
Hog Pgm-l 0.099
Hog Pgi-2 0.178
Hog Mdh-2 0.543
Hog Tpi-2 NA
Hog Aat-3 003*
High Pgm-l 0.070
High Pgi-2 <0.000*
High Mdh-2 0.864
High Tpi-2 0.90]
High Aat-3 0.062
Beaver Pgrn-l 0.592
Beaver Pgi-2 <0.000"'
Beaver Mdh-2 0.160
Beaver Tpi-2 0002*
Beaver Aat-3 0.350

* When the P value is less than 0.05 it indicates that the genotypes have frequencies
significantly different from Hardy-Weinberg equilibrium

38

Table 7 F statistics for allozymes among islands, estimating f (=Fis), F (=Fit),
and 9 (=Fst)

 

Locus Allele f F 9
Pgm-l All 0.02 0.02 0.01
2 0.02 0.02 0.01

l 0.02 0.02 0.01

Pgi-2 All 0.02 0.05 0.04
4 0.00 0.00 0.00

2 0.77 0.78 0.04

1 -0.18 -0.15 0.02

3 0.00 0.05 0.05

Mdh-2 All 0.05 0.06 0.01
3 0.14 0.14 -0.01

2 0.04 0.06 0.02

1 0.02 0.03 0.01

Tpi-2 All 0.13 0.13 0.00
l 0.13 0.13 0.00

2 0.13 0.13 0.00

 

Aat-3 All 0.15 0.14 —0.01

2 0.15 0.14 -0.01

l 0.15 0.14 -0.01

Overall 0.07 0.08 0.01
Upper bound 0.12 0.12 0.03
Lower bound 0.02 0.04 0.00

The 95 % confidence interval is derived from boot strapping over loci

39

Table 8 Fis of allozymes for loci and islands:

Locus Garden Hog High Beaver
Pgm-l -0.244 0.220 0.283 0.038
Pgi-2 -0.010 -0.149 -0.016 0.071
Mdh-2 0.114 0.073 -0.107 0.053
Tpi-2 -0.018 NA -0.015 0.206
Aat-3 0.197 0.284 0.318 0.067
All 0.039 0.099 0.070 0.063

40

Locus

Pgm-l‘

Pgi-2

Mdh-Z

Tpi-2
Aat-3

Garden Hog

0.324
0.261
0.515
0.052
0.437

Mean/Island 0.318

0.218
0.378
0.570
0.000
0.448
0.324

High
0.164
0.608
0.531
0.058
0.474
0.367

Beaver Mean/loci

0.270
0.396
0.490
0.045
0.443
0.329

Table 9 Gene diversity of loci and islands

41

0.244
0.41 1
0.527
0.039
0.451

Table 10 Genetic distance (below) and genetic identity (above) of allozymes among
islands
Ga .Ho Ifi Be
Garden 0.99 0.98 1.00
Hog 0.01 0.99 0.99
High 0.03 0.01 0.99
Beaver 0.00 0.01 0.01

42

Allozyme genetic diversity measures among plots, Tables 11-13

Genetic diversity measures among all plots are found in Tables 11-13. Table 11 contains
the results for the descriptive statistics. Descriptive measures were found to be similar
for islands with the mean among plots for measures of the proportion of loci polymorphic
(P mean = 0.83), average number of alleles per locus (A mean = 2.14) and polymorphic
locus (Ap mean = 2.38), expected proportion of heterozygotes (He mean = 0.32) and the
observed proportion of heterozygotes (Ho mean = 0.30). The only exceptions were in
plots with small sample sizes (less than 10 trees). In contrast, the measure of inbreeding
coefficient (f mean = 0.04) was high in plots Hog South East and High North East as well
as on Beaver Island in plots Barney's lake, Sloptown, and Airport, on Beaver Island, all
plots with high Fis values are located in close proximity to each other. The high Fis
values in plots on High and Hog Islands could be an effect of sampling. Table12 lists
the results for the standard F statistics. Significant deviations from zero were found for
overall values of Fit, (95% confidence interval 0.04-0.12) and for F st (95% confidence
interval 0.01-0.17). The Fit values for individual loci were very high for Tpi-2 and Aat-
3, which may have skewed the overall results among plots. The locus Pgi-2 had the
highest overall Fst value of any loci (Fst = 0.09). Table 13 shows the values for Nei’s
genetic distance and genetic identity. The values for Nei’s genetic distance were quite
low, ranging from -0.01 for many plots to 0.11 between Green's bay and Hog West.
Values for genetic identity were high, ranging from 0.99 to 1.00. with several measures
equal or greater than 1.0, with the lowest measure between High North East and Hog
West. Figure 3 shows the UPGMA phenogram of allozymes among plots. There was

little consistency of plots with their geographical locations.

43

The allele frequencies among plots can be found in the appendix

Table 11 Descriptive statistics for allozymes among plots including measures of,
sample size (n), proportion of loci polymorphic (P), average number of alleles per
locus (A) and polymorphic locus (Ap), expected proportion of heterozygotes (He),
observed proportion of heterozygotes (Ho), and inbreeding coefficient“).

3»
v-s

Population n P Ap He Ho

 

 

 

 

 

 

 

 

 

 

Garden 57 1.00 2.40 2.40 0.32 0.31 0.04
Hog-West 9 0.60 1.80 2.33 0.18 0.18 0.00
HogN-east 16 0.80 2.00 2.25 0.34 0.35 -0.03
HogM—east 16 0.80 2.20 2.50 0.34 0.33 0.05
Hogs—east 12 0.80 2.00 2.25 0.35 0.25 0.30
Highs—east 5 0.80 2.20 2.50 0.41 0.40 0.02
HighN-East 15 1.00 2.40 2.40 0.38 0.33 0.12
High West 14 1.00 2.40 2.40 0.32 0.33 -0.02
Powerline 15 0.80 2.00 2.25 0.25 0.23 0.09
Bluetrail 15 1.00 2.20 2.20 0.33 0.39 -0.19
FootBridge 11 0.80 2.00 2.25 0.27 0.25 0.05
Frenchbay 10 0.80 2.00 2.25 0.24 0.22 0.09
Greensbay 8 0.40 1.60 2.50 0.23 0.23 0.00
Wujek 17 0.80 2.20 2.50 0.30 0.27 0.09

CMU 18 0.80 2.20 2.50 0.29 0.29 0.00

DUDA 22 0.80 2.40 2.75 0.35 0.36 —0.05
Fontlake 21 1.00 2.20 2.20 0.33 0.40 -0.24
Barney'slake 20 0.80 2.20 2.50 0.39 0.34 0.12
Sloptown 18 0.80 2.40 2.75 0.37 0.33 0.11
Airport 21 1.00 2.00 2.00 0.34 0.26 0.26
:Mean 17 0.83 2.14 2.38 0.32 0.30 0.04

Table 12 F statistics for allozymes among plots, F statistics estimating f (=Fis), F
(=Fit), and 6 (=Fst)

 

 

Locus Allele f F 6
Pgm-l All 0.03 0.02 -0.01

2 0.03 0.02 -0.01

1 0.03 0.02 -0.01

Pgi-2 All -0.05 0.05 0.09

4 0.00 0.00 -0.01

2 0.77 0.78 0.04

1 -0.26 -0.16 0.08

3 -0.09 0.04 0.12

Mdh-2 All 0.04 0.06 0.02

3 0.12 0.14 0.02

2 0.04 0.06 0.01

l 0.01 0.03 0.02

Tpi-2 All 0.10 0.13 0.03

l 0.10 0.13 0.03

2 0.10 0.13 0.03

Aat-3 All 0.13 0.15 0.02

2 0.13 0.15 0.02

1 0.13 0.15 0.02

Overall 0.04 0.07 0.03
Upper bound 0.10 0.12 0.07
Lower bound -0.02 0.04 0.01

The 95 % confidence interval from boot strapping over loci

45

Table 13 Nei's Genetic distance (below) and genetic identity (above) of allozymes

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

among plots (in two blocks)
GI [Io-W HN-E IIM-E 318-3 818-3 HiN-E Hi-W PL 81'
Garden Island 0.99 0.99 0.99 0.99 0.98 0.97 0.96 0.99 1.00
Hog-West 0.01 0.96 0.98 0.97 0.95 0.93 0.95 1.00 0.99
HogN-east 0.01 0.04 1.00 1.01 1.00 1.01 0.99 0.97 0.96
HogM—east 0.01 0.02 0.00 1.00 1.02 0.98 0.96 0.98 0.99
Hogs-east 0.01 0.03-0.01 0.00 0.99 1.00 1.00 0.98 0.97
Highs-east 0.02 0.05 0.00—0.02 0.01 0.99 0.94 0.94 0.96
HighN-East 0.03 0.07-0.01 0.02 0.00 0.01 0.99 0.94 0.94
High West 0.04 0.05 0.01 0.04 0.00 0.07 0.01 0.97 0.94
Power line 0.01 0.00 0.03 0.02 0.02 0.06 0.06 0.03 1.00
Blue trail 0.00 0.01 0.04 0.01 0.03 0.04 0.07 0.07 0.00
FootBridge-0.01 0.01 0.00 0.00 0.02 0.01 0.02 0.04 0.01 0.01
French bay 0.00 0.03 0.03 0.03 0.04 0.04 0.04 0.06 0.01 0.01
Greens bay 0.03 0.06 0.04 0.02 0.06 0.00 0.05 0.11 0.06 0.03
Wujek 0.01 0.05 0.01 0.03 0.03 0.03 0.01 0.04 0.03 0.04
CMU -0.01 0.01 0.01 0.02 0.02 0.04 0.03 0.04 0.00 0.01
DUDA 0.00 0.02 0.00 0.01 0.00 0.01 0.01 0.02 0.01 0.02
Font lake 0.01 0.02 0.00 0.01—0.01 0.04 0.01 0.00 0.01 0.03
Barney‘s lake 0.04 0.08 0.00 0.02 0.00 0.01-0.01 0.01 0.07 0.08
Sloptown 0.00 0.03-0.01 0.00-0.01-0.01 0.00 0.01 0.02 0.03
Airport 0.01 0.05 0.00 0.03 0.01 0.04 0.00 0.02 0.04 0.05
as FBy ca wx CMU on n. BL su AT
Garden Island 1.01 1.00 0.97 0.99 1.01 1.00 0.99 0.96 1.00 0.99
Hog-West 0.99 0.97 0.94 0.95 0.99 0.98 0.98 0.92 0.97 0.95
HogN-east 1.00 0.97 0.96 0.99 0.99 1.00 1.00 1.00 1.01 1.00
HogM—east 1.00 0.97 0.98 0.97 0.98 0.99 0.99 0.98 1.00 0.97
Hogs-east 0.98 0.96 0.94 0.98 0.98 1.00 1.01 1.00 1.01 0.99
Highs-east 0.99 0.96 1.00 0.97 0.97 0.99 0.96 0.99 1.01 0.96
HighN-East 0.98 0.96 0.95 0.99 0.97 0.99 0.99 1.01 1.00 1.00
High West 0.96 0.94 0.90 0.96 0.97 0.98 1.00 0.99 0.99 0.98
Power line 0.99 0.99 0.94 0.97 1.00 0.99 0.99 0.94 0.98 0.96
Blue trail 0.99 0.99 0.97 0.97 0.99 0.98 0.97 0.93 0.97 0.95
Foot Bridge 1.01 0.98 1.00 1.01 1.01 0.99 0.96 1.00 0.99
French bay -0.01 0.98 1.00 1.00 1.00 0.97 0.93 0.98 0.98
Greens bay 0.02 0.02 0.96 0.95 0.96 0.93 0.93 0.96 0.94
Wujek 0.00 0.00 0.04 1.00 1.00 0.98 0.96 1.00 1.00
CMU -0.01 0.00 0.05 0.00 1.01 0.99 0.95 1.00 0.99
DUDA -0.01 0.00 0.04 0.00-0.01 1.00 0.98 1.01 1.00
Font lake 0.01 0.03 0.08 0.02 0.01 0.00 0.98 1.00 0.99
Barneys lake 0.04 0.07 0.07 0.04 0.05 0.02 0.02 1.00 0.98
Sloptown 0.00 0.02 0.04 0.00 0.00-0.01 0.00 0.00 0.99
Airport 0.01 0.02 0.06 0.00 0.01 0.00 0.01 0.02 0.01

 

46

Figure 3 UPGMA phenogram of allozymes among plots

+ ................... Garden Island
+1
+2 ------------------ Footbridge
I
+-——23 ------------------ CMU
I I
+-31 + ------------------- DUDA
II
+ .............. 34+ ----------------------- French bay
I I
I | + ------------------- WUjek
+ ---------------- 37 + ----- 30
I I + ___________________ Airport
I I
l l + ----------------------- Hog-WEST
I + ---------------- 33
I I + ................... Powerline
i +—--29
I + ................... Blue trail
I
, __________ 33 +.n_HogN-east
i 1 +4
I I IsanogS-east
I I I
I I + ----- zen—Font lake
I I I I
I I +2 +~u-Sloptown
I I I
39 I I +~n-HighN-East
I + ---------------------------------------------- 35----2e
I I +mn-Barneys lake
1 l
I + ---------- High WEST
I
I + ---------- HogM-east
I + -------------- 27
+ __________________________________________ 36 + ---------- Highs-east
I
+ ......................... Greens bay
I ---------------- I ---------------- I ---------------- I ---------------- I
0.015 0.011 0.007 0.004 0.000

47

Microsatellite genetic diversity measures for islands, Tables 14-17

The microsatellites genetic diversity measures for all islands are found in Tables 14-17.
Table 14 shows the haplotype frequencies. The most common haplotypes (numbers 1, 2
and 3) were found on all four islands. All haplotypes were found on Beaver Island, those
with a frequency of 0.005 were found in a single tree. The small islands had 4-6 rare
alleles (frequencies less than 0.1), and only one rare haplotype, number 6, was found on
all four islands. Table 15 contains the results for the descriptive statistics. Descriptive
measures were found to be consistent with the mean among islands for measures of
average number of haplotypes per locus (A mean = 10.5) and gene diversity (He mean =
0.80). Garden Island has the smallest population of white pine, and had the lowest
number of haplotypes (7). Both Hog and High Islands had nine haplotypes, but not all of
these were the same. Beaver Island had the largest ntunber of white pine trees of any
island, and the largest number of haplotypes (17). There were four private haplotypes
found on Beaver Island (haplotype numbers 11, 13, 16, 17). Each of these occurred in a
single tree. Table 16 lists the results for Fst and Rst. The Rst value among islands was
0.04. The overall Fst value was 0.01, but haplotype number 14 had an Fst value of 0.08.
Haplotype 14 had a relatively high frequency of 0.098, on High Island, and a singe tree
had haplotype 14 on Beaver Island. All other haplotypes had very low Fst values. Table
17 shows the values for Nei’s genetic distance and genetic identity. The values for Nei’s
genetic distance were higher for microsatellites than for allozymes, ranging from 0.02 to

0.13. Values for genetic identity were high, ranging from 0.88 to 098.

48

 

 

 

 

 

 

 

 

 

Table 14 Haplogg fr_e_guencies among islands

Haplotype Garden HoLHigh Beaver Mea_n_
N (49) (37) (41) (185)
1 0245 0.324 0.146 0.200 0.229
2 0.367 0.270 0.390 0.232 0.315
3 0.143 0.189 0.146 0.227 0.176
4 0.061 0.054 0.000 0.059 0.044
5 0.082 0.000 0.024 0.081 0.047
6 0.082 0.054 0.049 0.038 0.056
7 0.020 0.000 0.000 0.011 0.008
8 0.000 0.027 0.073 0.022 0.031
9 0.000 0.027 0.049 0.070 0.037
10 0.000 0.027 0.000 0.005 0.032
11 0.000 0.000 0.000 0.005 0.001
12 0.000 0.027 0.000 0.016 0.011
13 0.000 0.000 0.000 0.005 0.001
14 0.000 0.000 0.098 0.005 0.026
15 0.000 0.000 0.024 0.011 0.008
16 0.000 0.000 0.000 0.005 0.001
17 0.000 0.000 0.000 0.005 0.001

 

The most common haplotypes 1,2 and 3 were found on all four islands and over 72% of
the sampled trees consisted of one of the three most common haplotypes.

49

Table 15 Descriptive statistics for haplotypes among islands including measures of,
sample size (n), number of haplotypes (A), gene diversity (He)

 

 

a) Island n A He
Garden 49 7 0.78

Hog 40 9 0.79

High 42 9 0.78

Beaver 181 17 0.84

Mean 78.00 10.50 0.80

b) Private haplotypes among islands

Haplotype Frequency Found in

17 0.001 Beaver
16 0.001 Beaver
13 0.001 Beaver
1 1 0.001 Beaver

50

Table 16 Fst of haplotypes among islands

 

 

IflmkWHfi Ikt
17 -0.01
16 -0.01
13 -0.01
11 -0.01
15 -0.01
14 0.08
8 0.01
10 0.00
12 —0.01
9 0.01
6 -0.01
5 0.01
4 0.00
7 -0.01
3 0.00
2 0.02
1 0.01
Overall 0.01

51

Table 17 Nei's Genetic distance (below) and genetic identity (above) of haplotypes
among islands

Ga Ho Hi Be
Garden 0.98 0.95 0.95
Hog 0.02 0.88 0.95
High 0.05 0.13 0.89

Beaver 0.05 0.05 0.11

52

Microsatellite genetic diversity measures among plots, Tables 18-20

Microsatellite genetic diversity measures among all plots are found in Tables 18-20.
Table 18 contains the results for the descriptive statistics. Descriptive measures were
found to be consistent with the mean among plots for measures of average nrunber of
haplotypes per locus (A mean = 6.1) and gene diversity (He mean = 0.76). The only
exception was plot Hog East North which had only two haplotypes and a gene diversity
measure of 0.48, and this is most likely due to the low sample size. Four haplotypes were
found in a single plot, in a single tree. Haplotype 18 was found at the Airport plot,
haplotype 16 was found in the Dude plot, haplotype 13 was found only in the Wujek plot,
and haplotype eleven was found only in the Footbridge plot. Table19 lists the results for
Fst. The Rst value was 0.01. The overall Fst value was 0.03, but the haplotype number
14 had an Fst value of 0.08. Haplotype 14 had a high frequency (for a rare haplotype) on
High Island, and a singe tree had haplotype 14 on Beaver Island. Interestingly, the most
common haplotypes (number 1, 2 and 3) also had relatively high F st values (0.04, 0.04
and 0.07 respectively). All other haplotypes had very low Fst values. Table 20 shows the
values for Nei’s genetic distance and genetic identity. The values for Nei’s genetic
distance were variable, ranging fiom -0.05 to 1.91. Hog East Middle had the highest
value between Powerline with a value of 1.91. Hog East Middle also had several other
values greater than 1.0 between other plots on the archipelago. Hog East Middle had
several rare haplotypes which would have increased the measure of genetic distance.

Values for genetic identity were also variable, ranging from 0.11 to 1.02. Hog East

. Middle had the lowest values, suggesting it had the least number of haplotypes in

common with other plots. Interestingly, the highest genetic identity value was between

53

Garden Island and the tiny population Hog West. This may suggest that the prevailing
wind allowed Garden Island trees to pollinate the small number of trees in the Hog West
plot. Figure 4 shows the UPGMA phenogram of haplotypes among plots. There was
little consistency with grouping of plots with their geographical locations.

The Haplotype frequencies among plots can be found in the Appendix.

54

Table 18 Descriptive statistics for haplotypes among plots including measures of,
sample size (n), number of haplotypes (A), gene diversity (He)

 

 

 

a) Population n A He
Garden 49 7 0.78
Hog-West 10 4 0.76
Hog-East North 6 2 0.48
Hog-Midd 14 7 0.67
Hog-East South 10 4 0.74
High-East South 13 6 0.74
High-West 15 6 0.74
High-East North 14 7 0.83
Power line 12 6 0.83
Blue trail 19 6 0.66
Footbridge 16 7 0.77
French bay l7 8 0.79
Greens bay 18 7 0.82
Wujek 14 6 0.81
CMU 15 6 0.81
Duda 13 8 0.86
Font LK l7 7 0.83
Barney's LK 13 5 0.78
Sloptown 17 7 0.82
Airport 11 6 0.76
Mean 15.7 6.1 0.76
b) Private haplotypes among plots
Haplotype Frequency Foujd in
17 0.01 Airport (Beaver)
16 0.01 Duda (Beaver)
13 0.01 Wujek (Beaver)
11 0.01 Footbridge (Beaver)

55

Table 19 Fst of haplotypes among plots

Haplotype F st

 

17 0.03
16 0.01
13 0.01
11 0.00
15 0.00
14 0.08
8 -0.02
10 0.00
12 0.02
9 0.01
6 0.00
5 0.00
4 0.01
7 -0.02
3 0.07
2 0.04
1 0.04
Overall 0.03

Table 20 Nei's genetic distance (below) and genetic identity (above) of haplotypes

among plots (in two blocks)

GI Hi—W HN-E HM-E
0.93 0.86 0.62
0.57 0.85
0.11

Garden
Hog-West 0.08
Hog-EastNorth 0.15 0.57

Hog-Midd 0.47 0.16 2.16
Hog-EastSouth 0.07 0.34 0.00 1.09

HoS-E HOS-E HoN-E Hi-W PL

0.93 0.86 0.90 1.02
0.71 0.61 0.69 0.97
1.00 1.05 0.90 0.79
0.34 0.15 0.38 0.80

0.97 0.83 0.87

.90
.65
.93
.15
.93

BT
0.61
0.77
0.50
0.47
0.72

 

High-EastSouth 0.15 0.49-0.05 1.90
High~West 0.11 0.38 0.10 0.95

0.03
0.18 0.03

0.98 0.82
0.86

.99
.86

0000000

0.49
0.32

 

High-EastNorth-0.02 0.03 0.23 0.22
Powerline 0.10 0.43 0.07 1.91
Bluetrail 0.50 0.26 0.69 0.76

Footbridge 0.06 0.46 0.11 0.92
Frenchbay 0.35 0.31 0.25 1.50
Greendsbay 0.20 0.10 0.86 0.06
Wujek 0.12 0.21 0.08 0.79

CMU 0.25 0.22 0.45 0.67

0.14 0.20 0.15

0.07 0.01 0.16 0.24
0.33 0.72 1.15 0.36
0.18 0.11 0.10 0.07
0.16 0.18 0.60 0.32
0.37 0.67 0.49 0.06
0.03 0.05 0.26 0.01
0.26 0.44 0.70 0.09

0.79

0.72
0.07
0.16
0.65
0.23
0.31

0.69
0.49

1.05
0.08
0.61
0.21
0.04

 

Duda 0.18-0.01 0.96 0.08
FontLK 0.01 0.00 0.22 0.36

0.32 0.83 0.73 0.08
0.04 0.17 0.26-0.06

0.55
0.18

0.20
0.15

 

BarneysLK-0.01 0.01 0.19 0.30
Sloptown 0.02 0.31 0.18 0.54

0.09 0.19 0.17-0.10
0.08 0.18 0.07 0.05

0.21
0.12

0.22
0.83

 

Airport 0.16 0.63 0.14 1.00

FBFByGBWK
Garden 0.94 0.70 0.82 0.89

0.18 0.11 0.11 0.19

BL
1.01

(JHU’ DA. FL
0.78 0.84 0.99

0.25

SN
0.98

1.60

.AT
0.85

 

Hog-West 0.63 0.73 0.90 0.81
Hog-EastNorth 0.90 0.78 0.42 0.93

0.99
0.83

0.81 1.01 1.00
0.64 0.38 0.81

0.73
0.84

0.53
0.87

 

Hog-Midd 0.40 0.22 0.94 0.45
Hog-EastSouth 0.83 0.85 0.69 0.97
High-EastSouth 0.89 0.83 0.51 0.95
High-West 0.90 0.55 0.61 0.77

0.51 0.92 0.70
0.77 0.73 0.96
0.64 0.44 0.85
0.49 0.48 0.77

0.74
0.92
0.83
0.84

0.58
0.92
0.83
0.93

0.37
0.84
0.90
0.89

 

High-EastNorth 0.93 0.73 0.94 0.99
Powerline 0.93 0.85 0.52 0.80

0.91 0.92 1.06
0.73 0.58 0.84

1.11
0.81

0.95
0.89

0.83
0.78

 

Bluetrail 0.35 0.92 0.54 0.81
Footbridge 0.57 0.62 0.82

0.96 0.82 0.86
0.69 0.56 0.84

0.81
0.88

0.44
0.99

0.20
0.94

 

Frenchbay 0.57 0.47 0.94
Greens bay 0.47 0.75 0.74
Wujek 0.20 0.06 0.31

CMU 0.37 0.06 0.45 0.06

0.94 0.67 0.91
0.64 1.05 0.94
0.94 0.75 1.06

0.88 0.99

0.82
0.85
0.97
0.94

0.57
0.78
0.78
0.68

0.43
0.73
0.80
0.43

 

Duda 0.58 0.41-0.05 0.28
FontLK 0.17 0.09 0.06-0.05

0.91
1.06

0.12
0.01-0.02

1.02

0.75
0.88

0.50
0.78

 

BarneysLK 0.13 0.20 0.16 0.03
Sloptown 0.01 0.56 0.25 0.25
Airport 0.06 0.84 0.31 0.22

0.06 0.10-0.05
0.39 0.29 0.13 0.08
0.85 0.69 0.25 0.32

57

0.92

0.10

0.73
0.90

Figure 4 UPGMA phenogram of haplotypes among plots

+~Garden
+1
+22-High—EastNorth
H
+ ................ 2a~Barneys LK
I I
I I»_Wujek
| +25
I +—-F0nt LK
I

+ ..................................... 37 + ----- Hog-EastNorth
+4
azwmuHighEastSouth
II
+ .......... 30+ ----- Hog-EastSouth
I I
I + ------- Powerline
~36
I + ............. High-West
I I
I +~~44+ ---------- Footbridge
I I +26
I +a3 ---------- Sloptown
I I
I + ........... Airport
I
I
I

+ ...... Bluetrail
+—-29

1 ......... Frenchbay

+ ------------ Hog-W881:
+3

+31 ----------- Duda
H
+ ...................................................... 35 ------------ Greens bay

58

Allozyme genetic diversity measures within islands, Tables 21-30

Allozyme genetic diversity measures within Garden Island, Tables 21-22

Table 21 contains the results of the descriptive statistics for Garden Island. Descriptive
measures for all allozyme loci were found to be consistent with the mean within Garden
Island for measures of the average number of alleles per locus (A mean = 2.40), expected
proportion of heterozygotes (He mean = 0.32) and the observed proportion of
heterozygotes (Ho mean = 0.31) The locus Tpi-2 had values much lower than the mean
He (Tpi-2, He = 0.05) and Ho (Tpi-2, Ho = 0.05) This could be due to the low frequency
of the second Tpi-2 allele. The loci Mdh-Z had values much higher than the mean He and
Ho. The value for He for Mdh-2 was 0.53 and for Ho was 0.46. The measure of
inbreeding coefficient (0 was highest for Aat-3 (0.2) and the lowest for locus Pgm-l (-
0.24). Since there were no individual plots within Garden Island, the only F statistic that
could be estimated was Fis, found in Table 22. There was no significant deviation from

zero (95% confidence interval -0. 16 to 0.15).

59

Table 21 Descriptive statistics measured with allozymes within Garden Island
including measures of, sample size (n), proportion of loci polymorphic (P), average
number of alleles per locus (A), expected proportion of heterozygotes (He), observed
proportion of heterozygotes (Ho), and inbreeding coefficient (1)

 

 

Locus n A He Ho f
Pgm—l 57.00 2.00 0.32 0.40 -0.24
Pgi-2 57.00 3.00 0.26 0.26 -0.01
Mdh-2 57.00 3.00 0.51 0.46 0.11
Tpi-2 57.00 2.00 0.05 0.05 -0.02
Aat-3 57.00 2.00 0.44 0.35 0.20

All 57.00 2.40 0.32 0.31 0.04

60

Table 22 Fis measured with allozymes within Garden Island

 

 

Locus Allele f

Pgm-l All -0.24

2 -O.24

1 -0.24

Pgi-2 All -0.01

2 1.00

1 —0.14

3 -0.03

Mdh-2 All 0.11

3 0.23

2 0.09

1 0.08

Tpi-2 All -0.02

1 -0.02

2 -0.02

Aat-3 All 0.20

2 0.20

1 0.20

Overall 0.04
Upper bound 0.15
Lower bound -0.16

The 95 % confidence interval from boot strapping over loci

61

Allozyme genetic diversity measures within Hog Island, Tables 23-25

Table 23 contains the results for the descriptive statistics for Hog Island. Descriptive
measures were found to be consistent with the mean within Hog Island for measures of
the proportion of loci polymorphic (P mean = 0.75), average number of alleles per locus
(A mean = 2.0) and polymorphic locus (Ap mean = 2.3), expected proportion of
heterozygotes (He mean = 0.3) and the observed proportion of heterozygotes (Ho mean =
0.28) The only exception was the plot Hog West which only had nine samples. Hog
West expected proportion of heterozygotes (He)was only 0.18 and the Ho value was only
0.18. Hog East Middle was the only plot to have a single copy of the allele Pgi-2 allele
number 2. The measure of inbreeding coefficient (1) was highest for Hog South East
(0.30) which may have skewed the overall mean which was (0.09). Table 24 shows the
values for standard F statistics. There is no significant divergence from zero for Fis, Fit
or F st. Table 25 shows the values for Nei’s genetic distance and genetic identity. The
values for Nei’s genetic distance were quite low, ranging from -0.01 to 0.03. Values for

genetic identity were high, ranging from 0.96 to 1.01.

62

Table 23 Descriptive statistic measures of genetic diversity within Hog Island

including measures of, sample size (n), proportion of loci polymorphic (P), average
number of alleles per locus (A) and polymorphic locus (Ap), expected proportion of

heterozygotes (He), observed proportion of heterozygotes (Ho), and inbreeding

 

 

 

coefficient (f)
a) Population n P A Ap H H
Hog-West 9.00 0.60 1.80 2.33 0.18 0.18 0.00
HogN-East 16.00 0.80 2.00 2.25 0.34 0.35 0.03
HogM-East 16.00 0.80 2.20 2.50 0.34 0.32 0.05
Hogs-East 12.00 0.80 2.00 2.25 0.35 0.25 0.30
Mean 13.25 0.75 2.00 2.33 0.30 0.28 0.09
b) Private alleles within Hog Island
Locus Allele Frequency Found in
Pgi-2 2 0.03 Hog East-Middle

63

Table 24 F statistics measured using allozymes within Hog Island estimating the
parameters f (=F is), F (=Fit), and 9 (=Fst)

 

 

Locus Allele f F 8
Pgm-l All 0.24 0.21 -0.04
2 0.24 0.21 -0.04

1 0 24 0.21 —0 04

Pgi-2 All -0.25 -0.12 0.11
2 0.01 0.00 -0.01

l -0.31 -0.16 0.12

3 -0.20 -0.08 0.10

Mdh-2 All 0.09 0.07 -0.02
3 0.17 0.16 -0.01

2 0.12 0.09 -0.04

1 0.03 0.02 -0.01

Tpi-Z All *** *** ***
2 *** *** ***

Aat-3 All 0.27 0.29 0.02
2 0.27 0.29 0.02

1 0.27 0.29 0.02

Overall 0.09 0.10 0.02

Upper bound 0.26 0.27 0.08

Lower bound -0.13 -0.05 -0.03

The 95 % confidence interval from boot strapping over loci

Table 25 Genetic distance (below) and genetic identity (above) of allozymes within
Hog Island

Ho-W HN-E HM-E HS-E

Hog—WEST 0.96 0.98 0.97

HogN-east 0.04 1.00 1.01

HogM-east 0.02 0.00 1.00
Hogs-east 0.03-0.01 0.00

65

Allozyme genetic diversity measures within High Island, Tables 26-28

Table 26 contains the results for the descriptive statistics for High Island. Descriptive
measures were found to be consistent with the mean within plots for measures of the
proportion of loci polymorphic (P mean = 0.93 ), average number of alleles per locus (A
mean = 2.33) and polymorphic locus (Ap mean = 2.43), expected proportion of
heterozygotes (He mean = 0.37) and the observed proportion of heterozygotes (Ho mean
= 0.35). The measure of inbreeding coefficient (f mean = 0.04) was highest for High
North East (f = 0.12) which may have skewed the overall mean. Table 276 shows the
values for standard F statistics. There is no significant divergence from zero for Fis, Fit
or F st. Table 28 shows the values for Nei’s genetic distance and genetic identity. The
values for Nei’s genetic distance were quite low, ranging from 0.01 to 0.07. Values for

genetic identity were high, ranging from 0.94 to 0.99.

66

Table 26 Descriptive statistics measures within High Island including measures of,
sample size (n), proportion of loci polymorphic (P), average number of alleles per
locus (A) and polymorphic locus (Ap), expected proportion of heterozygotes (He),
observed proportion of heterozygotes (Ho), and inbreeding coefficient (1‘)

Population n P A Ap He Ho f
Highs-East 5.00 0.80 2.20 2.50 0.41 0.40 0.02
HighN-East 15.00 1.00 2.40 2.40 0.38 0.33 0.12
High West 14.00 1.00 2.40 2.40 0.32 0.33 -0.02
Mean 11.33 0.93 2.33 2.43 0.37 0.35 0.04

 

 

67

Table 27 F statistics measured with allozymes within High Island F statistics
estimates the parameters f (=Fis), F (=F it), and O (=Fst)

 

 

Locus Allele f F 9
Pgm-l All 0.28 0.28 0.00

2 0.28 0.28 0.00

1 0.28 0.28 0.00

Pgi-2 All -o.01 -0.02 0.00

l -0.41 -0.35 0.04

2 1.00 1.00 -0.10

3 -O.22 -0.22 0.00

Mdh-Z All -0.11 -0.10 0.01

3 -0.06 -0.10 -0.04

2 -0.13 -0.11 0.02

1 -0.12 -0.10 0.01

Tpi-2 All 0.01 -0.03 -0.04

1 0.01 -0.03 -0.04

2 0.01 -0.03 -0.04

Aat-3 All 0.26 0.35 0.12

1 0.26 0.35 0.12

2 0.26 0.35 0.12

Overall 0.05 0.08 0.03
Upper bound 0.25 0.32 0.09
Lower bound -0.07 -0.07 -0.01

The 95 % confidence interval from boot strapping over loci

68

Table 28 Nei's genetic distance (below) and genetic identity (above) of allozymes
within High Island

His-E HiN-E Hi-W

High S-east 0.99 0.94

High N-East 0.01 0.99
High West 0.07 0.01

69

Allozyme genetic diversity measures within Beaver Island, Tables 29-31

Table 29 contains the results for the descriptive statistics for Beaver Island. Descriptive
measures were found to be consistent with the mean within plots for measures of the
proportion of loci polymorphic (P mean = 0.82), average number of alleles per locus (A
mean = 2.12) and polymorphic locus (Ap mean = 2.39), expected proportion of
heterozygotes (He mean = 0.31) and the observed proportion of heterozygotes (Ho mean
= 0.3). The overall mean for Fis was 0.03. The measure of Fis was highest for Barney's
lake, Sloptown and Airport (Fis= 0.12, 0.11 and 0.26 respectively). While the measure of
Fis was lowest in the Blue trail and Duda plots (-0.19 and -0.24 respectively). Table 30
shows the values for standard F statistics. There was significant divergence from zero for
Fit, (confidence internal was 0.05 to 0.09) and for Fst (confidence interval was 0.01 to
0.08). Table 31 shows the values for Nei’s genetic distance and genetic identity. The
values for Nei’s genetic distance were quite low, ranging from -0.01 to 0.07. Values for
genetic identity were high, ranging from 0.93 to 1.01. Figure 5 shows the UPGMA
phenogram of allozymes among plots. There was some consistency of grouping of plots
with there geographical locations. Plots, Font Lake, Sloptown and Airport are located in
close proximity geographically and were grouped together in a branch. Plots Wujek,
CMU and Duda are located in close proximity geographically and were grouped together,
although French Bay and Foot bridge plots are distant plots in geographical area but

were also included in this branch.

70

Table 29 Descriptive statistics measured within Beaver Island including measures
of, sample size (n), proportion of loci polymorphic (P), average number of alleles per
locus (A) and polymorphic locus (AP), expected proportion of heterozygotes (He),
observed proportion of heterozygotes (Ho), and inbreeding coefficient (1)

 

 

 

 

 

 

 

Population n P A Ap He Ho f
Power line 15.00 0.80 2.00 2.25 0.25 0.23 0.09
Blue trail 15.00 1.00 2.20 2.20 0.33 0.39 -0.19

FootBridge 11.00 0.80 2.00 2.25 0.27 0.25 0.05
French bay 10.00 0.80 2.00 2.25 0.24 0.22 0.09
Greens bay 8.00 0.40 1.60 2.50 0.22 0.22 0.00

Wujek 17.00 0.80 2.20 2.50 0.30 0.27 0.09

CMU 18.00 0.80 2.20 2.50 0.29 0.29 0.00

DUDA 22.00 0.80 2.40 2.75 0.35 0.36 -0.05

Font lake 21.00 1.00 2.20 2.20 0.33 0.40 -0.24
Barney's lake 20.00 0.80 2.20 2.50 0.39 0.34 0.12
Sloptown 18.00 0.80 2.40 2.75 0.37 0.33 0.11
Airport 21.00 1.00 2.00 2.00 0.34 0.26 0.26
Mean 16.33 0.82 2.12 2.39 0.31 0.30 0.03

71

Table 30 F statistics measured within Beaver Island F statistics estimates the
parameters f (=Fis), F (=Fit), and 9 (=F st)

 

 

Locus Allele f F 9
Pgm-l All 0.05 0.04 -0.01
2 0.05 0.04 -0.01

1 0.05 0.04 -0.01

Pgi-2 All -0.02 0.08 0.10
4 0.01 0.00 -0.01

2 0.67 0.69 0.05

1 -0.23 -0.13 0.08

3 -0.04 0.09 0.13

Mdh-2 All 0.03 0.06 0.02
3 0.11 0.15 0.05

1 0.00 0.02 0.03

2 0.04 0.04 0.01

Tpi-2 All 0.16 0.21 0.05
l 0.16 0.21 0.05

2 0.16 0.21 0.05

Aat-3 All 0.05 0.07 0.02
2 0.05 0.07 0.02

1 0.05 0.07 0.02

Overall 0.03 0.07 0.04

Upper bound 0.06 0.09 0.08

Lower bound 0.00 0.05 0.01

The 95 % confidence interval from boot strapping over loci

72

Table 31 Genetic distance (below) and genetic identity (above) of allozymes within

 

 

 

 

 

Beaver Island

PL BT FB FBy GB WK CMU DA FL BL SN AT
Power line 1.00 0.99 0.99 0.94 0.97 1.00 0.99 0.99 0.94 0.98 0.96
Blue trail 0.00 0.99 0.99 0.97 0.97 0.99 0.98 0.97 0.93 0.97 0.95
Footbridge 0.01 0.01 1.01 0.98 1.00 1.01 1.01 0.99 0.96 1.00 0.99
French bay40.01 0.01-0.01 0.98 1.00 1.00 1.00 0.97 0.93 0.98 0.98
Greens bay 0.06 0.03 0.02 0.02 0.96 0.95 0.96 0.93 0.93 0.96 0.94
Wujek 0.03 0.04 0.00 0.00 0.04 .00 1.00 0.98 0.96 1.00 1.00
CMU 0.00 0.01-0.01 0.00 0.05 0.00 1.01 0.99 0.95 1.00 0.99
DUDA 0.01 0.02-0.01 0.00 0.04 0.00-0.01 1.00 0.98 1.01 1.00
Font lake 0.01 0.03 0.01 0.03 0.08 0.02 .01 0.00 0.98 1.00 0.99
Barney's 0.07 0.08 0.04 0.07 0.07 0.04 .05 0.02 0.02 1.00 0.98
Sloptown 0.02 0.03 0.00 0.02 0.04 0.00 .00-0.01 0.00 0.00 0.99
Airport 0.04 0.05 0.01 0.02 0.06 0.00 .01 0.00 0.01 0.02 0.01

 

73

Figure 5 UPGMA phenogram of plots within Beaver Island

+Powerline
+ --------------------------- 13

+Bluetrail

.eFootbridge
M

1 .............................. 21 +15Frenchbay
I H

i + ----------- 18-CMU
I I I
I

I

I I+—WUjek
I +16
+ ............. 20 +-DUDA
+ -------- 22 I
I waOUt lake
I + -------- 17

I
I
I +--19 +--Sloptown
I
I
I

+ ........... Airport

+ ........................................................... Barney's lake

74

Genetic diversity measured with haplotypes within islands Tables 32-41

Table 32 contains the results for the descriptive statistics for Garden Island. There was
only one plot on Garden Island so there is no mean value, and no estimation of F
statistics.

Table 32 Descriptive statistics measured within Garden Island including measures
of, sample size (n), number of haplotypes (A), gene diversity (He)

Island n A He
Garden 49 7 0.78

 

75

Genetic diversity measured with haplotypes within Hog Island
Tables 33-35

Table 33 contains the results for the descriptive statistics for Hog Island. Descriptive
measures were found to be consistent with the mean within plots for measures of average
number of haplotypes per locus (A) and gene diversity (He). The exception being Hog
East North which had a small sample size (6). The mean He was 0.66 and the He within
Hog East North was 0.48 also there were only two haplotypes in Hog East North when
the average number of haplotypes per plot was 4.25. There were several private
haplotypes within Hog Island, haplotype numbers 8, 10, 12, and 9 were found in single
trees in Hog East Middle, and haplotype number 6 was found in 2 of the 10 trees in Hog
West. Table 34 shows that the overall measure of Fst was quite high 0.13. Haplotype 1
had an F st value of 0.21 haplotype 2 had an overall value of 0. 19 and haplotype 6 had an
overall value of 0.1 1. The four private haplotypes found in Hog East Middle had
negative Fst values, possibly because they were found in single copies in one sample plot.
Table 35 shows the values for Nei’s genetic distance and genetic identity. The values for
Nei’s genetic distance were variable, ranging from 0.00 to 2.16. The high value (2.16)
was between Hog East Middle and Hog East North. While the low value (0.00) was
between Hog East North and Hog East South. Nei's genetic identity had a low between
Hog East Middle and Hog East North (0.1 l) and a high of 1.0 between. Hog East North

and Hog East South.

76

Table 33 Descriptive statistics measured within Hog Island including measures of,
sample size (n), number of haplotypes (A), gene diversity (He)

 

 

 

a) Population n A He
Hog-West 10 4 0.76
Hog-East North 6 2 0.48
Hog-Midd 14 7 0.67
Hog-East South 10 4 0.74
Mean 10.00 4.25 0.66

b) Private haplotypes within Hog Island

Haplotypes Frequency Found in

 

8 0.07 Hog-Midd
10 0.07 Hog-Midd
12 0.07 Hog-Midd
9 0.07 Hog-Midd
6 0.20 Hog—West

77

Table 34 Fst measured within Hog Island

 

 

Haplotype Fst
All 0.13

4 0.02

8 -0.04

10 -0.04

12 -0.04

9 -0.04

3 0.07

2 0.19

6 0.11

1 0.21
Overall 0.13

78

Table 35 Genetic distance (below) and genetic identity (above) of haplotypes within
Hog Island

Ho-W HN-E HM-E HS-E

Hog-West 0.57 0.85 0.71
Hog-East North 0.57 0.11 1.00
Hog-Midd 0.16 2.16 0.34

Hog-East South 0.34 0.00 1.09

79

Genetic diversity measured with haplotypes within High Island
Tables 36-38

Table 36 contains the results for the descriptive statistics for haplotypes within High
Island. Descriptive measures were found to be consistent with the mean among plots for
measures of average number of haplotypes per locus (A mean = 6.33) and gene diversity
(He Mean = 0.77). Table 37 shows that the overall measure of Fst was 0.00. Table 38
shows the values for Nei’s genetic distance and genetic identity. The values for Nei’s
genetic distance were variable but low, ranging from 0.03 to 0.2. Nei's genetic identity

was relatively high ranging from 0.82 to 0.98.

80

Table 36 Descriptive statistics measured within high island including measures of,
sample size (n), number of haplotypes (A), gene diversity (He)

 

 

Population n A He
High-East South 13 6 0.74
High—West 15 6 0.74
High—East North 14 7 0.83
Mean 14.0 6.3 0.77

81

Table 37 Fst measured within High Island

Haplotype Fst

 

 

All 0.00
5 0.00
15 0.00
1 0.09
14 0.05
9 -0.03
6 -0.04
8 —0.08
2 -0.03
3 -0.02

Overall 0.00

82

Table 38 Genetic distance (below) and genetic identity (above) of haplotypes within
High Island
HE-S Hi-W HE-N
High East South 0.98 0.82
High-West 0.03 0.86
High East North 0.20 0.15

83

Genetic diversity measured with haplotypes within Beaver Island
Tables 39—41

Table 39 contains the results for the descriptive statistics for haplotypes within Beaver
Island. Descriptive measures were found to be consistent with the mean among plots for
measures of average number of haplotypes per locus (A mean = 6.33) and gene diversity
(He mean = 0.8). The exceptions being Blue Trail, CMU and Barney's lake which had
values for the number of haplotypes (5 for all three) lower than the mean of 6.3. The He
value was lowest in Blue Trail, 0.63 compared to the overall mean of 0.80. There were
four plots which had private haplotypes, Airport had a single copy of haplotype l7, Duda
had a single copy of haplotype 16, Wujek had had 8 since copy of haplotype 13 and
Footbridge had a single copy of haplotype 11. Table 40 shows that the overall measure
of F st was 0.03. Haplotype number 3 had the highest individual Fst value of 0. 1 1. Table
41 shows the values for Nei’s genetic distance and genetic identity. The values for Nei’s
genetic distance were variable ranging fiom 0.07 to 1.60. The high value for Nei's
genetic distance was between Blue Trail and Airport. Nei's genetic identity was variable
with a low value of 0.35 to a high of 1.06. The low value (0.34) was between Blue Trail
and Footbridge which is interesting since the two plots are located in relatively close

proximity

84

Table 39 Descriptive statistics measured within Beaver Island including measures
of, sample size (n), number of haplotypes (A), gene diversity (He)

 

 

 

a) Population n A He
Power line 11 6 0.81
Blue trail 16 5 0.63
Footbridge 13 7 0.76
French bay 15 7 0.75
Greens bay 16 7 0.85
Wujek 10 6 0.86
CMU 14 5 0.78
Duda 11 8 0.88
Font LK 17 7 0.83
Barney's 13 5 0.78
Sloptown l7 7 0.82
Airport 9 6 0.84
Mean 13.50 6.33 0.80
b) Private haplotypes within Beaver Island
Haplotype Fregggncy Found in
17 0.09 Airport (Beaver)
16 0.08 Duda (Beaver)
13 0.07 Wujek (Beaver)
11 0.06 Footbridge (Beaver

85

Table 40 Fst measured within Beaver Island
Haplogpg Fst
All 0.03
17 0.04
14 -0.02
16 0.02
15 0.01
0.03
0.02
0.00
0.00
0.02
-0.02
0.01
-0.02
-0.01
0.03
-0.01
0.11
0.01
Overall 0.03

 

i—III-‘I—‘i—d

86

Table 41 Genetic distance (below) and genetic identity (above) of haplotypes within

 

 

 

 

 

Beaver Island

PL BL FE FBy GY WK CMU DA FL BL SN AT
Power line 0.49 0.93 0.85 0. 52 o. 80 0.73 0. 58 o. 84 0.81 0. 89 0.78
Blue trail 0.72 0.35 0.92 0.54 0.81 0.96 0.82 0.86 0.81 0.44 0.20
Footbridge 0.07 1.05 0.57 0.62 0.82 0.69 0.56 0.84 0.88 0.99 0.94
French bay 0.16 0.08 0.57 0.47 0.94 0.94 0.67 0.91 0.82 0.57 0.43
Greens bay 0.65 0.61 0.47 0.75 0.74 0.64 1.05 0.94 0.85 0.78 0.73
Wujek 0.23 0.21 0.20 0.06 0.31 0.94 0.75 1.06 0.97 0.78 0.80
CMU 0.31 0.04 0.37 0.06 0.45 0.06 0.88 0.99 0.94 0.68 0.43
Duda 0.55 0.20 0.58 0.41—0.05 0.28 0.12 1.02 0.91 0.75 0.50
Font LK 0.18 0.15 0.17 0.09 0.06-0.05 0.01-0.02 1.06 0.88 0.78
Barney's 0.21 0.22 0.13 0.20 0.16 0.03 0.06 0.10-0.05 0.92 0.73
Sloptown 0 12 0.83 0.01 0.56 0.25 0.25 0.39 0.29 0.13 0.08 0.90
Airport 0.25 1.60 0.06 0.84 0.31 0.22 0.85 0.69 0.25 0.32 0.10

 

87

Spatial autocorrelation analysis Tables 42 and 43

Table 42 contains the results from Moran's I statistic among all plots in the archipelago.
Alleles Pgrn-l, Pgi-2b, Pgi-2c and Mdh-2b had significant overall correlograms. The
SND showed significant positive spatial autocorrelations in distance class one (0-5127
meters) and in distance class 2 (5128m-8454m) Also, negative spatial autocorrelation was
detected in distance class 4 (10907m-13517m). The analysis was repeated without Pgi-
2b, which was a relatively low frequency allele. There was significant positive spatial
autocorrelation at the first distance class when excluding Pgi-2b and significant negative

spatial autocorrelation at the fifth distance class (13518m -18308m)

88

Table 42

Moran's I statistic for the archipelago

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

No. Pairs 27 27 27 27 27 27 28

Distance 5127 8454 10906 13517 18308 22944 30776 ph Mean

ciass' allele
frequenc
yamong
plots

PGM-1 0.16 -o.35 0.10 0.24 -0.60 -0.11 0.19 0.01 0.862

PGi-2a 0.06 0.20 0.00 -0.18 -0.18 -020 -0.07 0.39 0.190

PGi-2b 0.33 0.32 -0.02 -0.54 -0.22 -0.20 -0.03 0.01 0.053

PGi-Zc 0.28 0.36 0.10 -0.48 -0.33 -0.29 0.00 0.04 0.754

MDH-2a 0.17 -0.14 0.13 -0.11 -0.38 -0.15 0.11 0.18 0.629

MDH-2b 0.21 0.26 0.21 -0.28 -0.48 0.06 -0.34 0.04 0.267

MDH-2c 0.29 -0.21 -0.22 -0.08 -0.01 0.01 -0.14 0.12 0.104

AAT-2 -0.22 -0.01 0.08 -0.13 -0.14 0.08 -0.02 1.00 0.659

Average! 0.16 0.05 0.05 -0.2 -029 -0.1 -0.04

gvgrage 0.16 0.16 0.17 0.17 0.17 0.17 0.14

SND° 3.56 1.89 1.68 -2.44 -4.01 -0.08 0.25

p° 0.0002 0.03 0.056 0.007 0.0000 0.468 0.404

SND“ 2.95 1.14 1.72 -1.50 -309 -050 0.20

P“ 0.002 0.13 0.06 0.07 0.001 0.31 0.2

 

 

 

 

 

 

 

 

 

 

a Upper bound for distance class in meters

 

0 Overall correlogam shinifrcance

 

c including PGl-2b

 

 

d excluding PGi-2b

 

89

 

Table 43 contains the results from Moran's 1 statistic among all plots on Beaver Island.
Alleles Pgi-2a, Pgi-2b, Pgi-20 and Mdh-2a had significant overall correlograms. The
SND showed significant positive spatial autocorrelations in distance class one (0-3612
meters). In addition, negative spatial autocorrelation was detected in distance class 5
(9924 m-13394 m). The analysis was repeated without Pgi-2b, which was a relatively
low frequency allele. Similar results were found, there was significant positive spatial
autocorrelation at the first distance class excluding Pgi-2b and significant negative spatial

autocorrelation at the 5th distance class.

90

Table 43 Moran's I statistic for Beaver Island

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

No. Pairs 13 I 13 13 i 13 I 14 1
Distance 3612 6049 8056 9923 13394 ph Mean allele
class“ frequency
among
plots
PGM-1 0.14 0.04 -0.42 -0.22 0.00 0.27 0.848
Pol-2a 0.47 0.04 -o.05 -0.17 -0.70 0.01 0.162
PGI-2b 0.18 -0.35 0.23 0.02 -0.50 0.08 0.044
PGl-2c 0.51 0.03 0.01 -0.04 -0.90 0.00 0.789
MDH-2a -0.22 -0.10 -0.01 -0.50 0.35 0.08 0.667
MDH-2b -0.03 0.01 -0.07 -0.03 -0.33 0.61 0.211
MDH-2c 0.23 0.11 -0.28 -0.20 -0.30 0.44 0.111
AAT-2 -0.01 -0.21 0.13 -0.20 -0.16 0.74 0.667
Average 1 0.16 -0.05 -0.06 -0. 17 -0.32
Average 8.5. 0.23 0.23 0.22 0.22 0.20
SND" 3.05 0.73 0.44 -0.10 -3.77
p°< 0.00 0.77 0.67 0.46 0.00
SND“ 2.08 1.26 -0.08 -1.23 -2.61
p"< 0.00 0.11 0.47 0.10 0.01

 

 

 

 

 

 

 

 

a Upper bound for distance class in meters

 

b Overall correlogram significance

 

c inciudinngl-Zb

 

 

d excluding PGi-2b

 

91

 

Discussion

According to Hamrick and Godt, (1990) gymnosperms have a high number of
alleles per locus and high levels of genetic diversity. Hamrick and Godt, suggest this is
due to the relationship between genetic diversity and the characteristic of a species; long
lived, outcrossing, wind pollinated species have higher levels of allozyme variation
within populations and less variation among populations than species with other
combinations of traits. This study finds similar measures even among fragmented
populations (F st confidence interval among island 0.00 to 0.03; Rst among islands 0.04;
Nei’s genetic distance among islands 0.00 to 0.03; Nei’s genetic identity 0.98 to 1.0).

Furthermore, Hamrick and Godt ( 1990) suggest that the gene flow potential of a
species had the predominant influence on the partitioning of allozyme variation among
populations. Wide spread, long lived species with potential for long range gene
movement should have the highest levels of variation within populations. Again, this
study is consistent with this prediction.

It was hypothesized that there would be a loss of genetic diversity based on
information in the geological history of the area. The islands were most likely attached to
mainland Michigan around 10,000 y.b.p. The lake levels fluctuated greatly between eight
and six thousand years ago, but the levels would have maintained a height equal to or
greater than currently since 6,000 y.b.p. (Hansel, et al., 1985). The lake levels continued
to rise until around 4,000 y.b.p. during the Nipissing period, when lake levels were
estimated to have been about 7 meters higher than currently. After the Nipissing period,
the lake levels fluctuated but there was a trend toward lower lake levels until about 2,500

years ago, when the lake levels reached a height similar to today’s, which has been

92

maintained with minor fluctuations, less than 1m (Hansel, et al., 1985). The islands of
BIA have been fragmented for a relatively short time considering white pine has the
potential to be such a long lived species. Assuming it takes 20 years for a tree to reach
reproductive maturity, there have only been roughly 300 generations since isolation. For
a tree that can live centuries, this might not be along enough time for isolation and drift
to alter the genetic diversity.

The hypothesis was formed however, because white pine populations on Garden,
Hog and High Islands are found growing along the shores of Lake Michigan. White pine
is also quite commonly found growing along the Lake Michigan shore on Beaver Island.
White pine is only found growing in the sandy bays of the three smaller islands, and is
seldom found growing in rocky bays, or in upland areas. The high water level during the
Nipissing period would have greatly reduced the area available for white pine
reproduction on all islands, but the lower elevation above lake levels of Garden and Hog
islands should have resulted in a more extreme decrease in island size compared to
Beaver and High islands. The smaller islands size resulting from fluctuations lake levels
potentially reduced genetic diversity in a number of ways. First, there would be a loss of
genetic diversity if the populations had fallen below a size (in numbers) that could
support the diversity found before the population bottleneck. Second, if the populations
were large enough in number to support the original genetic diversity, the size in area
could have been small enough to lead to inbreeding, which would also reduce the genetic
diversity of the populations. The loss of diversity should have been especially high on
Hog and Garden Islands since both have a low elevation (less than 40 feet) above the

current lake level (Divins, 2002). Beaver and High Islands both have areas over 200 feet

93

above the current lake level, but large portions of Beaver and High Islands are less than
30 feet above current lake levels. In brief, the effects of the high water during the
Nipissing period should have caused a population bottleneck reducing the genetic
diversity of all organisms on the islands, even white pine which is long lived with
overlapping generations.

The results of the genetic diversity found within the BIA do not support the
hypothesis of a loss of genetic diversity. Both the allozyme and SSR markers displayed
high amounts of genetic diversity among the archipelago as well as relatively low but
significant levels of divergence. Similar studies have found high diversity within
populations, and low diversity among populations of conifers (Hiebert, and Hamrick,
1983; Hamrick and Godt, 1990; Vendrarnin et al., 1999; Thomas et al., 1999; Mariette, et
al., 2001; Ribeiro, et al., 2002 and others). High levels of genetic diversity within BIA
could result from several possibilities. First, perhaps the lake level estimates are wrong
and there was not a bottleneck. This is doubtful since the Lake Michigan basin has been
mapped (Divins, 2002) and basin topography agrees with the theory that the islands
would have been a part of mainland Michigan during low lake levels. F urtherrnore, there
are bluffs on the islands and throughout the Great Lakes region to support the high water
theory. Additionally, with isolation and fiagrnentation the remnant areas would decrease
in the number of species based on the theory of island biogeography. Garden and High
Islands supposedly have no American beech (F agus Americana) or eastern hemlock
(T suga Camdensis) which are both found in similar plant communities on Beaver Island,

suggesting that there was indeed a reduction in island size in the past.

94

Another possible cause for the high genetic diversity is that the populations of
white pine may have been larger or more widespread in the past than currently found on
BIA. If this is the case, it is possible there has been a loss of diversity, but relatively high
diversity still persists. The diversity measures for allozymes overlap those found in other
conifers (Hamrick and Godt, 1990). Moreover, the results in this study compare in
number of haplotypes with a continuous populations of white pine in Wisconsin (Craig
Echt. pers. comm.) Furthermore, allozyme allelic richness measured with the same
laboratory protocol is similar to that found in an old growth white pine forest on
mainland Michigan (Epperson and Chung, 2001). The similarity with Wisconsin and
mainland Michigan populations would suggest that these islands maintain genetic
diversity levels comparable with non-fragmented populations, not that they have high but
reduced genetic diversity relative to other conifers. One more possibility is that
migration has or is occurring to increase/maintain the genetic diversity in the populations.
Using microsatellites Ledig (1998) found that pollen mediated gene flow is effective in
preventing population differentiation. Prevention of differentiation could be
accomplished by gene flow among islands, which would also create high levels of genetic
diversity within populations.

Although the hypotheses of reduced genetic diversity was rejected, there are some
interesting results found in this study. One such result is the variation in allele and
haplotype frequencies among islands. The loci Tpi-2 was monomorphic only on Hog
Island. When looking at the different haplotype frequencies found among the islands,
the three most common haplotypes (numbers one two and three) are found on all islands.

However, some plots did not contain a single copy of one or two of the most common

95

haplotypes. High East West and Powerline had no haplotype number one. High East
Middle had no haplotype number two or number three. The Airport plot also had no
number three haplotype. The lack of the common haplotypes in some plots explains the
high F st values within Hog Island. With a bottleneck it is expected that the low
frequency alleles would be lost first since the probability of being lost is related to the
frequency of the alleles (Hatle and Clark, 1997). The fact that Tpi-2 was monomorphic
on one island, that the most common haplotypes are not found in all plots and rare
haplotypes are not found on all islands, suggests that there was a bottleneck and 1) that
occasional migration has occurred to reintroduce rare haplotypes 2) that mutations have
occurred to create new rare haplotypes 3) it is an artifact of sampling. Based on the
results of the spatial autocorrelation analysis (Tables 42 and 43), there does seem to be a
limit to migration, since distances greater than 8 km had no spatial autocorrelation, and
distances grater than 13.5 km had a negative spatial autocorrelation. However, the fact
that rare microsatellite markers are found scattered throughout the BIA suggests that
there has been low amounts of migration. It is unlikely that mutation is the cause of the
genetic diversity, since mutations should results in private haplotypes found among the
islands. One potential exception could be haplotype number 14. This haplotype had a
relatively high frequency on High Island, and was found in a single tree on Beaver Island.
It is possible this haplotype arose on High Island, and has since spread to Beaver Island.
If haplotype number 14 did arise on High Island and spread to Beaver Island, it would be
evidence for migration maintaining the genetic diversity within the archipelago. The fact
that there are low frequency alleles missing from some of the smaller islands and

common alleles missing from some plots could be a result of a past bottleneck.

96

Nei et al. (1975) suggested that heterozygosity could be maintained after a
population bottleneck but that the number of alleles would be reduced. If migration
occurs after the bottleneck this would increase the number of alleles in the population. It
is hypothesized that the effect of admixture via migration has resulted in the high levels
of genetic diversity. The genetic data analysis seems to be consistent with the geological
data, which suggests the islands were smaller in area during high lake levels resulting in a
population bottleneck for white pine, but the islands have since gotten larger and the
populations of white pine have grown in size (number) as the islands grew in area.
Wright’s F statistics among islands: The measure of F is and Fit suggest that some
inbreeding is occuning within the islands. This could be due to sampling trees that are
homozygotes for rare alleles. When low frequency alleles are found in trees as
homozygotes the F is and Fit values will be high.

The relatively large values for F is and Fst could be due to a number of factors
such as variable migration, establishment history or sampling error. High values for F is
and F st would suggest that there are low amounts of migration, while low values of Fis
and F st would suggest that there are high amounts of migration. The overall Fst values
suggest there is some divergence, since there was a small Fst value among islands. With
intennittent migration over distances, some plots may exchange pollen, and some may
not. Plots that have exchanged pollen will have low F is and F st values, while plots that
have not exchanged pollen over large distance would have high F is and F st values. The
local establishment history could also cause variable values to be attained for F st and Fis.

To explain, if a number of trees in a sample plot reach ecesis over a long period of

time, the possibility of long distance migration occurring at some point would be higher.

97

Consequently, there would be a low F is and F st values. If however, it was possible for
numerous individual trees to establish in a single year, then the possibility for siblings to
be grouped in an area would be high. If a sample plot contained many siblings then it
would appear that there is a high amount of inbreeding, and could also suggest limited
gene flow among plots. Finally, as with any study, there are limited resources available,
which is why sampling rather than a census was conducted. When sampling a population
there is always the possibility that the sample is not a true representative of the
populations.

N ei’s genetic distance: The measures found with Nei's genetic distance measures the
accumulated allele differences per locus between populations. Little differences were
found between islands. Low differentiation is most likely due to the effect of migration.
Nei’s genetic identity describes the proportion of alleles that are alike between
populations. This measure was high between populations suggesting little divergence,
which could be accomplished by high gene flow. Nei’s genetic distance measured
between plots did show some variability in the genetic distance. The highest value for
Nei's genetic distance was found with haplotype measures between Hog Middle East and
several other plots. High Middle East plot had several rare alleles, and had no trees with
the common haplotypes number one and two (Sample size in Hog East Middle is 14
trees). The sample plots with the exception of Garden Island are relatively small in
nrunber of trees, so it is not unexpected to find differences between areas with a low

number of samples.

98

Spatial autocorrelation: Moran’s I values found significant spatial autocorrelation
within 3 km. on Beaver Island, and within 5 km. among all islands. There was a negative
spatial autocorrelation at distances greater than 13.5 km. This would suggest that there is
a limit to the effect of pollen and seed migration. When populations are separated by
more than 13.5 kilometers genetic drifi, mutations and inbreeding could alter the genetic
diversity more than the effects of gene flow. Geographical separation with limited
migration should result in population differentiation and there is slight differentiation
between islands and a slight but significant differentiation among plots. When one looks
at the occurrence of the common and private alleles within and among islands, a low
migration and or mutation rate is suggested. It is possible that migration occurs enough
to prevent severe divergence, but not frequently enough to bring the frequencies of alleles

to a point where there is positive or no spatial autocorrelation at all distance class.

99

Conclusion
The genetic diversity measures are consistent with the geological history. It appears that
the islands were much smaller in the past, which resulted in a population bottleneck.
Since the bottleneck, the islands have grown in size, allowing for white pine populations
to increase in numbers and allow for admixture via pollen migration to increase the
genetic diversity within each island. The fact that the allozyme Tpi-2 was monomorpic on
one island, that rare haplotypes are missing from some islands, and that the most common
haplotypes are missing from some plots could be a result of genetic drift and infrequent
migration.

The islands as a whole maintain high levels of genetic diversity due to the effects
of mutation or migration. Mutation as the primary cause of the genetic diversity is
unlikely, since no private haplotypes were found on any of the smaller islands. Most
likely occasional long distance migration has occurred to maintain genetic diversity

within each population, and counter the effect of drift and inbreeding.

100

Appendix

The Diameter at Breat Heigt (D.B.H.) of Sampled trees

Garden D.B.H. D.B.H.
island Tree it cm Tree #

   

101

 

102

D.B.H.
Island Tree # cm

 

103

Beaver
island Tree #

Poweriine 200

French

Blue trail 225 37.40

 

104

D.B.H.
Tree # cm

292 35.60 CMU

 

105

D.B.H. D.B.H.
Treat! cm Tree# cm

Sloptown
Font Lake 380 32.80 Rd. 424 31.10

 

106

 

 

 

 

Appendix

Haplotype frequencies among plots continued

 

 

107

Appendix

Haplotype frequencies among plots continued

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

H

a E F ‘3

p O 1' e

' t e e

o n n

t B c ' w

r h s U

V I i

p d B B e C

9 g a a c M

S e y y k U

1 0.125 0.059 0.333 0.154 0.200
2 0.438 0.176 O. 167 0.154 0.133
3 0.063 0.412 0.056 0.385 0.333
4 0.000 0.000 0.167 0.000 0.000
5 0.188 0.059 0.000 0.000 0.200
6 0.000 0.118 0.056 0.000 0.000
7 0.000 0.000 0.000 0.000 0.000
8 0.000 0.059 0.056 0.000 0.000
9 0.063 0.059 0.167 0.154 0.067
10 0.063 0.000 0.000 0.000 0.000
1 1 0.063 0.000 0.000 0.000 0.000
12 0.000 0.059 0.000 0.000 0.000
13 0.000 0.000 0.000 0.077 0.000
14 0.000 0.000 0.000 0.000 0.000
1 5 0.000 0.000 0.000 0.077 0.067
16 0.000 0.000 0.000I 0.000 0.000
17 0.000 0.000 0.000] 0.000 0.000

= 13 15 16 10 14

 

108

 

Appendix

Haplotype frequencies among plots continued

 

Sta-I
3'0-I

:r-ocoar ~mmrn
nuoi :rc-I
3~COCD "mm!“
03—-—-rtb£012
--or--i arc—m

 

 

0.462 0.467 0.231 0.333 0.053

 

0.231 0.067 0.154 0.167 0.526

 

0.000 0.000 0.000 0.083 0.053

 

0.000 0.000 0.077 0.167 0.053

 

0.077 0.067 0.000 0.167 0.000

 

0.000 0.000 0.000 0.000 0.053

 

0.077 0.067 0.077 0.083 0.000

 

0.077 0.000 0.077 0.000 0.000

 

0.000 0.000 0.000 0.000 0.000

 

0.000 0.000 0.000 0.000 0.000

 

gigomqmmhunauduwsno-umz

0.000 0.000 0.000 0.000 0.000

 

 

 

 

 

13 0.000 0.000 0.000 0.000 0.000
14 0.077 0.200 0.000 0.000 0.000
15 0.000 0.000 0.077 0.000 0.000
16 0.000 0.000 0.000 0.000 0.000
17 0.000 0.000 0.000 0.000 0.000

 

 

 

 

 

 

 

13 15 14 11 15

 

109

 

Appendix

Haplotype frequencies among plots

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

H
H o H
G o g o
H a 9 9
a r E
d E a E
p e a s a
I n H s t s
O o t t
t i 9 M
s N i S
y I w o d o
p a e r d u
9 n - s t I t
s d t h e h
1 0.245 0.400 0.000 0.636 0.100
2 0.367 0.200 0.667 0.000 0.400
3 0.143 0.200 0.333 0.000 0.300
4 0.061 0.000 0.000 0.000 0.200
5 0.082 0.000 0.000 0.000 0.000
6 0.082 0.200 0.000 0.000 0.000
7 0.020 0.000 0.000 0.000 0.000
8 0.000 0.000 0.000 0.091 0.000
9 0.000 0.000 0.000 0.091 0.000
10 0.000 0.000 0.000 0.091 0.000
1 1 0.000 0.000 0.000 0.000 0.000
12 0.000 0.000 0.000 0.091 0.000
13 0.000 0.000 0.000 0.000 0.000
14 0.000 0.000 0.000 0.000 0.000
15 0.000 0.000 0.000 0.000 0.000
16 0.000 0.000 0.000 0.000 0.000
17 0.000 0.000 0.000 0.000 0.000
n= 49 10 6 14 10

 

 

 

 

 

 

110

 

Literature Cited

Beaulieu, J ., and J. Simon. 1994. Genetic structure and variability in Pinus strobus in
Quebec. Canadian Journal of Forest Research 24: 1726-1733.

Blake, J .G. 1991. Nested subsets and the distribution of birds on isolated woodlots.
Conservation biology 5:58-66.

Cheliak, W. M. AND Pitel, J. P. 1984. Technique for starch gel electrophoresis of
enzymes from forest tree species. Information Report PI-X-42. Petawawa
National Forestry Institute, Agriculture Canada, Canadian Forestry Service, pp. 1-
49. Chalk River, Ontario

Chiral, DD, 08. Owen, and J .P. Reganold. 2002. Natural Resource Conservation.
Pretice Hall NJ

Chung, M. G. AND Kang, S. S. 1994. Genetic variation and population structure in
Korean populations of Ewyajaponica (Theaceae). Am. J. Bot, 81, 1077-1082.

Cliff, A. D. and Ord, J .K. 1981. Spatial processes-methods and applications. Pion Ltd.
London.

Comps, B. D. Gomory, J. Letouzey, B. Thiebauf and R.J. Petie. Diverging trends
between heterozygosiy and allelic richness during post glacial colonization in the
European beech. Genetics 157:389-397.

Cooper DN, Krawczak M. ( 1993) Hum_an Gene MuLation. BIOS Scientific Publishers,
Oxford 402 pp.

Comuet, J .M. and G. Luikart. 2001. Description and power analysis of two tests for
detecting recent population bottlenecks from allele frequency data. Genetics
144:2001-2014

Davis, MB. 1981. Quaternary history and the stability of forest communities. In:
DC. West, H. H. Shugart and DB. Botkin, eds. Forest Succession: Concepts and
Applications. New York: Springer Verlag.

Daniel, G. and J. Sulivan. 1981. A Sierg Club Naturalist’s Guide to the North Woods.
Sierra club books. San Francisco 408 pp.

Divins D.L. 2002. Bathymetry of Lake Michigan CD—ROM on line sample
http://www.ngdc.noaa.gov/mgg/greatlakes/lakemich_cdrom/sample.htrnl

Echt C.S., P. May-Marquardt , M. Hseih, R. Zahorchak. 1996. Characterization of
Microsatellite markers in eastern white pine. Genome 39:1102-1108

lll

Echt C., and P. May-Marquardt. 1997. Survey of microsatellite DNA in pine. Genome
4019-17.

Epperson, B. and M. Chung. 2001. Spatial genetic structure of allozyme polymorphisms
within populations of Pinus strobus (Pinaceae). American Journal of Botany 88:
1006-1010

Frelich LE. and PB. Reich. 1995. Neighborhood effects, disturbance, and succession in
forests of the western great lakes region. Ecoscience. 2: 148-158.

Gillet, E. 1998. Diplogen: Qualitative inheritance analysis of zymograms and DNA
electropherograms in diploid individuals. Computer program and user’s manual
available at http://www.uni-forst.gwdg.de/fg/software.htm.

Goudet, J. 2001. F STAT, a program to estimate and test gene diversities and fixation
indices (version 2.9.3). Available from
http://www.unil.ch/izea/softwares/fstathtml. Updated from Goudet (1995)

Hamrick J .L. and M.J.W. Godt. 1990. Allozyme diversity in plant species. In Brown
A.H.D. Clegg, M.T. Kahler AL. and Weir, B.S. (eds) _P_l_ant Population Genetics.
Breeding and Genetic Resources. Sinauer Assoc. inc. Sunderland Mass.

Hancock, J .M. 1995. The contribution of slippage-like processes to genome evolution.
Journal of Molecular Evolution 41: 1038-1047.

Hale, WB and J .P. Margham. 1991. The Harper Collins Dictionary of Biology. Harper
Perennial. New York NY.

Hansel, A., D. Mickelson, A. Schneider, C. Larson. 1985 . Late Wisconsin and Holocene
history of the Lake Michigan basin. Geological Association of Canada Special
Paper 30

Hartl, D., and A. Clark. 1997. Principles of Population Genetics 3rd ed. Sinaur
associates inc. Sunderland MA.

Hartl D. L. and E. W. Jones. 1998. Genetics Principles and Analysis 4th ed. Jones and
Bartlett Publishers. Boston MA.

Hatt, R., J. Tyne, L. Stuart, C. Pope. A. Grobman. 1950. I_sl_and life: A Studv of the Lama
Vertebrates of the Islands of Eastern Lake Michgaar. Cranbook Institute of
Science Bulletin 27. ppl4.

Haufler, C. H. 1985. Enzyme variability and modes of evolution in Bommeria
(Pteridaceae). Syst. Bot, 10, 92-104.

112

 

Heame C.M., Ghosh S & Todd J .A. (1992). Microsatellites for linkage analysis of
genetic traits. Trends in Genetics, 8: 288-294.

Hiebert, RD. and J .L. Hamrick. 1983. Patterns and levels of genetic variation in great
basin bristlecone pine Pinus longaeva

Jacobson, G.L. The Palaeoecology of white pine (Pinus strobus) in Minnesota, Journal
of Ecology 67:697-726.

Ledig. F.T 1992. Human impacts on genetic diversity in forest ecosystems. Oikos
63: 87-108.

Ledi g F .T. 1998. Genetic Variation in Pinus. In Richardson D.M. (ed) Ecology and
Biogeogr_aphv of Pinus. Cambridge University Press Cambridge.

Lewis, PO. and D. Zaykin. 2001. Genetic Data Analysis. Computer program for the
analysis of allelic data. Ver. 1.0. (d16c) Free program distributed by the authors
over the Internet from http://lewiseeb.uconn.edu/leMshome/sofiware.html

Macdonale, G. M., L.C. Cwynar and C. Whitlock, 1998. Genetic Variation in Pinus. In
Richardson D.M. (ed) Ecology and Biogeoggaphy of Pinus. Cambridge
University Press Cambridge

Malecot G. 1948. Lemathematiques de l' heredite. Massorr, Paris
Mariette, S. D. Chagne, C. Pastuszka, A. Rafi'rn, C. Plomin and A. Kremer. 2001.

Genetic diversity within and among Pinus pinaster populations: comparison
between AF LO and microsatellite markers. Heredity 86:469-479.

Mitton, J. B., Linhart, Y. B., Sturgeon, K. B. and Hamrick, J. L. 1979. Allozyme
polymorphisms detected in mature needle tissue of ponderosa pine. Journal of
Heredity. 70: 86-89.

Morgante M & Olivieri AM. 1993. PCR-amplified microsatellites as markers in plant
genetics. The Plant Journal, 3: 175-182.

Mullis, K. and F. Falloona 1987. Specific synthesis of DNA in vitro via a polymerase
catalyzed chain reaction. Methods in Enzymology. 55:335-350.

Nei, M., T. Maruyama, and R. Chakrabority. 1975. The bottleneck effect and genetic
variability in populations. Evolution. 29: 1-10.

Nei, M. 1978 Estimation of average heterozygosity and genetic distance from a small
number of individuals. Genetics. 89:583-590.

113

 

Pielou, G. 1991. After the Ice Ag; The Return of Life to Glaciated North America. The
University of Chicago Press.

Powel, W. G.C. Machray, and J. Provan. 1996. Polymorphism revealed by simple
sequence repeats. Trends in Plant Science. 7:215-222.

Preston R.J. North American Trees. 4th ed Iowa state university press Ames 407 pp.

Queller, D.C., Strassmann J .E. & Hughes CR. 1993. Microsatellites and kinship. Trends
in Ecology and Evolution, 8: 285-288.

Rajora, O.P., M.H. Rahman, G.P. Buchert, and BF. Dancik. 2000. Microsatellite DNA
analysis of genetic effects of harvesting in old-growth eastern white pine (Pinus
strobus) in Ontario. Molecular Ecology 9: 339-348

Raven, P.H., R.F. Evert, and SE. Eichhom. 1992. Biology of Plants 5th ed Worth
Publishers, New York NY 791 pp.

Ribeiro M.M. S. Mariett, G.G. Vendramin, A.E. Szrnidt, C. Plomion and A. Kremer.
2002. Comparison of genetic diversity estimates within and among populations
of maritime pine using chloroplast simple-sequence repeat and amplified
fragment length polymorphism data. Molecular Ecology. 11:869-777.

Saiki, R.K., S.J. Scharf, F. Fallona, K.B. Mullis, G.T. Horn, H.A. Erlich and N.
Arrnheim. 1985. Enzymatic amplification of beta-globin sequences and

restriction site analysis for diagnosis of sickle cell anemia. Science. 230: 1350-
1354.

Saunders D.A., R.J. Hobbs and CR. Margules. 1991. Biological consequences of
ecosystem fragmentation: a review. Conservation Biology 5: 18-32.

Smith, W. and T. Hinckley, (eds). 1995. Ecophysiology of Coniferoujs Forests.
Academic press, New York. 338 pp.

Sokal R. and N. Oden. 1978. Spatial autocorrelation in biology. 1. Methodology.
Biological Journal of Linnean Society 10:199-249.

Sokal, R. 1979. Ecological parameters inferred from spatial correlograms, in G.P.Patil
and M. Rosenzweig (eds), Quantitative Ecology and Related Ecometrics.
International Cooperative Publishing House, Fairland, MD USA.

Soltis D., C. Haufler, D. Darrow, and G. Gastony. 1983. Starch gel electrophoresis of

ferns: a compilation of grinding buffers, gel and electrode buffers, and staining
schedules. American Fern Journal 73:9-27.

114

 

Swain, A. 1978. Environmental changes during the past 2000 years in North-Central
Wisconsin: analysis of pollen, charcoal and seeds from Varied Lake sediments.
Quaternary Research 10:55-68.

Thomas B.R., S.E. Macdonald, H. Hicks, D.L. Adams, and RB. Hodgetts. 1999. Effects
on reforestation methods on genetic diversity of lodgepole pine: and assessment

using microsatellite and randomly amplified polymorphic DNA markers.
Theoretical and Applied Genetics. 98:793-801.

Vendramin, G.G., Lelle, L.R., P. Rossi, and M. Morgante. 1996. A set of primers for
The amplification of 20 chloroplast microsatellites in Pinaceae. Molecular
Ecology. 5:595-598.

Vendramin, G.G., B. Degen, R.J. Petit, M. Anzidei, A. Madaghiele and B. Ziegenhagen.
1999. High level of variation at Abies alba chloroplast microsatellite loci in
Europe. Molecular Ecology. 8:1117-1126.

Walter R. and BK. Epperson. 2001. Geographic pattern of genetic variation in Pinus

resinosa:Area of greatest diversity is not the origin of postglacial populations.
Molecular Ecology 10: 103-1 1 l

Weber J .L. and PE. Wong. 1993. Mutation of human short tandem repeats. Human
Molecular Genetics. 2:1 123-1 128.

Whitney G. 1986. Relation of Michigan’s presettlement pine forests to substrate and
disturbance history. Ecology 67: 1548-1559.

Weir, B. 1996. Genetic Data Analysis 2nd ed. Sinauer associates inc. 422 pp.

Weeden, N. F. AND Wendel, J. F. 1989. Genetics of plant isozymes. In: Soltis, D. E. and
Soltis, P. S. (eds) Isozvmes in Plant Biology, pp. 46-72. Dioscorides Press,
Portland, OR.

Wendel, J. F. and Weeden, N. F . 1989. Visualization and interpretation of plant
isozymes. In: Soltis, D. E. and Soltis, P. S. (eds.) Isozvmes in Plant Biology, pp.
5-45. Dioscorides Press, Portland, OR.

Wright. S. 1943. Isolation by distance. Genetics 281114-138

Wright, S. 1946. Isolation by distance under diverse systems of mating. Genetics 31:
39-59

Wright, S. 1965. The interpretation of population structure by F-statistics with special
regard to systems of mating. Evolution 19:395-420.

115

 

Yeh, F .C. and Boyle, T.J.B. 1997. Population genetic analysis of co-dominant and
dominant markers and quantitative traits. Belgian Journal of Botany 129: 157.

116