.1. hflu h"...
4.: «kW

:x

fin}.

.:

3?
I.

t

m. z? )
«m3».
. D n.
, .VA him
4:. «it: . V
. . 3 £§#.§
, .9. . . Vituzstflw‘r-Wug
. . 5th.? 3....
.33 . . V .. . . 3.5.? H, 2
‘ A I
u . V . . n I. 3w
it... nil V V . . V V ,
shrughfia .. . . . , , .
I . A... V . V
x 5.1: n». V V
«kémnwpu . 5mg
..

Qua!

. all V
3419sz)...
vamfifiy

.2... .2
.d: .14...
{(1.539

... ._ ..
r. 3.. .
3. 3w
3.5...

 

 

 

 

 

 

V II!
n .ztn‘JV V

A. 34.»;
flu-mt.- 3
£51»!
..
E...

 

1 DO , LIBRARY
‘ ‘° Michigan State
University

This is to certify that the
thesis entitled

IMPACTS OF HERBIVORES AND PLANT COMMUNITIES ON
ESTABLISHMENT AND SPREAD OF ALLIAHIA PET IOLATA
(GARLIC MUSTARD) IN MICHIGAN

presented by

Jeffrey Adam Evans

has been accepted towards fulfillment
of the requirements for the

Master of Scien de ea in

 

Entomolryy
0;; A

MKjor Prof ssor' Sign e
2/924! 09

 

Date

MSU is an Affinnative Action/Equal Opportunity Institution

 

 

SUPP! WWTARY

 

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

 

memes

 

 

'1»!

:5"
[‘3
h»
r
h:

 

 

 

 

 

 

 

 

 

 

 

 

 

2/05 p:IC|RC/DateDue.indd-p.1

IMPACTS OF HERBIVORES AND PLANT COMMUNITIES ON
ESTABLISHMENT AND SPREAD OF ALLIARIA PET [OLA TA (GARLIC MUSTARD)
IN MICHIGAN

By

Jeffrey Adam Evans

A THESIS
Submitted to
Michigan State University

in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE
Department of Entomology

2006

ABSTRACT

IMPACTS OF HERBIVORES AND PLANT COMMUNITIES ON
ESTABLISHMENT AND SPREAD OF ALLIARIA PETIOLA TA
(GARLIC MUSTARD) IN MICHIGAN

By
J effiey Adam Evans

Alliaria petiolata (garlic mustard) (M. Bieb) Cavara and Grande) is shade-tolerant
biennial forb of European origin that is now widespread and invasive in North American
forest communities. As conventional control efforts against A. petiolata have failed,
research efforts are now focused on developing classical biological controls. The goal of
my research was to explore and interpret the biology-and invasion ecology of A. petiolata
in Michigan in support of current and future control efforts. Alliaria petiolata populations
were shown to be expanding at all sites studied with 59% of initially uninvaded sampling
quadrats becoming invaded after two years. 84.5% of the quadrats with A. petiolata
Showed evidence of herbivore browsing or other damage. However, damage estimates
were very low (2.9% of leaf area damaged) and were not correlated with A. petiolata
survival or fecundity. Alliaria petiolata may negatively impact native species where it
becomes dominant as it has at one site where it represented 57% of the total vegetation.
Native and exotic species richness of the sites were positively correlated (Spearman r, =
0.9729), but the relationship of quadrat-scale native species richness to A. petiolata
presence or absence reversed across a gradient from species-rich to species-poor Sites (P
= 0.0087). This suggests that neither biotic resistance nor enhancement alone sufficiently
explains A. petiolata’s invasion processes. Rather, A. petiolata appears to make tradeoffs

between its competitive ability and stress tolerance under different conditions.

Cepyright by
JEFFREY ADAM EVANS
2006

For Courtney

ACKNOWLEDGEMENTS:

I would like to thank Dr. Doug Landis for his insightful guidance, patience, and
constant support and encouragement over the last three years. This work would never
have been completed without him. I also wish to thank Dr. Doug Schemske and Dr. Deb
McCullough for their assistance and advice throughout this project.

This research would not have been possible without the generous support and
assistance of a great many people. For Sharing with me the overwhelming joy of counting
thousands of tiny plants in the rain, snow, and blistering heat that Michiganders refer to
casually as “weather” I owe my thanks to Doug Landis, Anna Fiedler, Mary Gardiner,
Alejandro Costamagna, Chris Sebolt, Bob MacDonald, Matthew Wood, Tara Lehman,
Chuck Stahlmann, Alissa Berro, Nik Vuljaj, Rachael Olson, Kevin Newhouse, Steve
Gottschalk, Erin Miller, Jacob McCarthy, John Davenport, Jessica Steffen, Ryan
Alderson, Mike Wayo, Dawn Richards, Charlie Richards, Emily Knoblock, and Courtney
Jones.

Rob Tempelman, Xuewen Huang, Chad Brassil, Brian McGill, Adam Davis,
Alejandro Costamagna, Kay Gross, Doug Schemske, Deb McCullough, and Doug Landis
provided frequent (continual) experimental, analytical, and statistical consultation, and
Jim Miller offered additional input to the analyses in Chapter 3. Plant identification was
possible with help fi'om Ed Voss and Tony Reznicek at the University of Michigan, Alan
Prather, Rachael Williams, Deborah Trock and Steve Stephenson at the Michigan State
University Herbarium, George Yatskievych at the Missouri Botanical Garden, Harvey

Ballard, Ellery Troyer and Bill Schneider. When complex electronics failed I received

iv

help fiom Lee Duynslager, Chuck Stahlmann and Shaun Langley. Thanks to Karen
Renner for use of her SeedBuro seed counter.

Access to research sites was graciously provided and made possible by Michigan
State University, Heidi Gray and the staff of the Femwood Botanic Garden, the Michigan
Department of Military and Veteran’s Affairs, Doug Pearsall and The Nature
Conservancy’s Michigan Chapter, Steve Norris at Lux Arbor, Earl Flegler and Ray
Fahlsing at the Michigan Department of Natural Resources, Greg Kowaleski at Russ
Forest, Dan DeLooff at Kent County Parks, Chip Franke at Ottawa County Parks, The
Shiawassee YMCA and their staff, Mike McCuistion and the staff of the Edward Lowe
Foundation, and the Gasinski Family.

This work was made possible with financial support from the Michigan State
University Plant Sciences Fellowship, the Michigan Department of Natural Resources,
the Michigan Department of Military and Veterans Affairs, and the United States EPA
STAR graduate fellowship. Extra special thanks go to Angela J emstad, Jill Kolp, Heather
Lenartson, and Linda Gallagher for making the Entomology Department and the
entomologists at MSU function every day.

For their perennial outstanding support from my first breath to this moment,
which I’m sure they can hardly believe has come, I offer my thanks and love to my
parents John and Carolyn Evans, and to my sister Jocelyn. I wouldn’t be here without
them.

Nearly as importantly, I’d never have said “my dog ate it” so often if it wasn’t
true. Thanks to Toby and Myelin for eating my telephones, various important computer

disks and early drafts of my thesis (unfavorable reviewers?), for generously keeping my

side of the bed warm when I was working late, and for continually fouling the air with
their unbridled, enthusiastic four-legged lust for life.

And finally, although empirical studies Show that She falls asleep in the sun
exactly 50% of the times she goes to the field to “help with data collection”, no one
deserves my love, respect, and thanks more than Courtney Jones for her boundless love
and support of all manners.

Thank you all,
Jeff Evans

East Lansing, February 23, 2006

TABLE OF CONTENTS

 

 

 

 

 

LIST OF TABLES - V ................ -- V- - -V---XII
LIST OF FIGURES . - V - ........ - XIV
BIOLOGICAL INVASION BY ALLIARIA PETIOIA TA (BIEB.) CAVARA AND
GRANDE: HISTORY, ECOLOGY, AND MANAGEMENT PROSPECTS .............. 1
Abstract ...... - -- - - V - - - - -- V - .......... 2
Introduction _ - V _ - - - - ........ V .......... 4
Study Species - - ...... V - .......... _ 5
Nomenclature and Anthropogenic Uses ...................................................................... 5
Alliaria petiolata Life Cycle ....................................................................................... 5
First Year Plants ...................................................................................................... 6
Seed and Germination Biology ........................................................................... 6
Seedlings and Rosettes ........................................................................................ 7
Overwinten'ng ..................................................................................................... 8
Second Year Plants .................................................................................................. 8
Growth and Description ...................................................................................... 8
Flowering and Pollination ................................................................................... 9
Seed Production ................................................................................................ 10
Response to Light Conditions ................................................................................... 11
Habitat ....................................................................................................................... 13
Range of Alliarz’a petiolata ........................................................................................ 14
Native Range ......................................................................................................... 14
Introduced Range .................................................................................................. 14
Genetic Variation ...................................................................................................... 15
Invasiveness of Alliaria petiolata ............................................................................. 16
Better Living through Chemistry .......................................................................... 18

vii

Allelopathy: Interactions with Plants ................................................................ 18

 

 

 

Defensive Chemistry: Interactions with Insects ................................................ 21
Interactions with White-Tailed Deer ..................................................................... 22
Differences between European and North American Alliaria petiolata populations
............................................................................................................................... 23

Management of Invasive Garlic Mustard 24
Conventional Controls ............................................................................................... 24
Prescribed Fire ....................................................................................................... 25
Chemical Control .................................................................................................. 26
Cutting ................................................................................................................... 27
Pulling ................................................................................................................... 28
Mowing ................................................................................................................. 28
Other ...................................................................................................................... 29
Biological Control ..................................................................................................... 29
Potential Agent: Phyllotreta ochripes ................................................................... 30
Potential Agents: Ceutorhynchus alliariae and C. roberti .................................... 31
Potential Agent: Ceutorhynchus constrictus ......................................................... 32
Potential Agent: Ceutorhynchus theonae .............................................................. 33
Potential Agent: Ceutorhynchus scrobr'collis ........................................................ 33
Interactions Between Potential Biocontrol Agents ............................................... 35
Biological Control Summary and Outlook ........................................................... 36
Predicting Biocontrol Requirements and Outcomes ............................................. 38
Research Goal and Objectives 39
Literature Cited 41

viii

PRE-RELEASE MONITORING OF ALLIARIA PETIOLA TA [GARLIC
MUSTARD (M. BIEB) CAVARA AND GRANDE] INVASION S AND IMPACTS
OF EXT ANT NATURAL ENEMIES IN SOUTHERN MICHIGAN FORESTS ....52

 

 

 

 

Abstract 53
Introduction 54
Study Species ............................................................................................................ 55
Objectives .................................................................................................................. 56
Methods and Materials 57
Site Selection ............................................................................................................. 57
Alliaria petiolata evaluations .................................................................................... 58
Site Descriptions ....................................................................................................... 59
Study Sites ............................................................................................................. 60
Analyses of Data ....................................................................................................... 65
Spread of Alliaria petiolata ................................................................................... 65
Estimation of Alliaria petiolata Fecundity ............................................................ 66
Calculation of Survival Probabilities .................................................................... 66
Sampling Error ...................................................................................................... 67
Herbivore Impacts on Alliaria petiolata ............................................................... 68
Results and Discussion 69
Expansion of Alliaria petiolata within sites .............................................................. 69
Estimation of Alliaria petiolata F ecundity ................................................................ 71
Alliaria petiolata Damage by herbivores .................................................................. 71
Spring damage ....................................................................................................... 71

Fall damage ........................................................................................................... 72
Extensively Damaged Quadrats ............................................................................ 73
Impacts of Damage ................................................................................................ 74

ix

Conclusions 75

 

 

 

 

 

 

Literature Cited 77
DYNAMICS OF ALLLARLA PE TIOLA TA (GARLIC MUSTARD) IN VASIONS IN
SOUTHERN MICHIGAN FORESTS 100

Abstract 101

Introduction 103

Biotic Resistance Hypotheses ................................................................................. 103
Biotic Enhancement Hypotheses (The Rich get richer) .......................................... 104
Interpreting Impacts of Invasions ............................................................................ 106
Study Species .......................................................................................................... 107
Community Response to Invasion and Control Efforts .......................................... 108
Objectives ................................................................................................................ 109
Methods and Materials ' - 109
Site Selection: .......................................................................................................... 109
Sampling Methods: ................................................................................................. 1 10
Analyses of Data ..................................................................................................... 111
Site Overviews and Characterization ............................................... 111
Effects of Deer .................................................................................................... 112
Metrics of Diversity ............................................................................................ 1 12
Simpson’s Diversity Index (-lnD) ................................................................... 112
Berger-Parker Index of Dominance (d) ........................................................... 1 13

F loristic Quality Assessment ........................................................................... 1 l3
Site-Level Diversity Comparisons ...................................................................... 114
Native vs. Exotic Species Richness ................................................................ 114
Differences Between Sites .............................................................................. 115
Alliaria petiolata Impacts and Invasion Processes ............................................. 116

Interaction of Site Species Richness and A. petiolata Presence ...................... 116

 

 

 

 

Quadrat-Scale Correlations ............................................................................. 1 16
Linking Invasion Processes across Scales ....................................................... 117
Predicting Invasion Probability ....................................................................... 117
Results and Discussion 119
Site Overviews ........................................................................................................ 120
Site Characterizations .............................................................................................. 120
Site-Level Diversity and Impacts ............................................................................ 122
Effects of Deer .................................................................................................... 122
Native vs. Exotic Species Richness ..................................................................... 123
Differences Between Sites ...................................................................................... 125
Alliaria petiolata Invasion Processes ...................................................................... 126
Effect of Site on A. petiolata Presence and Native Species Richness ................. 126
Quadrat Level Correlations ................................................................................. 126
Linking Invasion Patterns Across Scales ............................................................ 127
Predicting Invasion Probability ........................................................................... 129
Logistic Regression Analysis .......................................................................... 129
Patterns of Invasion and Possible Mechanisms ....................................................... 131
Summary and Conclusions 133
Potential Impacts ..................................................................................................... 134
Alliaria petiolata and Theories of Invasion Processes ............................................ 134
Literature Cited 137
APPENDIX 1 155
APPENDIX 2 164

 

xi

LIST OF TABLES
Table 1. Michigan sites selected for long-term monitoring of garlic mustard. . . . .

Table 2. Summary of data by site and sampling period 2*: 1 SE where applicable.
Percent cover data were estimated separately for seedling and adult A. petiolata
plants as well as overall in spring. Total A. petiolata cover in fall represents rosette
cover because no seedlings or adults are present at that time. Estimates of mean A.
petiolata height, fecundity and presence and extent of damage to A. petiolata plants
is calculated from quadrats that contained A. petiolata plants, although A. petiolata
percent cover and counts represent site means in both invaded and uninvaded
quadrats. During the six sampling periods from 2003 — 2005 there were a total of
960 quadrat observations fiom 20 quadrats at each of eight sites ........................

Table 3. Number of sampling quadrats at each site where live A. petiolata was
observed during each spring sampling period and percent change over time. Mean
values are i 1 SE. Means for percent change and relative percent change do not
include data from Femwood which was fully invaded during all years .................

Table 4. Univariate repeated measures ANOVA of the number of quadrats invaded
by A. petiolata over time within sites. Test excludes the Femwood site where initial
conditions prevented estimation ofspread..

Table 5. Dunn-Sidak adjusted comparisons of the number of invaded quadrats per
Site between years ..............................................................................

Table 6. Michigan sites selected for long-term monitoring of A. petiolata invasion,
county, GPS coordinates, identities and diameter at breast height (DBH)
measurements of principal canopy trees, and Michigan Natural Features Inventory
(2003) community types. Deer density estimates are county wide and may not
reflect local population sizes within the study sites (MDNR 2006) .....................

Table 7. Results of Floristic Quality Assessment of site inventories (all years) and
diversity indices for spring of 2005 data .....................................................

Table 8. Results of Kolmogorov-Smirnov two sample tests for differences in mean
(above) and summed (below) relative abundance distributions of species data.
Years 03, 04, and 05 Show results from 2003, 2004, and 2005 respectively. Results
above the diagonal are from fall data, and those below the diagonal are from Spring
data. Significant Dunn-Sidak adjusted P-values are Shown as: P _<_ 0.0018 = *d.
Unadjusted P-values are shown as: P S 0.05 = *, P S 0.01 = "”" .........................

Table 9. Two-way factorial GLM of quadrat native species richness by A. petiolata
presence or absence and Site as main effects ................................................

Table 10. Frequency of invasion events in 2004 in plots that were uninvaded in
2003 ...............................................................................................

xii

81

82

89

90

91

143

144

145

I46

I47

Table 11. Logistic regression analysis of invasion probability by native species
richness in 57 sampling quadrats that were not invaded by A. petiolata in spring
22()()23 ............................................................................................... .Ilié?

xiii

LIST OF FIGURES

Figure 1. Results of an 181 Web of Science query using the search terms "invasive
species OR biotic invasion OR biological invasion". Search performed on January
24, 2006 ..........................................................................................

Figure 2. Approximate known distribution of A. petiolata within Michigan's lower
peninsula (Shaded counties) (V 035 1985, J. Evans and D. Landis pers. obs ) and
locations of study sites (stars) .................................................................

Figure 3. Mean number of sampling quadrats per site (II = 20) where live A.
petiolata plants were observed. Femwood data are not included because all
quadrats were invaded there during all three years .......................................

Figure 4. The number of sampling quadrats containing live A. petiolata during
sampling has increased since 2003 at most sites during spring and fall ................

Figure 5. Mean percent A. petiolata cover per quadrat from all quadrats during
each sampling period. Solid line shows quadrats that were invaded in 2003, dotted
line Shows quadrats that were not invaded in 2003 but became invaded in 2004, and
the dashed line shows quadrats that became invaded In 2005. Cover estimates are
from spring only .................................................................................

Figure 6. Regression used to estimate fecundity of A. petiolata plants. Plants used
in this analysis were collected at multiple locations across southern Michigan ........

Figure 7. Frequency distribution of damage to A. petiolata foliage during each
sampling season. Damage was frequent but was rarely extensive ........................

Figure 8. The relationship between overwintering survival and damage to A.
petiolata, year, Site, and species richness is not significant. When the splitting
criteria are relaxed and site is dropped as a predictor, overwintering survival in
quadrats with holes-damage to A. petiolata was insignificantly higher than in those
without such damage. (PRE for overall model = 0.0852). Within quadrats that did
not have holes-damage to A. petiolata plants in fall, overwintering survival was
insignificantly higher in the winter of 2004 — 2005 than 2003 — 2004. Dashed lines
in figure indicate statistically insignificant splits ...........................................

xiv

51

92

93

94

95

96

97

98

Figure 9. When site is not included as a predictor of fecundity, the 24 quadrats in

which A. petiolata sustained greater than or equal to 2% damage to leaves had
insignificantly higher seedling survival than those with less than 2% damage.

Overall PRE fit = 0.0557. Dashed line indicates statistically insignificant split ....... 99

Figure 10. Correlation of site native species richness from 2003-2005 data and
county-wide deer-density. Spearman rank correlation coefficient and probability
are shown ........................................................................................ 149

Figure 11. Spearman rank correlation of exotic versus native species richness at

the site level. The number of exotic species observed at Lux Arbor (LA) increased
disproportionately following extensive soil and canopy disturbance during logging
activities in spring 2005. I excluded Lux Arbor from this analysis for this reason. . .. 150

Figure 12. Native species richness (mean 3: SE) in quadrats with and without A.

petiolata versus the total number of native species observed per site on the

sampling date. Differences between years in site Species richness reflect improved
sampling ability, but overall patterns are consistent across years. The reduced

variance in uninvaded quadrats over time at some sites was casued by A. petiolata’s
spread and the decreasing number of ininvaded quadrats ................................. 151

Figure 13. Quadrat native species richness versus A. petiolata relative abundance.
P-values S 0.1 from Spearman rank correlations are shown. Least square regression
lines are overlaid to indicate the linear trend from each correlation ...................... 152

Figure 14. Relationship of correlation coefficients (from figure 15) to site native
species richness. Spearman rank correlation coefficients are shown. Least square
regression lines are overlaid to indicate the linear trend from each correlation .......... 153

Figure 15. Predicted and observed outcomes of invasion in 2004 of quadrats that
were A. petiolata free in 2003. Lines represent predicted invasion probabilities and
95% confidence intervals. Quadrats that were invaded in 2004 are Show at the top
of the flame, while those that were not invaded are at the bottom. Observed data

points are jittered to reveal overlapping points ............................................. 154

XV

With the invasion of animals and plants. . ., it is the successful species that are
concerned. But there are enormously more invasions that never happen, or fail quite soon
or even after a good many years. They meet with resistance. It is this resistance, whether
by man or by nature or by man mobilizing nature in his support, that has now to be
examined: what it is and how it can be understood and when necessary manipulated and

increased when desired.

Charles S. Elton

1958

xvi

We can outcompete with anybody.

George W. Bush

Bay Shore, New York, Mar. 11, 200

xvii

Chapter 1

BIOLOGICAL INVASION BY ALLIARIA PE T [OLA TA (BIEB.) CAVARA AND
GRANDE: HISTORY, ECOLOGY, AND MANAGEMENT PROSPECTS

Jeffrey Adam Evans

ABSTRACT

Alliaria petiolata (garlic mustard) (M. Bieb) Cavara and Grande) is a shade-
tolerant biennial forb of European origin that is now widespread and invasive in forest
communities from the east coast of North America to Alaska. Alliaria petiolata’s owes its
success as an invader in part to its high fecundity and plastic response to changing
environmental conditions. Plants can grow to maturity under light conditions ranging
from the shade of closed forest canopies to near full-sun conditions. Flowers are typically
insect pollinated but are facultatively selfing if no pollinators visit them. Individual plants
can produce thousands of seeds, and fecundity is reduced only by 46% when 100% of
leaves are removed. Cutting A. petiolata stems effectively kills second year plants, but if
plants are out after A. petiolata has flowered seeds still reach maturity on the cut stalks.
Seed longevity in the soil can be at least eight years, implying that any successful control
efforts will need to be sustained on at least a decade scale until the seed bank is
exhausted. Although seedling and rosette survival rates are low at high densities, they are
substantially higher where populations are sparse. Thus, A. petiolata is an effective
colonizer of new areas. Allelochernicals produced in A. petiolata’s roots and leaves
effectively suppress germination of competitors’ seeds and prevent germination of
arbuscular mycorrhizal spores, thereby disrupting important mutualisms upon which
many later-successional native species are dependent. Other defensive chemistry deters
specialized herbivore feeding in A. petiolata’s introduced range, although a broad
community of specialists and generalists feed on it in Europe. Conventional control
efforts including cutting, pulling, prescribed fire, and herbicides have proven ineffective

against all but the smallest infestations. Research efforts are now focused on developing

classical biological controls for potential future releases, with a group of four weevils
(Coleoptera: Curculionidae) in the genus Ceutorhynchus now appearing the most
promising. Alliaria petiolata’s success as an invader results from its combined
demographic, phytochernical, and physiological properties, and effective future control
strategies will need to successfully address each of these. The goal of my research is to
explore and interpret the biology and invasion ecology of A. petiolata in Michigan in
support of current and future efforts to control its spread. Specific research objectives are

addressed in subsequent chapters.

 

INTRODUCTION

Biological invasions are a leading threat to global biodiversity (W ilcove et a1.
1998). As many as 50,000 non-indigenous species have been introduced to the United
States alone (Pirnentel et al. 2000), but globalization of economies, increased human
travel, and the task of regulating new introductions have torn down longstanding
geographical barriers to species dispersal and place this number within the realm of
possibility. Of these introductions, approximately 5000 plant species have become
established in natural systems and now comprise nearly one third of all wild flora
(Pimentel et al. 2000 and sources therein). The full scope of this phenomenon and its
potential consequences for human well being and natural systems alike are not firlly
understood. Increases in efforts to understand and address biological invasions (Figure 1)
have led to improved management approaches to invasions. Continued introductions and
developing fields within the sciences have highlighted the need to address invasions
across scales from molecular to global and with increasing analytical complexity. While
some invasions may pose substantial threats to ecosystems and resources required by
humans, they also present a rich arena in which to explore the dynamic processes of
community assemblage and structure, trophic interactions, and changes in each of these
across spatial and temporal gradients. I have reviewed the growing body of literature
related to the introduction and invasion of Alliaria petiolata (Bieb) Cavara and Grande
(Brassicaceae). From this I have attempted to interpret the dynamics and characteristics
that drive and limit its invasion process locally and evaluate its potential impacts on

native communities.

STUDY SPECIES
NOMENCLATURE AND ANTHROPOGENIC USES

Alliaria petiolata (Bieb) Cavara and Grande (Brassicaceae), is a globally
distributed, shade tolerant, cool-weather obligate biennial that is a frequent component of
forest understory and edge communities. Also know as garlic root, garlicwort, hedge-
garlic, jack-by-the-hedge, jack-in-the-bush, mustard root, poor-man's-mustard, and sauce-
alone, the most commonly used name in North America is “garlic mustard” (Nuzzo
1993b, Society for Economic Botany 1998). Most common names refer to the distinctive
garlic or onion odor of crushed leaves, its historic use as an edible green and cooking
herb, and its common distribution in edge habitats. Alliarz‘a petiolata is also used in
traditional remedies for numerous ailments. These include use as an expectorant, a
digestive aid, a subdofic (a medicine that causes sweating) and for treatment of wind
colic (pain or obstruction of the colon). Alliaria petiolata has higher vitamin C content by
weight than oranges and higher vitamin A content than spinach (Cavers et a1. 1979).
Older scientific names include Alliarz‘a alboi Sennen, Alliaria alliacea Britt. et Rendle,
Alliaria alliaria Huth, Alliaria matthioli Rupr.,Arabis petiolata M.B., Crucifera alliaria
E.H.L. Krause, Erysimum alliaceum Salisb., Erysimum alliarz'a L., Hesperis alliaria
Lam., Sisymbrium alliaria Scop., Sisymbrium oflicinalis DC, and in North America,

Alliaria ofi‘icinalis (Nuzzo 1993b, Hinz and Gerber 1998, Kerguélen 1999).

ALLIARIA PE NOLA TA LIFE CYCLE
Alliaria petiolata is an obligate biennial or winter annual in the Brassicaceae

(Cavers et al. 1979) that occurs primarily in temperate forest understory and edge

communities. First and second year plants are typically not intermixed within localized
patches, creating an effective alternation of generations in patches annually in newly
established populations. Over time, the seed bank moderates this effect somewhat and
first and second year plants are sometimes found in mixed patches. Strong suppression of
seedlings by adults keep the generations segregated locally in many areas, though

(W interer et a1. 2005). Seedlings that germinate under cover of second year plants have
very high mortality. In areas where it is invasive, A. petiolata spreads as a moving fi'ont
as satellite populations ahead of the core establish and fill out. This fi'ont typically
advances and retreats during alternate years advancing as much as 36 m and retreating by
up to 18 m during different years because of its biennial lifecycle. The net rate of spread

is usually positive, with an average advance of 5.4 m/y (Nuzzo 1999).

First Year Plants
Seed and Germination Biology

Seeds of A. petiolata require cold stratification prior to germination. Baskin and
Baskin (1992) studied the germination biology of A. petiolata and found that germination
rates are highest under natural, fluctuating winter temperatures ranging between 0.5 and
1 0°C. They report germination rates as high as 96 - 100% for buried seeds in an unheated
greenhouse. Germination peaks when mean daily temperatures range between —1.0 and
8.1°C. This finding is significant as seedlings of A. petiolata emerge at the end of winter
or in very early spring while the forest canopy is open and before other forbs germinate
and begin growing. Consequently, A. petiolata germinates under high light conditions

with little competition fi'om other plants. AIliaria petiolata seedlings grow rapidly and

form a low, tight canopy over the forest floor with densities as high as 20,000/m2
(Trimbur 1973).

In Great Britain, most seeds germinate after one season of cold stratification
(Roberts and Boddrell 1983). Seeds in Ontario, Canada, remain dormant for one to two
years before germinating with only 5 - 9% of seeds produced in a given year emerging to
form seedlings with only 2 - 4% of seedlings survive to flower (Cavers et al. 1979). Seeds
can remain viable for up to five years after production with viability in Great Britain
dropping off sharply after the first year to 1.4 - 24.1% in the second year and 0.1 - 1.5%
in years three, four and five (Roberts and Boddrell 1983). With seed production as high
as 107,580/m2 (Cavers et al. 1979), even this low rate of germination after several years’
dormancy could result in a substantial emergence of seedlings. ‘

Seed mass in A. petiolata varies between populations up to eightfold, within
populations fiom 2.5 - 7.5 fold, 2 - 3 fold within individual plants, and 1.4 - 1.8 fold
within individual fruits(Susko and Lovett-Doust 2000a). Variation in seed mass is
implicated in the timing of seedling emergence with smaller seeds germinating earlier
than larger ones. This may be a combined effect of the thinner seed coat and higher
surface to volume ratios of smaller seeds, both of which allow increased water absorption
and thereby break seed dormancy earlier. Earlier emerging seedlings produce longer
hypocotyls and grow taller than those fiom larger seeds, allowing them to partially shade

seedlings that emerge later. Thus, small seed size in A. petiolata may be advantageous.

Seedlings and Rosettes

Emerging seedlings present cotyledons with blades averaging 6 mm long on

petioles of equal length, and the hypocotyl averages 2 cm long. The first leaves are 1 - 5

cm diameter and roughly toothed on pubescent petioles. First year plants form low
rosettes of dark purple to green, kidney shaped leaves with scalloped edges 2 - 12 cm in
diameter on 3 - 10 cm long pubescent petioles (Cavers et a1. 1979). The early season
leaves produce a distinct garlic or onion odor when bruised, although its intensity
decreases as the leaves age during the summer. Mortality is highest (>50%) during the
seedling stage. Only 2 - 4% of seedlings that emerge will live to flower (Cavers et a1.

1979)

Overwintering

First year A. petiolata plants overwinter as green rosettes and are
photosynthetically active on days when the temperature is aboVe fi'eezing. Combined
seedling and winter mortality is high. Nuzzo (1993a, 1993b, 1993c, 1996) Showed that
approximately 20% of rosettes survive the winter. Rosette densities in the spring average
30-80/m2, although they can be as high as 450/m2. About 9% of the variance between fall
and spring rosette densities can be attributed to fall rosette density. A study of the
variation in demographic rates of A. petiolata populations across southern Michigan is

ongoing (Davis et al. 2005).

Second Year Plants
Growth and Description

Stern elongation in second year plants begins under high light conditions during
spring before canopy trees leaf out. Growth is rapid during bolting, with plants in Illinois

reported to increase in height an average of 1.9 cm/day (Anderson et a1. 1996). Basal

leaves of bolting and mature second year plants are the same as those of first year
rosettes. Cauline leaves are 3-8 cm long on pubescent petioles and become gradually
smaller towards the stem apex. Leaves are deltoid and roughly toothed. The stem is erect
and typically glabrous or with few hairs and grows as tall as 1.5 m (Hinz and Gerber
1998)

Flowering and Pollination

Timing of flower bud production is approximately synchronous with bolting and
occurs when plant height averages 7 .5 cm (Anderson et al. 1996). Flowers are grouped
into racemes at the apex of the stem and at some leaf axils. Mature flowers are white and
average 4-7 mm across with 4 spatulate petals 3-6 mm in length and 4 green sepals. Each
flower bears 4 long and 2 short sepals with one gland at the base of each pair of stamens
(Cavers et al. 1979).

Anderson et al. (1996) described flower development in six stages: (a) bud 2 mm
long with early development of white stripes along sepal margins; (b) bud 3 mm long
with white stripe enlarged along sepal margins; (c) bud 3.3 mm long prior to pushing off
of sepal cap (calyx cap) by the growing petals; ((1) cap stage where calyx cap is pushed
up the top of the petal as the petals expand - calyx cap covers flower for a short duration;
(e) anthesis, during which calyx cap is shed, petals expand and unfold, and the anthers
dehisce; (i) open flower stage when flower is completely open and pollen sacks are
exposed.

Flowers are reflective in the ultraviolet range, which may serve to attract insect
pollinators. Cruden et al. (1996) and Anderson et a1. (1996) studied the pollination and

breeding system of A. petiolata and reached contrasting conclusions. Whereas Cruden et

al. (1996) found A. petiolata in Iowa to be facultatively xenogamous (flowers normally
cross-pollinate, but are self fertile when pollinators are absent), Anderson et al. (1996)
found populations in central Illinois to be primarily autogamous with the majority of
pollination events occurring during the cap stage prior to anthesis. Flowers remain open
for two and infrequently three days. Nectar production and insect visitor frequency peak
on day one of flowering from 0915h to 1830b with the highest visitation rates from

1100h to 1600b (Cruden et al. 1996). Generalist pollinators are reported on A. petiolata in
Illinois (Anderson et al. 1996) and include Diptera: Syrphidae, and Hymenoptera:
Apidae, Andrenidae, Halictidae in Iowa (Cruden et al. 1996). Flowers in Iowa that are not
visited by insect pollinators self-pollinate (Cruden et al. 1996). From 1330h of day one
onward, stamens begin moving inward and eventually brush the anthers against the
stigma where they deposit from 30 to 750 pollen grains. However, insect pollinators
visited the majority of flowers at this site where self-fertilization is apparently a
secondary means of seed production. This contrast in breeding systems fiom Iowa and
Illinois may result fi'om genetic differences between populations that affect flowering

structure and phenology.

Seed Production

Seeds develop in siliques 2.5 - 6 cm long by 2 mm wide on stocky pedicles 4 - 6
mm long. Seeds typically number 10 to 20 per silique depending on the number of fruits
and alternate on either side of the sinus within each silique. Seeds are brown to black, 3
mm by 1 mm and can be cylindrical or ellipsoid with a transverse ridge close to the apex

(Cavers et al. 1979). The number of seeds and siliques is positively correlated with plant

10

height, and the ratio of seeds per silique reported to increase with plant height in some
systems (Smith et al. 2003), although the number of siliques is the best predictor of
fecundity in other studies (see Chapter 2). Susko and Lovett-Doust (1998) showed that
94% of ovules Show Signs of fertilization and begin development, and an average of 68%
of ovules develop into mature seeds. Seed rain by A. petiolata has been reported at
15,000 seeds/m2 (Anderson et al. 1996).

Fecundity can be reduced by several factors. Susko and Lovett-Doust (1998)
showed that damage to 50 - 75% of the root system in second year plants decreased the
number of developing mm on plants by 8 - 13% and decreased the number of fruits
reaching maturity by 4%. Damage to or removal of leaves significantly decreases fruit
set. Removal of 50% and 100% of cauline leaves caused a 25% and 46% reduction in
mature fi'uit production, respectively. However, the stems, fi'uits, and immature seeds of
A. petiolata are green and are likely photosynthetic and compensate for some degree of
defoliation. Significant numbers of mature seeds are still produced even when all cauline
leaves are removed (Susko and Lovett-Doust 1999). Other factors that influence seed
production include plant size and location of flowers within inflorescences (Susko and
Lovett-Doust 2000b). Plants senesce after seed set, although A. petiolata in Europe

perennates by the formation of adventitious buds (Cavers et al. 1979).

RESPONSE TO LIGHT CONDITIONS

Alliaria petiolata plants reach their seasonal Am (maximum photosynthetic rate)
in early spring prior to canopy leaf out and emergence of competing native ground layer
plants and when the solar irradiance reaching the forest floor at mid-day is greatest

(Myers and Anderson 2003). Numbers of leaves and dry biomass in first year plants are

11

positively correlated with higher light levels, lower plant densities, and with increased
nutrient availability (Meekins and McCarthy 2000). Shoot biomass allocation is greatest
among plants gown in higher densities under low light conditions. First year rosettes
have photosynthetic rates and stomatal conduction typical of other shade-adapted plants.
Under low light levels, 1'. e. plants raised under light conditions typical of a forest
understory (189 :l: 93 mol m'2 s"), attain higher Am and stomatal conduction than those
grown under full sun. Under high light conditions, the plants grown under full sun had
higher measured Am than those grown under shade. However, while Am and maximum
stomatal conduction are positively correlated with the light conditions, A. petiolata never
reaches photosynthetic rates or stomatal conduction comparable to full-sun-adapted
species. Plants grown under shade have higher chlorophyll content than those gown
under full sun (Dhillion and Anderson 1999, Meekins and McCarthy 2000). Myers et al.
(2005) found displays of similarly plasticity in in Amax and stomatal conductance
responses to variation in light conditions. They also found that plants grown under higher
light conditions (0 and 30% shade cloth) produced geater biomass and leaf mass than
those grown under lower light (60% shade cloth), but plants gown under low light had
higher leaf chlorophyll a and b concentrations. Thus, while it is shade tolerant and
obviously capable of gowing and competing successfully under a closed forest canopy,
A. petiolata seems optimally adapted to gowing in edge habitats or canopy gaps where
light levels are intermediate. Experimental removals of invasive Lonicera maackii (Amur
honeysuckle) shrubs and Acer platanoides (Norway maple) canopy trees has shown that

A. petiolata does respond positively to the formation of canopy gaps and can increase in

12

abundance even as other principal invaders are removed (Luken et al. 1997, Webb et a1.
2001)
HABITAT

Alliaria petiolata is a disturbance adapted species which profits from
anthropogenic and natural disturbances (Pyle 1995, J. Evans personal observation) and
can tolerate extraordinarily harsh gowing conditions. In a study of the metal content of
plants at a lead battery dump site where soil lead concentrations reached 140,500 mg/kg
Pb, the native plant Ambrosia artemisiifolia had lead concentrations up to 1695 mg/kg Pb
while lead was undetectable in A petiolata (Pichtel et al. 2000). Work by Byers and
Quinn (1987) and others (e. g. J. Evans Chapter 2) has shown that A. petiolata can
successfully colonize and reproduce across a wide range of habitat types ranging from
periodically inundated floodplain forests to dry oak-woods. Biomass of A. petiolata is
positively correlated with soil pH in Illinois (Anderson and Kelley 1995), although
survival rate is not. The effect of pH on gowth may be an effect of lower inorganic
nutrient availability that is typical of more acidic soils. Experiments by Meekins and
McCarthy (2001) indicated that plant performance, including germination, survival, plant
size and seed production, is higher in lowland forests where soil moisture is higher and
leaf litter per unit area is lower compared with upland forest plots. Performance is also
higher in edge plots where light availability is geater than in forest interior plots.
Removal of leaf litter does not significantly affect performance of A. petiolata, though.
However, plots with low densities of A. petiolata have sigrificantly geater survival rates,
plant size, and fruit production. These conditions are similar to those found in newly

colonized patches and may account for A. petiolata’s ability to Spread rapidly and

13

dominate new areas (Meekins and McCarthy 2002).

RANGE OF ALLIARIA PET/OLA TA
Native Range

Alliaria petiolata occurs throughout northern Europe south of 68°N from England
east to Czechoslovakia and from Sweden and Germany south to Italy. (Nuzzo 1993b,
2000). It is also a constituent of the native flora in Northern India and coastal North

Afiica (W elk et al. 2002).

Introduced Range

Alliaria petiolata has spread from its native range in Europe to Sri Lanka, New
Zealand (Bangerter 1985) and much of North America, although its occurance in Sri
Lanka has not been confirmed by recent floristic writings (Nuzzo 2000, Welk et a1.
2002). Alliaria petiolata was first reported in North America on Long Island, New York,
in 1868 (Nuzzo 1993a), where it was likely introduced by irnmigants fi'om the old world.
In North America A. petiolata is most abundant in New England and the Midwest with
populations now present in at least 34 US. States and 4 Canadian Provinces (N uzzo
2000) and appears to have become established near Juneau, Alaska, in 2001 (The Nature
Conservancy 2002, Ellen Anderson, USDA Forest Service personal communication
November, 2005).

Welk et a1. (2002) have predicted the equilibrium distribution of A. petiolata in
North America using a climate based model and give a detailed description of its
temperature and precipitation requirements throughout the year. The primary core

infestation matches climatic conditions with A. petiolata’s native range 100%. It is

14

expected to stretch fi'om Prince Edward Island west to Minnesota and from North
Carolina through Kentucky and Illinois west to Iowa. A second core zone in the Pacific
Northwest extends from southeastern Oregon northward along the Pacific coast to
southwestern British Columbia and fi'om the northern reaches of Idaho to northwestern
Montana. Concentric zones of decreasing probability of infestation extend outward from
core zones. Alliaria petiolata is still spreading though its introduced range in North
America, so predictions about its future distribution are considered preliminary. Some
populations of A. petiolata have been identified outside of the predicted core areas, in
Kansas and Alaska, for instance.

Peterson et al. (2003) created an ecological niche model to make predictions
about A. petiolata’s potential invasive range that explain its occurrence in Kansas. Using
geological, hydrological, climatic, and topogaphic data from A. petiolata’s native range
to calibrate the model, they predict a larger invasive range than Welk et al. (2002). Their
map of A. petiolata’s potential range includes nearly all of the continental United States
and Southern Canada exclusive of the high Rocky Mountains, almost all of Mexico, and

the Caribbean islands.

GENETIC VARIATION
Several goups have studied A. petiolata genetics. Meekins et al. (2001)
performed analyses of genetic variation within and between three North American and
eight European populations of A. petiolata. Analysis of genetic variance showed that
Variance was geatest among populations (61.0%) and that variance is substantially lower
between continents (16.3%) and within individual populations (22.7%). Their findings

indicate that sample populations fi'om Ohio, West Virginia, New York, and Kentucky

15

could possibly have originated fiom stock in the British Isles. Of the three native range
populations sampled, those fiom Belgium and The Netherlands were most similar, and
plants fiorn Scotland belonged in a separate goup. More recently Durka et a1. (2004)
isolated eight new microsatellite loci from A. petiolata. An inter-continental analysis
using these microsatellite loci indicated lower genetic variability in the introduced range
than in the native range, but diversity was high enough in the introduced range to suggest
that A. petiolata has been introduced to North America multiple times (Durka et al.
2005). Durka et al. (2005) found no evidence of a population bottleneck and similar, low
rates of heterozygosity on both continents, which is consistent with high observed rates of
self-fertilization. The populations they studied from North America shared the geatest
proportion of alleles with Northern and Central European and British Isles populations,

which were likely sources of introduction.

INVASIVENESS OF ALLIARIA PE T [OLA TA
Alliaria petiolata is a successful invader that has demonstrated its ability to

disrupt and restructure natural (e. g. Nuzzo 1993a, Meekins and McCarthy 1999, Nuzzo

1999) and urban (Y ost et al. 1991) forest communities throughout its North American
range. Numerous studies have attempted to dissect the nature and mechanisms of its
invasiveness. Alliaria petiolata appears to owe its invasiveness at least in part to its cold
tolerance. First, A. petiolata germinates and later resumes gowing during its second year
while temperatures are still low and before trees have leafed out and closed the forest
canopy. From studies on responses to different light levels (Dhillion and Anderson 1999,

Meekins and McCarthy 2000), we know that A. petiolata reaches its maximum

16

photosynthetic and gowth rates under conditions just short of full sun. Thus, A. petiolata
is able to gow rapidly for several weeks before other understory plants emerge and it has
an opportunity to overtop and shade them. Additionally, A. petiolata overwinters as a
geen rosette and is able to photosynthesize and gow whenever temperatures rise above
freezing. A gowing body of current research suggests that exotic earthworms increase
litter cycling rates (Bohlen et al. 2004, Hale et al. 2005a), alter soil structure (Hale et al.
2005b) n‘ansport and store weed seeds (Smith et al. 2005) and may also be correlated
with increased abundance of invasive plants (Kourtev et al. 1999) and A. petiolata
specifically (Maerz et a1. 2002, CM. Hale, Univ. of Minnesota, Duluth, personal
communications 2003). Although many negative impacts of A. petiolata on native
communities have been surmised, few studies have shown evidence of large-scale
impacts on invaded communities. Ground beetle (Coleoptera: Carabidae) assemblages,
species richness and other invertebrate prey abundances show no correlation with A.
petiolata presence or absence (Davalos and Blossey 2004). McCarthy (1997)
experimentally removed A. petiolata from forest communities and measured subsequent
changes in community composition for over two years. He concluded that diversity
increased as A. petiolata was removed, but this is somewhat confounded and difficult to
interpret because the sigr of the effect he observed changed multiple times over the
course of his study.
Meekins and McCarthy (1999) studied competitive interactions between A.
petiolata and several native species in Ohio and found that both Acer negundo and

Impatiens capensis are superior competitors to it but that Quercus prinus seedlings were

17

less successful competitors than A. petiolata. This implies that some oak forests may be

at risk of reduced regeneration rates through seedling suppression by A. petiolata.

Better Living through Chemistry

In coevolved communities, natural selection is expected to favor organisms with
mechanisms of reducing competitors’ fitness and competitive ability, while at the same
time competitors should be selected for their ability to tolerate those offenses. In plant
communities these interactions can be mediated through the production of secondary
chemical compounds. The Novel Weapons Hypothesis (Callaway and Ridenour 2004)
posits that when plants invade communities with which they Share little recent
evolutionary history, the invaded community is not likely equipped to tolerate the
invaders’ chemistry. Similarly, naive insect herbivores whose life cycles are closely tied
to relatives of an invading plant may be chemically attracted to the invader, but other
chemistry or plant properties render it unacceptable as a host, and it will serve as a
population sink if larvae cannot develop on it. Both of these phenomena have been shown

to occur where A. petiolata invades certain North American communities.

Allelopathy: Interactions with Plants

Numerous studies have found A. petiolata to be chemically well equipped for
both defensive and potentially offensive purposes which may contribute to its
invasiveness via several mechanisms. Allelopathy has a been cited as a cause of increased
invasive ability in several invasive plants including knapweeds (Centaurea spp.,

Acroptilon repens (L.)) (Bais et al. 2003, Grant et al. 2003, Weston and Duke 2003) and

18

Tree-of Heaven (Ailanthus altissima (Miller)) (Call and Nilsen 2003). McCarthy and
Hanson (1998) applied root and shoot extracts of A. petiolata to seeds of four target
species: radish, winter rye, hairy vetch, and lettuce. Radish germination rate and Shoot
biomass of rye were depressed by treatment with the extracts, but the authors concluded
that allelopathy is unlikely to be important in A. petiolata’s invasion ecology.

Vaughn and Berhow (1999) criticized McCarthy and Hanson’s methods and
pointed out that several highly phytotoxic products of glucosinolate hydrolysis are not
extractable as prepared. More specifically, the chemicals allyl isothiocyanate (AIT C) and
benzyl isothiocyanate (BzITC), which result fi'om the breakdown of their less toxic
parent compounds Sinigin and glucotropaeolin, are almost entirely insoluble in water.
The aqueous preparations used by McCarthy and Hanson (1998) would not likely have
contained these important toxins and would not be expected to affect the gowth of other
plants. Dichloromethane extracts of A. petiolata tissues prepared by Vaughn and Berhow
contained both AIT C and leT C as well as 2,3-epithiopropylnitrile. Solutions of AIT C
and BzITC and their parent compounds were lethal to cress (Lepidium sativum L.) and
wheat (Triticum aestivum L.) although the isothiocyanates were lethal at concentrations
an order of magritude more dilute than their parent glucosinolates. Concentrations of
these compounds vary seasonally in natural A. petiolata populations. Sinigin and
glucotropaeolin are undetectable in spring-harvested leaves and stems but are present
autumn-collected specimens. Sinigin is present in roots at similar concentrations in both

Spring and fall, but root-concentrations of glucotropaeolin are over three times as geat in

fall than in spring (Vaughn and Berhow 1999).

19

Work by Prati and Bossdorf (2004) has demonstrated the allelopathic ability of A.
petiolata by other means. They showed that germination rates of a North American native
plant were reduced in soils in which A. petiolata had been gown compared with control
soils and that germination of congeneric European plants was positively affected by A.
petiolata treated soils. Allelopatlric effects of A. petiolata specimens collected in North
America were gater than those of European specimens in concordance with Blossey and
NOtzold’s (1995) Evolution of Increased Competitive Ability hypothesis (EICA).
However, Prati and Bossdorf only measured allelopathic affects on one European and one
North American species. Further field trials involving geater species numbers should be
undertaken to increase confidence in their results.

Vaughn and Berhow (1999) demonstrated direct allelopathy by compounds
present in A. petiolata, but they also point out that the glucosinolates and glucosinolate

derivatives in their and other studies inhibit gowth of mycorrhizal fungi. They propose
that this could enable a mechanism for competitive superiority of A. petiolata over many
native plants that are dependent on arbuscular mycorrhizal fungi (AMF). Subsequent
experiments have explored this hypothesis. Aqueous leachates of whole A. petiolata
plants prevented germination of spores of the AMF Gigaspora rosea, reduced
germination rates of tomato seeds, and prevented association of tomato seeds with AMF
(Roberts and Anderson 2001). In natural communities, Roberts and Anderson (2001) also
identified a negative correlation between local A. petiolata density and soil mycorrhizal
inoculum potential which they suggest could negatively impact native plants that are
dependent on AMF associations. This has been verified in other systems where the

persistence of anti-mycorrhizal compounds in the soil prevented re-inoculation with AMF

20

for years after A. petiolata removal (Stinson and Klironomos 2005). This study showed
that later successional plants, which are more highly dependent than early successional
plants on AMF associations, are negatively affected by A. petiolata and suffer from
reduced gowth and regeneration in its presence. Interactions of this kind have the
potential to radically alter forest regeneration and successional trajectories. Stinson et al.
(2005) experimentally demonstrated negative relationships between A. petiolata
abundance and native species diversity and cover. These community responses could be
mediated through competition, direct allelopathy, AMF suppression by A. petiolata, or a
combination of mechanisms. Prati, Klironomos, Calloway and others are carrying out
additional studies on A. petiolata allelopathy and plant-soil feedback mechanisms (R.
Calloway, U. of Montana, personal communication January 22, 2004). In McCarthy’s
removal study (1997) it seems likely that although A. petiolata was removed from his
experimental plots, his results may have been influenced by the lasting suppression of

soil AMF by A. petiolata secondary chemicals as others have shown.

Defensive Chemistry: Interactions with Insects

In North America A. petiolata is utilized as a nectar source by the spring azure
butterfly Celastrina ladon (Y ahner 1998) and is accepted as a host for oviposition by the
native butterfly Pieris napi oleracea P. (Lepidoptera: Pieridae) (Porter 1994b). However,
1’. n. oleracea is unable to complete development on A. petiolata with most larvae dying
in the first or second instar (Porter 1994b). First instar larvae are deterred from feeding by
the presence of the cyanoallyl glycoside alliarinoside ((2Z)-4-()B-D-glucopyranosyloxy)-

2-butenenitrile) (Haribal and Renwick 2001, Haribal et a1. 2001, Renwick et a1. 2001)

21

while fourth instar larvae are deterred by the apigenin derivative isovitexin-6”-D-fl—
glucopyranoside (Haribal and Renwick 1998, Haribal et al. 1999, Renwick et al. 2001).
Concentrations of these secondary chemicals vary temporally with levels near zero in
June and July, which may explain why larval development and survival rates on A.
petiolata vary (Haribal and Renwick 2001). Larvae of the congeneric butterfly P. rapae
are stimulated to feed by extracts of glucosinolates from A. petiolata and other
Brassicaceaeous plants (Renwick and Lopez 1999). Renwick (2002) postulates that a
similar attraction leads P. n. oleracae to oviposit on A. petiolata, even though other plant
secondary chemicals inhibit larval development. Alliaria petiolata thereby serves as a
population sink for this native butterfly through a series of chemical “lures” and “traps”
(Renwick 2002). Porter (1994a) observed P. n. oleraceae’s uncOmmon congener P.
virginiensis ovipositing on A. petiolata. Where P. virginiensis 's primary hosts, Dentaria
spp. (Brassicaceae) are rare and A. petiolata is abundant, A. petiolata is a preferred host,
despite its unpalatability. If eggs are preferentially laid on unacceptable host plants, it is
possible that this species’ abundance will decline though a similar population sink

mechanism.

Interactions with White-Tailed Deer

White tailed deer (Odocoileus virginianus) (Boddaert) are important herbivores in
the fragnented agicultural landscapes of the Midwestem United States. Deer browse can
significantly alter the species composition of forest understory communities (Rooney
2001) and may disproportionately affect Liliaceae in some areas (Augustine 1997).
Estimates of deer densities in North America prior to European settlement ranged from 2

- 4.2 deer krn’2 (Alverson et al. 1988, Rooney 2001) while current deer densities in

22

Michigan are as high as 28.6 deer krn2 (MDNR 2006 Montcalm County). Kalisz et al.
(2003) found that trampling and selective browsing of native vegetation by deer
facilitated invasions by A. petiolata, but native species increased in abundance in areas
fiom which deer were excluded and were more competitive with A. petiolata. In
particular, the native species Trillium grandiflorum (Liliaceae) appeared to benefit fiom
protection from browsing while A. petiolata became less successful at establishing and
reproducing. Based on these findings a study evaluating the individual and combined
impacts of deer and A. petiolata on native plant communities was established in southern

Michigan and is ongoing (J. Evans unpublished).

Differences between European and North American Alliaria petiolata populations

Blossey and Nc’itzold’s (1995) Evolution of Increased Competitive Ability (EICA)
hypothesis proposed that in the absence of natural enemies, selection should favor
individual plants with increased resource allocation to gowth and decreased allocation to
energetically expensive defenses. Bossdorf et al. (2005) reviewed papers that address
various components of the EICA hypothesis in many invasive plant Species. Of the
studies that tested components of EICA on A. petiolata, nearly all either indicate zero
differences between native (European) and introduced (North American) populations or
they fail to support EICA. The only exceptions were Prati and Bossdorf’s (2004)
previously described allelopathy study and a study in which feeding rates of the European
specialist herbivore Ceutorhynchus scrobicollis (Coleoptera: Curculionidae) were geater
on introduced than on native A. petiolata populations (Bossdorf et al. 2004). This

indicates that A. petiolata in North America may have lost some ability to defend itself

23

against specialist herbivores and thus may be vulnerable to them if they are introduced as
biocontrol agents. However, the remaining body of work reviewed suggests that
introduced A. petiolata has not traded off defenses for an increase in competitive ability.
Rather, it Shows that the introduced populations are equally or less competitive than
European ones (Bossdorf et al. 2005) and are at least as well chemically defended as well
(Cipollini 2002, Cipollini et al. 2005). These findings may have important implications

for developing efforts to control A. petiolata invasions in North America.

MANAGEMENT OF INVASIVE GARLIC MUSTARD

Resource managers in North America perceive expanding A. petiolata populations
as threatening to many of their specific goals including conservation, recreation, and
wildlife management. Efforts to control or reverse the spread of A. petiolata both locally

and nationally have taken myriad approaches and are currently still being developed.

CONVENTIONAL CONTROLS

A number of conventional techniques have been developed in attempts to control
A. petiolata, including prescribed fire, chemical and mechanical controls. These methods
have proven insufficient in all but the smallest infestations. Additionally, treatments must
be repeated annually or semi-annually until the seed bank is exhausted to prevent re-

establishment after treatment.

Prescribed Fire
Several goups have explored the use of prescribed burning to control A. petiolata

and produced mixed results. Luken and Shea (2000) found no sigrificant effect of

24

 

 

 

burning on A. petiolata populations after three consecutive years of treatment in
Kentucky. Mid intensity fires (flames ~15 cm high) in late Spring that completely burn an
area can effectively reduce A. petiolata populations (Nuzzo 1991). However, fires often
burn patchily and do not eliminate all target vegetation, requiring further mechanical or
chemical treatment of remaining plants. Fall conditions in Nuzzo’s (1991) Illinois trials
were frequently too wet to produce a thorough burn, and low intensity fires (flames up to
3 cm) have no impact on A. petiolata populations.

Nuzzo et al. (1996) set repeated, controlled hot fires (flames >1 m) during three
consecutive springs and in alternate autumns and springs over four years. All plots were
free of A. petiolata following fire treatments but were quickly reinfested when treatment
was discontinued. Herb coverage increased 65 - 66% and species richness increased 50%
after two consecutive burns, although woody cover was not sigrificantly affected. Spring
and autumn high-intensity fires had equal effects on A. petiolata (Nuzzo et al. 1996,
Schwartz and Heim 1996), although Nuzzo (2000) points out that spring fires should be
most effective in reducing A. petiolata coverage because they affect both the newly
emerged seedlings and the rosette stage from the previous season’s cohort. Fires that do
not completely remove the litter layer may actually increase A. petiolata coverage by
promoting gowth of new flowering stems from axillary buds (Nuzzo et al. 1996).
Prescribed burning may be effective in controlling mid-sized A. petiolata populations if
used repeatedly and in combination with other methods. Fire is only recommended for

fire-adapted communities and may produce undesirable effects if used inappropriately.

25

 

Chemical Control

A number of herbicides have been explored for A. petiolata, including glyphosate
(Nuzzo 1991, 1996), 2,4-D (Rich Dunbar personal communication 1990 in Nuzzo 2000;
Nuzzo unpublished in Nuzzo 2000), 2,4-D plus Dicamba (Bill McClain personal
communication in Nuzzo 2000), Kihnor (a 2,4-D formulation) (RH. Brown personal
communication 1977 in Cavers et al. 1979), triclopyr (Rich Dunbar personal
communication 1990 in Nuzzo 2000), bentazon and acifluoren (Nuzzo 1994, 1996). 2,4-
D is not indicated for use on A. petiolata, although a 2,4-D ester formulation reduced A.
petiolata cover up to 70% when applied at 0.09 kg/ha during the gowing season (Nuzzo
2000). Likewise, acifluoren is not recommended (Nuzzo 2000). While triclopyr, 2,4-D
plus Dicamba, and Kilrnore have been used in limited trials, insufficient testing has been
conducted to recommend these products. More extensive data on glyphosate and
bentazon are available.

Dormant season broadcast (overspray) application of 1%, 2%, and 3% glyphosate
in late autumn or early spring reduced A. petiolata cover by over 95% without sigrificant
impact on other herbaceous species, although sedge cover was reduced by >83% (Nuzzo
1991, 1996). Carlson and Gorchov (2004) spot applied 1% glyphosate to A. petiolata in
late autumn in an old gowth Acer-Fagus forest and a second gowth Liriodendron forest
and found similar reductions in A. petiolata in subsequent years and observed increase
cover of native Spring ephernerals and increased reproduction of the native perennial
Phryma Ieptostachya (lopseed). They concluded that targeted treatment of A. petiolata
can positively impact native communities, but their methods are labor intensive and may

not be feasible for application at large spatial scales.

26

 

 

 

Application of bentazon (Basagan) at 0.09-0.18 kg/ha reduced coverage of
rosettes 90-95% (Nuzzo 1994), although dormant season application effects are unknown
(Nuzzo 1996). Bentazon is highly soluble in water has a low affinity for soil particles,
which creates the potential for goundwater contamination (Nuzzo 2000). This must be

weighed against its lower toxicity to non-target sedges and gasses.

Cutting

Cutting A. petiolata stems has been used with mixed success to control small
infestations. Nuzzo (1991) found that stems cut at gound level and at 10 cm showed 99%
and 71% mortality, respectively, compared with a natural mortality rate of 6% in control
plots. Seed production was reduced 100% and 98% respectively in plants cut at gound
level and at 10 cm above gound level. Plants cut during the height of bloom with
developing siliques have the least stored resources and are least likely to produce new
stems and recover fi'om cutting (Nuzzo 2000).

Viable seeds can develop on cut stems (Solis 1998). In his experiment, Solis
pulled A. petiolata stems in successive stages of development (budding plants, blooming
plants, early silique, and late siliques with developing seeds) and laid them to overwinter
in mesh-enclosed plots. The following spring A. petiolata seedlings had germinated in all
but his control plots, implying that seeds had developed to maturity on the pulled stems.
Solis recommends removing and destroying all pulled stems from the Site to prevent
further seed set. As an extension of this, Nuzzo (2000) recommends removing and
destroying cut stems as well. Stems can be cut with a weed whip, although care must be

taken to avoid damaging non-target species and to prevent spreading the cut stems, as

27

 

 

stated above (Nuzzo 2000).

Pulling

Pulling is an effective control technique in very small infestations (Nuzzo 2000).
When pulling plants, it is important to remove the upper portion of the roots as well as
the stem, since buds in the root crown can produce additional stems (Nuzzo 2000). All
pulled plants should be removed from the site as seed ripening continues even after plants
are pulled (Solis 1998). Repeatedly hand pulling of garlic mustard is reported to be
effective for control in small areas but has limitations. Because seeds remain viable in the
soil for up to five years, it is important to pull all garlic mustard plants in an area every
year until the seed bank is exhausted and seedlings no longer appear. This will require
multiple efforts each year as rosettes can continue to bolt and produce flowers over an
extended period (April-June). Pulling is advantageous over cutting in that it can be done
at any time during A. petiolata ’s lifecycle before seeds dehisce. Cutting the roots is

effective as well, although it is labor intensive (Nuzzo 2000).

Mowing

Mowing as a control method should be approached with caution (N uzzo 2000).
Mowing equipment may not be appropriate for use in sensitive areas as it is likely to
disturb the soil and damage desirable plants as well as A. petiolata. Nuzzo (2000)
suggests that while mowing appears to be an attractive option for areas like roadsides, it
could serve to spread seeds within and between sites if seeds are not removed from

equipment.

28

Other

Draining, dredging, and gazing have not been tested (N uzzo 2000), although
cattle reportedly do feed on A. petiolata (Cavers et al. 1979). Midsummer flooding is
capable of killing rosettes, but floodwaters apparently accelerate the invasion process by
spreading A. petiolata seeds (Nuzzo 1999). Nuzzo (2000) suggests that disking is

inappropriate for A. petiolata as the soil disturbance it creates damages desirable plants.

BIOLOGICAL CONTROL

Many conventional methods described above have been explored to control
Alliaria petiolata. However, A. petiolata’s high seed productiOn, persistent seed bank,
and tolerance for varied light conditions render it difficult to control, and land managers
are unable to successfully curb its spread. More powerful management tools are
necessary to reverse the trajectory of A. petiolata’s population gowth. In April, 1998 a
search for appropriate biological control organisms was launched to address this need
(Hinz and Gerber 1998). This effort is being carried out by CABI Bioscience in
Switzerland and at the University of Minnesota. It is funded by the USDA Forest Service
through Cornell University and the Minnesota Department of Natural Resources. Blossey
et al. (2001b) and Hinz and Gerber (2005) have summarized the search for biocontrol
agents. More complete information is found in CABI’S annual project reports (Hinz and
Gerber 1998, 2000, 2001, Gerber et al. 2002, Gerber et al. 2003, Gerber et al. 2004,
2005).

An initial survey of the literature by Hinz and Gerber (1998) found 69 species of

phytophagous insects and 17 fungi in Europe that are associated with A. petiolata. Of

29

these, 28 insect species were collected in Switzerland, Germany and Austria in 1998 and
1999, and 20 were reared (Hinz and Gerber 1998, 2000). Six species were identified as
potential biological controls and five have been subjected to further testing. Four species
belonging to the subfamily Ceutorhynchinae (Coleoptera: Curculionidae) are in the genus
Ceutorhynchus. These are: C. alliariae, C. roberti, C. constrictus, and C. scrobicollis. A
fifth Ceutorhynchus species, C. theonae, was studied in 2000 and 2001 but was difficult
to collect and rear and was thereafter discontinued (Gerber et a1. 2003). Phyllotreta
ochripes (Curtis) (Coleoptera: Chrysomelidae) has also been tested. The Sixth species,
Ophiomyia alliariae Hering (Diptera: Agomyzidae), was not found during initial or

subsequent surveys and has not been tested as a potential control agent.

Potential Agent: Phyllotreta ochripes

Phyllotreta ochripes (Curtis) (Coleoptera: Chrysomelidae) larvae feed beneath the
epidermis on roots or root crowns of Alliaria petiolata rosettes and bolting adults (Hinz
and Gerber 1998, Blossey et al. 2001b). Tests for host Specificity of this flea beetle were
conducted in 1999 and 2000 (Hinz and Gerber 2000, 2001) and were confirmed by
multiple choice tests in 2001 (Gerber et al. 2002). Results showed that P. ochripes is
capable of completing development on a number of Brassicaceae including Brassica spp.
and Rorippa spp.. In 2001 P. ochripes was determined to be to oligophagous to be

released in North America and further testing was suspended (Gerber et al. 2002).

30

Potential Agents: Ceutorhynchus alliariae and C. roberti

Ceutorhynchus alliariae Bristout and C. roberti Gyllenhal (Coleoptera:
Curculionidae) share similar life histories (Blossey et al. 2001b, Gerber et a1. 2002).
Larvae mine in stems and leaf petioles from March to May and pupate in the soil. Adults
emerge later the same summer and feed on leaves of A. petiolata. Adults overwinter in
the litter and soil, emerge early in the spring and soon begin ovipositing. Eggs of C.
alliariae are laid individually while those of C. roberti are laid in goups of up to nine in
A. petiolata stems and leaf petioles. Development takes from 1-3 weeks. Larvae hatch
and feed in stems and leaf petioles for approximately 7 weeks before third instar larvae
enter the soil to pupate. Adults emerge in June and feed on leaves. The adults of C.
alliariae are capable of flights of at least 1 km, although C. roberti adults have only been
observed flying short distances and infi'equently (Gerber et al. 2004). Both species are
univoltine and can be distinguished fi'om one another as adults by tarsal coloration,
although larvae are indistinguishable (Blossey et al. 2001b). Both can also survive for
more than two years and have a third oviposition season (Gerber et al. 2003).

In host specificity testing C. roberti has accepted 11 of 40 host species for
oviposition that were offered in sequential no-choice tests, and C. alliariae accepted 23
of 63 species - all within the family Brassicaceae. In single choice oviposition tests C.
roberti accepted 10 of 18 host plants offered and C. alliariae accepted 12 of 23. During
no choice development trials, C. roberti completed development on three of 22 host
species offered (Rorippa nasturtium-aquaticum (= Nasturtium oflicinalis), Thlaspi
arvense, and Peltaria alliacea - all Brassicaceae) and C. alliariae developed successfully

on two of 19 species offered (R. nasturtium-aquaticum and Thlaspi arvense) (Hinz and

31

Gerber 2005). Further host specificity testing of C. roberti and C. alliariae is being
continued at the University of Minnesota’s quarantine laboratory in St. Paul on several
additional North American plant species (Katovich et al. 2005). Both C. alliariae and C.
roberti are still currently considered candidate agents (Gerber et al. 2004, Gerber and

Hinz 2005, Gerber et al. 2005).

Potential Agent: Ceutorhynchus constrictus

Ceutorhynchus constrictus (Marsham) (Coleoptera: Curculionidae) has the
narrowest host range of the candidate biocontrol agents to date (Hinz and Gerber 2005).
Larvae feed on seeds fi'om May to July and then leave the host plant to pupate in the soil.
Adults emerge the following April to feed on leaves and mate. Females in fecundity trials
laid an average of 164 eggs from 21 May to 21 June (Hinz and Gerber 2000). Eggs hatch
after 15 to 20 days, and larvae exit seeds to pupate in the soil fi'om 28 to 44 days later.
Each larva destroys 2.5 i 0.34 seeds on average (Gerber et al. 2002).

Of 54 host plant species offered in no choice oviposition trials, C. constrictus
females accepted ten in the genera Arabis, Aurinia, Brassica, Barbarea, Rorippa, and
Cardamine (all Brassicaceae) (Gerber et al. 2004, 2005). Of these ten species only one
(Barbarea vulgaris) was accepted in single choice trials implying C. constrictus’s
preference for A. petiolata (Gerber et al. 2002, Gerber et al. 2004, 2005). Brassica nigra
alone supported development of C. constrictus to the adult stage in no-choice

development trials, but this Species was not attacked in open-field trials in 2003 which
suggests that it is not likely a normal field host of this insect. Additionally, C. constrictus

i s not a known pest on commercially produced B. nigra in Europe (Gerber and Hinz

32

2005, Hinz and Gerber 2005). Ceutorhynchus constrictus is the most common European
species of A. petiolata feeders in its genus. However, attack rates by C. constrictus on A.
petiolata in the field are typically low, reducing seed production by 0.3 - 6.4% in

Switzerland and southern Germany (Blossey et al. 2001b).

Potential Agent: Ceutorhynchus theonae

Ceutorhynchus theonae Korotyaev & Cholokava (Coleoptera: Curculionidae) is
the most recently identified potential biocontrol agent for A. petiolata. It was first
collected in Daghestan, Russia, in early 2000 and was reared in quarantine by CABI in
Switzerland through 2001 (Blossey et al. 2001b, Gerber et al. 2002). Larvae feed on
seeds of A. petiolata (Blossey et al. 2001b), although little else has been published on its
biology. Work on C. theonae was suspended after the 2001 field season because of

collection and rearing difficulties (Gerber et al. 2003).

Potential Agent: Ceutorhynchus scrobicollis

Ceutorhynchus scrobicollis Nerensheimer & Wagrer (Coleoptera: Curculionidae)
larvae feed in leaf petioles, buds, and root crowns of overwintering A. petiolata rosettes.
Larvae leave the plants to pupate in the soil by late April. Adults emerge fi'om May to
June and aestivate during summer. Females begin laying eggs in mid September and
oviposit continually through winter into Spring. Individual females can produce viable
eggs for at least three consecutive years, although few survive that long and fecundity
decreases with age. This may still have positive implications for establishment and spread

if this species is released (Gerber et al. 2002, Gerber et al. 2003, Gerber et al. 2004).

33

Ceutorhynchus scrobicollis was not found during initial collecting trips in Switzerland
and southern Germany in 1998, although it was collected Germany in subsequent years
(Hinz and Gerber 2000, 2001, Gerber et al. 2002, Gerber et al. 2004, 2005).
Ceutorhynchus scrobicollis oviposited on 35 of 73 host plant species offered in
sequential no-choice tests. Of these, six in the Brassicaceae and three in other families are
native to North America. These include Hydrophyllum virginianum (Hydrophyllaceae),
Viola sororia (V iolaceae), and Phlox divaricata (Polemoniaceae). Eggs were laid on the
surface of H. virginicum, V. sororia, and P. divaricata and desiccated prior to larval
emergence, whereas they are normally deposited in leaf petioles. The authors of these
studies raise doubts as to whether C. scrobicollis would complete development under
these conditions (Gerber et al. 2005, Hinz and Gerber 2005). All three non-Brassica
North American species co-occur with A. petiolata in Michigan at one or more A.
petiolata study Sites (Evans unpublished data, and see Chapter 3). Three North American
species (Draba sp., Cakile edentulata, and Lepidium virginianum, - all Brassicaceae) and
eight other species (all Brassicaceae) were accepted for oviposition during single-choice
oviposition tests in 2002 (Gerber et al. 2003). In larval development trials C. scrobicollis
adults emerged from five of 37 species offered. These included the three species that C.
roberti and C. alliariae accepted in similar trials (R. nasturtium-aquaticum, P. alliacea,
and T. arvense) (Gerber et a1. 2005). Experiments comparing C. scrobicollis’s feeding
behavior indicate that it prefers A. petiolata plants gown fiom North American seed over
European plants (Bossdorf et al. 2004). This implies that estimates of C. scrobicollis
feeding and infestation rates made on European A. petiolata may be low for North

America.

34

Interactions Between Potential Biocontrol Agents

Experiments conducted in 2000 and 2001 revealed strong but symmetrical intra-
and interspecific competitive interactions between C. alliarae and C. roberti (Gerber and
Hinz 2005). These studies showed that weevil fecundity and oviposition declined as
weevil density on A. petiolata increased (Gerber et al. 2002). Similarly, A. petiolata mean
height decreased linearly as the number of weevil pairs per plant increased (Hinz and
Gerber 2001). However, plant responses were not species-specific and appear to be a
function of weevil density alone, irrespective of species composition. The authors state
that C. alliarae had a geater impact on plant performance than C. roberti, but they
attribute this to C. roberti’s geater sensitivity to handling by experimenters and not to
higher damage rates by C. alliarae. Although mean A. petiolata height was negatively
impacted by feeding damage, the number of inflorescences per plant responded positively
to feeding as laterally buds were released from suppression. Wilting is induced at larval
densities of 20-30/shoot, although seed production is reduced at lower levels (Blossey et
al. 2001b).

CABI researchers conducted experiments to identify interactions between C.
scrobicollis and C. alliariae, which are temporally and spatially segegated on A.
petiolata (Gerber et al. 2003, Gerber and Hinz 2005). They were surprised to find that,
while C. scrobicollis-darnaged plants exhibited increased mortality, a 60% reduction in
biomass and a 48% reduction in fecundity compared with control plants, C. alliariae was
not negatively impacted by the presence or abundance of C. scrobicollis. These effects
and interactions were studied on both large and small A. petiolata rosettes. Larger

rosettes were better able to compensate for feeding damage than smaller rosettes by

35

producing multiple stems when attacked by one or both species, although C. scrobicollis
had a geater impact than C. alliariae. In contrast, the presence or abundance of C.
alliariae had no impact on A. petiolata mortality. Plants infested with C. alliariae alone
did have sigrificantly lower shoot heights, produced geater numbers of inflorescences
than uninfested plants and produced numerically but not statistically lower numbers of
seeds. The terms for interactions between the two weevil Species were insigrificant in
their analysis. Data from available reports may be insufficient or inappropriate as reliable
estimates of these insects’ impacts on A. petiolata demo gaphic rates in natural

communities.

Biological Control Summary and Outlook

Ceutorhynchus scrobicollis and C. alliariae are currently the furthest along in
host specificity testing. Of these C. scrobicollis has had the geatest impacts on A.
petiolata performance and is predicted to be the most effective of the four weevils if
released. Both of these two as well as C. roberti have been imported into the United
States to the quarantine facility at the University of Minnesota since 2003 for further
host-range testing on native North American plant species (Gerber et al. 2004, 2005,
Katovich et al. 2005). While work with C. scrobicollis has progessed in quarantine in
Minnesota, no tests have been conducted using C. alliariae or C. roberti because of
difficulties encountered in getting them to lay eggs (L.C. Skinner personal
communication 01/23/2006).

Of the plant species most frequently and commonly accepted for oviposition,
feeding, and development by the goup of potential biocontrol agents, T. arvense and P.

alliacea are both native to Europe and considered by some to be invasive in North

36

America. Various sources list R. nasturtium-aquaticum in North America as either native
(U SDA-NRCS 2006) or exotic (Gerber et al. 2004 and references therein). The assertion
that C. roberti and C. alliariae are unlikely to encounter these species under natural field
conditions is plausible for T. arvense and P. alliaciea, but A. petiolata and Rorippa spp.
are sympatric at one known location in Michigan where A. petiolata is being studied (J.
Evans personal observation). The consistent acceptance of Rorippa spp. by multiple
Ceutorhynchus Species in host-specificity trials is of concern to some, but no conclusions
have been reached yet regarding the potential for future agent releases to impact these
plant species in natural communities and whether those risks are acceptable (L.C. Skinner
personal communication 01/23/2006).

Petitions must be submitted to and approved by the USDA Technical Advisory
Group (TAG) on weed biological control before any proposed biological control agent
can be released in the United States. No petition for approval to release any of the
potential agents in North America has been submitted yet, although a proposal to release
C. scrobicollis seems a likely next step in the A. petiolata biocontrol progarn followed
by proposals for the leaf miners (C. alliariae and C. roberti) and finally the seed feeder
(C. constrictus) (Gerber and Hinz 2005). Gerber and Hinz (2005) have indicated that the
first three are predicted to be compatible with each other and could potentially be
released together. However, Hinz and Gerber (2005) propose that firrther testing of native
North American Thlaspi and Rorippa species and investigation of the phylogenetic
relationship of T. arvense, R. nasturtium-aquaticum, and P. alliacea be conducted prior
to submission of a release proposal to the USDA TAG as these questions are likely to be

raised by any review board.

37

If agents are approved, the first releases will be conducted by members of the
Cornell based consortium. A similar goup of state and federal collaborators carried out
the highly effective purple loosestrife biocontrol progarn across North America
(Katovich et al. 1999, Blossey et al. 2001a, Blossey et al. 2001c, Katovich et al. 2001,
Albright et al. 2004) and in Michigan (Kaufman and Landis 2000, Blossey et al. 2001a,
Sebolt and Landis 2002, Landis et al. 2003). Biocontrol agents for A. petiolata would be
made available to states with a demonstrated need, baseline A. petiolata population data

and the capability to execute successful biocontrol projects (Landis et al. 2004).

Predicting Biocontrol Requirements and Outcomes

A number of empirical and simulation studies have attempted to identify
vulnerabilities in A. petiolata populations that can be exploited by managers. Drayton and
Primack (1999) showed that extirpation of small populations can be achieved more
readily than in larger poulations. Realistically, what these authors referred to as large or
small “populations” were really isolated satellite patches of A. petiolata that ranged in
size up to ~ 2000 individuals. At the scale of a site (i.e. a forest) 2000 individuals can
occupy an area as small as several meters. Therefore, their findings imply that the only
hope of driving local populations extinct is by managing them as soon as they are
established. Their matrix population model of the A. petiolata lifecycle predicted that the
interfering with the transition fi'om seeds to rosettes should have the geatest impact on
population gowth. Rejmanek (2000) pointed out an error in their model. Preliminary
results of subsequent modeling efforts predict that the rosette to flowering transition and

any transitions affecting fecundity should have the geatest impact on population gowth,

38

and that these life stages would be the most fruitful targets (Davis et a1. 2005, Davis et al.

in review).

RESEARCH GOAL AND OBJECTIVES

Land managers in Michigan have identified controlling the spread of A. petiolata
populations as a priority. When initial efforts to control A. petiolata by conventional
means failed, managers and researchers in the state turned their attention to the
development of biological control strategies. The release of natural enemy insects into
novel environments has a number of associated risks that include both direct (e. g. Stiling
et al. 2004, Louda et a1. 2005) and indirect (e. g. Pearson and Callaway 2003, Ortega et al.
2004, Pearson and Callaway 2005) effects on non-target species Or communities. These
can be difficult or impossible to predict using current biocontrol agent screening
standards. Researchers hope to introduce better predictive tools to the biocontrol agent
screening process in the future, but these are still under development (Davis et al. 2005,
Davis et al. in review). Prior to accepting the unknown risks of introducing natural
enemies, resource managers want to know whether the spread of the invader warrants
such risk. Specifically, they want to know whether invasions by A. petiolata are
negatively impacting Michigan natural communities and whether or not existing
assemblages of herbivores, pathogens, or other processes that are present in Michigan
currently are controlling A. petiolata. If biocontrol agents are approved and released in
Michigan in the future, baseline data on the invaded communities where releases will be
made are necessary to evaluate the relative effectiveness and success of the biocontrol

agents once established (Blossey 1999).

39

The overall goal of my research was to explore and interpret the biology and
invasion ecology of A. petiolata in Michigan in support of current and firture efforts to
control its spread. To meet this goal, my specific research objectives were to (1)
document, describe, and characterize the natural communities that A. petiolata invades in
Michigan to allow future evaluation of potential biological control efforts, (2) determine
whether populations of A. petiolata are spreading in Michigan, (3) evaluate the impacts
of existing Michigan herbivore communities and diseases on the spread of A. petiolata,
and (4) evaluate the interactions between A. petiolata and native plant communities to
identify negative impacts of A. petiolata invasion and the response of A. petiolata
populations to native Michigan communities.

I have addressed these core research goals in work conduCted from 2003 - 2005 in

southern Michigan.

40

LITERATURE CITED

Albright, M. F., W. N. Harman, S. S. Fickbohm, H. Meehan, S. Groff, and T. Austin.
2004. Recovery of native flora and behavioral responses by Galerucella spp.
following biocontrol of purple loosestrife. American Midland Naturalist 152:248-
254.

Alverson, W. S., D. M. Waller, and S. L. Solheim. 1988. Forests too deer: Edge effects in
Northern Wisconsin. Conservation Biology 2:348-358.

Anderson, R. C., S. S. Dhillion, and T. M. Kelley. 1996. Aspects of the ecology of an
invasive plant, garlic mustard (Alliaria petiolata), in central Illinois. Restoration
Ecology 4:181-191.

Anderson, R. C., and T. M. Kelley. 1995. Growth of garlic mustard (Alliaria petiolata) in
native soils of different acidity. Transactions of the ILlinois State Academy of
Science 88:91-96.

Augustine, D. J. 1997. Grazing patterns and impacts of white-tailed deer in a fragnented
forest ecosystem. MS. Thesis. The University of Minnesota, St. Paul, MN.

Bais, H. P., R. Vepachedu, S. Gilroy, R. M. Callaway, and J. M. Vivanco. 2003.
Allelopathy and exotic plant invasion: From molecules and genes to Species
interactions. Science 301:1377-1380.

Bangerter, E. B. 1985. New and interesting records of adventive plants from the
Auckland Institute and Museum Herbarium: 11. Records of the Auckland Institute
and Museum 22:41-45.

Baskin, J. M., and C. C. Baskin. 1992. Seed-germination biology of the weedy biennial
Alliaria petiolata. Natural Areas Journal 12: 191-197.

Blossey, B. 1999. Before, during and after: The need for long-term monitoring in
invasive plant species management. Biological Invasions 1:301-311.

Blossey, B., R. Casagande, L. Tewksbury, D. A. Landis, R. N. Wiedenmann, and D. R.
Ellis. 2001a. Nontarget feeding of leaf-beetles introduced to control purple
loosestrife (Lythrum salicaria L.). Natural Areas Journal 21:368-377.

Blossey, B., and R. Nt’itzold. 1995. Evolution of increased competitive ability in invasive
nonindigenous plants - a hypothesis. Journal of Ecology 83:887-889.

Blossey, B., V. Nuzzo, H. Hinz, and E. Gerber. 2001b. Developing biological control of
Alliaria petiolata (M. Bieb.) Cavara and Grande (garlic mustard). Natural Areas
Journal 21:357-367.

41

Blossey, B., L. C. Skinner, and J. Taylor. 2001c. Impact and management of purple
loosestrife (Lythrum salicaria) in North America. Biodiversity and Conservation

10:1787-1807.

Bohlen, P. J., S. Scheu, C. M. Hale, M. A. McLean, S. Migge, P. M. Groffinan, and D.
Parkinson. 2004. Non-native invasive earthworms as agents of change in northern
temperate forests. Frontiers in Ecology and the Environment 2:427-435.

Bossdorf, O., H. Auge, L. Lafuma, W. E. Rogers, E. Siemann, and D. Prati. 2005.
Phenotypic and genetic differentiation between native and introduced plant

populations. Oecologia 144:1-1 1.

Bossdorf, O., S. Schroder, D. Prati, and H. Auge. 2004. Palatability and tolerance to
Simulated herbivory in native and introduced populations of Alliaria petiolata
(Brassicaceae). American Journal of Botany 91:856-862.

Byers, D. L., and J. A. Quinn. 1987. The effect of habitat variation in Alliaria petiolata
on life-history characteristics. American Journal of Botany 74:647-647.

Call, L. J ., and E. T. Nilsen. 2003. Analysis of spatial patterns and spatial association
between the invasive tree-of-heaven (Ailanthus altissima) and the native black
locust (Robinia pseudoacacia). American Midland Naturalist 150:1-14.

Callaway, R. M., and W. M. Ridenour. 2004. Novel weapons: invasive success and the
evolution of increased competitive ability. Frontiers in Ecology and the
Environment 2:436-443.

Carlson, A. M., and D. L. Gorchov. 2004. Effects of herbicide on the invasive biennial
Alliaria petiolata (garlic mustard) and initial responses of native plants in a
southwestern Ohio forest. Restoration Ecology 12:559-567.

Cavers, P. B., M. I. Heagy, and R. F. Kokron. 1979. Biology of Canadian weeds. 35.
Alliaria petiolata (M Bieb) Cavara and Grande. Canadian Journal of Plant

Science 59:217-229.

Cipollini, D. 2002. Variation in the expression of chemical defenses in Alliaria petiolata
(Brassicaceae) in the field and common garden. American Journal of Botany

89: 1422-1430.

Cipollini, D., J. Mbagwu, K. Barto, C. Hillstrorrr, and S. Enright. 2005. Expression of
constitutive and inducible chemical defenses in native and invasive populations of
Alliaria petiolata. Journal of Chemical Ecology 31: 1255-1267.

Cruden, R. W., A. M. McClain, and G. P. Shrivastava. 1996. Pollination biology and
breeding system of Alliaria petiolata (Brassicaceae). Bulletin of the Torrey
Botanical Club 123:273-280.

42

Davalos, A., and B. Blossey. 2004. Influence of the invasive herb garlic mustard (Alliaria
petiolata) on gound beetle (Coleoptera : Carabidae) assemblages. Environmental
Entomology 33:564-576.

Davis, A. S., D. A. Landis, and B. Blossey. in review. Demogaphic models inform agent
selection for Alliaria petiolata (M. Bieb.) Cavara and Grande (garlic mustard).

Davis, A. S., D. A. Landis, J. A. Evans, and D. W. Schemske. 2005. Matrix population
models to inform A. petiolata biocontrol agent selection (poster). in USDA NRI
Biology of Weedy and Invasive Plants PI Meeting, Washington, DC.

Dhillion, S. S., and R. C. Anderson. 1999. Growth and photosynthetic response of first-
year garlic mustard (Alliaria petiolata) to varied irradiance. Journal of the Torrey
Botanical Society 126:9-14.

Drayton, B., and R. B. Primack. 1999. Experimental extinction of garlic mustard (Alliaria
petiolata) populations: implications for weed science and conservation biology.
Biological Invasions 1:159-167.

Durka, W., O. Bossdorf, and B. Gautschi. 2004. Isolation and characterization of
microsatellite loci in the invasive Alliaria petiolata (Brassicaceae). Molecular
Ecology Notes 4:173-175.

Durka, W., O. Bossdorf, D. Prati, and H. Auge. 2005. Molecular evidence for multiple
introductions of garlic mustard (Alliaria petiolata, Brassicaceae) to North
America. Molecular Ecology 14:1697-1706.

Gerber, E., and H. Hinz. 2005. Biology, impact and interactions of potential biocontrol
agents on garlic mustard. Pages 4-6 in L. C. Skinner, editor. Symposium on the
biology, ecology, and management of garlic mustard (Alliaria petiolata) and
European buckthom (Rhamnus cathartica). USDA Forest Service, University of
Minnesotta, St. Paul, MN.

Gerber, E., H. Hinz, N. Guazzone, J. McKenney, S. Michler, and M. Zuefle. 2002.
Biological control of garlic mustard, Alliaria petiolata (Bieb.) Cavara & Grande.
Annual report 2001. Unpublished Report, CABI Bioscience Switzerland Centre,
Delemont, Switzerland.

Gerber, E., H. L. Hinz, and G. Cortat. 2004. Biological control of garlic mustard, Alliaria
petiolata. Annual report 2003. Unpublished Report, CABI Bioscience
Switzerland Centre, Delemont, Switzerland.

Gerber, E., H. L. Hinz, and G. Cortat. 2005. Biological control of garlic mustard, Alliaria
petiolata. Annual report 2004. Unpublished Report, CABI Bioscience
Switzerland Centre, Delemont, Switzerland.

43

Gerber, E., H. L. Hinz, M. Schat, G. Corat, N. Guazzone, and L. Lovis. 2003. Biological
control of garlic mustard, Alliaria petiolata. Annual report 2002. Unpublished
Report, CABI Bioscience Switzerland Centre, Delemont, Switzerland.

Grant, D. W., D. P. C. Peters, G. K. Beck, and H. D. Fraleigh. 2003. Influence of an
exotic species, Acroptilon repens (L.) DC. on seedling emergence and gowth of
native gasses. Plant Ecology 166:157-166.

Hale, C. M., L. E. Frelich, and P. B. Reich. 2005a. Exotic European earthworm invasion
dynamics in northern hardwood forests of Minnesota, USA. Ecological
Applications 15:848-860.

Hale, C. M., L. E. Frelich, P. B. Reich, and J. Pastor. 2005b. Effects of European
earthworm invasion on soil characteristics in northern hardwood forests of

Minnesota, USA. Ecosystems 8:911-927.

Haribal, M., and J. A. A. Renwick. 1998. Isovitexin 6 "-O-beta-D-glucopyranoside: A
feeding deterrent to Pieris napi oleracea from Alliaria petiolata. Phytochemistry

47 :1237-1240.

Haribal, M., and J. A. A. Renwick. 2001. Seasonal and population variation in flavonoid
and alliarinoside content of Alliaria petiolata. Journal of Chemical Ecology

27:1585-1594.

Haribal, M., J. A. A. Renwick, and A. Attygalle. 1999. A new sinapoyl derivative of
isovitexin 6 "-O-beta-D-glucopyranoside from Alliaria petiolata. Journal of

Natural Products 62: 1 79-1 80.

Haribal, M., Z. C. Yang, A. B. Attygalle, J. A. A. Renwick, and J. Meinwald. 2001. A
cyanoallyl glucoside from Alliaria petiolata, as a feeding deterrent for larvae of
Pieris napi oleracea. Journal of Natural Products 64:440-443.

Hinz, H., and E. Gerber. 2005. Summary of host Specificity testing conducted in
Switzerland and Brassicaceae taxonomy. Pages 12-16 in L. C. Skinner, editor.
Symposium on the biology, ecology, and management of garlic mustard (Alliaria
petiolata) and European buckthom (Rhamnus cathartica). USDA Forest Service,
University of Minnesotta, St. Paul, MN.

Hinz, H. L., and E. Gerber. 1998. Investigations on potential biological control agents of
garlic mustard, Alliaria petiolata (Bieb.) Cavara & Grande Annual Report 1998.
Unpublished Report, CABI Bioscience Switzerland Centre, Delemont,

Switzerland.

Hinz, H. L., and E. Gerber. 2000. Investigations on potential biological control agents of
garlic mustard, Alliaria petiolata (Bieb.) Cavara & Grande Annual Report 1999.
Unpublished Report, CABI Bioscience Switzerland Centre, Delemont,

Switzerland.

Hinz, H. L., and E. Gerber. 2001. Investigations on potential biological control agents of
garlic mustard, Alliaria petiolata (Bieb.) Cavara & Grande Annual Report 2000.
Unpublished Report, CABI Bioscience Switzerland Centre, Delemont,
Switzerland.

Kalisz, S., T. Knight, L. Smith, J. Dunn, and J. Davis. 2003. Recipe for invasive spread:
Venison with garlic mustard. in Invasive Plants in Natural and Managed Systems:
Linking Science and Management in conjunction with the 7th International
Conference on the Ecology and Management of Alien Plant Invasions, Wyndham
Bonaventure Resort, Ft. Lauderdale, FL.

Katovich, E. J. S., R. L. Becker, and D. W. Ragsdale. 1999. Effect of Galerucella spp. on
survival of purple loosestrife (Lythrum salicaria) roots and crowns. Weed Science
47 :360-365.

Katovich, E. J. S., R. L. Becker, D. W. Ragsdale, and L. C. Skinner. 2005. Host range
testing of garlic mustard (Alliaria petiolata) biocontrol insects in Minnesota.
Pages 17-18 in L. C. Skinner, editor. Symposium on the biology, ecology, and
management of garlic mustard (A lliaria petiolata) and European buckthom
(Rhamnus cathartica). USDA Forest Service, University of Minnesotta, St. Paul,
MN. ‘

Katovich, E. J. S., D. W. Ragsdale, L. C. Skinner, and R. L. Becker. 2001. Effect of
Galerucella spp. feeding on seed production in purple loosestrife. Weed Science
49:190-194.

Kaufman, L. N., and D. A. Landis. 2000. Host specificity testing of Galerucella
calmariensis L. (Coleoptera: Chrysomelidae) on wild and ornamental plant
species. Biological Control 18:157-164.

Kerguélen, M. 1999. Index dynonymique de la flore de France.
http://www.dijon.inra.fr/flore-france/indexhtrn. Institute National de la Recherche

Agonorrrique. Accessed February 5, 2006.

KOurtev, P., W. Huang, and .l. Ehrenfeld. 1999. Differences in earthworm densities and
nitrogen dynamics in soils under exotic and native plant species. Biological
Invasions 1:237-245.

Landis, D. A., D. W. Schemske, and A. S. Davis. 2004. Using demogaphic models to
assess biocontrol of an invasive plant. USDA NR1 Proposal.

Landis, D. A., D. C. Sebolt, M. J. Haas, and M. Klepinger. 2003. Establishment and
impact of Galerucella calmariensis L. (Coleoptera : Chrysomelidae) on Lythrum
salicaria L. and associated plant communities in Michigan. Biological Control

28:78-91.

45

Louda, S. M., T. A. Rand, A. E. Amett, A. S. McClay, K. Shea, and A. K. McEachern.
2005. Evaluation of ecological risk to populations of a threatened plant from an
invasive biocontrol insect. Ecological Applications 15:234-249.

Luken, J. O., L. M. Kuddes, and T. C. Tholerneier. 1997. Response of understory species
to gap formation and soil disturbance in Lonicera maackii thickets. Restoration
Ecology 5:229-235. ‘

Luken, J. O., and M. Shea. 2000. Repeated prescribed burning at Dinsmore Woods State
Nature Preserve (Kentucky, USA): Responses of the understory community.
Natural Areas Journal 20:150-158.

Maerz, J. C., B. Blossey, J. K. Liebherr, and V. A. Nuzzo. 2002.
http://www.invasiveplants.net/impsal.htm Accessed January 24, 2006.

McCarthy, B. C. 1997. Response of a forest understory community to experimental
removal of an invasive nonindigenous plant (Alliaria petiolata, Brassicaceae).
Pages 117-130 in J. W. Thieret, editor. Assessment and Management of Plant
Invasions. Springer, NY.

McCarthy, B. C., and S. L. Hanson. 1998. An assessment of the allelopathic potential of
the invasive weed Alliaria petiolata (Brassicaceae). Castanea 63:68-73.

MDNR. 2006. Proposed 2006-2010 Deer Population Goals.
http://www.michigan.gov/dnr/0,l607,7-153-10363_10856_10905-129948--
,00.htrnl. Michigan Department of Natural Resources.

Meekins, J. F., H. E. Ballard, and B. C. McCarthy. 2001. Genetic variation and molecular
biogeogaphy of a North American invasive plant Species (Alliaria petiolata,
Brassicaceae). International J oumal of Plant Sciences 162:161-169.

Meekins, J. F ., and B. C. McCarthy. 1999. Competitive ability ofAlliaria petiolata
(garlic mustard, brassicaceae), an invasive, nonindigenous forest herb.
International Journal of Plant Sciences 160:743-752.

Meekins, J. F., and B. C. McCarthy. 2000. Responses of the biennial forest herb Alliaria
petiolata to variation in population density, nutrient addition and light availability.
Journal of Ecology 88:447-463.

Meekins, J. F ., and B. C. McCarthy. 2001. Effect of environmental variation on the
invasive success of a nonindigenous forest herb. Ecological Applications
11:1336-1348.

Meekins, J. F., and B. C. McCarthy. 2002. Effect of population density on the
demogaphy of an invasive plant (Alliaria petiolata, Brassicaceae) population in a
Southeastern Ohio forest. American Midland Naturalist 147 2256-278.

46

Myers, C. V., and R. C. Anderson. 2003. Seasonal variation in photosynthetic rates
influences success of an invasive plant, garlic mustard (Alliaria petiolata).
American Midland Naturalist 150:231-245.

Myers, C. V., R. C. Anderson, and D. L. Byers. 2005. Influence of shading on the gowth
and leaf photosynthesis of the invasive non-indigenous plant garlic mustard
Alliaria petiolata (M. Bieb) Cavara and Grande gown under Simulated late-
winter to mid-spring conditions. Journal of the Torrey Botanical Society 132: 1-
10.

Nuzzo, V. A. 1991. Experimental control of garlic mustard [Alliaria petiolata (Bieb.)
Cavara & Grande] in Northern Illinois using fire, herbicide, and cutting. Natural
Areas Journal 11:158-167.

Nuzzo, V. A. 1993a. Current and historic distribution of garlic mustard (Alliaria
Petiolata) in Illinois. The Michigan Botanist 32:23-33.

Nuzzo, V. A. 1993b. Distribution and spread of the invasive biennial Alliaria petiolata
(garlic mustard) in North America. Pages 137-145 in B. N. McKnight, editor.
Biological pollution: The control and impact of invasive exotic species:
proceedings of a symposium held at the University Place Conference Center,
Indiana University-Purdue University at Indianapolis on October 25 & 26, 1991.
Indiana Academy of Science, Indianapolis.

Nuzzo, V. A. 1993c. Natural mortality of garlic mustard (Alliaria petiolata (Bieb.)
Cavara & Grande) rosettes. Natural Areas Journal 13:132-133.

Nuzzo, V. A. 1994. Response of garlic mustard (A lliaria-Petiolata Bieb - Cavara and
Grande ) to summer herbicide treatment. Natural Areas Journal 14:309-310.

Nuzzo, V. A. 1996. Impact of dormant season herbicide treatment on the alien herb garlic
mustard (Alliaria periolata[Bieb.] Cavara and Grande) and goundlayer
vegetation. Transactions of the Illinois State Academy of Science 89:25-36.

Nuzzo, V. A. 1999. Invasion pattern of the herb garlic mustard (Alliaria petiolata) in high
quality forests. Biological Invasions 1:169-179.

Nuzzo, V. A. 2000. Element stewardship abstract for Alliaria petiolata (Alliaria
oflicinalis) garlic mustard. The Nature Conservancy.

Nuzzo, V. A., W. McClain, and T. Strole. 1996. Fire impact on goundlayer flora in a
sand forest 1990-1994. American Midland Naturalist 136:207-221.

Ortega, Y. K., D. E. Pearson, and K. S. McKelvey. 2004. Effects of biological control
agents and exotic plant invasion on deer mouse populations. Ecological
Applications 14:241-253.

47

Pearson, D., and R. Callaway. 2003. Indirect effects of host-specific biological control
agents. Trends in Ecology and Evolution 18:456-461.

Pearson, D. E., and R. M. Callaway. 2005. Indirect nontarget effects of host-specific
biological control agents: Implications for biological control. Biological Control
35:288-298.

Peterson, A. T., M. Papes, and D. A. Kluza. 2003. Predicting the potential invasive
distributions of four alien plant species in North America. Weed Science 51:863-
868.

Pichtel, J ., K. Kuroiwa, and H. T. Sawyerr. 2000. Distribution of Pb, Cd and Ba in soils
and plants of two contaminated sites. Environmental Pollution 110: 1 71-1 78.

Pirnentel, D., L. Lach, R. Zuniga, and D. Morrison. 2000. Environmental and economic
costs of nonindigenous species in the United States. Bioscience 50:53-65.

Porter, A. 1994a. Implications of introduced garlic mustard (Alliaria petiolata) in the
habitat of Pieris virginiensis (Pieridae). Journal of the Lepidopterists' Society
48: 1 71-1 72.

Porter, A. 1994b. Native pierine butterfly (Pieridae) adapting to naturalized crucifer?
Journal of the Lepidopterists' Society 48:168-170.

Prati, D., and O. Bossdorf. 2004. Allelopathic inhibition of germination by Alliaria
petiolata (Brassicaceae). American Journal of Botany 91:285-288.

Pyle, L. L. 1995. Effects of disturbance on herbaceous exotic plant-species on the
floodplain of the Potomac River. American Midland Naturalist 134:244-253.

Rejmanek, M. 2000. On the use and misuse of transition matrices in plant population
biology. Biological Invasions 2:3 15-3 17.

Renwick, J. A. A. 2002. The chemical world of crucivores: lures, treats and traps.
Entomologia Experimentalis Et Applicata 104:35-42.

Renwick, J. A. A., and K. Lopez. 1999. Experience-based food consumption by larvae of
Pieris rapae: addiction to glucosinolates? Entomologia Experimentalis Et
Applicata 91:51-58.

Renwick, J. A. A., W. Q. Zhang, M. Haribal, A. B. Attygalle, and K. D. Lopez. 2001.
Dual chemical barriers protect a plant against different larval stages of an insect.
Journal of Chemical Ecology 27:1575-1583.

Roberts, H. A., and J. E. Boddrell. 1983. Seed survival and periodicity of seedling
emergence in 8 species of cruciferae. Annals of Applied Biology 103:301-304.

48

Roberts, K. J., and R. C. Anderson. 2001. Effect of garlic mustard Alliaria petiolata
(Beib. Cavara & Grande) extracts on plants and arbuscular mycorrhizal (AM)
fimgi. American Midland Naturalist 146: 146-1 52.

Rooney, T. P. 2001. Deer impacts on forest ecosystems: a North American perspective.
Forestry 74:201-208.

Schwartz, M. W., and J. R. Heim. 1996. Effects of a prescribed fire on degaded forest
vegetation. Natural Areas Journal 16:184-191.

Sebolt, D. C., and D. A. Landis. 2002. Neonate Galerucella calmariensis (Coleoptera :
Chrysomelidae) behavior on purple loosestrife (Lythrum salicaria) contributes to
reduced predation. Environmental Entomology 31:880-886.

Smith, G. R., H. A. Dingfelder, and D. A. Vaala. 2003. Effect of plant size and density on
garlic mustard reproduction. Northeastern Naturalist 10:269-276.

Smith, R. G., K. L. Gross, and S. Januchowski. 2005. Earthworms and weed seed
distribution in annual crops. Agicultural Ecosystems and Environment 108:363-
637.

Society for Economic Botany. 1998. Information request - Garlic mustard Alliaria
petiolata (Bieb.) Cavara & Grande: a crop with a future?
http://www.rbgkew.org.uk/SEB-UK/news0698.htrn. in Minutes of the Annual
Meeing of the Society for Economic Botany UK Chapter, Institute of
Archaeology, University College London.

Solis, K. 1998. Update: new results indicate flowering garlic mustard should be bagged
and destroyed (Wisconsin). Restoration and Management Notes 16:223-224.

Stiling, P., D. Moon, and D. Gordon. 2004. Endangered cactus restoration: Mitigating the
non-target effects of a biological control agent (Cactoblastis cactorum) in Florida.
Restoration Ecology 12:605-610.

Stinson, K., L. Durbin, K. Sylvan, and F. Lowenstein. 2005. Responses of a New
England forest community to increasing levels of invasion by garlic mustard. in
Annual meeting of the Ecological Society of America, Montreal, Canada.

Stinson, K., and J. N. Klironomos. 2005. Exotic plant invasion degades local
mycorrhizal association, alters community succession and limits restoration. in
Annual meeting of the Ecological Society of America, Montreal, Canada.

Susko, D. J ., and L. Lovett-Doust. 1998. Variable patterns of seed maturation and
abortion in Alliaria petiolata (Brassicaceae). Canadian Journal of Botany-Revue
Canadienne De Botanique 76:1677-1686.

Susko, D. J., and L. Lovett-Doust. 1999. Effects of resource availability, and fruit and
ovule position on components of fecundity in Alliaria petiolata (Brassicaceae).
New Phytologist 144:295-306.

49

 

 

Susko, D. J ., and L. Lovett-Doust. 2000a. Patterns of seed mass variation and their effects
on seedling traits in Alliaria petiolata (Brassicaceae). American Journal of Botany
87:56-66.

Susko, D. J ., and L. Lovett-Doust. 2000b. Plant-size and fruit-position effects on
reproductive allocation in Alliaria petiolata (Brassicaceae). Canadian Journal of
Botany-Revue Canadienne De Botanique 78:1398-1407.

The Nature Conservancy. 2002. http://tncweeds.ucdavis.edu/3uccess/ak001.html. in.

Trimbur, T. J. 1973. The ecological life history of Allaria ofiicinallis, a deciduous forest
"weed". MS. Thesis. The Ohio State University, Columbus, OH.

USDA-NRCS. 2006. The PLANTS Databse, Version 3.5 (http://plants.usda. gov). Data
compiled from various sources by Mark W. Skinner. National Plant Data Center,
Baton Rouge, LA 70874-4490 USA.

Vaughn, S. F ., and M. A. Berhow. 1999. Allelochernicals isolated from tissues of the
invasive weed garlic mustard (Alliaria petiolata). Journal of Chemical Ecology
25:2495-2504.

Webb, S. L., T. H. Pendergast, and M. E. Dwyer. 2001. Response of native and exotic
maple seedling banks to removal of the exotic, invasive Norway maple (Acer
platanoides). Journal of the Torrey Botanical Society 128: 141-149.

Welk, E., K. Schubert, and M. H. Hoffrnann. 2002. Present and potential distribution of
invasive garlic mustard (Alliaria petiolata) in North America. Diversity and
Distributions 8:219-233.

Weston, L. A., and S. 0. Duke. 2003. Weed and crop allelopathy. Critical Reviews in
Plant Sciences 22:367-389.

Wilcove, I). S., D. Rothstein, J. Dubow, A. Phillips, and E. Losos. 1998. Quantifying
threats to imperiled Species in the United States. Bioscience 48:607-615.

Winterer, J., M. C. Walsh, M. Poddar, J. W. Brennan, and S. M. Primak. 2005. Spatial
and temporal segegation of juvenile and mature garlic mustard plants (Alliaria
petiolata) in a central Pennsylvania woodland. American Midland Naturalist
153:209-216.

Yahner, R. H. 1998. Butterfly and skipper use of nectar sources in forested and
agicultural landscapes of Pennsylvania. Journal of the Pennsylvania Academy of
Science 71:104-108.

Yost, S. E., S. Antenen, and G. Hartvigsen. 1991. The Vegetation of the Wave Hill
Natural Area, Bronx, New York. Bulletin of the Torrey Botanical Club 118:312-
325.

50

 

 

Number of Papers Published

 

 

0 l_ _ l
1985 1990 1995 2000 2005
Year of Publication

Figure 1. Results of an 181 Web of Science query using the search terms "invasive
species OR biotic invasion OR biological invasion". Search performed on January 24,
2006.

 

51

 

 

Chapter 2

PRE-RELEASE MONITORING OF ALLMRIA PE TIOLA TA [GARLIC
MUSTARD (M. BIEB) CAVARA AND GRAN DE] INVASIONS AND IMPACTS
OF EXT ANT NATURAL ENEMIES IN SOUTHERN MICHIGAN F ORESTS

Jeffrey Adam Evans

52

ABSTRACT
When conventional controls strategies fail, managers of invasive species may
consider classical biological control approaches as potentially safe, cost effective, and
long-term solutions. To justify releasing natural enemies and to broadly improve the rigor
and safety of biological control, some authors have called for monitoring target species’
populations prior to releasing natural enemies. The data collected through such activities
permit evaluation of the invader’s impacts on native communities and create a reference
point for evaluating the effectiveness of biocontrol agents if they are later released. I
collected baseline data on populations of the invasive weed Alliaria petiolata (garlic
mustard) across its range in southern Michigan in advance of a classical biological
control progam. Alliaria petiolata populations were shown to be expanding at all Sites
studied with 59% of initially uninvaded sampling quadrats becoming invaded after two
years. 84.5% of the quadrats with A. petiolata showed evidence of herbivore browsing or
other damage. However, damage estimates were very low (2.9% of leaf area damaged)
and were not correlated with rosette or seedling survival or with fecundity. The data
presented here paint a portrait of an invasive weed that is spreading rapidly into new
habitats and is unchecked by extant natural enemies. Given the rapid expansion of A.
petiolata and the lack of significant herbivores, increasing herbivore damage with
introduced natural enemies may present a new opportunity to Slow or reverse the spread
of this invasive plant. If natural enemy agents are released in the future, these data will

provide a critical benchmark for evaluating their performance.

53

INTRODUCTION

The widespread ecological (W ilcove et al. 1998) and economic (U .S. Congess
1993, Pirnentel et al. 2000) impacts of invasive species on both natural and managed
systems have lead many managers to prioritize control of non-indigenous species. As
many as 5000 non-indigenous plant species are naturalized in the United States (Pimentel
et al. 2000) and present a set of unique management challenges and opportunities.
Conventional control methods, such as herbicide spraying and mechanical removal, are
not practical or effective on some target weeds. Practitioners of classical weed biological
control consider introductions of natural enemies to be an environmentally safe and
potentially effective management option. However, the rigor and safety of classical
biological control as currently practiced have been challenged by some authors
(Simberloff and Stiling 1996, McEvoy and Coombs 1999, Louda et al. 2003a, Louda et
al. 2003b, e. g. Louda et al. 2005, Pearson and Callaway 2005).

Often cited as problematic are the limited abilities of biological control
practitioners to justify implementing biological control progams for specific targets and
to evaluate the relative success of progams once established. This stems in part from a
historical lack of sufficient pre-release data collection as well as fiom a lack of post-
release follow-up monitoring (Blossey 1999). Blossey (1999) addresses these concerns
and points out that, for many invasive weed biocontrol progams, the evidence provided
to justify the release of new biocontrol agents has been largely anecdotal. There is a need
for biocontrol researchers to define pre and post agent-release monitoring goals that will
facilitate identification of negative impacts on invaded-community structure and/or

function as well as evaluation of agent performance and community response to weed

54

suppression. The biological control progam for Lythrum salicaria L. (purple loosestrife)
is an example of a well coordinated, successful post-release monitoring progam that was
based in part on Blossey’s (1999) recommendations and has facilitated follow-up
evaluation of the progam (Kaufman and Landis 2000, Blossey et al. 2001b, Katovich et

al. 2001, Landis et al. 2003).

STUDY SPECIES

Alliaria petiolata (M. Bieb) Cavara and Grande (garlic mustard) is a biennial
invasive weed of European origin. First recorded on Long Island, New York in 1868
(Nuzzo 1993), it is now widely distributed and invasive in North America (Nuzzo 1993,
2000). It is notable among invaders in its ability to penetrate high quality forest
understories as well as disturbed areas. Previous studies indicate that A. petiolata gowth
rates and fecundity are positively correlated with light availability (Meekins and
McCarthy 2000) and that A. petiolata responds positively to the formation of canopy gaps
(Luken et al. 1997). Current research suggests that exotic earthworms which increase
litter cycling rates may be correlated with increased A. petiolata abundance (Bohlen et al.
2004, Hale et al. 2005, CM. Hale personal communication 2003). Prati and Bossdorf
(2004) demonstrated that A. petiolata may have allelopathic properties which could be
mediated through suppression of arbuscular mycorrhizal fungi (Wolfe and Klironomos
2005).
I Alliaria petiolata seedlings emerge at high densities in early spring and gow over
the summer to form low rosettes of petiolate leaves. Seedling mortality over the summer
is high with fewer than 20% of seedlings surviving to the rosette stage in southern

Michigan in 2005 (Davis et al. 2005). Rosettes overwinter as geen plants and bolt,

55

flower (henceforth “adults” , senesce, and set seeds in late spring of the second year.
Overwintering rosette survival in southern Michigan ranged from 52 — 89% from 2004 —
2005 (Davis et al. 2004). Seeds are produced in siliques along the upper stem and are
released fi'om mid summer though mid autumn. Seeds can remain viable in the soil for at
least eight years (Nuzzo and Blossey unpublished data). The longevity of the seedbank
dictates that any effective control efforts will have to be sustained over many years until
the seed supply is exhausted. For all but the smallest infestations, this requirement is not
practically achievable for most managers.

A search for suitable biological control agents for A. petiolata was initiated in
1998 (Hinz and Gerber 1998, Blossey et al. 2001a) with efforts now focused on four
weevils in the genus Ceutorhynchus (Coleoptera: Curculionidae) that target multiple
stages in A. petiolata’s life cycle (Hinz and Gerber 2000, 2001, Gerber et a1. 2002,
Gerber et al. 2003, Gerber et al. 2004). Biological control agents are not yet available for
control of A. petiolata in North America. Thus, it is desirable to collect data on the target

weed and invaded communities in advance of the anticipated natural enemy releases.

OBJECTIVES

In 2003 permanent sampling quadrats were established at eight forests in southern
Michigan where A. petiolata occurred. My objectives were to (1) describe the study sites
and invaded communities prior to any biocontrol agents releases, (2) determine whether
Michigan A. petiolata populations are spreading within infested sites, and (3) measure the
degee to which existing herbivores are impacting A. petiolata populations. These data
will contribute to any future assessment of natural enemy releases. If biological control

agents for A. petiolata are approved for release in the firture, initial test releases may be

56

made at a subset of these sites to allow comparisons of pre and post-release community
dynamics and to evaluate the effectiveness of the agents. Here I present data on

populations of A. petiolata prior to the introduction of insect biological control agents.

METHODS AND MATERIALS
SITE SELECTION

Lab personnel established eight study sites within A. petiolata’s primary range in
Michigan’s southern Lower Peninsula (Landis et al. 2004). Criteria for site selection
included (1) forested lands > 2 ha in extent, (2) under state, federal, or other long-term
conservation management, (3) on which A. petiolata populations have been established
for at least four years, and (4) with protection fiom future disturbance or A. petiolata
management for at least ten years. In spring of 2003 we recorded GPS coordinates for
each site. We then marked 10 permanent 0.5 m2 sampling quadrats (0.5 x 1 m) along each
of two parallel, 100 m long transects spaced 10 m apart at seven sites and a single 200 m
transect with quadrats spaced 10 m part at the eighth site (Russ Forest) for a total or 20
quadrats per site. Site inventories included data on forest type (MN F1 2003), maturity
(diameter at breast height of principal overstory trees), and understory composition.
Because accurate records of species composition were not kept at these sites before the
initiation of this study, it is not possible to determine exactly how long A. petiolata had
been present at any of them prior to 2003, although the extent of the invasions and

anecdotal evidence fi'om managers indicated that they met the criteria listed above.

57

ALLIARIA PE TIOLA TA EVALUATIONS

I collected data on A. petiolata distribution and abundance in accordance with a
nationally standardized protocol (Nuzzo and Blossey unpublished). In spring (June) and
fall (Sept. — Nov.) of 2003 — 2005 I visited each site and recorded data from each quadrat
including: vegetation cover (A. petiolata total, A. petiolata by adult, seedling, and rosette
stage plants, total non-A. petiolata vegetation and non A petiolata vegetation by species),
counts of A. petiolata adults, seedlings and rosettes, percent cover of substrate (bare soil,
leaf litter, woody debris, and rock sum to 100%), and litter depth (cm). I recorded
damage to A. petiolata plants as the estimated percent of leaf area removed and identified
nine categories of damage to A. petiolata as either present or absent in each quadrat (leaf
mining, windowpaning, edge feeding, holes, spittle bug, scale damage, browse, disease,
and other). Finally, I recorded the height of and the number of siliques on each mature
second year plant during the spring sampling period.

In contrast to the methods outlined by Nuzzo and Blossey (unpublished), not all
sampling quadrats at each site contained A. petiolata at the initiation of the study. Rather,
the transects traversed the A. petiolata invasion front where possible. This was done to
allow us to measure spatial spread of A. petiolata populations within sites. At one site
(F emwood) this was not possible as all 20 quadrats there contained A. petiolata from the

outset of the study. Plant community data are discussed in Chapter 3.

SITE DESCRIPTIONS
The Michigan Natural Features Inventory (MN F I) has identified 74 community-

types that occur in Michigan (MNFI 2003). I used data on the identities, sizes, and

58

 

abundances of the principal canopy trees as well as physical features of the Sites and the
inventories of all gound-layer vascular plant species that occurred in the sampling
quadrats fiom June 2003 to October 2005 to describe each site in terms of the MNFI
community types. The General Land Office (GLO), which was established by the United
States federal government in 1785, systematically surveyed Michigan from 1816 through
1856 and made detailed records of soils, water resources, forests, and other natural
features (MN F1 2005). The MNF I has interpreted the GLO data and created maps of
early nineteenth century Michigan vegetation (MN PI 2005) with which I compared the
current communities. In the Site descriptions I have include an overview of the site
topogaphy, any important known or probable disturbances, important overstory and
understory plant species, and the native species and total species richness that I observed
in the sampling quadrats.

The eight sites selected are distributed across the southern four tiers of Michigan
counties and represent a diverse assortment of A. petiolata-invaded forest types ranging
fiom Southern Floodplain Forest to Dry Southern Forest (Figure 2, Table 1). A summary
of the data used in the analyses that follow is presented in Table 2 and Appendices 1 and
2.

Study Sites

Box Woodlot: Box Woodlot is an isolated mesic southern forest (MNFI 2003)
surrounded by crop fields with a gavel road along one edge. The overstory is dominated
by large Acer saccharum (sugar maple) and A. saccharinum (silver maple) with F agus
grandifolia (American beech) and A. rubrum (red maple) occuning less frequently.
Vegetation circa 1800 data shows the site as shrub swamp/emergent marsh and mesic

southern forest (MNFI 2005). Anthropogenic influences such as isolation, selective

59

cutting and drainage have affected changes in commmrity structure. Alliaria petiolata is
present throughout the site. Box Woodlot has the lowest plant Species richness of the
eight study sites with 17 species documented in the sampling quadrats, 16 of which are
native.

F emwood: The F emwood Botanic Garden Site is a mature dry-mesic southern
forest (MNFI 2003) adjacent to an open community of old fields and restored native
prairie. The overstory is dominated by Quercus alba (white oak), Q. rubra (red oak), Q.
velutina (black oak), and Prunus serotina (black cherry). Prior to settlement the site was
broadly classified as mesic southern forest (MNFI 2005). The sampling transects traverse
several steep, minor drainages. The transects are approximately perpendicular to the
primary slope of the hill and parallel to the forest edge 10 to 20 meters away. Alliaria
petiolata abundance is geatest along the forest boundary but dense populations persist
throughout the interior as well. Frequent deer trails, foot paths, and down-slope drainage
are likely means of A. petiolata seed dispersal into the forest interior. This Site has
intermediate species richness with 43 species occurring in the quadrats, 41 of which are
native. Alliaria petiolata had the largest spatial distribution at this site over the whole
study period and occurred in 20/20 sampling quadrats.

F ort Custer: Fort Custer is a gently sloping dry-mesic southern to dry southern
forest (MN F1 2003) whose canopy is dominated by large Q. velutina, Q. alba and Carya
ovata (shagbark hickory) and borders previously disturbed areas along a two-track
military access road. Pre-settlement vegetation surveys indicate the site as former mixed
oak savanna (MNF I 2005) the largest oaks still showing an open-gown canopy structure.

Suppression of fire has transformed the site to closed-canopy forest and led to loss of

60

understory prairie species, although infrequent savanna-like openings persist at the site.
Alliaria petiolata abundance is heaviest along the forest edge near the access road and in
the openings but penetrates the entire forest. Military vehicle and foot traffic along the
road and deer trails through the forest interior appear to be the primary means of A.
petiolata dispersal. Fort Custer had the second highest species richness of the eight study
sites with 58 Species recorded in the quadrats, 52 of which are native. Invasion by
Berberis thunbergii (Japanese barberry), Rosa multiflora (multiflora rose), and Lonicera
spp. (bush honeysuckles) is also occuning at this site.

Ives Road: The Ives Road Fen Preserve is a 267 hectare property owned and
managed by The Nature Conservancy that contains a mix of fen, restored native prairie,
southern floodplain forest, and dry-mesic southern forest habitats (MN F1 2005). The
study transects quarter a steeply sloping ecotone of mature dry-mesic southern forest that
separates the upland restored prairie from the southern floodplain forest. Vegetation circa
1800 data show the lowlands as mixed hardwood swamp bordered by black oak barrens
(MNF I 2005) The ecotone is too narrow to be resolved on the circa 1800 vegetation
maps. However, the dominant canopy trees include Q. alba, Q. rubra and C. ovata
indicating its coarse soil structure.

Alliaria petiolata stands appear most robust along the forest/prairie interface and
decline in stature and density as the transects descend towards the bottomlands. Alliaria
petiolata populations in the bottomlands are characterized by large, robust plants at
medium to high densities. The population on the floodplain is addressed in a separate
study (Davis et al. 2005). I recorded 48 species in the sampling quadrats at this site, 44 of

which are native. Alliaria petiolata is the most abundant non-native invasive plant

61

Species at the Ives Road site, but Lonicera spp., R. multiflora, Hesperis matronalis
(dame’s rocket), Euonymus alata (winged burning bush), and Ligustrum spp. (privet) also
occur.

Lux Arbor: Lux Arbor is characterized as a mature dry-mesic southern forest in a
bottomland gading up a hill into dry southern forest (MNFI 2003). The canopy is
dominated by mature Quercus velutina and Q. rubra with an understory of modest
species richness including several Rubus species (brambles) and Phytolacca americana
(pokeweed). Circa 1800 vegetation data shows the site classified as mixed oak savanna
(MNF I 2005). In spring 2005 logging activities led to major changes in canopy density
and woody debris at gound level. Most large trees were cleared resulting in geatly
increased light availability and substantial soil disturbance. In late spring 2005 I located
and re-marked quadrats damaged by logging equipment and continued with normal
sampling. Alliaria petiolata occurs throughout this site but is most abundant in the more
mesic lowland and at the crest of the hill than along the hillside. At Lux Arbor 38 species
have been recorded in the sampling quadrats, 34 of which are native. Continued
monitoring at this site may reveal information on A. petiolata’s response to disturbance
and changes in light availability at the population level.

Pinckney: The sampling area at Pinckney State Recreation Area is located in a
well-drained, mature, dry-mesic southern forest on a gently sloping hillside. Canopy
dominants in this system are large Quercus rubra, Q. alba, and Carya ovata, with a
diverse understory community including Cornusflorida and C. foemina (flowering and
gay dogwoods), Amelancheir spp. (serviceberry), and Sassafras albidus (sassafi'as). The

two transects are aligned transverse to the slope of the hill. Alliaria petiolata is present

62

throughout the Site, although abundance is heterogeneous and appears to track animal and
foot trails which likely serve as dispersal corridors. Human activities at Pinckney include
mountain biking and hiking along a trail approximately 10 -— 30 m from the study site and
hunting which draws limited foot traffic directly through the sampling area. At the
Pinckney Site 39 species occur in the quadrats, 38 of which are native.

Russ Forest: Russ Forest is an old gowth dry southern forest (MNFI 2003)
dominated by Quercus alba and Q. velutina. Large Acer saccharum, Prunus serotina, and
Liriodendron tulilpifera (tulip tree) trees are also present and A. saccharum constitutes
the majority of sub-canopy trees. Circa 1800 vegetation maps show the site as mixed oak
savanna and mesic southern forest (MNFI 2005). The site topogaphy is flat and level.
Two roads border Russ Forest along its northern and western edges. Alliaria populations
are well established and robust along the forest border to the north and diminish in
evenness and density towards the forest interior. This site has a single, 200 m long
transect consisting of 20 evenly spaced sampling quadrats that run parallel to the road
which is approximately 15 - 20 m to the north. Russ Forest has lower species richness
with 32 species observed in the sampling quadrats, 31 of which are native.

High winds produced during a storm event in the spring of 2004 caused a major
blow-down in the northwest corner of the forest. The core blow-down area was
completely deforested and is approximately 2 ha in Size. Six quadrats at the western end
of the transect are either in or near large treefall gaps created by the storm. Subsequent
salvage logging operations conducted with horse teams and conventional skidders

resulted in soil disturbance. Forest managers instructed logging crews to avoid the A.

63

petiolata study area and established a no-entry perimeter that extended approximately 20
m from the sampling areas for this and a separate study in the same forest.

Shiawassee: The Shiawassee YMCA Camp site is classified as southern
floodplain forest (MN F I 2003) and is located on the floodplain of the Shiawassee River.
Two parallel transects run fi'om the first bottom of the river plain, which is dominated by
Acer saccharinum and F raxinus pennsylvanica and has a relatively open canopy, up a
small rise to the second bottom of the river valley (Tepley et al. 2004), which is
dominated by Juglans nigra (black walnut) and adjoins a two-track service road/foot path
and Pinus sylvestris (scotch pine) plantation. Prior to settlement the site was classified as
mesic southern forest (MNF I 2005), although it is doubtful that the actual floodplain
would have supported that community type. Alliaria petiolata density is high throughout
the site. Plant densities and adult plant sizes are exceptionally high in the second and first
bottom floodplain areas, respectively. Second year A. petiolata plants on the first bottom
were typically multi-stemmed and produced high numbers of seeds. However, the lowest
areas, in which geater than 50% of the quadrats are located, are subject to periodic
flooding. Flooding eliminated all seedlings and new rosettes fi'om the first bottom areas
in the late Spring of 2004. Second year plants had already flowered and produced seeds,
although they had all been knocked down by the flood. I was able to measure heights and
estimate fecundity of the downed second year plants, although this required handling
plants to separate and measure matted stems. Shiawassee had the highest species richness
of all eight sites. I identified 59 species in the sampling quadrats at Shiawassee, 53 of

which were native.

64

ANALYSES OF DATA
Spread of Alliaria petiolata

A portion of the sampling quadrats at seven of the eight Sites had not yet been
invaded when the study was initiated in 2003. I coded each quadrat as either invaded or
uninvaded for each sampling period based on the presence of absence of live A. petiolata
plants. I tested for linear trends in the number of invaded quadrats per Site over time with
a repeated measures general linear model in SAS version 8.2 using the REPEATED
command in PROC GLM (SAS Institute 2001). I tested assumptions of compound
symmetry with Mauchly’s sphericity test applied to orthogonal components and
evaluated linear trends pending those findings. I tested the sigrificance of changes in the
mean number of invaded quadrats within-sites using Dunn-Sidak adjusted (Gotelli and
Ellison 2004) pairwise comparisons in SYSTAT version 11.0 (SYSTAT Inc. 2004). Both
first and second year A. petiolata plants were present during spring sampling, but second
year plants senesced each year prior to fall sampling. Because first and second year plants
were often spatially segegated, the fall data do not reflect the full distribution of A.
petiolata within each site. I used only spring data in the spatial-Spread analysis for this

reason, although I present the fall data gaphically.

Estimation ofAlliaria petiolata F ecundity

To estimate fecundity of A. petiolata plants, I collected 130 mature plants from
six locations in southern Michigan (Edward Lowe Foundation, Cassopolis; Gasinski
Farm, Springville; Holland State Park, Holland; Johnson State Park, Wyoming; Rose

Lake Wildlife Management Area, East Lansing; Shiawassee YMCA Camp, Bancroft) and

65

measured the height and number of siliques on each. I dissected the seeds from each plant
and counted them using an automated seed counter (SeedBuro model 801 Count-A-Pac

Seed Counter ®, SeedBuro Equipment Co., Chicago, IL). Linear regession of number of
seeds x plant'1 on number of siliques x plant'l (R Development Core Team 2004) allowed

fecundity estimates for plants with known numbers of siliques.

Calculation of Survival Probabilities

I calculated survival probabilities for seedling to rosette (“seedling survival”) and
rosette to adult plant (“rosette survival”) transitions for A. petiolata plants in each
sampling quadrat at each site. Seedling survival is expressed as the number of seedlings
observed during the spring sampling period divided into the number of rosettes observed
during the fall sampling period of the same year, giving the proportion of seedlings that
survived the summer. Rosette survival was similarly calculated by dividing the number of
rosettes observed during the fall sampling period into the number of flowering adult
plants observed during the spring of the following year.

Seedling mortality extends fiom the beginning of the germination period in late
March through the summer (Evans unpublished data). My sampling methods captured the
number of seedlings present during a single visit but did not account for seedling
mortality prior to spring sampling or germination and mortality of additional seeds
between spring and fall sampling. Thus, the estimates of seedling survival are useful for
between-site comparisons but are not true estimates of A. petiolata demogaphic
parameters. Because the study included three summers and two winters, there are three

estimates of seedling survival but only two estimates of rosette survival.

66

Sampling Error

I detected two forms of observational error in my data. There were 16 cases where
fewer seedlings were recorded in spring than the number of rosettes observed in fall and
nine similar cases where fewer rosettes were observed in fall than flowering plants the
following spring, which generated survival probabilities geater than one. Also there were
20 cases where rosettes were recorded where no seedling had been recorded in the spring
and nine cases where flowering plants were observed where no rosettes had been
recorded the previous fall (divide by zero error). These errors most often occurred where
A. petiolata density was lowest and the overlooked plant(s) represented a geater
proportion of the quadrat total. These 54 observations were omitted fi'om analyses. Future
estimation of sampling error may allow correction of these observations and allow

accounting for future errors.

Herbivore Impacts on Alliaria petiolata

I tested for impacts of herbivore damage on A. petiolata per capita fecundity and
both seedling and rosette survival using regession tree analyses. Regession trees are
distribution-free multivariate analyses that allow specification of both categorical and
continuous independent variables to predict continuous dependent variables. I specified
the presence or absence of the nine categories of damage to A. petiolata, estimated
percent leaf damage to A. petiolata, site, and year as independent variables in each
regession tree. I also included quadrat level species richness as an independent variable
to explore whether it was more predictive of survival and fecundity than herbivores or

other damage sources. I used data on species richness and damage to rosettes in fall to

67

separately predict overwintering rosette survival and fecundity in the following spring
which allowed for two winters’ data to be included (fall 2003 -— spring 2004 and fall 2004
- spring 2005). I used spring data to again identify predictors of fecundity and to identify
predictors of seedling survival.

The regession tree algorithm uses pre-specified criteria to divide the dataset into
increasingly homogeneous subgoups as measured in the independent variable and
reports the mean, standard deviation, and number of observations in each subgoup. The
proportion of reduction in error (PRE) is a measure of model fit equivalent to an R2
statistic that describes the fit of the overall model and the improvement in model fit
contributed by each split in the data. The stopping and splitting criteria can be adjusted to
avoid over-specification of the model and maximize interpretability of the final tree. In
each regession tree analysis I set the maximum number of splits to five, the minimum
PRE for each split to 0.05, the minimum proportion of the dataset partitioned at each Split
to 0.05, and the minimum number of observations in each terminal goup to five. The
data are split until the maximum number of splits is reached or until further splits do not

meet the other three stopping criteria. All errors shown are i one standard error.

RESULTS AND DISCUSSION
EXPANSION OF ALLIARIA PE TIOLA TA WITHIN SITES

The spatial distribution of A. petiolata increased at seven of the eight sites from
2003 to 2005 (Figure 3). At the Femwood Site all 20 quadrats were originally invaded,
and detection of spread was not possible. All 20 quadrats remained invaded for the

duration of the study at this site, and Femwood was thus excluded from analyses of

68

spread. In the spring of 2003, the number of sampling quadrats (n = 20) that contained A.
petiolata at the seven other sites ranged fi'om 7 — 18 (mean = 11.9) (Table 3). By spring
of 2004, this range had shifted to 11 — 18 (mean = 15.9) quadrats invaded per site and to
11 - 20 (mean = 16.4) by spring of 2005. Across these seven sites, the mean percent
increase in the number of quadrats invaded fiom spring of 2003 to spring of 2005 was
45.9 d: 12.1% (range 5.6 — 100%). An average of 59.0 i 9.0% (range 30.8 — 100%) of
initially uninvaded quadrats became invaded over this same period (Table 3).

The change in the number of invaded quadrats per Site over time was sigrificant
in a repeated measures analysis (Table 4). The assumptions of sphericity and
homogeneity of variance were both satisfied (Mauchly’s criterion = 0.8143, df= 2, x2 =
1.0269, P = 0.5984, Levene’s Test P = 0.4238). The main effect of year on the number
of invaded quadrats was sigrificant indicating a change in the distribution of A. petiolata
within sites over time (F2, 12 = 11.8575, P = 0.0014). Although change in the number of
invaded quadrats per site was not sigrificant during either one-year time step, over the
two year period fiom spring 2003 to Spring 2005 the increase was sigrificant and positive
(Table 5, D-S adjusted P = 0.0154) at 45.9% increase per 2 years. While it would be
desirable to estimate the rate of A. petiolata spread either within sites or across the
landscape, these data are not suited to that purpose. Nuzzo’s (1999) study of A. petiolata
spread in Illinois concluded that populations expanded at a mean rate of approximately
5.4 m/year, which may be similar to the rate of spread at some of these sites and may be
lower than the rates I saw at others.

These findings offer quantitative support for the frequent observation that A.

petiolata populations almost invariably expand within Sites once established. Populations

69

do fluctuate in density from year to year (Figure 4) which may result from density
dependent effects, competition between first and second year plants (W interer et al.
2005), or response to variable environmental conditions and interactions with the
receiving community. The sharp decline in A. petiolata abundance at Shiawassee in fall
of 2004 resulted fiom the drowning of most seedlings during the flood that spring. This
was offset by the large emergence of seedlings the following year. Seasonal variation in
abundance (T able 2) is primarily an artifact of A. petiolata’s biennial life cycle. Both
seedlings and adults are present during the spring sampling period, but adults senesce in
mid-summer leaving just rosettes in the fall counts. The increasing number of invaded
quadrats coupled with lower percent cover in newly invaded quadrats explains the
apparent overall decline in A. petiolata cover over time (Figure 5), although the decreases
in mean A. petiolata cover from 2004 to 2005 may be due in part to environmental
conditions such as reduced precipitation in 2005. Over time, I expect that newly invaded
quadrats should increase in A. petiolata cover to the same levels as quadrats that were

initially invaded.

ESTIMATION OF ALLIARIA PETIOLA TA FECUNDITY

The ratio of seeds per plant to siliques per plant was invariant across Sites (Figure
6). I counted the seeds fi'om 132 A. petiolata plants collected from multiple locations
across southern Michigan. Plants ranged in number of siliques from zero to 266 and had
from zero to 3,864 seeds. The linear regession of the number of seeds versus siliques had
a slope of 14.587 (R2 = 0.9806), meaning that each siliques contained an average of 14.6

seeds.

70

ALLIARIA PETIOLA TA DAMAGE BY HERBIVORES

I observed damage to A. petiolata plants 536 out of the 631 times (84.9%) that
sampling quadrats contained A. petiolata across all sites and years (of 960 total quadrat
observations). However, the mean proportion of A. petiolata leaf area damaged or
consumed per quadrat was estimated to be only 2.9 i 2.2% across all sampling dates, and

incidence of more substantial damage was infrequent (Figure 7).

Spring damage

Within the subset of quadrats that contained A. petiolata plants across all sites and
years, I observed leaf edge feeding damage in an average of 41 .6% (range 13.0 - 62.5%)
quadrats sampling”, leaf hole damage in 79.4% (range by site 48.1 — 98.2%), and
windowpaning in 7.1% (range 0 — 35.0%) of sampling quadrats. I recorded browse by
larger herbivores (i.e. deer, woodchucks) at four sites with damage occuning in 2.8 — 5%
of quadrats. The majority of sampling quadrats at the Shiawassee site are located on the
Shiawassee river floodplain, which was substantially flooded during the spring of 2004.
This accounts for the high A. petiolata seedling mortality observed during that season,
which was recorded here as “other” damage.

Diseases on A. petiolata in Michigan: I observed diseased plants at one site in
spring 2003 and at three sites in spring 2005 with 1.7 — 8% of quadrats containing
diseased plants within those sites. Plants fiom Ives Road in spring of 2005 had virus-like
symptoms but tested negative for cucumber mosaic virus (CMV). Diseased plants were

stunted with unusual gowth patterns that included highly convoluted leaf surfaces and

71

siliques. Plants with these symptoms were typically gouped close together within a site
and were seen at Russ Forest and at the Kellogg Biological Station Bird Sanctuary in
Hickory Comers, MI. Wilted plants in Springville, MI (approximately 20 km west-
northwest of the Ives Road site) tested positive for Pythium sp. (personal communication
Jan Byrne, Mich. State Univ. Plant Disease Diaglostician, Diagrostic Services May 18,
2005), and fungal gowths that caused weakening of A. petiolata stems at a site
approximately 6 km south of Russ Forest were identified as Sclerotinia sclerotiorum
(white mold) by Pat Hart (personal communication. Mich. State Univ. Department of

Plant Pathology, May, 2004).

Fall damage

The damage types I observed in the spring were also most common in the fall.
Edge feeding damage occurred in an average of 65.0% (range = 47.8 - 80.0%) quadrats
sitel sampling", leaf hole damage in 75.7% (range 44.1 — 97.4%), and windowpaning in
22.6% (range 0 - 32.1%) of quadrats site‘1 sampling". I saw evidence of browse only
once during fall sampling at one site and disease only twice. Diseased plants at Lux
Arbor appeared to be virally infected as described above, but those at Shiawassee were

only yellowed and not wilted.

Extensively Damaged Quadrats

In most quadrats A. petiolata was not accepted for sustained feeding by
herbivores, and feeding damage was therefore limited to “tasting” followed by rejection.
However, the few quadrats in which A. petiolata was more extensively damaged are of

special interest because they suggest the possible existence of local populations of

72

herbivores that are more accepting of A. petiolata. With the exception of flood damage at
Shiawassee in spring 2004, there were only 33 quadrats with geater than 10% leaf area
damaged, 28 of which were observed during fall sampling. Most of these represented
feeding in quadrats containing a small number of A. petiolata plants which may give a
false impression of extensive damage. Nearly all quadrats with high damage estimates
had holes and edge-feeding damage.

The most interesting cases were in four quadrats: two each fi'om Femwood in fall
2003 and Lux Arbor in spring 2004 with higher A. petiolata cover (10 — 45%) which
sustained 10 — 20% leaf area damage. Each or these four quadrats had damage from edge
feeding insects and holes from other herbivorous invertebrates, and one at Lux Arbor had
been browsed by deer. The extensive edge and hole damage invone quadrat at Lux Arbor
(20% damage in a quadrat with 45% A. petiolata cover) may be worth monitoring in the
filture. If local populations of herbivores at some locations are capable of feeding on A.
petiolata and taking advantage of it as an abundant food source, it is possible that they
could multiply and spread to other locations.

Despite the widespread presence of herbivore damage, total leaf area removed
averaged 2.1% (range 0.3 — 9.5%) across all Sites and years in spring and 4% (range 0.9 —-
8.9%) in fall. The highest damage estimate represents the effects of the flood at

Shiawassee.

Impacts of Damage

Regession tree analyses indicated non-negative or no relationships between
survival or fecundity estimates and the presence or extent of A. petiolata damage. Alliaria

petiolata fecundity and overwintering rosette survival were not correlated with any

73

measures of A. petiolata damage in fall, year, or site when Site was included as an
independent variable. Removing the term for site and relaxing the splitting criteria to
allow splits to improve fit (PRE) by 0.025 rather than 0.05 and lowering the minimum
terminal goup size to 3 from 5 did allow estimation of effects on overwintering survival,
but all findings were insiglificant (Figure 8). The insigrificant model as fit showed
higher survival where holes damage was present, and among quadrats without holes-
darnage survival was insignificantly higher in the winter of 2004 - 2005 than 2003 —
2004. Using spring damage data to predict fecundity was again not sigrificant as was the
relationship with seedling survival. Relaxing the fit criteria still did not allow fitting of a
model to the predict fecundity. However, when site was removed as a predictor, seedling
survival was positively correlated with percent damage to A. petiolata, but the fit was not
sigrificant (Figure 9, PRE = 0.0557) and only partitioned out 24 of 306 observations.
These analyses Show that the impacts of herbivores, browsing animals, and other
forms of damage to A. petiolata as well as Site species richness and differences between
Sites are not Siglificantly correlated with A. petiolata survival or fecundity. This suggests
that although A. petiolata plants were minimally fed upon in the majority of quadrats, this
feeding had positive impact on A. petiolata performance if any. It is possible that A.
petiolata overcompensated for damage, but because the correlations were not sigrificant,

further speculation about their nature is not warranted.

CONCLUSIONS

Alliaria petiolata populations are expanding within invaded forests in Michigan.

This trend is sigrificant across seven locations where measurement of spread was

74

possible. At an eighth sampling location the population had completely invaded the study
area fiom the initiation of sampling in spring of 2003. Although I was not able to detect
increase in the spatial distribution of A. petiolata at this site, I was able to Show that the
distribution has not decreased over the course of the study.

Surprisingly, I found that A. petiolata plants are almost universally fed upon by
herbivores across southern Michigan. However, damage to plants rarely exceeded 2% of
the total leaf area in a quadrat. Although widespread, this feeding appears to have no
impact on A. petiolata survival or fecundity as I measured it. Because I only sampled
each site twice annually, the measures of survival that I used are simplified and do not
represent the true demogaphic rates of these transitions. However, because all sampling
was done in a short time period each year, using these rates for comparisons between
sites is appropriate. Variation in A. petiolata demogaphic parameters across southern
Michigan is the focus of other ongoing research and can be better used to make
predictions about A. petiolata population gowth (Davis et al. 2005).

The data presented here paint a portrait of an invasive weed that is spreading
rapidly into new habitats and is unchecked by natural enemies. These data show that, in
Michigan, damage to A. petiolata fi'om herbivores is not biologically sigrificant.
Increasing herbivore damage to invading A. petiolata populations with introduced natural
enemies may present a new opportunity to slow or reverse its spread. Given the potential
for A. petiolata to cause harm to the communities that it invades as others have shown
(e. g. Stinson and Klironomos 2005) and the ineffectiveness of conventional controls,

classical biological control agents may be recommended for A. petiolata in Michigan if

75

agents are approved for release in the future. If natural enemy agents are released, these

data will provide a usefirl benchmark for evaluating their performance.

76

LITERATURE CITED

Blossey, B. 1999. Before, during and after: The need for long-term monitoring in
invasive plant species management. Biological Invasions 1:301-311.

Blossey, B., V. Nuzzo, H. Hinz, and E. Gerber. 2001a. Developing biological control of
Alliaria petiolata (M. Bieb.) Cavara and Grande (garlic mustard). Natural Areas
Journal 21:357-367.

Blossey, B., L. C. Skinner, and J. Taylor. 2001b. Impact and management of purple
loosestrife (Lythrum salicaria) in North America. Biodiversity and Conservation
10:1787-1807.

Bohlen, P. J., S. Scheu, C. M. Hale, M. A. McLean, S. Migge, P. M. Groffinan, and D.
Parkinson. 2004. Non-native invasive earthworms as agents of change in northern
temperate forests. Frontiers in Ecology and the Environment 2:427-435.

Davis, A. S., P. M. Dixon, and M. Liebman. 2004. Using matrix models to determine
cropping system effects on annual weed demogaphy. Ecological Applications
14:655-668.

Davis, A. S., D. A. Landis, J. A. Evans, and D. W. Schemske. 2005. Matrix population
models to inform A. petiolata biocontrol agent selection (poster). in USDA NR1
Biology of Weedy and Invasive Plants PI Meeting, Washington, DC.

Gerber, E., H. Hinz, N. Guazzone, J. McKenney, S. Michler, and M. Zuefle. 2002.
Biological control of garlic mustard, Alliaria petiolata (Bieb.) Cavara & Grande.
Annual report 2001. Unpublished Report, CABI Bioscience Switzerland Centre,
Delemont, Switzerland.

Gerber, E., H. L. Hinz, and G. Cortat. 2004. Biological control of garlic mustard, Alliaria
petiolata. Annual report 2003. Unpublished Report, CABI Bioscience
Switzerland Centre, Delemont, Switzerland.

Gerber, E., H. L. Hinz, M. Schat, G. Corat, N. Guazzone, and L. Lovis. 2003. Biological
control of garlic mustard, Alliaria petiolata. Annual report 2002. Unpublished
Report, CABI Bioscience Switzerland Centre, Delemont, Switzerland.

Gotelli, N. J., and A. M. Ellison. 2004. A primer of ecological statistics. Sinauer
Associates, Inc. Publishers, Sunderland, Mass.

Hale, C. M., L. E. F relich, and P. B. Reich. 2005. Exotic European earthworm invasion
dynamics in northern hardwood forests of Minnesota, USA. Ecological
Applications 15:848-860.

Hinz, H. L., and E. Gerber. 1998. Investigations on potential biological control agents of
garlic mustard, Alliaria petiolata (Bieb.) Cavara & Grande Annual Report 1998.

77

Unpublished Report, CABI Bioscience Switzerland Centre, Delemont,
Switzerland.

Hinz, H. L., and E. Gerber. 2000. Investigations on potential biological control agents of
garlic mustard, Alliaria petiolata (Bieb.) Cavara & Grande Annual Report 1999.
Unpublished Report, CABI Bioscience Switzerland Centre, Delemont,
Switzerland.

Hinz, H. L., and E. Gerber. 2001. Investigations on potential biological control agents of
garlic mustard, Alliaria petiolata (Bieb.) Cavara & Grande Annual Report 2000.
Unpublished Report, CABI Bioscience Switzerland Centre, Delemont,
Switzerland.

Katovich, E. J. S., D. W. Ragsdale, L. C. Skinner, and R. L. Becker. 2001. Effect of
Galerucella spp. feeding on seed production in purple loosestrife. Weed Science
49: 1 90-1 94.

Kaufman, L. N., and D. A. Landis. 2000. Host specificity testing of Galerucella
calmariensis L. (Coleoptera: Chrysomelidae) on wild and ornamental plant
Species. Biological Control 18:157-164.

Landis, D. A., D. W. Schemske, and J. A. Evans. 2004. Evaluating the potential for
biological control of garlic mustard (Alliaria petiolata (Bieb.) Cavara & Grande)
in Michigan. 2003 Progess Report. Michigan State University, East Lansing.

Landis, D. A., D. C. Sebolt, M. J. Haas, and M. Klepinger. 2003. Establishment and
impact of Galerucella calmariensis L. (Coleoptera : Chrysomelidae) on Lythrum
salicaria L. and associated plant communities in Michigan. Biological Control
28:78-91.

Louda, S. M., A. E. Amett, T. A. Rand, and F. L. Russell. 2003a. Invasiveness of some
biological control insects and adequacy of their ecological risk assessment and
regulation. Conservation Biology 17:73-82.

Louda, S. M., R. W. Pemberton, M. T. Johnson, and P. A. Follett. 2003b. Nontarget
effects - The Achilles' Heel of biological control? Retrospective analyses to
reduce risk associated with biocontrol introductions. Annual Review of
Entomology 48:365-396.

Louda, S. M., T. A. Rand, F. L. Russell, and A. E. Amett. 2005. Assessment of
ecological risks in weed biocontrol: Input from retrospective ecological analyses.
Biological Control 35:253-264.

Luken, J. O., L. M. Kuddes, and T. C. Tholerneier. 1997. Response of understory species
to gap formation and soil disturbance in Lonicera maackii thickets. Restoration
Ecology 5:229-235.

78

McEvoy, P. B., and E. M. Coombs. 1999. Biological control of plant invaders: Regional
patterns, field experiments, and structured population models. Ecological
Applications 9:3 87-401.

Meekins, J. F., and B. C. McCarthy. 2000. Responses of the biennial forest herb Alliaria
petiolata to variation in population density, nutrient addition and light availability.
Journal of Ecology 88:447-463.

MNFI. 2003. Michigan's natural communities draft list and descriptions. Michigan
Natural Features Inventory, Lansing, MI.

MNFI. 2005. Vegetation circa 1800 maps
http://web4.msue.msu.edu/mnfi/data/veg1 800.cfm.

Nuzzo, V., and B. Blossey. unpublished. Garlic mustard monitoring protocol.
http://www.invasiveplants.net/invasiveplants/InvasivePlants/GarlicMustard/ garlic
Mustardasp.

Nuzzo, V. A. 1993. Current and historic distribution of garlic mustard (Alliaria Petiolata)
in Illinois. The Michigan Botanist 32:23-33.

Nuzzo, V. A. 1999. Invasion pattern of the herb garlic mustard (Alliaria petiolata) in high
quality forests. Biological Invasions 1:169-179.

Nuzzo, V. A. 2000. Element stewardship abstract for Alliaria petiolata (Alliaria
ofi‘icinalis) garlic mustard. The Nature Conservancy.

Pearson, D. E., and R. M. Callaway. 2005. Indirect nontarget effects of host-specific
biological control agents: Implications for biological control. Biological Control
35:288-298.

Pimentel, D., L. Lach, R. Zuniga, and D. Morrison. 2000. Environmental and economic
costs of nonindigenous species in the United States. Bioscience 50:53-65.

Prati, D., and O. Bossdorf. 2004. Allelopathic inhibition of germination by Alliaria
petiolata (Brassicaceae). American Journal of Botany 91:285-288.

R Development Core Team. 2004. R: A language and environment for statistical
computing. Version 1.9.1. R Foundation for Statistical Computing. Vienna,
Austria.

SAS Institute. 2001. The SAS System Version 8.2. Cary, NC.

Sirnberloff, D., and P. Stiling. 1996. Risks of species introduced for biological control.
Biological Conservation 78:185-192.

79

Stinson, K., and J. N. Klironomos. 2005. Exotic plant invasion degades local
mycorrhizal association, alters community succession and limits restoration. in
Annual meeting of the Ecological Society of America, Montreal, Canada.

SYSTAT Inc. 2004. SYSTAT for Windows. Version 11.0. Chicago, IL.

Tepley, A. J ., J. G. Cohen, and L. Huberty. 2004. Natural Community abstract for
southern floodplain forest. Natural Features Inventory, Lansing, MI.

US. Congess. 1993. Harmful non-indigenous species in the United States, OTA-F-565.
Office of Technology Assessment, Washington, DC.

Voss, E. G. 1985. Michigan flora. A guide to the identification and occurrence of the
native and naturalized seed-plants of the state. Part II. Dicots (Saururaceae-
Comaceae). lst edition. Cranbrook Institute of Science, Bloomingtion Hills, MI.

Wilcove, D. S., D. Rothstein, J. Dubow, A. Phillips, and E. Losos. 1998. Quantifying
threats to imperiled species in the United States. Bioscience 48:607-615.

Winterer, J ., M. C. Walsh, M. Poddar, J. W. Brennan, and S. M. Primak. 2005. Spatial
and temporal segegation of juvenile and mature garlic mustard plants (Alliaria
petiolata) in a central Pennsylvania woodland. American Midland Naturalist
153:209-216.

Wolfe, B. E., and J. N. Klironomos. 2005. Breaking new gound: Soil communities and
exotic plant invasion. Bioscience 55:477-487.

80

 

 
 

680m finance—m Sflsom
Sachem Eofisom .CQ

«mecca EDS—Sm 28255
“8.8m Eofisom DOBRO—24CQ

Sachem Eofisom 28.2le

«mecca Eofisom BESTEQ

“mecca Eofisom Bmozlba

“mecca 282 Eofisom
2:... reggae fizz

aw Sansone .89 e833... .Ams

been :83 5.3 as 8.. .se seas use;
.5 .33 essences .33 2&8 sea
9.8 8e .53 .84 as Page

v.83 A3 st use as... .38 use case
as are e2 .3: 2&8 ans. .23 been
seas .9: sea; seas Ame: as sea;
as page

has Ame as use as; .8” e8 one 8m
been enemas a8 e2 A E use an?
64 is

3:33 x85 .GS 292: Ema .35 bee—o5
seamen. .e: .E a8 as; 58.18 seam
gone x83 .48

x83 .va Bane .se as nausea... .s:
Refine: Ave .8 .8 ass. #83 .se
.5 one see .83 are e2 Ass one an?
an as

as an...» is season .5 .eg use as
.8” .va 2&8 age. .3 .es 298 awe...

0mm $3.37

ommmogfiw-

034386?

omvmoovww-
0 ~ Emmeaw-

osmommmdw-

immcvfiew-

o _ memwvdwl

omfiwmwwNv

$529?

08834.?

oowwmavdv
ooomcwe. :4

swamomdv

otgcwév

33.30.?

:5 use $2... 22.5 Ease: Sane...— seas...—

commeBeEm
memo
:ofiwfifiq
cam

oaneoA

8.8823—

scream

enema 8683 am
@550 25.2 3a

cemmeBeEm
amazon mmsm

>22er
.82 55

need 83

DEC .E

ecoBEom

3838 came» .«o wetoEoE Enorwe£ com Babe—om 8% 53:22 A 03:.

81

 

 

so 4 so no 4 as so 4 S no 4 M: to 4 S no 4 S no 4 3 so 4 3 Ass 5894:: as:
o 4 as o 4 as E 4 es 2 4 3 e 4 as o 4 as so 4 3 o 4 es senesce seem
as 4 we: a 4 new as 4 is we 4 3e 3 4 2n 3. 4 ”.8 ea 4 QB e 4 new $5883
3 4 e: m 4 on 3 4 3 es 4 e: 2 4 4.2 as 4 as 3 4 3 cm 4 S; Ascenseo 8°?
3 4 9:. ”a 4 Na 3 4 can a 4 in we 4 new 5 4 2: ed 4 me e 4 was 35.8 :8
E 4 Sm 3. 4 2m Sn 4 5:. a 4 ...: e... 4 ”.2 on 4 in w... 4 n: as 4 ”.2 Esau Seems/Eco
new 4 23 SN 4 use can 4 3.2 a? 4 32 n: 4 we. new 4 an: in 4 was a: 4 23 reasons serves
2: 4 Sn E 4 3m 3 4 can no 4 9% an 4 EN 3 4 3.4 2 4 man we 4 4.: Essence asses .w as:
3 4 n: 2 4 2 S 4 2 as 4 a: on 4 2 3 4 an 2. 4 q: n... 4 S roses . sees...
an” 4 4.3 on 4 an ad 4 3m a: 4 4.2 3 4 2: M42 4 3N an 4 New 9: 4 «.8 scenes . seesaw
3. 4 2 as 4 2 S 4 2 3 4 ...N S 4 3 Z 4 2 S 4 m... 3 4 1.. Trans . 3.3...
3 4 3 o o o o o o 4 as o Asvnmaeacsao
3 4 S e o 3 4 S o o o 4 3. o 4 as g unseen 88.5
o o 4 .3 o c 4 3. o o o 4 ed o 4 3. es Banana 8325
S 4 as 3 4 so 3 4 so 3 4 as 3 4 ed 3 4 as 3 4 mo 3 4 as Asvnmaéo 83m
3 4 no 3 4 so 3 4 no 3 4 3 3 4 to 3 4 no 3 4 no 3 4 es Asvnwaeac mesons seem
S 4 3 o 4 ed 3 4 so 3 4 no 3 4 ..o 3 4 no 3 4 3 o 4 as assuage 843853
3. 4 as 2 4 S 2 4 Z 2 4 an S 4 3 no 4 2 S 4 2 2 4 2 A3882 seaweeacaea
ea 4 4...: 3 4 3 ...N 4 on an 4 we as 4 an as 4 3 NB 4 ”.2 2 4 2. A5580 36.3
as 4 a: _ 4 ea 2 4 2 3 4 3. S 4 2 cm 4 on .... 4 q: as 4 3 3.03023...
3 4 SN 2 4 2 as 4 3. ma 4 ..s f 4 an S 4 as S 4 can as 4 as .260 asses .4. Bee
3 a 8 we a S m: s 835 e530 co Benz
8.33.5 sees ass Siege .82 5: as 8: ease as Bases“. .28; 3m 25 sec
8:558 83c seesaw 5..

dose. Emma mo :03 as 38936 cm Bob 80%;on wage
as .8 E2 4. an; 2% Bow 1 88 see assess means... an as need .eeesae assesses sea Bass 58 a means as
88892 $550 was .550 Beacon Susanna .V ammonia .353 SSSBQ .v. 3:338 35 3293: 88m cone—no.8 mm 3.83 Sanctum
.v. 9 newsman mo E85 can Snowman e5 EBA—doom .Ewmo: SSSSR .V :38 we 835$ .25“ 35 as «88E one 3.23 .8 mwfizooom
o: 8:83 525 0338 35¢.an zfl E b>oo BREWER .V 38,—. mew—mm E .356 we :63 we 35E SSouem .V. :35 was meager.
com b03834. 338me ob? Sue .5>oo Boobs 6382956 econ? mm _ a venom 33953 new 8% .3 8% .«o b.2556 .N 03:.

82

 

 

3 4 S 4.4 4 2 4.4 4 4.4 4.4 4 4.4 4.4 4 4.4 no 4 S :4 4 4.4 4.4 4 4.4 €445.45 5:: :82
:4 4 :4 4 o 4._ 4 S o 4 4.4 o :4 4 4.4 4 $4.38 vtom
2 4 4.4 4._ 4 4.8 4.4 4 4.44 4.4 4 $4 4.4 4 4.44 4.4 4 4.44 4.4 4 4.44 4.4 4 4.44 3428443
4._ 4 4.4 2 4 4.4 3 4 4.4 2 4 4.2 N 4 4.4 4.4 4 S: 4.4 4 4.2 4.4 4 4.4 34:38 483
4.4 4 4.3 4.4 4 S 4.4 4 4.2 S 4 4.4 4.4 4 4.44 4.4 4 3 4.4 4 4.2 4.4 4 4.44 358264
4.4 4 24 4.4 4 444 4.4 4 4.44 4 4 4: 4.4 4 34 4.4 4 4.44 4.4 4 4.44 4.4 4 4.2 $428 84843, 550
4.42 4 4.444 4.44 4 4.444 4.9“ 4 4.4: 34 4 4.444 4.: 4 4.2: 4.2 4 4.44 4.44 4 4444 4.44 4 4.4”: £45.84 536.3
4.4 4 4.44 4.4 4 $4 4.4 4 4.44 4.2 4 4.44 4 4 4.44 4.4 4 $4 4.4 4 4.44 _.4 4 4.44 483.446: 334.424 .44 :82
4.44 4 4.5 4._ 4 4.4 4.4 4 4.4 4.4 4 4.4 4.4 4 4.4 4.4 4 4.: 4.4. 4 4.44 4.4 4 4.: 1484.44.43.4on
N4 4 4.44 4.4 4 4._ 3 4 4.4 2 4 Z Z 4 4.4 4.4 4 34 4._ 4 4.4 3 4 4.4 1484.84.33
S 4 3 o o o o o o 4 4.3444488 550
3 4 3 o o o o o 4 4 $444.55 8845
o o o o o o o 4 £4 444.59 838m
4.4 4 4.4 o 4 3 3 4 4.4 :4 4 4.4 ..o 4 4.4 3 4 4.4 3 4 4.4 4 4 3 3 ounces 8.4:
o o o o o o o 4 £4 44485 45484 448
4 o o 4 c o o o g 44454— 4543833
o 4 4.4 o 4 4.4 S 4 4.4 3 4 4.4 S 4 no 4 4 4.4 o 4 4.4 4 4 4.4 89682 $444853
4.4 4 4.4N 4.4 4 4.4 4._ 4 2 4.4 4 4.4 4.4 4 4.4 4.4 4 4.4 4 4 4.44 4.4 4 3 4.4.4.28 45.484
4.4 4 4.: 2 4 4.4 4.4 4 4.4 3 4 _.4 4.4 4 4.4 4.. 4 2 3. 4 4.: 4.4 4 4.2 4.4468834.
4.4 4 4.44 4.4 4 3 Z 4 4.4 4.4. 4 4.4 4.4 4 4.4 4.4 4 4.2 4.4 4 4.44 «.4 4 4.2 3.260 43924 .2544
4_ a S a M: 4 8 4_ 8442: 4.8490 44 4444.42
84443424 .884 443. 482454 342 43 4m .8: .345 4.4 483844 42483 48m 25 54a
88 45am

425480 .4 2.3.

83

 

 

3 a I. to a 3 no a 3 no a M: to a 2 3 a M: 3 a E to a 3 2.3 58935 :82
o o o S a S o a 3 o 3 a 3 o §v§oo x8m
we a. mi 3. a q; 3 a no» 5 a a: as a 20 E. a 3w 3 a 2m x... a n2 $530.33
.3 « 0.3 we a n: S « no _ a 3 v a «2 an a 3 E a E E « MEN 39:28 825
2. a <8 3 a 3 Z a 92 to a h: 2 « EN 3 a 3 3. a n: 3. a can §V§8=om
2. a «.3 5 « v.2 3 a 3m 3. a 3: 3 a 3:. 2 a can 3 .4” EN 3. a v.2 .260 8:89:28
3” a $2 ”.2 a 3: new a mo: v.3 « 3% o mm a 2% ”S a 3mm 9: a ..8 988$ £965
3% a :2. v a 3 3 a we as a 3;. a 3 a :3 S a ”.2 E a 2: €32»? sssg i=8:
3:. a 2w ”.2 a 3m 92 a new 3." a 12. I: a 3:. we a 3m Q: a t: 0.: a 3m ..msmdzwssoom
3 a a: S a no 3 a to no a 2 o 3 a 2 3” a 5.2 3 a 3 Trans . $33
3 a E o o o o o S a 3 o 303.5 25
o o o o o o c o 303.5 a;
o 3 a 3 o 3 a S o o 3 a No 3 a S Aéowsfin 838m
3 a S 3 a we 3 « .3 o a 3 3 a no a a S S a .3 3 a 3 Ga gunman 86m
8 a no 3 a to S a no 3 a 3 S a to S a we 3 a to c a 3 g omega magnum omum
o o o o o o o o 3 page 053853
Nd a SN 2. a 3 no a M3 2 a an o a to o a 2 o a 3 3 a 3 @852 é owaaaeg
S a q: n a 5 3“ a me an a q: 3 a 2 N a 3 3 a 3m 5,. a «.2 .260 ”£88
2 a v.8 3 a do 3 a mo 2. a 3 o 3 a as 2 a S: 3 a 3 580.33
3 a 2m 3 a a... E” a q: 3 a n: 3. g 2. an a 5.2 S a can a...” a 5.2 $550 Susi .V Bow
2 3 2 S 2 : om E BEE 923.0% .385
8355 35m .....3 $205 se< 5: .3 a: 5:5 .5 83,63— 3.25? 3m 25 38
38 9.5m

3580 .n as:

84

 

 

3 H 3 3 H 3 3 H S 3 H 2 3 H 3 3 H 3 3 H 3 3 H 3 @3535: :82
o o o 3 H 3 3 H 3 c 3 H 3 c @3550 “.83
a... H 3.: 3 H 3.3 3 H 3» 3 H 3.: 3 H 3“ 2 H 3° 2 H 33 S H 3.3 3:283
S H 3 3 H 3 3 H 3 3 H 3 m H «.2 2 H 3 3 H 3 S H 3 3:38 803
3 H S: 3 H 3 3 H 3 e H 3.2 3 H 3 S H 3 3 H 3 w.— H 3 35838
3 H 33 3 H 33 3 H 3 3 H S: 3 H 3». 3 H 3.3 m H 33 v H 3_ 355m 8:8»; 550
33 H «.3 3_ H NR 33 H 23 33 H 3.3 E H 3m 3 H 3m 3“ H 32 n2 H v.8 £38m £30.03
3 H 33 3 H 3.8 a H v.3 3 H «.35 3 H 33 3 H 33 3 H 3.. o H ...: @3233 539.3 .V :82
v.2 H S: E H 3 3.: H 3« 3. H 3 3 H w: w: H 33 «.3 H 3.83 3_ H 33 33.83383
3 H 3 2 H 3 2 H 3 3 H w: 3 H 3 3 H 3 3 H 3 3 H 3 $833 . 2.32
o o o o o o o o 3:33 550
3 H 3 o o o o o 3 H 3 3 H 3 3359 23,35
3 o o o o o o c 3039 832m
3 H 3 3 H 3 3 H 3 c H 3 3 H 3 3 H 3 3 H 3 3 H 3 3 3.59 as:
3 H 3 3 H 3 3 H 3 c H 3 3 H 3 3 H 3 3 H 3 3 H 3 §Voma§n 338m 3m
3 H 3 c 3 H 3 3 H 3 o 3 H 3 c 3 H 3 333.3“ 833853
o H 3 3 H S 3 H 3 3 H 3 3 H 3 3 H 3 c H 3 o H 3 3552 3385.33
3 H 3.2 3 H 2 2 H 3 3 H E 3 H 3 3 H 3 3 H .3 _ H 3 33830 338%
3 H 3 _ H 3 3 H S 3 H 3 3 H 3 3 H 3 3 H 3 3 H 3.2 3:28.33
3 H 33 2 H 3 S H E 3 H 3 3.. H 3 3 H 3 3 H E 3 H 3_ 3.28 3523 .353
M: 2 8 2 a : 8 a BEE 923.0 .8 3.52
8333 .85 2.x 363m .82 x3 ,3. 8: .326 .E 83:3 .383 3m 25 35
88. 3%

8:580 .3 2.:

85

 

 

3 H 3 3 H 3 3 H E 3 H 3 3 H 3 3 H 3 3 H 3 3 H 3 Es 595 8:: :82
o H 3 c o 3 H 3 o o 3 H 3 c 33526 “.83
3 H 3_ 3 H 33 3 H 33 3 H 2m 3 H 3 S H 33 ~ H 33 3 H 3 338803
3 H 3.: 3 H _.: 3 H 3: 3 H 5.: 3 H 33 S H 3 _ H 3 3 H 3 3:260 885
3 H 3.8 3 H E 3 H 3.3 3 H 33 3 H 3.8 3 H 3 3 H 3 3 H 3 3380 =3
3 H 3.3 3 H 3: 3 H 3.2 3 H 3 3 H 33 S H 3 3 H E 3 H 3.: 3:28 =o38w0>§5
2: H 53 3 H 3 3 H M: 3 H 3 3 H 3 3 H 3 3 H n: 3 H 3 Trans . 8:83

o o o o o o o o 30ng 25

o c o o o o o o 3 0383 0385

o o o o o c o o 3 3&5 832m

o H 3 o o o H 3 3 H 3 3 H 3 3 H 3 3 H 3 A3 385 83:

o o o o o o 3 H 3 3 H 3 3 $959 95.8w 3m

3 o o c o o o c 3 03am 03°33
3 H 2 o o 3 H 3 3 H 3 3 H 3 _ H 3 3 H as 3882 33853

3 H 3: 3 H 3 3 H 3 3 H 2 3 H 3 3 H 3 3 H 3 2 H 3 3380 £83
.650 8583 .V 33.
2 N h h m m cm a 832: «.835 go Benz
8333 3cm 33 bios." .32 x3 .3 8: 53.6 .E Beacon .383 6m 25 58
383 am

625980 .N 03.5.

86

 

 

3 H 3 3 H 2 3 H 3 3 H 2 3 H 3 3 H 3 3 H E 3 H 3 @3589 3: =8:
3 o H 3 o 2 H 3 3 H 3 o 3 H 3 o 580 383
3 H 3.3. 3 H 33 3 H 3.3 3 H 3.. w H 33 3 H 33 3 H 33 3 H SN 3803
3 H 3 3 H 3 S H 3 3 H 3 3 H 3. 2 H 3 S H 3 3 H 3N .260 383
3 H 3 3 H 33 3 H 33 3 H 3.3 w H 33 3 H 33 3 H 3 3 H 33 5538
3 H 3.. 3 H 3_ 3 H 3_ 3 H 3 3 H 3_ 3 H 33 3 H 3.: 3 H 3 .550 833325
3 H ..2 3 H 3 3 H 3 3_ H 33 3 H 3.: 3 H 3 3 H 3 3 H 33 33.8 . 8:33.
o o o 3 o o o o 3 385 350
o o o 3 H 3 o o 3 o 3 3:39 883
3 o c o o o 3 H 3 c 3 3&5 838m
3 H 3 3 H 3 3 H 3 c H 3 3 H 3 3 H 3 3 H 3 3 H 3 3 033 83m
3 H 3 3 H 3 3 H 3 o H 3 3 H 3 c H A: 3 H 3 3 H 3 33 page 338". 3m
3 3 o c o c o c 3 385 053533
3 H 2 3 H 3 3 H 3 _ H 3 3 H 2 o H 3 3 H 3 3 H 3 3882 3333
3 H 3 3 H 2 3 H 3 3 H 3: 3 H 3 3 H 3 3 H 2 I H 3. 330 0:83
.850 33333 .V 38.
n 2 z 2 M: 3 2 : 332: 93% L332
8353 3cm 9.3 353.3 H32 .3 .3 8: L.225 .E 80253 8683 6m 25 35
38 .3

82680 .n «3:.

87

 

 

3 fl 2 S a Z 3 « f 8 a 2 é a .8 No a I do a E 3 a no €359.93: :82
3 a S o tn « nm _ a 2 ..o a 3 3 a 3 S a no 3 a 3 $3.260 zoom
3. a _.2 E a 3m 3 a ode we « moo a.“ a QR S. a a: E a 93 2. a n: $058884
3 a 3 3 a o6 2 a 3. an a 2: Z a as S a 2 ad a 3. M: a 92 §ro>oo 8o?
3 a a: 2 a 3. mm a v.8 S « EN 3 a n: 3‘ a N2 2 a 3m 3 a n3 §Vb>oo=om
w « man an « <2 2 a 3 3 « can 2 a S 2 a 93 _ a 3. S « 3 30:38 =o_.8wo>.§o
3 a we. 3 a 3 E a 3 3 a an 2 a 3 on a S: no a 3m 3 a ,5 Trans . 838m
o o o o o o o o Agvowasfi 350
S a 3 o o o o o o o g omega 0895
o o o o o o o o g amaze 838m
o « S S a no 3 a no 3 a mo 3 a E 3 a no 3 a 3 3 a no 3 Duncan 86:
o a S S a no 2. a to c « 3 S a «o 3 a 8 o a 3 No a no 3 Duncan 380m 03m
3 u." no 3 a no 3 fl 2 3 a no 3 « mo 3 a 3 3 a .2 o g amaze 08%,853
2 a on 3 um 3 n a E n « n: 5. a E .3 a ma 3 a 3 ed a 3 @582 é amazes;
3 a «a 3 a 3 S a 3 2 a 3 3 a 3 ed a S we a «a 3 a 3 3380 £82
.850 8322‘ .V 38.
S N 2 2 E 2 8 N BEE agomo 5:52
08:55 H88» and saxofi h32 .3 .3. 8: .326 .E Bose: 6683 6m 25 88
88 .3

8.580 .N 2.:

88

fire—25 08: @28me 05 mat—6 3935 895 85 $8230 893:8: 3325 no 838a 2: fl omega Eugen a

 

 

 

 

. to»:— 0
mo Em 05 an @332: .8383: 05 9 033—0.— vouwomg 3:35 0:5 05 maids 9595 begin“ .«0 hon—SE 05 a owgo 05 830:2: owgo “Goa—HQ Dims—“Hm a
ed a can So «.3 2x: 3% 9% mom 3 oi 83 T 88
of a can 98 3:- 0.2: 92. 58 ed 3 9% 88 T 38
”.3 a 3m 5: n? is 0.8 9%. ”.8 ed gow 38 T 88
pg ”wage
Z: a a? fix v.3 32 So 3 En S. 3: 88 T 88
3 a 3 Q8 3. 9mm 3. a: od 3 9m 88. T 38
NS .4“ m: E 9mm 98 ME 3- 3m ad ad 38 T 88
“g omega 3%?
S a. t M: 2 cm 2 e : cm a 88
S a E a 3 2 2 t : om M: 38
S um 2 E a S a M: p cm 2 88
3520 882:
932 083335 «moan $3— 55an SE 55 6m 83 .5630 am vooBEom 8:503 xom 30>

 

as“; % wag 38% 3a £3 623 Egg
89a 33 3.22: 8: ow omega 2583 9622 v5 omgao $8.89 8m 882 .mm ~ a as 83.3 :82 .08: ~96 owSEo E389
can carom @5953 9:ka :08 wgv 3530 mm? 8392mm .V 3: 20:3 8% some “a flagv mam—mag .«o 89:52 .m 93:.

89

Table 4. Univariate repeated measures AN OVA of the number of quadrats invaded by A.
petiolata over time within sites. Test excludes the Femwood site where initial conditions
prevented estimation of spread.

 

 

Source SS df MS F P
Year 79.2381 2 39.6190 11.8575 0.0014
Error 40.0952 12 3.3413

 

90

Table 5. Dunn-Sidak adjusted comparisons of the number of invaded quadrats per site

 

 

between years.

Within Subject Mean Difference Std Error Of P Lower Upper

Factor Comparing: Between Levels Difference 95% CI 95% C1
2003 H 2004 -3.4286 1.088 , 0.0582 -6.9904 0.1333
2004 H 2005 -1.1429 0.7377 0.4330 -3.558 1.2723
2003 H 2005 -4.5714 1.0659 0.0154 -8.0609 -1.082

 

91

 

 

 

 

 

 

 

 

 

 

 

 

Figure 2. Approximate known distribution of A. petiolata within Michigan's lower
peninsula (shaded counties) (Voss 1985, J. Evans and D. Landis pers. obs ) and
locations of study sites (stars).

92

 

N
O
.3

 

 

 

.93
'(75

3’

E

'U

(U

3

'0 1O '— ‘—
Q)

'0

(U

>

E

J I l
2003 2004 2005
YEAR

Figure 3. Mean number of sampling quadrats per site (n = 20) where live A. petiolata
plants were observed. Femwood data are not included because all quadrats were invaded

there during all three years.

93

 

Box Woodlot

Femwood .

Ft Custer.

Ives Rd

Lux Arbor

ens Jed SlBJanC) pope/M

Pinckney

Russ Forest:

Shiawassee .

 

 

 

 

1

 

200320042005 200320042005

Figure 4. The number of sampling quadrats containing live A. petiolata during sampling
has increased since 2003 at most sites during spring and fall.

94

 

100- -
80- -

40— .—

Mean Percent A. petiolata Cover
0)
c:
T
|

 

 

 

2003 2004 2005

Figure 5. Mean percent A. petiolata cover per quadrat from all quadrats during each
sampling period. Solid line shows quadrats that were invaded in 2003, dotted line shows
quadrats that were not invaded in 2003 but became invaded in 2004, and the dashed line
shows quadrats that became invaded in 2005. Cover estimates are from spring only.

95

 

      
 

 

 

 

 

4000 , 1
Seeds = 14.587 x siliques — 33.3306
R2 = 0.9806
*5 3000 —- _.
£9
a.
8 2000 —
(I)
U
(D
(D
(D 1000 —
0 100 200 300

Siliques per Plant

Figure 6. Regression used to estimate fecundity of A. petiolata plants. Plants used in this
analysis were collected at multiple locations across southern Michigan.

96

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

150 I I I I I I“ I I j I I T I T I r I I
Spring 2003 — Spring 2004 - fl Spring 2005
100” Median: 0.5000 _ :1 Median: 0.5000 .. _ MedIan:1.0000 ~
Mean :0.2750 ‘ Mean: 3.6437 Mean: 0-3750
4 _
50_ _ _ _ _ _
E 0 1 1 1 1 1‘ rig—1 1 1—1 l_.—-1 1 1 1 L 1
8 O 20 4O 60 80 100 0 20 4O 60 80 100 O 20 40 60 80 100
O 150 I I I I I I I I I I I I I I I I I I
Fall 2003 Fall 2004 I Fall 2005 ‘
10% Median: 0.0000 _ Median: 0.5000 g _ Median: 1.0000 7
Mean:1 .3469 " Mean:1.0500 _ Mean:4.4062 ‘
50*- 1- : 1- :
O 1 _1 1 1 1 1 1 1 j—Ji—l L“ L__ 1— 1‘
0 20 4O 60 80 100 0 2O 40 60 80 100 O 20 4O 60 80 100

 

 

 

 

 

 

 

 

 

 

 

Percent of Leaf Area Removed or Damaged

Figure 7. Frequency distribution of damage to A. petiolata foliage during each sampling
season. Damage was fiequent but was rarely extensive.

97

WINTEFLS RV

 

Mean=0.3977
SD=0.3130
N=164

 

 

 

 

HOLES_DAN

IAG E<1 .0000

 

 

Mean=0.2790
SD=0.3280
N=46

 

 

Mean=0.4440
SD=0.2957
N=1 18

 

 

 

 

 

 

 

YEAR<2004.0000

 

 

 

 

Mean=0.1746 Mean=0.3748
SD=0.2193 SD=0.3829
N=22 N=24

 

 

 

 

 

 

Figure 8. The relationship between overwintering survival and damage to A. petiolata,
year, site, and species richness is not significant. When the splitting criteria are relaxed
and site is dropped as a predictor, overwintering survival in quadrats with holes-damage
to A. petiolata was insignificantly higher than in those without such damage. (PRE for
overall model = 0.0852). Within quadrats that did not have holes-damage to A. petiolata
plants in fall, overwintering survival was insignificantly higher in the winter of 2004 —
2005 than 2003 — 2004. Dashed lines in figure indicate statistically insignificant splits.

98

SEEDLINGSRV

 

Mean=0.2598
SD=0.2841
N=306

 

 

 

 

A
LEAF_ATT.+CK<2.0000

 

 

 

Mean=0.2403 Mean=0.4893
SD=0.2695 SD=0.3510
N=282 N=24

 

 

 

 

 

 

Figure 9. When site is not included as a predictor of fecundity, the 24 quadrats in which
A. petiolata sustained greater than or equal to 2% damage to leaves had insignificantly
higher seedling survival than those with less than 2% damage. Overall PRE fit = 0.0557.
Dashed line indicates statistically insignificant split.

99

Chapter 3

DYNAMICS OF ALLIARIA PE TIOLA TA (GARLIC MUSTARD) INVASIONS
IN SOUTHERN MICHIGAN FORESTS

Jeffrey Adam Evans

100

ABSTRACT

Invasive species are frequently cited a among the greatest threats to global
biodiversity, although quantifying such impacts is challenging. Authors have variously
argued that invasions have strong or weak impacts on native communities, and that the
relationship between exotic and native plant species richness or abundance is positive
(biotic enhancement), negative (biotic resistance), or insignificant. Others have asserted
that the relationships are simply a fitnction of the scale at which studies are conducted.
Those who favor biotic resistance cite resource competition with native species and
stochastic survival and mortality as the primary mechanisms limiting or propelling
invasions. Proponents of biotic enhancement show that native and exotic species richness
are ofien positively correlated and covary with extrinsic environmental factors such as
climate, resource inputs, and geological features, thus proving that exotic and native
species are functionally similar. As the scale of comparison increases fiom local to
regional, the relationship of native to exotic diversity typically changes from negative to
positive as intrinsic factors that structure processes at the quadrat scale are overwhelmed
by extrinsic factors that dominate local to regional processes. While the native and exotic
species richness of the sites were positively correlated (P = 0.0002), the relationship
between native species richness and A. petiolata abundance within sites varied as a
function of site species richness (P = 0.0165) fi'om insignificant at species-poor sites to
negative at species-rich sites. Most interestingly, I found that the relationship of local
scale native species richness to A. petiolata presence or absence reversed from species

rich to species poor sites (P = 0.0087). In species-rich sites where resource availability to

101

plants is presumed to be greatest, quadrats with A. petiolata had significantly lower
species richness, implying that biotic resistance may be limiting establishment there.
Altemately, at species-poor sites where resource availability is expected to be limiting,
quadrats with A. petiolata had significantly higher species richness. This suggests that
neither biotic resistance nor enhancement alone sufficiently explains A. petiolata’s
invasion processes. Rather, they serve as endpoints along a continuum of tradeoffs
between resource competition and stress tolerance. This enriches current models that

have only found such reversals across gradients of scale.

102

INTRODUCTION

Biological invasions have been implicated in the disruption and alteration of
many biological systems through a broad range of mechanisms. Invaders compete with
natives for resources (Mack et al. 2000), alter habitat structure (e. g. Benoit and Askins
1999, Bohlen et a1. 2004, Hale et a1. 2005) and chemistry (e. g. Callaway and Aschehoug
2000, Bais et al. 2003), and sometimes alter successional processes via changes to soil
fungal communities and fire intensity or frequencies (e. g. D'Antonio and Vitousek 1992,
Stinson et al. 2005). These changes to both natural and managed systems have significant
implications for biotic communities (W ilcove et al. 1998) and substantial associated
economic costs (U .8. Congress 1993, Pimentel et al. 2000, Pimentel 2005).
Understanding whether communities differ in their susceptibility to being invaded and
whether invaders will negatively impact native communities if they become established is
of key concern to both researchers and managers alike. But predictions of the functional
and numerical relationships between invading species and the composition of invaded

communities vary widely.

BIOTIC RESISTANCE HYPOTHESES

Much of the discussion on invasibility has centered on the biotic resistance
hypothesis which predicts that species-rich communities should be more resistant to
invasion than species-poor communities (termed 'ecological resistance' by Elton 1958).
This expectation that diversity limits invasions is explained in part by niche partitioning
and mass effects and has been supported empirically in some systems (Knops et al. 1999,

Levine and D'Antonio 1999, Stachowicz et al. 1999, Naeem et a1. 2000). Tilman’s

103

stochastic niche model (Tilman 2004) similarly predicts that invasion probability
decreases exponentially as species richness increases by incorporating both stochastic
(e. g. Hubbell 2001) and competitive (e.g. Tilrnan 1976) elements from earlier models.
This outcome is predicted because (1) different species require different but overlapping
resources, (2) invader propagule dispersal is influenced by stochastic processes (i.e. the
movements of floodwaters, rodents, bird excrement, etc. dictate where propagules are
deposited), (3) propagules can only become established where the resources they require
are available, (4) established species use resources and make them unavailable to
invaders, and (5) the more established species that are present and using resources, the
lower the probability is of an invader’s propagule arriving exactly where its required
resources are present and unused. If the new species survives these largely stochastic
dispersal obstacles, the invader will only establish if its R“ (Tilman 1976, 2004) is equal
to or lower than those of its competitors. That is, the invader must be able to reduce the
availability of a limiting resource below the level required by competing species. If it
does, it will then displace any competing species that are limited by that resource.
Tilman’s (2004) predictions are not radically different from those of other biotic
resistance models, but the concert of stochasticity and competitive interactions in a single

model is more biologically plausible than either element operating alone.

BIOTIC ENHANCEMENT HYPOTHESES (THE RICH GET RICHER)

Other studies contradict the expected negative relationship between native and
exotic species richness or cover (Levine and D'Antonio 1999, Stohlgren et al. 1999a,
Stohlgren et al. 2002, Stohlgren et al. 2003, Stohlgren et al. 2005). This pattern suggests

that invasive species are functionally equivalent to native species in their resource

104

requirements and that resource rich habitats which support large numbers of native
species should be highly invasible (e. g. Stohlgren et al. 2003). Stohlgren (1999a) and
Stohlgren et al. (2003) have argued that the negative relationship “paradigm” is supported
primarily by studies conducted at the l-m2 scale or smaller and even then only in some
systems. At increasingly larger spatial scales this negative correlation typically breaks
down such that the overall relationship between native and non-native species richness or
cover is positive, i.e. “the rich get richer” (Stohlgren et al. 2003). Shea and Chesson
(2002) suggest that conflicting patterns of native versus exotic abundance observed at
different spatial scales may stem from variation in extrinsic factors such as climate,
geology, and rates of system-wide resource inputs at larger scales which overwhelm the
interspecific interactions that dominate local or plot-scale processes. Empirical evidence
from croftonweed (Eupatorium adenophorum) invasion in southwest China is supportive
of this proposal and, additionally, indicates that the relationship between native diversity
and invasion susceptibility also changes with time since invasion (Lu and Ma 2005). This
implies that the sampling scale may affect the significance or direction of any
conclusions drawn in studies of invasive species impacts on native communities.

These conclusions are not entirely explanatory. Stohlgren et al. (1998, 1999a,
1999b, 2001, 2002) found positive relationships between the absolute cover or biomass of
native and exotic species at many spatial scales. However, analysis of the relative
abundances of species (calculated from the data in Stohlgren et al.’s works cited above)
shows that there is either no relationship or a negative relationship between native species
richness and exotic species relative cover at spatial scales from 14000 m2 (Lundholrn

and Larson 2004), which is exactly the opposite of Stohlgren et al.’s general conclusions.

.105

The positive relationships observed from studies of absolute native and exotic cover
indicate that both groups of species respond broadly to the same environmental
conditions, but the neutral to negative relationship of relative exotic cover to native
richness indicates first that absolute abundance or number of invaders should not be
equated with dominance by them, and second that species rich communities may have
some ability to resist invasions (Lundholm and Larson 2004). Thus, although biotic
resistance is not predicted to prevent invasion fi’om occurring, it has been shown to play a
role in lowering establishment probability, rates of spread and impacts (Levine et al.

2004).

INTERPRETING IMPACTS OF INVASIONS

Quantifying the impacts of invasive species on indigenous communities can be
difficult. Often there is no pre-invasion dataset to compare with the post-invasion
community or the impacts are ambiguous (e.g. Koenig 2003). In unmanipulated natural
systems there may be strong correlations between invading species and patterns of native
species abundance, but causality cannot be assigned. Causal relations might, however, be
inferred where the biology of the invader is well understood. For example, the invasive
weed Centaurea maculosa DC. (Asteraceae) has been widely studied and shown to
interact with native plant assemblages in part via allelopathic root exudates (Bais et al.
2003). Kedzie-Webb et al. (2001) revealed significant negative relationships between C.
maculosa abundance and native grass abundance, richness, and diversity, but could not
conclude whether C. maculosa caused reduction of native grass richness and abundance
or whether it invaded and become abundant in areas that were already lower in native

species richness. There is value in identifying these patterns in natural communities,

106

though, as they serve to guide future research efforts that can distinguish cause from
effect. Additionally, analyses of natural systems are not subject to the criticism generally
directed at studies of artificial communities and may shed light on true natural

associations between species.

STUDY SPECIES

Alliaria petiolata (M. Bieb) Cavara and Grande (garlic mustard) is an invasive
weed of European origin. First recorded on Long Island, New York in 1868 (Nuzzo
1993), it is now widely distributed and invasive in North America (Nuzzo 1993, 2000). It
is notable among invaders in its ability to penetrate high quality forest understories as
well as disturbed areas. Previous studies indicate that A. petiolata growth rates and
fecundity are positively correlated with light availability (Meekins and McCarthy 2000)
and that A. petiolata populations respond positively to the formation of canopy gaps
(Luken et al. 1997, Webb et al. 2001). Alliaria petiolata interactions with native species
may be mediated through soil communities. Ongoing research suggests that exotic
earthworms which increase litter cycling rates may be correlated with increased A.
petiolata abundance (Maerz et al. 2002, CM. Hale, Univ. of Minnesota, Duluth, pers.
comm, 2003, Bohlen et al. 2004, Hale et al. 2005). Additionally, A. petiolata is
arnycorrhizal and has allelopathic properties that disrupt associations between native
plants and their arbuscular mycorrhizal fungi (ABF) (Roberts and Anderson 2001,
Stinson et al. 2005, Wolfe and Klironomos 2005) and directly suppress germination of

sympatric species (Roberts and Anderson 2001, Prati and Bossdorf 2004).

107

COMMUNITY RESPONSE TO INVASION AND CONTROL EFFORTS

Land managers in many systems report negative responses of native plant
communities to increasing abundance of A. petiolata. Anecdotal evidence of this type is
widespread, though few studies have addressed this assertion. McCarthy (1997)
concluded that diversity increased as A. petiolata was experimentally removed, but the
sign of the effect he observed changed multiple times over the course of his study. Recent
findings demonstrating that the impacts of A. petiolata on ABF persist for years after A.
petiolata extirpation (Stinson et al. 2005) suggest that McCarthy’s study plots were likely
still subject to the legacy effects of A. petiolata even after it had been removed. Because
of the difficulty in interpreting the lingering effects of an invader after its removal, an
alternative approach may be to observe A. petiolata as it invades new areas. One can then
assess whether communities that are invaded differ from those that remain uninvaded
fiom the outset, and whether communities that become invaded diverge from their
uninvaded counterparts.

There is substantial interest in developing effective control strategies for
naturalized A. petiolata populations in North America. Conventional control methods
such as pulling, cutting, spraying, and burning are ineffective on all but the smallest
infestations (Nuzzo 1991, 1994, Nuzzo et al. 1996). Current research efforts are therefore
focused on classical biological control and host specificity testing of European insect
natural enemies (Blossey et al. 2001, Hinz and Gerber 2005). My study was designed to
establish benchmark, pre-release data at multiple A. petiolata invaded locations to
support and allow evaluation of the effectiveness of biological control efforts if they are
implemented in the future. Pre—release baseline community data such as these are rich

with ecological information and can provide added insight into the processes directing A.

108

petiolata invasions in our area. Because no manipulations were performed in this study
all interpretations are strictly correlative. However, no other datasets of A. petiolata
populations and community dynamics that I know of exist with this scale of replication.
Thus, correlative analysis is appropriate in this case as it may allow quantitative support

for the phenomena that many have casually described.

OBJECTIVES

From April, 2003 through October, 2005 I followed the progression of invasion
by naturalized A. petiolata populations at eight sites in the southern Lower Peninsula of
Michigan. The objectives of this study were to (1) characterize the plant species
assemblages at these eight locations prior to the introduction of biological control agents,
(2) identify potential impacts of A. petiolata invasions on these communities, and (3) to
evaluate patterns of A. petiolata invasions within the broader framework of plant invasion

theory.

METHODS AND MATERIALS
SITE SELECTION:

I utilized eight previously established study sites within A. petiolata’s primary
range in Michigan’s southern Lower Peninsula (Landis et al. 2004). Criteria for site
selection included (1) forested lands > 2 ha in extent, (2) under state, federal, or other
long-term conservation management, (3) on which A. petiolata populations have been
established for at least four years, and (4) with protection from future disturbance or A.

petiolata management for at least ten years. In spring of 2003 10 permanent 0.5 m2

109

sampling quadrats (0.5 x 1 m) were marked along each of two parallel, 100 m long
transects spaced 10 m apart at seven sites and along one 200 m long transect with 20
sampling quadrats at the eight site (Russ Forest) for a total or 20 quadrats per site, and
GPS coordinates were recorded for each site. Transects traversed the A. petiolata
invasion front where possible. This was done to allow us to measure spatial spread of A.
petiolata populations within sites. At one site (Femwood) this was not possible, and all
20 quadrats there contained A. petiolata from the outset of the study. Site inventories
included data on forest type (MNFI 2003) and maturity (diameter at breast height of
principal overstory trees), and overall plant community composition. Sites are described
in greater detail in Chapter 2. Accurate records of species composition were not kept at
these sites before the initiation of this study. It is therefore not possible to determine
exactly how long A. petiolata had been present at any of the sites prior to 2003, although
the extent of the invasions and anecdotal evidence from managers indicated that they met

the criteria listed above.

SAMPLING METHODS:

I collected data on A. petiolata distribution and abundance based on a nationally
standardized protocol developed for pre-biocontrol-release monitoring of A. petiolata
populations (Nuzzo and Blossey unpublished). In spring (June) and fall of (Sept. — Nov.)
of 2003 — 2005 I visited each site and recorded vegetation percent-cover from each
quadrat including A. petiolata, and non-A petiolata vegetation by species. Specimens of
species that were not identifiable were collected near the quadrats for later identification.
Estimates of total cover often exceeded 100% because each species’ cover was estimated

separately and the layers of vegetation frequently overlapped.

110

ANALYSES OF DATA
Site Overviews and Characterization

To characterize the sites I compiled inventories of all ground layer vascular plant
species (up to one meter height) for each site fi'om the six combined sampling dates. I
used the Michigan Natural Features Inventory (MNFI) community types (MNFI 2003)
and calculated the negative logarithm of Simpson’s diversity index (-lnD), the Berger-
Parker index of dominance (d), and Floristic Quality Assessment indices (described
below) to characterize the ground-layer plant species composition and distribution at the
eight study sites. All biodiversity indices were calculated using formulas in Magurran
(2004) with the exception of the Floristic Quality Assessment indices (Taft et al. 1997,
Herman et al. 2001). Metrics of diversity which include A. petiolata-data become
overwhelmed at high A. petiolata abundances while information from the residual
community is forced into the tails of the species distribution. Therefore, I analyzed
indices calculated for the residual community without A. petiolata except where noted.
Calculations in which A. petiolata data were included are subscripted with the letters
“GM” (for Garlic Mustard) to distinguish them from those that were not. Indices
calculated from native species data only are subscripted with the letter “N” (for Native).
For analyses of transect-scale patterns indices were calculated from the summed data
from all quadrats within each site (20 quadrats x 0.5 m2 = 10 m2). Only first year plants
were censused during fall sampling because second year plants had already senesced.

Therefore, only spring data were used in my analyses.

11]

Effects of Deer

Previous studies have shown that browsing by white-tailed deer [Odocoz'leus
virginianus (Boddaert)] exacerbates the negative impacts of A. petiolata invasions on
native plant communities (Kalisz et al. 2003). I evaluated the relationship between site
native species richness and county deer density estimates fi'om the Michigan Department
of Natural Resources (MDNR 2006) using Spearman rank correlations with PROC
CORR in SAS version 8.2 (SAS Institute 2001) to identify any potentially confounding

effects of deer density on native species and A. petiolata spread.

Metrics of Diversity
Simpson 's Diversity Index (-lnD)

Simpson’s diversity index (D) is a measure of the probability that any two
individuals chosen randomly fiom a sample belong to the same species. Simpson’s D is
weighed significantly towards the most abundant species but is not very sensitive to
differences in species richness which makes it an appropriate measure for comparisons
between sites with different numbers of species (Magurran 2004). In this analysis,
percent cover was used in the calculations as the measure of abundance. The negative
logarithm of Simpson’s D (-lnD) ranges from 0 to +00, with higher values representing
increasing diversity, and is recommended over the commonly-used l/D for its improved
variance properties and similar ease of interpretation (Magurran 2004 and references

therein).

112

Berger-Parker Index of Dominance (d)
The Berger-Parker index of dominance (d) is a measure of the relative abundance
of the most abundant species. Values of d range from US (perfect evenness) to 1

(monoculture).

F loristic Quality Assessment

Floristic Quality Assessment is an established analytical system for interpreting
the overall integrity of plant communities based on the conservativeness of the plant
species present (Taft et al. 1997, Herman et al. 2001). All species native to the study
range are assigned a coefficient of conservatism (C) ranging from 0 — 10. Common native
species that show no specific affinity to undisturbed habitats are assigned a C of 0, while
rare native species that are only found in the highest quality undisturbed natural areas are
assigned a C of 10. Non-native species (also referred to as adventive) are given a C of 0,

as they do not ad to the quality a site’s flora. Each site inventory is summarized by its
mean C value (C ) and Floristic Quality Index (F Q1), which is calculated as:

FQI = "CK/E
where S is the number of species in the sample and C is the mean coefficient of
conservatism for the sample. The F Q] is a weighted metric that allows for comparison
between samples of different species richness (i.e. between sites).

I conducted F QA analyses using the Floristic Quality Analysis computer program
and Michigan Flora database (Wilhelm and Masters 1999, Herman et al. 2001). Species
lists fi'om the six sampling periods were combined to create species inventories for each

site and calculate FQA statistics for overall site compositions using the F QA “inventory”

113

program. Additionally, F QA statistics were calculated for each site during each sampling
period using the F QA “transect” program, which weights calculations of FQA statistics
by species fi'equency and abundance. The fiequency (number of quadrats x site"),
relative frequency (number of observations x total number of observations"), cover (sum
of cover for all quadrats within site), relative cover, and relative importance value (RI V)
were calculated for each species. R] V is calculated as one-half the sum of the relative

frequency and relative cover for each species.

The site characterizations are summarized by Michigan natural community type,
(MNFI 2003), known disturbances, native and total species richness, F QA statistics,
Simpson’s —lnD, the Berger-Parker index of dominance (d), and both seasonal and

overall species inventories.

Site-Level Diversity Comparisons
Native vs. Exotic Species Richness

The most robust prediction of Stohlgren et al.’s (2003) “rich get richer” model is
that native and exotic species richness are positively correlated at larger spatial scales. I
tested this with a Spearman rank correlation analysis of site exotic species richness and
site native species richness using the site inventory data (SAS Institute 2001). Following
this, I considered patterns of diversity and evenness between sites and whether

anthropogenic or natural sites features logically correspond with them.

114

Diflerences Between Sites

I used Kolmogorov-Smirnov two-sample tests to compare patterns of species
abundance distributions between sites. The Kolmogorov-Smirnov two-sample test is used
to compare two sets of paired data (paired by rank number) and evaluate the significance
of the maximum difference within pairs (Magurran 2004). With an experiment-wise error
rate of a = 0.5, the Dunn-Sidak correction for multiple comparisons (Gotelli and Ellison
2004) between the 28 possible pairs of sites within each season was (1’ = 0.0018. Because
the test can only compare two distributions at a time, I created two summary distributions
for each site. In the first I ranked species separately within each quadrat and then
calculated the mean relative abundance of each rank within the site. In the second
distribution 1 summed the abundances of each species within each site across the 20
replicate quadrats and then ranked their relative abundances. The mean distribution
characterizes the relative abundances of species at the quadrat scale (0.5 m2) which is
closer to the scale at which inter- and intraspecific interactions occur, while the summed
distribution shows patterns at the scale of the transects (10 m2). In the former, the
species’ identities are not tied to their ranks, whereas they are in the latter distribution.
Kolmogorov-Smirnov tests were run using cumulative relative species abundances rather
than absolute abundances to allow comparisons between sites with different numbers of

species and total amounts of vegetation cover.
Alliaria petiolata Impacts and Invasion Processes

1 evaluated the potential impacts of A. petiolata invasions on native communities

and the possible roles of biotic resistance and facilitation in regulating invasions. This

115

was a two step process in which I first analyzed patterns of invasion and diversity within
sites by relating patterns of native species richness to A. petiolata presence or abundance
at the quadrat scale. To test whether invasion patterns within sites are correlated with
site-level patterns of diversity, I then related these quadrat-scale findings to the native

species richness of the sites.

Interaction of Site Species Richness and A. petiolata Presence

To test the hypothesis that A. petiolata presence is negatively correlated with
native species richness at the site level, I used a two way factorial GLM with A. petiolata
presence or absence and site as main effects and tested for the significance of the
interaction term (SYSTAT Inc. 2004) for each year. A significant interaction between site
and A. petiolata presence would indicate that A. petiolata is related to native species
richness differently at sites with different properties. The Femwood site was completely
invaded during all years, and Pinckney was completely invaded in 2005, so these sites

were dropped form the analyses of those years, respectively.

Quadrat-Scale Correlations

Many studies have found negative correlations between native and exotic
diversity or cover at small spatial scales (Shea and Chesson 2002). I tested the
relationship between native species richness and A. petiolata abundance within sites
using Spearman rank correlations constructed fi'om quadrat data for each spring sampling

date. I tested these relationships with both absolute (mean A. petiolata percent cover) and

116

relative A. petiolata cover successively because of possible discrepancies between these

two metrics (Lundhohn and Larson 2004).

Linking Invasion Processes across Scales

To evaluate the hypothesis that diversity and A. petiolata abundance within sites
are related as a function of site characteristics, I used Spearman rank correlations to test
the associations between the quadrat scale correlation coefficients (fi'om the preceding
section) and native species richness from each site. Significant findings in these
relationships would suggest that patterns of A. petiolata invasion vary as a function of
intrinsic site or community characteristics and could be generalized to explain broader

patterns of invasibility.

Predicting Invasion Probability

Observation of patterns of new invasions in natural communities is a strong
correlative test of biotic resistance. Fifty-seven sampling quadrats distributed across
seven of the eight sites were located where A. petiolata had not yet invaded in 2003. By
spring of 2004 some of these quadrats had become invaded by A. petiolata. The data
collected fi’om these 57 quadrats afforded opportunities to explore hypotheses about the
A. petiolata invasion process in Southern Michigan. Specifically, I hypothesized that
increasing species richness would be predictive of a decreasing probability of invasion by
A. petiolata.

Within the 57 uninvaded quadrats, I modeled the probability of a quadrat

becoming invaded with a logistic regression using the presence or absence of A. petiolata

117

in 2004 as a binomial response variable and either native or total species richness as a
continuous predictor variable. In this model, the presence of A. petiolata in 2004
represented an “event” outcome while the absence of A. petiolata in 2004 was a “non-
event”. I hypothesized that, at the quadrat level, invasion probability would be negatively
related to native or total species richness, which would lend support to the general biotic
resistance framework. I first tested for significant blocking effects of site and then pooled
the 57 observations. 1 used PROC LOGISTIC (SAS Institute 2001) to carry out a series
of logistic regressions which were then evaluated based on multiple criteria (Peng and So
1998, Peng et al. 2002, Peng and So 2002).

Overall fit was evaluated by comparing the logistic models with intercept-only
(null) models using Likelihood ratio, Score, and Wald tests. If the logistic model had
greater predictive power than the null model, Wald’s 36 tests were then used to test the
significance of individual predictor parameters. The fit of the model to the data was
evaluated with several goodness-of-fit tests. The Hosmer-Lemeshow statistic is an
inferential Pearson’s )8 statistic that tests the null hypothesis that the model is a good fit
to the data. The model is said to fit the data well when P > (1. Additionally four measures
of association are reported: Kendall’s Tau-a, Goodman-Kruskal’s Gamma, Somers’s Dyx,
and the c statistic. Each of these estimates the degree of association between the outcome
and the predictor. Kendall’s tau-a is a rank-order correlation coefficient that does not
adjust for ties. The Gamma and Somer’s Dyx estimate the percent reduction in prediction
errors made using the logistic model versus chance alone. (a Gamma of 0.25 would thus
mean a 25% reduction in predictive error using a given model). Dy, may represent an

improvement over Gamma in that independent and dependent variables can be specified.

118

Finally, the c statistic is the proportion of paired observations with different outcomes
(i.e. one quadrat is invaded while one is not) for which the model correctly predicts a
higher probability for “event” observations (invaded) than the probability of “non-event”
observations (not invaded) (Peng et al. 2002). For example, a c of 0.65 would mean that
for all pairs of observations that had different invasion outcomes, the model correctly
predicts the quadrat that became invaded 65% of the time. Values of c range from 0.5 to

1.0.

RESULTS AND DISCUSSION

Our ability to differentiate plant species improved substantially between 2003 and
2004 and again between 2004 and 2005. We recorded 95, 116, and 131 species in 2003,
2004, and 2005 respectively with 146 individual species observed during all three years.
Thus, species data from 2003 (not collected by the author) are not reasonably comparable
to data from other years, and I have not made comparisons of species diversity between
years for this reason. However, the data are consistent within years and do permit inter-
site comparisons. Data from spring of 2005 are used to illustrate specific points when
appropriate. Estimates of A. petiolata abundance and total non-A. petiolata vegetation

cover are reliable for evaluation of changes over time over time.

SITE OVERVIEWS
The eight study sites varied fi'om Southern Floodplain Forest to Dry-Mesic and
Dry Southern Forests (MNF I 2003) (Table 6). Site species richness ranged from 17 to 59

total species occurring in the sampling quadrats (Table 7) including A. petiolata, which is

119

present at all sites. Of these, native species accounted for 81.6 to 96.9% of the total
species observed with perennial forbs representing the single most common growth form.
Species inventories for each site are provided (Appendix 1) as are species lists with
abundance data from each sampling period (Appendix 2). I found earthworms at all sites
where I have looked for them (Box Woodlot, Femwood, Ives Rd., Lux Arbor, Russ
Forest, Shiawassee) and expect that they are also present at the remaining sites. Complete
descriptions of the eight sites are presented in Chapter 2 while selected features and

diversity information are discussed below. Site abbreviations given here are used in the

figures.

SITE CHARACTERIZATIONS

Box Woodlot (BW): Box Woodlot is a remnant Southern Mesic Forest (MNF I
2003) surrounded by agricultural fields. It has the lowest overall diversity of all sites by
multiple measures. In spring of 2005 A. petiolata occurred in 19 of 20 quadrats had the
highest relative importance value (RI V).

F ernwood Botanic Garden (F W): Femwood is a Dry—Mesic Southern Forest
(MNF I 2003) with high species richness and is the only site where A. petiolata occurred
in 20 of 20 quadrats during all years.

F ort Custer Military Training Facility (F C): Fort Custer is a Dry-Mesic Southern
Forest (MN F I 2003) whose occasional open-grown oaks indicate that it is likely a
degraded oak savanna. It has the second highest species richness of all the sites. In spring
of 2005 A. petiolata occurred in 11 of 20 quadrats and had the highest R] V.

Ives Road Fen Preserve (1R): The site at Ives Road is a Dry-Mesic Southern

Forest (MNFI 2003) situated between a restored upland prairie and a Southern Floodplain

120

Forest and has an intermediate level of plant diversity. In spring of 2005 A. petiolata
occurred in 19 of 20 quadrats and had the second highest RI V (9.5) after the native
perennial forb Sanicula gregaria (R1 V= 14).

Lux Arbor (LA): Lux Arbor is a Dry-Mesic/Dry Southern Forest (MNFI 2003)
that has a low level of plant diversity relative to most other sites and has the lowest
percentage of native species. Logging activity at the study site in spring 2005 caused
substantial disturbance to the under- and overstory communities. In spring of 2005 A.
petiolata occurred in 15 of 20 quadrats and had the highest RI V of all species.

Pinckney State Recreation Area (PR): The Pinckney site is a high quality Dry-
Mesic Southern Forest (MNFI 2003) with intermediate diversity but a high proportion of
native species. In spring of 2005 A. petiolata occurred in 20 of 20 quadrats and had the
highest RI V of all species.

Russ Forest (RF): The Russ Forest site is an old growth Dry Southern Forest
(MNFI 2003) with lower species richness but a high proportion of native species. In
spring of 2005 A. petiolata occurred in 20 of 20 quadrats and had the second highest RI V
(11.1) after the native tree Acer saccharum (RIV: 33.7).

Shiawassee YMCA Camp (SH): The Shiawassee site is a Southern Floodplain
Forest (MN F1 2003). The sampling transects extend from the drier second bottom of the
Shiawassee River valley onto the first bottom floodplain (Tepley et al. 2004). Shiawassee
had the highest total number of species of all sites but was dominated by adventives.
Quadrats on the first bottom were subjected to extensive flooding following heavy rains
in spring of 2004 which resulted in significant changes in species composition and

abundance. The relative abundance of another invasive plant, Lysimachia nummularia,

121

more than doubled by spring of 2005 following disturbance from the 2004 floods (RI V =
25.6) and displaced A. petiolata as the most important plant (RIV= 16.6). In spring of
2005 A. petiolata occurred in 18 of 20 quadrats but was less abundant than it had been in

previous years;

SITE-LEVEL DIVERSITY AND IMPACTS
Effects of Deer

The density of deer populations at the county level (Table 6) was negatively but
insignificantly correlated with site native species richness (Figure 10, Spearman rank
correlation r, = -0.6347, P = 0.0909). Although deer browse may serve to reduce native
plant diversity or richness and selectively facilitate A. petiolata invasions as others have
found (Kalisz et al. 2003), deer cannot be held directly responsible for the patterns of
plant communities observed in these data. First, because these data are observational, I
cannot interpret causality from them. Second, deer are known to respond positively to
habitat fiagmentation (Horsley et al. 2003) while plant diversity is know to respond
negatively (Higgins et al. 2003). This suggests that although deer can directly reduce
plant species richness, the relationship between deer density and plant diversity could be
derived from larger scale patterns of land use and habitat fragmentation. A separate
factorial study of the impacts of deer browse and A. petiolata invasions on native plant
communities in southern Michigan is ongoing and will address these concerns directly in

the future (Evans unpublished).

122

Native vs. Exotic Species Richness

Native and exotic species richness (SR) are positively correlated at the site level
(Figure 11) (Spearman rank correlation = 0.9728, P < 0.05) in keeping with observations
fi'om many other systems. As Stohlgren et al. (e. g. 1999a, 2003) have pointed out, this
trend is indicative of the relative suitability of the sites to plant growth in general and
shows that invasive species respond to the same site factors as natives. Other site factors
described below are broadly correlated with diversity. Lux Arbor was excluded fi'om this
analysis because a disproportionately large number of exotic species appeared there in
2005 following disturbance fiom logging activities that other sites did not experience.

Distance from public roads appears to separate species rich sites from species
poor ones and those with high diversity (-lnDc;M) from those with low diversity. Box
Woodlot, Lux Arbor, and Russ Forest, which have the fewest native species and low -
lnDGM, are each bordered along one or more sides by public or frequently used roads
from which the sampling transects are visible. In contrast, Shiawassee and Ft. Custer are
the most species rich sites and are much less accessible. The remaining sites with
intermediate richness are accessible to the public by foot except for Ives Rd., which is
closed to the public. Shiawassee, which has the highest total species richness, also has
low diversity (-lnDGM) and high dominance (dGM). While the other sites that share these
properties are easily accessible to human as described above, Shiawassee is less
accessible but is subjected to natural disturbances from periodic flooding which has led to

dominance by two disturbance adapted invasive plants (A. petiolata and L. nummularia).

The mean conservativeness (C ) and F Q1 of native species (Table 2) are not as clearly

separated by proximity to roads or disturbance, although F Q1 is more often lower in the

123

species poor sites. This shows that although proximity to roads is correlated with species
richness, it is not necessarily indicative of which species are present.

The Femwood site stands apart from the others in its extent of A. petiolata
invasion within the sampling area. Alliaria petiolata dominated the local community with
a relative cover of 57.4% in spring of 2005 (Appendix 2). Femwood had the lowest
diversity and highest dominance of all sites as measured by -lnDGM and dGM (calculated
with A. petiolata) (Table 7). However, when A. petiolata was excluded from these
metrics this site had the highest diversity and evenness within the residual community by
both measures. The difference between -lnD and -lnDGM at Femwood in spring of 2005
was 2.9015, which is greater than at any other site. This means that the probability of two
individual selected randomly from the community belonging to the same species is over

ez'9015 ) when A. petiolata is included than for the residual community

18 times greater (
without A. petiolata. The other species present, though numerous, neither become
abundant relative to A. petiolata nor differ in their abundances relative to each other. The
invasion at the F emwood study area appears to be more advanced than at other sites as
indicated by the spatial and vegetative dominance of A. petiolata. The dampening of
variance in native species abundance was not observed to this extent elsewhere, and I
believe that this may be an impact of A. petiolata invasion. Because Allee effects can
prevent some species fiom persisting as very small populations (Allee 1931 in Begon et

al. 1996) I expect that some native species may be lost at this site if A. petiolata remains

abundant.

124

DIFFERENCES BETWEEN SITES

The relative distribution of species abundance was similar across sites at the
quadrat scale but less so at the site scale (Table 8). Dunn-Sidak adjusted Kolmogorov-
Smimov two-sample tests for cumulative abundance of the mean rank-abundance data
show few significant differences between the distributions at the quadrat scale (Table 8,
top). No tests were significant within the 2003 data. The only significant differences in
2004 — 2005 were between sites with the highest high diversity (i.e. Ft. Custer) and those
with the lowest diversity (i.e. Box Woodlot and Lux Arbor). This implies that, at the
quadrat scale, the relative distributions of species that A. petiolata interacts with are
similar across sites.

Tests for differences in the distribution of the summed data were significant
between the three least species rich sites (Box Woodlot, Russ Forest, and Lux Arbor) and
many of the more species rich sites at the transect scale (Table 8, bottom). Box Woodlot
in particular, which had the lowest species richness, F Q1, and C (Table 7), stands out as
different from the greatest number of sites across all sampling dates. Fort Custer, which
had the highest F Q1 and the most evenly distributed species, was significantly different
from sites with lower species richness and sites such as Shiawassee that had highly
skewed species distributions. While at the quadrat scale there were few differences in the
relative distributions of species, sites differed greatly at the site-scale in ways that

correspond with observations.

125

ALLIARIA PE TIOLA TA INVASION PROCESSES
Effect of Site on A. petiolata Presence and Native Species Richness

Native species richness in the quadrats varied by site and A. petiolata presence or
absence, and the interaction between site and A. petiolatapresence was significant
(Figure 12, Table 9). The main effect of site was significant during all three years, and
the main effect of A. petiolata presence on native species richness was significant in 2004
and marginally significant in 2005. The most interesting finding fi'om these tests, though,
was the significance of the interaction between site and A. petiolata presence in 2004 and
2005. In the sites with higher species richness native species richness was higher in
quadrats where A. petiolata was absent, while at sites with the lowest species richness,
native species richness was highest in quadrats where A. petiolata was present. This
unexpected finding indicates that different processes are affecting A. petiolata at these
different sites. Specifically, it suggests that competition with residents may be reducing
the probability of A. petiolata establishment at the species rich sites, while at the species
poor sites resource limitation or environmental stress may be more important in

restricting A. petiolata establishment probability than competition.

Quadrat Level Correlations

The relationship of quadrat scale native species richness to relative A. petiolata
abundance (Figure 13) was variable and showed similar patterns to the categorical
analysis of A. petiolata presence or absence. The number of significant relationships
(from Spearman rank correlations at a = 0.05) increased from two in the spring of 2003

(Ft. Custer, Ives Rd.) to five in the spring of 2004 (F emwood, Ft. Custer, Ives Rd.,

126

Pinckney, Shiawassee) and six in spring of 2005 (Femwood, Ft. Custer., Ives Rd., Lux
Arbor, Pinckney, Shiawassee). The negative coefficients of these correlations suggest
that biotic resistance may play a role in directing A. petiolata invasions at some sites.
Contrary to the expectation set by Lundholm and Larson (2004), the results from
this and the next analysis lead to similar overall conclusions using absolute or relative
abundance data, although the correlations that used relative abundance data were a better
fit in several cases and were statistically significant more often. I have only presented the

relative abundance analyses in full for this reason.

Linking Invasion Patterns Across Scales

The varied patterns of A. petiolata invasion within sites are explained in part by
the native species richnesses of the sites. I evaluated the relationship of the Spearman
rank correlation coefficients from the quadrat scale correlations (Figure 13) from
previous section) with site-level native species richness using Spearman rank
correlations. These were significant during both 2004 and 2005 (Figure 14, P = 0.1361 ,
0.0165, and 0.0396 for 2003 - 2005 respectively). The analyses that used absolute A.
petiolata abundance (not shown) were similar, although the relationships were were only
significant during spring 2004 and were marginally significant in 2003 and 2005 (P =
0.0687, 0.0138, and 0.0621 for 2003 — 2005 respectively). The correlations between sites
Species richness and the correlation coefficients were negative in all cases for both
absolute and relative A. petiolata abundance.

These analyses suggest two general patterns. First, while both absolute and
relative abundances measures of A. petiolata are correlated with native species richness

in similar ways, the relative abundance model provides a better fit to the data. Because

127

the absolute abundance measure is unbounded by species richness, quadrats with any
number of species can have high absolute A. petiolata cover, but relative abundance must
decrease as a function of species richness. Thus, the variance in species richness should
be less when expressed as a function of relative A. petiolata abundance than of absolute
A. petiolata abundance and make for a better predictive model.

Second and more importantly, this shows that the relationship between quadrat
native species richness and A. petiolata abundance becomes more significant and
negative as the sites’ native species pools increase. There are two possible explanations
for this. Either A. petiolata causes species richness to decrease as it becomes more
abundant within quadrats at the more species rich sites or those quadrats with greater
species richness are resistant to A. petiolata invasion or dominance. This latter
explanation is in keeping with observations of biotic resistance in other systems at the
quadrat scale (Levine et a1. 2004). That the relationship is only significant at sites with
higher species richness suggests that there may be a threshold of species richness below
which biotic resistance is not effective or detectable. When coupled with the factorial
analysis of A. petiolata presence versus absence across the eight sites, the evidence favors
concluding that A. petiolata is responding to variation in species richness much more than
it is causing it because of differences in species richness between invaded and uninvaded
quadrats. Because the data do not allow analysis of change over time I cannot interpret
the causality of this relationship. However, observation of new invasions in previously
uninvaded quadrats may highlight correlations between species richness and invasion

probability.

128

Predicting Invasion Probability
Logistic Regression Analysis

Within the 57 initially uninvaded quadrats 33 became invaded by spring of 2004,
while 24 did not (Table 10). I used logistic regressions to test the hypothesis that native
and total species richness would be predictive of invasion probability within the quadrats
that were initially uninvaded and found that the probability of a quadrat becoming
invaded increased as its species richness decreased. The quasi-complete separation of
data from Pinckney prohibited asymptotic maximum likelihood estimation of logistic
regression parameters. However, exact logistic regression can estimate parameters using
alternative methods for data with complete or quasi-complete separation and can be
executed with the “exact” statement in PROC LOGISTIC (Derr 2000, SAS Institute
2001). Exact logistic regression with terms for invasion outcome in 2004 (‘Invade 2004’)
as the dependent variable and site as the independent variable was significant (Score test
statistic = 12.5051, Exact P = 0.03 75) when all seven sites are included (Femwood was
excluded from all invasion analyses because 20 of 20 quadrats were initially invaded).
However, the Pinckney site alone appeared to drive this relationship, as its removal
rendered insignificant both the exact regression and the conventional maximum
likelihood estimated logistic regression, which could then be run without separation
problems, of Invade 2004 on site, (Exact regression: Score test statistic 3.7315, Exact P =
0.6370; Maximum likelihood estimated logistic regression: Wald’s x2 = 3.5914, df = 5, P
= 0.6096). Site effects were not included in subsequent models based on these findings,

although data from the Pinckney site were retained to avoid bias from data selection.

129

To test for the relationship between initial native or total species richness and A.
petiolata invasion probability, I fit single-predictor logistic models to the data from the
57 initially uninvaded plots and found that:

( 1) Predicted logit of (Invade 2004) = 1.5780 + (—0.2556)*Native SR 2003
and:

(2) Predicted logit of (Invade 2004) = 1.5299 + (—0.2362)*Total SR 2003
Differences in model fit estimates and predictions were negligible, so only the fit
parameters for native species richness are given (Table 11) as they were marginally more
significant. These models indicate that the log of the odds of a sampling quadrat being
invaded in 2004 was negatively related to the native (P = 0.0296) (Figure 15) and total (P
= 0.0339) species richness in that quadrat in 2003. Put otherwise, the higher a quadrat’s
species richness was in 2003, the lower its probability of being invaded the following
year was. For an increase in native richness of one species, the probability of invasion
correspondingly decreased from 1.0 to 0.7746 (#2556 Table l 1). Similarly, for an
increase in native richness of five species, the odds of invasion decrease from 1.0 to
0.2786 (e5’[-0.2556] Table 11). Repeating the test in the same quadrats with species data
from 2005, when our plant identification ability had improved showed that:

Predicted logit of (Invade 2004) = 1.5990 + (—0.1908)*Species Richness 2005
This result was more highly significant (P = 0.0117), and the sign of the estimated effect
parameter is concurrent with the test of the 2003 data. The consistency of these finding
over several years as these invasions have progressed suggests that the differences
between these sets of quadrats did not necessarily result from A. petiolata invasion but,

rather, that they directed it.

130

PATTERNS OF INVASION AND POSSIBLE MECHANISMS

The changing relationship between A. petiolata presence (Figure 12) or
abundance (Figure 13) and native species richness across the eight sites suggests that
properties of either the sites themselves or the resident communities influence the relative
success of A. petiolata’s invasions. While species richness was negatively correlated with
A. petiolata abundance in the most species rich sites (Figures 13 & 14) and invasion
probability in 2004, at sites with low diversity there was a positive association between
species richness and A. petiolata presence and zero relationship with A. petiolata relative
abundance. This suggests a potential dichotomy between the “rich get richer” hypothesis
(Stohlgren et al. 2003), which predicts that areas with greater numbers of species and
higher resource availability are thus more likely to be invaded or colonized by additional
species, and biotic resistance, which predicts that those same areas should not be invasion
prone. The finding that exotic and native species richness are positively correlated at the
site scale (Figure 11) suggests that extrinsic factors like resource availability or habitat
stability make some sites generally better suited to plant growth than others. If resource
availability limits establishment probability, one would expect A. petiolata establishment
to be greatest within sites where the availability of resources was greatest. But
simultaneously, species richness within sites is likely positively correlated with resource
availability. At the sites with higher species richness, the increased survival probability
expected in resource rich areas is thus offset by increased competition from the resident
community, which is richer at those same locations. Others have shown that as species
richness increases, the availability of resources becomes locally reduced as the resident

species consume them (N aeem et al. 2000), and crowding by dense stands of residents

131

reduces suitable germination sites for arriving seeds (Kennedy et al. 2002) as per
Tilman’s R* (Tilman 1976) and stochastic niche (Tilman 2004) theories. True to these
findings, A. petiolata more fi'equently establishes within species rich sites in areas of
lower species richness. Conversely, at species poor site (i.e. Box Woodlot), the opposite
occurs, and A. petiolata establishes first in the areas with the greatest species richness
(Figure 12). The lower diversity of both native and exotic species (Figure 11) at these
sites suggests that resources or physiological stress likely limit plant growth, and that
survival probability within these sites is lowest where species richness is lowest.
Alliaria petiolata’s variable patterns of establishment did not prevent it from
spreading into new areas over time (Chapter 2). Within the species-rich sites, A. petiolata
first established in the species poor areas where competition was least limiting. Given
enough time, propagule pressure fiom the established invader populations should
eventually overwhelm the residents and allow A. petiolata to spread into areas where
establishment and survival probabilities were potentially limited by increased
competition with resident species. If A. petiolata is able to weaken competitors by
disrupting their mycorrhizal associations (Roberts and Anderson 2001, Stinson and
Klironomos 2005) and through direct allelopathy, it is possible that in time it could come
to dominate even the most “resistant” areas. At the species-poor sites, A. petiolata
initially became established where resource availability and competition were likely
greatest, but because the diversity in these sites was so low, competition was not
significant enough to limit or prevent establishment. Once established, populations of A.
petiolata could similarly produce a steady seed rain into the remaining lower diversity

areas within the sites where resource limitation likely reduced the probability of

132

establishment. Over time, mass effects of this continuous seed production would likely
result in some A. petiolata plants becoming established even in these less favorable areas
and ultimately spreading as I have observed (Chapter 2).

Biotic resistance (generally) appears to be especially important in limiting or
slowing A. petiolata invasions in sites with higher species richness, but resource
limitation seems to resist establishment of native and exotic species alike in sites with
lower species richness. Tilman’s (2004) stochastic niche theory predicts the biotic
resistance observed in the species rich sites, and Stohlgren et al’s (2003) “rich get richer”
hypothesis predicts what I observed at the lowest diversity sites, but I believe that the
progressive inversion of the relationship between local native richness and invader
presence or abundance across a larger-scale natural diversity gradient is novel and

suggests a possible union between the two hypotheses.

SUMMARY AND CONCLUSIONS

Populations of A. petiolata that invade natural areas in Michigan exhibit complex
interactions with resident species assemblages that vary both spatially and temporally. At
larger scales the relationship of native and exotic species richness is positive (Figure 11)
as others have previously shown. But in some communities I found a negative correlation
between A. petiolata abundance and native species richness which suggests that either A.
petiolata invasions lead to reductions in diversity at the quadrat scale (0.5 m2), that biotic
resistance locally impedes the invasion process, or that a combination of the two

scenarios is played out over time.

133

POTENTIAL IMPACTS

The patterns of native species richness and A. petiolata abundance at the
Femwood site where A. petiolata dominates the community indicate that A. petiolata is
likely negatively affecting native plant populations there. Allelopathic, competitive, and
antimycorrhizal effects are expected to be greatest where A. petiolata is widespread and
has high absolute abundance (Roberts and Anderson 2001). If A. petiolata invasions do
have negative impacts such as local extinctions, Femwood may be the best place to begin

looking for them among my study sites in future years.

ALLIARIA PETIOLA TA AND THEORIES OF INVASION PROCESSES

The size of sites’ species pools dictates the relationship between A. petiolata
presence or abundance and native species richness within quadrats. As the site-level
species pool increased, the relationship between quadrat-scale species richness and A.
petiolata abundance became increasingly negative. As this happened, species richness
shifted fi'om being greatest where A. petiolata was present to being greatest where it was
absent. One possible explanation of this correlation which would be in keeping with
Tilman’s stochastic niche theory (Tilman 2004) is that biotic resistance, mediated by
resource competition, is only biologically significant or detectable above a minimum
threshold of species richness. Higher species richness at the site level should indicate a
greater breadth of resources present but should be simultaneously correlated with a
greater partitioning of those resources between species. The odds of establishment thus
diminish as local species richness increases and the available portion of the resource pool

is reduced. The alternate hypothesis is that native species richness was more uniformly

134

distributed prior to A. petiolata invasion and that A. petiolata has significantly reduced
diversity where it has become abundant. However, my observation that the probability of
A. petiolata invading is greatest where local species richness is lowest indicates that
biotic resistance is primarily responsible for this pattern. My data collection did not
predate the initial invasions of these sites, so I cannot say which, if either, is correct. If A.
petiolata does in fact reduce diversity or richness, I should see the negative relationship
between quadrat-scale species richness and A. petiolata abundance degrade within the
more species rich sites as A. petiolata continues to spread.

A hardy species that can survive where resources are limiting to most others and
that has a plastic response to variation in resource availability should be able to (l)
colonize less favorable areas that other species cannot, (2) potentially perform better
where those resources are more abundant, and (3) increase when rare. Alliaria petiolata
appears to fit this profile as it has invaded even the least and most diverse sites and has
spread throughout them over time. If species richness is indicative of resource abundance
this indicates that A. petiolata is making tradeoffs between its competitive ability and
overcoming obstacles to establishment. Over time, natural selection should variably favor
hardier individuals with greater tolerance of resource stress (survivors) at species poor
sites and individuals with greater competitive abilities (competitors) at species rich sites.
This hypothesis could be tested empirically in the future by growing A. petiolata from
high and low diversity sites in a common garden experiment or by exchanging seeds
between species rich and species poor sites and evaluating their survival and fitness

across a range of conditions. Additional measurements of resources such as light, water,

135

space availability and soil properties at the eight current study sites could provide further
tests.

Taken as a whole, this suggests that biotic resistance and enhancement (the rich
get richer) are not diametrically opposing concepts, but, rather, are endpoints along a
continuum of tradeoffs between resource competition and stress tolerance against a
background of stochastic establishment and survival probabilities. Continued monitoring
at these eight sites may further elucidate the complex relationships between diversity,
competition, and invasibility and enrich current models which have only found such

tradeoffs in plant invasions across gradients of scale.

136

LITERATURE CITED

Allee, W. C. 1931. Animal aggregations. A study in general sociology. University of
Chicago Press, Chicago.

Bais, H. P., R. Vepachedu, S. Gilroy, R. M. Callaway, and J. M. Vivanco. 2003.
Allelopathy and exotic plant invasion: From molecules and genes to species
interactions. Science 301: 1377-1380.

Begon, M., J. L. Harper, and C. R. Townsend. 1996. Ecology : individuals, populations,
and communities, 3rd edition. Blackwell Science, Oxford, UK; Cambridge,
Mass.

Benoit, L. K., and R. A. Askins. 1999. Impact of the spread of Phragmites on the
distribution of birds in Connecticut tidal marshes. Wetlands 19:194-208.

Blossey, B., V. Nuzzo, H. Hinz, and E. Gerber. 2001. Developing biological control of
Alliaria petiolata (M. Bieb.) Cavara and Grande (garlic mustard). Natural Areas
Journal 21:357-367.

Bohlen, P. J., S. Scheu, C. M. Hale, M. A. McLean, S. Migge, P. M. Groffinan, and D.
Parkinson. 2004. Non-native invasive earthworms as agents of change in northern
temperate forests. Frontiers in Ecology and the Environment 2:427-435.

Callaway, R. M., and E. T. Aschehoug. 2000. Invasive plants versus their new and old
neighbors: a mechanism for exotic invasion. Science 290:521-523.

D'Antonio, C. M., and P. M. Vitousek.‘1992. Biological invasions by exotic grasses, the
grass/fire cycle, and global change. Annual Review of Ecology and Systematics
23:63-87.

Derr, R. E. 2000. Performing exact logistic regression with the SAS system. in SUGUI.

Elton, C. S. 1958. The ecology of invasions by animals and plants. Methuen & Co. Ltd.,
London.

Gotelli, N. J., and A. M. Ellison. 2004. A primer of ecological statistics. Sinauer
Associates, Inc. Publishers, Sunderland, Mass.

Hale, C. M., L. E. Frelich, and P. B. Reich. 2005. Exotic European earthworm invasion
dynamics in northern hardwood forests of Minnesota, USA. Ecological
Applications 15:848-860.

Herman, K. D., L. A. Masters, M. R. Penskar, A. A. Reznicek, G. S. Wilhelm, W. W.
Brodovich, and K. P. Gardiner. 2001. Floristic quality assessment with wetland
categories and examples of computer applications for the state of Michigan -
Revised, 2nd edition. Michigan Department of Natural Resources, Wildlife,
Natural Heritage Program, Lansing, MI.

137

Higgins, S. 1., S. Lavorel, and E. Revilla. 2003. Estimating plant migration rates under
habitat loss and fragmentation. Oikos 101:354-366.

Hinz, H., and E. Gerber. 2005. Summary of host specificity testing conducted in
Switzerland and Brassicaceae taxonomy. Pages 12-16 in L. C. Skinner, editor.
Symposium on the biology, ecology, and management of garlic mustard (Alliaria
petiolata) and European buckthom (Rhamnus cathartica). USDA Forest Service,
University of Minnesotta, St. Paul, MN.

Horsley, S. B., S. L. Stout, and D. S. DeCalesta. 2003. White-tailed deer impact on the
vegetation dynamics of a northern hardwood forest. Ecological Applications
13:98-1 18.

Hubbell, S. P. 2001. The unified neutral theory of biodiversity and biogeography.
Princeton University Press, Princeton, NJ.

Kalisz, S., T. Knight, L. Smith, J. Dunn, and J. Davis. 2003. Recipe for invasive spread:
venison with garlic mustard. in Invasive Plants in Natural and Managed Systems:
Linking Science and Management in conjunction with the 7th International
Conference on the Ecology and Management of Alien Plant Invasions, Wyndham
Bonaventure Resort, Ft. Lauderdale, F L.

Kedzie-Webb, S. A., R. L. Sheley, J. J. Borkowski, and J. J. Jacobs. 2001. Relationships
between Centaurea maculosa and indigenous plant assemblages. Western North
American Naturalist 61:43-49.

Kennedy, T. A., S. Naeem, K. M. Howe, J. M. H. Knops, D. Tilman, and P. Reich. 2002.
Biodiversity as a barrier to ecological invasion. Nature 417 :636-638.

Knops, J. M. H., D. Tilman, N. M. Haddad, S. Naeem, C. E. Mitchell, J. Haarstad, M. E.
Ritchie, K. M. Howe, P. B. Reich, E. Siemann, and J. Grotlr. 1999. Effects of
plant species richness on invasion dynamics, disease outbreaks, insect abundances
and diversity. Ecology Letters 2:286-293.

Koenig, W. D. 2003. European starlings and their effect on native cavity-nesting birds.
Conservation Biology 17:1134-1140.

Landis, D. A., D. W. Schemske, and J. A. Evans. 2004. Evaluating the potential for
biological control of garlic mustard (Alliaria petiolata (Bieb.) Cavara & Grande)
in Michigan. 2003 Progress Report. Michigan State University, East Lansing.

Levine, J. M., P. B. Adler, and S. G. Yelenik. 2004. A meta-analysis of biotic resistance
to exotic plant invasions. Ecology Letters 7 :97 5-989.

Levine, J. M., and C. M. D'Antonio. 1999. Elton revisited: a review of evidence linking
diversity and invasibility. Oikos 87: 1 5-26.

138

Lu, Z. J ., and K. P. Ma. 2005. Scale dependent relationships between native plant
diversity and the invasion of croftonweed (Eupatorium adenophorum) in
Southwest China. Weed Science 53:600-604.

Luken, J. O., L. M. Kuddes, and T. C. Tholemeier. 1997. Response of understory species
to gap formation and soil disturbance in Lonicera maackii thickets. Restoration
Ecology 5:229-235.

Lundholm, J. T., and D. W. Larson. 2004. Dominance as an overlooked measure of
invader success. Biological Invasions 6:505-510.

Mack, R. N., D. Sirnberloff, W. M. Lonsdale, H. Evans, M. Clout, and F . A. Bazzaz.
2000. Biotic invasions: Causes, epidemiology, global consequences, and control.
Ecological Applications 10:689-710.

Maerz, J. C., B. Blossey, J. K. Liebherr, and V. A. Nuzzo. 2002.
http://www.invasiveplants.net/impsal.htm Accessed January 24, 2006.

Magurran, A. E. 2004. Measuring biological diversity. Blackwell Pub., Malden, MA.

McCarthy, B. C. 1997. Response of a forest understory community to experimental
removal of an invasive nonindigenous plant (Alliaria petiolata, Brassicaceae).
Pages 117-130 in J. W. Thieret, editor. Assessment and Management of Plant
Invasions. Springer, NY.

MDNR. 2006. Proposed 2006-2010 deer population goals.
http://www.michigan.gov/dnr/0,1607,7-153-10363_10856_10905-129948--
,00.html. Michigan Department of Natural Resources.

Meekins, J. F., and B. C. McCarthy. 2000. Responses of the biennial forest herb Alliaria
petiolata to variation in population density, nutrient addition and light availability.
Journal of Ecology 88:447-463.

MNFI. 2003. Michigan's natural communities draft list and descriptions. Michigan
Natural Features Inventory, Lansing, MI.

Naeem, S., J. M. H. Knops, D. Tilman, K. M. Howe, T. Kennedy, and S. Gale. 2000.
Plant diversity increases resistance to invasion in the absence of covarying
extrinsic factors. Oikos 91:97-108.

Nuzzo, V., and B. Blossey. unpublished. Garlic mustard monitoring protocol.
http://www.invasiveplants.net/invasiveplants/InvasivePlants/GarlicMustard/ garlic
Mustard.asp.

Nuzzo, V. A. 1991. Experimental control of garlic mustard [Alliaria petiolata (Bieb.)
Cavara & Grande] in Northern Illinois using fire, herbicide, and cutting. Natural
Areas Journal 11:158-167.

139

 

Nuzzo, V. A. 1993. Current and historic distribution of garlic mustard (Alliaria Petiolata)
in Illinois. The Michigan Botanist 32:23-33.

Nuzzo, V. A. 1994. Response of garlic mustard (Alliaria-petiolata Bieb - Cavara and
Grande ) to summer herbicide treatment. Natural Areas Journal 14:309-310.

Nuzzo, V. A. 2000. Element stewardship abstract for Alliaria petiolata (Alliaria
ofiicinalis) garlic mustard. The Nature Conservancy.

Nuzzo, V. A., W. McClain, and T. Strole. 1996. Fire impact on groundlayer flora in a
sand forest 1990-1994. American Midland Naturalist 136:207-221.

1

Peng, C. Y. J ., K. L. Lee, and G. M. Ingersoll. 2002. An introduction to logistic
regression analysis and reporting. Journal of Educational Research 96:3-14.

"fi‘

Peng, C.-Y. J ., and T.-S. H. So. 1998. If there is a will, there is a way: Getting around
defaults of PROC LOGISTIC. Pages 243-252 in MidWest SAS Users Group
Conference. SAS Institute, Cary, NC.

Peng, C.-Y. J ., and T.-S. H. So. 2002. Logistic regression analysis and reporting: A
primer. Understanding Statistics 1:31-70.

Pimentel, D. 2005. Aquatic nuisance species in the New York State Canal and Hudson
River systems and the Great Lakes Basin: An economic and environmental
assessment. Environmental Management 35:692-701.

Pimentel, D., L. Lach, R. Zuniga, and D. Morrison. 2000. Environmental and economic
costs of nonindigenous species in the United States. Bioscience 50:53-65.

Prati, D., and O. Bossdorf. 2004. Allelopathic inhibition of germination by Alliaria
petiolata (Brassicaceae). American Journal of Botany 91:285-288.

Roberts, K. J ., and R. C. Anderson. 2001. Effect of garlic mustard Alliaria petiolata
(Beib. Cavara & Grande) extracts on plants and arbuscular mycorrhizal (AM)
firngi. American Midland Naturalist 146:146-152.

SAS Institute. 2001. The SAS System Version 8.2. Cary, NC.

Shea, K., and P. Chesson. 2002. Community ecology theory as a framework for
biological invasions. Trends in Ecology & Evolution 17:170-176.

Stachowicz, J. J ., R. B. Whitlatch, and R. W. Osman. 1999. Species diversity and
invasion resistance in a marine ecosystem. Science 286:1577-1579.

Stinson, K., and J. N. Klironomos. 2005. Exotic plant invasion degrades local
mycorrhizal association, alters community succession and limits restoration. in
Annual meeting of the Ecological Society of America, Montreal, Canada.

140

Stinson, K. A., S. A. Campbell, J. R. Powell, B. E. Wolfe, R. M. Callaway, G. C. Thelen,
S. G. Hallett, D. Prati, and J. N. Klironomos. 2005. Exotic plant invasion degrades
local mycorrhizal association, alters community succession and limits restoration.
(Poster). in 90th Annual Meeting of the Ecological Society of America, Montreal,
QC, Canada.

Stohlgren, T. J ., D. Barnett, C. Flather, J. Kartesz, and B. Peterjohn. 2005. Plant species
invasions along the latitudinal gradient in the United States. Ecology 86:2298-
2309.

Stohlgren, T. J ., D. T. Barnett, and J. Kartesz. 2003. The rich get richer: patterns of plant
invasions in the United States. Frontiers in Ecology and the Environment 1:11-14.

Stohlgren, T. J ., D. Binkley, G. W. Chong, M. A. Kalkhan, L. D. Schell, K. A. Bull, Y.
Otsuki, G. Newman, M. Bashkin, and Y. Son. 1999a. Exotic plant species invade
hot spots of native plant diversity. Ecological Monographs 69:25-46.

Stohlgren, T. J., K. A. Bull, Y. Otsuki, C. A. Villa, and M. Lee. 1998. Riparian zones as
havens for exotic plant species in the central grasslands. Plant Ecology 138:113-
125.

Stohlgren, T. J., G. w. Chong, L. D. Schell, K. A. Rimar, Y. Otsuki, M. Lee, M. A.
Kalkhan, and C. A. Villa. 2002. Assessing vulnerability to invasion by nonnative
plant species at multiple spatial scales. Environmental Management 29:566-577.

Stohlgren, T. J ., Y. Otsuki, C. A. Villa, M. Lee, and J. Belnap. 2001. Patterns of plant
invasions: A case example in native species hotspots and rare habitats. Biological
Invasions 3:37-50.

Stohlgren, T. J ., L. D. Schell, and B. Vanden Heuvel. 1999b. How grazing and soil
quality affect native and exotic plant diversity in rocky mountain grasslands.
Ecological Applications 9:45-64.

SYSTAT Inc. 2004. SYSTAT for Windows. Version 11.0. Chicago, IL.

Taft, J. B., G. S. Wilhelm, D. M. Ladd, and L. A. Masters. 1997. Floristic quality
assessment for vegetation in Illinois, a method for assessing vegetation integrityz.
Erigenia 15:3-23.

Tepley, A. J ., J. G. Cohen, and L. Huberty. 2004. Natural community abstract for
southern floodplain forest. Natural Features Inventory, Lansing, MI.

Tilman, D. 1976. Ecological competition between algae - experimental confirmation of
resource-based competition theory. Science 192:463-465.

141

 

Tilman, D. 2004. Niche tradeoffs, neutrality, and community structure: A stochastic
theory of resource competition, invasion, and community assembly. Proceedings
of the National Academy of Sciences of the United States of America 101:10854-
10861 .

US. Congress. 1993. Harmful non-indigenous species in the United States, OTA-F-565.
Office of Technology Assessment, Washington, DC.

Webb, S. L., T. H. Pendergast, and M. E. Dwyer. 2001. Response of native and exotic
maple seedling banks to removal of the exotic, invasive Norway maple (Acer
platanoides). Journal of the Torrey Botanical Society 128:141-149.

Wilcove, D. S., D. Rothstein, J. Dubow, A. Phillips, and E. Losos. 1998. Quantifying
threats to imperiled species in the United States. Bioscience 48:607-615.

Wilhelm, G. S., and L. A. Masters. 1999. Floristic quality assessment (ver. 1.0) with
Michigan database. Conservation Research Institute, Elrnhurst, IL.

Wolfe, B. E., and J. N. Klironomos. 2005. Breaking new ground: Soil communities and
exotic plant invasion. Bioscience 55:477-487.

142

 

 

88 B882 .88 Boise .68
ES 8% 58 ea 8. £8 8%; x83

8: seem £288: 883% A; .88 8.58:8 58 238 new .3338 .3288. 8.833% 8.3.2.3
$8 8a £3 .82. 58 .630

a: 38m Seesaw be 3283 .fi .28 v.8 233 .28 88 82m 0383.3- 0225.8. .26 Seem a;
88 x8 B. .8: 23a ewe .28 p.23

...: Seem Seesaw 2826a x83 .58 5%... 6.83 @218 23% 08825.8 0883.8. 8.355 35.85
“8.5m G$ Ego

new Seesaw b92333 8% AS .88 x8 SE... .8” .88 x8 Be $383.3- 083%.? ham .82. .3

2: seem Seesaw amozéa been 83me 3.8 B. .28 x8 22.5 LESS- .883; 8.53 33. a:
6.188
3%; ..83 .68 238 am... .68 ESE

8: Sea 805:8 eazém Mafia... .3» .E x8 as; .28 x8 82m 28228 0888.9. 885.3 .226 a
ES 82.. 18

has .38 38.2 5.8 as Sofia... .a:
FEES; .ce .3 .m8 3%... x83 .80

 

2. Case Seam 282.3 .5 8E .53 .88 x8 S. 5.8 3.8 22.5 228.38 :33. 3 Stem 8038......
am .88
Ame macaw ARV 2083.? AS .ecv mac .25
Q: mac... as: eoesom .sw .8 2&8 52L. .3 .8 2&8 Ema 0388.3- 0338.? Enema 8:83 Sm
«Ea—teen 23,—. 3:55:80 E22 A89 fine 82,—. .3230 325.39 ovata—3‘— 33an 3550 0:32 8%

 

 

 

.Goom M29): 83m .895 65 E55 mofim sewage?“ Roo— fioce we: .38 93 cars .958 0.8 83:58 3656
Son $25 b83588 Amoomv .6852: 838m 3532 newfiomz 93 £95 .3058 3&0th we 38803808 3.58 E30: .305
“a 6825:. ES nova—BE $635208 $6 558 £282: 836.5% .Vmo wqtouaofi €2-32 how 3528 8% gwfioaz .8 03am.

143

....zo: ...... 3.5.3..
850.5% .V wag—05 00.00% :0 .8 93 $9533 05 330.00% .V 3088 00.00% :0 .8 300 38 meta. 88.. 0033200

0.03 00065 0.5 000.2. .Avoom 5285 005.588 mo x005 .0013 bwbm u m. 3.005 .0935 0.588% .«0 ESEwB 3.80:
03.03: H Q5: 6.2.. 5.3 cowatomnsmv .25 860% 03.0: 8.. 2... $3833 05 860% :0 .8 $.00.» :3 02.82.02: 0:. 5.03
89m 00.03030 0.03 .0065 00.5 0.05. 0.09: 3:030 0:885 u NQK .Emumgmaoo «0 000.0508 8.0:. n 0 $8.30... 00.00am u Mm

144

 

 

$.20 .83 a... .3... ..N ...N ... ... a. N. 8.3%....
~32 8.2. 8.... .3... ... a SN ... ... N. .m ....o. ..3.
8:... $2... 3...... 3%.. 0: a... 0.. ... ... w. 35.2....
8&2 8.2... 5... ...... 0.2 a: ..N ... w. ... .82 .3
.82 .32 8.2 E ...N 5.. ...N 0.. ... 8. s. .3. ...;
9.2... .2 ..o 8.3 .8... N... ...N 0.... o... m. N. ....6 ....
3...... 8:2 83.. 2...... ....N 5.. ... ... 8. ... .838...
:83 883 ...... as...” 0.2 o... .... ... S 2 .2825 .am

...... u ...... Q..- 9..- .9. 2.0m 0 .0 5. 2% 2...

 

...... .8. ..o 8...... .... .86... 0.32,... ...... ...... :8 80982.. 2.... 8.8.8.... 3.35 ......oz .6 038m ... .3...

 

33..

3., N»...

E...

V.,

oommmBaEm
620m 35%
553:5

H354. 55

 

*3.

 

‘29

F.

 

......

F.

 

.1.

P.

 

is"

 

...... .3...

F.

F. m»...

 

F.

 

P”

cm 83
5650 .E
cock/Eon

8:503 xom
3.5 ESE;
038325
awoken 33m
38—05;

 

“one 55

 

 

.3...

P.

 

Nu...

 

 

 

P.

*

 

 

 

 

 

mo

3“ 83
83:0 .E
uooBEom

8:503 xom
3.5 :32
58>

-oN
2%

 

 

V

 

 

 

 

 

8 mo

 

 

 

 

 

mo

mo

 

 

 

mo

 

 

 

 

 

 

 

 

99539,“ HWS

18910:! 83113 v
O
Kenyan”

JOQJV an] V

IPH SEMI

Jalsno .JJ ‘3'
poo/“wad V
o

.... u :5 w m ,_. u 85 w m ”a :32: 9a 8%: 338:: .3. u 285

w R ”mm 950% 2a mos—«>4 wound—m3 fiEwéSfi ESmefi .83 watam 89¢ 0.8 383% 05 323 805 98 .Smw Em 88m 03

383:. 05 30% $38M Egcoommou 8cm can .38 .88 Sch 83%.— 32? no 98 .3 .8 was”; Saw 860% mo $83335va
8525.? 3322 ©5226 3583 28 8253 S88 5 3058mm? 8m 38“ 29:3 95 >o§m$80mo§ov~ mo 338% .n «3:.

145

Table 9. Two-way factorial GLM of quadrat native species richness by A. petiolata
presence or absence and Site as main efi‘ects.

 

 

 

 

Source Sum—of-Squares df Mean-Square F -ratio P

20032|
A. petiolata Presence 4.9124 1 4.9124 1.9365 0.1665
Site 316.1744 6 52.6957 20.7726 0.0000
A. petiolata x Site 24.9381 6 4.1563 1.6384 0.1419
Error 319.6357 126 2.5368

2004a
A. petiolata Presence 19.3326 1 19.3326 7.5952 0.0067
Site 757.2076 6 126.2013 49.5804 0.0000
A. petiolata x Site 46.0254 6 7.6709 3.0136 0.0087
Error 320.7187 126 2.5454

2005mb
A. petiolata Presence 12.5649 1 12.5649 3.1243 0.0800
Site 744.3778 5 148.8756 37.0185 0.0000
A. petiolata x Site 59.1424 5 11.8285 2.9412 0.0158
Error 434.3386 108 4.0217

 

a Femwood was excluded from analysis because all 20 quadrats were invaded.
b Pinckney was excluded from analysis because all 20 quadrats were invaded.

146

Table 10. Frequency of invasion events in 2004 in plots that were uninvaded in 2003.

 

 

 

Site Uninvaded Invaded Total
Box Woodlot l 3 4
F t. Custer 9 4 13
Ives Rd. 1 1 2
Lux Arbor 4 7 11
Pinckney 0 10 10
Russ Forest 6 5 11
Shiawassee 3 3 6
Total 24 33 57

 

 

147

Table 11. Logistic regression analysis of invasion probability by native species richness
in 57 sampling quadrats that were not invaded by A. petiolata in spring 2003.

 

 

 

 

Predictor [3 SE B Wald's x2 df P e " (odds ratio)
Constant 1.5780 0.6518 5.8601 1 0.0155 NA
Native Species Richness -0.2556 0.1175 4.7296 1 0.0296 0.774
Test x2 df P
Overall Model Evaluation

Likelihood ratio test 5.2418 1 0.0220

Score test 5.0948 1 0.0240

Wald test 4.7296 1 0.0296
Goodness-of-fit test

Hosmer & Lemeshow 8.8263 6 0.1836

 

Note SAS code: [PROC LOGISTIC; MODEL INV ADE_04=NS_03/CT ABLE PPROB = (0.1 T0 1.0 BY 0.1)
LACKFIT RSQ;]. Cox and Snell R 2 = 0.0879. Nagelkerke R2 (Max rescaled R 2 ) = 0.1181. Kendall’s Tau-a = 0.2377
Goodman-Kruskal Gamma = 0.3499. Somers’s D yx = 0.1768. c-statistic = 66.0%. NA = not applicable.

148

 

 

30 _ I l 1 I l i__
LAO rs='0.6347
P=0.0909
20 r- —
BWO 0 PR
RFC FC 00 SH r]
10 - 0 IR —4 '
ELFO 0 FW
0"] J l l L 1“

 

 

Site Native Species Richness

 

 

1O 20 3O 40 . 5O 60
Deer per km2

Figure 10. Correlation of site native species richness from 2003-2005 data and county-
wide deer-density. Spearman rank correlation coefficient and probability are shown.

149

 

 

8 j I i I I I _
g 7 _. LA 0 ‘-
8
{g 6 - FC 00 SH -
E 5 __ rs = 0.9729 _
g 4 P =0.0002 0
'3 without Lux Arbor 'R “
a? 3" PR '"
g 2 - o 0 FW —
0
I11 1 — BWo RF 0 _
0 " I I I I I I

 

 

 

10 20 30 4O 50 60
Native Species Richness

Figure 11. Spearman rank correlation of exotic versus native species richness at the site
level. The number of exotic species observed at Lux Arbor (LA) increased
disproportionately following extensive soil and canopy disturbance during logging
activities in spring 2005. I excluded Lux Arbor from this analysis for this reason.

150

 

90.60% 000023: me 03:50 $500,800 05 0:0 002% 0.33330

.V .3 003000 003 0030 0:80 00 0:5 0>o 38003 00035:: E 00§E> 0003000 2:. .0000» 3800 E03608 0%.. 080309 =E0>o
:5 5:50 mam—@800 @3895 000—000 0000530 860% 0:0 3 0.30% c0030; 0005.0me .0000 $5950 05 :0 0:0 009 002030
860% 0300: me 0098:: 39 05 389 SSSSQ ..V 3223 0:0 55 306020 E Amm a." 0025 8230i 0.200% 0.50 Z .N— earn

ww0ccoE 00808 0282 05
0400800892 0 0400800832 0 300800892 0

_a__4___q_a _ _ ..fin _ ....n.
7 LI II
(I.

 

O

\\H
"E
I I 30 I

V

 

ma

I w :00“. m. I

co

 

2080 0588 .< u

 

l

I
J
N
F

20005 0.03.03 .< n IIIII

moom

 

 

 

l

P

sseuuogu segoeds eAgteN teipeno

 

 

 

 

 

 

 

(D
,—

_ _ _ b _ _

 

151

 

BoxVWodd

 

eouepunqv aAgieIeu manned 'v

       

 

 

4.3121382812130231‘215
Quadrat Native Species Richness

Figure 13. Quadrat native species richness versus A. petiolata relative abundance. P-
values 5 0.1 from Spearman rank correlations are shown. Least square regression lines
are overlaid to indicate the linear trend from each correlation.

152

Spring 200 Spring 2004 Spring 2005

- - *f ' ' I"0.5

 

I

           

rs = ~0.5749 rs = -0.8024 . rs = -0.7306
P1P = 01.136] 1 1" h: = 01.01615 l hip: 01.039? 1 1‘ '1-0

010203040010203040010203040
Site Native Species Richness

 

 

 

 

 

 

tuegoggeoo uoueIeuoo

 

Figure 14. Relationship of correlation coefficients (from figure 13) to site native species
richness. Spearman rank correlation coefficients are shown. Least square regression lines
are overlaid to indicate the linear trend fiom each correlation.

153

 

 

 

Probability of Invasion

 

 

VAGD-t-XO

A D X
0'0_l [‘5 l <1<1§ZIA<II><1 IXX? XXX XXI“

O 2 4 6 8 1O
Qiadrat Native Species Richness

 

Figure 15. Predicted and observed outcomes of invasion in 2004 of quadrats that were
A. petiolata free in 2003. Lines represent predicted invasion probabilities and 95%
confidence intervals. Quadrats that were invaded in 2004 are show at the top of the
frame, while those that were not invaded are at the bottom. Observed data points are

jittered to reveal overlapping points.

154

 

APPENDIX 1

155

Appendix 1. Species inventories for all sites from Spring 2003 - Fall 2005. Capitalized
scientific names indicate non-native species. Values for C and W indicate coefficients
of conservatism and wetness used in floristic quality assessment. Prefixes under
physiognomy describe life cycle of species as biennial (B), perennial (P), or annual (A)
and growth form as herbaceous (H) or woody (W).

 

 

 

Scientific Name Common Name C W PHYS
Box Woodlot

Acer saccharinum SILVER MAPLE 2 -3 Nt Tree
Acer saccharum SUGAR MAPLE 5 3 Nt Tree
ALLIARIA PE NOLA TA GARLIC MUSTARD "‘ 0 Ad B-Forb
Arisaema triphyllum JACK-IN-THE-PULPIT 5 -2 Nt P-Forb
Circaea Iutetiana ENCHANTER'S-NIGHTSHADE 2 3 Nt P-Forb
Erythronium americanum YELLOW TROUT LILY 5 S Nt P-Forb
F raxinus pennsylvanica RED ASH 2 -3 Nt Tree
Geum canadense WHITE AVENS l O Nt P-Forb
Parthenocissus quinquefolia VIRGINIA CREEPER 5 1 Nt W-Vine
Prunus serotina WILD BLACK CHERRY 2 3 Nt Tree
Quercus alba WHITE OAK 5 3 Nt Tree
Quercus macrocarpa BUR OAK 5 l Nt Tree
Ribes cynosbati PRICKLY or WILD GOOSEBERRY 4 S Nt Shrub
Rubus allegheniensis COMMON BLACKBERRY 1 2 Nt Shrub
Sambucus racemosa RED-BERRIED ELDER 3 2 Nt Shrub
T oxicodendron radicans POISON-IVY 2 -l Nt W-Vine
Vitis riparia RIVERBANK GRAPE 3 -2 Nt W-Vine
Femwood

ALLIARIA PE T [OLA TA GARLIC MUSTARD * 0 Ad B-F orb
Arisaema triphyllum JACK-IN-THE-PULPIT 5 -2 Nt P-Forb
Asimina triloba PAWPAW 9 0 Nt Tree
Aster cordifolius HEART-LEAVED ASTER 4 S Nt P-Forb
Carex blanda SEDGE l O Nt P-Sedge
Catya ovata SHAGBARK HICKORY 5 3 Nt Tree
Celtis occidentalis HACKBERRY 5 1 Nt Tree
Circaea lutetiana ENCHANTER'S-NIGHTSHADE 2 3 Nt P-Forb
Dentaria laciniata CUT-LEAVED TOOTHWORT 5 3 Nt P-F orb
Euonymus obovata RUNNING STRAWBERRY BUSH 5 S Nt Shrub
Eupatorium rugosum WHITE SNAKEROOT 4 3 Nt P-Forb
F raxinus americana WHITE ASH 5 3 Nt Tree

F raxinus pennsylvanica RED ASH 2 -3 Nt Tree
Galium aparine ANNUAL BEDSTRAW 0 3 Nt A-Forb
Galium asprellum ROUGH BEDSTRAW 5 -S Nt P-Forb
Geum canadense WHITE AVENS l 0 Nt P-F orb
Glyceria striata F OWL MANNA GRASS 4 -S Nt P-Grass
Hepatica americana ROUND-LOBED HEPATICA 6 S Nt P-Forb
Hydrophyllum appendiculatum GREAT WATERLEAF 7 S Nt P-Forb
Hystrix panda BOTTLEBRUSH GRASS 5 S Nt P-Grass
Juglans nigra BLACK WALNUT 5 3 Nt Tree
LAM! UM PURPURE UM PURPLE DEAD-NETTLE * 5 Ad A-Forb
Lindera benzoin SPICEBUSH 7 -2 Nt Shrub
Osmorhiza cIaytonii HAIRY SWEET-CICELY 4 4 Nt P-Forb
Osmorhiza Iongistylis SMOOTH SWEET-CICELY 3 4 Nt P-F orb

156

 

Appendix 1. (cont'd)

 

 

Scientific Name Common Name C W PHYS
Parthenocissus quinquefolia VIRGINIA CREEPER 5 1 Nt W-Vine
Pilea pumila CLEARWEED 5 -3 Nt A-Forb
Podophyllum peltatum MAY APPLE 3 3 Nt P-F orb
Polygonatum biflorum SOLOMON-SEAL 4 3 Nt P-Forb
Polygonatum pubescens DOWNY SOLOMON SEAL 5 S Nt P-F orb
Polygonum virginianum JUMPSEED 4 O Nt P-F orb
Prunus serotina WILD BLACK CHERRY 2 3 Nt Tree
Quercus alba WHITE OAK 5 3 Nt Tree
Ranunculus abortivus SMALL-FLOWERED 0 -2 Nt A-Forb
Ribes cynosbati PRICKLY or WILD GOOSEBERRY 4 S Nt Shrub
ROSA MULHFLORA MULTIFLORA ROSE * 3 Ad Shrub
Sanicula gregaria BLACK SNAKEROOT 2 -1 Nt P-F orb
Sassafi'as albidum SASSAFRAS 5 3 Nt Tree
Smilacina racemosa FALSE SPIKENARD 5 3 Nt P-F orb
Smilax tamnoides BRISTLY GREEN-BRIER 5 O Nt W-Vine
Solidago caesia BLUE-STEMMED GOLDENROD 7 3 Nt P-Forb
T oxicodendron radicans POISON-IVY 2 -1 Nt W-Vine
Trillium grandiflorum COMMON TRILLIUM 5 S Nt P-Forb
Ulmus americana AMERICAN ELM l -2 Nt Tree
Viola canadensis CANADA VIOLET 5 S Nt P-F orb
Viola cucullata MARSH VIOLET 5 -S Nt P-F orb
Viola pubescens YELLOW VIOLET 4 4 Nt P-Forb
Viola sororia COMMON BLUE VIOLET l 1 Nt P-Forb
Vitis riparia RIVERBANK GRAPE 3 -2 Nt W-Vine
Ft. Custer
Acer rubrum RED MAPLE l O Nt Tree
Acer saccharum SUGAR MAPLE 5 3 Nt Tree
ALLIARLA PE TIOLA TA GARLIC MUSTARD * 0 Ad B-F orb
Amelanchier laevis SMOOTH SHADBUSH 4 S Nt Tree
Amphicatpaea bracteata HOG-PEANUT 5 O Nt A-F orb
Arisaema triphyllum JACK-IN-THE-PULPIT 5 -2 Nt P-F orb
Aster lanceolatus EASTERN LINED ASTER 2 -3 Nt P-Forb
BERBERIS IHUNBERGII JAPANESE BARBERRY * 4 Ad Shrub
Carex cephalophora SEDGE 3 3 Nt P-Sedge
Celtis occidentalis HACKBERRY 5 1 Nt Tree
Cinna arumlinacea WOOD REEDGRASS 7 -3 Nt P-Grass
Circaea lutetiana ENCHANTER'S-NIGHTSHADE 2 3 Nt P-Forb
Camus foemina GRAY DOGWOOD l -2 Nt Shrub
Cryptotaenia canadensis HONEWORT 2 O Nt P-Forb
Desmodium glutinosum CLUSTERED—LEAVED TICK- 5 S Nt P-F orb
TREF OIL
Desmodium nudiflorum NAKED TICK-TREFOIL 7 S Nt P-Forb
Elymus villosus SILKY WILD-RYE 5 3 Nt P—Grass
F estuca subverticillata NODDING F ESCUE 5 2 Nt P-Grass
F raxinus pennsylvanica RED ASH 2 -3 Nt Tree
Galium aparine ANNUAL BEDSTRAW O 3 Nt A-Forb
Galium concinnum SHINING BEDSTRAW 5 3 Nt P-F orb
Galium pilosum HAIRY BEDSTRAW 6 S Nt P-F orb
Galium trzflorum FRAGRANT BEDSTRAW 4 2 Nt P-Forb

157

 

Appendix 1. (cont'd)

 

 

 

Scientific Name Common Name C W PHYS
Geum canadense WHITE AVENS 1 O Nt P-F orb
Impatiens pallida PALE TOUCH-ME-NOT 6 -3 Nt A-F orb
Lindera benzoin SPICEBUSH 7 -2 Nt Shrub
Onoclea sensibilis SENSITIVE FERN 2 -3 Nt Fern
Osmorhiza claytonii HAIRY SWEET-CICELY 4 4 Nt P-Forb
Osmorhiza longistylis SMOOTH SWEET-CICELY 3 4 Nt P-Forb
Parthenocissus quinquefolia VIRGINIA CREEPER 5 l Nt W-Vine
Phlox divaricata WOODLAND PHLOX 5 3 Nt P-Forb
Phryma Ieptostachya LOPSEED 4 S Nt P-F orb
Pilea pumila CLEARWEED 5 -3 Nt A-F orb
Polygonatum pubescens DOWNY SOLOMON SEAL 5 S Nt P-Forb
Polygonum virginianum JUMPSEED 4 O Nt P-Forb
Prunus serotina WILD BLACK CHERRY 2 3 Nt Tree
Quercus alba WHITE OAK 5 3 Nt Tree
Quercus velutina BLACK OAK 6 S Nt Tree
RHAMVUS CA IHART [CA COMMON BUCKTHORN * 3 Ad Tree
Ribes cynosbati PRICKLY or WILD GOOSEBERRY 4 S Nt Shrub
ROSA MULTIFLORA MULTIFLORA ROSE * 3 Ad Shrub
Rubus allegheniensis COMMON BLACKBERRY 1 2 Nt Shrub
Rubus occidentalis BLACK RASPBERRY l S Nt Shrub
R UMEX OB TUSIF 0L1 US BITTER DOCK “ -3 Ad P-F orb
Sambucus racemosa RED—BERRIED ELDER 3 2 Nt Shrub
Sanicula gregaria BLACK SNAKEROOT 2 -l Nt P-F orb
Sassafi'as albidum SASSAFRAS 5 3 Nt Tree
Senecio obovatus ROUND-LEAVED RAGWORT 10 4 Nt P-F orb
Smilacina racemosa FALSE SPIKENARD 5 3 Nt P-Forb
Smilax lasioneura CARRION-FLOWER 5 S Nt W-Vine
Solidago caesia BLUE-STEMMED GOLDENROD 7 3 Nt P-Forb
SIELLARIA MEDIA COMMON CHICKWEED " 3 Ad A-Forb
T oxicodendron radicans POISON-IVY 2 -l Nt W—Vine
Ulmus americana AMERICAN ELM l -2 Nt Tree
Urtica dioica NETTLE 1 -l Nt P-Forb
Viola canadensis CANADA VIOLET 5 S Nt P-Forb
Viola pubescens YELLOW VIOLET 4 4 Nt P-F orb
Ives Road

Acer rubrum RED MAPLE l 0 Nt Tree
Acer saccharum SUGAR MAPLE S 3 Nt Tree
Agrimonia pubescens SOFT AGRIMONY 5 S Nt P-Forb
ALLIARIA PE NOLA TA GARLIC MUSTARD "‘ 0 Ad B-Forb
Allium canadense WILD GARLIC 4 3 Nt P-F orb
Allium tricoccum WILD LEEK 5 2 Nt P-Forb
Anlsaema triphyllum JACK-IN-THE-PULPIT 5 -2 Nt P-F orb
Carex blanda SEDGE l O Nt P-Sedge
Carex grisea SEDGE 3 -3 Nt P-Sedge
Carex hirtifolia SEDGE 5 S Nt P-Sedge
Carya ovata SHAGBARK HICKORY 5 3 Nt Tree
Celtis occidentalis HACKBERRY 5 1 Nt Tree
Circaea lutetiana ENCHANTER'S-NIGHTSHADE 2 3 Nt P-Forb
Claytonia virginica SPRING-BEAUTY 4 3 Nt P-Forb

158

 

 

Appnedix 1. (cot'd)

 

 

Scientific Name Common Name C W PHYS
Cryptotaenia canadensis HONEWORT 2 0 Nt P-Forb
F estuca subverticillata NODDING F ESCUE 5 2 Nt P-Grass
F raxinus americana WHITE ASH 5 3 Nt Tree
F raxinus pennsylvanica RED ASH 2 -3 Nt Tree
Geum canadense WHITE AVENS 1 O Nt P-Forb
HESPERIS MATRONALIS DAME'S ROCKET * 5 Ad P-Forb
Laportea canadensis WOOD NETTLE 4 -3 Nt P-Forb
Lindera benzoin SPICEBUSH 7 -2 Nt Shrub
LONICERA MAACKII AMUR HONEYSUCKLE " 5 Ad Shrub
Moms rubra RED MULBERRY 9 l Nt Tree
Osmorhiza claytonii HAIRY SWEET-CICELY 4 4 Nt P-F orb
Osmorhiza longistylis SMOOTH SWEEToCICELY 3 4 Nt P-Forb
Panhenocissus quinquefolia VIRGINIA CREEPER 5 1 Nt W-Vine
Pilea pumila CLEARWEED 5 -3 Nt A-Forb
Podophyllum peltatum MAY APPLE 3 3 Nt P-F orb
Polygonatum pubescens DOWNY SOLOMON SEAL 5 5 Nt P-Forb
Polygonum virginianum JUMPSEED 4 0 Nt P-Forb
Prunus serotina WILD BLACK CHERRY 2 3 Nt Tree
Prunus virginiana CHOKE CHERRY 2 l Nt Shrub
Ribes cynosbati PRICKLY or WILD GOOSEBERRY 4 S Nt Shrub
ROSA MULTIFLORA MULTIFLORA ROSE " 3 Ad Shrub
Sambucus canadensis ELDERBERRY 3 -2 Nt Shrub
Sambucus racemosa RED-BERRIED ELDER 3 2 Nt Shrub
Sanicula gregaria BLACK SNAKEROOT 2 -l Nt P-Forb
Smilacina racemosa FALSE SPIKENARD 5 3 Nt P-Forb
Smilax tamnoides BRISTLY GREEN-BRIER 5 0 Nt W-Vine
Symplocarpusfoetidus SKUNK-CABBAGE 6 -S Nt P-Forb
T oxicodendron radicans POISON-IVY 2 -l Nt W-Vine
Trillium grandiflorum COMMON TRILLIUM 5 S Nt P-F orb
Ulmus americana AMERICAN ELM l -2 Nt Tree
Viola cucullata MARSH VIOLET 5 -5 Nt P-Forb
Viola pubescens YELLOW VIOLET 4 4 Nt P-Forb
Viola striata CREAM VIOLET 5 -3 Nt P-Forb
Vitis riparia RIVERBANK GRAPE 3 -2 Nt W-Vine
Lux Arbor
Acer rubrum RED MAPLE l 0 Nt Tree
Acer saccharum SUGAR MAPLE 5 3 Nt Tree
Actaea pachypoda DOLL'S-EYES 7 5 Nt P-F orb
ALLIARIA PE T IOLA TA GARLIC MUSTARD " 0 Ad B-Forb
Arisaema triphyllum JACK-IN-THE-PULPIT 5 -2 Nt P-Forb
Carex cephalophora SEDGE 3 3 Nt P-Sedge
CI-ENOPODI UM ALBUM LAMB'S QUARTERS "' 1 Ad A—Forb
Circaea lutetiana ENCHANTER'S-NIGHTSHADE 2 3 Nt P-F orb
Erechtites hieracifolia FIREWEED 2 3 Nt A-Forb
ERUCA VESICARIA ROCKET SALAD; GARDEN " 5 Ad A-Forb
SALAD
Galium aparine ANNUAL BEDSTRAW 0 3 Nt A-Forb
Galium tnflorum FRAGRANT BEDSTRAW 4 2 Nt P-F orb
Geranium maculatum WILD GERANIUM 4 3 Nt P-Forb

159

 

 

Appendix 1. (cont'd)

 

 

Scientific Name Common Name C W PHYS
Geranium robertianum HERB ROBERT 3 S Nt A-Forb
Geum canadense WHITE AVENS 1 O Nt P-F orb
Juncus tenuis PATH RUSH l O Nt P-Forb
LE ONUR US CARDIA CA MOTHERWORT * 5 Ad P-F orb
Oxalis striata COMMON YELLOW WOOD- 0 3 Nt P-F orb
SORREL
Parthenocissus quinquefolia VIRGINIA CREEPER 5 l Nt W-Vine
Phytolacca americana POKEWEED 2 l Nt P-Forb
Pilea pumila CLEARWEED 5 -3 Nt A-Forb
Podophyllum peltatum MAY APPLE 3 3 Nt P-Forb
Polygonum virginianum JUMPSEED 4 O Nt P-F orb
Prunus serotina WILD BLACK CHERRY 2 3 Nt Tree
Quercus velutina BLACK OAK 6 S Nt Tree
Ribes cynosbati PRICKLY or WILD GOOSEBERRY 4 5 Nt Shrub
Rubus allegheniensis COMMON BLACKBERRY l 2 Nt Shrub
Rubus occidentalis BLACK RASPBERRY l S Nt Shrub
RUMEX OBTUSIFOLIUS BITTER DOCK " -3 Ad P-Forb
Senecio aureus GOLDEN RAGWORT 5 -3 Nt P-Forb
Solidago caesia BLUE-STEMMED GOLDENROD 7 3 Nt P-Forb
T ARAM C UM OF FICINALE COMMON DANDELION * 3 Ad P-Forb
Halictrum darycarpum PURPLE MEADOW-RUE 3 -2 Nt P-Forb
T oxicodendron radicans POISON-IVY 2 -l Nt W—Vine
ZRIF 0L1 UM REPENS WHITE CLOVER “ 2 Ad P-Forb
Urtica dioica NETTLE l -l Nt P-Forb
Viola pubescens YELLOW VIOLET 4 4 Nt P-Forb
Vitis riparia RIVERBANK GRAPE 3 -2 Nt W-Vine
Pinckney
Acer negundo BOX ELDER O -2 Nt Tree
Acer rubrum RED MAPLE l 0 Nt Tree
Acer saccharum SUGAR MAPLE 5 3 Nt Tree
ALLIARIA PE NOLA TA GARLIC MUSTARD "' 0 Ad B—Forb
Amelanchier arborea JUN EBERRY 4 3 Nt Tree
Arisaema triphyllum J ACK-IN-THE-PULPIT 5 -2 Nt P-Forb
Carex pensylvanica SEDGE 4 5 Nt P-Sedge
Carex rosea CURLY-STYLED WOOD SEDGE 2 5 Nt P-Sedge
Carya ovata SHAGBARK HICKORY 5 3 Nt Tree
Celtis occidentalis HACKBERRY 5 l Nt Tree
Circaea lutetiana ENCHANTER’S-NIGHT SHADE 2 3 Nt P-Forb
Cornusfoemina GRAY DOGWOOD 1 -2 Nt Shrub
Desmodium glutinosum CLUSTERED—LEAVED TICK- 5 5 Nt P-Forb
TREFOIL
Dioscorea villosa WILD YAM 4 1 Nt P-Forb
F raxinus americana WHITE ASH 5 3 Nt Tree
Galium aparine ANNUAL BEDSTRAW O 3 Nt A-Forb
Galium concinnum SHINING BEDSTRAW 5 3 Nt P-Forb
Geranium maculatum WILD GERANIUM 4 3 Nt P-Forb
Geum canadense WHITE AVENS l O Nt P-Forb
Hepatica americana ROUND-LOBED HEPATICA 6 5 Nt P-Forb
Parthenocissus quinquefolia VIRGINIA CREEPER 5 1 Nt W-Vine

160

_-—a'

 

I'-

 

Appendix 1. (cont'd)

 

 

 

Scientific Name Common Name C W PHYS
Podophyllum peltatum MAY APPLE 3 3 Nt P-F orb
Potentilla simplex OLD-FIELD CINQUEF OIL 2 4 Nt P-F orb
Prunus serotina WILD BLACK CHERRY 2 3 Nt Tree
Prunus virginiana CHOKE CHERRY 2 l Nt Shrub
Quercus alba WHITE OAK 5 3 Nt Tree
Quercus rubra RED OAK 5 3 Nt Tree
ROSA MULTIFLORA MULTIFLORA ROSE * 3 Ad Shrub
Rubus allegheniensis COMMON BLACKBERRY l 2 Nt Shrub
Rubusflagellaris NORTHERN DEWBERRY l 4 Nt Shrub
Sassafias albidum SASSAFRAS 5 3 Nt Tree
Smilacina racemosa FALSE SPIKENARD 5 3 Nt P-Forb
Ihalictrum dioicum EARLY MEADOW-RUE 6 2 Nt P-Forb
T oxicodendron radicans POISON-IVY 2 -l Nt W-Vine
Trillium grandiflorum COMMON TRILLIUM 5 5 Nt P-F orb
Uvularia grandiflora BELLWORT 5 5 Nt P-Forb
Viburnum opulus var. americanum HIGI-IBUSH CRANBERRY 5 -3 Nt Shrub
Viola pubescens YELLOW VIOLET 4 4 Nt P-Forb
Vitis aestivalis SUMMER GRAPE 6 3 Nt W-Vine
Russ Forest

Acer rubrum RED MAPLE l O Nt Tree
Acer saccharum SUGAR MAPLE 5 3 Nt Tree
ALLIARIA PE TIOLA TA GARLIC MUSTARD “ 0 Ad B-Forb
Carexjamesii JAMES' SEDGE 8 S Nt P-Sedge
Celtis occidentalis HACKBERRY 5 l Nt Tree
Circaea lutetiana ENCHANTER'S-NIGHTSHADE 2 3 Nt P-F orb
Claytonia virginica SPRING-BEAUTY 4 3 Nt P-F orb
Dentaria laciniata CUT-LEAVED TOOTHWORT 5 3 Nt P-Forb
Desmodium nudiflorum NAKED TICK-TREFOIL 7 5 Nt P-Forb
Dicentra canadensis SQUIRREL CORN 7 5 Nt P-Forb
Euonymus obovata RUNNING STRAWBERRY BUSH 5 S Nt Shrub
F raxinus pennsylvanica RED ASH 2 -3 Nt Tree
Galium aparine ANNUAL BEDSTRAW 0 3 Nt A-Forb
Geum canadense WHITE AVENS l 0 Nt P-Forb
Impatiens capensis SPOTTED TOUCH-ME-NOT 2 -3 Nt A-Forb
Lindera benzoin SPICEBUSH 7 -2 Nt Shrub
Maianthemum canadense CANADA MAYFLOWER 4 O Nt P-Forb
Monotropa uniflora INDIAN PIPE 5 3 Nt P-Forb
Osmorhiza claytonii HAIRY SWEET-CICELY 4 4 Nt P-Forb
Osmorhiza longistylis SMOOTH SWEET-CICELY 3 4 Nt P-F orb
Parthenocissus quinquefolia VIRGINIA CREEPER 5 1 Nt W-Vine
Phytolacca americana POKEWEED 2 1 Nt P-Forb
Polygonatum pubescens DOWNY SOLOMON SEAL 5 5 Nt P-Forb
Prunus serotina WILD BLACK CHERRY 2 3 Nt Tree
Quercus alba WHITE OAK 5 3 Nt Tree
Rubus allegheniensis COMMON BLACKBERRY 1 2 Nt Shrub
Smilacina racemosa FALSE SPIKENARD 5 3 Nt P-Forb
T oxicodendron radicans POISON-IVY 2 -l Nt W-Vine
Ulmus americana AMERICAN ELM l -2 Nt Tree

161

 

Appendix 1. (cont'd)

 

 

Scientific Name Common Name C W PHYS
Viola cucullata MARSH VIOLET 5 -5 Nt P—Forb
Viola rostrata LONG-SPURRED VIOLET 6 3 Nt P-F orb
Viola striata CREAM VIOLET 5 -3 Nt P-F orb
Shiawassee

Acer negundo BOX ELDER O -2 Nt Tree
Acer saccharinum SILVER MAPLE 2 -3 Nt Tree
ALLIARIA PE HOLA TA GARLIC MUSTARD * 0 Ad B-Forb
Allium canadense WILD GARLIC 4 3 Nt P-Forb
Anemone canadensis CANADA ANEMONE 4 -3 Nt P-F orb
Arisaema dracontium GREEN DRAGON 8 -3 Nt P-Forb
Arisaema triphyllum JACK-IN-THE-PULPIT 5 -2 Nt P-F orb
Aster Ianceolatus EASTERN LINED ASTER 2 -3 Nt P-F orb
Aster ontarionis ONTARIO ASTER 6 O Nt P-Forb
Boehmeria cylindrica FALSE NETTLE 5 -S Nt P-Forb
BRASSICA NIGRA BLACK MUSTARD * 5 Ad A-Forb
Carex blanda SEDGE l 0 Nt P-Sedge
Carex grayi SEDGE 6 -4 Nt P-Sedge
Carex grisea SEDGE 3 -3 Nt P-Sedge
Celtis occidentalis HACKBERRY 5 1 Nt Tree
Circaea lutetiana ENCHANTER'S-NIGHTSHADE 2 3 Nt P-Forb
Crataegus punctata DOTTED HAWTHORN ' l S Nt Tree
Dioscorea villosa WILD YAM 4 1 Nt P-Forb
Elymus villosus SILKY WILD-RYE 5 3 Nt P-Grass
Elymus virginicus VIRGINIA WLD-RYE 4 -2 Nt P-Grass
Erigeron philadelphicus MARSH FLEABANE 2 -3 Nt P-Forb
Eupatorium rugosum WHITE SNAKEROOT 4 3 Nt P-Forb
F raxinus pennsylvanica RED ASH 2 -3 Nt Tree
Galium aparine ANNUAL BEDSTRAW O 3 Nt A-Forb
Geranium maculatum WILD GERANIUM 4 3 Nt P-Forb
Geum canadense WHITE AVENS 1 O Nt P-Forb
Impatiens capensis SPOTTED TOUCH-ME-NOT 2 -3 Nt A-F orb
Juglans nigra BLACK WALNUT 5 3 Nt Tree
Laportea canadensis WOOD NETTLE 4 -3 Nt P-Forb
Lobelia siphilitica GREAT BLUE LOBELIA 4 -4 Nt P-F orb
LONICERA MAACKII AMUR HONEYSUCKLE * 5 Ad Shrub
Lysimachia ciliata FRINGED LOOSESTRIFE 4 -3 Nt P-F orb
LYSIMA CHIA NUMMULARIA MONEYWORT " -4 Ad P-Forb
Parthenocissus quinquefolia VIRGINIA CREEPER 5 l Nt W-Vine
Pilea pumila CLEARWEED 5 -3 Nt A-Forb
Polygonatum biflorum SOLOMON-SEAL 4 3 Nt P-Forb
Polygonatum pubescens DOWNY SOLOMON SEAL 5 5 Nt P-Forb
Polygonum virginianum JUMPSEED 4 0 Nt P-Forb
Prunus serotina WILD BLACK CHERRY 2 3 Nt Tree
Quercus velutina BLACK OAK 6 5 Nt Tree
Ranunculus hispidus SWAMP BUTTERCUP 5 0 Nt P-F orb
Ranunculus recurvatus HOOKED CROWFOOT 5 -3 Nt A-Forb
Ribes americanum WILD BLACK CURRANT 6 -3 Nt Shrub
Ribes cynosbati PRICKLY or WILD GOOSEBERRY 4 5 Nt Shrub
ROSA MULTIFLORA MULTIFLORA ROSE "‘ 3 Ad Shrub

162

 

 

Appendix 1.(cont'd)

 

 

Scientific Name Common Name C W PHYS
Rubus allegheniensis COMMON BLACKBERRY l 2 Nt Shrub
Rubus occidentalis BLACK RASPBERRY l 5 Nt Shrub
Sambucus canadensis ELDERBERRY 3 -2 Nt Shrub
Sanicula gregaria BLACK SNAKEROOT 2 -l Nt P-Forb
Smilacina racemosa FALSE SPIKENARD 5 3 Nt P-Forb
Solidago gigantea LATE GOLDENROD 3 -3 Nt P-Forb
TARA)“ CUM OFFICINALE COMMON DANDELION * 3 Ad P-Forb
T eucrium canadense WOOD SAGE 4 -2 Nt P-Forb
Thalictrum dasycarpum PURPLE MEADOW-RUE 3 -2 Nt P-Forb
T oxicodendron radicans POISON-IVY 2 -l Nt W—Vine
Ulmus americana AMERICAN ELM l -2 Nt Tree
Urtica dioica NETTLE l -l Nt P-Forb
Viola cucullata MARSH VIOLET 5 -5 Nt P-Forb
Vitis riparia RIVERBANK GRAPE 3 -2 Nt W-Vine

 

163

APPENDIX 2

164

 

 

 

 

 

 

N.N N.N N2: N N 22822-0 3. o .. 022502 02-2220 20 30:00 ENS-2.22..
N 22.2: N.N 222 2 8F 22 N- N 22.22 0022 8222223222 3.352
N2 02 2.2N N2 2. 225-3 22 2 N 22020520 2220225 522522205: 32285220822
NN N.NN 2.N2. NN N use 22 N N 02.22 220% 2222283 New
N222 22.2.2
N2 N2. N.N 2 2 225-3 22 N- N 02520 0272020222 eras-2 2.22:
2 N22 N.N 2 2 8P 22 N- N 0222 205% 2222:3822 5%
2 N22 N 2 2 use 22 N N 520220 020.30 0223 82283. .3552
N2 N2 N.N 2 2 225-3 22 2- N 092-28202 2.28.22: 2282.202082on
>2 2.2 1N N 2 002,—- 22 2 m M<O gm 0%0002002 Nae-20$
N.N N.N 2 N2 N 28.2-2 22 o 2 N203. 0.2.2223 0.3%38 .500
2.2. 22.2. N.N N 2. 225-3 22 2 N 220200220 2220225 202225.232: 3228202250
2.2. N.N N2. 222 N 822. 22 N- N 22922 0022 822.30%.22 33¢
N.NN N.NN N.NN 222 2 2 8P .z N N 02.22 220% 23382. 0%
N.NN N.NN N22 N22 2 2282-0 32 c .. 022502 02222220 E 302.522 253.2?
222 22.222
N.N N22 N.N 2 2 822. 272 N N 5220220 02050 0223 2.228% 3:30
5222000800
N.N 2.2 N.N N 2 225% 22 N 22 0.223 8 50202222 2238220 822.222
N.N N.N N.N N N 22.5% 22 N 2 >520NU20<20 202200 22302522522: 2.2252
N.N 2.22 E N2 N 25 22 N- N 08. 0022 825223.22 33¢
2..NN N.NN SN 22. N 22822-0 3. c .. 92502 055.0 20302.50 2303022
N.NN 22.20 N.N.- oN2 2 2 8; 22 N N 02.22 220% 5.22288 28‘
82 22.2.2
#020003 x00
202 x2000 0252 N200 0202 N252 22 0 02.2 202200 0222 02-02-2028

 

 

 

 

 

 

.2000? n >2, 0220 £00835: u 3 .522200200 u m .3555 n m .5005 u .222
53200300 II- 0< .0350 n N Z 20200502 2200223222230 .5022: 223025 0220003050422 .93: 0:52, 00228-200022 032052 0222 0220 A>OUMV
.5>00 032202 0222 .6525 2000000222 0355.2 05 .908 $000000 :0 80¢ 20>00 .20 20% 05 .5va 222 0022-580 83 22 880000 m0

50222022 02.: Amémv 850% 05 .20 0528 0.:— 025 8.20.2 fiBEw 05 65V N880? .20 20020E000 06 ADV 2222222020800 .5 22555000 05
800502 850% 2500 20.2 002208.235 5220225024 28525 :0 E 030% 020 850% 0350.000 .20 8222022 ocueomom .800 225225884222
@2525 00222-203 523 0200 $0382 258 mam-226 025 2500 802.2 $800220 3:083 222 003030 850% 00 82.20252: .N N=22—0027..

165

 

 

166

2 3 N2 2 2 025 .2 N- N 02.22 05% 555.528... 002..
2 2.22 S 2 2 00: .2 N N 5002020 02030 0223 5500. 5.52.2
2 20 N2 2 2 85-3 .2 N- N 0.200 022200030 ...-5.2.: ...:
2.2 N.N N.2 N 2 .225 .2 N 2 522000020220 20200 5505050022.. 5.5.0
500000800

...2 N.N N2 0 2 .235 .2 N 2. 0223 ... 5.2020200 25.2850 0.220
2.N N.N N.N N N .2000 .2 N 2 20.2 0.003 0502.05... .500
N.N N2 N.N N2 N 8.2-N .2 N- N 02 000 855.2800 5.5.0.0
N... N.N N.N NN N .2300 .2 N- N 2.2-200-002-22.205 552.252.... 5503222..
N.N N.N N.N 2 N .2300 .2 N N 00222800220202.2520 5.5052 0.050
N.N2 N.N2 N.N2 0N N2 oe-N .2 N N 02.22 20% 5.5.2.0.... 002..
N.N2 N.NN N.N2 82 N 225.3 .2 2 N 00.20000 2222005 55055.0 502850550
N2... NN N.NN NNN N2 .2800 22 N .. 92-502 02020 2.0 3020.2 2222-2
38 0:22.20
2 222 N2 2 2 .2500 .2 N N 5222 5022 302-205 5.50.05: 555.500

2 2.22 S 2 2 8: .2 N N 5000220 0202-20 0223 552% 5.55.2 .
N.2 N.N N2 N 2 .2800 .2 N 2 N202 0203 08055.0 .500
N2 N.N 3 N 2 2.5% .2 N 2 5520020220 20200 5205520022.. 5.5.22
N.N N2 N.N 2.2 N N.N-2-0 .2 N- N 2.2-20.205-22-205 553.25. 5505050
N.N N2 N.N 2.2 N .022. .2 N- N 222 000 85220-500 5555-0

500000800

N.N N.N N.N NN N 2.5% .2 N 5 0.223 ... 5.020020 5.58.50 0.2.0
N.N N2 N2 2 N2 N .2800 .2 N N 02080022000520.0020 20.052 .0850
E2 N.NN N.N2 22 N 25.3 .2 2 N 00.20000 222005 52005552. 30.80.2550
N.NN N.NN N.NN £2 ..2 8: .2 N N 02.22 20% 55528.... 50..
N.NN N.N2. N.NN 02 N 02 .2000 22 N .. 02.602 0220 .0 302.2 20232
N.N 2.2 N.N 2 2 2.5-3 .2 2- N 52.20200 58.2.2 5.202.852
N.N 2.2 N.N 2 2 2.52m .2 N N 00020 002000-000 58.00.: 5.5.552
N.N 2.2 N.N 2 2 .022. .2 N N 20 02-2223 5.22.. 50050
88 0.02%
N.N 2.2 N.N 2 2 .2500 .2 N 2 202 02-22-23 0502.38 .500
.52 .2000 -. 0002.2 - N200 - .052 .. 022.2 5 0 0222 202200 02<2 0202,2028

 

 

 

6.203 .N 2:50.542

 

 

 

N 2 N.N 2. N .052 22 N .. 0N022 5202025202 222022202 2.2022
N 2 N.N 2. N 28.2.0 .2 2. 2. .5205 302205 580...... 5......
N.N N.N 2..2 N2 2 8.2- 22 N N 2202. 0:223 8885.. 5.58.5
N.N N N.N N N N.N-2-2 .2 N- N 00032020 52.5.... 82....
N.N N N.N N N .2800 .2 N 2. 802200220. 02223 .580... 55.50580
N.N N.N 2.2 222 2 .2300 .2 2. N 520020-0003N 582m 0.252.082 8555.0
N.N 2 2.2. N N 8: .2 2 N 502200020222 555808 .500
N.N N2 2.2. N N 8.2- .2 N N 5220020222 002002220. 5...... .250
N.N N.N 2.2. N N 20.2-0 .2 N N 2022.202 00002-02000 88205.. 8.8822
552000800
N.N N.N 2.2. 22 N 2.5% .2 N 2. 0223 ... 520202220 2.8850 8.2.22
N.N N.N 2.2 NN 2 8.0 .2 N N 32.2320 8...... 825......
N2. N... 2.2. N2 N 28.2-0 .2 N 2. 0008202 55555.... 5.5.08.2
N.N N.N N.N 22 N 28.2-0 .2 2. 2. 5200209005 5022 .5558 82.2556
N.N N.N N.N N N 050.0 .2 N N 8200 080002-500 52.8.2 ...-.022
N.N N.N 2.2. NN N 8; .2 N- 2 220 202022.. 88.05.. 55.20
N.N N.N N.N N2 N N.N-2-0 .2 N 2 N203. 0203 0502.88 5.00
N.NN 2.2. 2.NN NN2 NN 230-0 22 N .. 92502 02220 2.2-20220.2 2202.222
NNNN 2250
00032000
2 N.N N2 2 2 8: .2 N- N 02.22 220.322 55.5528... .82.
2.2 2 E N 2 .2800 .2 N- N 2.202220022522022 55252.2... 5585.2.
. 5220000800
N N.N 2..N N N 2.NN .2 N 2. 0223 ... 520202.220 2.8850 8.2.22
N.N 2 2..N N N .2800 .2 N N 02080022000522.0020 55.8.... 8850
N 22.22 N.N 2. N .2800 .2 N 2 202.. 0.2223 058.58 5.00
N.N 2.2. 2..N NN N 8: 22 N- N 02 00.22 8.5.5.20.500 85.8.2
N.2 2 2..N N.NN N2 N2 8: .2 N N 5020220 020220 0223 8.5.0.. 55.5..
N22 N.N2 N.N2 NN N 85.3 .2 2 N 00.20000 2222005 525085.... 58.88.2550
N.N2 2.NN N.N2 22 2 222 8: .2 N N 02.22 220% 5.55288 .82..
N.N2. N.N2. N.NN N2.N N2 N.N-2-0 22 o .. 92-502 0220 22 22020.2 22222222..
.52, 2.0%. -. 0252 22.0.0..- 020,0 . N.N2-E2 .52. 0. 0222 202200 0222 020252028

 

 

222582 .N 22.8.2.2

167

 

33 :0..—

 

 

 

 
 

N.N ...N ..2 2 2 N.N-.2-.. .2 2- N 5022028 0202.20 5.50.2... 5.8.52
N.N N.N .... 2 2 25.2-.. .2 N- N 3525000 200022 5.2.05.5 5......0
N.N ...N .... 2 2 ......2-0 .2 N N 2022.202 00002-0200... 88.05.. 8.5.0..
N.N ...N .2 2 2 8.5 .2 N N 5.200220 202.20 0223 55...... 55.5..
N.N ...N .... 2 2 N.N-.2-.. .2 N- N 5205 N.N-.22 5.....88 5......
N.N ...N ..2 2 2 85-3 .2 N- N 0.2.20 0222000322 5.5.... ......
N.N ...N .... 2 2 ......2-2. .2 N- N 00032020 ....5.... 8....
.2 N.N ..2 N 2 85-3 .2 N N 00200-20020 5258.20 .855... 5.50
N2 N2 ..2 N 2 8.5 .2 N N .5223 020220 80.5 55.0.5
.2 N.N N.N N N 25.2-.. .2 .. .. 50205 30220..- ..0885. 5......
N2 N.N .... N 2 25.2-.. .2 N .. 500220028 02223 5.550... 55555.5...
5220000800
N.N N2 .... N N ......a .2 N .. 0223 ... 52020222.. .5888 8N...
N.N N2 .... .. N 858... .2 N . 000% .....5... .050
N.N N.N 2... N N .250-.. .2 2 2 2.0205 0020 202200 5.5.58 5......
N.N N2 N.N .. .. 2.5-3 .2 2 N 00.20020 52.0.25 5.505.555 5.5.88.5...
N.N N.N N.N N N ......2-.. .2 .. N 520020-5005 22.50% 5.8.5.... 8.5.5.0
N.N N.N N.N 2N N 25.2 .2 N- N 2280002.; 5.08.. 555......
N.N N.N N.N N N 850-.. .2 N N 8520 080002.500 5.5.... 580.
N.N 2.N. N.N NN .. 8.5 .2 2 N 522220002020 5.55808 5.00
N.N N.N. .... N.. 2 8.5 .2 N- N 22 0022 858.558.. 55......
N... .22 N.N NN N 8.5 .2 N N 3<.23<.2 8...... 555......
N. N2 .... 2.. N 8.5 .2 N- . 220 2202202,. 88.55.. 55.0
.9 N N.N. N2 N2 N.N".-. .2 N . 202.. 02223 88.55... 5.00
...N. N.N N.N. NN N. ......2-0 .2 N _. 0.2502 02220 ...N 302.... 5.22.2
..NNN 22....
N.N N.N .... . . 008m-.. .2 N . 0008 5.55.. .050
N.N N.N .... . _ 85-3 .2 . N 00.20000 22220.25 5.5.0.555... 58.88.55...
N2 N2 ...2 N . ......2-.. .2 N N 09228202220202.2520 5.5.0.... 58.....0
N2 2 N.N N N 59.... .2 . 2 50205 0020 202200 85.8 5......
. .5. .60... r. 02...... . .1200 - 0.2.5 .E... ...... 0 0222 202200 0222 020252028

 

2......82 .N 2.5252

168

 

 

 

 

 

N.N N.N ..N N0 N .....0-.. .2 N N 52006-500550 050020 5.5.5.5. 55555.0
N.N N.N ..N NN N .250-.. .2 N N 02205202000522.2020 855.... 8850
N.N ..N N.N NN N. 25.2-.. .2 N N 52006-500550 5020 55.55... 8......5.0
N22 .... 02. NN. N. 85>-..» .2 2 N 0000000 222200; 555.55.... 5.58558...
N.NN N.NN N.N. NNN NN 25.20 ..2 N .. 0025002 022020 ..N 220.50.. 2.2.2.2....
NNNN 0......0
N.N N.N N2 2 2 .250-.. .2 2- N 500000220 00220 55080 5.5.5.0
N.N N.N N2 . . N50-.. .2 N N 02005202000522.2020 855.... 8850
N.N N.N N2 2 . .....0-.. .2 N N 2055200.. 00002-02000 88.55.. 8.55..
N.N N.N N2 2 . 8.5 .2 N N 500020 00220 0223 8.55. 5.5...
N.N N.N N2 2 . 25...-.. .2 N N 52006-50030 250020 5.5505. 8......5.0
N.N N.N N2 . . 25.2-.. .2 N- N 502025 20022 55.8... 55..
N.N N.N N2 . . .....0-.. .2 N N 5020.5 302205 588...... 5......
... ... N2 N . 8.5.2 N N 50000... 00200200 5...... 5050
N2 N2 N2 N 2 05.2-.. .2 N N 2200-2020200 5.55.... 558505....
N.. ..N N2 N 2 2.5.0 .2 N- N 000000.00 5.88 55.8...
5000000000
02 ... N.N N N ......20 .2 N N 0223 ... 520020.. .5885 ......0
N2 ... N.N N N 250-2 .2 N N 32050000 220222 8.5.... 5550
N N2 N.N N N .0280-.. .2 N . 00000 555.. 550
N.N N.N N2 N . 8.5.2 N- N 2202 000 8.5.5.8... 55.5...
N.N N.N N.N N N 8.5 .2 N N 32032.. 8...... 8.5....
N.N N.N N.N N N .250-.. .2 . . 5020.5 0020 202200 55...... 55..
N.N N.N N2 N. . 8.5.2 N N .202 0522.5 88.55.. 55.5....
N.N N.N N.N N N 8.5-3 .2 N N 00200-20000 5250.00 8585... 55.50
..N N.N N.N N N 8.5.2 N- . 220 220.0022 88.55.. 55.0
..N N.N N.N N N ....0-.. .2 N N 00200 00000020500 55.... 8......
N.N N.N .... N. N 25.2-.. .2 N . 02052 0.2.0.3 .5888 5.50
..N N.N N.N 0. N 05>-..» .2 . N 0000000 222200; 555.55.... 5.585555
N.N. N.N. N.N NN N 8.5.2 . N 5000000222 5.55558 ...50
N.NN N.NN N.NN NN NN .....0-0 ..2 N . 0025002 0.2020 ..N 220.50.. 2.0.222...
.5. .000 00.0. ..00 0.....- 0...... .... , 0 , 0222203560 0222 0.0520200

 

 
 
 

. . .....NN-N: 0.0.5.202

169

 

 

 

 

N.N N.N N.N NN N_ eon-.. .z N N mgNEoENam-Nimuz-N 8285 885

N.N N.N N.N. N2 2 85.? Hz _ N ammmmxu SEEP» £8888 3:88NEN

NN N.NN N.N. go ON eon-m 2 c .. 925:: 0320 E 323m .538?

2.NN Nunav-

I. 3 N.N N _ 83.? Hz a N mama-2E5 552m 83822 82$.
N.N S N.N _ _ seam Hz N N >§mmBSPN @2222 S888 §§§m
to 3 N.N N _ 8.: Hz N N 5.0 NEE» 83 288$
to 3 N.N _ _ eon-NV .z N- c amNmBOE-qizm 2:88: 3888:
N.N S N.N _ _ eon.-.— Nz .- .N NNN-55 303m» 8888 so:
N.N ..o N.N _ _ eon-N Hz - N 538275 Mufim £888 3838-

Emma-N800

N.N N.N N.N N _ DEN Hz N v as, 8 >565: 8888 N88

N.N S N.N N _ 85-3 2 _- N 2-28.2 8888 8§88NN£

N.N 3 N.N N _ 0N8: .2 o _ H53 883 88m

N.N Nd N.N N _ eon-N .z N N «GE-Em: ammo-.29. 88.88 888:

N.N N.N N.N N N 8:- 2 N N NNN-N NEE; 88.55 358$
N.N N.N N.N N _ 9:5 3 N .. mNom 5545-52 ESE-S: was:
N.N N.N 3 N N as: “2 N v q<mmiozoqu 588% 258838;
2 N.N N.N N N 85.? “z N- N mafia Ema-ENE 8.23: 8.x

2 N.N N.N v N eon-N .z N N Scam-595 NEE» 28%.: antenna-N

N.N N 3 N_ N 8; ,z N- _ 24m 252mg? 885$ .355

2 N._ N.N NN N eon-N .z N N "Ewan-5% .255 5538338 255?}:
S N.N _.N : N ea”: 2 N _ SSS mam 20228 8.288 as:
_.N _ ..N 2 v eon: “z N N 1% 20.20qu >238 888.8 saEcNSom

N.N N._ _.N 8 N ecu-N 2 o N 85.52 28558.: §8§£

N.N N.N 3 8 N 8: Hz N N N$E<NN<N 283% 8x85
N.N N.N _.N NN N 8:- ,z _ N bammmoé £3888 889
N.N N.N _.N NN v eon-N ,2 N- N pagan-$3-53 as}? 8388‘

N.N N.N N.N N N 3.5... .z N N NNN-dc mmammmd-Nom seam 359$
N.N N._ N.N 2 N ecu-4N .z N c BEE-mm 1:22.6- 8E8 23:8
3 N... N.N S N eon: 2 N N mam? N22 5888 §=§88m

5 N8: 3% N8 9E Em k o 832 20228 322 uENzBuN

 

 

6.83 .N 5.52:?

170

 

 

OomZmQ-HOG

 

  
 

N.N N.N N.N N N eon-AN 2 N N $228355 888 9885-
NN N.N N.N N N eon-N 2 N N ZEN-5F 20228 888883 8.5.5
N.N N.N N.N N N NNN-N-N 2 N N .5305 £225 N.N8888 NNN:
N.N N.N N.N N N NNN-N-N 2 N- N NNN-NOS NNN-22 8888 £9:
N.NN N.N N.N N N 83-? 2 N- N ”N.NN-mo xz-«NNNNNSNN 8&2 8.:
N.NN N.N N.N N N 8: 2 N N 2865 Ema-«NNN N88 N85
N.N N.N N.N N N N.N-N-N 2 N N Qm-NNNNNNNNB 256 8858888 883888:
N.N N.N N.N N N ecu-N 2 N N Scam-NOON 533-50 8888 888-5
N.N N.NN N.N N N £82 2 N- N ommmao-N-NNN-NNSN 25.88 388%
N.NN N.N N.N N N 85-3 2 N N Nmam-mexo 5.52m 83888 82%
bammmmoow

N.N N.N N.N N N NNN-NNN 2 N N 35 a EVEN-NNN .8888 8.8
N N.N N.N NNN N ...st 2 N- N NNNDNNNNUNNN 888 N88:
N.N N.N N.N N N ecu-N .2 o N ONE-NEE 88858.5 88838;
N.N N.NN N.N N N ecu-N 2 N N 225:8 NNN-2.N 8888 8.38m
N.N N.N N.N NN N ecu-N 2 N N imN-zozo-NON 88% 888885
N.N N.N N.N N N N.N-N-N- 2 N o BEEN-NNN 1:272 8.28 888
N.N N.N N.N NN N eon-N 2 N N >NmoNu-Nmm3N 5005 88.83 88888
N.N N.N N.N NN N 8: 2 N N NNN-N NNN-NNN? 88.88N 888m
N.N N.N N.N NN N 8; 2 N N 353$ 83.8 8.8.8
N.N N.N N.N N N ecu-N 2 N N Suva-Ema NENSN .8888 888:6
N.N N.N N.N NNN N 8:. 2 N N bN-NNmmvNofiN 88888 889
N N.N N.N N N eon-N 2 N- N Sou-52% MEN-NNN N.N-s88 3888
N.N N.N N.N NN N 8.; 2 N- N 2.5 252% 88.58 385
N.N N N.N NN N N.NN-N 2 N N NNN-NOS Bog-NNN.» 88888 3...:
N.N N.N N.N NN N ea”: 2 N- N NNN-NNNN-BNHZNJNNNS 8888.8 888%
N.N N N.N 2 N eon-N 2 N N NNN-NOS ma-NNN 20228 8.88 so:
N.N N.N N.N NN N N.NN-N-N 2 N N 3.5. >52 8888 §=888m
N N.N N.N NN N eon-N 2 N N NzNN>< BE? 8:888 88¢
N.N N.N N.N NN N N gem-N Nz N N NNN-No :NsmmmNN-Nom 583 8.58:
NE NNN-um: -, NNN-é 88. , 9E. ENE - 8 - w $22 20228 $22 oNNNszNuN

 

6.83 .N 553%

171

 

 

 

 

N.N N.N N.N NN N NENNN 2 N- N 28828 888 8883
N.N N.N N.N N N N.NN-NNN 3. N .. BEEN-N28 85:: 293588 2053
N.N N.N N.N N N eon-N 2 N- N N8~vaN<zN 205m 8888 3888
N.N N.N N.N N N NNBN-N 2 N N 80% 883 8N0
N.N N.N N.N NN N 8.; .2 N N 3<N3<N 888 88.8
N.N N.N N.N NN N N.NN-N .2 N N NNN-m-NNNNNNNNB .280 8838888 8:888:
N.N N.N N N N 2.N-N .2 N N >880 850 82, 8:88 88m
N.N N.N N N N N.N-N-N. 2 N- N 088280 888 8N:
N.N N.N N.N N N N.NN-N .2 N N NN03=N8N 838-50 8883 8880
N.N N.N N.N N N N.N-N-N 2 N N SmuNQNNNN-NBN NNN-NNN 8898 88880
N.N N.N N NN N 8.: 2 N N >§m§0<m 88888 880
N.N N N.N NN N N.NN-N .2 N- N NNN-588-2385 8888 888N-
N.N N.N N.N NN N ecu-N .2 N N N80; 3088 88,88 8.:
N.N N.N N.N NN N Nam-N 2 N N 92288 NNN-Ni 8888 888%
N.N N.N N.N NN N 8: 2 N- N 28 28282 88.88 885

N N.N N.N NN N N.NN-N 2 N N NNN-N05 88 20228 888 NB:
N N.N N.N 2 N eon-4N 2 N N 35:88 N<Dzz< 8.28 8.80
N.N N.N N.N NN N NNN-N-N 2 N N 8.72 :2 N888 8:885
N.N N.N N 2 N 85¢ 2 N N 856 NNNBNmm-N-N-Nom 388 N.N-8N:
N.N N.N N.N NN N N.NN-N .2 N N m8N502-N.NmE§02m 8888 8880
N.N N.N N.N NN N N N.NN-N Nz N N 586988 E82N 828.83 88820
N.N N.N N NN N Nam-N 2 N N sz>< BE? 8888 s80
N.NN N.N N N.NN NNN NN 85-3 .2 N N 8880 57:85 888858 8888er
N.NN N.NN N.NN NNN NN N.NN-m 3.. N . 92502 0820 E 305: ENE-NNN
NNNN N.NNEN

N.N N.N N.N N N Nam-N 2 N N Noni-”NNN 088-0208 88.88 8:88:
N.N N.N N.N N N N.NN-N 2 N N «N.NN-N 838-58: 8888 8%
N.N N.N N.N N N 0N8: Z N N 80% 882 8N0
N.N N.N N.N N N NNN-N-N 2 N N 88-52% NNN-NNN? 888: 8.583
N.N N.N N.N N N Nam-N- ? N ._ m-FNNNz-sza 882 >580sz 2823
N.N N.N N.N N N 8: Nz N N 5280 203m 85, 8.88 85E

5 N00: 8% N00 92 Si 8 0 m2<z 20228 m2<z BEN-288

 

6.88 .N £88.?

172

 

 

 

 

N N.N N.N NN NN NNN-N-N 2 N- N N00NNNUN<2N 205m NNNNNEN NNENNNN
N.N N.N N.N NN N NENNN .2 N N >§mmNN$N 2050 N.NNNNNNNNNNN 25%
N.N N.NN N.N NN NN N.NN-N .2 N N 203N202 NNNNNNNNS 5.32980

N.NN N.NN NN NNN NN N.NN-N 2 N N 586-25% E82N NNNNNNNNNN 33350
NNNN NNN-N
N.N N.N N.N N N 25 .2 N N N20 205m 83% 30:0
N.N N.N N.N N N eon-N .2 N N 83.22 53:558.... 238N232
N.N N.N N.N N N 85-3 2 N- N >>N20NNoN 8822 NENNNNNSNNNN
N.N N.N N.N N N NNN-N 2 N- N 24m 2<02m2< 3.85% 2.25
N.N N.N N.NN N N 85-3 .2 N N «mam-8 $2525 N.NNNNNNNNNN NNNNNSNNNEN
N.NN N.NN N.NN NN N NBNNN 2 N N bammxofim 20228 NNNNNNNNNNNNNN .....Né
N.NN N.NN NN NN NN N.NN-N .2 N N 588985 E82N NNNNNNNNNN $3050
N.NN NN N.NN NN NN NNN-N-N 2 N N >NNN0N0-NNNNN3N 5N2: ......NNNNNN 33829-0
N.NN N.NN N.NN NN N NNN-N-NN 3. N ._ 02582 035.0 E 3028 NEE-N.NN.-
NNNN NNN-N
«N.N-N8 502
N.N N.N N.N N N 8802 2 N- N NNN-20 <22<2 N38 N.NN;- N288
N.N N.N N.N N N N.NN-N 2 N- N NNN-N09 2922 33398 NNN:
N.N N.N N.N N N N.NN-N .2 N N 205-5: 20228 sENSNNN& SSE
00228-80
N.N N.N N.N N N N.NN-N 2 N N 0828308 NNN-NNN NNNNNNNN-
N.N N.N N.N N N NNSNN 3. N . 38 5555-52 ENS-NE .52 NNN-02
N.N N.N N.N N N N.NN-N .2 N N ..N-2.2.22 088-0208 NNNNEEN 823$
N.N N.N N.N N N 8; 2 N N 50205 828.32 598 N08
N.N N.N N.N N N N.NN-N. .2 N- N 0m-NNNN30-NhNkN-NNN2N 2.....NENNN 38:23
N.N N.N N.N N N N.NN-N 2 N N N8NNNNNN<2N N523 N.NN-NNN; 2:538
N.N N.N N.N N N 2.3-3 2 N N ”mam-288 5.508 8235. 3N;-
N.N N.N N.N N N N.NN-N .2 N N 2.NN-20288 ...ENNNN sage-NNN;
N.N N.N N.N NN N 8; 2 N N :3 NNN-E3 9.8.55 3:222
NENNNNNNNNN000
N.N N.N N.N N N 952 2 N N 83 No 52002 €226 NNNNNN
NE N002 02% N8 92 N32 2. 0 NN2<2 202200 m2<Z 0NNNN2N8N

 

 

N283 .N 525...?

173

 

 

N.N N.N N.N N N N.N-N-.. .2 N N 5808-828 5222 .280... 8.82.0
N.N N.N N.N N N N.N2-.. .2 N N 35.808 5222 2.8.... 2......0
N.N N.N N.N N N 8: .2 N N 20.22 02.. 2.5.... .8..-
N.N N.N NN N N 2.N-..-... .2 N N 35508 2022... 82...... 2......0
N.N N N.N N N 880-.. .2 N N 288... 028002 22.2.2.8... 828..
N.N N.N N.N N N 880-2 .2 N N 25.2-00.3 528 88...... 822....
N.N ..N N.N N N 8.2 .2 N N 282028 2088 8.8. 8.88822
N.N N.N NN N N 2.5-3 .2 N N 2220 8228 .8888 8.:
8880800

N.N N.N NN N N 855 .2 N N 0.23 ... >802... 288...... .8...
N.N N.N N.N N N 2.5.3 .2 N N 8308-20820 888.8. 8.22

N N.N N.N N N N.N2-.. .2 N N 8005 202220 2882.8 N.N...

N N.N N.N N N 2.5.3 .2 N- N 332080.. 88...... ......2...............N
NN NN N.N NN N N.N2-.. .2 N- N 02.82 8...... 8......
N.N N.N N.N N N 8.2.0 .2 N . 0.2802 0.820 ..N ...-.05... .88....
NN N.N NN NN N ...-N .2 N- N 28 220.822 88.52.. 825
N.N N.N ..N N N 8.2 .2 N N 00.22 200m 2.22.88 8..
NN NN NN N. N 858 .2 N .. N502 5.00-.5002 22082-52 ..N-0..
N.N N.N N.N 2 N 8.0 .2 N N 52820 2028 00.3 2.88 8.3....

002220000

N.N N.N N.N N. N N.N2-.. .2 N N 022228-208 ....88 88......-
NN N N.N NN N 255 .2 N N >8280<8 202200 ...-8288.... 8......
N.N N.N N.N N N N.N-+2 .2 N- N 083.2200 2.2.... 8....
N.N NN N.N N . N N.N-N-.. .2 N N 325N000 02.88 28288 2......0
N.N N.N N.N NN N N.N2-.. .2 N N 0.088.202- 0022 2822...... 228280

N N.N N.N NN N 855 .2 N- N 083000 .220 8.2.8880

N N.N N.N NN N 8.: .2 N N 20 2.223 ..N... 8880
N.N N.N N.N N. N 2.53 .2 N N 8880 52.025 .....8............. 88.88.82
N.N N.N N.N NN N N.N2-.. .2 N N N202 0.223 .8888 280
N.N ..N ..N NN N N.N2-2 .2 N N 5228002 28.8... 828222..
N.N N.N N.N NN N 252 .2 N- N 2880.8 2288 882-.
.5. 28.. 022.. 200 9.2 .22.. ... 0 2222 20228 2222 0.22-2.08

 

0.8.. .N 232......

174

 

 

 

NN N.N N.N N N 2.5-3 .2 N- N 2.2-20.202 88...... 8.288.on
N N N.N .N N .....N .2 N N 020 2223 ..N... 88.80
N.N N N.N N. N 8: .2 N- N 202 220.202... 2.822.. 82.0
N.N N.N N.N N. N N.N2-.0 .2 N N 0002222202 00222 2282...... 22.8280
N.N N.N N.N N. N 8.2 .2 N N 322020 20.200 00.3 2......8 8......
222202800
N.N N.N N.N NN N 2.2.2 .2 N N 00.3 ... 202002.. ......828 8....
..N N.N N.N NN N 082.... .2 N N 5222-002 28...... 8.2.8.222...
N.N N.N N.N NN N 282-... .2 N- N 002322000 ....2.... 8......
N.N N.N N.N NN N .252 .2 N- N 0003000 .220 8.282.220
N.N N.N N.N N. N. .082... .2 N N 202.. 00.023 ...-.....88 280
N.N N.N N.N NN N ....2m .2 N . 22222222 20200 2.22.28. 8......
N N.N N NN . N N.N2-.. .2 N- N 20022025 20200 .288... 282......
N.N N.N N.N NN N ......NN .2 N N 220220200 202200 288.828.... 8......
N.N N.N N.N NN NN 2.2.0 .2 N . 022502 0.020 ..N 3020.. 2230-2.
N.N N.N N.N NN NN 2.5-3 .2 N N 2220020 $2.025 222.22.... 88.8.2.2...
N.N N.N N NN N N 252 .2 N- N 28000.2 282... ......20
N.N N.N N.N NN NN N.N2-.. .2 N N 2030202 282...... 8.28.220
..N N N. NN N.NN NN NN N.N2-.. .2 N N 2000.0-2003N 2Noo2m 2.2.2.... 8.2.82.0
NNNN ......
N.N N.N N.N N N 8F .2 N- N 2.2 002 8.28.8.2... 82......
N.N ..N N.N . . .....2... .2 N- N 22002-052020... 2.2.2.2.. 82.8....
N.N N.N N.N N N 282-2 .2 N- N 22.5... 00200 222.522 8....822 ......
N.N N.N N.N N N N.N2-.. .z N N 002N220. 2.22.2.2... 22.22....
N.N N.N N.N N N 820-.. .2 N- N NN<200202 0003 88.222... 22.0
N.N N.N N.N N . N.N2-2 .2 N- .. 2000 20220 202022220 20202
N.N N.N N.N N N N.N2-.. .2 N N .2005 300002 888...... ..N...
N.N N.N N.N N N .822 .2 N N 0000N 82.8.2.8. 82.0
N.N N.N N.N N N .....N .2 N N 22.23% 22...... 8.2.82
N.N N.N N.N N N 28.2 .2 N N 22000 002202.002 8.28... 8.2.2.2
N.N N.N N.N N N N.N2-.. .2 N N 000N200 ......8.8...... .22....
2.2 2002 02.22 ..00 L. 02... 2.0.... ... 0 0222 202200 0222 0N22-20N0N

 

6.280 .N 5.8.0.2

175

 

 

 

N2 2: 2: N: N_ eon-.. .z e N 586-3%; 5005 333s 33856
N.N. Ni N.N. NNN M: 85-3 “2 _ N ”Emma 575%, sawing 338=§§m
5; NS N.N. NNN 2 ea: 2 v v 5853mm; 53.: E239 33856
NSN usam

Nd 3 N.N _ _ ”2.N .z N- N mm< mum Bauigmmgsét
N.N 3 Nd _ _ magi “z N N mam: ozanz susgrmsa Seam
N.N S N.N N _ 8; 3 N . 595.25 20228 5:me 5 Nuzzwmz
Nd S Nd N N .35 Hz N N «mam agmmdma «8582 332%
N.N 3 N5 _ _ ea: Hz v N 5305 303m.» $833 so:
Nd S Nd N _ ONBNN Hz N N mwomm Eofisufis p98
Nd S N.N N _ 36d .2 N- N Nméoammm 803 3833.3. 53.6
.3 N.N N.N N N 833 .z N N mm30di©2¢<o 2:823 555
2. N5 N5 N _ eon: Hz N N 83.84 383% 33$
.3 N.N Nd N _ as: Hz N N 2% 20238 >238 $833 seggwbé
Nd 2U N.N N _ eon: Hz e 2 333-9233 Nagse 9.33%

992348
N.N 8 N5 N N 93: Hz N N 5223328 588 6335
8 N.N I N N eon: .z N N mgmemoENagosz guess 385
8 N5 2 N N eon: .z N- N NE? 875 533m 33883 saw
3 Nd 2 N N 380.; .z N N 82.33 SEN 33? 3.3m
3 N5 N.N N _ as: “2 N N 35595 555 23.83 23:8
dommmfi

N.N Z Nd N _ e8-.. Hz N N axon am><5dmmfimao 2383» §§§~Q
N.N to ... N N 255 3 N ._ 39. 5.33352 E3553: EON
_ E N.N 2 _ as: “2 7 N mEmz 83% 8:5
N._ N._ 2 N N 85-3 Hz N N mac «gm £3358 a:
N._ E 2 e N eon: 3 N- . v.08 NEE NENQENENQ 523
N._ S E N N 8; Hz o _ 3.22 92 .SEE 3%
E N.N N.N N v 88¢ .z N v 35:83 5:202 seeks $.56
S 2 N.N N v 8; .z N N 5.22 «<08 3.2382. 5%
S N.N 2 t N 023 Hz N v mmameém 50035 :35 §N§N§V
5 N82 9% N8 92 93$ 3 o 32 20228 m2<z 233298

 

@253 .N 55...?

176

 

 

 

FIJI-[141“...
N... ..N N.N . . N.N-N-N. .2 N- N 2535.30 588 8....
N.N ..N N.N N . eon-.. .2 N N 352.8% >22: 8885. 88.5
N.N ..N N.N N . £8... .2 N N -.<N.N 2020.8 >238 888.8. 888888.
GONE-”.38
N.N N.N N.N N . N.NN-.. .2 N N N.N.Ezm-NNNBN. N.88 888.8
N.N N.N N... N . 85-3 .2 .- N >>.-zoN.o.. 88.88 88888.5
N.N N.N N.N N . 80”.-.. .2 .- . ”NE-".2 88.8 8.8:
N.N N.N N.N N. . £88 .2 N N 352.8% 252sz- 8.:u8 88.5
N... N... N.N N. . EN”. .2 N- N 2am... NSENZNN 8.8.88 8.8.6
N.N N.N N.N N. . 88> .2 N N Esme-..N .5005 885 8.8885
N.N N.N N.. N N NE-.. .2 N N dongs-Nu; 9522 888.88 8888qu
N... N.N N.N N. . N.NN-N 3. N- .. Noon. NN.-Em NENENNENS ESE.
N.N N.N N.. N N N.NN-.. .2 N . N2m>< 5.23 88888 889
. N.N N. .. N NNN-[.-N .2 N N mgNNmoE-NNEUZN. 8.888. 885
... N.. N.N NN . NENN 3. N . >§mmx<m N.N-”.255 893%me N.N.mmmmm
N.. ... N. N. N 8:- .2 N N 3.22 ..N-0P0. 88888 88-
N.. N.N N N N N.NN-.. .2 .- N 59.52% 231:. 2888 5888.
N.. N.N N N. N eon-N. .2 N N 528.800: 38.88. N8888msw
N.. N.. N.. NN N £55 .2 N .. N.NS. SSE; 38-553: .82.
N.. N.N N.N N N 880... .2 N N NEE-=3 >v.-..N ...-88...; Nae-N.N.
N N. N.N NN N 88> .2 N N 55 N523 38 888$
N.N N.. N.N ..N N NEN .2 N- . 900308 >56 888N888
N.N N.N N.N ..N N 8; .2 N- . 2.2. 2<oEm2< 88:8... .385
N.N N.N N NN N 8: .2 N N ham-mo No.82. 3.3 8.888 85$.
N.N ..N N.N NN N N.N-N-N .2 N N 2.83ng 88888 8:88.85
>N-Nmmmmooc

N.N N.N ..N NN N 85.. .2 N N 5.3 8 >562... 8888 8N...
N N.N ..N NN N 85 .2 N- N .5385 8888 N88:
N.N N.N ..N NN. N NEN .2 N . 5.8.5.33 20228 888888.... 88..
N.N ..N N.N NN N. N.N-N-.. .2 N N 35.592. Eng 88825 88.5
N.N N.N. N.N 8N N eon-N. .2 N .. 92522 0.520 x. ENE-NE ENS-N..-
5 38. 95. N8 92 9.5 8 Pal m2<2 20228 N922 0.”...N2m6N

 

.988. .N 8.82.3.

177

 

 

 

 

N.N N.N N.N N N .8. .2 N- N 2.N... N.>.....szN 88.88 8.8.0
N.N N.N N.N N N eon... .2 N- N 52--”.2288 0.... 8.8.. 88.88.
N.N N.N N.N N. N N.N-.-.. ..N. N- . .000 NN.-EN. NBS-.3200 x08...

. N.N N. N N N.NN-.. .2 N N cum-N02... 888.8... 888......
... N.N N. N N .30... .2 N N .508... 02.0002 88.8888. 888-.
N. N. N.N NN N NEN B. N .. N.Nom 5.0.5.0.). 8.0.2.3.... .00..
N. N.N N. N N 8... .2 N . 3.3,. 0.... 888. 8..
N. N. N.N NN N NN.-N .2 N- . 2... 250.02.. 8888 8s...
N. N.N N.N N N N8”.-. .2 N N NN.-.05 30...... 888...... N.N...
N. N.N N.N N. N N.N-.-.. .2 N N 525300.. 8888.. 8808.88...
N. N. N. N. N 2.5.3 .2 .- N 22080.. 88.88 8.88888
N. N. N. NN N 8: .2 N N N.N-0 .353 88 8880

35.0.88
N.N N.N N. NN N NENN .2 N N ...-:3 a 5.0.5... .8888 8N...
N.N N. N.N NN N N.NN-.. .2 N N dong-0.0: 0.022 28.8.8: 8.8880
00.23.00
N.N N.N N. NN N ..N"... .2 N N 08.23800... 888 88.3..
N.N N.N N.N NN N Na... .2 N . 35.05.05. 202.200 888888.... 88..
N.N ..N N.N NN N NN.-N .2 N N 35.8.0 .03.. 3.3 8.88 852..
N.N N.N N.N NN N 85 .2 N- N 5.2.005 8.88.. 888..
..N N.N N.N NN N 855 .2 N- . 0003000 35.0 8.88888
N.N N. N NN N. N.NN-.. .2 N N 00<2NN=0.2-N...E.2<:02N 8888. 88.0
N.N N.N N. N. . N N.N..N .2 N . >§mmNN§ 5.050 88888.. 88..

N N.N N.N NN N. N.NN-.. .2 N N 3550.2. 55.05.". 88.8. 8.80
N.N N.N N.N NN N. N.N.-. .2 .- N 80.82% .030 c.2888 N.888
..N N.N N.N NN N. N.NN-.. .2 N N 0.03020: ....8888 888.000
N.N N.N N.N NN N. N.NN-.. .2 N N 5.0.0-505 5005 808.8. 8.88.0

N N.N N.N NN. N. 2.5.3 .2 . N ...-.....056 52.0.9 8888.8 88.8885.
N. N... N.N NN. N. N.N”... .2 N N 5.0.0-505 5...... 898.... 8.88.0

..N. ..N ..N N. N .. N.N-.-N. .... N . 925...). 05.20 .... 30...... 5.3-...-
NNNN NN.-EN
. ...... N08. 08... .. . «Sm , 0...... NN.... .\ ..N 0 0.5.2 202.200 .822 0...N20.0N

 

5.008 .N 50:00.3.

178

 

 

 

 

M: N: N._ NN v eon-N Hz N N dong-20F Razz seing 2323.3
N._ N._ _.N a N eomé Hz a N SEE-mo: 33283 ageing
N.N N._ N.N _N N as”: ,2 c _ m2m>< NEE» mgnga .ssc
N.N N.N N NN N seam “2 N- N 9003009 >55 338N§Eou
N.N S N.N. ..N 2 8: “2 N N 5555 mofim N35 38$ :35
N.N N._ N.N a N_ flan-N “z N N waggimagozm 33225 835
bammmmooo

N.N N.N N 8 N 255 .z N v a; a 52053 c3859 NEE
N.N N.N N.N NV 2 £5 Hz N- N mmbmmoam scams Buns-N
N.N. v N.N a. N. ecu-m Hz _- N 53825 xofim 3.5?» 5835-
N.N N.N N.N NV : ecu-N .z N .- Bgmomm Nzéofim sis? 53:6
N.N N.N N.N 8 N spam “2 N _ bammaofim 20228 NENENNMNNB 33
N.N N N NN E eon-m “z e N Emu—09mm; 500% £535 fifosao
_.N _.N _.N 9: e eon-.. Hz o N 502620: a?%28 3:33»me
N.N N.N _.N N: a eon-m Hz v v SmBQNmN—BN >fi<m €230 8332.6
N N.N _.N _N_ a 85-3 .2 _ N «Emma <20? £Q§§é gagefité
N2 2 N.N N2 : ecu-m 3 o _. 925:: 0356 E 305E 533w
NSN «...-am

Nd Nd Nd N _ eon-m .z o _ N2m>< NEE» manage 2&6
N5 3 N6 _ _. eon-m .z N N BEEN—m b5: 583 2536
N5 3 Nd N _ N85-.. Hz N- N mméoammx nook §§B§ 33.6
N6 2. N6 N _ 8; Hz _ N bammmném NNN-Buses €30
N... N5 N6 N _ eon-N Hz N N NE-«zNgmm Nada 38:88 ESSEN-
N.o 3 N6 N _ eon-N Hz N- N ME? 82: 582m 33885 $3.-
Nd 3 Nd _ _ 8: .z N N 5.22 558 53388. L8“.-
2. N.N Nd v _ 95m 3 v _. ham-92m N.NN-N255 NNuxmmzbE mama-“NNN
to Nd Nd v _ eon-m .z N N xo-EN 923003 manta-cu 83
N5 N6 Nd N _ 83-3 .2 N N «god-zogo £523 32%
Nd Nd Nd N _ 8; 2 N v Esme-Em 582m ass gmmfiasusw
N.N N5 3 N N 2.3.3 ,z N N mac $223 $3.83 3.:
N.N Nd N... v N as”: Hz N- N ammBMZm-B SEE 83
NE Nook . , 3%., - N8 92 NNE N. u Ez- 20228 $22. BEEN-5m.

 
  

-€.aos..N may?

179

 

 

 

 

N.NN N.NN N.NN NNN NN eon: 2 N- N NoommVN-«zm Mufim NE?» 53.5.
NNNNN NNN-.—

552 NN.—>—

N.NN N.N N.NN N N omNNoN-N E N N momma £30328 Hg0
N.N N.NN N.N N N N8: z N N ngNEN NNN-NE 55:83 ENNNNEN
N.NN N.N N.N N N eon-N 2 N- N NE? 323 zmeN<m 33883 s:-
N.NN N.N N.N N N eon-N. 3. N _. 895650 202200 2002 532.2
N.NN N.NN N.NN N N 8; E o N 5.22 mm: :30: 0%-
N.N N.NN N.NN N N Nam-AN 2 N- N ENE-E3065 as}? Emafiw
N.NN N.N N.NN N N 380... .2 N N mam-"N.N 025002 353533 85$
N.N 2N N.NN N N ecu-N 2 N N 2% 20238 >238 389.3% sagombom
N.N N.NN N.N N. N Nam-N 2 N- _. M000 NNN-SE 23002030 SSS:
N.NN N3 N.NN N N £52 2 N N «ma-NNN magma-0mm 2382 3328
NS .3 N.NN N N eon-N 2 N N BEEN-NNN N22: saga 2530
N.NN N.NN N.NN N N 83-3 .2 N N ago-2-20320 £32223 35%
N.NN N.N N.NN N N 8; 2 N N 3.22 «<08 53383 0%
N.NN N.N N.NN N N 3802 .2 N- N NN.-2000mm 0003 SNENES 350
N.NN N.NN N.N N N ecu-N- 2 N- N 5202-503 ma; 2.5% $235
N.N N.NN N.N N N 33¢ 2 N N 82-0-03 NON-NNN 38:: 3:30
N.N N.NN N.N N N 83-? 2 N N NNN-3N0 $228 332:8 N.NN:
N.NN N.NN N.NN NNN N ea 2 N- N 78E m>EmzNNN €032 «08:0
N.NN N.N N.N N. N N.NN-N 2 N. N NNN-N05 302m; 383R 3...:

00505400

N.NN N.NN N.NN N N N N8”: 2 N N 0822BN-m0-Nm 588 $25
N.N N.NN N.NN N N N 85-3 .2 N- N £285.. 38%.: NNEENBNNBN
N.N N.NN N.N N N. eon-N .2 N N. 000204 398830 32$
N.N N.N N.NN NN N 8: 2 N N. $00032 582m 005 NNNNNN§N§V
N.N N.N N.N NN v 8: 2 N- N 22 z<02m2< 285% .355
N.N N.N N.N NN N 8; 2 N N 020 BE? 2% 33:0
N.N N.N N.N NN N 255 NE N . mmom SEN-552 ESE-S2 26x
N.N N.N N.N NNN v £9: 2 N v mum-4.022 53358: §§§£
NE N000 052 N00 0% ER 2 0 m2<z 202200 m2<z 020-288

 

383 .N 55...?

180

 

 

 

N.N N.NN N.N N N 808-0 2 N N 0000N 33003 880
N.N N.N N.N N N 85-3 .2 N- N 53-20000 880.8 80888.80
N.N N.N N.N N N e000 2 N N 00.2005 030 N.N-8282 88050
500000800

N.N N.N N.N N N NaNNN 2 N N. 0036 500200 08850 800
N N.NN N.N N N 8; 2 N N 5000000400 £580.08 N.NN-.0

N.N N.N N.N N N 030-0 .2 N N 0000 003 $88.8 53.50
N.N N.N N.N N N N.NN-N-NN 2 N- N 000300.000 5.88. 800
N.N N.N N.N N N 030-0 2 N- N N085 000<2 3.583 N.NN:
N.N N.N N.N N N 85-3 2 N N 0000000 £2605 80883. 3082850
N N.N N.N N N 88.0-0 2 N- N 0000N 80:0 880

N.N N.N N.N. NN N. .055 3. N .. 0N00 00005-52 000000002 N000
N.N. N N.N NN N 030-0 2 N- N 5005 2.000 30:: so:
N.N. N.N N.N NNN N 050-0 2 N N .203. 0:03 N.N-N888 .500
N N.N N.N. NN N. £00 2 N- N. 00502 0003 NNENEN SE03

N.N N.N N.N NN N 80- .2 N N 500000 0030 003 8.088 8.5.0
N.NN N.N N.NN NN NN 080-0 3. NN .. 00532 05000 $30000 0003-N.NN.-
NNN 03. E: N: N.N 080-0 .2 N- N N00000<2N 00000 03080 03.80
2.8 NN.-0

N.N N.NN N N N N800 2 N N 052005 03.00 N.N-8582 880:0
N.N N.NN N N N 85 2 N N 09.. 0:03 88.28.. 958:0
N.N N N N N 85-3 2 N- N 02-20000 88.0.2 80888..on

500000.800

N.N N.NN N.N. N N NaNNN 2 N N. 0033 500000 088:0 80.0
N.N N.N N.N. N N NNN-.00 2 N- N 0000N 8.80 .880
N.N N.N N.N N N 080-0 .2 N- N B005 502 833.8 so:
N.N N N.NNN N N 080-0 3 N .. 005002 030-00 N0 30:00 20330
N.N N.N N.N NN N 080-0 2 N- N 5005 2<000 805. so:
N.N N. 02 NN N 8:. 2 N N 500000 0050 003 8.828 35:0
N.NN NN N.N NN N. 080-0 2 N- N. 00.202 0003 088%... SE03
N.N.N 2 02 NN N NENNN 3.. N .. 080 «0000552 0000-00-52 0000
E0 0000 0000 N00 000 0000 .0 0 0202 202200 02<2 0N0NN-20N0N

 

0.80 .N 8.52:.

181

 

 

 

 

N.N N.N N.N N N 80- .2 N N 00.22 008 88888,. 8N.
N.N N.N N.N N N 080-0 2 N N 20200... 0.000 88.830 88.800
N.N N.N N.N N. N 85 2 N N 05. 0003 88.8.8 8880
N.N N.N N.N N N 8.0.0 2 N- N 50005 000.2 8.088 .8:
N.N N.N N.N NN N 85-3 2 N- N 53.20000 8888 88888.80
N.N N.N N.N N N 080-0 .2 N. N 50005 300005 88080 .8:
N.N N.N N.N N N .800 2 N- N 50005 20000 N88 .8:
N.N N.N N.N NN N NNN-N-N 2 N N 0<0N 202008 52300 88.88 8888080
N.N N.N N.N N N 808-0 .2 N- N 0000N 880 8N0
N.N N.N N.N NN N 025 2 N N 500000 0050 003 888 880
N.N N.N N.N NN N 85-3 2 N N 0000000 570005 888880. 88.8850
N.N N.N N.N NN N 080-0 2 N N 0000... 5.2 88880 808880
N.N N.N N.N NN N N 82.0 .2 N N 000.2005 030.0 88.88 8.8080
N.N N.N N.N NN 2 8.0.0 2 N N 00<0w5002-m.0052<0020 888.0 8.8.0
N.N N.N N.N NN N 080-0 2 N- N 00.502 0003 08828 8:83

N.NN NN N.N NNN N N 8:0 3. N ._ 080 «00005002 000000002 N000
N.N N N.N N N.N N NNN NN 080-0 2 N- N NNN-50-005-70-005 80808 3883.
N.NN N.NN NN NNN NN 050-0 3.. N . 00502 0300 N.N 30000 803000
N.NN N. N N N.N N NNN NN N800 .2 N- N 500000<2N 00<00 N.N-8080 0888

NNNN N.NN-NNN
N.N N.N N N N NENNN .2 N N 00000 0000000000 8888 8828
N.N N.N N N N 080-0 2 N. N 50005 300005 8808.8 NNN:
N.N N.N N N N 8.5 2 N- N 200 2<0N002< 88.:88 885
N.N N.N N N N 808-0 2 N N 0000N 883 8N0
N.N N.N N N N 08.0.0 2 N N 000.? 5.02 8.880 808880
N.N N.N N N N 85-3 2 N N 0000020000 50500 88888 .880
N.N N.N N N N NNN-N-N 2 N N 5000N0-5003N 05002N 80080 83850
N.N N.N N N N 080-0 2 N N 0.0% 202008 52300 88880 8888080
N.N N.N N.N N N 080-0 2 N N 50030200 08888 888.080
N.N N.N N.N N N 925 .2 N N 00.22 0.008 88888 8N.
N.N N.N N.N N N 025 2 N N 05.. 0503 88.8.8 8:880
E0 0000 0000 000 00.0 0000 .0 0 02<2 2W2200 02-2 005208

 

6.083 .N 0.88%.

182

1....

 

 

III-q

183

 

 

 

N N.N N.N N N NNN-0-0 .2 N- N 00000502000 8880808800800
5000000000
N N.N N.N N N N800 2 N N. 0003 8 5000000 08008.0 800
N.N N.N N.N N. N 95 2 N N 5000000<0 0088008 080
00 N.N N.N 0 N N800 .2 N N 020>< 0.003 08888 880
N.N N.N N.N NN N. 8.0-0 2 N N 00030200 08888 888000
N N.N 00 NN N N800 2 N- N 000005 000002 08088 8.00
N.N N.N N.N NN N 800 2 N N 500000 00<00 0003 8.088 0880
N.N N.N N.N N N 0000 .2 N- N 5000000000 08888 88880
N.N N.N 00 NN N NNN-0-0 .2 N- N 00005 N.NN-0000 888 800
N.N N.N N.N NN N 030-0 2 N N 000005 3000005 880880 000
N.N N.N N.N NN N 0080-0 2 N- N 00000 8.80 880
N.N N.N N.N NN N 80.0 .2 N N 0000200000 000<0 8888 808080
N.N N.N N.N NN N 83-3 .2 N N 0000000 00200005 888808 83880080
N.N N.N N.N NN N 0:00 .2 N N 000000 00000 8808.: 880
N.N N.N N.N NN N NNN-0-0 .2 N N 03000000200050.0020 888.0 8800
N.N N.N N.N NN N £00 .2 N- N 000002 0003 08880 8883
N.N N.N N.N NN N N N300 .2 N N 000 20020000 052300 880.800 8888.300
N N.N N.N :- N 0:00 3. N _. 0000 000000.000: 000000.00: N000
N.N N.N N.N NN N 80.0 02 N N 0000< 5.02 08800 8003880
2 N.N N N N0 0N 030-0 2 N- N 0000000000005 80080.8 8.8.80
N.NN N.NN N N NNN NN 080-0 3.. N . 0050002 0000-00 N0 0000000 00000000
N.NN N.NN N.N NNN NN N800 2 N- N 000000.020 00.000 08080 08.080
NNNN 00.000
N.N N.N N.N N N 0080.0 02 N 00000 883 880
N.N N.N N.N N N 205.3 02 N 00.00 02000020 880.: 800
N.N N.N N.N N N N300 2 N 0000 0003 8880.8 .8000
N.N N.N N.N N N 0080-0 2 N 00000 88.83 80.0
N.N N.N N.N N N 0.0.0 2 N 0030200 08888 8880000
N.N N.N N.N N N £00 2 N 00000200 80880000 8800080
N.N N.N N.N N N 800 2 N 200 2<0N002< 88.808 .385
- 0N0, . 000.0 M000- 0000:. 0 00,22 202200 0522 0000020000

 

0.880 .N 00.803.

 

 

 

 

N.N 00 00 2 N 880.0 02 N- N 00000 80.80.N 880
N.N N.N N.N N N N 800-0 02 N N 0000 0003 8088.8 8.800
N.N N.N N.N 2 N 8.0.0 2 N N 0400 2020000 52300 88088 8888080
N.N N.N N.N S N NNN-0-0 2 N N 00005 300005 88880 000
N.N N.N N.N 0 N NNN-0-0 N2 N N 50300-020000 88.800 080000
N.N N.N N.N NN N 0000 2 N N 500000 00000 800000 8880
N.N N.N N.N NN N N800 02 N- N 00005 2.0000 888 000
N N.N N.N NN N 800-0 02 N N 00030200 08888 08880000
N.N N.N N.N NN N 85-3 .2 N N 0000000 0020005 888808 80.088080
N.N N N.N 0N N N 80.0 02 N N 05.200000 000<0 8888 88080
N.N N.N N.N NN 2 8.00 02 N N 03000000200002.0520 8808 8800
N.N N.N N.N NN N 8.00 02 N- N 000002 0003 08888 88003
N.N N.N N.N NN NN 800 02 N N 500000 0050 0003 8.88 8880
N.N 0.0 N N0 0 N80 NE N . 0000 NN.-000000002 000000 00000 N000
N.N N.NN 00 NN N 8.00 02 N N 0000< 53.0 8808 8808880
N.N N N.N N0 0N N800 .2 N- N 00000000000008 88808 8.800..
N.N N.N N.NN NN- NN N800 N8 N .. 0000000: 0000.00 00 0000000 00000000
NN N.NN N.N- NNN NN N800 2 N- N 000000<20 00<00 0.8880 08.880
NNNN 0:000

N.N N.N N.N N N 800 02 N- N 200 20000000200 88.88 085
N.N N.N N.N N N 800 2 N N 000...: 000 888: 88
N.N N.N N.N N N 880-0 02 N N 00000 8888 880
N.N N.N N.N N N 800 02 N N 00< 00003 88.88:. 88.880
N.N N.N N.N N N N800 2 N N. 00000200 8800000 888080
N.N N.N N.N N N 880 02 N N 00000 0000000000 888 8.0880
N.N N.N N.N N N 85-3 02 N N 00000-20000 5000000 0.8.0888 8080
N.N N.N N.N N N 880-0 2 N N 00000 883 880
N.N N.N N.N N N N808 2 N- N 00030<000 0.880 N80
N.N N.N N.N N N 8.0.0 2 N N 520000000 0000 880880 088.800
N.N N.N N.N N N 85-3 02 N- N 52.20000 88.88 88888080
0.N N.N N.N 0 N N000.0 2 N N 50000000030 000020 000080 88880
w 00.0 000,0 .. z .0000. 0.00 000- 0000-00,, 0 02.02. 2020000- 0022 0000020000

 

00.800 .N 00.80000

184

 

 

 

 

 

N N.N N.N NN N N800 .2 N N 00000200 88888.8 88880
N.N N.N N N.N NN N N.N-0-0 3. N- .. 0000 0055 00000800000 0.0800
N.N N.NN N.N NN N 0:00 2 N N 5000000<0 00.000 088888 880

N.NN N N.NN NN N N800 .2 N N 03000002000020.0020 8888 8800
N.NN N.NN N.NN NN 0 80-0 NN. N . 00<0002 0N00<0 N0 000N000 000N300
N.NN N.NN N.N N NN N 0800 .2 N N 500000050 202200 0888008 880
N.NN N.NN N.NN N N N N 85-3 2 N N 0000000 320005 88888 808880
NNNN NNN-.0

00000 000

N.N N.N N.N N N 800 02 N- N 200 2<0N002< 8888 88S
N.N N.N N.N N N 800 .2 N N 000.02 0<000 88888 8N.
N.N N.N N.N N N 880-0 2 N N 00000 883 880
N.N N.N N.N N N 85-3 .2 N N 00000-20000 5000000 N80888 80.80
N.N N.N N.N N N 80.0 .2 N N 00000200 88888.8 88880
N.N N.N N.N N N 050-0 02 N N 5202000... 0000 88880 888.800
N.N N.N N.N N N N800 .2 N N 030.00 0003 N8888 8800.
N.N N.N N.N N N N800 3.. N ._ 000000 002.00 0000200080 00000000
N.N N.N N.N N N 800 2 N N 50000002 000 8N8 888
N.N N.N N.N N N 8000 .2 N N 00000 8888 880
N.N N.N N.N N N 0500 2 N- N 00030<000 0.80 N80
N.N N.N N.N N N 880-0 2 N N 000000 0200002 888888 8880
N.N N.N N.N N N 80-0 .2 N N 50000000030 000020 008.80 888800
N.N N.N N.N N N 80-0 02 N N 50006-00030 5020 0888 888800

N N.N N.N N N 800 .2 N- N 00< 000 8888880 88880

N N.N N.N 0 N 85-3 .2 N- N 5>N-200N00 88.88 8888880

N N.N N.N 0 N 80-0 .2 N- N 00<00<0-02000 88808888800
N.N N N.N N N 050.0 .2 N N 020>< 00003 08888 880

5000000000
N.N N.N N.N N N N N80 2 N N 0003 No 5000000 888 800
N.N N N.N N N 80.0 2 N- N 00005 00002 8888 000
N.N N.N 00 N N 80.0 2 N N 2000000 202200 8800880 880.80
000 0000 0000 000 000 0000 00 0 02<2 202200 02.2 00.002050

 

0.080 .N 008000.

185

 

 

 

 

N.N N.N N.N N N N800 3 N .. 20500200 202200 0002000000 200 000000
N.N N.N N.N 2 N N800 2 N N 020>< 00003 88888 280
N.N N.N N.N N N 85-3 2 N- N 0000 02000030 880: 8.:
N.N N.N N.N N N 85-3 2 N- N 53.20000 88.08 0888880
N.N N.N N.N NN N N800 2 N N 0003000 0.80.88: 8.0880
N.N N.N N.N 2 N 080-0 2 N N 00.000000200002020 88.88 8800
N.N N.N N.N NN N N800 2 N N 00000200 $80583 $80080
N.NN N.NN N.N N0 N 880.2 N N 500000030 202200 08080008 880
N.NN N.NN N.NN 0N NN N800 3 N ._ 005.002 050-00 00 30000-0 0003-EN.
N.NN N.NN N.NN NN NN 85-3 .2 N N 0000000 570005 580858 88.88080
N.NN N.NN N.N NNN 0 080-0 2 N N 00030000 $8.85 888800
NNNN NNN-0

N.N N.N N N N 0500 .2 N- N 00030000 0080 800
N.N N.N N N N 85-3 2 N- N 0000 0500030 080.: 8.0
N.N N.N N N N 80 .2 N N 0.00 0030 888; 08830
N.N N.N N N N 080-0 .2 N N 3.0000000 02<00<00 £8000 $.80
N.N N.N N N N 80 2 N N 00002 0<000 288888 80

5000000000
N.N N.N N N N 880 2 N N 003 No 500000 088:0 800
N.N N.N N N N N N800 2 N N 00000 5<2 5880 20000880
N N.N 0 S N N800 2 N N 020>< 00003 8:888 s80
N.N N.N N NN N 880 2 N N 500000030 202200 0808.088 .380
N.N N.N N NN N 080-0 2 N N 00030000 88.85 888000
N.N N.N NN NN N 880 2 N N 5000000<0 00<00 882888 880
N.N N.N NN N N N 85-3 2 N N 0000000 52605 0808830 80088050
N.NN N.NN NN NNN NN 080-0 NE N . 00<0002 020-00 00 300000 3000-NN.
NNNN N10
N.N N.N N.N N N 800 2 N N 0.00 0050 8.008 888w
5000000000

N.N N.N N.N N N N030 2 N N 0003 8 500200 088:0 800
N.N N.N N.N N N 800 2 N N 3500000 0<022< 888 20.80
E0 0000 0000 000 000 .0000 0 0 1 0202 202200 02.0.2 0N0N020N00

 

 

  

   

 

6.050 .N 0.88.?

 

186

 

 

 

 

N.N N N.N N N 800 .2 N N 500000 00.00 0003 888 85:0
N N.N N.N N N 080-0 02 N N 0003000 88.58 888800
N.N N.N N.N N N N800 02 N- N 0003900 200000 888 2880
N N.N N.N N N 800 2 N N 0.00 00<00 8.808 8880

5000000000
N.N N.N N.N 0 N 880 2 N N 0003 No 5000000 08080 08.00
N.N N.N N.N NN N 080-0 02 N N 0000022 880808 888080
0 0 N.N NN N 080-0 02 N N 0000< 502 8880 8080880
N.N N.N N.N 0N N 880 .2 N N 500000050 202200 08888008 880
N.N N.N N.NN NN 0 8.0.0 02 N N 00<0000002-0.0002<0020 8888 8800
N.NN N.NN N.NN NNN N 080-0 2 N .. 00.000002 020.00 E 300000 00000000
N.NN N.NN N.NN NNN NN 85-3 .2 N N 0000000 520005 8888.08 8888880
NNNN 0:000

000200000
N.N N.N N.N N N N800 02 N N 00220000000 588 88.80
N.N N.N N.N N N 8.0 .2 N N 500000 00.000 003 8.808 880
N.N N.N N.N N N 080-00 02 N- N 00030000 880 80.0
N.N N.N N.N N N 80.0 .2 N N 0000 00<0 080 88%
N.N N.N N.N N N 080-0 02 N N 2002000 0003 880808 88880
N.N N.N N.N N N 080-0 02 N N 305000 02<00<00 88008 8.80
N.N N.N N.N N N 080-00 3 N .. 00000-000 0.02<0 00000200000020.0000
N.N N.N N.N N N 8.00 .2 N N 0002 000 .888 8N-
N.N N.N N.N N N 080-0 2 N- N 00002 8.80 8.80
N.N N.N N.N N N 820 02 N N 5000000<0 00<00 0088.88 880
N.N N.N N.N N N 8.0 .2 N N 0<0 00000 8.808 8880
N.N N.N N.N N N .0800 N.N N .. 0003000002 N0 30000 00002000
N.N N.N N.N N N N800 3 N _. 00>000 0.003 020000 20200000
N.N N.N N.N N N 880-0 2 N N 00000 8800808 880
N0 N.N N.N N N .0500 02 N N 000000 0000 88.888 88880

5000000000
N.N N.N N.N N N .080 02 N N 0003 8 5000000 .8880 800
N00 -. 0.00.0 - 0000. 00.10. . .000 00.00 0 -0- 0202202200. 02.02 000020000

 
  

187

 

 

 

 

 

N.NN N.NN NN NNN 2 805-3 .2 N N 0000000 020005 88858 8888080
N.NN N.NN NN NNN NN 080-0 08 N .. 005002 000050 E 500000 300.0000
NNNN 0:000
N.N N.N N.N N N .800 2 N- N 00030500 880 800
N.N N.N N.N N N 83.3 02 N- N 53.20000 88080 8.888080
N.N N.N N.N N N 280 02 N N 5000005 0050 08888 880
N N.N N.N N N 85-3 .2 N- N 0.200 02.000030 0080.: 8.0
N N.N N.N N N 8; 02 N N 0<0 0050 8.808 8880
N.N N.N N.N N N 030-0 2 N N 3550000 020000.000 8800.0 .880
N.N N.N N.N N N 80- .2 N N 00080 000 8808 8%.
N.N N.N N.N N N 050-0 .2 N N 2002-0000 0003 88088 88880
N N.N N.N N N NNN-0-0 02 N N 020>< 0003 88888 880
N N.N N.N N N 805 02 N N 500000 0050 0003 8.88 0880
5000000000
N.N N.N N.N NN N 0.80 2 N N 0003 No 5000000 .0880 800
N.N N.N N.N NN N N.NN-0 2 N N 00030000 88.288 88080.0
N N.N N.N NN N 030-0 02 N N 00500002000200.0020 88880 N880
N.N N.N N.N NN N 080-0 2 N N 00000200 8888.: 88880
N.N N.N N.N NN N 80.0 02 N N 0000.0 5.02 8880 8000880
N.NN N.NN N.N. NNN N 0.80 .2 N N 500000050 202200 088888 880
N.NN N.NN N.NN NNN NN 83-3 .2 N N 0000000 0020003 88888 8008880
N.NN N.NN N.NN NNN NN 080-0 30 N .. 005002 0N00<0 E 30000 000.300
NNNN 0.8.00
N.N N.N N.N N N 080-0 30 N . 00>000 0003 020000 20000000
N.N N.N N.N N N 080-0 02 N N 00005 300005 88880 8.:
000000
N.N N.N N.N N N 080-0 .2 N N -0003 300005 202200 808 880
N.N N.N N.N N N 030-0 02 N N 0050-0.0000 80.00080 880
N N.N N.N N N 030-0 2 N- N 00000-005955 50000.8 8880
N.N N.N N.N N N .0500 .2 N N 5000005 0050 88888 880
N.N N N.N N N 80.0 02 N N 3550000 5250-000 8880 8.80
E0 0000 0000 000 000 0000 0 0 0322 2020000 0022 0000520000

 

00.880 .N 00803

188

 

 

 

 

N N.N N.N NN N 80-0 2 N N 2002500 0003 88088 8880
N.N N.N N.N NN N 80-0 3. N .. 005002 0N00<0 E 0000000 0000.000
N.NN N.NN N.N NN N 80. 2 N N 00.0 0.003 88.88 8080
N.NN N.NN N.NN 0N 2 80 .2 N N 500000 0050 0003 8.88 880
N.NN N.NN N.NN NN NN 805-3 2 N- N 53.20000 88.8: 8888080
NNNN 80
50200200
N.N N.N N.N N N 80.0 02 N- N 00030052 000000 88888 88880
N.N N.N N.N N N 80-0 08 N . 20000025 202200 0002000000 200 0.00000
N.N N.N N.N N N 080-0 30 N .. 0003000502 00 000000 00 002000
N.N N.N N.N N N 030-0 .2 N N 3550000 525050 8808 8.80
0.00000
N.N N.N N.N N N 00.0 NE N .. 2000<0 005.00 000000 00000000 00000
N.N N.N N.N N N 8.5 2 N N 000002 000 88: 80
N.N N.N N.N N N 833 2 N- N 00.000 0250030 080.: 8.0
N.N N.N N.N N N 0300 2 N N 500000000 0050 88888 .380
N.N N.N N.N N N 80-0 2 N N 2002500 0003 88088. 8880
N.N N.N N.N N N 80 .2 N N 0.00 0050 8.88 880
N.N N.N N .N N N 805 .2 N N 500000 0050 003 8.088 880
N0 N.N N.N N N .800 .2 N N 000000 0000 28.888 8880
N0 N.N N.N N N 8,-0-0 N8 N . 00>000 0.003 020000 20000000
N.N N.N N.N N N 800 .2 N- N 00030500 0.80 800
00 N.N N.N N N 80-5.. 02 N N 0003000 88880 88.80
N.N N.N N.N N N 80-0 .2 N N 020>< 0.003 8888 :80
5000000000
N.N N N.N NN N 0.80 .2 N N 0003 No 5000000 .8880 800
N.N N.N N.N NN N N.N-.00 .2 N N 00<00N0002-0.000.2<0020 8880 8800
N.N N.N N.N NN N 80-0 .2 N N 00000200 88808 88080
N.N NN N.N NN N 80-0 .2 N N 00030000 88.880 888800
N.N N.N N.N NN N 80-0 .2 N N 000.? 5<2 8880 8000880
N.NN N.NN N.N NNN N .800 2 N N 500000050 202200 088888 880
N.N-00 . - N.N-00.0 .. .. 000.00 00.0 0.0 0000 0.0. .0. . @2002 202200 Egg

 
  

- NNN-.00 .N 88.2

189

 

 

 

 

N.N N.N N N N §5 .2 N N 500000900 038088 080
N.N N.N N N N .0800 .2 N N 2.05 0003 8800 88820
N.N N N N N 080-0 02 N N 0000500000 0000002000 88.8.5 8.880
N.N N N N N 080-0 .2 N N 020>< 05003 88888 580
N.N N N N N .0300 .2 N- N 0003000 5.00 805808580
N.N N.N N N N 080-0 .2 N N 3.0050000 0202000 $5088 2030
N.N N.N N.N NN N 805 02 N N 500000 00<00 0003 8.088 85:0
00000005
N.N N.N N NN N 080-0 02 N N -0005 00200000005030 2883» 208280
N.N N.N N.N N N 853 .2 N- N 550200000 88.008 80888080
N.N N.N N.N NN N 0800 .2 N N 50000300 20005002 0030050080
N.N N.NN N.N NN N 025 02 N N 5000000 00<00<00 s88 NN.N-0
N.N N N.NN N NN N ~25 02 N N 00< 05003 88085 8:080
N.NN N.N N.NN NN NN 080-0 N.N. N .. 00.05002 0300 N0 300000 000000-00
NN 0N N NN N 0080.0 .2 N N 00000 8.080080 880
NNNN NN.-.0
N.N N.N N.N N N 080-0 02 N N 0000200000 00000 88588 8080.80
N.N N.N N.N N N 030-0 02 N N 00<0050002-0.0052<0020 88.830 8800
00000
N.N N.N N.N N N 0080-0 .2 N N 0003 00055050000 880 880
N.N N.N N.N N N 805 .2 N N 0<0 05003 808 88:0
N.N N N.N N N 03.0-0 02 N N 3<050000 0202000 238088 53.80
N N.N N.N 2 N 050 02 N- N 0003000 5<00 888003580
N.N N.N N.N NN N 85-3 02 N N 0000000 32000005 N.NN-0858 3088.050
N.N N.N N.N N N 080-0 02 N N 0000000200 00000000 000qu 80.0880
N.N N.N N.N NN N 050 NE N .. 0000 0000005002 000000000: 0000
N.N N N.N 0 N 025 02 N N 000000 0<000 588880. 80
N.N N.N N.N N N 080-0 02 N N 00030902 5000 28.8.0 5:802:
N.N N.N N.N 0N N NNN-0-0 02 N N 500005 300005 888000 NNN:
N.N N.N N.N NN N 0300 .2 N N 50000300 20005002 08805380
N.N N.N N.N NN N 0080.0 .2 N N 00000 8.080080 830
000 0000 0000 000 000 00000 0 0 00002 202200 0022 0000520000

 

 

0.080 .N 850%.

190

 

 

N.N N.N N.N N N 8.5 .2 N N 020 05003 88 8880
N.N N.N N.N N N 080-0 02 N N <0N5<000 00000-02000 88.888 8880
N.N N.N N.N N N 880 2 N- N 500002<00 0000000 8888.888 .88 8808 8888.:
N.N N.N N.N N N 080-0 02 N N 3<050000 0202000 88888 8880
N.N N.N N.N N N 850 2 N N 500000 00000 88808.8 88:0
N.N N.N N.N N N 880 02 N N 5000300 200055002 888000880
N.N N.N N.N NN N 985 02 N N 0<0 000 8.88 8880
N.N N.N N.N N N 050 .2 N- N 0003000 5.000 888-08880
N.N N.N N.N 2 N 83.3 .2 N N 0000000 52605 88088.08 880888880
00000005

N.N N.N N.N 2 N 88.0 02 N N .0055 00><00-000050000 888.880 888880
N.N N.N N N: N 880 .2 N N 500000030 202200 88888808 880
N.N N.N N.N NN N 8080-0 2 N N 00000 88800880 880
N.N N.N NN NN 0 85-3 .2 N- N 53.20000 88.088 88888880
N NN N NN N 885 2 N N 5000000 00<00<00 888 8580
N.N N.NN N.N NN N 285 02 N N 00.0 05003 88.58 8880
N.N N N.N N.NN E N. 080-0 3. N .. 0045002 0300 00 300000 0005000
N.N N N.NN N.N NN N 885 02 N N 500000 00<00 0003 8888 8880
NNNN N30

N.N N.N N.N N N .0200 .2 N N 500000 00000 88808 88:0
N.N N.N N.N N N 080-00 .2 N N 305800 0<022< 8.88 8880
N.N N.N N.N N N 85-3 02 N N 00<00 0020200 88.888 8.:
N.N N.N N.N N N 8.5 02 N N 0.00 05003 88 8880
N.N N.N N.N N N 853 02 N N 0000000 0020003 888808 880888880
N.N N.N N.N N N 080-0 .2 N N 0000000200 0000-000 80088 8.8880
N.N N.N N.N N N N.N-0-0 .2 N N 000309002 500<0 888.8 888.880
N.N N.N N.N N N 025 2 N N 0<00<00<0 88.88 88.0880
N0 0N N.N N N 0.500 3. N ._ 0000 00000005002 000000000: 0000

00000

N.N N.N N.N N N 880.0 02 N N 0003 00055050000 88 8880
N.N N0 00 N N 850 02 N- N 500002<00 0000000 8888.588 .88 8808 8888.0
000 . 0.01010! 0000 000 000 .0000 .8- 0 , 0022 20.000000 -EV0NN0-5-NM0N0M

 

 

 

0.880 .N 08888

191

 

 

 

 

 

 
  

N.NN N.NN N.NN NNN NN NNN-0-0 2 N .. 00.0502 0505 E 000000 808000
NNNN 080%
N.N N.N N N N 030-0 2 N- N 000000053003 88888 8888.80
N.N N.N N N N NNN-0-0 2 N N 000020000 035 88888 8.88.80
N.N N.N N N N .800 2 N N 55000 0000002000 88.888 8880
N.N N.N N N N 8:00 2 N- N 0003000 5<00 8888,8880
N.N N.N N N N 080-0 2 N N 3<05N000 0202N0N 88888 8880
N.N N.N N N N 80 2 N N 00.22 000 8888 880
N.N N.N N N N N.NN-0 2 N N 000000200 0000-000 888.8 888880
N.N N.N N NN N NNN-0-0 2 N N 000.2 52 88880 888880
N.N N.N N NN N NNN-0-0 2 N N 000-300<02 500.0 888.8 8888.0
N.N N.N N.N NN N 0080-0 2 N N 00000 8888880 880
N.N N.N N NN N NENNN 2 N N 50000300 20000002 888083880
N.N N N.NN NN NN 080:0 2 N N 3<0NN000 0<022< 8.88 8880
N.N N.N N.N NN N 8: 2 N N 0? 0.003 88.888 88880
N.N N.N N.N NN N 800 2 N N 03550200055020 88.88 8880
N N N.N NN N 83-3 2 N N 0000000 5200005 888888 88.088880
N N N.NN NN 2 085-3 2 N- N 52-20000 8888 8888880
N.N N.N N N.N NNN N 80. 2 N N 500000 0050 0003 8.888 8880
N.NN N.NN N.NN NNN 2 080-0 3 N . 005002 0300 .0 30000 808000
N.NN N.NN N.NN NNN NN NNN-0-0 2 N N 20N2<000 0003 88888. 88.880
NNNN N.NN-NNN
N.N N.N N.N N N 8; .2 N- N 00000 000 88:08 8N-
N.N N.N N.N N N 8; 2 N N 00.22 000 8888 880
N.N N.N N.N N N NNN-N .2 N N 50000sz 8888 8.888880
0000N
N.N N.N N.N N N 0080-0 .2 N N 0003 0005550000 88 880
N.N N.N N.N N N NNN-0-0 2 N N 2030000 888880 8.8880
N N.N N.N N N NNN-N.NN 80 N . 080 500.0502 000000008 .000
N.N N.N N.N N N 080-0 2 N N 000000200 00000000 8888 88880
N.N N.N N.N N N NNN-0-0 2 N N N20>< 05003 88888 880
, 00 00.0.0 . 0000 , 800- 000 _ v0.0.0 8.- m 02.2 202200 02<2 0N0N-N20N0N

 

 

8.880 .N 08882

192

mODmm

 

 
  

N.N N.N N.N N N 0080-0 2 N N 0003 00050050000 88 880
N.N N .N N.N N N 80 2 N N 040 0003 88 8880
N.N N.N N.N N N 080-0 2 N N NON-2000 00000-02000 88.888 8880
N.N N.N N.N N N 8.0 2 N N 500000000 888888 880
N.N N.N N.N N N 80 .2 N N 0<00<00<0 8888 088080
N.N N.N N.N N N 285-3 .2 N N 00800 002200 88808 .80
N.N N.N N.N N N 800 2 N N 2000000 202200 8888888 880.80
N.N N.N N.N N N 080-0 2 N N 000020000 00.00 88888 8.8880
N.N N.N N.N N N .880 3 N _. 0000 0000000002 800000008 N000
N.N N.N N.N NN N .880 2 N- N 500002<00 00000000 888888 .88 .888 8888.0
N.N N.N N.N N N 800 2 N N 0000000200 0000-000 88088 88880
N.N N.N N.N 2 N 080-0 .2 N N 3000000 0202000 88888 8880
N.N N.N N.N NN N 080-0 2 N N 0082 52 88880 8808880
N.N N.N N.N NN N NNN-0-0 2 N N 020>< 0.003 88888 880
N.N N.N N.N NN N 8800 .2 N- N 0003000 5<00 88808880
0000000

N.N N.N N.N NN N 800 2 N N .000 00>000-000000000 888.880 888880
N N.N N.N 0N N 8.0.0 .2 N- N 0000000020008 88308 8880.88
N.N N.N N.N NN N .880 2 N N 500000030 202200 88888088 0880
N.N N N.N NN N N080-0 .2 N N 00000 88800880 880
N.N N.N N.N NN N 800 2 N N 00030902 500<0 888.8 8888.0
N.N N.N N.N NN N 80- 2 N N 5000000 00<00<00 N88 880
N.N N.N N.N 0N N 0080 .2 N N 50000300 20000002 888088880
N.N N.N N.N NN N 80 2 N N 500000 00.000 003 8.888 8880
N.N N.N N.N NN NN NNN-N-NN 2 N N 30000000 0<0zz< 8.88 8880
N N.N N.N 0N N 80 2 N N 00< 0.003 88.888 8880
N.N N.N N.N NN N 885-3 2 N N 0000000 5700005 888888 8888880
N.N N N.N 0N N N 8.0.0 2 N N 00000000200050.8520 8888 8880
N N N.NN N.N NNN 2 085-3 .2 N- N 52.20000 8888 88888880
N.NN N.NN N.NN NNN NN NNN-0-0 2 N N 200200000 003 888888 88880
x00 800-00 0000: N 0,0 0.00 00.00 .8- 0 02.02. 202200. E

6.880 .N 088...?

193

 

N.N N.N N.N N N 80 .2 N N 0<0 0.003 88 .8880
N.N N.N N.N N N N800 2 N N 2<5 0003 888. 88880
N.N N.N N.N N N 80 2 N- N 00000 000 88808 8%
N.N N.N N.N N N 080-0 .2 N N 008200000 00.20 88888 8.8880
N.N N.N N.N N N 08.0.0 .2 N N 00000000 0000000 80088 888880
N.N N.N N.N N N 880-0 2 N N 300000 0207000 88888 8880
N.N N N.N N N 80 .2 N N 0<00<00<0 8888 88880
N.N N N.N N N 880 .2 N- N 500002-000 0000000 888.888 .88 888 8888.:
N.N N.N N.N N N 08.0-0 2 N N 2000005 202200 8888880 888.80
N.N N.N N.N N N 8800 .2 N- N 00000000200000 88800.8 8888.80
N.N N N.N N N 8.80 3. N .. 0000 0000000002 8000-00008 8000
00000
N.N N.N N.N N N 8080-0 .2 N N 0003 00050-50000 88 880
N0 N.N N.N N N 0880 .2 N N 500000 00000 88808 N880
N.N N.N N.N 2 N .880 .2 N- N 0003000 5.00 88888880
N.N N.N N.N NN N 8800 .2 N N 000300402 5000 888.8 8.8880
N.N N.N N.N NN N 8080-0 .2 N N 00000 8888880 880
00000000
N.N N.N N.N NN N 8800 .2 N N -000 0030000000000 888.880 888880
N.N N.N N.N NN N 80 2 N N 5000000 00<00<00 N88 880
N.N N.N N.N NN N 080-0 .2 N N 00.0.0 52 88880 8880880
N.N N.N N.N 2 N 8.0 .2 N N 500000 00<00 0003 8.888 8880
N.N N.N N.N NN N .880 .2 N N 50000300 2000-N002 8.8888880
N.N N.N N.N NN N 80 .2 N N 00< 0003 88.888 8880
N N.N N.N NN N N 0800 .2 N N 00<00N00N2.0.00-Nz<0020 8888 8880
N.N N.N N.NN NN NN NNN-N-NN .2 N N 30.00000 0<022< 8.88 8880
N.NN N.N N N N0 NN 85-3 2 N- N 52.20000 8888 88888880
N.NN NN N.N NNN 2 885-3 .2 N N 0000000 5200005 888888 88888880
N.NN N.NN N.NN NNN NN 080-0 .2 N N 2002-0000 0003 888888 88880
N.NN N.NN N.NN NNN NN 08.00 80 N .. 005002 2000.0 .0 000000 8080-0.-
NNNN 08.80
00 0000 0000 000 000 0.000 8 0 022 202200 02<2 00020000

 

 

8.880 .N 08.80%

194

W:

 

 

 

N.N N.N N.N N N NNN-N 2 N N 0022 000 8.8808 880
N N N.N N N N NNN-0-0 .z N N 5005 000000-0200 8888 N8:
N.N N.N N.N N N 080-0 .2 N N 020>2 0.003 88888 880
N.N N.N N.N N N 8: 02 N N 000000 00200 0003 8.888 8880
N.N N.N N.N N N 030-0 2 N N 00000000030 >020 08808 888800
N.N N.NN N.N NN N 800 .2 N N 020 00.003 88 8880
N.N N.NN N.N NN N 8: 2 N- N 002 000 88800880 88800
N.N 0. N N N.N NN N 800- 2 N- N 200 22000022 88.8888 8800
N.NN 0.0 N.NN NN NN 030-0 2 N . 002N002 000020 N0 0000000 200000000
N.NN NN N.NN NNN NN 02.0 2 N N 0022 02000 88888 080.-
. NNNN N000

N.N N.N N.N N N 0200 .2 N N 0000000200 202200 8888008108 8880
N0 N.N N.N N N 0080-0 .2 N 0 00000 .00220 00888880
N.N N.N N.N N N 050.0 0.2 N .. 002-N002 000020 00 300.000 000000.02
N.N N.N N.N N N 080-0 2 N N 00003 0000000200 88088 88.:
N.N N.N N.N N N N.NN-0-0 2 N N 020>2 0.003 88088 880
N.N N.N N.N N N 80.0. .z N N 000000 00200 0003 808-8 8880
N.N N.N N.N N N 800. 2 N N 020 0.003 88 8.880
N.N N N.N N N 8.: 2 N- N 200 200022 888.588 8800
N.N N.N N.N N N 85-3 2 N N 0000000 200003 888888 808888880
N N.N N.N NN N 0880 .2 N N 0000032000 0202200 8888 88880
N.N N.N N.N N N 050-0 2 N N 00000000030 >020 08808 888800
N.N N.N N.N 0 N 050-0 Nz N N 000000090030 05020 8088080 80.08800
N.N N.N N.N 0N N NNN-N .0 N- N 002 000 88800880 8.0.88.0
N.NN N.NN N.NN NNN NN 020 02 N N 00.22 02000 8.8888 882
NNNN NNN-..N

000000 0000

N.N N.N N.N N N 080-0 02 N N 0003000 88000880 8.8085
N.N N.N N.N N N 030-0 02 N N 200.2000 00000-02000 88.8888 888800
N.N N.N N.N N N 800 02 N N 00.22 000 80.88 0800
N.N N.N N.N N N 205-3 .2 N N 0200 002200 808.888 0.88
000 . 80.00 - 100.00% - 0.0 w, 00.0,- 00000 8 0 - - 0.202 20.2200 -0222 0000-20000

 

 

 
 

 

.0NN.NN@..N-0N=_N002

195

 

 

 

 
  
 

, eaflmv .N 522.2

33 N.NN-N.NN

N.NN N.NN N.N N N eon-N 2 N- N NNN-NOS NNNNENNN «3:88 so:
N.NN N.NN N.N N N eon-N 2 N N 5535-08 swig-So 33:3 3.22:3
N.NN EN N.N N N eon-N 2 N. N EBB-Ema Eoozm NNN-cans 8335.6
N.NN N.NN N.N N N eon-N 2 N N NNN-«ZBNNNN NNN-NE Sagas «582%
N N.NN N.N N N NNNENN 2 N N >§m§u<Nm 202N200 NNN§§NNNNN§ 35

N N.NN N.N N N spam 2 N N ENNNmmBSN-NN ozszNNNN 83% Nae-seam
N.N N.NN N.N N N 8; .2 N N ban-Eu MEN-NNN NNN; 328% 352m
N.N N.NN N.N N N oNBNgN .2 N N momma .NmES mania-Eu
N.N N.NN N.N N N eon-N E o N N2m>< NEE» 3%?8 sew
N.N N.N N.N N N N as”: 2 N N mgmhmoNz-NREZNNNNuzn-N 853:: 885
N.N N.N N.N N N eon-N z N N NNN-NOS omNN-NNNNNNdzo-N 3222 so:
N.N N.N N.N N: N eon: E N N -N<mN zozo-NON >238 nae-3E £52938
N.N N.NN N.N- N N 3:- Nz N N M5 ENE» SN: 3.2%
N.N N.N N.N NN N 0252 N- N NNN-NNN 252% 2885 3:5
N S N.N N.N N 02» .2 N- N :3 ANNE §§§§§ 33¢
N.N N.NN N.N x: N 85.? 2 N N «NNN-”NBC <NzNoNNN> cNNo-NNNSNNN §NNN8§§N
N.N N.N N.N NN N eon-m NE o .. 92522 2520 E SEN-NE SEN-NV
N.N N.N EN NN NN £93 2 N 0 3538mm N<Dzz< N.NN-NR: $.56
N.N: N.N N.NN NNN NNN Nam-N 2 N- N. SmoNo-BNEN N555 N.NN-939 33826
N.NN N.NN N.NN NNN NN 8: 2 N N MEN-«z «<08 52382 33.-
NNxNN NNNEN

N.NN N.NN N.N N N 85-3 .2 N- N E-zoNNoN 38.3 NNEBNNNSNS-N
N.NN N.NN N.N N N .NaNNN 2 N- N NNNDmmuNNN 553 $33
N.NN N.NN NNN N N eon-N 2 N N- NNmBCNhN-NSNN N.NN-<25 3%28 §s§§§
N.NN N.NN NNN N N eon-N z N- N 5.55 SEN-Eu §§ :3:
N N.NN N.N N N Nam-N 2 N N ESE-VG? 9822 330%: saNEsN-NQ
N.N N.NN NNN N N NaNNN .2 N N twang-EN 02223 328.. §§N§m
N.N N.NN N.N N N 95,-N-N .2 N N dam zozo-NoN >238 Emugi sags-$8
N.N N.NN N.N N N Nam-N 2 N N QmmBmvNoN 385$ 883::
N.N N.N. N.N NN N own-NNN 2 N N monmN .NmES NNasuNNENG
, EN .. Noun, SEN - N61,».- -- MEN I NNN-NR, ARI-w Biz. ”Nb-gnu. N922 2.NN-EBB

196

 

 

 

  

   

 

N.N NNN N.N NNN N ”Naomi .2 N N momma .NmNzS N.NNNENNEG
N.N N N.N N N 8.; .2 N N 5285 MUN-NNN NN-NN3 3:83 352m
N.N N.N N.N. N N Nam-N 2 N N NzNN>< NNN-i3 a§§8 sac
N.NN N.N N.N NN N eon-AN .2 N N Emma-Ema ENE: Nae-9N9 33858
N.N N N N.N NNN N NNaNNN E N N ENNNNNUNo-«NNN 20228 NNENNNNNNNNNNN 25%
N.N N.NN N.N N.N N 8; .2 N- N NNN< NNNE SNNNENNENNN 33¢
N N.N N.N NNN N 85-3 .2 N N NNNNNNmmNNu <NzNoNNN> «NNN-333 §NNN8§NN§N
NNN N.N N.N-N NN NN eon-(N .2 N N 3$NNNNNmm #325. «53% $.36
N.N N N.N N.NN 8 NN N.NN-m 3 N .. 925:2 NNN-:36 NNN Est-NE 353%
N.NN NNN N.NN NNN NNN 8: .2 N N 5.22 «<08 5&383 New
. Nch 9:25

N.N N.NN N.N N N NNN-N-N 2 N N ago-EN; <NN<z<o NNN.NNNNNNNNNNS §s§§32
N.N N.N N.N N N ecu-N .2 N N dothNNN.vNuN-N ANNE-«z gains: 25353
N.N N.NN N.N N N 8; .2 N N bammxoé NNNstNNNac N36
N.N N.N N.N N N ecu-N E N N N<NNN zozo-NoN >238 ENNNNNE segoNNNom
N N.N N.N N N £55 .2 N N ENNNNNUNUNN-NNN 20228 NNN-NNNNNNNNNNNN 3%:
N.N N.N N.N N N 8: 2 N- N NNN-NNN 252% 288% .355
N.N N.NN N.N N N N.NN-N .2 N N 503509- mm><mN-So 35:33 3.22:3
NNN N.N N.N N N NNN-N-N 2 N N NNN-Emma NNN-NE :85qu gNustN.
NNN N.N N.N N N eon-N .2 N N N2m>< NENN3 N.N-$38 =58
N.N N.N N.N N N 8: .2 N N N.NN-«z amNN 22:: NN.-NN.-
N N.N N.N N. N 8; .2 N N NNN-NBS 293m 55 358% 3:5
N.N N.N N.N N N Nam-N .2 N N NNN-NOS 2.NN-023 SEEN so:
N.N N.N N.N N N 8: .2 N N N.N-No NNN-53 33 gemsm
N.N N.N N.N 8 N eon: 2 N N mgNEoNzNamNzfiNUZm 2:223 3826
N.N N.N N.NN N N N N as: 2 N N 3538mm N<Nsz< N533 2&8
N.N N.N N N:- NN N. 95 .2 N- N NNN-N am: 8N§NNN=§N 33¢
N.N N.N N.N NN N 85-3 2 N N NNNNNmmNNo 525m; NNNNNNNNNNNN §NNN8§N§N
N.N N.N- N.NN NN N N ea": .2 N- N- NNNNoNoE-NNEN 555 $939 32%ch
N.NN N.N.N N.NN NN N.N NNN-N-NN 3 N .. 92532 0556 NNN 303mm 534.:-
NNN N.NN. NN NNN NNN 8: 2 N N 3.22 558 5:388 $2.-
SN , Now-NN. 1.9% KNB . 592 . Si N o N922 20228 mz<z uNNNNzNNNuN

 

5.83 .N 55%?

197

 

 

 

 

- N.N.NNNNN .N €8.58

N.N N.N N 83.32 N mom-NNN .88 889
N.N N.N N NNN-N-N 2 N mgmchNz-N.NNEZ<NNUZN 8888 885
N.N N.N NN NNN”: 2 N NooNNN-UN-«zm 5.53 5888 88888
N.N N.N NN £55 .2 N NSENNNNNNSN N88 88888 888
NN N.N N.N 8: .2 N NNN< NN.-NNN 888888 8882
N.N N.N N Nam-N 2 N N2m>< N.N-:23 88888 58b
N N NN 83-3 2 N 8-2022 88.88 288888:
NSNNNNNNNNNNooo
N.N N.N 8 NaNNN .2 N NN-NN3 3 >88: 8888 N88
N.N N.NN NN Nam-N .2 N max-30882 NNN-”SN 5888.8 2888.:
N.N N.N NNN 855 2 N- N 280 205m 85 288.88 882
N.N N.NN NNN 2.NN/-3 2 N N ”NNN-”N80 885 8.8888 83888£
N.NN N.NN NNN N.NN-N 2 N- N 982880 BE 888.8 8888
NN NN N8 eoN-N NE N1 _. 3038202 83-58852 :5 ENE-N
N.NN N.NN NNN Nam-m 3- N . 925:2 NNN-:20 NNN ENE-NE Ems-N.NN.-
NNNN NNN-N
NNNNNNNSNSNNN
N.N N.N N N eon-N N2 N N- 83082: .8223 88888 28888:
N.N N.N N N 8% .2 N N NNNNNNNNNN3$NNN 02228 88% 8§=8m
N.N N.N N N NNN-N-N .2 N N 280 N853 NNN-N888 888.5
N.N N.N N N eon-N .2 N N NNoNNNNN-N-NNUNN 9522 88888: 88858
N.N N.N N N Nam-N 2 N N 553508 BEEN-So 8888 888a
N.N N.N N N N.NN-N .2 N N. EN-mmdzN-NNN 88.8: 8886
N.N N.N N N NNN-N .2 N N NNN-o NNN-53 8:N 888m
N.N N.N N N Nam-N 2 N N 92288 NNN-3N 8888 888N-
N.N N.N N.N N N.NN-NNN .2 N- N 8282-528 NNN-EONN N888 888.8
N.N N.N N.N N eonN-N 2 N N 8<NNNNNNoN2-N.NNm.Nz<NNozNN 8888 885
N.N N.N N.N N N.N-N-N 2 N N N85 8288 >238 .8888 888:5
N.N N N.N N 8: .2 N- N NNN-NNN 232% 8.888 825
N.N N.N N.N N ecu-AN .2 N N Nm-NoS 98%-028 N888 N8:
N.N N.N N.N N N8": 2 N- N >NmuNu-NNNm3N 58% 88.83 83886
SN- , 88ml N8- ----- 8W8. NN , m- ,imNNN-N-nggou- , 882 uNNNszNNoN

 

198

 

 

N.N N.N N N N N.NN-N 2 N «PENN oEEzo 88.288 888
N.N N.N N N N NaNNN .2 N NNNNNNmmNNmn-Nm 88888 88888.
N.N N.N N NN N N.NN-NNN .2 N 89596 Bacon 3388 888882
N.N N.N N N N ”NNN-N-N .2 N mom-"N.N .88» 88w
bammmmooo
N.N N.N N N N 855 2 N ad? 8 >565: 8888 88x
N.N N.N N N N Nam-N .2 N mmmmaza 88888.: 8:885
N.N N N N N N £55 .2 N >5NNNNEN$N 203m 888888 88:
N.N N.N N.N N N 83-? 2 N NNN-Emma SEEN; 888888 838888an
N.N N.N N NN N N8:- 2 N :3 9mm 88:88.88N 888$
N.N N.N N NN N NNN-N-N .2 N 8% coo? N8888 88.588
N.N N N.N N N N 85-3 .2 N E-zoNNoN 88.88 8888888
N.N N.N N.N NN N NENNN 2 N 55:50 MEN-NNN 35 888.88 83
N.N N N.N NN N NNN”: .z N- N mam-N383 50258 888 888883
N.N N.N N.N NN N.NN-m N.N. N .. NEE-NNN: 0556 E SEN-NE 8338
N.N N N.NN NN N.NN-N .2 N- N mam-303m: m-NNNNNPN 838888 8888:
N.N N.N N.N N NN N N eon-N 2 N- N 98538 53 888%N 8888
N.N N.N N.NN NN NN N.NN-N .2 N N N2m>< BE? 8888 88w
N.NN N.NN N.N N NNN N N NNN": 3 N1 ._ 203N502 Saw-588:2 :5 8583
NNNN NN.-N
N.N N.N N N N 8.:- Nz N N 50533 NNNFNONN 888% $889
N.N N.N N N N Nam-N .2 N N 3mm; 888888 888.888
N.N N.N N N N 8; 2 N- N Ed. 252% 88888 885
N.N N.N N N N NNN-N-N 2 N- N NNN-NENNNNNEQUS 8838.8 888.28
N.N N.N N N N Nam-N 2 N N ”NB-NNN 02520 88.288 888
N.N N.N N N N NNNoN-N 2 N N monmm 888 88w
N.N N.N N N N N.NN-«N 2 N N agmomm N<Dzz< 8.28 888m
N.N N.N N N N Nam-N 2 N N .SBNEB .923 8888 88888:
N.N N.N N N N N.NN-N .2 N N imN-zozo-NON 8888.8 8888888
N.N N.N N N N N.NN-N 2 N- N m-NN-Nmz 88.8 8.8:
N.N N.N N N N N.NN-N .2 N- N N.NN-N05 Nam-«NNN 8888 NE:
SN . - NNN-NNN, . . 55,8 , , NN.-N. .98. w, , , -.N-Em 8 in: \gzzNNEmo ., N922. BEEN-NNN

   

 

 

 
 

 

‘ -8883 £88de

199

 

 

 

 

- ,, ‘ Aamoov Nag?

N.N 3 3 N _ as: “2 N mgNEoENRBEEZm 35.25 885
N.N 3 N5 N N eon-.. Hz N mEmz mmfla 8233 uEESN
Nd Nd Nd N _ DEN Hz _ >§mmmo§m 20228 afimsgwmzu 3%
N3 N5 3 N _ as: “2 v SENS; mam 2mg 833%. £33
N._ No N.N N N as”: “2 e 83.32 53:323.» saaombom
E to N.N N N 983‘ .z N 50:06 ammoom 338.2 38:35»:
E Nd _.N v v omBm§z N momma BE 325
N Nd _.N : v 05>; “z N Eczema; €3.88 €§§8§N
N 3 _.N N_ v £522 N «my? 0350 3.6.238 .53
N.N ..N N.N “N N 8: Hz N a? 82 Saueagmm 35¢
N.N 3 S. 2 N ea: .2 v Emmmgq nmozwa 5%.; £82.23
N.N N 3 N2 N as: Hz v mo<m moo? manages 3:85
S N.N ..N N 955 .2 N 55:50 205m 35 séatmsu 83
N.N N.N N.N NN eon: “z N maméoaaz 3&5; 2:885» sfiéufi
N.N 3 N.N NN 853 “z N mmmmmmo SEES SEESE §a§§r£
N.N N.N NE NV ea: 2 _ sz>< NEE» gangs 596
N.N 3 N.N NN £8-.— .2 N 922348 ES 83%» egg
N.N_ N_ v.2 NM: as”; .2 .. 925:: 2520 E 395m 533w
N.NN N.NN N.N 5 ea": 2 . 505N202 VEEEEE Emu ENQN
£8 =5.—

3 ..o _ _ eon; Hz N mz<m<md £922 §N§m33q notmtm
N5 3 _ _ 02th N 3.32 ”NE/am s§t§§ 8%
N3 Nd _ N as: Hz N Mme? 9N2: zmmpmé 35883 La?
N.N 3 _ _ oNBNéZ _ mcamm §§3 95
N5 3 _ _ 8.; Hz N 52,23 205m Ems Ecuax
N.N Nd _ N eon: Hz e Sumac; mam 550 832%» £83
3 3 _ _ 8; Hz N >§m=o Mofim 35 358% 33$
N.N 3 _ _ 8P .z N $5 545 83»; 35$
3 Nd _ N 8: Hz N 33m 5.02% 283% 325
3 Nd _ N .3: .z N 9305 :92: 33398 :3:
NE- imoow mg 1N8! ....NEN 9. . Ezlzméow , m2? £5288

 

200

 

 

 

 

   

   

   

   

N N.N N.N NN N 8:. 2 N- N NNN-N NNNE 8.38338N 338$

N.N N.NN N N N 83-3 2 N- N E-zoNNoN .3838 83883on
N.N N.NN N N N NNBN-N N2 3. N momma 38w 836
N.N N.NN N N N eon-N 2 N- N :NSN-m=N-7NN.VNU<N 33335 338N3-
N.N N.NN N.N N N 8:. 2 N- o awn-NNN xom 8883 8N.-

N.N N N NNN N 95$ 2 N N tam-ENE 203m 338388 33a

N.N N.NN N.N N N. ecu-N 2 N- N zoo-«ma 556 33883 Euufiw
N.N N.NN N.N N N Nam-N.N 2 N o 35FNNNNNNN N527? 3.:23 3336
N.N N.N N.N N N 83-3 .2 N- N mac 25E 3.238 3:
N N.NN N.N NN N eon-m .2 o N N2m>< NEE» 83888 38¢

N.N N.N N.N N N N3": E N v NNN-:20 35 83888 333
N.N N.N N.N NN N N3”: 2 N N 228sz NNN-2N 88388 883%
N.N N; N NN N £55 .2 N- N 5528 205m 35 338.88 83
N.N N.N N.N NN NN eon-AN 2 N N mgmNmoNz.N.mm.N.z<NNozNN 8888 885
N.N N.N N.N NNN z eon-N .2 N- N mam-303m: 5.22 8.8853 3883:
N.N N.N N.N NNN NN 2.5-3 2 N N «Emma 52555, 333838 3388§£
N.N N.N N.N NNN NN eon-N .2 N- N nomszn-Noo NNN-3 83%..» 88:8
N.NN N.NN N.N N.NN NNN NNN-N-N N8 3 _. EoEmzosN Saw-533:2 £5 £3qu
N.NN NN N.N NNN z eon-m 3 N .. NNN-«5:2 NNN-:20 EEQNNE NEE-N.NN.-
NSN N.NENN

ZN N.N N.N N N eon-NNN 2 N- N BEEN-«Bo 33:3 8N:
to N.N N.NN N N 880d 2 N- N. Bad-N5 526m; 28333, 333m
to N.N N.N N N 8; 2 N N bammxoé 338388 N36
N.N N.NN N.NN N N 0N8: 2 N N moomm 3383 839
N.N N.NN N.N N N 8: .2 N- o amp-NNN xom 8.383 8N.-
N.o N.NN N.N N N eon-N 2 N N. 222mm 35 333383 .8388
N.NN N.NN N.N N- N NNN-N-N 2 N- N .535 Nam-«2 33:88 .83.:
N.NN N.N N.NN N N 8; 2 N- N N.N-NNN 2.N-02% 8888 .335

Emmmmmoom

N.NN N.NN N.NN N N 355 2 N N. 35 3 55:5 38838 83
E N! mm»: -. 0E --NNo-N WE LNME 3 u :NNEN-zz-oé-B . gsznNNNzNNNoN

     

,- ctr-NE .N 5833

201

 

 

 

N.N N.N N.N NN N 8; .2 N- N NNN< GEN ..........>N.NN..8..N ......NNE
N.N N.N N.N 2 N 85-3 .2 N- N 22-2022 888...... 8.888..ch
N.N N.N N.N NN N NENNN .2 N N NNN-NNNNNNNNNDN 23.-NNN NNNs8N....-.8 88:
N.N N.N N.N N N. ecu-N .2 N N. 2.23 ad? 88888 .55?
N.N N.N N.N NN N N.NN-N .2 N- N 2855 5B6 23.88.83 .....88N.
N.N N.N N.N NN N 82:2 N- N «ma-NNN 20m 888...... 88.
N N.N N.N NN N N.NN-N .2 N N BN<2NUNNNN NNN-N.NN 8828... 8.82.8.
N. N.N N.N NN. N ecu-N .2 N N N2m>< NEE» 88888 58¢
N.N. N.N N.N NN N NENNN .2 N- N 55250 23.-NNN GEN 588.28... 8...:
N.N N.N N.N N.N NN N.N-N-N .2 N- N mam-303m: N.N-NEE 8.88.3. 538%:
N.N N.N N.NN NN N.N N.NN-N .2 N N mgmezoNz-NpNn-szsNuzn-N 88.8.... 88.6
N.N N.N N.N NN.N NN NNN-N.N .2 N- N 9223-50 35 8888 88:8
N..N N.NN N.N NNN 2 2.3-3 .2 N N NEH-N56 <N2NoNNN> ......8858 88.8885
NN NN N NNN NN N8”: 3. N1 _. ZONE-N202 SEN-SEEN N.N2-NEE:
N.NN N..N.N N.NN NNN NN N.N-N-NN N... N _. 925:2 NNN-220 N.N 305$ ENE-NN.
N.NNN NN.—N.NN

N..N N.N N.N N N mac-N .2 N N MEN-NNN; 5:5 3.5.... 3.3-N
N.N N.N N.N N N NNBNN .2 N- N moamm 8N...» 8.39
..N N.N N.N N N N.NN-... 3. N .. 925:2 293m 562 SEN-Em
N.N N.N N.N N N N.NN-N. .2 N- N 82-82-2022 NNNNN-SNN NN.-8.8 .888:
N.N N.N N.N N N N.NN-N .2 N N N<mN 20238 .238 88.88% sage-NEE
N.N N.N N.N N N N.NN-N .2 N N SEE-Sm EBN 838.... 83.3.3.5
N.N N.N N.N N N eon-N .2 N- N Noomm2<2N 2315 8888 3888
..N N.N N.N N N N.NN-N 3. N .. 203323 20228 ..N-SEES :8 $33
N.N N.N N.N N N 888.22 N N momma 8.83 8.8
N.N N.N N.N S N Nam-N .2 N- N NNN-NOS 2922 ......88 N.NN:
N.N N.N N.N 2 N N8”: .2 N N. SEEN-«2N NNN-NNN? .888... 5.8225
N.N N.N N.N 2 N N.NN-N .2 N- N. NNN-N.Nmz moo? NN.-8.888 8:83
N.N N.N N.N N. N NNN-N-N .2 N- N. m202m2< <n<2<u 28888 88.8%

ENNmmmNooo

N.N N.N N.N N N £55 .2 N N. 35 8 5202: .8888 89.:
E ,, NEE N.NN-E N.N-..N, Em ‘ E NN., ‘ .N -E- $5.2 uNNNszNoN

5.83 .N N.NNNNNNN.

202

 

 

  
 

   

N.N N.N N.N NN N N.NN-N .2 N- N. 5% 0003 .8888 ......88N
N.N N.N N.N NN N 855 .2 N N wémmNNSN 0.03m N......8....8.. 8......
N.N N.N N.N NN N N.NN-.. .2 N- N. @883 0.52:: 8...... 8.85.3
N.N N.N N.N NN N ......N-N .2 N N Nan-N.N 2020N0N >238 888......- 25888....
N.N N.N N.N N N N.NN-.. .2 N N. 055.0 0.23 .8888 5......
N.N N.N N.N N. 2 N.NN-.. .2 N- N 20055 2NNBN0 8.888.. 8.88....
N.N N N.N NN N 82$ .2 N- N 55250 0.3-NNN 083 588.82.. 8E
N.N N.N N.N NN NN N.NN-.. .2 N N m0<=NN=0N2-N.NNm-N2<202m 88...... 88....0
N.N N N.N NN NN N.NN-.. .2 N N N2m>< NNN-NNN? .8888 s80
N.N N N.N NN NN N.NN-N .2 N- N mam-303m: 3.25.. $88.08 58.82:
N.N N.N N.N NN NN N.NN-N .2 N- N 00x2m0N00 m5: 8.8...8 8888
N.N N.N N.N NN. NN 2.5-3 .2 N N NNN-N.Nmmmu 52825 5.8888... 88.88.28.
N.NN N.NN N.NN N... N. N.NN-m N... N .. 9.552 0520 N.NN-ENE SEN-N.N
N.NN N.NN N.N NNN N. N.NN-.. N... N- . 50.5.20: 223E302 SEVEN-N.N

NNNN NN....N

N.N N.N N.N N N N.NN-N .2 N N N.NN-NNN. 022920 28.28... 8.8
N.N N.N N.N N N 888.. .2 N N 80% 88... 880
N.N N.N N.N N N ......N-.. .2 N N SENSE 88.58.... .....8».......
N.N N.N N.N N N 880-.. .2 N- N BEAN-=3 32.9.5 .838... 88....
N.N N.N N.N N N 2.3.? .2 N- N NNN-<20 2282.52 38...... ......N
N.N N.N N.N N N ......a .2 N- N ENNNNNNNNNNNNNNN ...-N888 8.88....
N.N N.N N.N N N N.NN-.. .2 N- N 80.5020 05.002 8.888. 8.885...
N.N N.N N.N N N ......N-.. .2 N- N N.NN-NNN..-BN.N-7NN-20<. 83...... 888......
N.N N.N N.N N N N.NN-.. .2 N- N N.NN-N05 N.NN-$2 3888 ......NN
N.N N.N N.N N N N.N-N-N .2 N N $02893 NNN-«>5 88.8... 8383......
N.N N.N N.N N N N.NN-N .2 N- N m-Emz 8...... 8.80

>§mmmN000

N.N N.N N.N N N ......NN .2 N N 083 ... 5202.. 28850 8E
N N.N N.N N N N.NN-N .2 N- N E25800. NNN-N022... 8...... 8.82.3
N N.N N.N N N N.NN-N .2 N N 2.; 083 8...... 8.88.0
N.N N.N N.N NN N N.NN-.. .2 N- N ...-0% 0003 N....--.......... 2......8N
, E .00.. - EE! ....100. | - 0E- .NNE. ..N .0 E2 202,200 NNN-2 02:298.

. ......NSN .N N.N...NNNN.

   

203

 

 

204

m0 N.0 0.0 N N 0m00m-m NZ 0 N mODmm 0.0003 0.0000
n .0 N0 0.0 N N Pam-m NZ N 0 N>N<> Gd? 0000...» 00000095
m0 N.0 0.0 N N 0086-0 .2 N- .0 gm-fld? EOE 00.0.3.3... 030.0%
m0 N.0 0.0 N N Atom-m .2 N- N mEmZ 00.0.0 00.0.5
m0 N.0 0.0 N N 02H .2 m N Wampum-NU vNU<-Nm DAN? 00.00000 000.2%
m0 N.0 0.0 N N 030m N0< m ... MmOm gnu-NE; VZONKE-ND: 0.00%
00 N0 0.0 N N Duo-N-NN .2 N- 0 PNNNADNN-mNNH-E-MU<N. 5.5.03.5 0.0000?‘
>m¢mmmm08
0.0 m0 0.0 m N 030m .2 m N. Gd...» ..0 >AMUNME 000000.00 000.5
00 N0 0.0 m N 02-N- NZ m 0 H3223 VNU<-Nm 00%? 000%.:-
v.0 N.0 0.0 m N Atom-N.N .2 m- N HOZ-mNZ-NNUDO-N. QEONNm 0.000%00 000.0055
0.0 m0 0.0 m N A505 .2 N- m >§mm¢mflqm 0.0000008 000000.00.
0.0 0.0 0.0 m N 030m .2 N N >§mm¥u<qm 203950 00003003.? 0000K
0.0 0.0 0.0 m N 0000. .2 m- N 3%... mid 00~00>~A0000Q 0000.0.Q
m .0 m0 0.0 0 N NN.—0.0-0 .2 m- m Hum-NOS 3mg 0.020000 0.0.3
0.0 0.0 0.0 0N N Atom-a. .2 m- m HOOmBOmU vaNOO: 03030000.. 00~000000-
0.0 N.0 N.N v N Eon-m NZ 0 v Qmmmg 0.000..0..M....> .0000MAN0R
0.0 N.0 N.N N N Eon-«N .2 m 0 BEN-whim A<DZZ< 00....0Nm0 50.00%
0.0 N.0 N.N m N owNvom-NN NZ 0- 0 mOQmNm 090% 0000.0
0.0 N 0.0 0N N 030m 3.. m ... vaNUDm>mZONN MON)? NQDVV: ENUNZOQ
N.N v.0 0N w m 00N>$> .2 N- N E-ZOmNONN ...-000.000.. 000000000NR0N
E -. E... 00%. . 00.0.. . . 0E . .. - M.,...Nm. .m- m .. E2 202200. 82.02 2.NN-206.0

 
  
 

NN....SN .N €8.50

   

IIHEWIMIQLMEIEEWJEEM[Wu