HOST PLANT INTERACTIONS BETWEEN EMERALD ASH BORER AND
FIVE FRAXINUS SPECIES.

By
Sara R. Tanis

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Forestry - Doctor of Philosophy

2013

ABSTRACT
HOST PLANT INTERACTIONS BETWEEN EMERALD ASH BORER AND FIVE
FRAXINUS SPECIES.
By
Sara R. Tanis
Agrilus planipennis Fairmaire (Coleoptera: Buprestidae), a secondary pest of stressed or
declining ash (Fraxinus) trees in its native Asia, is the most destructive forest insect to ever
invade North America. To date, tens of millions of ash trees have been killed in 19 states and
two Canadian provinces. North American ash species are generally less resistant to A.
planipennis than Asian species. Fraxinus quadrangulata Michx. and Fraxinus americana L.
were originally abundant in two woodlots in southeast Michigan, but survival rates of the two
species following A. planipennis invasion differed profoundly. Overall, 63% and 71% of F.
quadrangulata trees were alive at the sites. In contrast, only 16% (all <11 cm diameter) of the F.
americana trees survived at one site and all were dead at the second site.
Urban trees are often planted in confined spaces and may experience varying degrees of
environmental stress. Fertilizer and paclobutrazol (PB), a gibberellin inhibitor, may be applied
to enhance tree vigor. Effects of fertilizer and PB were assessed on the physiology and growth
of North American (F. americana and F. quadrangulata) and Asian (Fraxinus mandshurica
Rupr.) ash trees in a plantation established in 2006. Fraxinus quadrangulata had 33% less radial
growth, 13% more root biomass and 25% lower foliar nitrogen than F. americana or F.
mandshurica. Ratios of pre: post treatment radial growth in control and fertilized trees were 33
and 43% higher, respectively, than PB trees. Aboveground growth of F. quadrangulata and F.
mandshurica were reduced by PB treatment while F. americana was unaffected. Fertilized F.

quadrangulata trees had higher relative chlorophyll content and nitrogen concentrations than
control or PB trees, but other species did not respond to fertilizer.
Suitability of the ash species for adult A. planipennis feeding and survival and larval
density was also determined. Survival of adult A. planipennis caged on F. quadrangulata was
lower and beetles consumed less leaf area than beetles caged on other species. North American
Fraxinus nigra Marsh. and Fraxinus pennsylvanica Marsh. trees were heavily colonized by A.
planipennis larvae. In contrast, F. quadrangulata and F. mandshurica were rarely colonized,
while F. americana were moderately colonized. Results confirm North American F.
quadrangulata has resistance to A. planipennis similar to F. mandshurica.
Systemic trunk injected insecticides are often used to protect ash trees from A.
planipennis, but wounds and injury are a concern. We examined basal trunk sections from 22 F.
pennsylvanica and 24 F. americana trees macro-injected with a low or medium rate of
emamectin benzoate in 2008 only or 2008+2009. Only 12 of the 233 injection sites had evidence
of injury and there was no sign of pathogen infection. Confocal laser scanning microscopy
(CLSM) examination showed xylem discoloration was not indicative of tissue damage. Crosssectional area of earlywood lumen was 55% larger in F. quadrangulata than in F. pennsylvanica
trees. Interspecific differences in xylem anatomy likely influence efficiency of trunk injections.
Phenolic compounds and reactive oxygen species (ROS) are produced by plants in
response to injury, stress or insect damage. Autofluorescence indicative of phenolics was
detected in F. quadrangulata and F. mandshurica phloem, but not in F. pennsylvanica phloem.
Fluorescence of ROS was detected in all the phloem of all three ash species, but F. mandshurica
produced three times more than the other species. A continuous layer of sclerenchymatous cells
was observed in F. quadrangulata and F. mandshurica and may affect A. planipennis larvae.

This one is for you, Muff.

iv

ACKNOWLEDGMENTS
I thank my advisor, Dr. Deborah G. McCullough for her guidance and assistance during
this venture. Deb broadened my horizons and introduced me to the amazing worlds of
entomology and forestry. I also thank my guidance committee: Dr. Bert M. Cregg (for
encouraging me to “press on”), Dr. Therese M. Poland and Dr. David E. Rothstein.
These projects would not have been possible without the help and support of my lab
mates: Andrea Anulewicz, Elliott Berlin, Jacob Bournay, Stephen Burr, Tara Dell, Nicholas
Gooch, Russ Kibet, Emily Pastula, Kyle Redilla, Molly Robinett, Heather Surface, Andrew
Tluczek and James Wieferich. You are my friends and my co-workers. Thank you.
I thank Dr. Melinda Frame from the MSU Center for Advanced Microscopy for her
assistance with my microscopy projects and Paul Bloese and Randy Klevickas from the MSU
Tree Research Center for their willingness to help at the plantation.
I have the most wonderful parents - thank you mom and Dad - I’ll always be your Tiger.
Thank you Lo, for the finals week care packages and for being an amazing “little sister”. Salad.
Thanks also to my Aunt Bonnie and Uncle Norm, my “sister” Michelle, “brother” Jim and my
niece Lauren, my friend and inspiration Dr. Bridget Behe and my cousins Amy and Evy for their
unwavering love and encouragement.
I thank Dr. Wendy F. Klooster – my best friend forever (and ever) – for always being just
a text message away.
Last but not least, I thank my husband Dave for never doubting for one minute that I
could do this, Hazie, Chuck, Ziva, Edison and Franklin for their feline assistance and Timber and
Sunshine for wagging their tails every time I walk in the door. In memory of Gabriel Tanis – I
still miss you every day.

v

TABLE OF CONTENTS
LIST OF TABLES.......................................................................................................................viii
LIST OF FIGURES........................................................................................................................xi
INTRODUCTION………………………………………………………………………………...1
CHAPTER ONE
Differential persistence of blue ash and white ash following emerald ash borer invasion..............4
INTRODUCTION………………………………………………………………………………...5
METHODS…………………………………………………………………………………..........8
Study Sites…………………………………………………………………………….......8
Ash Inventory……………………………………………………………………………..8
Statistical Analysis…………………………………………………………………….....10
RESULTS………………………………………………………………………………………..11
DISCUSSION………………………………………………………………………………........15
TABLES……………………………………………………………………………………........21
FIGURES………………………………………………………………………………………..23
CHAPTER TWO
Effects of fertilizer and paclobutrazol on the physiology and growth of five ash species............26
INTRODUCTION……………………………………………………………………………….27
METHODS………………………………………………………………………………………31
Ash Plantation…………………………………………………………………………...31
Treatments……………………………………………………………………………….32
Variables Measured……………………………………………………………………...32
Biomass Allocation……………………………………………………………………....33
Growth…………………………………………………………………………………...34
Statistical Analysis……………………………………………………………………….35
RESULTS……………………………………………………………………………………......38
Growth………………………………………………………………………… ………..38
Biomass Allocation………………………………………………………………………40
Gas Exchange……………………………………………………………………………42
Relative Chlorophyll Content……………………………………………………………42
Nutrient Content……………………………………………………………………….....43
DISCUSSION……………………………………………………………………………………47
TABLES…………………………………………………………………………………………54
FIGURES………………………………………………………………………………………..66
CHAPTER THREE
Effects of paclobutrazol and fertilization on host resistance and suitability of five Fraxinus
species to Agrilus planipennis…………………….…………………………………………......70
INTRODUCTION…………………………………………………………………………….....71
METHODS………………………………………………………………………………………76
Ash Plantation……………………………………………………………………………76
vi

Adult Agrilus planipennis………………………………………………………..………77
Leaf Feeding Bioassays………………………………………………………………….77
Mortality Bioassays………………………………………………………………...........79
Larval Gallery Density……………………………………………………………….…..79
Statistical Analysis…………………………………………………………………….....81
RESULTS……………………………………………………………………………………......82
Leaf Area and Weight Consumed per Beetle Day……………………………………….82
Adult Agrilus planipennis Mortality……………………………………………………..83
Larval Gallery Density…………………………………………………………………...85
DISCUSSION……………………………………………………………………………………88
TABLES…………………………………………………………………………………………95
FIGURES………………………………………………………………………………...…….102
CHAPTER FOUR
Evaluation of ash xylem discoloration and architecture associated with macro-injections
of a systemic insecticide………………………………………..................................................105
INTRODUCTION……………………………………………………………………………..106
METHODS………………………………………………………………………………….....111
Trunk Injection Damage………………………………………………………………..111
Xylem Architecture…………………………………………………………………….113
RESULTS………………………………………………………………………………………117
Trunk Injection Damage………………………………………………………………..117
Xylem Architecture…………………………………………………………………….119
DISCUSSION………………………………………………………………………………….120
TABLES……………………………………………………………………………………......126
FIGURES……………………………………………………………………………………....127
CHAPTER FIVE
Examination of the phloem of three ash species using fluorescent confocal laser scanning
microscopy ……………………………………………………………………...135
INTRODUCTION……………………………………………………………………………..136
METHODS…………………………………………………………………………………….140
Ash Plantation…………………………………………………………………………..140
Tissue Sampling………………………………………………………………………...140
Confocal Laser Scanning Microscopy………………………………………………….141
RESULTS………………………………………………………………………………………144
Phenolic Compound Autofluorescence………………………………………………...144
Reactive Oxygen Species (ROS) Production…………………………………………..144
Phloem Anatomy……………………………………………………………………….145
DISCUSSION………………………………………………………………………………….146
FIGURES………………………………………………………………………………………150
LITERATURE CITED…………………………………………………………………………153

vii

LIST OF TABLES
Table 1.1: Number of trees, mean (±SE) diameter at breast height (DBH), total basal area and
1
Relative Importance Values for the five most common tree genera recorded in
variable radius plots in woodlots in Plymouth and Superior Township, Michigan.
Values for Fraxinus include dead and live trees.…………………………………..21
Table 1.2: Number of living blue ash and white ash trees by canopy dieback and diameter
classes in woodlots surveyed in Plymouth and Superior Township, Michigan….....22
Table 2.1: Results of analysis of variance to assess effects of treatment on shoot growth and
1
ratios of pre-: post-treatment shoot growth on blue, white and Manchurian ash trees
treated with paclobutrazol, fertilizer or left as untreated control….………….…….54
1

Table 2.2: Average (±SE) shoot length (cm) in 2008-2010 and ratios of pre-: post-treatment
growth of blue ash, white ash and Manchurian ash trees treated with fertilizer,
paclobutrazol or left as untreated controls (2011). Measurements were taken in
October 2011 using annual leaf scars. Within years, different letters indicate a
significant (α≤0.05) difference between species (a,b,c). Among average (±SE) shoot
length (cm) in 2011 and pre-: post-treatment ratios, different letters indicate a
significant (α≤0.05) difference between treatments within species (x,y)..………….55
Table 2.3: Results of analysis of variance to assess effects of treatment on annual radial growth
1

and ratios of pre-:post-treatment growth of blue ash, white ash and Manchurian ash
trees treated with paclobutrazol, fertilizer or left as untreated control.……………..56
1

Table 2.4: Mean (±SE) radial growth (mm) from 2008-2011 and ratios of pre-: post-treatment
radial growth of blue ash, white ash and Manchurian ash trees treated with fertilizer,
paclobutrazol or left as untreated controls. Treatment effects were not significant
(P>0.05) in 2008-2010. Within years, different letters indicate significant (α≤0.05)
differences between species (a,b,c) and treatments (x,y)…………………………...57
Table 2.5: Results of analysis of variance to assess effects of species and treatment on leaf,
aboveground woody, fine root, coarse root, total root and total biomass and leaf,
aboveground woody, and ratios of root biomass: total biomass and fine: coarse root
of blue ash, white ash and Manchurian ash trees treated with paclobutrazol, fertilizer
or left as untreated control. Trees were harvested in October 2011……………….58
Table 2.6: Mean (±SE) biomass (kg) of leaves, total roots, fine roots, coarse roots aboveground
woody and total biomass of blue ash, white ash and Manchurian ash trees treated
with fertilizer, paclobutrazol or left as untreated controls. Biomass was harvested in
October 2011. Among plant parts, different letters indicate a significant (α≤0.05)
difference between species (leaves, total roots, fine roots and coarse roots) (a,b,c) or
within species and treatments (aboveground woody and total biomass) (x,y).
viii

Treatment effects on leaves, total roots, fine roots and coarse roots were not
significant (ns). ……………………………………………………..….…………..59
Table 2.7: Results of analysis of variance of effects of species, treatment and date on
photosynthesis and transpiration rates and relative chlorophyll content of black,
blue, green, white and Manchurian ash trees (2010) or blue ash, white ash and
Manchurian ash trees (2011) treated with paclobutrazol, fertilizer or left as untreated
control. Trees were sampled in June and August 2010 and 2011. …….………….60
2

-2 -1

Table 2.8: Mean (±SE) photosynthesis rates (μmol CO m s ), transpiration rates (mmol H2O
-2 -1 1

2

m s ) and relative chlorophyll content of black ash, blue ash, green ash, white
ash and Manchurian ash trees (2010) or blue ash, white ash and Manchurian ash
trees (2011) treated with fertilizer, paclobutrazol (PB) or left as untreated controls.
Measurements were taken in June and August of 2010 and 2011. Within species,
different letters indicate a significant (α=0.05) difference between treatments.……61
Table 2. 9: Results from analysis of variance of effects of ash species and treatment on macro
nutrients (percentage dry weight) in foliage of black ash, blue ash, green ash, white
ash and Manchurian ash trees sampled in July 2010 or July 2011. ...………………63
Table 2.10: Mean (±SE) percentage of macronutrients in foliage of black ash, blue ash, green
ash, white ash and Manchurian ash trees treated with paclobutrazol, fertilizer or left
as untreated control sampled in July 2010 and July 2011. Different letters indicate
significant (α ≤ 0.05) differences in nutrient concentration among species (a,b,c) or
treatments (x,y)………………………………………….…………………………65
Table 3.1: Results of repeated measures analysis of variance to assess effects of species,
treatment and day (2010) and species, treatment, day and leaf type (2011) on
survival when Agrilus planipennis adults were caged on excised black, blue, green,
white or Manchurian ash trees (2010) and excised or intact blue, white and
Manchurian ash trees (2011) treated with paclobutrazol, fertilizer or left as untreated
control……………………………………………………………………………….95
Table 3.2: P values generated by repeated measures least significant difference tests used to
compare mean number of live Agrilus planipennis caged with leaves from black,
blue, green, white or Manchurian ash leaves over a 14 day period. Bioassays were
performed in June 2010. An “*” denotes significant differences in beetle survival
(α=0.05) between host species………………………………...…………………....96
Table 3.3: P values generated by repeated measures least significant difference tests comparing
mean number of live Agrilus planipennis caged with leaves from blue, white or
Manchurian ash leaves over a 14 day period. Bioassays were performed in June
2011. An “*” denotes significant differences in beetle survival (α=0.05) between
host species………………………………………………………………………....97

ix

Table 3.4: Mean (±SE) number of surviving Agrilus planipennis adults caged with excised
black, blue, green, white or Manchurian ash leaves in June 2010 or excised or intact
blue, white or Manchurian leaves in June 2011. There were two male and two
female A. planipennis per cage. Letters indicate differences among species and
treatments (α=0.05)…………………………………………………...…………….98
Table 3.5: Mean (±SE) number of Agrilus planipennis adult exit holes, parasitized larvae,
2

woodpecker attacks, living larvae, dead larvae and total galleries per m on five
species of ash trees (N=21 trees per species) and P values generated by least
significant difference tests comparing mean densities. An “*” denotes significant
differences between species (α=0.05)……………………………………………...99
Table 3.6: Number of Agrilus planipennis emerged adults from 2009 and 2010 larval cohorts
and total larvae from the 2010 and 2011 larval cohorts including live and dead early
instar larvae (first, second and third), total late instar larvae (fourth and prepupae)
including those parasitized by Atanycolus spp. or killed by woodpeckers found on
five Fraxinus species in a plantation, N=21 trees per species. Larvae from the 2010
cohort were progeny from A. planipennis adults that emerged in 2009. Larvae from
the 2011 cohort were progeny from adults that emerged in 2010. Trees were
harvested before larvae from the 2011 cohort could pupate (Fall 2011)……..…101
Table 4.1: Results of analysis of variance to assess differences between green ash and white ash,
insecticide rate (low rate = 0.1 g a.i. per 2.5 cm DBH or medium-high rate = 0.4 g
a.i. per 2.5 cm DBH), injection frequency (2008 or 2008 + 2009) and their
interactions on the number of trunk injection sites with new wood growing over
them or secondary wounds and tissue discoloration depth, width and height…….126

x

LIST OF FIGURES
Figure 1.1: Number of live and dead blue ash and white ash trees by diameter class
in woodlots in Plymouth (A) and Superior Township (B) Michigan. ……………23
Figure 1.2: Distribution of live (A) and dead (B) blue ash (black points) and white ash
(grey points) trees in 50 × 50 m grid cells in a woodlot in Plymouth,
Michigan. Point size is proportional to tree DBH (cm). Living blue ash
trees ranged from 2.5 - 45.0 cm in DBH. Dead blue ash trees ranged
from 2.5 - 50.0 cm in DBH. Living white ash trees ranged from 2.5 –
10.9 cm in DBH. Dead white ash trees ranged from 2.5 - 96.0 cm in
DBH. Gray quadrants represent areas where no Fraxinus spp. trees
were present. ……………………………………………………………………...24
Figure 1.3: Distribution of live (A) and dead (B) blue ash (black points) and white
ash (grey points) trees in 50 × 50 m grid cells in a woodlot in Superior
Township, Michigan. Point size is proportional to tree DBH (cm). Living
blue ash trees ranged from 2.5 - 46.5 cm in DBH. Dead blue ash trees
ranged from 2.5 - 31.0 cm in DBH. There were no living white ash trees.
Dead white ash trees ranged from 2.5 - 43.5 cm in DBH. Gray quadrants
represent areas where no Fraxinus spp. trees were present. ……………………..25
Figure 2.1: Percentage of biomass allocated to roots, leaves and aboveground woody material
(shoots and bole) in blue ash, white ash and Manchurian ash trees (A) and ash trees
treated with paclobutrazol, fertilizer or left as untreated controls (B).……………66
Figure 2.2: Relationship between root biomass (kg) and tree caliper (mm) of blue ash, white ash
and Manchurian ash trees harvested in October 2012 from a plantation in Okemos,
Michigan. ………………………………………………………………………….67
Figure 2.3: Biomass vector analysis of nitrogen for blue ash (diamonds), white ash (triangles)
and Manchurian ash (squares) trees treated with paclobtrazol (black, PB) or
fertilizer (gray, F). Dry weight is the dry weight of all leaves, twigs, boles and
roots………………………………………………………………………………68
Figure 2.4: Biomass vector analysis of phosphorous for blue ash (diamonds), white ash
(triangles) and Manchurian ash (squares) trees treated with paclobtrazol (black,
PB) or fertilizer (gray, F)……..……………………………………………...…69
2

Figure 3.1: Mean (±SE) leaf area (cm ) consumed per beetle day by Agrilus planipennis caged
with (A) excised leaves from black, blue, green, white or Manchurian ash trees in
2010 or (B) excised or intact leaves from blue, white or Manchurian ash in 2011.
Letters indicate differences among species and leaf types (α=0.05).……………102
Figure 3.2: Mean (±SE) number of surviving Agrilus planipennis adults caged with (A) black,
blue, green, white or Manchurian ash leaves in 2010 or (B) blue, white or
xi

Manchurian ash in 2011. An “*” denotes significant differences among species on
that day (α=0.05). ………………………………………………………………..103
2

Figure 3.3: Mean (±SE) number of Agrilus planipennis larval galleries per m of debarked
phloem on black, blue, green, white or Manchurian ash trees harvested in 2011.
Letters indicate differences among species (α=0.05)……………………....……104
Figure 4.1: A cross section of a Fraxinus pennsylvanica tree injected in 2008 + 2009 with a low
rate (0.1g a.i. per 2.54 cm at diameter at breast height) of emamectin benzoate
®
®
(TREE-äge ) applied with a QUIK-jet system. The cross section was cut 10.2 cm
above 2008 injection sites. Discoloration from 2008 and 2009 injections, and
injection wounds from 2009 are visible. Xylem is present around wounds and
discolored tissues. Cracking in the cross secti+on was caused by drying, not the
injection process…………………………………………………………………127
Figure 4.2: Transverse cross section of Fraxinus quadrangulata earlywood xylem tissue dyed
with safranin orange. The image was obtained with a confocal laser scanning
microscope equipped with helium-neon (λ=543 nm) and argon (λ=488 nm) lasers
and a 10 dry objective for magnification. The three-dimensional image consists
of 19 slices obtained 11.5 µm apart.……………………………………………..128
Figure 4.3: Transverse cross section of Fraxinus pennsylvanica earlywood xylem tissue. The
image was obtained with a confocal laser scanning microscope equipped with
helium-neon (λ=543 nm) and argon (λ=488 nm) lasers and a 10 dry objective for
magnification. The three-dimensional image consists of 15 slices obtained 5.0 µm
apart…………………………………………………………………………….…129
Figure 4.4: Radial cross section of a Fraxinus quadrangulata xylem element and surrounding
cells from a tree injected with 15 ml of safranin orange. The image was obtained
using a confocal laser scanning microscope equipped with helium-neon (λ=543 nm)
and argon (λ=488 nm) lasers, a 20 dry objective and 1.8 zoom were used for
magnification. The three-dimensional image consists of 87 slices obtained 1.0 µm
apart……………………………………………………………………………….130
Figure 4.5: Radial cross section of a Fraxinus pennsylvanica xylem element and surrounding
cells of a tree injected with 15ml of safranin orange. The image was obtained using
a confocal laser scanning microscope equipped with helium-neon (λ=543 nm) and
argon (λ=488 nm) lasers, a 20 dry objective and 1.8 zoom were used for
magnification. The three-dimensional image consists of 48 slices obtained 1.0 µm
apart……………………………………………………………………………….131
Figure 4.6: Transverse cross section of Fraxinus pennsylvanica discolored by trunk injected
emamectin benzoate. The image was obtained using a confocal laser scanning
microscope equipped with helium-neon (λ=543 nm) and argon (λ=488 nm) lasers.

xii

The three-dimensional image consists of eight slices obtained 14.5 µm apart. A
10 dry objective was used for magnification………….……………………….132
Figure 4.7: Radial cross section of a Fraxinus quadrangulata xylem element and surrounding
cells from a tree injected with 15ml of safranin orange. The image was obtained
using a confocal laser scanning microscope equipped with helium-neon (λ=543 nm)
and argon (λ=488 nm) lasers. The three-dimensional image consists of 54 slices
obtained 2.0 µm apart. A 20 dry objective was used for magnification.....……133
Figure 4.8: Radial cross section of a Fraxinus pennsylvanica xylem element and surrounding
cells from a tree injected with 15ml of safranin orange. The image was obtained
using a confocal laser scanning microscope equipped with helium-neon (λ=543 nm)
and argon (λ=488 nm) lasers. The three-dimensional image consists of 39 slices
obtained 1.0 µm apart. A 20 dry objective was used for magnification.. ..…….134
Figure 5.1: Transverse samples of blue ash (B1, B2), green ash (G1, G2) and Manchurian ash
(M1, M2) phloem. Samples were illuminated with a helium-neon laser (λ= 543 nm)
to show basic phloem structure (B1, G1, M1) or a blue Diode laser (λ=405) to detect
phenolic compound autofluorescence (B2, G2, M2). Representative images were
obtained with a confocal laser scanning microscope equipped with a 10 dry
objective. Three dimensional images of blue ash, green ash or Manchurian ash
consist of 79, 24 or 34 slices, respectively. Slices were obtained 11.5 µm apart. For
interpretation of the references to color in this and all other figures, the reader is
referred to the electronic version of this dissertation…………………………….…150
Figure 5.2: Transverse cross sections of (A) blue ash, (B) green ash and (C) Manchurian ash
phloem tissues. Tissues were incubated with Amplex® Red peroxidase assay and
illuminated with argon (λ=488) and helium neon (λ= 543 nm) lasers.
Representative images were obtained with a confocal laser scanning microscope
equipped with a 10 dry objective. The three dimensional images consist of 24, 11
or 12 slices, respectively, obtained 5.0 µm apart. Fluorescence profiles correspond
to tissues located under corresponding yellow horizontal or vertical lines. Lines
were placed in areas where fluorescence was strongest………………………….151
Figure 5.3: Transverse cross sections of (A) blue ash, (B) green ash and (C) Manchurian ash
phloem tissues. Tissues were dyed with safranin orange (1%) and illuminated with
argon (λ=488) and helium neon (λ= 543 nm) lasers. The three dimensional images
consisted of 61, 47 and 56 slices, respectively, taken 5.0 µm apart. Representative
images were obtained with a confocal laser scanning microscope equipped with a
10 dry objective for magnification. Tissues directly above the orange arc or the
orange arrows indicate sclerenchymatous tissues………………………...………152

xiii

INTRODUCTION
Agrilus planipennis Fairmaire (Coleoptera: Buprestidae), a secondary pest of stressed or
declining ash (Fraxinus) trees in its native Asia, has become the most destructive forest insect to
ever invade North America (Aukema et al. 2011; Kovacs et al. 2011). To date, tens of millions
of Fraxinus trees have been killed in 19 states and two Canadian provinces (Pugh et al. 2011;
EAB.info 2013).
North American ash species are generally less resistant to A. planipennis than Asian ash
species, which may be attributable to the lack of a co-evolutionary history (Bryant et al. 1994;
Gandhi and Herms 2010a; Nielsen et al. 2011) and the scarcity of native phloem-feeding
congeners (Drooz 1985). In large, forested tracts of parkland in southeastern Michigan, the
origin of A. planipennis in North America (Siegert et al. 2009), mortality rates of North
American Fraxinus pennsylvanica Marsh., Fraxinus americana L., and Fraxinus nigra Marsh.
trees exceeded 99% (Herms et al. 2009; Gandhi and Herms 2010b; Knight et al. 2010). In 2009,
We revisited two woodlots where Fraxinus quadrangulata Michx. and F. americana trees were
originally sampled in 2004–2005 by Anulewicz et al. (2007). We observed large, apparently
healthy specimens of F. quadrangulata in both areas, which was surprising given the nearly
complete mortality of native ash trees in these counties and across southeastern Michigan. In
Chapter One, we assess tree survival by diameter class of F. quadrangulata and F. americana in
the two naturally regenerated woodlots.
In the Midwestern United States, Fraxinus trees can comprise 20 to 50% of the urban
forest canopy (MacFarlane and Meyer 2003; Poland and McCullough 2006). Like other
landscape trees, urban ash trees are often planted in locations with limited space and may
experience varying degrees and types of environmental stress (Chaney 2005a). Fertilizer and

1

paclobutrazol (PB), a gibberellin inhibitor, may be applied to enhance tree vigor. In Chapter
Two, I assess effects of fertilizer and PB on the physiology and growth of four North American
(F. americana, F. nigra, F. pennsylvanica and F. quadrangulata) and one Asian (Fraxinus
mandshurica Rupr.) ash species growing in a plantation established in 2006.
Ash species are evolutionarily diverse (Wallender 2008) and have varying levels of
inherent resistance to A. planipennis (Rebek et al. 2008; Anulewicz et al. 2007; Tanis and
McCullough 2012). Fraxinus nigra and F. pennsylvanica are highly preferred by A. planipennis
(Cappaert et al. 2005, Limback 2010) while F. americana is considered moderately preferred
(Anulewicz et al. 2007). To date, F. quadrangulata appears to be the least preferred North
American species (Anulewicz et al. 2007, 2008; Tanis and McCullough 2012). In Asia, A.
planipennis functions as a secondary pest, typically attacking stressed ash trees (Liu et al. 2003).
In North America, A. planipennis colonizes both healthy and stressed Fraxinus trees (Poland and
McCullough 2006), but preferentially attacks stressed trees (McCullough et al. 2009a, 2009b).
Many tree care companies promote plant growth regulator treatments to reduce tree growth and
ameliorate stress (Chaney 2005a, 2005b). In Chapter Three, I assess differences in A.
planipennis host preference and suitability of the four North and the Asian ash species and
among trees treated with PB, fertilizer or left as untreated controls in the controlled plantation.
Systemic insecticides applied as trunk injections are often used to protect landscape ash
trees from A. planipennis, but wounds and injury are a concern particularly for macro-injection
methods. TREE-äge, a product with emamectin benzoate as the active ingredient, is increasing
used because it provides ash trees with multiple years of protection against A. planipennis
(McCullough et al. 2011). Trunk injected insecticides can also cause internal tissue discoloration
(Mota-Sanchez et al. 2009; Tanis et al. 2012) which may be indicative of tissue wounding (Smith

2

and Lewis 2005). In Chapter Four, I evaluate possible injury and tissue discoloration associated
with emamectin benzoate injection sites on F. pennsylvanica and F. americana trees injected in
2008 only or in 2008 and again in 2009.
Interspecific variability in xylem architecture may affect the efficiency of trunk injections
(Mota-Sanchez et al. 2009; Tanis et al. 2012). Uptake of trunk injected insecticides can be
slowed by friction as products pass through the xylem lumen or through the bordered pits. In a
related study in Chapter Four, I examined cross sectional area of earlywood vessels and bordered
pit densities in F. quadrangulata and F. pennsylvanica with confocal laser scanning microscopy
(CLSM) to determine if xylem anatomy could influence trunk injection success.
Interspecific variability in ash resistance to A. planipennis has sparked interest in ash
physiology and defensive chemistry. Fraxinus quadrangulata does not share an evolutionary
history with A. planipennis, yet it has resistance similar to F. mandshurica, a species that coevolved with EAB in China. In contrast, F. pennsylvanica is highly vulnerable to A. planipennis
infestation. Phenolic compounds and reactive oxygen species (ROS) are often produced by
plants in response to injury, stress or insect damage. Agrilus planipennis larvae feed on the
phloem and cambium tissues of Fraxinus trees (Cappaert et al. 2005) and could be affected by
interspecific differences in phloem defensive chemical production or anatomy. In Chapter Five,
I narrow the scope of my host preference research down to the tissue level. I use CLSM to
assess interspecific differences in phenolic compound autofluorescence, ROS production and
phloem architecture of F. pennsylvanica, F. quadrangulata and F. mandshurica.

3

CHAPTER ONE
Differential persistence of blue ash and white ash following emerald ash borer invasion
Tanis, S.R. and D.G. McCullough. 2012. Differential persistence of blue ash and white ash
following emerald ash borer invasion. Canadian Journal of Forest Research. 42(8):
1542-1550.
ABSTRACT
Catastrophic mortality of North American ash (Fraxinus spp.) caused by Agrilus planipennis
Fairmaire has been attributed to the lack of coevolved resistance between native ash species and
this Asian invader. Although A. planipennis host preference or tree resistance can vary, all North
American ash species are presumably highly vulnerable to A. planipennis. WE inventoried live
and dead blue ash (Fraxinus quadrangulata Michx.) and white ash (Fraxinus americana L.) in
two southeastern Michigan woodlots several years after the A. planipennis invasion to assess
their survival. Agrilus planipennis populations in this area peaked in approximately 2005, and
the region is now characterized by nearly complete ash mortality. At the Plymouth site, 71% of
the original 380 blue ash were alive, whereas only 29 saplings of the original 187 white ash were
alive. At the Superior Township site, 63% of the original 210 blue ash were living, whereas all
125 white ash were dead. More than 80% of the blue ash had evidence of previous A.
planipennis colonization, but 87% appeared healthy in 2011. Tree diameter did not consistently
affect survival, and live and dead trees of both species were distributed across sites, indicating
that differential survival was not attributable to localized conditions.

4

INTRODUCTION
Emerald ash borer, Agrilus planipennis Fairmaire (Coleoptera: Buprestidae), a phloemboring beetle native to Asia (Chinese Academy of Science, Institute of Zoology 1986; Yu 1992),
was first identified in North America in southeastern Michigan, USA, and southern Ontario,
Canada, in 2002 (Cappaert et al. 2005). Tens of millions of ash (Fraxinus spp.) trees in forested
and urban settings have been killed by A. planipennis in Michigan alone, and infestations have
been found in 14 additional states and in Quebec, Canada (Emerald Ash Borer Information
Website, www.emeraldashborer.info, accessed 6 March 2012). Recent analyses showed that A.
planipennis is the most costly forest insect to invade the US, with annual economic impacts in
excess of US $900 billion (Aukema et al. 2011; Kovacs et al. 2011).
Ash trees are injured by A. planipennis larvae feeding on phloem and cambium in
serpentine galleries (Cappaert et al. 2005). On most trees, A. planipennis initially colonizes the
upper canopy (Cappaert et al. 2005; McCullough and Siegert 2007a). As A. planipennis
densities build, larval feeding eventually leads to mortality of the entire tree. External evidence
of A. planipennis infestation includes distinctive D shaped holes left by emerging A. planipennis
beetles, larger holes left by woodpeckers preying on late instar larvae, cracks in the outer bark
above larval galleries, epicormic shoots on large branches or the trunk, and progressive canopy
dieback (Cappaert et al. 2005; Poland and McCullough 2006; McCullough et al. 2009a).
North American ash species are generally less resistant to A. planipennis than Asian ash
species, which may be attributable to the lack of a co-evolutionary history (Bryant et al. 1994;
Lieutier 2008; Gandhi and Herms 2010a; Nielsen et al. 2011) and the scarcity of native phloemfeeding congeners (Drooz 1985). In its native range, A. planipennis is considered a secondary
pest and typically colonizes only declining or stressed ash trees (Liu et al. 2007; Wang et al.
5

2010; Cipollini et al. 2011). In North America, A. planipennis preferentially colonizes stressed
ash (McCullough et al. 2009a, 2009b) but will readily colonize and kill healthy ash trees
(Cappaert et al. 2005; Poland and McCullough 2006). In large, forested tracts of parkland in
southeastern Michigan, mortality rates of North American green ash (Fraxinus pennsylvanica
Marsh.), white ash (Fraxinus americana L.), and black ash (Fraxinus nigra Marsh.) trees
exceeded 99% by 2009 (Herms et al. 2009; Gandhi and Herms 2010b; Knight et al. 2010).
To date, populations of A. planipennis have encountered five native ash species in eastern
North America, and although all appear to be suitable hosts, interspecific variation in A.
planipennis host preference or ash tree resistance has been observed in field settings (Cappaert et
al. 2005; Anulewicz et al. 2007). Preferential colonization of some North American species was
first noted by Anulewicz et al. (2007), who quantified A. planipennis exit holes and canopy
condition annually in sites in southeastern Michigan from 2003 to 2005. They found that when
green ash and white ash trees co-occurred, green ash trees were preferentially colonized, and as
they succumbed, A. planipennis densities increased on white ash trees.
In 2004 and 2005, Anulewicz et al. (2007) compared A. planipennis densities on white
ash and blue ash (Fraxinus quadrangulata Michx.) trees in two southeastern Michigan woodlots
in which both species were relatively abundant. They reported that average densities of A.
planipennis exit holes and woodpecker attacks were six fold higher on white ash trees than on
blue ash trees. Some white ash trees in both woodlots had been killed by A. planipennis before
sampling began in 2004 and others were declining in 2005 (Anulewicz et al. 2007). They noted
that 57% of the blue ash trees sampled in 2005 had A. planipennis exit holes, bark cracks, or
holes left by woodpeckers preying on larvae (A.C. Anulewicz and D.G. McCullough,
unpublished data).

6

Other studies have similarly suggested that blue ash may be less preferred by or less
suitable for A. planipennis than other North American ash species. In a laboratory trial with
clipped leaves, adult A. planipennis beetles appeared to become more discriminatory over the
course of 48 h and ultimately spent less time and consumed less blue ash and Manchurian ash
foliage compared with four other ash species (Pureswaran and Poland 2009). Similarly, when
adult A. planipennis were caged with intact leaves on live trees for 14 days, beetle survival was
higher on white ash trees (67%) than on blue ash trees (42%) (Tanis and McCullough 2012).
In 2009, we revisited the two woodlots where blue ash and white ash trees were
originally sampled in 2004–2005 by Anulewicz et al. (2007). We observed large, apparently
healthy specimens of blue ash in both areas, which was surprising given the nearly complete
mortality of native ash in these counties and across southeastern Michigan (Herms et al. 2009;
Gandhi and Herms 2010b; Knight et al. 2010). In 2010–2011, we returned to the two woodlots
to systematically quantify and compare survival and canopy condition of blue ash and white ash
trees. We assessed size, distribution, and basal area of live and dead ash trees of both species in
each woodlot to evaluate the potential influence of site-related variables on tree survival or
condition.

7

METHODS
Study sites
Ash trees were inventoried in a 6.8 ha woodlot in Plymouth Township, Wayne Co.,
Michigan, and a 4.8 ha woodlot in Superior Township, Washtenaw Co., Michigan. Soils at both
sites are predominantly silty clay loam with 2%–6% slopes (U.S. Department of Agriculture
2012b). Understory vegetation in the two sites included herbaceous ephemerals (Trillium
grandiflorum (Michx.) Salisb., Erythronium americanum Ker, and Podophyllum peltatum L.)
and occasional areas of shrubs (Asimina triloba (L.) Dunal, Crataegus spp., and Rosa multiflora
Thunb.).
Ash inventory
In September 2009, we identified and flagged all Fraxinus spp. trees within each woodlot
with colored tape to indicate whether the trees were alive or dead. Live trees had at least one
living shoot or epicormic sprout on the trunk or branches. In October 2009, following leaf drop,
12 contiguous 50 × 50 m cells were established in each woodlot to facilitate overstory
inventories. Plots were excluded from low lying, swampy areas where no trees occurred. We
recorded species and GPS coordinates (eTrex Legend, Garmin Ltd., Olathe, Kansas, USA) and
measured diameter at breast height (DBH) (1.3 m aboveground) for all living and dead (standing
and down) ash trees ≥2.5 cm DBH. A total of 17 dead, fallen ash trees (three at Plymouth; 14 at
Superior Township) were too decayed for species determination and were excluded from maps
and analyses. An estimated 25 dead white ash trees had been removed and used for firewood by
the owners of the Superior Township property in 2007–2009 and could not be inventoried. The
property owner at the Plymouth site indicated that no trees (regardless of species) had been
removed for firewood. Dead trees were examined and bark was removed when necessary to

8

assess larval galleries and confirm that death was caused by A. planipennis. Distribution of live
and dead blue ash and white ash trees was mapped using GPS coordinates (R statistical software;
R Foundation for Statistical Computing, Vienna, Austria).
In May 2010, following leaf expansion, we returned to evaluate canopy condition of live
ash trees. Canopy dieback was visually estimated (10% increments) for live ash in both sites
following protocols of Zarnoch et al. (2004), where dieback represents the percentage of the tree
crown comprised of dead branches, excluding natural branch pruning. Dieback ratings were
completed by one person to ensure consistency. Trees were grouped by canopy dieback classes
for presentation. Trees with <30% dieback were considered to be relatively healthy, those with
31%–60% dieback exhibited “moderate” canopy dieback, and trees with >60% canopy dieback
were considered to be severely declining and unlikely to recover (Herms et al. 2009).
Variable radius plots and relative importance values
In July 2011, we established a variable radius plot (10 BAF wedge prism) in the center of
each 50 × 50 m cell at each site. We recorded species and measured DBH of live and dead
(standing and down) trees ≥2.5 cm DBH and visually estimated canopy dieback for live trees
within each plot. Data from these plots were consolidated to estimate basal area by genus
(including ash and non-ash trees) and total basal area for each site. Relative importance values
were calculated for the five most abundant genera at each site. Relative importance values
represent the contribution of a given genus to the overstory and were calculated as the sum of the
relative frequency (number of plots containing the genus as a percentage of the total number of
occurrences for all genera within all plots) + relative density (number of individuals of the genus
as a percentage of the total number of individuals of all genera) + relative dominance (total basal
area of the genus as a percentage of the total basal area for all genera) (Kent and Coker 1992).

9

Statistical analysis
To simplify presentation, ash trees were grouped by DBH into four classes corresponding
to young recruits (≤10 cm), small trees (10.1–20 cm), pole-sized trees (20.1–30 cm), and
merchantable trees (>30.1 cm). We used a binomial t test to determine if DBH differed between
ash species and between living and dead blue ash trees in each site. Linear relationships between
DBH and canopy dieback were evaluated for live blue ash in both sites. Data were analyzed
using SAS statistical software (SAS Institute Inc., Cary, North Carolina) at α of P < 0.05.

10

RESULTS
Blue ash and white ash trees were originally abundant in the two woodlots, but survival
rates of the two species following invasion by A. planipennis differed profoundly. We
inventoried a total of 567 and 335 Fraxinus spp. trees in the Plymouth and Superior Township
sites, respectively, in the 50 × 50 m cells. At the Plymouth site, 270 of the 380 blue ash trees
(63%) survived, including at least 50% of the blue ash trees in all diameter classes (Fig. 1.1a). In
contrast, only 29 of 187 white ash trees (16%) at the Plymouth site survived, and all were <11
cm DBH (Fig. 1.1a). At the Superior Township site, 132 of the 210 blue ash trees (71%) were
alive, including at least 50% of trees in each diameter class, while all 125 white ash trees were
dead (Fig. 1.1b). Evidence of previous A. planipennis infestation was observed on 83% and 87%
of live blue ash trees at the Plymouth and Superior Township sites, respectively, and 10 of the 29
live white ash at the Plymouth site.
Both blue ash and white ash trees were originally distributed throughout each woodlot.
In Plymouth, blue ash trees originally occurred in all 12 grid cells and live blue ash were present
in 11 of the cells (Fig. 1.2). Dead white ash occurred in every cell, and the 29 live white ash
trees were scattered across 10 of the 12 grid cells (Fig. 1.2). At the Superior Township site, live
and dead blue ash and dead white ash trees were present in all 12 grid cells (Fig. 1.3).
We calculated total Fraxinus spp. basal area using the ash inventory data from the grid
2

–1

–1

2

cells. At the Plymouth site, 62% (2.5 m • ha ) of the total blue ash basal area (4.1 m • ha )
2

–1

was alive. White ash trees in the Plymouth site originally accounted for 70% (9.7 m • ha ) of
2

–1

the total (living and dead) Fraxinus spp. basal area (13.7 m • ha ), but the 29 live white ash
cumulatively accounted for <1% of the Fraxinus spp. basal area. At the Superior Township site,

11

2

–1

–1

2

81% (3.5 m • ha ) of the total blue ash basal area (4.3 m • ha ) was alive. Before the A.
2

–1

planipennis invasion, white ash comprised at least 44% (3.4 m • ha ) of the total (living and
2

–1

dead) Fraxinus spp. basal area (7.63 m • ha ) in this site, but all of these trees were killed.
Survival of the ash species at the two sites was not consistently related to tree DBH. At
the Plymouth site, the overall mean (± standard error (SE)) DBH of all blue ash trees (live and
dead), which averaged 8.9 ± 0.38 cm, was smaller than the overall mean of all white ash trees,
which averaged 20.7 ± 1.13 cm (t[1,564] = 12.32; P < 0.001). Dead blue ash trees, which
averaged 11.0 ± 0.70 cm DBH, were larger than live blue ash, which averaged 8.1 ± 0.45 cm
DBH (t[1,379] = 3.58; P < 0.001). Most of the live blue ash (94%) and dead blue ash (92%) were
≤20 cm DBH (Fig. 1.1a). Only 2% and 4% of the live and dead blue ash trees, respectively,
were >30 cm DBH (Fig. 1.1a). The largest living blue ash tree was 45.0 cm DBH, while the
largest dead blue ash tree was 50.0 cm DBH. Average DBH of dead white ash at the Plymouth
site was 23.7 ± 1.17 cm, and dead white ash trees were represented in all DBH classes (Fig.
1.1a). All 29 live white ash trees at the Plymouth site were <11 cm (Fig. 1.1a), with an average
DBH of 4.7 ± 0.35 cm. Approximately 50% of the dead white ash were relatively small (<20 cm
DBH), but 24% were at least 30 cm DBH and the largest dead white ash tree was 96.0 cm DBH.
At the Superior Township site, DBH of the dead white ash trees averaged (±SE) 16.0 ±
0.89 cm, which was larger than the overall mean of all blue ash trees (live and dead), which
averaged 13.4 ± 0.06 cm (t[1,333] = 2.36; P = 0.019). Average DBH of live blue ash was 15.2 ±
0.88 cm, which was larger than the DBH of dead blue ash, which averaged 10.2 ± 0.57 cm
(t[1,208] = 4.10; P < 0.001). Most of the live blue ash (67%) and nearly all of the dead blue ash

12

(98%) trees were <20 cm DBH (Fig. 1.1b). Only 7% of the live blue ash trees and a single dead
blue ash tree were >30 cm DBH. The largest living blue ash tree was 46.5 cm DBH, while the
largest dead blue ash tree was 31.0 cm DBH. Overall, 69% of the dead white ash trees were <20
cm DBH. Only 12% of the dead white ash trees were >30.1 cm DBH (Fig. 1.1b), and the largest
dead white ash was 43.5 cm DBH.
The two woodlots included a mix of northern hardwood species, and until the A.
planipennis invasion, both blue ash and white ash comprised a substantial portion of the
overstory. We measured 34 Fraxinus spp. and 114 non-Fraxinus spp. trees at the Plymouth site
and 22 Fraxinus spp. and 129 non-Fraxinus spp. trees at the Superior Township site (Table 1.1).
2

–1

Total basal area estimated from variable radius plots was 24.2 and 16.4 m • ha

in the

Plymouth and Superior Township sites, respectively. Overall, Fraxinus spp. ranked as the
second and fourth highest contributing genus, accounting for 9% and 7% of the total basal area at
the Plymouth and Superior Township sites, respectively (Table 1.1). Overstory composition at
the sites was similar and dominated by Acer spp., Quercus spp., Tilia americana L., and Carya
species. We also encountered a few stems of Ulmus americana L., Juglans nigra L., Ostrya
virginiana (Mill.) K. Koch, and Prunus serotina Ehrh. at both sites. Two standing non-ash trees
were dead (one U. americana and one O. virginiana) at the Superior Township site. No standing
non-ash trees were dead in the plots at the Plymouth site.
Previous A. planipennis larval feeding resulted in some level of canopy dieback on most
of the live blue ash trees at the two sites, particularly those trees that were large enough (e.g.,
>10.1 cm DBH) to have been colonized when A. planipennis densities were peaking. At the
Plymouth site, canopy dieback was minimal (<10%) on 131 of the 204 (64%) small blue ash
trees (<10 cm DBH), while only 8% of these small trees had severe dieback (>60%) (Table 1.2).
13

Dieback on these small trees averaged (± SE) 23% ± 1.7%. Of the 50 blue ash trees that were
10.1–20 cm DBH, 16% had negligible dieback and 20% had severe dieback. For the 6% of the
blue ash trees that were >20.1 cm DBH, canopy dieback averaged 43.8% ± 3.3%. Only two of
these 16 trees had severe dieback, whereas dieback was low to moderate on the other 14 trees
(Table 1.2). For live blue ash trees at Plymouth, DBH was linearly related to canopy dieback (Y
= 38.108 + 0.169x) (F[1,268] = 39.15; P < 0.001), although the regression explained relatively
2

little variation (R = 0.127). Ten of the 29 small live white ash trees in the Plymouth site had
evidence of previous A. planipennis colonization. Canopy dieback was minimal (<10%),
however, on 24 of the trees, and only a single tree had severe dieback.
At the Superior Township site, canopy dieback was minimal (<10%) on 18 of the 89
(20%) blue ash trees that were ≤20 cm DBH and severe (>60%) on 21 of these trees (24%)
(Table 1.2). Canopy dieback for blue ash trees ≤20 cm DBH averaged 44% ± 3.1%. Of the 34
blue ash trees in the 20.1– 30 cm DBH class, most (83%) had low to moderate dieback (Table
1.2). Only eight blue ash trees at Superior Township were >30 cm DBH. Three of these trees
had minimal dieback, five trees had low to moderate dieback, and one tree (41.2 cm DBH) had
severe dieback (Table 1.2). Canopy dieback of blue ash trees at the Superior Township site was
not related to DBH (P = 0.46).
Canopy dieback on the non-Fraxinus trees assessed in the variable radius plots was
negligible. None of the non-ash trees had more than 10% canopy dieback at either site.

14

DISCUSSION
The long-term persistence of blue ash in southeastern Michigan, where high densities of
A. planipennis have resulted in nearly complete mortality of other native ash species, is one of
very few bright spots in the saga of the A. planipennis invasion in North America. Results from
the original 2004–2005 survey of these sites led Anulewicz et al. (2007) to conclude that
although white ash trees would succumb sooner, A. planipennis densities would continue to build
on blue ash trees until the trees succumbed, eventually leading to the catastrophic mortality
observed in forested areas across southeastern Michigan (Herms et al. 2009; Gandhi and Herms
2010b; Knight et al. 2010). We were surprised, therefore, to find that at least 60% of the blue
ash trees at both sites were not only alive in 2011, but most appeared to be healthy. In contrast,
every white ash tree in the Superior Township site was killed and the few surviving white ash
trees in the Plymouth site were likely too small to have been colonized during the peak of the A.
planipennis invasion. All of the dead ash trees in the two sites had been fully colonized by A.
planipennis; larval galleries literally covered the trunks and branches.
In our experience, populations of A. planipennis are typically established in a site for at
least three to four years before severe decline or mortality becomes apparent (Siegert et al. 2009;
Pugh et al. 2011). Moreover, in newly infested sites with low A. planipennis densities, a high
proportion of larvae requires two years for development (Siegert et al. 2010; Tluczek et al.
2011), which slows the population growth rate (Mercader et al. 2011a, 2011b). Evidence
strongly suggests, therefore, that A. planipennis populations were established in our sites at least
two years before this invasive pest was first identified in North America in 2002. The
unexpectedly high survival of blue ash trees 10 or more years after invasion, therefore, indicates

15

that this species is more resistant than any other North American ash species that A. planipennis
populations have encountered to date.
We evaluated several factors to determine whether exposure to sun, local growing
conditions, or other site-related variables were responsible for the differential mortality rates of
the two ash species. Soil at the two sites consisted primarily of silty clay loam, which is suitable
for both Fraxinus species (Burns and Honkala 1990; Farrer 1995). Previous observations have
shown that adult A. planipennis activity, including oviposition, is consistently higher on trees
exposed to full or nearly full sun than on shaded trees (Yu 1992; McCullough et al. 2009a,
2009b; Poland et al. 2011). Stress can alter volatiles emitted by ash trees, increasing their
attraction to A. planipennis (Rodriguez-Saona et al. 2006; McCullough et al. 2009a, 2009b; Chen
and Poland 2009). If white ash trees experienced poorer growing conditions (e.g., flooded,
compacted soil, etc.) than blue ash trees, for example, survival rates could reflect differential
attraction of A. planipennis to stressed versus healthier trees. Both blue ash and white ash trees,
however, were relatively abundant in the two woodlots, and exposure and environmental
conditions were similar. Live and dead ash trees of both species were distributed throughout
both sites, and there were no apparent spatial patterns in ash mortality for either species.
Furthermore, non-ash trees in both woodlots were healthy; only two dead non-ash trees were
encountered in the plots in either site, and canopy decline was minimal on the non-ash overstory
trees.
Similarly, there was no evidence that tree size consistently influenced mortality rates.
Average DBH of live blue ash was slightly (3 cm) smaller than that of dead blue ash trees at the
Plymouth site, but live blue ash were somewhat larger (5 cm), on average, than dead blue ash at
the Superior Township site. Dead white ash were represented in all diameter classes at both

16

sites. Although we cannot determine whether the rate of white ash mortality differed among size
classes, ultimately, all white ash trees were killed, with the exception of the 29 saplings at the
Plymouth site. The unexpected persistence of blue ash trees in these sites, therefore, does not
appear to be merely a function of differences between ash species in growing conditions or size.
It is especially notable that at least 60%–70% of the blue ash trees in both woodlots
survived despite A. planipennis densities that were likely very high during the peak of the
invasion. Using methods of McCullough and Siegert (2007a), we estimated that 114,750 and
53,980 A. planipennis beetles could have been produced by the dead white ash trees at the
Plymouth and Superior Township sites, respectively, before they succumbed. An additional
24,680 and 15,050 A. planipennis adults could have emerged from the dead blue ash trees at
Plymouth and Superior Township sites, respectively. Given these numbers, it is not surprising
that nearly all of the surviving blue ash trees had some evidence of A. planipennis colonization.
For example, at least 80% of the live blue ash trees >10 cm DBH in both sites had more than
10% canopy dieback. Most live blue ash appeared to be healthy in 2011, however, and less than
13% of these trees had severe dieback.
Although naturally regenerated blue ash and white ash trees were present in both
woodlots, the two species differ considerably in terms of distribution, silvicultural traits, and
morphological features. For example, distribution of blue ash is centered in Kentucky,
Tennessee, and Missouri (USDA Hardiness Zones 4–7) (U.S. Department of Agriculture 2012a),
where it occurs most commonly in mixed stands on limestone outcroppings (Bryant et al. 1980;
Pallardy et al.1988; Prasad et al. 2007). Blue ash trees in the sites that we sampled, therefore,
were growing near the northern edge of their native range. In contrast, the distribution of white
ash extends across much of eastern North America, from Nova Scotia, Canada, south to northern

17

Florida (USDA hardiness Zones 3–9) (Burns and Honkala 1990). Whether blue ash will
demonstrate similarly strong resistance to A. planipennis in other areas of its native range,
particularly where it is relatively abundant, remains to be seen. Populations of A. planipennis
were identified in Missouri in 2008, Kentucky in 2009, and Tennessee in 2010 (Emerald Ash
Borer Information Website, www.emeraldashborer.info, accessed 6 March 2012), but to date,
infestation and mortality rates for specific ash species have not been reported from these areas.
Morphological differences between blue ash and white ash trees are also apparent. Blue ash and
white ash were widely separated in a phylogenetic tree constructed for Fraxinus species based on
ribosomal internal transcribed spacers and floral morphological characteristics (Wallander 2008).
White ash, a dioecious species, was placed in the Melloides section, along with eight other North
American species. Blue ash, along with F. dipetala and F. anomala, species native to the
southwestern US, were placed in the Dipetalae section and are characterized by monoecious
flowers and four distinctive cork ridges on woody shoots. Whether the other North American
Dipetalae ash species are also relatively resistant to A. planipennis remains unknown.
Identification of specific traits or mechanisms underlying blue ash resistance could play a
key role in the population dynamics of A. planipennis and survival of the genus Fraxinus in
North America. Adult A. planipennis females feed on ash leaves for at least 14 days before
oviposition begins and continue to leaf feed throughout their life span (Cappaert et al. 2005),
laying a few eggs between bouts of feeding. Therefore, selection of hosts for leaf feeding by
adult A. planipennis females likely has a strong effect on selection of hosts for oviposition
(Mercader et al. 2011a). To date, most studies of Fraxinus resistance to A. planipennis have
compared Asian ash species with North American green ash, white ash, and black ash (Eyles et
al. 2007; Chen et al. 2011; Rebek et al. 2008; Whitehill et al. 2011). Pureswaran and Poland

18

(2009) reported that adult A. planipennis leaf consumption was lower when beetles were caged
with blue ash and Manchurian ash leaves compared with leaves from green, white, or black ash
trees. However, this study used detached leaves, perhaps negating potential inducible defensive
responses. In a 14-day field trial, survival of A. planipennis beetles allowed to feed on intact
leaves of live trees was lower on blue ash than on white ash (Tanis and McCullough 2012).
More than 40% of the beetles that fed on blue ash, however, survived the maturation feeding
period, and females could presumably have begun laying eggs.
Although foliar traits or defenses may influence host selection by adult A. planipennis,
most (87%) living blue ash trees in the two woodlots exhibited some evidence of previous A.
planipennis colonization. It seems likely, therefore, that phloem-based compounds or other
factors must affect the suitability of ash species for A. planipennis larval development or
survival. When Anulewicz et al. (2007) surveyed these woodlots, they noted the presence of
callus tissue, i.e., wound periderm, associated with some A. planipennis larval galleries on blue
ash trees. Rapid production of wound periderm functions as a defense mechanism against
Agrilus anxius Gory larvae, a North American species that colonizes paper birch (Betula
papyrifera Marsh.) (Nielsen et al. 2011; Miller et al. 1991). We similarly observed wound
periderm on trees in these sites and on ash trees in other areas. In our experience, however, this
tissue seldom affects developing larvae and is rare on trees colonized by high densities of A.
planipennis larvae. Whether fewer eggs were laid on blue ash trees in the sites that we sampled
or whether most larvae failed to develop and emerge successfully from blue ash trees remains
unknown.
The persistence of blue ash through the A. planipennis invasion has implications for both
A. planipennis management and future research. Green ash and black ash trees, for example, are

19

more preferred and suitable hosts for A. planipennis than white ash (Cappaert et al. 2005; Chen
and Poland 2010; Limback 2010), while our results show that blue ash is less preferred or a less
suitable host for A. planipennis than white ash. Selecting green ash or black ash trees for traps or
use as girdled trap trees, therefore, is presumably more likely to facilitate detection of A.
planipennis than using white ash or blue ash (McCullough and Siegert 2007b). Resistance of
blue ash may also have implications for biological control efforts underway across much of the
A. planipennis infestation in the United States. Federal agencies have invested substantially in
the importation, rearing, and release of Asian parasitoids for biological control of A. planipennis
(Bauer et al. 2008; Duan et al. 2011, 2012). Native parasitoids, particularly Atanycolus spp.,
have also begun to parasitize A. planipennis larvae, sometimes accounting for high mortality in
localized areas (Cappaert and McCullough 2009; Duan et al. 2009, 2011; Taylor et al. 2012). In
areas with a substantial component of blue ash, the combination of relatively resistant hosts and
specialized parasitoids could perhaps maintain A. planipennis populations at substantially lower
densities than those typical of sites dominated by green, white, or black ash and may warrant
evaluation (Barclay 1987; Barclay and Li 1991; Suckling et al. 2012). Additional studies to
assess morphological features and constitutive or inducible defensesthat confer resistance in blue
ash by acting on A. planipennis adults or larvae could potentially be propagated or enhanced in
other ash species. Evaluating potential resistance of F. dipetala and F. anomala, the two North
American ash species most closely related to blue ash (Wallander 2008), would also be useful.

20

1

Table 1.1: Number of trees, mean (±SE) diameter at breast height (DBH), total basal area and Relative Importance Values for the
five most common tree genera recorded in variable radius plots in woodlots in Plymouth and Superior Township, Michigan.
Values for Fraxinus include dead and live trees.

Genus

Number of Trees

Quercus
Fraxinus
Acer
Carya
Tilia

40
34
26
17
11

Acer
Tilia
Quercus
Fraxinus
Carya

54
31
19
22
9

DBH (cm)

Basal
2
Area(m /ha)

Plymouth Site
51.2 ± 5.2
17.4 ± 5.3
34.6 ± 3.0
38.4 ± 3.8
37.2 ± 6.5
Superior Township Site
37.0 ± 2.1
37.1 ± 2.3
44.2 ±3.9
24.0 ± 1.9
30.1 ± 4.6

1

Relative Importance Value

11.1
2.2
2.9
2.3
1.6

94.3
53.5
40.3
35.1
26.4

6.8
3.7
3.3
1.1
0.8

1

59.5
40.7
33.3
24.5
16.0

Relative Importance Values reflect the contribution of a given genus to the overstory and were calculated as the sum of the relative
frequency + relative density + relative dominance following Kent and Coker (1992).

21

Table 1.2: Number of living blue ash and white ash trees by canopy dieback and diameter
classes in woodlots surveyed in Plymouth and Superior Township, Michigan.
1

Canopy Dieback Class
≤10%
11-30
31-60
>60
≤10%
11-30
31-60
>60
≤10%
11-30
31-60
>60

Ash Diameter Class (cm DBH )
≤10.0
10.1 - 20.0
20.1 - 30.0
> 30.1
Plymouth - Living Blue Ash
131
8
0
0
32
16
1
2
24
16
8
3
17
10
2
0
Plymouth - Living White Ash
23
1
0
0
4
0
0
0
0
0
0
0
1
0
0
0
Superior Township - Living Blue Ash
10
8
4
3
11
8
11
3
19
12
19
2
13
8
0
1

1

DBH refers to diameter at breast height, measured 1.3 m aboveground

22

Figure 1.1: Number of live and dead blue ash and white ash trees by diameter class in woodlots
in Plymouth (A) and Superior Township (B) Michigan.

23

50
meters

Figure 1.2: Distribution of live (A) and dead (B) blue ash (black points) and white ash (grey points) trees in 50 × 50 m grid cells in a
woodlot in Plymouth, Michigan. Point size is proportional to tree DBH (cm). Living blue ash trees ranged from 2.5 - 45.0
cm in DBH. Dead blue ash trees ranged from 2.5 - 50.0 cm in DBH. Living white ash trees ranged from 2.5 - 10.9 cm in
DBH. Dead white ash trees ranged from 2.5 - 96.0 cm in DBH. Gray quadrants represent areas where no Fraxinus spp.
trees were present.

24

50
meters

Figure 1.3: Distribution of live (A) and dead (B) blue ash (black points) and white ash (grey points) trees in 50 × 50 m grid cells in a
woodlot in Superior Township, Michigan. Point size is proportional to tree DBH (cm). Living blue ash trees ranged from
2.5 - 46.5 cm in DBH. Dead blue ash trees ranged from 2.5 - 31.0 cm in DBH. There were no living white ash trees.
Dead white ash trees ranged from 2.5 - 43.5 cm in DBH. Gray quadrants represent areas where no Fraxinus spp. trees
were present.

25

CHAPTER TWO
Effects of fertilizer and paclobutrazol on the physiology
and growth of five ash species
ABSTRACT
Ash (Fraxinus spp.) trees are widely planted in urban areas across the United States
where they experience varying degrees and types of stress. Fertilizer and paclobutrazol (PB), a
gibberellin inhibitor, may be applied to enhance tree vigor. Fertilization typically increases leaf
area and radial growth but reduces root: shoot ratios, while PB typically reduces leaf area and
radial growth but increases root: shoot ratios. In 2010, 105 ash trees in a common garden were
assigned to one of three treatments: fertilization (Harrell’s Pro blend with Micros, 19-5-10), PB
®

application (Shortstop , soil drench) or untreated control. Leaves from control or fertilized were
similar in size, but they were 19 and 25% larger than leaves from PB trees, respectively. Gas
exchange did not vary among species or treatments. Ratios of pre: post treatment radial growth
were similar between white ash (Fraxinus americana) and Manchurian ash (F. mandshurica)
trees, but were 39 and 33% higher, respectively, than blue ash (F. quadrangulata) tree ratios.
Ratios of pre: post treatment radial growth were similar between control and fertilized trees, but
were 33 and 43% higher, respectively, than PB tree ratios. Ratios of root: total biomass were 29
and 13% higher in blue ash trees than white ash and Manchurians ash trees, respectively
indicating that blue ash trees allocate more growth to their root systems than white ash or
Manchurian ash trees. Trees treated with PB had 9 or 10% higher ratios of root: total biomass
than control or fertilized trees, respectively. I found PB treatment shifts growth from above to
below ground resources, therefore applications could extend the period before trees outgrow
available space, enabling trees to persist in confined spaces.

26

INTRODUCTION
Urban trees are exposed to a variety of environmental stresses. To mitigate stress and
enhance vigor, urban trees may be fertilized or treated with paclobutrazol (PB), a growth
regulator. Nitrogen (N) fertilizer typically increases shoot growth and leaf production but may
reduce root: shoot ratios (Herms 2002, Struve 2002, Scharenbroch and Lloyd 2004, Hasselkus
and Schulte 2010). In contrast, PB, a gibberellin inhibitor, typically reduces shoot growth and
leaf area but may increase root: shoot ratios (Bai et al 2004, Chaney 2005a,b, Martinez-Trinidad
et al. 2011). Both treatments can potentially affect foliar nutrient levels, water potential and gas
exchange (Werner 1993, Remphrey and Davidson 1996, Premachandra et al. 1997, Arzani and
Roosta 2004, Scharenbroch and Lloyd 2004).
Since the early 20th century, benefits of fertilization for urban trees have been widely
disputed (Miller 1998, Herms 2002, Ferrini and Baietto 2006). Fertilizers have been
recommended to ensure that urban trees receive adequate nutrients, increase growth and vigor,
and maintain or enhance aesthetic qualities (Herms 2002, Scharenbroch and Lloyd 2004,
Hasselkus and Schulte 2010). Urban soils may lack nitrogen (N), the most common limiting
nutrient for temperate zone trees (Vitousek and Howarth 1991, Aber 1992, Harris 1992), because
humans continually alter the N cycle by removing leaf litter and displacing soils during
infrastructure development (Scharenbroch and Lloyd 2004). However, many urban substrates
contain levels of N and other nutrients adequate to support tree growth (Harris et al. 2008,
Watson 2010). In these situations, fertilization is not cost effective and nutrient runoff can be
harmful to surrounding environments (Watson 2010). High soil N levels can increase N
concentrations in trees which may affect insect performance. High N concentrations in plant
tissues (leaves, phloem, etc.) may make them more attractive to insect pests (e.g., aphids) (Kytö
1996). To determine if fertilization is necessary and reduce the potential negative effects of
27

over-fertilization, ideally, tree care professionals should develop species and site specific
fertilizer prescriptions based on tree species, time of year, climate, current tree nutrient status and
site conditions (Scharenbroch and Lloyd 2004, Ferrini and Baietto 2006, Harris et al. 2008).
Utility companies began using PB in the late 1970s (Breedlove et al. 1989, Davis and
Curry 1991, Redding et al. 1994) to reduce aboveground tree growth and the pruning necessary
to ensure lines remained clear of entangling branches (Chaney 2005b). Paclobutrazol has more
recently been adopted by the landscape industry to similarly reduce pruning and maintenance
costs, particularly for trees growing in confined spaces (e.g., next to buildings) (Chaney 2005a).
Application of PB can affect a variety of tree processes, including foliar nutrition (Rieger and
Scalabrelli 1990, Yelenosky et al. 1995, Huett et al. 1997, Navarro et al. 2009), gas exchange
and water use efficiency (Fletcher and Gilley 2000). In addition, PB reportedly affects tree
response to salt and drought stress (Wang and Steffens 1987, Abu El-Kashab et al. 1997, Chaney
2005a,b).
In the Midwestern United States, ash trees often comprise 20-50% of the urban canopy
(Schoon 1993, MacFarlane and Meyer 2003, Percival et al. 2006, Poland and McCullough 2006).
Abundance of ash in landscapes reflects both the desirable traits of the genus and the previous
invasion by Dutch elm disease. In the late 20th century, ash trees became exceedingly popular as
replacements for American elm (Ulmus americana L.) trees killed by Dutch elm disease
(MacFarlane and Meyer 2005, Poland and McCullough 2006). Ash tolerate compacted soils,
drought and soil salinity (Dobson 1991, MacFarlane and Meyer 2005, Percival et al. 2006) and
perform well in challenging urban locations (e.g., along streets and in parking lot islands and
sidewalk cubes) (Schoon 1993, MacFarlane and Meyer 2005, Percival et al. 2006, Poland and
McCullough 2006). Ash trees are also easy to transplant and require little maintenance (e.g.,

28

pruning, irrigation or fertilization) (Dirr 1998). Until the arrival of emerald ash borer (EAB)
(Agrilus planipennis Fairmaire) in the early 1990s, ash was also popular because it was
threatened by relatively few insect pests and is one of few genera not preferred by gypsy moth
(Lymantria dispar L.) (Liebold et al. 1995).
Several North American ash species are planted in urban areas. Green ash (Fraxinus
pennsylvanica Marsh.), which has the widest natural distribution of North American species, is
also the most prevalent ash in urban landscapes (MacFarlane and Meyer 2003). Cultivars of
white ash (Fraxinus americana L.), the second most common species (MacFarlane and Meyer
2003), are marketed for their distinct purple fall color and ability to bear snow and ice (uky.edu
2012). Black ash (Fraxinus nigra Marsh.) and blue ash (Fraxinus quadrangulata Michx.) are
also planted in some areas of the United States. In extreme northern climates or sites with poor
drainage, black ash and ‘Northern Treasure’, a hybrid of black ash and Manchurian ash, perform
well. Cultivars of blue ash grow slowly, tolerate a variety of soils and are considered relatively
drought tolerant (Dirr 1998). Exotic ash species are also used in urban landscapes, including
Manchurian ash (Fraxinus mandshurica Rupr.) which is native to Asia and Excelsior ash
(Fraxinus excelsior L.) which is native to Europe. Identification of EAB in 2002 dramatically
altered the future of urban ash trees in North America.
Native to Asia, EAB has killed tens of millions of ash trees in 19 states and two Canadian
provinces (eab.info Accessed 25 April 2013) and is the most destructive forest insect to invade
North America (Aukema et al. 2011, Kovacs et al. 2011). In Asia, A. planipennis functions as a
secondary pest, typically attacking stressed ash trees (Lyons et al. 2009). In North America, it
can colonize both healthy and stressed ash trees (Poland and McCullough 2006), but will
preferentially attack stressed trees (McCullough et al. 2009a, 2009b). Interest in ash tree

29

response to EAB and cultural amendments designed to ameliorate stress related to infestation has
subsequently stimulated research on ash tree physiology.
Several studies have assessed effects of fertilization on ash trees both in urban and forest
settings, but response is variable. Fertilization did not increase leaf area or stem growth of
established urban green ash trees in one study (Watson 2010), while Remphrey and Davidson
(1996) found it increased the height of containerized green ash trees at one site but decreased
growth at a second site. Fertilization of newly established F. excelsior did not affect shoot
growth but did increase net photosynthesis and foliar N concentration (Ferrini and Baietto 2006).
Only a few studies have examined effects of PB on ash and results vary, depending on
the ash species or growth parameters assessed. Watson (2004) reported that PB application
reduced twig growth and leaf area of established green ash trees but on white ash, PB had no
effect on lateral shoot growth (Sterrett and Tworkoski 1987, Bai et al. 2004) or suppressed only
the central leader (Bai et al. 2004).
Our objective was to evaluate effects of fertilization or PB applications on North
American black ash, blue ash, green ash and white ash and Asian Manchurian ash growing in a
common location and under otherwise similar conditions. These species were selected because
they are planted in urban landscapes, are evolutionarily diverse (Wallender 2008) and have
varying levels of inherent resistance to EAB (Cappaert et al. 2005, Anulewicz et al. 2007, Rebek
et al. 2008, Pureswaran and Poland 2009, Chen and Poland 2010, Limback 2010, Tanis and
McCullough 2012). I measured a range of growth, gas exchange, nutrient concentration and
biomass allocation variables to determine if species respond consistently or differently to either
fertilizer or PB application.

30

METHODS
Ash Plantation
On 26 April 2006, 225 ash trees were planted in a randomized complete block design at
the Michigan State University Tree Research Center (TRC) in Okemos, Ingham County,
Michigan. The plantation consisted of 45 trees each of five species: black ash, blue ash, green
ash, white ash and Manchurian ash. Trees arrived from nurseries as bare root liners (black and
Manchurian) or balled and burlapped (blue, green and white). Trees were planted 2.4 m (8 feet)
apart. Tree size varied among species at the time of planting: caliper diameter (15 cm
aboveground) of green ash trees ranged from 6.5 – 4.0 cm (mean (±SE) caliper =6.27 ± 0.06 cm)
and white ash trees ranged from 6.0 to 4.1 cm (mean caliper = 5.06 ± 0.06) (Poplar Farms
Nursery and Equipment, Waterman, IL, USA), ‘Fall Gold’ black ash trees ranged from 5.2 to 3.2
cm (mean caliper =4.29 ± 0.08 cm) (Bailey Nurseries, St. Paul, MN, USA), blue ash trees ranged
from 6.0 to 3.0 cm (mean caliper =4.62 ± 0.11 cm) (Wirkus Nurseries, Clinton, WI, USA), and
‘Mancana’ Manchurian ash trees ranged from 2.8 to 2.0 cm (mean caliper =5.06 ± 0.06 cm)
(Bailey Nurseries, St. Paul, MN, USA). Sixteen blue ash trees died during the 2008-09 winter
and were replaced in May 2009 with 5 cm caliper bare root trees (Connon Nursery, Waterdown,
ON, Canada). All trees were shipped from nurseries outside of the known EAB host range and
were uninfested when they were planted. Appropriate permits were acquired from state and
federal regulatory agencies prior to shipment and planting. Trees were watered twice per week
or as needed with drip irrigation.
Trees were fertilized annually from 2006-2009 with a top dressing of Harrel’s Pro-Blend
with Micros (19-5-10) (Harrell’s, Lakeland, FL, USA) at a rate of 70 g (5.5 ounces N) of product
per tree. In June 2006 and 2007, trees were protected from EAB with a cover spray of Tempo

31

®

SC Ultra (cyfluthrin, 11.8%, Bayer Environmental Science, Research Triangle Park, NC,
USA). In April 2008 and 2009, trees were wrapped with plastic tree wrap (Jobe's TreeWrap Pro,
Easy Gardener, Inc., Waco, Texas, USA) overlaid with screen mesh to deter EAB infestation.
Treatments
On 26 May 2010, seven trees of each species (35 trees per treatment) were randomly
assigned to one of three treatments: fertilization, PB application or untreated control. Fertilized
trees received 350 g of Harrell’s Pro-Blend with Micros (19-5-10) (Harrel’s, Lakeland, FL,
®

USA) distributed evenly around the base of each tree. ShortStop (Paclobutrazol 22.3%,
Zhejiang Tide Crop Science Co., LTD, Irvine, CA, USA) was applied as a soil drench. Diluted
product was applied evenly according to labeled instructions (200 ml diluted product, 4 g a.i., per
2.54 cm DBH) in a shallow trench, approximately 12 cm deep, around the base of each tree. All
tree boles and branches larger than 4 cm diameter were again wrapped from May to September
with tree wrap overlaid with wire mesh to minimize EAB infestation. On 23 May 2011, trees
fertilized in 2010 received a second application. Paclobutrazol, which typically affects tree
growth the year after application, was not reapplied in 2011. Tree wrap was not applied in 2011.
Variables Measured
Photosynthesis and transpiration rates of all trees were measured on a mature, sunexposed leaf using a Li-Cor 6400 portable photosynthesis meter (Li-Cor, Lincoln, Nebraska,
USA) on 8 June and 18 August 2010 and 3 June and 9 August 2011 between 11:00 and 16:00.
To reduce daily variation in light exposure, the leaf chamber was illuminated with an LED light
-1 -1

source set to provide 1,500 µmoles quantum flux m s . Leaf chamber block temperature (°C)
was set to correspond to the highest temperature predicted for the day to ensure consistent

32

temperature across daily measurements. Measurements were recorded when the real time
graphical displays indicated photosynthesis and transpiration rates had stabilized.
Relative chlorophyll content was assessed on four sun-exposed lower-canopy leaves from
each tree at each of cardinal direction on 8 June and 18 August 2010 and 3 June, 7 July and 7
August 2011 using a handheld Minolta SPAD-502 meter (Spectrum Technologies, Plainfield, IL,
USA). Relative chlorophyll content for each tree was determined by averaging the four readings.
To assess individual leaf area and foliar nutrients, four mid-canopy, sun-exposed leaves
were removed from each of the cardinal directions on 7 July 2010 and 13 July 2011. Four
randomly selected leaflets were removed from petioles and pooled by tree. Individual leaf area
®

was assessed using a flatbed scanner ((Epson Perfection 4490 Photo, Epson America, Inc.,
®

Longbeach, CA) and WinFOLIA software (Regent Instruments, Quebec, Canada). Individual
leaf area was then averaged by tree for each sample date. To assess foliar nutrients, pooled
leaves from each tree were oven dried (75°C) for 72 hours, ground and sent to Water’s
Agriculture Laboratory (Camilla, GA, USA) for foliar macro-nutrient (N, P, K and Ca) and
micro-nutrient (Mg, S, B, Zn, Mn, Fe and Cu) analysis.
Biomass Allocation
By August 2011, every black ash and 30 of the 35 green ash trees had been killed by
EAB so these species were excluded from evaluations of biomass allocation. Blue ash, white ash
and Manchurian ash trees were all still alive in 2011 and were destructively harvested from
September to October 2011. Trees were felled 10 cm above graft unions on grafted trees (white
ash and Manchurian ash) or approximately 15 cm above the first buttress root on blue ash trees.
Leaves and branches were removed from each tree, bagged separately and stored until they could
be oven dried.
33

On 19 and 20 October 2011, a skid steer mounted with a hydraulic tree spade was used to
3

extract the root ball and a consistent quantity of soil surrounding each tree (1.1 m ). Roots of
one fertilized white ash tree could not be removed because the tree spade could not penetrate the
soil to the required depth. Between 26 and 28 October 2011, soil was removed from extracted
roots by first penetrating the smooth soil surface with a pitchfork, then using a compressor®

driven air spade (Air-Spade , Guardair Corporation, Chicopee, MA, USA). Remaining soil
particles were subsequently removed with a garden hose. While air-spades do not typically
detach fine roots (Nadezhdina and Čermák 2003), I carefully sifted soil removed by the air-spade
to recover roots that might have been disconnected. Intact root masses were stored outdoors
between plastic tarps until they could be processed individually into coarse (> 2 mm) or fine
roots (≤ 2 mm) using hand pruners. All leaves, twigs, dissected roots, tree boles and root balls
were oven dried to constant weight and weighed.
I predicted root, leaf and woody biomass would vary among ash species and treatments.
However, I recognize inter- and intra-specific differences in initial, pre-treatment, tree size could
skew biomass allocation results. To account for differences in initial tree size among species, I
calculated ratios of biomass components (leaves, roots, woody above-ground) to total biomass
(the sum of all components), in addition to actual biomass for each tree.
Growth
During harvest, annual growth of the terminal shoot was recorded on the four lowest,
living branches and averaged by tree. Leaf scars were used to determine annual shoot growth for
two years before and two years after treatment.
Cross sections (1 cm thick) from the base of each tree were cut 10 to 15 cm above
ground, oven dried, lightly sanded, sealed with polyurethane (to visually clarify tree rings) and
34

scanned using a flatbed scanner. Tree ring increments were measured with tree ring analysis
®

®

programs (Coo-Recorder and CDendro , Cybis Elektronik and Data, Saltsjöbaden, Sweden).
The four most recent years of radial growth were measured at the top, bottom, right and left of
each cross section (16 measurements per tree) and averaged by year. I calculated ratios of pretreatment (2008 and 2009): post treatment (2010 and 2011) radial growth to standardize
measurements.
Statistical Analysis:
Data were analyzed using SAS statistical software (SAS Institute, Inc. 1989). Variables
measured in 2010 are presented for all five ash species, while data from 2011 are presented for
blue ash, white ash and Manchurian ash trees only because of nearly complete mortality of black
ash and green ash trees. Assumptions of normality were tested with residual plots and the
Shapiro-Wilk test (Shapiro and Wilk 1965). Leaf area was normalized using a log
transformation and foliar nutrient concentrations (N, P, K, Ca, Mg, S, B, Zn, Mn, B, Fe and Cu)
were normalized using log (x+2) transformations (Ott and Longnecker 2001). Two-way
ANOVA was performed to assess effects of ash species, treatment and interactions between the
two factors on shoot and radial growth, leaf area, biomass, relative chlorophyll content, foliar
nutrient concentrations and gas exchange rates. Relative chlorophyll content and gas exchange
rates were analyzed using repeated measures ANOVA. When ANOVA results were significant
(α ≤ 0.05), Fisher’s protected least significant difference (LSD) multi-comparison test with
Tukey’s adjustment was applied to shoot and radial growth, leaf area, gas exchange and nutrient
data sets (Ott and Longnecker 2001).
In 2011, simple linear regression was used to determine if tree caliper was a significant
predictor of harvested root biomass. To account for differences in initial tree size, ratios were
35

calculated for aboveground, leaf, root and total biomass. Biomass allocation data (root: total
biomass, aboveground woody: total biomass and canopy: total biomass ratios) were natural log
(ln) transformed before computing ratios, standard errors (SE), and ANOVA. Root: total
biomass ratios were calculated by dividing total root biomass (fine + coarse root weights) by
total biomass (canopy weight + twig weight + bole weight + coarse root weight + fine root
weight). Aboveground woody: total biomass ratios were calculated by dividing aboveground
woody biomass (twigs + boles) by total biomass. Canopy: total biomass ratios were calculated
by dividing leaf weight by total biomass.
Vector analysis of PB or fertilizer effects on total biomass nutrient concentration was
performed using methods developed by Haase and Rose (1995). For this analysis, total dry
weight (roots, leaves and aboveground woody biomass), nutrient concentration (percentage of
each foliar element) and nutrient content (percentage nutrient concentration as a function of tree
dry weight) are considered relative values that have been divided by corresponding values from
control trees (the mean of control tree variables within species) and multiplied by 100. For
example, foliar N levels of blue ash trees treated with PB or fertilizer were compared to the mean
foliar N content of blue ash control trees. I then replaced control values with a normalized
control always equal to 100, enabling comparisons to be made to a common base. Relative
nutrient concentration is thus graphically represented as a function of relative total tree dry
weight and relative foliar nutrient content. For presentation purposes, I refer to relative dry
weight as “biomass”. Relative biomass, nutrient content and nutrient concentration were
analyzed using 2-way ANOVA. Where significant for treatment or the species

treatment

interaction (α= 0.05), Dunnet’s adjustment was used in LS means tests to determine significant

36

differences between treated ash species and the normalized control. If ANOVA results were
significant for species only, vector analysis was not performed.

37

RESULTS
Growth
From 2008-2010, mean (±SE) shoot growth (cm) was different among species (Table
2.1). In 2008 and 2009, shoot growth of blue ash was more than 50% lower than that of white
ash and in 2010, blue ash shoots grew 89% less than white ash shoots grew (Table 2.2). In 2008,
mean (±SE) shoot growth of blue ash was similar to that of Manchurian ash, but in 2009 and
2010, blue ash shoot growth was 86 and 53% less than white ash and Manchurian ash shoot
growth, respectively. In 2008, Manchurian ash shoots grew 77% less than white ash, but in 2009
and 2010, shoot growth was similar between the two species (Table 2.2).
In 2011, two growing seasons after PB treatment, shoot growth and ratios of pre-: posttreatment shoot growth were affected by species, treatment and the interaction (Table 2.2).
Manchurian ash trees treated with PB had 74 and 70% less shoot growth than fertilized or control
Manchurian ash trees, respectively. In addition, ratios of pre-: post-treatment shoot growth of
Manchurian ash trees were 70 and 64% higher for control and fertilized trees, respectively than
Manchurian ash PB trees. Blue ash and white ash shoot growth and ratios of pre-: post-treatment
shoot growth were unaffected by treatments (Table 2.2).
Interspecific differences in radial growth were apparent during each sample year (20082011). In 2008, radial growth of blue ash and white ash trees were similar and both were lower
than Manchurian ash (Table 2.4). From 2009-2011, blue ash tree radial growth was at least 50%
lower than white ash and Manchurian ash, which were similar (Table 2.4).
Radial growth was unaffected by treatments until 2011 (Table 2.3). Fertilized ash trees
had 40% and 20% more radial growth than PB and control trees, respectively (Table 2.4). Ratios
of pre: post treatment radial growth were affected by species and treatment but not the

38

interaction (Table 2.3). Ratios of pre: post treatment radial growth were 39 and 33% higher in
white ash and Manchurian ash trees, respectively, than in blue ash but were similar to each other
(Table 2.4). Ratios of pre: post treatment radial growth in control and fertilized trees were
similar but were 33 and 43% higher than PB trees (Table 2.4).
In 2010, individual leaf area varied among species (F=14.4; df=4,83; P<0.001) but
treatment (P=0.16) and the interaction (P=0.20) were not significant. Area of individual white
2

ash leaves averaged 29.2 ± 2.34 cm and were at least 41% larger (P<0.001 for all species) than
black ash, blue ash and Manchurian ash leaves which averaged 17.2 ± 2.23, 17.2 ± 1.25 and 17.3
2

2

± 0.9 cm respectively. Green ash leaves averaged 12.8 ± 0.97 cm and were 44% smaller
(P<0.001) than white ash leaves and 26% smaller than leaves on black ash (P=0.046), blue ash
(P=0.038) and Manchurian ash leaves (P=0.009).
In 2011, individual leaf area varied among species (F=6.87; df=2, 42; P=0.003) and
treatments (F=3.10; df=2,42; P=0.043) but the interaction was not significant (P=0.31). Area of
2

white ash leaves averaged 29.1 ± 2.57 cm and which was 34% larger (P<0.001) than blue ash
2

2

(19.2 ± 1.10 cm ) and 23% larger (P=0.023) than Manchurian ash leaves (22.6 ± 1.36 cm ).
Blue ash and Manchurian ash leaf area were similar (P=0.17). Leaves from trees treated with PB
2

had an average leaf area of 19.8 ± 1.96 cm which was 19% smaller (P=0.010) than leaves from
2

control trees (24.5 ± 1.61 cm ) and 25% smaller (P=0.003) than leaves from fertilized trees (26.3
2

± 2.32 cm ). The leaf area of leaves from fertilized and control trees were similar (P=0.73).

39

Biomass Allocation
Root Biomass
Total root biomass was affected by species, but not treatment or the interaction (Table
2.5). Ratios of root: total biomass were affected by species and treatments but not the interaction
(Table 2.5). Blue ash trees had the lowest average root biomass (3.98 ± 0.40 kg) but had 29 and
13% higher average ratios of root: total biomass than white ash (average biomass = 8.65 ± 1.46
kg) or Manchurian ash (average biomass = 4.95 ± 0.35 kg) trees, respectively (Fig. 2.1A),
indicating blue ash trees allocated more carbon to root systems compared to white ash or
2

Manchurian ash trees. Root biomass extracted was related (R =0.432) to tree caliper (Y = 1448.15 + 47.95x) (F[1,46] = 35.03; P < 0.001) (Fig. 2.2).
Blue ash trees had a higher percentage of fine roots than white ash or Manchurian ash
trees. Fine root biomass averaged 0.75 ± 0.09 kg per tree for blue ash, 0.64 ± 0.09 kg per tree for
white ash and 0.46 ± 0.03 kg per tree for Manchurian ash trees and while the raw data was not
variable, ratios of fine: coarse roots varied among species (Table 2.5). The ratio of fine: coarse
roots of blue ash averaged 13.2% ± 0.02 which was higher (P <0.001) than the ratios for white
ash (5.3% ± 0.01) and Manchurian ash (3.6% ± 0.03) (Table 2.6).
Overall, trees treated with PB had the greatest root biomass (6.13 ± 1.23 kg) and the ratio
of root: total biomass was almost 10% higher (P<0.001) than fertilized or control trees (Fig.
2.1B), which averaged 5.92 ± 0.80 kg and 4.67 ± 0.52 kg of root biomass, respectively. The
ratio of root: total biomass of fertilized trees was similar (P=0.17) to control trees (Fig 2.1B).
Leaf Biomass
Foliar biomass of blue ash trees averaged 0.87 ± 0.21 kg which was 73% and 59% less
than white ash (3.28 ± 0.30 kg) and Manchurian ash (2.11 ± 0.24 kg), respectively (Table 2.6).
40

Leaf weight varied among species, while ratios of leaf: total biomass were variable among
species and treatments (Table 2.5). Blue ash ratio of leaf: total biomass was 6% smaller
(P<0.001) than white ash or Manchurian ash. Ratios of leaf: total biomass of white ash and
Manchurian ash trees were similar (P=0.88) (Fig. 2.1A).
While the PB trees had a smaller (P=0.010) ratio of leaf: total biomass than fertilized
trees, it was only a 2% difference. Ratio of leaf: total biomass for control trees was not different
from PB (P=0.13) or fertilized trees (P=0.31) (Fig. 2.1B).
Aboveground Woody Biomass
Given that radial growth was reduced by PB treatment, it is not surprising that trees
treated with PB had lower ratios of aboveground woody: total biomass. Aboveground woody
biomass was affected by species and the species

treatment interaction and ratios of

aboveground woody: total biomass were affected by species and treatments (Table 2.5). White
ash trees had the most aboveground woody biomass (14.57 ± 1.11 kg), blue ash trees had the
least (3.86 ± 0.73 g), while Manchurian ash trees were intermediate (6.58 ± 0.58 g) (Table 2.6).
Consequently, ratios of aboveground woody: total biomass of white ash trees were 13 and 7%
higher (P=0.001, P=0.050) than blue ash and Manchurian ash trees, respectively. Ratios of
aboveground woody: total biomass for blue ash and Manchurian ash trees were similar (P=0.14)
(Fig. 2.1A). Among treatments, ratios of aboveground woody: total biomass of PB trees were
7% lower (P=0.009) than control trees and 10% lower (P =0.050) than fertilized trees. Ratios for
fertilized and control trees were similar (P=0.35) (Fig. 2.1B).

41

Gas Exchange
In 2010, photosynthesis rates were affected by species, the species
species

treatment and

date interactions (Table 2.7). Transpiration rates were affected by species, treatment,

date, and the species

treatment and species

date interactions (Table 2.7). There were few

effects of treatment within species in 2010 (Table 2.8). In June, fertilized and PB blue ash trees
had higher photosynthesis rates than blue ash control trees. In August, fertilized and control
green ash trees had higher photosynthesis and transpiration rates than green ash PB trees (Table
2.8). However, it is important to note that the majority of PB (71%) green ash and fertilized
(71%) and PB (57%) black ash trees failed to leaf out in 2011as a result of EAB infestation.
Poor tree health as a result of infestation likely accounts for the low gas exchange rates exhibited
by black ash and green ash trees in these treatments in August 2010 (Table 2.8). In 2011, gas
exchange rates increased over time but within species, there were no treatment effects (Table
2.8).
Relative Chlorophyll Content
Treatment effects on relative chlorophyll content were variable but generally, chlorophyll
content increased between June and August in both growing seasons (Table 2.8). In 2010,
chlorophyll content was affected by species, treatment, date and the species
date and treatment

treatment, species

date interactions (Table 2.7). In June 2010, there were no within species

treatment effects. In August 2010, fertilized blue ash trees had higher chlorophyll content than
blue ash control and PB trees. White ash control and fertilized trees had higher chlorophyll
content than white ash PB trees (Table 2.8). There were no differences among treatments for
Manchurian ash trees,

42

In 2011, with the exception of blue ash, treatment effects on chlorophyll content were
inconsistent. Mean (±SE) chlorophyll content was affected by species, treatment, date, species
treatment, species

date and treatment

date interactions (Table 2.7). In June, fertilized blue

ash trees had higher chlorophyll content than blue ash control and PB trees, which did not differ
from each other. No other within-species treatment effects were apparent (Table 2.8). In July,
chlorophyll content of fertilized blue ash and white ash trees were higher than blue ash and white
ash control and PB trees and Manchurian ash control trees had lower chlorophyll content than
PB and fertilized Manchurian ash trees. By August, fertilized blue ash trees still had higher
chlorophyll content than blue ash control or PB trees, but no treatment effects persisted among
white ash or Manchurian ash trees (Table 2.8).
Nutrient Content
In 2010, foliar concentrations of N, P, K, Mg and Ca in foliage differed among ash
species (Table 2.9) and treatment affected N and P concentrations (Table 2.9). Micronutrients,
S, Zn, Mn, Fe, B and Cu were not affected by species or treatments. Foliar N concentration was
lower in blue ash (1.94% ± 0.086) than all other species (Table 2.10). Blue ash also had lower
foliar P concentrations than black, green and white ash and lower foliar K concentrations than
white and Manchurian ash (Table 2.10). In contrast, Mg concentration in blue ash foliage
averaged 0.65 % ± 0.061, which was two to three times higher than in any other species. Foliar
nutrient concentrations in green ash and black ash were similar among N, P, K, Mg and Ca
(Table 2.10). White ash foliage had higher N concentration than black ash, while K
concentration was lower in white ash than in black ash. Manchurian ash had higher foliar N, P,
and K concentrations than blue ash, but lower Mg and Ca concentrations. Magnesium

43

concentration was higher in Manchurian ash than in black ash, green ash and white ash (Table
2.10).
In 2010, species and treatment affected N and P concentrations (Table 2.9). Nitrogen
concentration of fertilized and control tree averaged 2.59% ± 0.077 and 2.68% ± 0.090,
respectively, which was 9 and 5% higher than PB trees which averaged 2.45% ± 0.091. Nitrogen
concentration was similar between fertilized and control trees. Phosphorous concentrations
averaged 1.43% ± 0.036 which was higher than P concentrations for white ash and Manchurian
ash trees which averaged 1.33% ± 0.033 and 1.33% ±0.030, respectively. Concentrations of P
were not different between blue ash and Manchurian ash trees.
In 2011, blue ash control and PB trees had lower foliar N concentrations than the other
ash trees in the study (Table 2.10). Overall, N concentrations were affected by species, treatment
and the interaction (Table 2.9). Nitrogen concentrations were higher in fertilized blue ash trees
(2.2% ± 0.08) than in blue ash control (1.7% ± 0.12) or PB (1.7% ± 0.13) trees, which were
similar. Among white ash and Manchurian ash trees, there were no differences among
treatments (Table 2.10). Phosphorous concentration was affected by species and treatment but
not the interaction (Table 2.9). Foliar P concentration of Manchurian ash was 15 and 48% higher
than in white ash or blue ash foliage, respectively. Among treatments, P concentration was 29
and 20% higher in control tree foliage (0.21% ± 0.019) than in foliage from fertilized (0.15% ±
0.010) or PB (0.17% ± 0.011) trees, respectively. Potassium, Mg, and Ca were affected by
species only (Table 2.9). Potassium concentrations were 31% higher in Manchurian ash trees
than in blue ash trees but concentrations in Manchurian ash were similar to white ash (Table
2.10). Blue ash foliage had 64% and 44% higher Mg concentration and 45 and 29% higher Ca

44

concentration than white ash or Manchurian ash foliage, respectively (Table 2.10). As in 2010,
S, Zn, Mn, Fe, B and Cu concentrations were not affected by species or treatments.
In 2011, I took advantage of our total tree harvest and used vector analysis to further
assess treatment effects on N and P concentrations. In this analysis, percent relative dry weight
(the “Z” axis) is consistent across diagrams because the same trees were used for each nutrient.
Relative biomass varied among species (F=7.0; df=2, 27; P=0.006) and treatments (F=7.2; df=2,
27; P =0.005) but the interaction was not significant (P=0.19). When compared to normalized
controls, biomass was lowest (- 41%) for blue ash PB trees and highest (+ 25%) for blue ash
fertilized trees (Figs. 2.2 and 2.3). White ash PB and fertilized trees had more biomass than
normalized controls, while biomass of fertilized and PB Manchurian ash trees was not different
from the normalized control (Figs. 2.3 and 2.4).
Relative N concentration (the “Y” axis) was affected by species (F=5.15; df=2, 27;
P=0.013), treatment (F=7.92; df=2, 27; P =0.009) and the species

treatment interaction

(F=7.21; df=2, 27; P=0.003). Relative N content (the “X” axis) was affected by treatment
(F=5.70; df=2, 27; P=0.024) and the species

treatment interaction (F=6.33; df=2, 27; P=0.006)

but not species (P=0.25). The highest relative N concentration was recorded for fertilized blue
ash trees with values that ranged from +15 to +38% (Fig. 2.3). Relative N concentrations of all
other trees (PB blue ash and PB and fertilized white ash and Manchurian ash trees) were not
different from the normalized control (Fig. 2.3). Relative N content was also variable. Blue ash
PB trees had 45% less relative N content and blue ash fertilized trees had 57% more relative N
content than normalized controls (Fig. 2.3). The lower relative N content in blue ash PB trees,
however, is offset by less biomass. In contrast, fertilized blue ash trees had higher relative N
content, relative N concentration and biomass than the normalized control. An increase in all

45

three parameters indicated trees were responding to N deficiency. Manchurian ash and white ash
PB and fertilized trees received sufficient N to maintain relative N concentrations.
Vector analysis indicated PB treatment could result in an antagonistic effect on P
concentration, whereas fertilization could result in P dilution because of excess growth. Relative
P concentration was affected by species (F=5.85; df=2, 27; P =0.008) and treatment (F=5.69;
df=2, 27; P =0.009) but not the species

treatment interaction (P =0.09). Compared to the

normalized control, all blue ash and Manchurian ash trees had insufficient relative P
concentrations, -11 and -32%, respectively. Relative P content was variable among species and
treatments, ranging from -48.4% for blue ash PB trees to +38.5% for white ash PB trees (Fig.
2.4). Blue ash and Manchurian ash PB trees exhibited a decrease in all three relative parameters,
indicating the PB might have an antagonistic effect on relative P content. In other words, trees in
our plantation were responding phosphorous deficiency induced by PB treatment. In contrast,
fertilized blue ash and Manchurian ash trees experienced an increase in biomass, a decrease in
relative P concentration and no change in relative P content. The increase in biomass and
decrease in P concentration indicated trees were responding to a dilution effect (Fig. 2.4). White
ash trees displayed an increase in biomass and relative P content, but there was no change in
relative P concentration, indicating that they had sufficient P.

46

DISCUSSION
Because ash trees are common in urban landscapes and experience varying degrees and
types of environmental stress, understanding effects of treatments designed to enhance tree vigor
is critical for arborists and landscape professionals. My results indicate that effects of fertilizer
and PB varied among ash species which is not surprising given their evolutionary diversity
(Wallander 2008) and tendency to occupy diverse ecological niches (USDA plants.gov,
Accessed 14 April 2013). While other studies (Bai et al. 2004, Bai et al. 2005) and mine indicate
white ash might be relatively unresponsive to PB, I found blue ash and Manchurian ash were
sensitive to PB treatment.
I recognize that our study was somewhat limited by the size of the trees. Several studies
have assessed the effects of PB on similarly sized/aged ash trees (Sterrett and Tworkoski 1987,
Bai et al. 2004, Watson 2004) and even though I could find no studies that examined PB or
similar compounds on ash trees larger than the ones used in my study, I suspect that PB
treatment effects would likely carry over to larger trees. Most treatment effects of PB are the
result of changes in allocation. Stem and radial growth reduction is a well-documented effect of
PB (Sterrett and Tworkoski 1986, Watson 2001, Arzani and Roosta 2004, Bai et al. 2004, Yadav
et al. 2005, Haas 2012) and has been recorded for a wide variety of plants ranging from turf grass
to citrus (Yelenosky et al. 1995) and landscape trees (Chaney 2005 a, b). Growth is reduced by
PB at the cellular level and should presumably carry over to large trees. Treated plants generally
have the same number of cells in stems and leaves as control plants, but their cells are shorter.
Smaller cells reduce shoot and radial growth, as well as leaf area and thickness (Yelenosky et al.,
1995, Bai et al., 2004, Yadav 2005). In addition, several studies have shown that PB affects

47

growth of larger trees in a variety of genera including Quercus spp. (Redding et al. 1994, Watson
1996) and Acer spp. (Redding et al. 1994, Burch and Wells 1995).
Effects of fertilization in this study were not as dramatic as I might have expected. There
is general consensus that fertilization enhances tree growth (Schulte and Whitcomb 1975, Struve
2002, Wang et al. 2012), but not all landscape trees require fertilization (Struve 2002, Ferrini and
Baietto 2006, Harris et al. 2008, Watson 2010, Chorbadjian et al. 2011, Wang et al. 2012). I
hypothesized initially that fertilization would increase ash shoot growth, but shoot growth of
fertilized ash trees in our plantation was similar to control trees. Radial growth was higher in
fertilized trees than in control trees in 2011, but differences were small and probably not
biologically relevant. Ratios of pre: post treatment shoot and radial growth did not differ
between fertilized or control trees, which is not surprising given all trees used in this study were
fertilized from 2006-2009. Similarities between control and fertilized trees indicate that our ash
trees, like many urban landscape trees, were growing in an environment that provided adequate
nutrients. In hindsight, perhaps I should have included a nutrient deficiency treatment in the
study (for example, adding a pine bark amendment to the plantation soil (Gilman 2004)).
Urban trees planted in locations with limited space for growth can become stressed from
lack of resources when they overgrow their limited space (Chaney 2005a). In this study, PB
treatment reduced shoot growth of Manchurian ash trees by 70% and radial growth of ash trees
suggesting PB would likely increase the health of ash trees planted in confined, urban areas. Our
study illustrates how species within the same genus can react differently to PB treatment.
Manchurian ash was the only species to have reduced shoot growth but the 16 blue ash trees
planted in 2009 might have been suffering from transplant shock during that year, which could
have influenced pre-treatment shoot growth. Other studies also report that white ash trees are

48

relatively insensitive to PB treatment (Bai et al. 2004, Bai et al. 2005). In a study that compared
flurprimidol (a triazole gibberellin inhibitor) and PB trunk injection on growth of black walnut
(Juglans nigra) and white ash, researchers found that white ash exhibited reduced growth when
treated with flurprimidol but not when treated with PB (Sterrett and Tworkoski 1987).
Fertilization often increases relative chlorophyll content but fertilization effects on gas
exchange rates are variable among species (Loh et al. 2002, Klooster et al. 2012). In a study
using seedling Manchurian ash, Koike and Sanada (1989) found that photosynthesis rates
increased when seedlings were fertilized but Klooster et al. (2012) found that net photosynthesis
rates of a variety of pot-in-pot hardwood species were unaffected by fertilization. In our study,
fertilization did not affect the photosynthesis rates of white ash or Manchurian ash. Fertilized
blue ash had higher photosynthesis rates than control trees, but only on one sample date (August
2010). Fertilization consistently increased relative chlorophyll content in blue ash leaves and
fertilized white ash and Manchurian ash leaves compared to control white ash and Manchurian
ash leaves in July 2011, but the trend had disappeared by August.
Applications of PB generally increase chlorophyll concentration and reduce leaf area
(Sharma et al. 2009) but effects on gas exchange rates are variable (Wieland and Wample 1985,
Rieger and Scalabrelli 1990, Fletcher and Gilley 2000, Percival and AlBalushi 2007 ). In a study
using English (Quercus robur L.) and evergreen oak (Quercus ilex L.), Percival and AlBalushi
(2007) found that chlorophyll content increased, leaf area decreased and photosynthetic
efficiency was 60 to 100% higher in PB trees compared to control trees. In our study, PB
reduced leaf area but relative chlorophyll content and gas exchange rates were rarely different
between PB and control trees within any species.

49

Because of expense, time constraints and the equipment necessary for excavation, there
are few studies that evaluate whole tree coarse-root or fine-root biomass (Vanninen and Mäkelä
1999, Lavigne and Krasowski 2007 Drexhage and Colin 2001). When extracting root systems, a
quantity of the fine root system is typically unaccounted for because roots are severed and left
behind in the soil. This can lead to standard errors ≥ 99% (Vanninen and Mäkelä 1999). Some
researchers correct for this loss by adjusting root biomass with a defined percentage of root loss
(for example, Goff et al. 2001 multiplied all root biomass data by 13%), others extract
subsamples of the fine root system and use statistics to extrapolate total root biomass (Yin et al.
1989). I recognize that although I did not harvest 100% of the roots from trees in our plantation,
I did standardize our study by excavating a consistent volume of soil for each tree.
Fine root biomass generally decreases on sites with nutritious soils and cultural
applications of fertilizer can potentially decrease fine root production (Gower and Vitousek
1989, Vanninen and Mäkelä 1999). In this study, root biomass of fertilized and control trees
were similar. I could find no reports in the literature of typical root biomass for ash species, but
our results were similar to other studies that indicate fine roots generally comprise 2 to 10% of
total tree biomass (blue ash trees ratio of fine roots: total biomass (8.7%), white ash (2.4%) and
Manchurian ash (3.4%)) (Vanninen and Mäkelä 1999, Hendrick and Pregitzer 1993, Bolte et al.
2004).
When PB reduces shoot and radial growth, shifts in biomass partitioning often result, but
effects vary by genera and species (Watson 1996, 2000, 2004, Fletcher and Gilley 2000, Yadav
et al. 2005,). Reductions in leaf biomass have been reported for orchard and landscape trees with
concomitant increases in root biomass (George and Nissen 2002, Bai et al. 2004). For a biomass
allocation shift toward root production to take place, two things must happen. Growth of shoots
50

and/or leaves must be reduced and the amount of photosynthate produced by trees must increase
or remain steady. This creates an excess of available carbohydrates that can be used for root
production. In large transplanted green ash trees, root dry weight and length were not affected
by PB treatment (Watson 2004). In contrast, PB increased pin oak (Quercus palustris Münchh)
fine root density in the upper 5 cm of the soil profile by 64% (Watson 1996). In peach, fine root
length was reduced with increasing PB concentrations, but root: shoot ratios increased when
trees were treated with ≥ 0.10 mg PB per liter (Reiger and Scalabrelli 1990). While other studies
have not shown significant effects of PB on ash tree roots (Sterrett and Tworkoski 1986, Watson
2004), in 2011, one year after application, I found that PB increased root biomass. I found that
PB reduced aboveground woody material of ash trees by 10% compared to control trees, while
photosynthesis rates remained the same. As a result, ratios of root: total biomass of PB trees
increased by 10%. Fertilized trees had larger canopies (+2%) and more aboveground woody
biomass (+7%) than PB trees but again, photosynthesis rates were not different among treated
trees. The decrease of 9% in aboveground biomass of PB trees was therefore reflected in a 9%
increase in root biomass.
I assessed effects of species and treatments on foliar micro- and macro-nutrient
concentrations and then used vector analysis to assess how treatments were affecting nutrient
status and growth response. Vector analysis has been used in several fertilizer and nutrient
studies (Imo and Timmer 1997, Labrecque et al. 1998, Isaac et al. 2007) to pin point “dilution
effects, nutrient imbalances, and element interactions” (Haase and Rose 1995). In our study,
fertilizer effects on foliar nutrient concentrations were most noticeable in blue ash. In 2011,
fertilized blue ash trees had higher relative chlorophyll content and N concentrations than control
trees. Nutrient vector analysis further indicated that fertilized blue ash trees were responding to

51

N deficiency. Blue ash occurs naturally on limestone outcropping (plants.gov 2013, Barnes and
Wagner 2003) and despite their ability to tolerate poor sites, our study indicated blue ash growth
responded positively to fertilization at our site. For the fertilized white ash and Manchurian ash
trees, N was not limiting. Vector analysis further indicated that P was limiting for fertilized blue
ash and Manchurian ash trees, which were responding to a dilution effect caused by increased
biomass. White ash trees received adequate N and P in our study.
PB effects on nutrient concentrations are known to vary among species and genera. In a
variety of citrus cultivars and Arbutus unedo L. (an evergreen Ericaceous shrub), PB increased N
and Ca and some micro-nutrient concentrations (Yelenosky et al. 1995, Navarro et al. 2009). In
some peach cultivars, however, PB increased Ca and Mg, but lowered N and P (Rieger and
Scalabrelli 1990, Huett et al. 1997). While PB application did not decrease N concentrations in
our ash species it did cause an antagonistic effect (PB treatment was reducing P concentration)
on P content in both blue ash and Manchurian ash trees. Phosphorous is a key component of the
chemical energy produced in chloroplasts which is used for a variety of processes including
growth and synthesis of proteins and nucleic acids (Mengel and Kirkby 1982). In ash trees, P
deficiency can result in reduced growth, root conductivity and leaf: total biomass ratios
(Anderson et al. 1989). In fruit trees, P deficiency can also cause aborted shoot buds, discolored
leaves and early leaf senescence (Mengel and Kirkby 1982). While a reduction in growth as a
result of P deficiency would not be considered an adverse effect in PB treated trees, discolored
leaves and poor root conductivity could potentially reduce urban ash tree health and aesthetics.
Further investigation is necessary to evaluate long term effects of PB on P uptake and
implications of P deficiency on ash tree health.

52

I evaluated a variety of growth, physiological and nutrient effects of fertilizer and PB on
an evolutionarily diverse group of ash species. Fertilization of ash trees is common in urban
areas. Homeowners and tree care companies apply fertilizers to enhance aesthetics and improve
tree vigor. Our study site contained adequate nutrition for white ash and Manchurian ash, but
blue ash responded positively to additional nutrients. Paclobutrazol is gaining popularity among
arborists and is marketed in EAB infested areas to reduce ash tree stress. It is unfortunate that I
was unable to determine the effects of PB on green ash as it is the most common ash species in
urban areas of the United States. However, I found PB reduces radial growth and shifts biomass
allocation from above to below ground resources of ash trees. Therefore, PB application could
extend the period before trees outgrow available space, enabling trees to persist in confined
spaces.

53

Table 2.1: Results of analysis of variance to assess effects of treatment on shoot growth and
1
ratios of pre-: post-treatment shoot growth on blue, white and Manchurian ash trees
treated with paclobutrazol, fertilizer or left as untreated control.

F Value
Effect
Species
Treatment
Species × Treatment

35.69
1.43
1.93

Species
Treatment
Species × Treatment

8.64
0.95
0.56

Species
Treatment
Species × Treatment

40.18
2.18
0.95

Species
Treatment
Species × Treatment

30.52
9.92
4.93

Species
Treatment
Species × Treatment

14.72
5.34
2.59

Shoot Growth
Degrees of Freedom
2008
2,44
2,44
4,44
2009
2,44
2,44
4,44
2010
2,44
2,44
4,44
2011
2,44
2,44
4,44
Pre-: Post-treatment Ratio
2,44
2,44
4,44

1

P value
<0.001*
0.252
0.121
<0.001*
0.400
0.700
<0.001*
0.125
0.442
<0.001*
<0.001*
0.002*
<0.001*
0.008*
0.050*

Pre-: post-treatment ratio = (growth 2008 + growth 2009) / (growth 2010 + growth 2011).

54

1

Table 2.2: Average (±SE) shoot length (cm) in 2008-2010 and ratios of pre-: post-treatment
growth of blue ash, white ash and Manchurian ash trees treated with fertilizer,
paclobutrazol or left as untreated controls (2011). Measurements were taken in
October 2011 using annual leaf scars. Within years, different letters indicate a
significant (α≤0.05) difference between species (a,b,c). Among average (±SE) shoot
length (cm) in 2011 and pre-: post-treatment ratios, different letters indicate a
significant (α≤0.05) difference between treatments within species (x,y).
Year
Ash species

2008

2009
b

Blue

15.0± 2.37

Manchurian

13.1 ± 1.27

White

30.6 ± 1.76

Ash
Species

Treatment

b
a

2010

13.4 ± 2.37

29.1 ± 3.36
31.1 ± 3.07

4.4 ± 1.08

Fertilizer

7.9 ± 1.15
4.0 ± 0.73

31.8 ± 4.06

Fertilizer

37.0 ± 5.84

Paclobutrazol

9.7 ± 2.10

Control

25.7 ± 1.26

Fertilizer

25.5 ± 2.59

Paclobutrazol

21.1 ± 4.30

1

5.2 ± 1.22

b

37.0 ± 4.58
44.4 ± 1.80

a
a

Pre: Post Treatment Ratio

Control

White

a

x

Paclobutrazol
Manchurian

a

2011

Control
Blue

b

x
x
x
x

y
x
x
x

0.51 ± 0.24
0.43 ± 0.13
0.43 ± 0.13
2.54 ± 0.55
2.15 ± 0.43
0.76 ± 0.16
1.57 ± 0.36
1.18 ± 0.09
0.99 ± 0.18

x
x
x
x
x
y
x
x
x

Pre-: post-treatment ratio = (growth 2008 + growth 2009) / (growth 2010 + growth 2011).

55

Table 2.3: Results of analysis of variance to assess effects of treatment on annual radial growth
1

and ratios of pre-:post-treatment growth of blue ash, white ash and Manchurian ash
trees treated with paclobutrazol, fertilizer or left as untreated control.

F Value
Effect
Species
Treatment
Species × Treatment

3.35
0.10
1.05

Species
Treatment
Species × Treatment

16.33
0.16
0.27

Species
Treatment
Species × Treatment

25.42
3.07
1.21

Species
Treatment
Species × Treatment

30.52
9.92
4.93

Species
Treatment
Species × Treatment

17.11
18.53
0.78

Annual Increment Growth
Degrees of Freedom
2008
2,44
2,44
4,44
2009
2,44
2,44
4,44
2010
2,44
2,44
4,44
2011
2,44
2,44
4,44
Pre-: Post-treatment Ratio
2,44
2,44
4,44

1

P value
0.044*
0.905
0.392
<0.001*
0.853
0.421
<0.001*
0.057
0.318
<0.001*
<0.001*
0.002*
<0.001*
<0.001*
0.542

Pre-: post-treatment ratio = (growth 2008 + growth 2009) / (growth 2010 + growth 2011).

56

1

Table 2.4: Mean (±SE) radial growth (mm) from 2008-2011 and ratios of pre-: post-treatment radial growth of blue ash, white ash
and Manchurian ash trees treated with fertilizer, paclobutrazol or left as untreated controls. Treatment effects were not
significant (P>0.05) in 2008-2010. Within years, different letters indicate significant (α≤0.05) differences between
species (a,b,c) and treatments (x,y).
Year
2009

2008

Blue

2.44 ± 0.38

Manchurian

3.51 ± 0.212

White
Treatment

3.11 ± 0.265

Control

NS

NS

NS

4.01 ± 0.56

Fertilizer

NS

NS

NS

5.04 ± 0.41

Paclobutrazol

NS

NS

NS

3.05 ± 0.54

b
a

b

1.94 ± 0.40

2010

Pre: Post Treatment
Ratio

Ash species

b

4.06 ± 0.19

a

3.58 ± 0.305

a

1.76 ± 0.34

2011
b

4.42 ± 0.35
4.57 ± 0.29

1

a
a

2.34 ± 0.48

b

5.49 ± 0.44
4.86 ± 0.30

a
a

y

x
y

Pre-: post-treatment ratio = (growth 2008 + growth 2009) / (growth 2010 + growth 2011).

57

0.92 ± 0.99

b

1.37 ± 0.10
1.51 ± 0.09
1.29 ± 0.11
1.52 ± 0.08
0.87 ± 0.10

a
a

x
x
y

Table 2.5: Results of analysis of variance to assess effects of species and treatment on leaf, aboveground woody, fine root, coarse
root, total root and total biomass and leaf, aboveground woody, and ratios of root biomass: total biomass and fine: coarse
root of blue ash, white ash and Manchurian ash trees treated with paclobutrazol, fertilizer or left as untreated control.
Trees were harvested in October 2011.
Degrees of
Degrees of
P Value
F Value
Freedom P Value
F Value
Freedom
Effect
Leaves
Leaf: Total Biomass
Species
25.54
2, 41
<0.001*
25.94
2, 41
<0.001*
Treatment
1.15
2, 41
0.329
3.68
2, 41
0.034*
Species × Treatment
1.30
4, 41
0.285
1.95
4, 41
0.110
Aboveground Woody
Aboveground Woody: Total Biomass
Species
32.77
2, 41
<0.001*
4.77
2, 41
0.014*
Treatment
0.98
2, 41
0.385
3.42
2, 41
0.042*
Species × Treatment
2.75
4, 41
0.040*
1.02
4, 41
0.569
Fine Roots
Total Roots: Total Biomass
Species
1.91
2, 41
0.162
5.64
2, 41
0.007*
Treatment
1.15
2, 41
0.326
17.46
2, 41
<0.001*
Species × Treatment
1.30
4, 41
0.287
0.88
4, 41
0.480
Coarse Roots
Fine Roots: Coarse Roots
Species
11.92
2, 41
<0.001*
15.72
2,41
<0.001*
Treatment
1.78
2, 41
0.182
0.97
2, 41
0.430
Species × Treatment
2.01
4, 41
0.112
1.04
4, 41
0.350
Total Roots
Species
10.54
2, 41
<0.001*
Treatment
1.64
2, 41
0.207
Species × Treatment
2.11
4, 41
0.097
Total Biomass
Species
38.18
2, 41
<0.001*
Treatment
1.06
2, 41
0.358
Species × Treatment
2.71
4, 41
0.043*
58

Table 2.6: Mean (±SE) biomass (kg) of leaves, total roots, fine roots, coarse roots aboveground woody and total biomass of blue ash,
white ash and Manchurian ash trees treated with fertilizer, paclobutrazol or left as untreated controls. Biomass was
harvested in October 2011. Among plant parts, different letters indicate a significant (α≤0.05) difference between species
(leaves, total roots, fine roots and coarse roots) (a,b,c) or within species and treatments (aboveground woody and total
biomass) (x,y). Treatment effects on leaves, total roots, fine roots and coarse roots were not significant (ns).

Ash Species

Biomass Allocation (kg)
Fine Roots
Coarse Roots

Leaves

Blue

0.87 ± 0.21

White

3.28 ± 0.28

Manchurian
Treatment

2.11 ± 0.23

Control

1.96 ± 0.32

Fertilizer

2.21 ± 0.32

Paclobutrazol

1.60 ± 0.36

Ash Species

c
a

b

ns
ns
ns

Treatment

0.75 ± 0.09
0.64 ± 0.09
0.46 ± 0.03
0.58 ± 0.09
0.69 ± 0.08
0.57 ± 0.07

a
a
a

ns
ns
ns

Aboveground Woody Biomass
x

Control

2.03 ± 0.25
9.90 ± 3.24

Fertilizer

14.61 ± 1.68

Paclobutrazol

16.74 ± 0.91

Control
Manchurian

5.18 ± 1.26

Control
White

Fertilizer
Paclobutrazol

Blue

4.21 ± 1.78

7.29 ± 1.02

Fertilizer

6.60 ± 1.13

Paclobutrazol

5.55 ± 0.73
59

x
y
y
x
x

x
x
x

3.23 ± 0.41
4.54 ± 0.36

b
b

7.51 ± 1.13
4.09 ± 0.54
5.23 ± 0.78
5.12 ± 0.88

a

ns
ns
ns

Total Biomass
9.22 ± 3.33

x

11.27 ± 2.02
5.32 ± 0.47

x

x

20.23 ± 2.63
26.75 ± 4.51
31.67 ± 3.38
14.09 ± 1.75
13.78 ± 1.27

x
x
x
x

xy

12.84 ± 1.42

y

Total Roots
3.983 ± 0.40
8.65 ± 1.46
4.95 ± 0.35
4.67 ± 0.52
5.92 ± 0.80
6.13 ± 1.23

b

b
b

ns
ns
ns

Table 2.7: Results of analysis of variance of effects of species, treatment and date on photosynthesis and transpiration rates and
relative chlorophyll content of black, blue, green, white and Manchurian ash trees (2010) or blue ash, white ash and
Manchurian ash trees (2011) treated with paclobutrazol, fertilizer or left as untreated control. Trees were sampled in June
and August 2010 and 2011.

Photosynthesis

Transpiration

Relative
Chlorophyll
Content

Effect
Species
Treatment
Species × Treatment
Date
Species × Date
Treatment × Date
Species × Treatment × Date
Species
Treatment
Species × Treatment
Date
Species × Date
Treatment × Date
Species × Treatment × Date
Species
Treatment
Species × Treatment
Date
Species × Date
Treatment × Date
Species × Treatment × Date

2010
Degrees of Freedom
F Value
(num, den)
8.28
4,124
0.24
2, 124
2.95
8,124
2.89
1, 124
4.83
4, 124
0.70
2,124
1.13
8, 124
12.41
4,124
3.67
2, 124
3.35
8,124
10.47
1, 124
4.54
4, 124
1.06
2,124
0.40
8, 124
99.89
4,124
7.24
2, 124
3.33
8,124
60.82
1, 124
7.24
4, 124
4.94
2,124
0.57
8, 124

60

P Value
<0.001*
0.384
0.005*
0.099
0.001*
0.711
0.708
<0.001*
0.028*
0.002*
0.002*
0.002*
0.469
0.939
<0.001*
<0.001*
<0.001*
<0.001*
<0.001*
0.008
0.739

2011
Degrees of Freedom
F Value
(num, den)
26.93
2, 106
0.53
2, 106
0.55
4, 106
25.44
1, 106
3.05
2, 106
2.27
2, 106
1.29
4, 106
23.31
2, 106
0.91
2, 106
0.25
4, 106
121
1, 106
10.72
2, 106
2.56
2, 106
0.57
4, 106
119.4
2, 106
52.61
2, 106
18.85
4, 106
339.57
1, 106
12.23
2, 106
4.1
2, 106
0.78
4, 106

P Value
<0.001*
0.588
0.702
<0.001*
0.051
0.108
0.278
<0.001*
0.407
0.911
<0.001*
<0.001*
0.082
0.682
<0.001*
<0.001*
<0.001*
<0.001*
<0.001*
0.003*
0.624

2

-2 -1

-2 -1 1

Table 2.8: Mean (±SE) photosynthesis rates (μmol CO m s ), transpiration rates (mmol H2O m s ) and relative chlorophyll
2

content of black ash, blue ash, green ash, white ash and Manchurian ash trees (2010) or blue ash, white ash and
Manchurian ash trees (2011) treated with fertilizer, paclobutrazol (PB) or left as untreated controls. Measurements were
taken in June and August of 2010 and 2011. Within species, different letters indicate a significant (α=0.05) difference
between treatments.
2010
Photosynthesis
Ash
Species

Blue

Control
Fertilizer
PB

June
9.5 ± 2.44
7.8 ± 0.04
4.5 ± 3.0

August
5.0 ± 2.77
3.3 ± 1.65
3.3 ± 1.20

Control

Black

6.9 ± 0.93

3.87 ± 1.38

Fertilizer

10.3 ± 0.65

8.3 ± 2.51

Control

10.3 ± 1.79 16.8 ± 1.31
9.0 ± 1.99

a

ab

12.4 ± 1.72 6.6 ± 2.14

Fertilizer

June
0.17 ± 0.062
0.18 ± 0.038
0.07 ± 0.024

b

PB
Green

Relative Chlorophyll
Content

Transpiration

15.3 ± 1.02

a
a

b

August
0.17 ± 0.075
0.11 ± 0.04
0.10 ± 0.026

June
38.1 ± 1.62
30.9 ± 2.50
33.2 ± 1.60

August
45.6 ± 1.46
44.1 ± 2.62
39.3 ± 1.78

0.23 ± 0.042

0.10 ± 0.054

37.0 ± 1.02

34.2 ± 1.54

0.25 ± 0.029

0.23 ± 0.083

35.9 ± 0.83

37.3 ± 1.56

0.21 ± 0.047

0.18 ± 0.060

36.1 ± 0.91

34.1 ± 1.65

43.8 ± 1.12

47.8 ± 1.49

46.9 ± 1.04

51.6 ± 1.59

0.35 ± 0.012
0.36 ± 0.061

a
a

b

0.49 ± 0.053
0.47 ± 0.052

a
a

b

a

b
a

6.9 ± 3.33
12.5 ± 1.02
12.2 ± 1.32

0.09 ± 0.048
0.27 ± 0.025
0.28 ± 0.033

0.21 ± 0.102
0.40 ± 0.053
0.42 ± 0.069

43.6 ± 1.02
31.1 ± 0.88
30.5 ± 1.18

45.3 ± 1.61
36.1 ± 1.15
36.9 ± 1.04

9.9 ± 1.30

12.8 ± 0.90

0.18 ± 0.036

0.48 ± 0.022

32.5 ± 1.01

36.6 ± 0.76

Control
White

7.2 ± 2.85
10.3 ± 1.01
10.7 ± 1.52

PB

Manchurian

PB
Control
Fertilizer

7.8 ± 1.47

9.0 ± 1.44

0.16 ± 0.006

0.31 ± 0.072

43.3 ± 0.72

45.9 ± 0.87

Fertilizer

11.0 ± 1.50

10.9 ± 1.68

0.27 ± 0.020

0.36 ± 0.059

42.6 ± 0.80

45.9 ± 0.72

PB

11.8 ± 2.31

11.7 ± 0.46

0.28 ± 0.053

0.42 ± 0.040

41.4 ± 0.75

41.6 ± 1.27

61

a
a

b

Table 2.8 (cont’d)

Photosynthesis
June
August

2011
Transpiration
June
August

Relative Chlorophyll Content
June
July
August
a

Control

0.29 ± 0.022

0.40 ± 0.018

27.2 ± 0.63

Fertilizer

12.5 ± 0.99

13.1 ± 1.08

0.30 ± 0.028

0.40 ± 0.040

31.8 ± 0.36

12.5 ± 0.64

11.8 ± 1.54

0.30 ± 0.017

0.34 ± 0.062

28.4 ± 0.58

Control
Manchurian

14.0 ± 0.84

PB

Blue

11.3 ± 0.75

15.7 ± 0.56

17.1 ± 0.72

0.38 ± 0.021

0.61 ± 0.018

26.9 ± 0.34

Fertilizer
PB
Control

White

Fertilizer
PB

14.2 ± 1.31
16.4 ± 0.96
11.8 ± 1.16
11.7 ± 1.09
11.5 ± 1.55

20.1 ± 0.71
17.9 ± 0.66
16.5 ± 0.62
16.2 ± 0.96
13.9 ± 2.13

0.33 ± 0.035
0.39 ± 0.037
0.27 ± 0.032
0.22 ± 0.025
0.24 ± 0.052

0.65 ± 0.015
0.59 ± 0.023
0.55 ± 0.021
0.54 ± 0.053
0.45 ± 0.085

1

b
a

27.7 ± 0.34
28.7 ± 0.33
29.9 ± 0.66
31.1 ± 0.57
31.3 ± 0.89

Photosynthesis and transpiration rates were measure with a Li-Cor 6400 portable photosynthesis meter.

2

Relative chlorophyll content was measured with a handheld Minolta SPAD-502 meter.

62

31.1 ± 1.01
39.0 ± 0.59

b

31.3 ± 0.95
33.1 ± 0.73
35.7 ± 0.40
36.2 ± 0.63

a
a

b
b

37.5 ± 1.10
42.2 ± 0.62

a

a

b

39.3 ± 1.11

a

31.9 ± 1.12
39.3 ± 0.66

a

b

32.1 ± 1.08

a

36.3 ± 0.43
37.0 ± 0.55
37.8 ± 0.42
41.3 ± 0.61
42.9 ± 0.61
40.9 ± 1.04

Table 2.9: Results from analysis of variance of effects of ash species and treatment on macro
nutrients (percentage dry weight) in foliage of black ash, blue ash, green ash, white
ash and Manchurian ash trees sampled in July 2010 or July 2011.
F Value

df (num, den)
2010

P Value

Species

28.00

4, 75

<0.001*

Treatment
Species × Treatment

3.94
0.46

2, 75
8, 75

0.024*
0.877

Species

10.58

4, 75

<0.001*

Treatment
Species × Treatment

6.58
2.02

2, 75
8, 75

0.002*

Potassium

Species
Treatment
Species × Treatment

14.96
1.34
0.76

4, 75
2, 75
8, 75

<0.001*
0.268
0.636

Calcium

Species
Treatment
Species × Treatment

32.71
1.33
1.01

4, 75
2, 75
8, 75

<0.001*
0.270
0.437

Magnesium

Species
Treatment
Species × Treatment

29.17
0.77
1.09

4, 75
2, 75
8, 75

<0.001*
0.466
0.380

Nutrient
Nitrogen

Phosphorous

Effect

63

0.055

Table 2.9 (cont’d)
2011
Species

43.34

2, 41

<0.001*

Treatment

3.29

2, 41

0.047*

Species × Treatment

3.50

4, 41

0.015*

Species

26.48

2, 41

<0.001*

Treatment
Species × Treatment

6.53
1.93

2, 41
4, 41

0.003*
0.123

Potassium

Species
Treatment
Species × Treatment

8.26
0.89
0.26

2, 41
2, 41
4, 41

<0.001*
0.418
0.904

Calcium

Species
Treatment
Species × Treatment

32.88
0.22
0.72

2, 41
2, 41
4, 41

<0.001*
0.807
0.586

Magnesium

Species
Treatment
Species × Treatment

31.09
0.10
0.18

2, 41
2, 41
4, 41

<0.001*
0.909
0.946

Nitrogen

Phosphorous

64

Table 2.10: Mean (±SE) percentage of macronutrients in foliage of black ash, blue ash, green ash, white ash and Manchurian ash trees
treated with paclobutrazol, fertilizer or left as untreated control sampled in July 2010 and July 2011. Different letters
indicate significant (α ≤ 0.05) differences in nutrient concentration among species (a,b,c) or treatments (x,y).
Ash Species

Nitrogen

Phosphorous

Potassium

Magnesium

Calcium

2010
Black

2.95 ± 0.125

Blue

1.94 ± 0.086

Green

2.87 ± 0.066

White

2.59 ± 0.050

Manchurian

2.69 ± 0.063

a
c

ab
b

ab

0.28 ± 0.027

ab

0.95 ± 0.122

c

0.17 ± 0.020
0.33 ± 0.031
0.23 ± 0.014
0.27 ± 0.014

c

1.01 ± 0.163

a

1.35 ± 0.104

bc
ab

1.61 ± 0.076

c

bc
ab

1.99 ± 0.045

a

b

0.30 ± 0.131

0.65 ± 0.061

a

b

0.23 ± 0.015

b

0.22 ± 0.013

b

0.29 ± 0.017

1.76 ± 0.120

bc

3.12 ± 0.165
1.60 ± 0.068
1.69 ± 0.070
2.11 ± 0.093

a
c
c

b

2011
0.13 ± 0.006

Blue
Control

1.72 ± 0.128

Fertilizer

2.19 ± 0.085

Paclobutrazol

1.66 ± 0.120

Fertilizer

2.42 ± 0.061

Paclobutrazol

2.46 ± 0.030

b

Fertilizer

2.53 ± 0.145

Paclobutrazol

2.55 ± 0.061

a

2.53 ± 0.126

a

1.32 ± 0.041

ab

b

0.20 ± 0.013

1.39 ± 0.052

b

x
x
x

0.22 ± 0.015
2.56 ± 0.061

0.59 ± 0.047

x

Manchurian
Control

b

y

0.17 ± 0.006
2.43 ± 0.113

1.07 ± 0.115

x

White
Control

c

a

1.56 ± 0.062

x
x
x

65

a

0.33 ± 0.019

b

1.63 ± 0.088

b

A

B

Figure 2.1: Percentage of biomass allocated to roots, leaves and aboveground woody material (shoots and bole) in blue ash, white ash
and Manchurian ash trees (A) and ash trees treated with paclobutrazol, fertilizer or left as untreated controls (B).

66

120

Tree Caliper (mm)

100
80
60
40
2

R =0.432
Y = -1448.15 + 47.95x
P < 0.001

20
0
0

1

2

3

4

5

Root Dry Weight (kg)
Figure 2.2: Relationship between root biomass (kg) and tree caliper (mm) of blue ash, white ash
and Manchurian ash trees harvested in October 2012 from a plantation in Okemos,
Michigan.

67

Figure 2.3: Biomass vector analysis of nitrogen for blue ash (diamonds), white ash (triangles)
and Manchurian ash (squares) trees treated with paclobtrazol (black, PB) or
fertilizer (gray, F). Dry weight is the dry weight of all leaves, twigs, boles and
roots.

68

Figure 2.4: Biomass vector analysis of phosphorous for blue ash (diamonds), white ash
(triangles) and Manchurian ash (squares) trees treated with paclobtrazol (black,
PB) or fertilizer (gray, F).

69

CHAPTER THREE
Effects of paclobutrazol and fertilization on host resistance and suitability of five
Fraxinus species to Agrilus planipennis

ABSTRACT
Agrilus planipennis Fairmaire is the most destructive forest insect to ever invade North
America. All Fraxinus spp. (ash) assessed to date are suitable hosts, but A. planipennis exhibits
interspecific and intraspecific host preferences. In conjunction with insecticide applications,
fertilizer and paclobutrazol (PB) treatments may be recommended to ensure ash trees remain
healthy. I assessed differences in host preference and suitability among 1) four North American
(Fraxinus nigra Marsh., F. quadrangulata Michx., F. pennsylvanica Marsh. and F. americana
L.) and one Asian ash species (F. mandshurica Rupr.) and 2) trees treated with PB, fertilizer or
left as untreated controls. Species effects overshadowed most treatment effects. In 2010, adult
beetle survival was highest when beetles were caged with black ash (53%), intermediate for
beetles caged with green, white and Manchurian ash (30%, 32%, or 30%, respectively) and
lowest (14%) for beetles caged with blue ash. When beetles were caged on intact blue ash and
white ash leaves, they consumed 42 and 31% more leaf area per beetle day than beetles caged on
excised blue ash and white ash leaves but survival was similar within species. In contrast, leaf
area consumed per beetle day was similar between beetles caged on intact or excised Manchurian
ash leaves, but survival was 11% higher for beetles caged on excised Manchurian ash foliage
than for beetles caged on intact leaves from the same tree. In 2011, larval gallery density
2

(galleries per m ) was highest on black (235.9±36.41) and green ash (220.1±39.77), intermediate
on white ash (40.7±11.61), and lowest on blue and Manchurian ash (2.0±0.98 and 1.5±0.67,
respectively). Results indicate that blue ash and Manchurian ash trees are more resistant to A.
planipennis than black ash, green ash or white ash.
70

INTRODUCTION
Even though it is a secondary pest in its native range (Yu 1992, Gao et al. 2004), Agrilus
planipennis Fairmaire (emerald ash borer) is the most destructive forest insect to ever invade
North America (Aukema et al. 2011, Kovacs et al. 2011). Tens of millions of ash trees in
Michigan alone have been killed by A. planipennis since it was identified in 2002 and to date,
infestations have been found in 19 states and two Canadian provinces (Cappaert et al. 2005,
EAB.info 2013). An estimated 8 billion ash trees occur in forested settings across the United
States (EAB.info 2013). Annualized economic impacts of A. planipennis were recently
projected to exceed 900 billion USD by 2020 (Aukema et al. 2011, Kovacs et al. 2011). The cost
of treating or removing even 50% of the landscape ash trees in affected US cities will likely
exceed 10.5 billion USD by 2019 (Kovacs et al. 2011).
Ash species are evolutionarily diverse (Wallender 2008) and have varying levels of
inherent resistance to A. planipennis (Rebek et al. 2008, Duan et al. 2012, Tanis and McCullough
2012). The Asian ash species, Manchurian ash (Fraxinus mandshurica Rupr.) and Chinese ash
(Fraxinus chinensis Roxb.), are relatively resistant to A. planipennis and are attacked only when
stressed (Liu et al. 2003, Duan et al. 2012). In contrast, North American species, lack a coevolutionary history with A. planipennis (Bryant et al. 1994, Lieutier 2008, Gandhi and Herms
2010, Nielsen et al. 2011). This lack of evolved tree resistance and the absence of native
congeners (Drooz 1985) and natural enemies likely accounts for the widespread ash mortality
associated with A. planipennis (EAB.info 2012). While all ash species assessed to date can be
colonized, A. planipennis prefer some species over others (Anulewicz et al. 2007, Rebek et al.
2008, Limback 2010, Tanis and McCullough 2012). In northeast China, A. planipennis
preferentially attacked two North American species, green ash (Fraxinus pennsylvanica Marsh.)

71

and velvet ash (Fraxinus velutina Torrey) rather than native Manchurian or Chinese ash (Duan et
al. 2012). In a common garden in southeast Michigan, white ash (Fraxinus americana L.), green
ash and a black ash (Fraxinus nigra Marshall)

Manchurian ash hybrid (‘Northern Treasure’)

all succumbed to A. planipennis infestation before Manchurian ash (Rebek et al. 2008). Among
North American ash species, black ash and green ash are highly preferred by A. planipennis
(Cappaert et al. 2005, Limback 2010) while white ash is considered moderately preferred
(Anulewicz et al. 2007). To date, blue ash (Fraxinus quadrangulata Michx.) appears to be the
most resistant North American species (Anulewicz et al. 2007, 2008, Pureswaran and Poland
2009). In areas where blue ash and white ash co-occurred in southeast Michigan, 60-70% of
blue ash of all diameter classes survived the A. planipennis invasion compared to <5% of white
ash (Tanis and McCullough 2012).
Intraspecific A. planipennis host preference is also apparent (McCullough et al. 2009a,
2009b). In Asia, A. planipennis functions as a secondary pest, typically attacking stressed ash
trees (Lyons et al. 2009). In North America, it can colonize both healthy and stressed ash trees
(Poland and McCullough 2006), but will preferentially attack stressed trees (McCullough et al.
2009a, 2009b). Larval development in stressed trees takes one year, but larvae require two years
to develop on healthy trees (Siegert et al. 2010, Tluczek et al. 2011).
Since the North American discovery of A. planipennis in 2002, researchers have been
working to identify or improve control options for valuable ash landscape trees. Systemic
insecticides, applied via trunk injection or soil applications are most commonly used (Herms et
al. 2009, McCullough et al. 2011). Some researchers and tree care professionals also
recommend fertilization to ensure ash trees remain healthy and receive adequate nutrients (Rebek
and Smitley 2005). Although enhanced vigor could enable trees to resist and/or tolerate A.

72

planipennis infestation, there have been no studies to assess the effects of fertilizer on A.
planipennis host preference or host suitability.
Effects of fertilization on urban trees have been widely disputed since the early 1920’s
(Miller 1998, Marshall et al. 2003, Herms 2002, Ferrini and Baietto 2006). Fertilization has been
promoted as a means to increase tree resistance (reviewed by Waring and Cobb 1992), while
Raupp et al. (2010) stated “fertilization almost always decreases tree resistance to herbivores”.
Fertilizer application can affect growth rates, volatile production, and in some plants, the
production of chemical defenses (Gouingené and Turlings 2002). Growth and development of
phytophagous insects are often limited by nitrogen (N) availability (Scriber and Slansky 1998)
and most insects prefer N rich over N limited foliage (Reviewed by Herms 2002, Schoonhoven
et al. 2005). However, fertilization may also enhance defensive chemical production and
decrease herbivore survival or reproduction (Coley et al. 1985, McCullough and Kulman 1991).
Whether fertilization affects ash resistance or A. planipennis host selection behavior has not been
examined.
Many tree care companies promote plant growth regulator treatments to reduce tree
growth and ameliorate stress (Chaney 2005a, 2005b). Paclobutrazol (PB), a triazole gibberellin
inhibitor, is considered one of the most potent growth inhibitors in its class (Yadav 2005). It
reduces aboveground plant growth by inhibiting sterols (Dalziel and Lawrence 1984) and blocks
the gibberellin biosynthesis pathway (Raese and Burts 1983). Changes in gibberellin production
alter source - sink relationships within trees by reducing aboveground growth and increasing
stored carbohydrates (Ramina and Tonutti 1985, Davis and Shankhla 1987). Paclobutrazol,
commonly applied to landscape trees as a soil drench (Chaney 2005a, 2005b), can persist in soil
up to three years (Yadav 2005) and there is typically a one season lag between application and

73

measurable changes in growth (Bai et al. 2004). Although it was originally used by utility
companies to reduce pruning and maintenance of trees and shrubs in right of ways (Mann 1995,
Chaney 2005a, 2005b), PB is commonly marketed in the landscape industry as a means to
increase tree vigor (Chaney 2005a, 2005b). In addition to reducing growth, PB increases
tolerance of cold (Haas 2012), drought (Davis and Shankhla 1987, Marshall et al. 2000, Percival
and AlBalushi 2007) and salt (Haas 2012); chlorophyll concentration (Yadav 2005); stored
carbohydrates (Fletcher et al. 2000); root: shoot ratio; cell division (not cell elongation)
(Wiesman and Lavee, 1994); and fine root biomass (Fletcher et al. 2000, Chaney 2005a, 2005b,
Haas 2012).
In a concurrent study, I found radial growth, biomass allocation, relative chlorophyll
content and nutrient concentrations were affected by fertilization and PB treatments. One year
after application, trees treated with PB had reduced radial growth and biomass allocation shifted
from aboveground to belowground (Tanis, Chapter 2). Blue ash and Manchurian ash trees were
relatively sensitive, and responded to PB treatment in almost every parameter assessed. In
contrast, white ash was relatively unaffected, which is consistent with previous reports (Sterrett
and Tworkoski 1987, Bai et al. 2004). Trees rarely responded to fertilization with the exception
of blue ash; fertilized trees had higher relative chlorophyll content and nutrient concentrations
than control or PB trees (Tanis, Chapter 2).
Although inter- and intraspecific A. planipennis host preference is apparent, it is not clear
whether the susceptibility of ash species is driven by adult beetles, that select trees for feeding
and oviposition, or by the ability of larvae to feed and develop, or both. In this study, I
hypothesized that blue ash and Manchurian ash trees will be less suitable hosts for adult A.
planipennis than black ash, green ash and white ash and that A. planipennis larval densities will

74

reflect host suitability or adult beetle host preference. I also hypothesized trees treated with PB
will be more resistant to A. planipennis infestation than fertilized or control trees. In oak, trees
that have moderate to high starch reserves (and theoretically more energy for the production of
defensive chemicals) tend to resist attack by two-lined chestnut borer (Agrilus bilineatus Weber)
(Dunn et al. 1987), a North American A. planipennis congener. Our concurrent study indicated
that PB shifts biomass allocation from aboveground to belowground which suggests PB could
increase resources available for defense against A. planipennis.

75

METHODS
Ash Plantation
In 2007, 225 ash trees were purchased from commercial nurseries and planted (2.4 m
apart) in a randomized complete block design at the Michigan State University Tree Research
Center (TRC) in Okemos, Michigan, Ingham County. The plantation consisted of 45 trees of
each species: black ash, blue ash, green ash, white ash and Manchurian ash. Each block
consisted of one tree of each species. Trees were grown as either bare root (black ash and
Manchurian ash) or balled and burlapped stock (green ash, white ash and blue ash). Tree size
varied among species (Tanis, Chapter 2). All trees were shipped from outside of A. planipennis
quarantine zones in cooperation with state and federal agencies and were not infested when they
were planted. During the 2008-09 winter, 16 blue ash trees died from non- A. planipennis causes
and were replaced with bare root trees in spring 2009. Throughout the experiment, trees were
watered twice per week or as needed with drip irrigation. Trees were fertilized in July (20062009) with a top dressing of Harrel’s Pro-Blend with Micros (19-5-10) (Harrell’s, Lakeland, FL,
USA) at a rate of 70 g (5.5 ounces N) per tree.
®

In May 2006 and 2007, trees were sprayed with a foliar application of Tempo (Bayer
Environmental Science, New Jersey, USA) to prevent A. planipennis infestation. In May 2008 2010 rather than apply insecticide, tree boles were wrapped with tree wrap (Jobe's TreeWrap Pro,
Easy Gardener, Inc., Waco, Texas, USA) over laid with screen mesh to deter A. planipennis
infestation. Tree wrap was removed every September.
On 26 May 2010, seven trees of each species were randomly assigned to one of the
following treatments: control, fertilizer or PB (35 trees per treatment). Control trees received no
chemical applications. Fertilized trees received 350 g (a higher rate than above) of Harrell’s Pro76

Blend with Micros (19-5-10) (Harrel’s, Lakeland, FL, USA) sprinkled at the base of each tree.
®

ShortStop (Paclobutrazol 22.3%, Zhejiang Tide Crop Science Co., LTD, Irvine, CA, USA) was
applied as a soil drench. Diluted product was applied evenly according to label instructions (200
ml diluted product, 4 grams a.i., per 2.54 cm DBH) in a shallow trench, approximately 12 cm
deep, around the base of each tree. All trees were protected with wrap during the summer of
2010 as described above.
Trees fertilized in 2010 received a second application on 23 May 2011. Because PB
applications typically persist in soils for two to three years, PB trees did not receive additional
treatments. In 2011, trees were left exposed to wild A. planipennis populations.
Adult A. planipennis
Adult A. planipennis used for this study were reared in the laboratory from infested ash
®

logs harvested in autumn 2009 or 2010. Logs were cut to 60 cm, sealed with Bailey’s seal
(Bailey’s, Laytonville, CA USA) and placed into coolers maintained at 4º C. Logs were
removed from cold-storage approximately three weeks before experiments began and placed in
rearing tubes (ambient temperature). After two weeks, rearing tubes were checked daily. As
beetles emerged, they were collected, separated by sex and placed directly into bioassays.
Leaf feeding bioassays
In July 2010 and 2011, I assessed effects of ash species and treatments on adult A.
planipennis foliage consumption and survival using excised (2010 and 2011) or intact (2011
only) leaves from each tree. In June 2010, a single sun-exposed leaf was removed from each
tree, sealed in a plastic bag, placed in a cooler and brought back to the laboratory. Upon arrival,
a section with 4-5 leaflets still attached to the petiole was removed (whole leaves were too large
®

to fit into petri dishes), weighed and scanned with a flatbed scanner (Epson Perfection 4490
77

Photo, Epson America, Inc., Longbeach, CA) to determine pre-feeding leaf weight and area.
®

Leaf area was assessed using WinFOLIA software (Regent Instruments, Quebec, Canada).
After obtaining initial measurements, leaf petioles were inserted into a water pick, to slow
desiccation, and placed in a Petri dish.
Four newly emerged beetles (two males and two females) were immediately added to
each Petri dish. Petri dishes with beetles were maintained in growth chambers with 16:8 hr light:
dark, 75% humidity and 25°C. Beetles were allowed to feed for four days and the number of
surviving A. planipennis was recorded daily. On the fourth day, the leaf was removed from the
Petri dish, re-weighed and scanned to quantify leaf weight and area consumed. Leaf weight and
area consumed were standardized using “beetle days”. On day four, beetle days were calculated
for each sample by tallying the number of surviving beetles from each day. For example, if all
beetles survived in the Petri dish, the number of beetle days would be 16 (four beetles alive on
each of four days). Leaf weight and area consumed per beetle day were calculated by dividing
2

the leaf weight (g) or area consumed (cm ) by the number of beetle days. Bioassays were
repeated in June 2011.
In spring 2011, I developed an inexpensive, novel cage that allowed me to cage beetles
with intact leaves still attached to trees. Cages consisted of two foam dessert plates (15 cm
diameter, Gordon Food Service, Wyoming, MI, USA). To allow airflow and sunlight to
penetrate the cage, I removed a 10 cm diameter section from the center of each plate and glued a
piece of screen mesh over the center to prevent beetles from escaping. I then stapled two plates
together, top to top. A small opening, approximately 10 cm long, was left to allow for the
addition of beetles and to enable me to move cages from leaf to leaf. Cages were attached to

78

undamaged leaves, beetles were added and the opening along the edge was sealed with three
small binder clips. Cages were secured to trees with twine to reduce wind damage.
In July 2011, a cage containing four newly emerged beetles (two males and two females)
was placed on a single undamaged sun-exposed leaf of each tree. The number of surviving
beetles was recorded daily for four days then the leaf was removed, sealed in a plastic bag,
returned to the lab and scanned with a flatbed scanner. Each leaf scan was printed twice. One
copy was used to quantify pre-feeding leaf area, the other to quantify post-feeding leaf area. To
determine pre-feeding area, I carefully filled in the area of the leaf that was missing as a result of
A. planipennis feeding. Post feeding images were not altered. Both images were re-scanned
with WinFOLIA® software and area consumed was determined by subtracting post-feeding area
2

2

(cm ) from pre-feeding area (cm ). Leaf area consumed per beetle day was recorded.
Mortality Bioassays
After emergence, adult A. planipennis typically feed for two weeks before they begin
mating. In 2010 and 2011, bioassays from (excised and intact) were continued for 10 additional
days to determine how many beetles would survive the maturation period. After the first four
days, beetles in Petri dishes were given a new excised leaf from the same tree every other day for
the 10 remaining days (14 days total). Beetles and cages in the plantation (2011 only) were
moved to a new leaf on the same tree every other day for 10 additional days (14 days total).
Number of surviving beetles was recorded when foliage was changed or cages were moved.
Larval Gallery Density
Local populations of A. planipennis were very high and despite our efforts to protect trees
from infestation, 66% of black ash trees and 71% of green ash trees were killed in 2010. By
August 2011, all but one of the remaining black ash and green ash trees had severe canopy
79

dieback (>80%). I harvested and debarked dead black ash (14 of 21) and green ash (15 of 21)
trees in May 2011 to ensure they did not rot over the summer. Remaining trees were left
exposed to the local wild A. planipennis populations during the 2011 growing season, then were
harvested and debarked in the fall. For each tree, I assessed galleries made by the 2010 and 2011
A. planipennis larval cohorts.
On trees debarked in spring 2011, I counted the number of exit holes created by larvae
from the 2009 cohort, but because trees were dead and many were starting to desiccate, I could
not record with confidence the number or state of larvae from the 2009 cohort. Larval condition
(dead or alive), larval stage (first, second and third instar larvae were classified as early instars,
fourth instar and prepupae larvae were classified as late instars) were recorded for larvae from
the 2010 cohort. I also assessed woodpecker or parasitoid attacks on the 2010 larval cohort.
Parasitoid attacks were recorded when an Atanycolus spp. cocoon was present in the A.
planipennis larval gallery or when an Atanycolus spp. larva was attached to an A. planipennis
larva. All of the larvae recorded from trees debarked in the spring of 2011 were from the 2010
larval cohort.
For trees debarked in fall 2011, I distinguished between galleries made by the 2010 and
2011 cohorts using methods described by Tluzcek et al. (2011). Larvae from the 2010 cohort
were assessed for larval condition and larval stage as described above. Holes made by
woodpeckers removing EAB larvae in 2010 were differentiated from holes made in 2011 by the
state of the bark surrounding the hole. Holes created by birds in 2011 were still green and fresh
when trees were harvested. Atanycolus spp. cocoons from 2010 contained an adult emergence
hole, whereas 2011 cocoons still contained an Atanycolus spp. pupae. Larval cohort was

80

assigned for parasitized larvae or larvae removed by woodpeckers based on the appearance of the
gallery (Tluzcek et al. 2011).
Statistical Analysis
All data were analyzed using SAS statistical software (SAS Institute, Inc. 2003). Results
from 2010 are presented for all five ash species, while 2011 results are presented for blue ash,
white ash and Manchurian ash only. Assumptions of normality were tested with residual plots
and the Shapiro-Wilk test (Shapiro and Wilk 1965). Leaf weight and leaf area consumed per
beetle day were normalized using log transformations (Ott and Longnecker 2001). Two-way
ANOVA was performed to evaluate effects of ash species, treatments and interactions between
the two factors on changes in leaf weight (2010 and 2011) and leaf area consumed per beetle day
(2010). In 2011, three-way ANOVA was used to evaluate effects of ash species, treatments, leaf
type (intact or excised) and interactions of the three factors on leaf area consumed per beetle day.
In 2011, repeated measures ANOVA was used to evaluate effects of ash species, treatments, leaf
type, days after treatment and the interactions on beetle survival.
The number of exit holes, parasitoid and woodpecker attacks, density of living and dead
larvae, and total gallery density, could not be normalized using transformations. Friedman's
nonparametric two-way ANOVA was used to analyze all gallery data (number of total, alive and
dead larvae, adult A. planipennis exit holes, woodpecker attacks and parasitized larvae). When
species or treatment effects were significant (α ≤ 0.05), I used Fisher’s non-parametric protected
least significant difference (LSD) multi-comparison test with Tukey’s adjustment to determine
differences among means (Ott and Longnecker 2001).

81

RESULTS
Leaf Area and Weight Consumed per Beetle Day
2010 – Excised Leaves
Area consumed per beetle day of excised leaves (2010 and 2011) varied among ash
species (Figures 3.1A). In 2010, area of excised leaves consumed per beetle day was affected by
species (F=14.33; df=4,89; P<0.0001) but not treatment (P=0.63) or the interaction (P=0.31). In
2010, area of excised leaves consumed per beetle day of white ash and Manchurian ash was over
twice as high as the area of excised leaves consumed per beetle day of blue ash (Fig. 3.1A). Leaf
weight consumed per beetle day (excised leaves) did not differ among species (P=0.54),
treatments (P=0.43) or their interactions (P=0.81).
2011 – Excised and Intact Leaves
In 2011, I caged beetles on intact or excised foliage for the feeding and survival
bioassays. Leaf area consumed per beetle day varied by species (F=45.01; df=2, 111;
P<0.0001), leaf type (excised or intact) (F=52.31; df=1; 111; P<0.001) and the species
leaf type interaction (F=9.87; df=2, 111; P=0.001) but not treatment (P=0.06). Species
treatment (P=0.38), treatment

leaf type (P=0.17) and species

treatment

leaf type (P=0.19)

interactions were also not significant. Beetles caged with excised leaves from blue ash trees
consumed less foliage than beetles caged with any other foliage type (Fig. 3.1B). Beetles
consumed 42 and 31% more (P<0.001) leaf area when caged with intact white ash and blue ash
leaves, respectively, than beetles caged with excised white ash and blue ash leaves (Fig. 3.1B).
Leaf area consumed per beetle day was similar (P=0.07) for excised and intact Manchurian ash
foliage (Fig. 3.1B). As in 2010, leaf weight consumed per beetle day (excised leaves only) did
not differ among species (P=0.66) treatments (P=0.36), or the interaction (P=0.87).

82

Adult A. planipennis Mortality
In 2010, beetle survival was affected by species, day, species

treatment and species

day interactions (Table 3.1). In 2011, I caged beetles on intact, as well as, excised leaves. Beetle
survival was affected by species, day and species
treatment

leaf type and species

treatment

day, species

leaf type (excised or intact),

leaf type interactions (Table 3.1).

In 2010, at least 38% more beetles survived for 14 days when caged with excised black ash
leaves than beetles caged with excised leaves from the other four ash species (Table 3.2).
Species effects were first apparent on day six when 48 and 38% of beetles caged with excised
blue ash or Manchurian ash leaves were dead, respectively, compared to 18, 35 and 23% of
beetles caged with excised black ash, green ash or white ash foliage (Fig. 3.2A). At the end of
the 14 day trial, survival was highest (53%) for beetles caged with excised black ash,
intermediate for beetles caged with excised green ash (30%), white ash (32%) or Manchurian ash
(30%) foliage, and lowest (15%) for beetles caged with excised blue ash foliage (Table 3.2).
In 2011, six days after emergence, 85% of beetles caged with white ash and Manchurian
ash foliage (intact + excised) were alive compared to 65% of beetles caged with blue ash foliage.
This pattern was consistent for the remaining eight days of the bioassays (Fig. 3.2B).
Throughout the 14 day trial, survival rates were similar between beetles caged with white ash
and Manchurian ash leaves (Table 3.3). At least 73% of beetles caged with white ash or
Manchurian ash foliage survived the two week maturation period. In contrast, only 33% of
beetles caged with blue ash foliage survived 14 days (Fig. 3.2B).
2010 – Excised Leaves
Within species, treatment effects on survival of beetles caged with excised foliage were
variable (Table 3.4). There were no treatment effects on survival when beetles consumed

83

excised black ash leaves. For beetles caged with excised blue ash leaves, survival was highest on
leaves from control trees (72%) and lowest on leaves from fertilized (55%) or PB trees (53%)
(Table 3.4). Among beetles caged with excised green ash or Manchurian ash leaves, survival
was higher for A. planipennis caged with leaves from PB trees (74% and 74%) than beetles
caged with leaves from fertilized (63% and 69%) or control (73% and 58%) green ash or
Manchurian ash trees, respectively (Table 3.4). Beetles caged with excised white ash leaves had
higher survival when caged with leaves from fertilized (81%) or PB (76%) trees than beetles
caged with leaves from control (60%) trees, respectively.
2011 - Excised Leaves
Treatment effects on beetle survival were evident when beetles were caged with excised
foliage, but results varied among species (Table 3.4). Survival was unaffected by treatment
when beetles were caged with excised blue ash or Manchurian ash leaves (Table 3.4). However,
when beetles were caged with excised leaves from white ash PB trees, survival was 11% higher
than when beetles were caged with leaves from white ash control or fertilized trees (Table 3.4).
2011 – Intact Leaves
Treatment and species effects on survival were also variable among beetles caged on
intact leaves (Table 3.4). When beetles were caged on intact leaves on blue ash PB and control
trees, survival was 15 and 18% higher, respectively, than when beetles were caged on intact
leaves on fertilized blue ash trees. In contrast, survival of beetles caged with intact leaves on PB
white ash trees was 11% lower than survival of beetles caged on intact leaves on control trees.
Beetle survival on intact white ash leaves from fertilized trees (85%) did not differ from beetle
survival on intact leaves from PB (78%) or control (88%) white ash trees. Survival was

84

unaffected by treatment when beetles were caged on intact leaves from Manchurian ash trees
(Table 3.4).
2011 - Excised vs. Intact Leaves
Beetle survival was similar between beetles caged with excised foliage from blue ash
control and PB trees and intact foliage from control or PB blue ash trees (Table 3.4). Among
beetles caged with white ash foliage, survival was higher for beetles caged with excised foliage
from PB trees than for beetles caged with all other types of white ash foliage (Table 3.4). Across
all three treatments, beetles caged with excised Manchurian ash foliage had higher survival than
beetles caged with intact Manchurian ash foliage (Table 3.4).
Larval Gallery Density
A total of 4,016 larval galleries were examined in this study when trees were debarked in
2011. Only, 14% of the galleries produced an adult A. planipennis as evidenced by an adult
emergence hole. Among all ash species, 22% of galleries contained live larvae, 8% of these
were early instar larvae (first, second or third instars) and 92% were late instar larvae (fourth
instar or prepupae). Only 11% of galleries contained dead larvae, 61% of these were early
instars and 39% were late instars. Across all species, 54% of larvae were killed by either
woodpeckers or Atanycolus spp. I also assessed the densities of dead and living larvae, adult A.
planipennis exit holes and evidence of woodpecker and parasitoid attacks by species (Table 3.5).
2

Larval density varied among species (χ =67.53; df=4,104; P<0.001) but not treatments
(P=0.72) or the interaction (P=0.93). Black ash and green ash trees were heavily infested (Fig.
3.3) and had larval densities so high (235.9 ± 36.41 and 220.1 ± 39.77, respectively) that 18
black ash and 11 green ash trees failed to leaf out in 2011. Larval densities on blue ash and
Manchurian ash trees were significantly lower than on black ash or green ash and ranged from 0
85

2

to 15 and 0 to 13 galleries per m , respectively (Fig. 3.3). Fifteen blue ash and 13 Manchurian
ash trees had no A. planipennis galleries. Larval density on white ash trees was moderate (Fig.
2

2

3.5) ranging from 0 to 178 galleries per m . Three white ash trees contained > 89 larvae per m ,
a density high enough to result in tree death (McCullough and Siegert 2007), but five trees had
2

<10 galleries per m and two trees had no evidence of A. planipennis infestation.
2009 Larval Cohort
Black ash and green ash trees were infested by the 2009 larval cohort. Adult emergence
holes from the 2009 larval cohort comprised 15 and 6% of the total larval galleries found on
black ash and green ash trees, respectively (Table 3.6).
2010 Larval Cohort
Among black ash trees, 5% of galleries made by the 2010 larval cohort were associated
with an adult emergence hole, 21% contained live larvae and 74% were associated with dead
larvae. Woodpecker predation and Atanycolus spp. parasitism accounted for 62% of the dead
larvae (Table 3.6). There were no larvae from the 2010 cohort on blue ash trees (Table 3.6).
Among green ash trees, 14% of galleries made by the 2010 larval cohort contained an adult
emergence hole, 17% contained live larvae and 69% were associated with dead larvae. Among
the dead larvae, 56% were killed by either woodpeckers or Atanycolus spp. (Table 3.6). White
ash trees had 79% fewer galleries from the 2010 larval cohort than black ash or green ash trees.
Among the 175 galleries found on white ash trees, 7% contained an exit hole, 28% contained live
larvae and 65% were associated with dead larvae. While woodpecker predation accounted for
91% of the dead larvae on white ash trees, there was no evidence that Atanycolus spp. parasitized

86

any of the larvae from the 2010 cohort (Table 3.6). Manchurian ash trees contained no larvae
from the 2010 larval cohort (Table 3.6).
2011 Larval Cohort
In 2011, only seven black ash and six green ash trees remained in the plantation for
infestation by the 2011 larval cohort. Among the 304 galleries examined on black ash trees, 8%
contained live larvae and 92% were associated with dead larvae. Of the dead larvae, 96% were
killed by woodpeckers or Atanycolus spp. (Table 3.6). Among blue ash trees, 41% of larval
galleries contained live larvae and 59% were associated with dead larvae. Woodpecker
predation and parasitism by Atanycolus spp. accounted for 50% of the dead larvae (Table 3.6).
Among the 952 galleries found on green ash trees, 39% contained live larvae and 61% were
associated with dead larvae, of these, 85% were killed by woodpeckers or Atanycolus spp. (Table
3.6). The number of larval galleries on white ash trees increased 45% from 2010 to 2011. Of the
319 galleries assessed, 38% contained live larvae and 62% contained dead larvae. Contrary to
2010 when parasitism by Atanycolus spp. did not occur, 79% of the dead larvae from the 2011
cohort were parasitized by Atanycolus spp. and 21% were killed by woodpeckers (Table 3.6).
Among the 10 galleries found on Manchurian ash trees, 40% were alive and 60% were dead.
Neither woodpeckers nor Atanycolus spp. killed any of the larvae on Manchurian ash trees
(Table 3.6).

87

DISCUSSION
In this study, ash species was the main driver behind A. planipennis adult host suitability
and larval gallery density. Several studies indicate A. planipennis adults prefer some ash species
over others, but the mechanisms underlying preference are not fully understood. In a study using
excised leaves and caged beetles, A. planipennis spent less time on and consumed less area of
Manchurian ash and blue ash leaves when given a choice between them and black ash, green ash
and white ash leaves (Pureswaran and Poland 2009). In my no choice bioassays, A. planipennis
caged with Manchurian ash leaves consumed as much or more leaf area as beetles caged with
black ash, green ash or white ash leaves. In a concurrent study, I determined that foliar N
concentration of Manchurian ash was similar to black ash, green ash and white ash (Tanis,
Chapter Two). Results of previous studies suggest foliar N concentrations might influence A.
planipennis adult host selection (Pureswaran and Poland 2009, Chen et al. 2011) because N is
often the limiting nutrient for folivorous insects (Scriber and Slansky 1981). Although A.
planipennis adults may vacate Manchurian ash leaves in favor of other available species
(Pureswaran and Poland 2009), our foliar nutrition data (Tanis, Chapter 2) suggests they are
probably not reacting to N levels in Manchurian ash foliage. In contrast to Manchurian ash, blue
ash foliar N concentrations were consistently lower than in the other four ash species (Tanis,
Chapter 2), but A. planipennis adults did not exhibit a compensatory feeding response as might
be expected (Simpson and Simpson 1990). Beetles caged with blue ash foliage consumed less
leaf area and died at higher rates than beetles caged with the other ash species. This pattern
suggests blue ash foliage may contain compounds that are toxic or trigger an A. planipennis
antifeedant response.

88

Induced foliar defenses can be activated in a variety of ways including leaf excision
(Schmelz et al. 2001) and insect feeding (Crook and Maestro 2010). Previous assessments of
adult A. planipennis feeding, host preference and survival have typically used captive A.
planipennis caged with excised foliage (Chen and Poland 2009a, 2009b, Pureswaran and Poland
2009). Removing ash leaves from trees, however, could trigger localized defensive responses
within leaves or short-circuit inducible defenses in trees responding to A. planipennis feeding. In
2011, beetles caged with excised leaves consumed less blue ash and white ash leaf area than
beetles caged with intact leaves on blue and white ash trees. In my feeding bioassays, some
excised leaves became desiccated even though they were kept in a carefully controlled
environment. I suggest leaf area consumed per beetle day was low when beetles were caged
with excised leaves simply because leaf quality and moisture were low. Effects of leaf excision
on beetle survival were highly variable among species and treatments and no consistent patterns
were apparent. Leaf area consumed was similar between beetles caged with excised Manchurian
ash leaves and beetles caged on intact Manchurian leaves, but A. planipennis survival was
consistently higher across treatments when beetles were caged on intact Manchurian ash leaves
than when caged with excised Manchurian ash leaves from the same tree.
In North America, A. planipennis attack only ash trees (Anulewicz et al. 2007, 2008,
Rebek et al. 2008). To locate suitable hosts, A. planipennis process a complex combination of
visual and olfactory cues that likely act as both attractants and repellents (Rodriguez-Saona et al.
2006, Pureswaran and Poland 2009). Host tree attributes (stress and species) not only influence
adult behavior, but can also have important impacts on larval distribution, development and
survival. For example, most larvae require two years to develop on healthy trees while larvae

89

develop in one year on stressed or heavily infested trees (McCullough and Siegert 2007, Tluczek
et al. 2011).
A combination of biotic and abiotic factors influence host suitability and subsequently
female feeding and oviposition preferences. In my study, A. planipennis adults consumed less
leaf area and fewer beetles survived (14% in 2010, 33% in 2011) when caged with blue ash
leaves (excised or intact) than when caged with black ash, green ash, white ash or Manchurian
ash leaves. This is consistent with a study by Pureswaran and Poland (2009) that indicated A.
planipennis will avoid blue ash leaves when given a choice between blue ash and black ash,
green ash or white ash leaves. Given the unsuitability of blue ash leaves as an adult food source,
it is perhaps not surprising that 71% of blue ash trees in our plantation had no larval galleries. In
fact, larval density on blue ash, a North American species, was similar to that of Manchurian ash,
an Asian ash species that shares an evolutionary history with A. planipennis.
While host suitability might explain why larval densities on blue ash trees were low (2.0
2

± 0.98 galleries per m ), it does not explain differences in gallery density among the other
species where relationships among leaf area consumed, beetle survival and larval gallery
densities were inconsistent. In 2010, survival was highest for beetles caged with excised black
ash leaves (53%). Beetles caged with excised white ash and Manchurian ash leaves consumed
more leaf area than beetles caged with black ash leaves, but survival was 42% lower. In 2011,
beetles caged with intact white ash leaves consumed more leaf area than beetles caged with
intact Manchurian ash leaves (I did not perform bioassays on black ash or green ash trees in 2011
due to poor tree health) but beetle survival was high (at least 73%) regardless of whether beetles
were caged with white or Manchurian ash leaves. If oviposition choice were driven by adult host
suitability, I might expect larval densities to be similar among black ash, green ash, white ash
90

and Manchurian ash because adult survival was relatively similar among these host species.
Instead, gallery densities on black ash and green ash trees were at least 82 and 99% higher than
gallery densities on white ash and Manchurian ash trees, respectively. While this does not
negate the hypothesis that female host preference is driving larval densities, it does imply that
females are making oviposition choices unrelated to the suitability of ash species as a food
source.
When removing the bark from ash trees, I was extremely careful to examine individual
larvae and galleries. On black ash and green ash trees, larval densities were so high (235.9 ±
2

36.41 and 220.1 ± 39.77 galleries per m , respectively) that bark fell away from the bole when I
was removing trees from the plantation. In contrast, only 17 larvae (41% alive, 29% dead) were
found on the blue ash trees, and 10 larvae (40% alive, 60% dead) were found on the Manchurian
ash trees. During the debarking process, I examined the outer sections of bark and phloem to
determine if neonate larvae were dying before they were able to penetrate the cambium. I found
three first instar larvae on Manchurian ash that had died before they reached the cambium but the
two early instar larval cadavers found on blue ash trees had died as third instars. It is important
to note I was unable to find evidence (eggs, tissue discoloration or neonate cadavers) that female
A. planipennis oviposit on blue ash and Manchurian ash trees at rates similar to green ash and
black ash. I therefore suspect A. planipennis females are avoiding blue ash and Manchurian ash
trees for oviposition. Among the dead larvae, I was unable to discern the cause of death but
there was no evidence of parasitism or woodpecker predation. It is possible neonate larvae were
killed by toxins as they began to chew into tree bark, but there were so few larvae on any of the
Manchurian ash trees that further assessment using artificially positioned eggs would be
necessary to confirm or negate this hypothesis.
91

There were no indications that any of the ash species responded to infestation by
callusing over tunneling larvae, a mechanism healthy North American paper birch (Betula
papyrifera Marsh.) trees use to defend phloem from a native A. planipennis congener, bronze
birch borer (Agrilus anxius Gory) (Nielsen et al. 2011, Miller et al. 1991). In contrast, Duan et
al. (2010) reported 32.0 to 41.1% and 17.5 to 21.5% of experimentally established and wild A.
planipennis larvae, respectively, were killed by callus tissue produced by healthy green ash trees.
While they report tree resistance in the form of callusing over was the “major source of (A.
planipennis larval) mortality” in their study, it was not observed on any of the 4,016 galleries
examined in my study. Perhaps it was not observed because my trees were smaller than the ones
used in their study, but McCullough et al. have debarked hundreds of ash trees for a variety of
studies and callusing over of larvae has not been observed in any of the healthy ash trees they
have examined (D.G.M. unpublished data).
Adult A. planipennis often prefer to attack open grown trees (McCullough et al. 2009a,
2009b). Trees in our plantation were evenly spaced and most received full sun, as a result, tree
location within the plantation did not appear to hinder or encourage A. planipennis infestation.
Blue ash and Manchurian ash trees with zero galleries were located adjacent to black ash or
green ash trees with 35 to 125 galleries per tree. Within individual blocks (five trees, one of each
species), there was not a single case where black ash or green ash trees did not have higher
gallery densities than blue ash, white ash and Manchurian ash trees.
Use of PB to enhance resistance to insect pests has been evaluated for several insects on
an assortment of host plants. Enhanced insect resistance is typically attributed to an array of
secondary responses that reflect altered growth patterns, but the exact mechanisms of resistance
are rarely identified and resistance effects are variable and generally unpredictable. In paper

92

birch (Betula papyrifera Marshall), PB reduced radial growth and height and increased specific
leaf mass and foliar tannin concentrations the year after PB application but not the following
year. White-marked tussock moth (Orgyia leucostigma Smith) and gypsy moth (Lymantria
dispar Linnaeus) survival, however, was reduced two years after treatment, but not the year that
specific leaf mass and tannin concentrations increased (Chobadjian et al. 2011). In pear (Pyrus
spp.), PB reduced pear psylla (Cacopsylla bidens Sulc.) oviposition and survival and
development of nymphs. The authors originally hypothesized that differences in psylla
performance would result from changes in foliar N concentrations, PB treatment, however, had
no effect on foliar N concentrations, indicating that some other mechanism was acting on the
psylla populations (Shaltiel-Harpaz et al. 2010).
In my study, ash species effects on A. planipennis host preference and host suitability
overshadow most treatment effects. In my concurrent study, I showed that PB reduces
aboveground growth which in theory might have provided additional resources for defensive
chemical production but the results of my adult A. planipennis feeding and mortality bioassays
were highly variable and no clear patterns emerged.
The majority of black ash and green ash trees in the plantation were heavily infested the
first summer of our study and nearly all of them had died by August 2011. When I harvested the
2

majority of black ash and green ash trees in spring 2011, they had 240 and 89 galleries per m ,
respectively, made by the 2010 larval cohort. Because larval densities were so high during the
summer of PB application (2010), and because effects of PB typically do not become apparent
during the year of application, there were likely no PB treatment effects on A. planipennis larval
gallery densities for black ash and green ash. It is unfortunate that green ash trees were so
heavily attacked before PB effects could become apparent as green ash is the most popular ash
93

tree in North American landscapes and urban areas (MacFarlane and Meyer 2003). The opposite
scenario occurred with blue ash and Manchurian ash trees. Gallery densities on these species
were so low that treatment effects would have been irrelevant.
Even without the burden of A. planipennis infestation, urban ash trees experience an
array of environmental stresses (Poland and McCullough 2006). They are planted in confined
areas (parking lot islands and sidewalk cubes) and subjected to drought, salt stress, soil
compaction and nutrient deficiency (MacFarlane and Meyer 2005). Treating trees with PB in
confined locations could reduce growth and increase root mass (Tanis, Chapter Two) and
ultimately extend the period before trees outgrow available space. Healthy ash trees are less
attractive to A. planipennis adults (McCullough et al. 2009a, 2009b) and presumably, trees
treated with PB might have enhanced resistance to A. planipennis simply because they are
healthier than their untreated counterparts.
Trees in our plantation were not treated with insecticide for two years prior to and the two
years during our experiments, and yet almost 30% had no evidence of A. planipennis infestation.
Many blue ash (43%) and Manchurian ash (37%) trees were completely free of A. planipennis
galleries and 95% had <5 galleries (two blue ash trees had six galleries). In contrast, surrounding
black ash and green ash trees were heavily infested and dying despite efforts to carefully protect
them with physical barriers. My study confirms reports of blue ash and Manchurian ash inherent
resistance (Rebek et al. 2007, Anulewicz et al. 2007, Tanis and McCullough 2012), but further
investigation is necessary to identify resistance mechanisms.

94

Table 3.1: Results of repeated measures analysis of variance to assess effects of species,
treatment and day (2010) and species, treatment, day and leaf type (2011) on
survival when Agrilus planipennis adults were caged on excised black, blue, green,
white or Manchurian ash trees (2010) and excised or intact blue, white and
Manchurian ash trees (2011) treated with paclobutrazol, fertilizer or left as untreated
control.
F Value
23.03
0.11
10.94
114.79
1.91
0.86
0.76

Degrees of Freedom
2010
4, 874
2, 874
8, 874
9, 874
36, 874
18, 874
72, 874

126.13
2.36
0.27
62.11
10.66
0.42
0.17
0.18
26.32
4.20
6.32
0.58
0.99
0.59
0.45

2011
2, 999
2,999
4, 999
8, 999
16, 999
16, 999
32, 999
1, 999
2, 999
2, 999
4, 999
8, 999
16, 999
16, 999
32, 999

Effect
Species
Treatment
Species × Treatment
Day
Species × Day
Treatment × Day
Species × Treatment × Day

Species
Treatment
Species × Treatment
Day
Species × Day
Treatment × Day
Species × Treatment × Day
Leaf Type
Species × Leaf Type
Treatment × Leaf Type
Species × Treatment × Leaf Type
Day × Leaf Type
Species × Day × Leaf Type
Treatment × Day × Leaf Type
Species × Treatment × Day × Leaf Type

95

P Value
<0.001*
0.894
0.001*
<0.001*
0.001*
0.630
0.929

<0.001*
0.095
0.899
<0.001*
<0.001*
0.978
1.000
0.675
<0.001*
0.015*
<0.001*
0.796
0.460
0.893
0.997

Table 3.2: P values generated by repeated measures least significant difference tests used to
compare mean number of live Agrilus planipennis caged with leaves from black,
blue, green, white or Manchurian ash leaves over a 14 day period. Bioassays were
performed in June 2010. An “*” denotes significant differences in beetle survival
(α=0.05) between host species.

Ash Species

Black

Blue

Green

White

Day

Blue

1
2
3
4
6
8
10
12
14
1
2
3
4
6
8
10
12
14
1
2
3
4
6
8
10
12
14
1
2
3
4
6
8
10
12
14

P Values
Green
White

0.868
0.856
0.820
0.237
<0.001*
<0.001*
<0.001*
<0.001*
<0.001*

0.869
0.742
0.610
0.683
0.028*
0.012*
<0.001*
<0.001*
0.005*
1.000
0.878
0.452
0.107
0.081
0.120
0.478
0.041*
0.020*

96

0.760
0.988
0.610
0.572
0.369
0.031*
0.015*
0.021*
0.008*
0.630
0.867
0.452
0.076*
0.002*
0.047*
0.013*
0.001*
0.013*
0.634
0.751
1.000
0.874
0.179
0.691
0.081
0.153
0.874

Manchurian
0.869
0.988
0.844
0.925
0.007*
0.003*
<0.001*
<0.001*
0.005*
1.000
0.867
0.666
0.196
0.218
0.297
0.821
0.060
0.020*
1.000
0.751
0.751
0.751
0.616
0.616
0.634
0.874
1.000
0.634
1.000
0.751
0.634
0.064
0.365
0.030*
0.112
0.874

Table 3.3: P values generated by repeated measures least significant difference tests comparing
mean number of live Agrilus planipennis caged with leaves from blue, white or
Manchurian ash leaves over a 14 day period. Bioassays were performed in June
2011. An “*” denotes significant differences in beetle survival (α=0.05) between
host species.

Ash Species Day
1
2
3
4
Blue
6
8
10
12
14
1
2
3
4
White
6
8
10
12
14

P values
White
Manchurian
0.781
0.781
1.000
0.487
0.266
0.165
0.331
0.126
<0.001*
<0.001*
<0.001*
<0.001*
<0.001*
<0.001*
<0.001*
<0.001*
<0.001*
<0.001*
0.578
0.487
0.781
0.578
0.677
0.890
0.211
0.165
0.071

97

Table 3.4: Mean (±SE) number of surviving Agrilus planipennis adults caged with excised
black, blue, green, white or Manchurian ash leaves in June 2010 or excised or intact
blue, white or Manchurian leaves in June 2011. There were two male and two
female A. planipennis per cage. Letters indicate differences among species and
treatments (α=0.05).
Ash Species

Control

Fertilizer

Black

3.4 ± 0.16

Blue

2.9 ± 0.16

Green

2.9 ± 0.18

White

2.4 ± 0.18

Manchurian

2.3 ± 0.22

Ash Species

Control

Intact

2.7 ± 0.15

Excised

3.4 ± 0.09

Intact

3.5 ± 0.09

Excised

3.3 ± 0.10

Intact

3.9 ± 0.04

Manchurian

a
b
b

de

2.8 ± 0.15

White

a

3.4 ± 0.10
2.2 ± 0.18
2.5 ± 0.18
3.3 ± 0.14
2.7 ± 0.16

a
b
b
a

ab

Fertilizer

Excised

Blue

a

e

bc
b

bc
a

2.9 ± 0.16

de

2.3 ± 0.20
3.4 ± 0.11
3.4 ± 0.11
3.3 ± 0.11

bc
bc
bc

3.7 ± 0.06

98

f

a

Paclobutrazol
3.1 ± 0.15
2.1 ± 0.19
3.0 ± 0.15
3.0 ± 0.15
2.9 ± 0.13

a
b
a
a
a

Paclobutrazol
2.7 ± 0.17
2.8 ± 0.17

de
de

3.8 ± 0.06
3.1 ± 0.10
3.4 ± 0.11

a
c

bc

3.8 ± 0.05

a

Table 3.5: Mean (±SE) number of Agrilus planipennis adult exit holes, parasitized larvae,
2

woodpecker attacks, living larvae, dead larvae and total galleries per m on five
species of ash trees (N=21 trees per species) and P values generated by least
significant difference tests comparing mean densities. An “*” denotes significant
differences between species (α=0.05).

P Values
Green
White
Adult Exit Holes

Blue
Ash Species

Mean (±SE)

Black

40.1 ± 8.61

Blue

0.0 ± 0.00

Green
White
Manchurian

30.4 ± 8.04
1.0 ± 0.39
0.0 ± 0.00

<0.001*

1.000
<0.001*

Manchurian

<0.001*

<0.001*

1.000
<0.001*

1.000
<0.001*
1.000

Parasitized Larvae
Black

96.5 ± 21.3

Blue

0.043*

<0.001*

<0.001*

0.038*

0.4 ± 0.27

Green
White
Manchurian

<0.001*

0.963

1.000
0.043*
0.962

46.5 ± 15.64
11.3 ± 6.77
0.0 ± 0.00

0.205

Woodpecker Attacks
Black

26.6 ± 4.91

Blue

0.2 ± 0.20

Green
White
Manchurian

0.013*

0.733

0.245

<0.001*

0.922
<0.001*

1.000
<0.001*
0.924

61.9 ± 14.90
10.2 ± 4.93
0.0 ± 0.00

Black

38.9 ± 9.38

Blue

1.0 ± 0.60

Green
White
Manchurian

0.229

0.115

Live Larvae
0.544
0.404
0.001*
0.971

63.9 ± 23.65
10.3 ± 3.02
0.6 ± 0.53

0.013*

99

0.126
1.000
0.002*
0.970

Table 3.5 (cont’d)
Dead Larvae (cause unknown)
Black
Blue
Green
White
Manchurian

30.5 ± 11.37
0.4 ± 0.29
14.6 ± 5.78
2.1 ± 0.56
0.6 ± 0.36

0.002*

0.353
0.286

0.006*
0.999
0.438

0.003*
1.000
0.329
1.000

Total Galleries
Black

235.9 ± 36.41

Blue

2.0 ± 0.98

Green

220.1 ± 39.77

White
Manchurian

<0.001*

0.821
<0.001*

40.7 ± 11.61
1.5 ± 0.67

<0.001*

<0.001*

<0.001*

0.681
<0.001*

<0.001*

<0.001*

100

Table 3.6: Number of Agrilus planipennis emerged adults from 2009 and 2010 larval cohorts and total larvae from the 2010 and 2011
larval cohorts including live and dead early instar larvae (first, second and third), total late instar larvae (fourth and
prepupae) including those parasitized by Atanycolus spp. or killed by woodpeckers found on five Fraxinus species in a
plantation, N=21 trees per species. Larvae from the 2010 cohort were progeny from A. planipennis adults that emerged in
2009. Larvae from the 2011 cohort were progeny from adults that emerged in 2010. Trees were harvested before larvae
from the 2011 cohort could pupate (Fall 2011).
Black Ash

Blue Ash

Green Ash

White Ash

Manchurian Ash

7.6
191

9.9
0

13.6
135

15.3
0

8.8
0

2

Phloem Area (m )
Emerged Adults, 2009 larval cohort

Larvae, 2010 cohort
Instars 1-3
4th Instars + Prepupae
Parasitized
Woodpecker Predation

Live
175
7
168

Emerged Adults, 2010 larval cohort

Larvae, 2011 cohort
Instars 1-3
4th Instars + Prepupae
Parasitized
Woodpecker Predation

Dead
614
104
510
345
145

Live
0
0
0

37

Live
23
0
23

Dead
0
0
0
0
0

Live
206
0
206

0

Dead
281
29
252
220
21

Live
7
4
3

101

Dead
830
38
792
170
505

Live
49
0
49

170

Dead
10
2
8
3
2

Live
367
8
359

Dead
114
9
105
0
104

Live
0
0
0

12

Dead
585
71
514
293
205

Live
122
54
68

Dead
0
0
0
0
0
0

Dead
197
19
178
155
15

Live
4
0
4

Dead
6
3
3
0
0

Leaf Area Consumed per Beetleday (cm2)

1.2

a
a

1.0
b

0.8
b
0.6
c
0.4
0.2
0.0
Black

Blue

A

Green

White

Manchurian

Ash Species

Leaf Area Consumed per Beetleday (cm2)

1.2

B

Intact
1.0

Excised
a

0.8
b

0.6

bc

c
0.4

bc

d

0.2

0.0
Blue

White
Ash Species
2

Manchurian

Figure 3.1: Mean (±SE) leaf area (cm ) consumed per beetle day by Agrilus planipennis caged
with (A) excised leaves from black, blue, green, white or Manchurian ash trees in
2010 or (B) excised or intact leaves from blue, white or Manchurian ash in 2011.
Letters indicate differences among species and leaf types (α=0.05).

102

Number of Surviving Beetles

4

*

*
*

3

*

*

Black
Blue

2

Green
White

1

Manchurian

0
1

2

3

Number of Surviving Beetles

A

4

6

8

10

12

14

Days After Emergence

4

*

*

*

*

*

3
Blue
2

White
Manchurian

1

0
1

B

2

3

4

6

8

10

12

14

Days After Emergence

Figure 3.2: Mean (±SE) number of surviving Agrilus planipennis adults caged with (A) black,
blue, green, white or Manchurian ash leaves in 2010 or (B) blue, white or
Manchurian ash in 2011. An “*” denotes significant differences among species on
that day (α=0.05).
103

Total Larval Galleries per m2

300

a

a

250
200
150
100

b
50
c

c

0
Black

Blue

Green

White

Manchurian

Ash Species
2

Figure 3.3: Mean (±SE) number of Agrilus planipennis larval galleries per m of debarked
phloem on black, blue, green, white or Manchurian ash trees harvested in 2011.
Letters indicate differences among species (α=0.05).

104

CHAPTER FOUR
Evaluation of ash xylem discoloration and architecture associated with macro-injections
of a systemic insecticide
ABSTRACT
Emerald ash borer (EAB, Agrilus planipennis Fairmaire) (Coleoptera: Buprestidae) was
first identified near Detroit, Michigan in 2002. It has killed tens of millions of ash (Fraxinus)
trees in 19 states and two Canadian provinces. Systemic insecticides applied as trunk injections
are often used to protect ash trees from EAB, but wounds and injury are a concern. I assessed
bark cracks and wound closure on 22 green ash (Fraxinus pennsylvanica Marsh.) and 24 white
ash (Fraxinus americana L.) trees macro-injected with either a low or medium-high rate of
emamectin benzoate in either 2008 or 2008+2009. Of the 233 injections sites assessed, only 12
(5%) injection sites had associated bark cracks and there was no evidence of pathogen infection.
On 39 (85%) of the 46 trees, new xylem was healing over injection sites. Ash species,
insecticide application rates and injection frequency had little effect on incidence of bark cracks
or wound closure. Xylem discoloration occurred around all injection sites, but it was likely
caused by a marker dye and was not indicative of tissue damage. In a related study, I examined
xylem anatomy in blue ash (Fraxinus quadrangulata Michx.) and green ash trees (caliper
diameter=77.3±2.73 and 81.0±2.52 mm, respectively). Trunk injected insecticides can
experience friction created by passing through xylem lumen or bordered pits. I assessed these
structures with CLSM to determine if xylem anatomy could influence insecticide uptake. Cross
sectional area of earlywood lumen was 55% larger and hydraulic conductivity was 67% greater
in green ash than blue ash, however, densities of bordered pits were similar between the two
species.

105

INTRODUCTION
Emerald ash borer (EAB, Agrilus planipennis Fairmaire) (Coleoptera: Buprestidae), a
phloem boring insect pest from Asia, was first identified near Detroit, Michigan in 2002
(Cappaert et al. 2005). Since its arrival, EAB has killed over 40 million ash (Fraxinus spp.) trees
in southeast Michigan alone and tens of millions in 18 additional states and two Canadian
provinces (EAB.info 2013). A variety of systemic insecticides are available to protect ash trees
from EAB, including products with azadirachtin, dinotefuran, emamectin benzoate and
imidacloprid as their active ingredients (Herms et al. 2009, McKenzie et al. 2010). Logistical
constraints, environmental concerns and cost typically limit use of these products to high-value,
urban ash trees (Mercader et al. 2011).
Most products used to protect ash trees from EAB are applied systemically as soil
drenches or injections, basal bark sprays or trunk injections (Herms et al. 2009). Ash trees are
ring porous (Dickison 2000) and systemic products are translocated upward through the xylem
tissues in the outer sapwood to the canopy (Herms et al. 2009; Mota-Sanchez et al. 2009; Choat
et al. 2007). Soil applications must be applied annually and contain the neonicotinoids
imidacloprid or dinotefuran. Dinotefuran can also be applied as a basal trunk spray, but annual
treatment is needed (Herms et al. 2009). Environmental contamination from soil and basal bark
applications can be a concern, however, especially near open water or sites with high water
tables. Neonicotinoids can persist in aquatic environments and potentially harm aquatic
invertebrates (Young-Song et al., 1997; Tišler et al. 2009). Systemic products that are injected
directly into the base of tree trunks eliminate insecticide drift and minimize non-target or
environmental effects (Sanchez-Zamora and Escobar 2000, Sur and Stork 2003, Herms et al.
®

2009, Hahn et al. 2012). Emamectin benzoate, sold as TREE-äge (4.0% emamectin benzoate,

106

Syngenta Crop Protection, Inc., Greensboro, NC, USA), an avermectin product, has become
especially popular in many EAB infested areas (Smitley et al. 2010; McCullough et al. 2011;
Mercader et al. 2011). In a large scale study, TREE-äge applied in the spring provided >99%
EAB control for at least two years (Smitley et al. 2010; McCullough et al. 2011).
Potential wounds from trunk injections, however, can be a concern for homeowners and
tree care professionals (Doccola et al. 2011). Wounding from the holes made during the
application process can sever conductive tissues and may provide an entry point for pathogens
(Shigo 1977; Lawson and Dahlsten 2003; Smith and Lewis 2006). Two types of trunk injections
are generally used to apply insecticides: micro-injections are applied by drilling holes ≤ 0.5 cm
(3/16”) in diameter whereas macro-injections are applied by drilling holes ≥ 1.0 cm (3/8”) in
diameter (Costonis 1981). Micro-injections generally deliver 1 to 3 ml of product per injection
site and have been used for many years (Costonis 1981). In contrast, macro-injections like those
®

performed by the TREE IV Micro-Infusion system (Arborjet, Inc. Woburn, MA, USA), are
relatively new. Macro-injection systems were designed to inject high rates and volumes of
product into large trees. While they require larger drill holes, the higher volume injections
improve within tree product distribution (Doccoloa and Wild 2012).
Pressure associated with trunk injection varies among application systems and can
®

potentially cause tree damage. The Wedgle Direct-Inject micro-injection system
(ArborSystems, Omaha, Nebraska, USA), does not require drilling but the injection pressure
created during injection can be high enough ≥ 334 kPa (≥50 PSI) to physically separate the bark
from the cambium (Smith and Lewis 2006). The J.J. Mauget Company (Arcadia, CA, USA)
®

uses a closed system, micro-injection capsule that is pre-loaded with an imidacloprid (Imicide ,

107

10% imidacloprid, J.J. Mauget Co., Arcadia, CA, USA). These capsules are inserted into 0.3 cm
(1/8”) diameter holes drilled into the base of trees. Insecticide is injected into the sapwood under
approximately 41 kPa (6 PSI) of pressure (Mauget.com, Accessed 14 April 2013). TREE-äge is
typically injected with a TREE IV Micro-Infusion

®

, or a QUIK-jet® system (Arborjet, Inc.

Woburn, MA, USA). Holes are drilled around the base of trees with a 0.95 cm drill bit (3/8”)
®

and Arbor plugs are inserted to prevent leaking and seal the drill hole (Arborjet.com, Accessed
14 April 2013). Insecticides are injected with a needle into the sapwood under approximately
310 kPa (45 PSI) or 1379 kPa (200 PSI) of pressure for the TREE IV or QUIK-jet, respectively.
TREE-äge is efficacious for up to three years, which means substantially fewer injection sites
compared to products that must be applied annually (Smitley et al. 2010; Doccola et al. 2011;
McCullough et al. 2011).
In addition to wounds made by drilling, trunk injected insecticides can also cause internal
tissue discoloration in multiple genera, including ash (Shigo et al. 1977, Smith and Lewis 2006,
Mota-Sanchez et al. 2009, Doccola et al. 2011, Tanis et al. 2012). When maple and ash trees
were injected with imidacloprid, discoloration in tissues adjacent to injection wounds was
considered a wound compartmentalization response (Smith and Lewis 2006). If tissue
discoloration is indicative of compartmentalization, insecticide distribution through the tree
could be compromised because barrier zone tissues rarely translocate (Shigo 1977). Whether
discoloration is indicative of wounding, however, is questionable. In a study where 14C®

imidacloprid (Imicide , J.J. Mauget Co., Arcadia, CA, USA) was injected into small ash trees
(Mean DBH = 4.5 cm), discolored areas adjacent to injection sites had 75 to 300 times higher
imidacloprid equivalent concentrations than adjacent unstained tissues (Mota-Sanchez et al.

108

2009, Tanis et al. 2012). Whether cells were viable, however, is unknown. Tissue discoloration
was also observed in four green ash (Fraxinus pennsylvanica Marsh.) trees (mean DBH = 27.6
cm) treated with TREE-äge but Doccola et al. (2011) reported discolored areas remained “firm”
and no symptoms of infection or deterioration were observed four years after the injections.
While their study indicated that discoloration was not indicative of tree damage, their sample
size was small and the integrity of xylem cells was not evaluated.
Although trunk injection systems and insecticide formulations have been continually
refined, potential effects on hydraulic conductivity and tree response to macro-injection remain
relatively unevaluated. Understanding the effects of macro-injections on ash trees is especially
important given the increasingly wide use of TREE-äge for EAB protection. Objectives of our
study were to determine if 1) TREE-äge trunk injections created external bark cracks or necrosis
in green ash or white ash trees; 2) discoloration in the xylem was indicative of tree injury and 3)
whether wounding varied between ash species, application rates or frequency of TREE-äge
injections
Success of trunk injections can also be affected by an array of physical parameters (tree
health, size, previous injury), environmental factors (air temperature, soil moisture) and cultural
practices (application timing, delivery method) (Smith and Lewis 2005, Doccola et al. 2006,
Mota-Sanchez et al. 2009, Tanis et al. 2012). Uptake and translocation of systemic products is
further influenced by root, xylem and crown anatomy (Tatter et al., 1998, Young, 2002, MotaSanchez et al. 2009, Tanis et al. 2012).
Interspecific variability may also affect the rate at which systemic products move into
trees. Among individual ash trees, injection efficiency and efficacy can be highly variable
among trees depending on soil moisture and weather related factors that affect transpiration and

109

uptake (Mota-Sanchez et al. 2009; Tanis et al. 2012). Blue ash (Fraxinus quadrangulata
Michx.) trees, for example, are reportedly difficult to trunk inject compared to green ash or white
ash trees. Xylem architecture is known to play an important role in the translocation of
insecticides within ash trees (Tanis et al. 2012), therefore, I hypothesize that interspecific
differences in xylem anatomy might affect trunk injected insecticide movement in trees. Trunk
injected products are subject to two sources of friction: 1) resistance created by vertical passage
through xylem lumen, the hollow portion of the xylem vessel elements and 2) resistance
resulting from lateral passage in the xylem through the bordered pits (Choat et al. 2007). Our
second goal was to measure blue ash and green ash earlywood lumen area and bordered pit
densities using confocal laser scanning microscopy (CLSM).

110

METHODS
Trunk Injection Potential Injury
I evaluated injection sites on 46 ash trees growing in four locations in Ionia, Genesee and
Clinton Counties in central Michigan. The 22 green ash trees ranged in size from 4.3 to 11.5 cm
diameter at breast height (DBH, 1.4 meters aboveground) (mean DBH= 7.13 ± 0.45 cm) and the
24 white ash ranged in size from 4.2 to 8.9 cm DBH (mean DBH = 6.90 ± 0.24 cm).
Between 21 and 23 May 2008 all 46 trees were injected with TREE-äge (4.0%
emamectin benzoate, Syngenta Crop Protection, Inc., Greensboro, NC, USA). A total of 24 trees
were treated using the lowest label application rate (EB-low) (0.1 g a.i. per 2.5 cm DBH) applied
with a QUIK-jet system (Arborjet, Inc. Woburn, MA, USA). The other 22 were treated with a
medium-high rate (EB-high) (0.4 g a.i. per 2.5 cm DBH) plus an equal amount of water applied
with a TREE IV Micro-Infusion system (Arborjet, Inc. Woburn, MA, USA). The number of
injections sites was determined by tree DBH, according to label directions. Injection sites were
evenly spaced and care was taken to avoid placing injection sites near injured or dead tissue.
Injection sites were drilled into the base of each tree with a 0.95 cm drill bit (3/8”). Arbor plugs
(#4, 0.95 cm (3/8”), Arborjet, Inc. Woburn, MA, USA) were carefully inserted with a set tool
and plastic mallet, according to label directions.
On 2-4 June 2009, 22 trees, including 11 EB-high and 11 EB-low trees were re-treated.
Holes were drilled as above, approximately 5 cm above and offset from 2008 injections,
according to label directions. The remaining 24 trees, including 11 EB-high and 13 EB-low trees
were not re-treated. Trees were felled between 10 October and 29 December 2009. After
felling, the basal 50 cm of each bole was brought to the laboratory and cut into 5.1 cm thick

111

cross sections. Cross sections were reassembled after cutting and reference points were
established to ensure proper orientation throughout the experiment.
I examined a total of 239 injection sites on 301 cross sections from the 46 trees. Six
injection sites had no associated discoloration (one 2008 EB-high, one 2008 EB-low tree, four
2008+2009 EB-high trees) because the applicator determined plugs were set improperly and no
insecticide was injected. These six injection sites were excluded from further analyses. Of the
233 injection sites examined, 50 were from the 11 EB-high trees treated in 2008, 77 were from
the 11 EB-high trees treated in 2008 + 2009, 42 were from the 13 EB-low trees treated in 2008
and 64 were from the 11 EB-low trees treated in 2008 + 2009. I assessed discoloration depth,
width and height for each injection site. Discoloration depth (mm) was the longest distance
between the outside edge of discolorations (near the bark) to the inside edge (near the central
pith) (Fig. 4.1). Discoloration width (mm) was the widest distance between left and right edges
of the discoloration associated with a single site (Fig. 4.1). Discoloration height (cm) was the
vertical distance discoloration persisted above and below the basal cross sections. I also
measured bark thickness at two opposite points on each cross section.
I visually assessed all injection sites to evaluate their condition (Fig 4.1). I assumed
xylem produced during the same year trees were injected was latewood (trees were injected in
May) while xylem produced the year after injections was both earlywood and latewood
(Dickison 2000). I visually examined the cross sections and recorded external bark cracks,
internal xylem necrosis and evidence of pathogen infection.
Transverse tissue samples (approximately 5

3 mm) were removed from the

discolored/unstained tissue interface (Fig. 4.1) of randomly selected injection sites on 20
different trees using a razor blade. I removed 5 to 10 samples from each injection site to ensure

112

that I would have quality samples. After removal, samples were immediately placed into fixing
solution (50% ethanol, glycerol, formaldehyde, 18:1:1) for at least 72 hours. After fixation,
samples were placed in a solution of safranin orange dye (1% safranin, Carolina Biological
Supply Company, Burlington, NC, USA) for 15 minutes, then dehydrated through 30, 50, 75, 90
and 100% ethanol (Decon Labs, Inc., King of Prussia, PA, USA), then rehydrated backward
through the same solutions. Samples remained in each alcohol solution for 10 minutes.
Solutions were changed multiple times during each cycle to remove excess dye. After
rehydrating, samples were placed in deionized water for 15 minutes then processed through 25,
50 and 75% glycerol solutions (1 hour each with frequent decanting) and finally left overnight in
100% glycerol. Samples were viewed with an Olympus Fluoview 1000® confocal laser scanning
microscope (Olympus Corporation, Tokyo, Japan) the following day.
I used confocal laser scanning microscopy (CLSM) to evaluate sapwood integrity. All
tissue samples were illuminated with argon (λ=543 nm) and helium-neon (λ=488 nm) lasers.
Single channel z-scan collections were recorded (Hutzler et al. 1989). A 10 dry objective was
used to magnify transverse samples removed from the discolored/unstained interface of trees
trunk injected with TREE -äge. Total slice depth ranged from 10.2 to 14.5 µm and 7 to 21 slices
per sample were obtained depending on optimal image acquisition parameters.
Xylem Architecture
I assessed xylem architecture in young blue ash and green ash trees growing in a common
garden setting at the Michigan State University Tree Research Center (MSU TRC) in Okemos,
Michigan (Ingham County). Trees which arrived as whips (≤1cm diameter) were planted in
April 2006 and protected from EAB colonization using a cover spray in 2006-2007 and by
physical barriers from 2008-2012). Trees were watered twice per week or as needed with drip

113

irrigation and fertilized annually with a top dressing of Harrel’s Pro-Blend with Micros (19-5-10)
(Harrell’s, Lakeland, FL, USA).
On 12 July 2012, three blue ash (mean (±SE) caliper diameter=77.3 ± 2.73 mm) and three
green ash trees (81.0 ± 2.52 mm) (all trees were 7 years old) were injected with safranin orange
dye (1% safranin, Carolina Biological Supply Company, Burlington, NC, USA) through two
injection sites on opposite sides at the base of each tree using a 0.95 cm drill bit (3/8”), 20 cm
above the ground. Arbor plugs (#4) were carefully inserted into each injection site as described
above. I used the QUIK-jet system (Arborjet, Inc. Woburn, MA, USA) to inject 15 ml of
safranin into each injection site (30 ml per tree). I was unable to inject dye into the sites on two
of the blue ash trees, so I drilled two new injection sites on each tree, 5 cm above and offset from
the original injection sites and successfully injected the dye. After injections were complete,
trees were left untouched for 24 hours, they were then felled, returned immediately to the lab and
dissected to assess within tree translocation. I also removed two cross sections (1 cm thick,
each) from each bole, 10 cm above the injection sites.
Each cross section was dissected with a chisel to isolate wood stained by the injected
safranin. Because earlywood vessel size can vary from year to year depending on environmental
conditions (Dickison 2000), these stained portions were further dissected to isolate stained xylem
produced in 2011. I removed several thin (approximately 200 µm thick) transverse and
tangential cross sections with a razor blade. After removal, samples were dehydrated with
alcohol and cleared with glycerol as described above.
Samples were viewed with the Olympus Fluoview 1000® confocal laser scanning
microscope the following day. Transverse samples taken from trees injected with safranin were
magnified with a 10 dry objective. Slice depth ranged from 5.0 to 11.5 µm and the number of

114

slices ranged from 14 to 22 per sample. Blue ash and green ash xylem elements on tangential
samples were magnified with a 20 dry objective. Total slice depth was 1.0 to 2.0 µm and the
number of slices ranged from 38 to 62 slices per sample.

Within these samples, bordered pits

were further magnified with a 1.8 zoom, slice depth was 1.0 µm for all samples and the number
of slices ranged from 44 to 95 per sample.
I used CLSM to determine the cross sectional area of all earlywood lumen that were
larger than 30 µm in each sample (these were the vessels that translocated a majority of the dye).
I examined 48 transverse samples (six trees, two injections per tree, two cross sections per tree,
two transverse samples from each injection stain on each cross section) from earlywood
produced in 2011. I used Olympus Fluoview® software (FV10-ASW Viewer 3.0, Olympus
Corporation, Tokyo, Japan) to determine the length (µm) and width (µm) of large earlywood
vessels in each image (image size = 1.6 mm2). I then calculated lumen cross-sectional area using
the formula ellipse area = π

(length / 2)

(width / 2) (Figs. 4.2 and 4.3). Total cross sectional

area of large earlywood lumen was obtained by summing the cross sectional areas.
To assess average density of bordered pits, I identified 48 vessels (six trees, two
injections per tree, two cross sections per tree, one tangential sample per injection on each cross
section, two vessels within each tangential sample) and assessed the number of bordered pits
within two 50

50 µm sections of each vessel (Figs. 4.4 and 4.5). I calculated the bordered pit

density per vessel by averaging the number of bordered pits within the two 50

50 µm sections.

Statistical analysis
Data were analyzed using SAS statistical software (SAS Institute, Inc. 1989).
Assumptions of normality were tested with residual plots and the Shapiro-Wilk test (Shapiro and
Wilk 1965). Width and length of discolored tissue on cross sections from trees treated with

115

TREE-äge were normalized using log transformations (Ott and Longnecker 2001), while all
other variables met assumptions of normality. Three-way ANOVA was performed to assess
effects of ash species (green or white), TREE-äge rate (EB-low or EB-high) and number of
applications (2008 only or 2008 + 2009) and their interactions on the presence of xylem laid over
injection sites, presence of bark cracks or necrosis and discoloration depth, width and height.
When ANOVA results were significant (α ≤ 0.05), Fisher’s protected least significant difference
(LSD) multi-comparison tests were applied with Tukey’s adjustment for unbalanced data sets
(Ott and Longnecker 2001) to determine differences between means. For the blue ash and green
ash trees injected with dye, I used Student’s t tests to determine if earlywood lumen cross
sectional area, cross sectional area of all earlywood lumen (>30 µm) per image, density of
earlywood vessels or density of bordered pits differed between blue ash and green ash trees.

116

RESULTS
Trunk Injection Potential Injury
Diameter at breast height of green ash and white ash trees used for this study ranged from
10.7 to 29.2 cm but DBH was similar among treatment combinations (EB low or EB high trees
treated in 2008 or 2008+2009) (P=0.65). The number of injections on EB-high and EB-low trees
treated only in 2008 was similar and averaged 4.6 ± 0.53 and 3.3 ± 0.24 injections per tree,
respectively. The number of injection sites on EB-high and EB-low trees treated in 2008 + 2009
was also similar and averaged 7.4 ± 0.54 and 5.8 ± 0.54 injections per tree, respectively.
Number of injections did not vary between species (P=0.75) nor was the species

insecticide

rate interaction significant (P=0.37).
Tree species, insecticide application rate, injection frequency and the interactions of the
factors did not affect whether or not xylem was growing over drilled injection sites (Table 4.1).
Xylem was healing over 134 (58%) of the injection sites, but 42% of the injection sites were still
completely open. Among the 46 trees examined in this study, 39 (85%) had new xylem growing
over at least one injection site and 37 trees (80%) had new tissue growing over multiple injection
sites.
I assessed 12 trunk injuries near injection sites on six trees (13%), the remaining 40 trees
had no evidence of injury. All of the injuries were vertical bark cracks (six bark cracks on green
ash trees, six bark cracks on white ash trees). Among the 13 EB-low trees that were treated in
one year (2008), three trees had bark cracks, but no trees had internal necrosis or cracking.
Among the 11 EB-low trees that received two treatments (2008 + 2009), one tree had bark cracks
but none of the trees had internal necrosis or cracking. Among the 11 EB-high trees treated in
one year (2008), one tree had bark cracks but no trees had internal necrosis or cracking. Among

117

the 11 EB-high trees treated in two years (2008 + 2009) one tree had bark cracks, but no trees
had internal necrosis or cracking. The number of bark cracks was higher (Table 4.1) in trees that
received only one treatment in 2008 (nine bark cracks) than trees that received treatments in
2008+2009 (three bark cracks). There were no significant interactions among species,
insecticide rate or number of injections (Table 4.1). One white ash EB-low tree treated in 2008
had seven injections sites, four of which had associated bark cracks.
Depth and width of discolored xylem were not affected by species, treatment rates,
number of applications or their interactions (Table 4.1). Average (±SE) discoloration depth was
31.6 ± 3.25 mm while average (±SE) discoloration width was 13.7 ± 1.22 mm. Discoloration
height was 63% higher (P<0.001) in EB-high trees (32.3 ± 2.4 cm) than in EB-low trees (12.1 ±
0.96 cm) (Table 4.1) but was similar among species and between trees treated in 2008 or 2008 +
2009. No interactions were significant (Table 4.1). At least 67% of trees had discolored tissue
below the basal felling cut, which was at ground level.
There was no evidence that tissue damage was associated with discoloration among any
of the samples examined with CLSM (Fig. 4.6). Cell lysis was not present in any of the samples
and there were no visible symptoms of tissue damage or infection. I scrutinized areas where
discolored/unstained tissue interfaces were well-defined, as these areas might be indicative of a
barrier zone (Shigo 1984), but I saw no evidence that trees attempted to compartmentalize
around discolored tissues (Fig. 4.6). When preparing tissues for examination with CLSM, much
if not all of the discoloration disappeared from the xylem tissue. In addition, discoloration
associated with 2008 injection sites or 2009 injection sites overlapped and were continuous
through cross sections indicating tissues discolored in 2008 were translocating product in 2009.

118

Xylem Architecture
The safranin orange dye injected in July 2012 was translocated primarily through large
earlywood vessels (> 30µm) formed in 2011 and 2012. Dye was present in the terminal leader
and lateral branches of all six trees within 24 hours of injection. Cross sectional area of large
earlywood lumen was 55% smaller in blue ash trees than green ash trees (t [1,689] = 13.48; P <
2

0.001). Cross sectional area of earlywood lumen in blue ash averaged 0.14 ± 0.02 mm per
2

lumen compared to 0.31 ± 0.02 mm in green ash. There were more (t [1,47] = 11.52; P < 0.001)
earlywood vessels per image in the blue ash images (19 ± 1.02 per image) than in green ash
images (12.4 ± 0.57 per image), but the total cross sectional area occupied by all of the large
earlywood vessels in an image was lower (t [1,47] = 7.41; P < 0.001) in blue ash (141.3 ± 20.0
2

2

µm ) than in green ash (308.6 ± 19.2 µm ).
The density of bordered pits was similar between species (P = 0.13) and averaged 76.3 ±
2

3.9 and 70.5 ± 2.2 pits per 2,500 µm for blue ash and green ash, respectively. I observed that
helical thickening was present in all blue ash vessels (Fig. 4.7) but only 45% of green ash
vessels. The degree of helical thickening was also much more pronounced in blue ash (Fig. 4.7)
than in green ash (Fig. 4.8).

119

DISCUSSION
Over the next ten years, approximately 37.9 million ash trees in urban areas across the
United States are projected to be within areas infested by EAB (Kovaks et al. 2010). When EAB
was first discovered in 2002, many landscape ash trees were removed. Treatment success was
variable and the logistics and cost of annual treatment were impractical (Herms et al. 2009).
Registration of TREE-äge in 2010 and evidence of highly effective control, however, have
reduced the costs and logistical constraints associated with treatment. As a result, large mature
ash trees, and the ecosystem services they provide, are being preserved in EAB infested areas
(Smitley et al 2010; Doccola et al. 2011; McCullough et al. 2011; McCullough and Mercader
2012).
Our study indicates that damage to ash trees is rare when trees are trunk injected with
TREE-äge using QUIIK-jet or TREE IV Micro-Infusion systems. Among the 233 injection sites
on the 46 white ash and green ash trees assessed in this study, only 5% of injection sites had
evidence of bark cracks and none had disease symptoms. Our results are consistent with those of
Doccola et al. (2011) who found no “cracking, oozing or decay” on 4 green ash trees (DBH 21.5
to 36.3 cm DBH; mean = 27.6 cm) injected with emamectin benzoate (the four trees were treated
in 2005, and two of these were retreated in 2008) (TREE-äge, injected with a TREE IV MicroInfusion system). Unlike imidacloprid products that require annual applications, TREE-äge is
efficacious for up to three years (Smitley et al. 2010; McCullough et al. 2011). Reduced
application frequency means fewer opportunities for wounding.
Many studies have reported the existence of tissue discoloration adjacent to trunk
injection sites, but the integrity of discolored tissue has not been previously assessed (Smith and
Lewis 2006; Mota-Sanchez et al. 2009; Doccola et al. 2010; Tanis et al. 2012). In this study,

120

multiple lines of evidence show that discoloration surrounding injection sites was not indicative
of tissue damage. Results from observations with CLSM determined the integrity of xylem
tissues immediately surrounding injection sites was sound regardless of ash species, insecticide
rate or the number of injections. In addition, when I prepared samples for examination with
CLSM, much, if not all, of the discoloration was removed when samples were soaked in alcohol.
If discoloration was indicative of wounding, I would have expected it to persist. I also found
discolored tissues from 2008 injections effectively translocated products injected in 2009, which
would not have been possible if discoloration represented barrier zone formation (Shigo 1977).
There were also six “blank” injection sites (no insecticide was injected) with no associated
discoloration further suggesting the insecticide formulation was the source of discoloration. If
wounding had been the cause, tissues around these blank sites should have also been discolored.
In hindsight, it would have been advantageous to inject water to assess the wound response
exclusively associated with drilling and injection pressure. In other macro-injected ash trees,
insecticide concentration in discolored tissues was much higher than in adjacent unstained tissues
(Mota-Sanchez et al. 2009, Tanis et al. 2012) and a previous study on green ash trees, discolored
tissues were sound even four years after injection (Doccola et al. 2011).
Tissue discoloration in trunk injected trees is most likely caused by a “marker dye”.
Marker dyes are typically inert and are used to color pesticides so that applicators can see and
safely handle the products they are applying (Pepling et al 1997). The identity of the blue
marker dye used in the TREE-äge formulation is protected by Arborjet’s proprietary patent.
In my study and others, tissue discoloration diminished with tree height above the
injection site (Mota-Sanchez et al. 2009; Tanis et al. 2012). As I followed discoloration upward
through the tree cross sections, discolored areas became smaller and paler. This is likely because

121

the volume of translocated insecticide decreased with tree height. Previous studies indicate that
a reservoir of trunk injected insecticide can persist for several years around injection sites but
these studies used high concentrations of imidacloprid a product that may have low solubility.
This suggests that not all of the injected product is actually translocated into the canopy in a
given year (Mota-Sanchez et al. 2009). In our study, EB-high trees received at least 2 more
liquid than EB-low trees. As a result, discoloration height was over twice as high in EB-high
trees as it was in EB-low trees, even though EB-low trees were likely injected under greater
pressures. This suggests vertical discoloration was a function of the volume of insecticide that
was translocated to the canopy. In contrast, discoloration width and length did not vary between
EB-high and EB-low trees. This suggests the lateral movement of insecticides in the lower 50
cm of the trees was unaffected by volume or the pressure applied during application.
I identified interspecific variations in blue ash and green ash xylem anatomy that could
influence uptake of trunk injected insecticides. While blue ash appears relatively resistant to
EAB compared to other North American ash species (Tanis and McCullough 2012), blue ash
trees will likely require insecticide treatments during peak densities of EAB invasion. Arborists
have noted the difficulty of injecting blue ash relative to other ash species and I similarly
experienced this difference. I was unable to inject dye into four of our original six injections
sites on blue ash trees and had to create additional injection sites to apply the dye. After
investigating the internal anatomy of blue ash and green ash xylem, multiple factors likely
contribute to injection efficiency.
During translocation, trunk injected products are subject to friction created by passage
through the lumen and pit fields (Choat et al. 2007). North American ash trees are ring porous,
straight grained and in spring, when hydraulic pressure is low, ash trees transport water through

122

large earlywood vessels (Dickison 2000). As the growing season progresses, production of the
number of large vessels drops and trees produce smaller vessels (latewood) that can transport
water under high hydraulic pressure (Choat et al. 2007). The Hagen-Poiseuille equation states
that hydraulic conductivity, the ease with which water can flow through a plant, is proportional
to the sum of vessel radii raised to the fourth power. Xylem conductivity therefore is a function
of lumen diameter and frequency (Choat et al. 2007, Kiten et al. 2010). Applying the HagenPousille equation to determine the hydraulic conductivity of earlywood vessels in our study
(Figs. 4.2 and 4.3) indicates green ash hydraulic conductivity was 67% greater than that of blue
ash, despite the fact that the density of large earlywood vessels in blue ash was twice as high as
green ash. Earlywood vessel size can be influenced by tree age, environment and genetic
variability (Dickison 2000). Our trees were all the same age (seven years old), received the same
cultural amendments, and samples were removed from tissues produced in the same year (2011),
indicating differences in cross sectional area of lumen were species dependent.
While lumen transport water vertically, bordered pits, the gateway between xylem
conduits, play a critical role in controlling the lateral movement of water (Zwieniecki et al. 2001,
Kiten et al. 2004, 2009 Domec et al. 2006, Choat et al. 2007). I originally hypothesized that
density of bordered pits in blue ash would be substantially less than in green ash because
bordered pits can account for 50% of the resistance undergone by injected insecticides (Choat et
al. 2007). This was not the case, however, densities of bordered pits were similar between the
two species.
I observed that earlywood vessel elements in blue ash had a greater incidence of helical
thickening than green ash vessels. Helical thickenings are lignin deposits that develop in
earlywood vessels. When thickening crosses back and forth across xylem elements, as seen in

123

this study, the term helical or spiral thickening is used (Toole and Toole 2004). The extent of
thickening, and the patterns that appear as a result of thickening, can differ among species.
Helical thickening generally increases with tissue age (Bailey 1944) and can block water
movement (Kiten et al. 2004). In our study, the extent of helical thickening in blue ash
earlywood vessels was striking. In some cases, blue ash vessels appeared completely blocked.
Differences in helical thickening between blue ash and green ash samples could not be attributed
to environment or tissue age. The abundance of helical thickening in blue ash xylem likely
contributes to the difficulty of blue ash trunk injection.
Injection of systemic insecticides for EAB control generally occur in the spring, allowing
time for products to translocate to the canopy of ash trees where EAB adults are feeding on
leaves (Herms et al. 2009). I injected dye in July to assess the anatomy of xylem tissues that
were involved in translocation. While the timing of our dye injection was later than what is
typically recommended for optimal EAB control (Herms et al. 2009; McCullough et al. 2011) I
examined lumen size and bordered pit density in mature xylem produced by trees in 2011
Given the continued spread of EAB across North America, understanding effects of trunk
injection systems and products on ash tree health is important. In this study, I found green ash
and white ash trees were rarely harmed by the trunk injection process. Internal discoloration of
xylem tissues after trunk injection was previously assumed to be indicative of an ash tree
compartmentalization response (Smith and Lewis 2005). Our results indicate discoloration was
likely caused by inert dyes found in the TREE-äge formulation. It is also important to
understand how ash tree anatomy can affect the uptake of trunk injected insecticides. I found
xylem characteristics of blue ash and green ash trees likely contribute to interspecific differences
in uptake of trunk injected insecticide. Trees of many genera receive trunk injections, including

124

Acer, Populas and Quercus, perhaps further investigation of interspecific differences in xylem
anatomy in these species could advance our general understanding of the trunk injection process.

125

Table 4.1: Results of analysis of variance to assess differences between green ash and white ash, insecticide rate (low rate = 0.1 g a.i.
per 2.5 cm DBH or medium-high rate = 0.4 g a.i. per 2.5 cm DBH), injection frequency (2008 or 2008 + 2009) and their
interactions on the number of trunk injection sites with new wood growing over them or secondary wounds and tissue
discoloration depth, width and height.

Effect
Species (Spp)
Insecticide Rate (Ins Rate)
Spp × Ins Rate
Injection Frequency (Inj Freq)
Spps × Inj Freq
Ins Rate × Inj Freq
Spp × Ins Rate × Inj Freq

Effect
Species (Spp)
Insecticide Rate (Ins Rate)
Spp × Ins Rate
Injection Frequency (Inj Freq)
Spps × Inj Freq
Ins Rate × Inj Freq
Spp × Ins Rate × Inj Freq

New Wood
F Value P value
1.10
0.300
1.66
0.205
0.05
0.817
1.97
0.05
0.51
0.04

0.169
0.828
0.481
0.839

Discoloration
Depth
F Value P value
0.33
0.571
0.01
0.37
0.60
0.59
1.20
0.32

0.934
0.548
0.445
0.447
0.280
0.576

Secondary Wounds
F Value P value
1.18
0.285
1.47
0.233
0.00
0.950
0.024*
5.57
1.18
0.285
1.47
0.233
0.00
0.950
Discoloration
Width
F Value P value
1.20
0.281
3.09
0.49
0.05
1.16
0.66
1.20

0.087
0.487
0.821
0.289
0.422
0.281

126

Discoloration
Height
F Value P value
0.27
0.605
<0.001*
61.66
0.26
0.616
0.34
0.561
0.80
0.377
0.03
0.875
0.21
0.649

Discolored/unstained
tissue interface of
2009 injections
Discoloration length
(L) and width(W)

2009 Growth ring
2008 Injection discoloration
2008 Growth ring
2009 Xylem growing
over injection wounds

Figure 4.1: A cross section of a Fraxinus pennsylvanica tree injected in 2008 + 2009 with a low
rate (0.1g a.i. per 2.54 cm at diameter at breast height) of emamectin benzoate
®
®
(TREE-äge ) applied with a QUIK-jet system. The cross section was cut 10.2 cm
above 2008 injection sites. Discoloration from 2008 and 2009 injections, and
injection wounds from 2009 are visible. Xylem is present around wounds and
discolored tissues. Cracking in the cross secti+on was caused by drying, not the
injection process.

127

Xylem ray

Helical thickenings
Helical thickenings

Earlywood xylem vessel

Figure 4.2: Transverse cross section of Fraxinus quadrangulata earlywood xylem tissue dyed
with safranin orange. The image was obtained with a confocal laser scanning
microscope equipped with helium-neon (λ=543 nm) and argon (λ=488 nm) lasers
and a 10 dry objective for magnification. The three-dimensional image consists
of 19 slices obtained 11.5 µm apart.

128

Earlywood
xylem
vessels

Helical thickening

Latewood
xylem vessel
Xylem ray

Figure 4.3: Transverse cross section of Fraxinus pennsylvanica earlywood xylem tissue. The
image was obtained with a confocal laser scanning microscope equipped with
helium-neon (λ=543 nm) and argon (λ=488 nm) lasers and a 10 dry objective for
magnification. The three-dimensional image consists of 15 slices obtained 5.0 µm
apart.

129

Parenchyma cell

Bordered pits
End wall
remnant

50μm

50μm

Helical thickening

50μm
Figure 4.4: Radial cross section of a Fraxinus quadrangulata xylem element and surrounding
cells from a tree injected with 15 ml of safranin orange. The image was obtained
using a confocal laser scanning microscope equipped with helium-neon (λ=543 nm)
and argon (λ=488 nm) lasers, a 20 dry objective and 1.8 zoom were used for
magnification. The three-dimensional image consists of 87 slices obtained 1.0 µm
apart.

130

Parenchyma cell
Bordered pits

End wall
remnant
50μm
50μm

Wood fibers

50μm

Figure 4.5: Radial cross section of a Fraxinus pennsylvanica xylem element and surrounding
cells of a tree injected with 15ml of safranin orange. The image was obtained using
a confocal laser scanning microscope equipped with helium-neon (λ=543 nm) and
argon (λ=488 nm) lasers, a 20 dry objective and 1.8 zoom were used for
magnification. The three-dimensional image consists of 48 slices obtained 1.0 µm
apart.

131

Latewood xylem vessel

Boundary between
earlywood and
latewood
Earlywood xylem vessel

Xylem ray cells

100μm

Figure 4.6: Transverse cross section of Fraxinus pennsylvanica discolored by trunk injected
emamectin benzoate. The image was obtained using a confocal laser scanning
microscope equipped with helium-neon (λ=543 nm) and argon (λ=488 nm) lasers.
The three-dimensional image consists of eight slices obtained 14.5 µm apart. A
10 dry objective was used for magnification.

132

Bordered pits

End wall
remnant

Helical thickenings
Parenchyma cell

Helical thickenings

Wood fibers

Figure 4.7: Radial cross section of a Fraxinus quadrangulata xylem element and surrounding
cells from a tree injected with 15ml of safranin orange. The image was obtained
using a confocal laser scanning microscope equipped with helium-neon (λ=543 nm)
and argon (λ=488 nm) lasers. The three-dimensional image consists of 54 slices
obtained 2.0 µm apart. A 20 dry objective was used for magnification.

133

Wood fibers
Parenchyma cells

Xylem ray cells

Pit fields

Helical
thickening
Wood fibers
End wall remnant

Figure 4.8: Radial cross section of a Fraxinus pennsylvanica xylem element and surrounding
cells from a tree injected with 15ml of safranin orange. The image was obtained
using a confocal laser scanning microscope equipped with helium-neon (λ=543 nm)
and argon (λ=488 nm) lasers. The three-dimensional image consists of 39 slices
obtained 1.0 µm apart. A 20 dry objective was used for magnification.

134

CHAPTER FIVE
Examination of the phloem of three ash species using fluorescent confocal
laser scanning microscopy
ABSTRACT
Variability in ash (Fraxinus spp.) tree resistance to emerald ash borer (EAB, Agrilus
planipennis Fairmaire), an exotic pest introduced from Asia, has sparked interest in ash
physiology and defensive chemistry. Blue ash (Fraxinus quadrangulata Michx.), which is
native to North America, does not share an evolutionary history with EAB, yet it has resistance
equal to that of Manchurian ash (Fraxinus mandshurica Rupr.), a species that co-evolved with
EAB in Asia. In contrast, green ash (Fraxinus pennsylvanica Marsh.) is highly preferred by
EAB. Phenolic compounds and reactive oxygen species (ROS) are often produced by plants
when they are injured. These compounds may serve as defensive compounds (e.g., antifeedants), signal molecules or defensive compound precursors. Larval stages of EAB develop
under the bark of ash trees, therefore interspecific differences in phloem anatomy and defensive
chemical production could play an important role in tree resistance. I used confocal laser
scanning microscopy (CLSM) to detect phenolic compounds, assess ROS production and
examine the architecture of blue ash, green ash and Manchurian ash phloem. Autofluorescence
indicates that phenolic compounds were present in blue ash and Manchurian ash phloem but
there were no detectable levels of phenolic compound autofluorescence in green ash phloem. All
three ash species produced ROS, but Manchurian ash trees produced at least three times more
than blue ash and green ash trees. In addition to chemical defenses, blue ash and Manchurian
ash phloem had a continuous layer of sclerenchymatous cells that could affect EAB larvae. The
layer of sclerenchymatous cells was also present in green ash but it was not continuous.

135

INTRODUCTION
Emerald ash borer (EAB, Agrilus planipennis Fairmaire) (Coleoptera: Buprestidae) was
first identified in North America near Detroit, Michigan in 2002 (Cappaert et al. 2005). A native
of Asia, EAB has killed tens of millions of ash (Fraxinus spp.) trees in 19 states and two
Canadian provinces (EAB.info 2013). In Michigan, EAB adults generally emerge in mid-May.
Females feed for approximately three weeks before they begin laying eggs in the cracks and
crevices of ash bark in mid-June (Poland and McCullough 2006). Ash tree canopies are fully
flushed during this time (Barnes and Wagner 1981) and phloem is actively transporting nutrients
from the canopy to the roots (Dickison 2000). First instar EAB larvae tunnel through the bark
into the phloem and cambium where they feed on the nutrient rich tissues and develop through a
total of four instars (Poland and McCullough 2006). Tree death occurs when galleries made by
tunneling larvae cut off water and nutrient conducting pathways (Cappaert et al. 2005).
Ash species have varying levels of inherent resistance to EAB (Anulewicz et al. 2007;
Rebek et al. 2008; Duan et al. 2012; Tanis and McCullough 2012). Manchurian ash (Fraxinus
mandshurica Rupr.), an Asian species and blue ash (Fraxinus quadrangulata, Michx.), a North
American species, are relatively resistant to EAB (Liu et al. 2003, Duan et al. 2012, Tanis and
McCullough 2012; Tanis, Chapter three). In southeast Michigan woodlots, blue ash survival
was > 60% in 2011, several years after EAB populations peaked in the area (Tanis and
McCullough 2012). In contrast, green ash (Fraxinus pennsylvanica Marsh.), another North
American species, is heavily attacked by EAB and killed. In forested parklands in southeast
Michigan and northwest Ohio, green ash mortality exceeded 99% in 2009 (Herms et al. 2009;
Gandhi and Herms 2010b; Knight et al. 2010).

136

Blue ash, green ash and Manchurian ash are evolutionarily diverse. On a phylogenetic
tree based on floral morphology, blue ash, a monoecius species, is placed in the Dipetalae section
along with Fraxinus dipetala Hook. & Arn. and Fraxinus anomala Torr. ex S.Watson, both of
which are native to the southwestern United States (Wallander 2008). Green ash, a dioecious
species, is placed in the Melloides section with eight other North American species. Manchurian
ash, also a dioecious species, is placed in the Fraxinus section along with black ash (Fraxinus
nigra Marsh.) (Wallander 2008), a species that is highly susceptible to EAB (McCullough 2013).
Variability in ash tree resistance has sparked interest in ash physiology and defensive chemistry.
Ash tree resistance to EAB likely involves a suite of defensive mechanisms (Cipollini et
al. 2011), including the production of chemical defenses in the phloem, where EAB larvae feed
and develop (Eyles et al. 2007, Cipollini et al. 2011, Whitehill et al. 2012). Blue ash, green ash
and Manchurian ash have distinct phloem chemistry profiles. For example, several lignin
compounds that could function as larval antifeedants or growth retardants are unique to
Manchurian ash (Cipollini et al. 2011, Whitehill et al. 2012).
Confocal laser scanning microscopy (CLSM) can be used to detect the autofluorescence
of defensive compounds in living cells (Roshchina 2012). Autofluorescence is a natural
phenomenon that occurs when organic compounds emit a specific wavelength of light after
absorbing light from a different wavelength. For example, chloroplasts emit green light when
they have absorbed red light. Autofluorescence that is detected by CLSM can serve as a
diagnostic indicator (Roshchina 2012) since many phenolic compounds (e.g., hydroxycinnamic
acids, coumarins, stilbenes and styrlypyrones) autofluoresce when illuminated with blue lasers
(Hutzler et al. 1998; Pfündel et al. 2006; Roshchina 2012). In Norway spruce (Picea abies L.),
CLSM autofluorescence techniques were used to determine the location of phenolic compounds

137

in needles. As expected, autofluorescence was recorded in phloem near resin ducts, but high
concentrations of phenolic compounds were also present in the cell walls of the xylem (Hutzler
et al. 1998). While ash phloem chemistry has been the focus of considerable research (Eyles et
al. 2007, Cipollini et al. 2011, Whitehill et al. 2011, 2012, Hill et al. 2012), the anatomical
features that may affect larvae directly or indirectly through chemical defense production remain
unexplored. My first objective was to assess phenolic autofluorescence in the phloem of
resistant blue ash and Manchurian ash and highly susceptible green ash using CLSM.
In addition to defensive phenolic compounds, ash trees also likely produce a variety of
reactive oxygen species (ROS) when injured (Rajarapu et al. 2011). Reactive oxygen species are
biologically active compounds that contain oxygen. Peroxide, a common ROS produced by
plants, can serve as a defensive compound, defensive compound precursor or a signal for injury
(Maffei et al. 2007). When insects feed on plant tissues, high concentrations of ROS typically
flood the injured area in a phenomenon called an “oxidative burst” (Maffei et al. 2007). When
insects ingest these injured tissues, high concentrations of ROS can cause cellular oxidative
stress (lipid, DNA and/or protein degradation), which could potentially kill the insect. In order
to counteract cellular oxidation, insects must produce antioxidant enzymes to detoxify the ROS
compounds before their tissues are damaged (Rajarapu et al. 2011). Rajarapu et al. (2011)
suggest that the ability of EAB to produce antioxidant enzymes is at least partially responsible
for their “rampant invasion of North American ash”, but the extent to which different ash species
produce ROS is unknown.
®

The production of ROS in living tissues can be detected with CLSM. Amplex Red
(Invitrogen, Molelcular Probes Inc., Eugene OR, USA), a fluorescent probe that can be absorbed
by living cells (Maffei and Bossi 2006, Maffei et al. 2007, Ozawa et al. 2009, Bricchi et al.
138

2010). The reaction that occurs between Amplex Red and ROS produces a compound called
resorufin in a 1: 1 peroxide: resorufin ratio. When resorufin is produced, it emits red light that
can be detected with CLSM (Invitrogen.com, Accessed 14 April 2012). My second objective
was to assess ROS production in blue ash, green ash and Manchurian ash using Amplex Red and
CLSM.
Relatively few studies have examined how anatomical characteristics of deciduous trees
might contribute to defense against phloem boring larvae (Dunn et al. 1990, Cipollini et al.
2011). However, there is evidence for anatomical resistance mechanisms (i.e., lignification of
attacked tissue) (Heering, 1956, Lieutier, 2008, Dunn et al., 1986, Hanley et al., 2007). While
debarking blue ash and Manchurian ash trees for a concurrent study, I observed a distinct layer
of cells that could act as a barrier to tunneling larvae. The cell layer was visually apparent and so
hard that drawknives glanced off the layer instead of penetrating into the cambium. Many
studies incorporate CLSM to examine the internal architecture of wood because high resolution,
three-dimensional images can be acquired with relatively little tissue preparation (Kitin et al.
2004, 2009). My third objective was to examine phloem architecture of blue ash, green ash and
Manchurian ash with CLSM to more thoroughly evaluate variations in phloem structure that
could potentially affect EAB larvae.

139

METHODS
Ash Plantation
On 26 April 2006, blue ash, green ash and Manchurian ash trees were planted (2.4 m
apart) at the Michigan State University Tree Research Center in Okemos, Michigan (Ingham
County) (Tanis, Chapter Four). Trees were fertilized annually and watered twice per week or as
needed with drip irrigation. All trees used in our experiments were seven years old at the time of
experiments.
Tissue Sampling
Phenolic Compound Autofluorescence
On 31 July 2012, four tissue samples consisting of phloem, cambium and sapwood
(approximately 2.0

2.0

0.5 cm) were removed from breast height (1.4 m aboveground) with

a drawknife from the cardinal directions from three blue ash, green ash and Manchurian ash trees
(four samples per tree, 36 total samples). Samples were sealed in plastic bags, placed in an
opaque box to reduce exposure to light and immediately transported to the Michigan State
University Microscopy Center. Upon arrival, two thin cross sections approximately 200 µm
thick were removed from the transverse plane of each phloem sample (eight sections per tree, 72
total samples) with a razor blade. Samples were immediately analyzed for blue auto-florescence
with CLSM, they did not receive any dye treatments.
Reactive Oxygen Species (ROS) Production
Six to 10 additional cross sections were removed from each phloem sample as described
above and incubated in Amplex Red reagent in the dark for 1 hr per Amplex Red hydrogen
peroxide/peroxidase assay kit instructions (Bricchi et al. 2010, Invitrogen.com 2012). I
incubated an excess of cross sections in order to ensure I had multiple high quality sections to

140

choose from. The two thinnest cross sections from each phloem sample (eight sections from
each tree, 72 total samples) were analyzed using CLSM.
Phloem Anatomy
While evaluating phloem cross sections for autofluorescence, I also assessed potential
differences in anatomy. The unique cells observed while debarking blue ash and Manchurian
ash trees during a concurrent study were clearly visible through the oculars, but I was unable to
obtain a high resolution image because samples were living and the cytoplasm had not been
cleared. Therefore, I dyed, dehydrated and cleared the phloem cross sections that were used to
assess autofluorescence. Immediately after autofluorescence observations were complete, cross
sections were placed into fixing solution (50% ethanol, glycerol, formaldehyde, 18:1:1) for at
least 72 hours. After fixation, cross sections were placed in a solution of safranin orange dye
(1% safranin, Carolina Biological Supply Company, Burlington, NC, USA) for 15 minutes, then
dehydrated through 30, 50, 75, 90 and 100% ethanol (Decon Labs, Inc., King of Prussia, PA,
USA), then rehydrated backward through the same solutions. Cross sections remained in each
alcohol solution for 10 minutes. Solutions were changed multiple times during each cycle to
remove excess dye. After rehydrating, cross sections were placed in de-ionized water for 15
minutes then processed through 25, 50 and 75% glycerol solutions (1 hour each with frequent
decanting) and finally left overnight in 100% glycerol. Images were obtained with a confocal
microscope the next day.
Confocal Laser Scanning Microscopy
Confocal laser scanning microscopy images are obtained by raster scanning, a process
that uses lasers to scan point by point through a series of slices from the highest to the lowest
focused portions of the sample (Wymer et al. 1999). The number of slices (defined by the user)

141

depends on the thickness of the sample and the resolution necessary to produce an image of the
required quality (Hutzler et al. 1998). To obtain a three dimensional image, multiple slices
(called a z-scan) are compiled by software programs that detect and amass only the focused
portions of each slice (Wymer et al. 1999). After compiling an image, vertical and horizontal
fluorescent profiles can be overlain to show the frequency and intensity of fluorescence. I used
®

an Olympus Fluoview 1000 (Olympus Corporation, Tokyo, Japan) confocal microscope
equipped with a 10

dry objective for magnification throughout all experiments. The total area
2

of each image was 1.59 mm .
Phenolic Compound Autofluorescence
Phloem cross sections that were assessed for autofluorescence were illuminated with a
blue Diode (λ=405) laser to detect autofluorescence of phenolic compounds and a helium-neon
(λ=543) laser to show the basic structure of each sample. Three-dimensional images were
produced for each phloem cross section using a double channel z-scan. The double channel zscanning option enables the user to assign colors to images that have been produced by different
lasers. I chose blue to label phenolic autofluorescence (detected by the blue Diode laser) and
orange to label the structural architecture of the phloem (detected by the helium-neon laser)
(Figure 5.1). Colors are completely arbitrary and are used to clarify presentation. The number
of slices ranged from 24 to 81 per sample while slice depth was 11.5 µm for all samples.
Reactive Oxygen Species (ROS) Production
Phloem cross sections incubated with Amplex Red were illuminated with argon (λ=488)
and helium-neon (λ=543) lasers, a laser combination that detects the red fluorescence produced
when Amplex Red reacts with peroxide (Ozawa et al. 2009). Three-dimensional images were
produced for each phloem cross section and red was used to label areas where fluorescence was
142

present (Fig. 5.2). Slice depth was 5.0 µm for all images and the number of slices ranged from
10 to 31 per sample. Vertical and horizontal fluorescence frequency profiles were added to each
image to provide a reference for the intensity and distribution of fluorescence (Fig. 5.2).
Phloem Anatomy
Cross sections used to assess differences in phloem anatomy were illuminated with argon
and helium-neon lasers, a laser combination that detects the red fluorescence produced by
safranin dye. A three dimensional image was produced for each phloem cross section. Images
consisted of 34 to 79 slices per sample, taken 5.0 µm apart.

143

RESULTS
Phenolic Compound Autofluorescence
Phenolic autofluorescence was detected in blue ash and Manchurian ash phloem samples
but not in green ash phloem samples (Fig. 5.1). Phenolic autofluorescence was detected in 40 to
50% of the phloem area of every blue ash phloem sample. Autofluorescence located 290 to 370
µm (mean depth (±SE) = 323.3 ± 12.02 µm) beneath the outside edge of the bark but never
occurred within the bark (Fig. 1B1 and 1B2). Phenolic compound autofluorescence was never
detected in green ash phloem samples; all images were completely black (Fig. 1G2). In all
Manchurian ash samples, at least 70% of the phloem area contained autofluorescing phenolic
compounds (Fig 1.M2). These phenolic compounds were distributed throughout all samples and
were always present in the bark.
Reactive Oxygen Species (ROS) Production
Production of ROS was recorded in all phloem samples regardless of ash species.
Distribution and concentration of ROS, however, varied considerably among species. Where
ROS were produced within the phloem samples, the tissues appear bright red (Fig. 5.2).
Anomalies in fluorescent frequency profiles correspond to areas within each phloem sample
where ROS fluorescence was detected. These fluorescent profiles indicated by the horizontal
and vertical traces (Fig. 5.2), suggest the production of ROS occurred in localized areas within
the blue ash and green ash phloem samples (Figs. 5.2A and 5.2B). In contrast, ROS production
was more or less continuous in the Manchurian ash phloem (Fig. 5.2C). The images confirm the
patterns indicated by the frequency traces. Blue ash and green ash ROS fluorescence appeared
much more mottled, whereas ROS fluorescence appeared in two bands in the Manchurian ash
phloem (Fig. 5.2).

144

2

The area (µm ) where ROS were produced comprised 13.3 ± 1.2 % and 16.1 ± 2.6% of
the total blue ash and green ash phloem area, respectively. In contrast, the area where ROS were
produced in Manchurian ash phloem samples comprised 72.2 ± 3.2 %, which was 82 and 78%
higher than blue ash or green ash, respectively.
Phloem Anatomy
The layer of cells I observed when peeling blue ash and Manchurian ash trees was visible
in phloem tissues when I assessed phenolic autofluorescence. I determined that the layer of cells
was a sclerenchymatous cylinder as described by Wilson et al. (1983). Sclerenchymatous
cylinders can be either continuous or semi-continuous. Continuous cylinders are comprised of
interconnecting cells that occur in a ring. Semi-continuous cylinders are frequently interrupted
by parenchyma cells and groups of sclerenchyma cells look more like bundles rather than a
cylinder.
The blue ash sclerenchymatous cylinder was continuous (Fig. 5.3A). It was located 390
to 560 µm (mean (±SE) depth = 461.7 ± 18.3 µm) from the outside edge of the bark and was 40
to 75 µm thick (mean thickness = 52.0 ± 5.8 µm). In contrast, the green ash sclerenchymatous
cylinder was semi-continuous (Fig. 5.3B). The bundles of sclerenchyma occurred 290 to 490 µm
(mean depth = 330.0 ± 10.4 µm) from the outside edge of the bark and were between 60 and 95
µm thick (mean thickness = 82.0 ± 6.6 µm). As expected from our previous observations, the
Manchurian ash sclerenchymatous cylinder was also continuous in nature (Fig. 5.3C). It was
located 420 to 570 µm (mean depth = 495.0 ± 11.6 µm) beneath the outside edge of the bark and
was between 80 and 140 µm thick (mean thickness = 103 ± 5.4 µm).

145

DISCUSSION
In his classic review of tree resistance, Hanover (1975) suggests multiple ways trees can
defend themselves against insect pests. Resistant trees might produce defensive compounds to
attract insect predators or repel pests, they may have anatomical structures to deter insects,
and/or their nutritional content might be such that they are less preferred by or unsuitable for
insect pests. Any one of these mechanisms, or a combination thereof, could influence adult and
larval insect stages by altering feeding behavior, survivability and oviposition preference
(Hanover 1975). I found blue ash and Manchurian ash trees produce defensive chemicals and
have anatomical features in the phloem that could affect their resistance to EAB.
Detecting the autofluorescence of phenolic compounds in plant tissues with CLSM is a
new technique that is gaining popularity (Hutzler et al. 1998). Many naturally occurring
compounds produce autofluorescence of a known wavelength. Illumination of tissues with a
blue light source has been used successfully to identify the autofluorescence of coumarin and
polyresinol compounds in Norway spruce (Hutzler et al. 1998). Studies that have assessed the
phloem chemistry of ash species indicate these two classes of compounds are unique to resistant
ash species and could be harmful to EAB larvae (Whitehill et al. 2012). I chose to use blue
autofluorescence techniques to assess the distribution of phenolic compounds in ash phloem
because autofluorescence can detect an array of defensive compounds. This is important because
the compounds in ash that confer resistance to EAB remain unknown (Eyles et al. 2007,
Cipollini et al. 2011, Whitehill et al. 2012). Phenolic autofluorescence was intense in
Manchurian ash phloem and bark. It was present to a lesser degree in blue ash phloem, but was
completely absent in the blue ash bark. The complete lack of phenolic compound
autofluorescence in the green ash phloem was unexpected. Multiple studies have identified

146

phenolic compounds in green ash phloem (Eyles et al. 2007; Cipollini et al. 2011; Whitehill et al.
2012). Intraspecific differences in phenolic chemistries could be highly variable depending on
tree age, phenology and genotype (Cipollini et al. 2011; Whitehill et al. 2012). My study
provides visual evidence of the presence and distribution of phenolic compounds in Manchurian
ash and blue ash phloem and bark, but the exact identity of those compounds is unknown. The
CLSM technique apparently did not detect all phenolic compounds, as evidenced by the lack of
autofluorescence in green ash phloem tissues. The autofluorescence detected by CLSM,
however, indicates one or more phenolic compounds were present in blue ash and Manchurian
ash phloem that were not present in green ash phloem. Blue ash and Manchurian ash trees are
relatively resistant to EAB while green ash is highly susceptible, further research is necessary to
determine the identity of the autofluorescing compounds.
Many plants produce reactive oxygen species in response to herbivory (Peltonen et al.
2005, Witzell and Martin 2008). These compounds can serve directly as defensive compounds
but can also serve as signals that turn on stress-tolerance genes (Potters et al. 2007). Through
coevolution, many insects have developed antioxidant enzymes to metabolize and detoxify ROS.
Rajarapu et al (2011) suggest the ability to produce these enzymes is a trait often exhibited by
invasive insects, which could partially explain how EAB has been able to attack and infest a
variety of ash species native to North America.
Reactive oxygen species are one of the first compounds produced when plants are injured
(Bi and Felton 1995). Fluorescent CLSM has been used to detect and visualize ROS production
in leaves injured by mechanical means and feeding insects (Maffei and Bossi 2006, Maffei et al.
2007, Ozawa et al. 2009, Bricchi et al. 2010), but to our knowledge, CLSM has not been used to
visualize ROS production in ash phloem. I induced ROS production by cutting into the phloem

147

with a drawknife. Although this does not exactly mimic the injuries created by tunneling EAB
larvae, I feel the injury response was similar. Production of ROS occurred in all of the ash
species I assessed. While blue ash and green ash trees produced detectable levels of ROS, it was
highly localized in the phloem tissues and occurred in <20% of the phloem area. In contrast,
ROS appeared in two large swathes of tissue in Manchurian ash and occurred in almost 80% of
the phloem area.
In addition to phenolic and ROS chemical defenses, blue ash and Manchurian ash trees
may also utilize anatomical defenses in the form of a sclerenchymatous layer of tissue in the
phloem. Continuous or semi-continuous sclerenchyma cylinders have been described for a
variety of woody species (Wilson et al. 1983). In Populas tremula L. (Norberg and Meier 1966)
the sclerenchyma cylinders were described as crystalline and appeared to be highly organized.
In healthy, mature lilac (Syringa vulgaris L.), the sclerenchyma cylinder regenerated itself when
tissues were mechanically wounded but as trees became stressed they lost their regenerative
ability (Wilson et al. 1983).
Many studies suggest anatomical defenses can serve as important mechanisms for tree
resistance, but few studies actually investigate those mechanisms in hardwood trees (Dunn et al.
1986; Hanley et al. 2007; Mithöfer et al. 2009; Cippollini et al. 2011). In our concurrent host
preference study, the density of larvae on blue ash and Manchurian ash was so low (1.9 ± 0.93
2

and 1.5 ± 0.66 galleries per m of phloem, respectively) I was unable to determine how EAB
might be affected by the cylinder. Further investigation using artificially placed EAB eggs on
blue ash and Manchurian ash trees would be necessary to determine how larvae are affected, but
my assessment suggests the sclerenchyma layer could act as an obstacle for tunneling EAB
larvae.
148

Blue ash continues to be an anomaly among the North American ashes. I identified
phenolic compounds and a continuous sclerenchymatous cylinder in blue ash phloem, despite the
fact it has not coevolved with any North American EAB congeners (Drooz 1985). In contrast, I
detected no phenolic compound autofluorescence in green ash phloem and the sclerenchyma
cylinder was frequently interrupted with parenchyma cells. While ROS production was
comparable between blue ash and green ash, it was extremely low in both species compared to
Manchurian ash. Results indicate that upon hatching, neonate larvae likely encounter phenolic
compounds almost immediately when they begin to chew into the outer bark of Manchurian ash.
If larvae persist, they will continue to encounter defensive compounds and the sclerenchymatous
cylinder in Manchurian ash phloem. Larvae have no way to avoid or escape these tissues and are
subject to the defensive compounds they encounter therein. Although my results provide
evidence that blue ash and Manchurian ash trees are unique among ash trees commonly planted
in eastern North America, further research is necessary to determine how their phloem
characteristics and chemistry specifically affect EAB larvae.

149

Green

Blue

Manchurian

Bark
Bark
Phloem Surface

Bark

B1

200µm

G1

200µm

M1

200µm

Bark
Phenolic
Autofluorescence

Bark

B2

200µm

G2

200µm

M2

200µm

Figure 5.1: Transverse samples of blue ash (B1, B2), green ash (G1, G2) and Manchurian ash (M1, M2) phloem. Samples were
illuminated with a helium-neon laser (λ= 543 nm) to show basic phloem structure (B1, G1, M1) or a blue Diode laser
(λ=405) to detect phenolic compound autofluorescence (B2, G2, M2). Representative images were obtained with a
confocal laser scanning microscope equipped with a 10 dry objective. Three dimensional images of blue ash, green ash
or Manchurian ash consist of 79, 24 or 34 slices, respectively. Slices were obtained 11.5 µm apart. For interpretation of
the references to color in this and all other figures, the reader is referred to the electronic version of this dissertation.

150

A

200µm

B

200µm

C

200µm

Figure 5.2: Transverse cross sections of (A) blue ash, (B) green ash and (C) Manchurian ash phloem tissues. Tissues were incubated
with Amplex® Red peroxidase assay and illuminated with argon (λ=488) and helium neon (λ= 543 nm) lasers.
Representative images were obtained with a confocal laser scanning microscope equipped with a 10 dry objective. The
three dimensional images consist of 24, 11 or 12 slices, respectively, obtained 5.0 µm apart. Fluorescence profiles
correspond to tissues located under corresponding yellow horizontal or vertical lines. Lines were placed in areas where
fluorescence was strongest.

151

A

A

B

200µm

C

200µm

B

C

200µm

Figure 5.3: Transverse cross sections of (A) blue ash, (B) green ash and (C) Manchurian ash phloem tissues. Tissues were dyed with
safranin orange (1%) and illuminated with argon (λ=488) and helium neon (λ= 543 nm) lasers. The three dimensional
images consisted of 61, 47 and 56 slices, respectively, taken 5.0 µm apart. Representative images were obtained with a
confocal laser scanning microscope equipped with a 10 dry objective for magnification. Tissues directly above the
orange arc or the orange arrows indicate sclerenchymatous tissues.

152

LITERATURE CITED

153

LITERATURE CITED
Aber, J.D. 1992. Nitrogen cycling and nitrogen saturation in temperate forest ecosystems.
Trends Ecol. Evol. 7: 220-223.
Andersen, C.P., E.I. Sucoff, R.K. Dixon and A.H. Markhart III. 1989. Effects of phosphorus
deficiency on root hydraulic conductivity in Fraxinus pennsylvanica. Canadian Journal of
Botany, 1989, 67: 472-476.
Anulewicz, A.C., McCullough, D.G., and Cappaert, D.L. 2007. Emerald ash borer (Agrilus
planipennis) density and canopy dieback in three North American ash species.
Arboricult. Urban Forest. 33: 338–349.
Anulewicz, A.C., D.G. McCullough, D.L. Cappaert and T.M. Poland. 2008. Host range of the
emerald ash borer (Agrilus planipennis Fairmaire) (Coleoptera: Buprestidae) in North
America: results of multiple-choice field experiments. Environmental entomology 37:
230-241.
Arzani K., and H.R. Roosta. 2004. Effects of paclobutrazol on vegetative and reproductive
growth and leaf mineral content of mature apricot (Prunus armeniaca L.) trees. J. Agric.
Sci. Technol. 6: 43-55.
Aukema, J.E., Leung, B., Kovacs, K., Chivers, C., Britton, K.O., Englin, J., Frankel, S.J., Haight,
R.G., Holmes, T.P., Liebhold, A. M., McCullough, D.G., and Von Holle, B. 2011.
Economic impacts of non-native forest insects in the continental United States. PLoS
ONE, 6(9): e24587. doi:10.1371/journal.pone.0024587.
Bai, S., W. Chaney and Y. Qi. 2004. Response of cambial and shoot growth in trees treated with
paclobutrazol. Journal of Arboriculture 30: 137-145.
Bai, S., W.R. Chaney and Y. Qi. 2005. Wound closure in trees affected by paclobutrazol. Journal
of Arboriculture 31: 137-145.
Bailey, I.W. 1944. The development of vessels in angiosperms and its significance in
morphological research. American Journal of Botany 1944: 421-428.
Barclay, H.J. 1987. Combining methods of pest control: complementarity of methods and a
guiding principle. Nat. Resour. Model. 2: 299–323.
Barclay, H.J., and Li, C. 1991. Combining methods of pest control: minimizing cost during the
control program. Theor. Popul. Biol. 40(1): 105–123.
Barnes, B.V. and W.H. Wagner Jr. 2003. Michigan trees: revised and revisited. University of
Michigan Press, Ann Arbor, MI, USA.

154

Bauer, L.S., Liu, H., Miller, D., and Gould, J. 2008. Developing a classical biological control
program for Agrilus planipennis (Coleoptera: Buprestidae), an invasive ash pest in North
America. Newsl. Mich. Entomol. Soc. 53: 38–39.
Bi, J.L and G. W. Felton. 1995. Foliar oxidative stress and insect herbivory: primary compounds,
secondary metabolites, and reactive oxygen species as components of induced resistance.
Journal of Chemical Ecology 21: 1511-1530.
Bolte, A., T. Rahmann, M. Kuhr, P. Pogoda, D. Murach and K.V. Gadow. 2004. Relationships
between tree dimension and coarse root biomass in mixed stands of European beech
(Fagus sylvatica L.) and Norway spruce (Picea abies [L.] Karst.). Plant and Soil 264: 111.
Breedlove D.A., H.A. Holt and W.R. Chaney. 1989. Tree growth regulators: an annotated
bibliography. Purdue University Agriculture Experiment Station Resource Bulletin 989,
135p.
Bricchi, I., M. Leitner, M. Foti, A. Mithöfer, W. Boland and M. Maffei. 2010. Robotic
mechanical wounding (MecWorm) versus herbivore-induced responses: early signaling
and volatile emission in Lima bean (Phaseolus lunatus L.). Planta 232: 719-729.
Bryant, W.S., Wharton, M.E., Martin, W.H., and Varner, J.B. 1980. The blue ash – oak savanna
– woodland, a remnant of resettlement vegetation in the inner bluegrass of Kentucky.
Castanea J. S. Appalachian Bot. Club, 45: 149–165.
Bryant, J.P., Swihart, R.K., Reichardt, P.B., and Newton, L. 1994. Biogeography of woody plant
chemical defense against snowshoe hare browsing: comparison of Alaska and eastern
North America.Oikos, 70(3): 385–395.
Burch, P.L. and R.H. Wells. 1995. Tree growth evaluated 10 years after application of
paclobutrazol tree growth regulator. Down to Earth 50: Available online at:
V=http://www.treegrowthregulator.com/documents/Profile_Pac_article.pdf [Accessed 1
April 2013]
Burggraaf, P.D. 1972. Some observations on the course of the vessels in the wood of Fraxinus
excelsior L. Acta Bot Neerlandica 21: 32–47 (1972).
Burns, R.M., and Honkala, B.H. (Technical coordinators). 1990. Silvics of North America: 1.
Conifers; 2. Hardwoods. USDA Forest Service, Washington, D.C., Agriculture
Handbook 654.
Cappaert, D., McCullough, D.G., Poland, T.M., and Siegert, N.W. 2005. Emerald ash borer in
North America: a research and regulatory challenge. Am. Entomologist, 51: 152–165.

155

Cappaert, D., and McCullough, D.G. 2009. Occurrence and seasonal abundance of Atanycolus
cappaerti (Hymenoptera: Braconidae), a native parasitoid of emerald ash borer, Agrilus
planipennis (Coleoptera: Buprestidae). Great Lakes Entomol. 42: 16–29.
Carrow, J. R. and K. Graham. 1968. Nitrogen fertilization of the host tree and population growth
of the balsam woolly aphid, Adelges piceae (Homoptera: Adelgidae). Can. Entomol. 100:
478-485.
Carrow, J.R. and R.E. Betts. 1973. Effects of different foliar-applied nitrogen fertilizers on
Balsam woolly aphid. Can. J. For. Res. 3: 122-139.
Chaney, W.R. 2005a. A paclobutrazol treatment can leave a tree more stress tolerant. Golfdom,
Advanstar Publication. Available online at:
http://www.sepro.com/documents/profile_article.pdf [Accessed 16 January 2013].
Chaney, W.R. 2005b. Growth retardants: a promising tool for managing urban trees. Purdue
University extension bulletin FNR-252-W.
Chen, Y. and T.M. Poland. 2009a. Interactive influence of leaf age, light intensity, and girdling
on green ash foliar chemistry and emerald ash borer development. J Chem Ecol 35: 806–
815.
Chen, Y. and T.M. Poland. 2009b. Biotic and abiotic factors affect green ash volatile production
and emerald ash borer adult feeding preference. Environ. Entomol. 38: 1756-1764.
Chen, Y., and Poland, T.M. 2010. Nutritional and defensive chemistry of three North American
ash species: possible roles in host performance by emerald ash borer adults. Great Lakes
Entomol. 43: 20–33.
Chen, Y., T. Ciaramitaro and T.M. Poland. 2011a. Moisture content and nutrition as selection
forces for emerald ash borer larval feeding behaviour. Ecological Entomology 36: 344354.
Chen, Y., J.G.A. Whitehill, P. Bonello, and T.M. Poland. 2011b. Differential response in foliar
chemistry of three ash species to emerald ash borer adult feeding. J. Chem. Ecol. 37(1):
29–39.
Chen, Y., J.G.A. Whitehill, P. Bonello and T.M. Poland. 2011c. Feeding by emerald ash borer
larvae induces systemic changes in black ash foliar chemistry. Phytochemistry 72: 1990–
1998

Chinese Academy of Science, Institute of Zoology. 1986. Agrilus marcopoli Obenberger
[online]. In Agriculture insects of China (part I). p. 445. China Agriculture Press, Beijing,
China. Available at http://www.ncrs.fs.fed.us/dev/4501/eab/downloads/biologyCAS1986.pdf [online translation accessed 20 December 11].

156

Choat, B., A.R. Cobb and S. Jansen. 2007. Structure and function of bordered pits: new
discoveries and impacts on whole-plant hydraulic function. New Phytologist 177: 608626.
Chorbadjian, R.A., P. Bonello and D.A. Herms. 2011. Effect of the growth regulator
paclobutrazol and fertilization on defensive chemistry and herbivore resistance of
Austrian pine (Pinus nigra) and paper birch (Betula papyrifera). Arboriculture and Urban
Forestry 37: 278-287.
Cipollini, D., Q. Wang, J.G.A. Whitehill, J.R. Powell, P. Bonello, and D.A. Herms. 2011.
Distinguishing defensive characteristics in the phloem of ash species resistant and
susceptible to emerald ash borer. J. Chem. Ecol. 37(5): 450–459.
Clifford, D.R., L.R. Cooke, P. Gendle and M.E. Holgate. 1977. Performance of carbendazim
formulations injected for the control of Dutch elm disease. Annals of Applied Biology
85: 153-156.
Coffelt, M.A., P.B. Schultz and T.J. Banko. 1993. Tree growth regulator influences orangestriped oakworm (Lepidoptera: Saturniidae) development and survival. Ecology and
Behavior: Journal of Economic Entomology 86: 1446-1452.
Coley, P.D., J.P. Bryant and F.S. Chapin. 1985. Resource availability and plant antiherbivore
defense. Science 230: 895–899.
Crook, D.J. and V.C. Mastro. 2010. Chemical ecology of the emerald ash borer Agrilus
planipennis. J. Chem. Ecol. 36:101-12.
Dalziel, J. and D.K. Lawrence. 1984. Biochemical and biological effects of kaurene oxidase
inhibitors, such as paclobutrazol. Biochemical aspects of synthetic and naturally
occurring PGRs. British Plant Growth Regulators 11: 43-57.
Davis, T.D., N. Shankhla, R.H. Walser and A. Upadhyaya. 1985. Promotion of adventitious root
formation on cuttings by paclobutrazol. Hortic Sci 20: 883-884.
Davis, T.D. and E.A. Curry. 1991. Chemical regulation of vegetative growth. Critical reviews in
Plant Sciences 10: 151-188.
De Boer, A.H and V. Volkov. 2003. Logistics of water and salt transport through the plant:
structure and functioning of the xylem. Plant, Cell and Environment 26: 87-101.
Dickison, W.C. 2000. Integrative plant anatomy. Academic Press. San Diego, CA.
Dirr, M.A. 1998. Manual of Woody Landscape Plants: Their Identification, Ornamental
Characteristics, Culture, Propagation and Uses. 5th edition Stipes Pub Llc., Champaign,
IL, USA. 1187 pp.

157

Dobson, M.C. 1991. De-icing salt damage to trees and shrubs. Forestry Commission Bulletin
101.
Doccola, J.L., D.R. Smitley, T.W. Davis, J.J. Aiken, and P.M. Wild. 2011. Tree wound
responses following systemic injection treatments in Green ash (Fraxinus pennsylvanica
Marsh.) as determined by destructive autopsy. Arboriculture and Urban Forestry 37: 612.
Doccola, J. J. and P.M. Wild. 2012. Tree Injection as an Alternative Method of Insecticide
Application. Insecticides – basics and other applications: 61.
Domec, J., B. Lachenbruch, and F.C. Meinzer. 2006. Bordered pit structure and function
determine spatial patterns of air seeding thresholds in xylem of Douglas-fir (Psuedotsuga
menziesii; Pinaceae) trees. American Journal of Botany 93: 1588-1600.
Drexhage, M. and F. Colin. 2001. Estimating root system biomass from breast‐height diameters.
Forestry 74: 491-497.
Drooz, A.T. 1985. Insects of eastern forests. USDA Forest Service, Washington, D.C., Misc.
Publ. 1426.
Duan, J.J., Fuester, R.W., Wildonger, J., Taylor, P.B., Barth, S., and Spichiger, S.E. 2009.
Parasitoids attacking the emerald ash borer (Coleoptera: Buprestidae) in western
Pennsylvania. Fla. Entomol. 92(4): 588–592.
Duan, J.J., M.D. Ulyshen, L.S. Bauer, J. Gould and R.V. Driesche. 2010. Measuring the impact
of biotic factors on populations of immature emerald ash borers (Coleoptera:
Buprestidae). Environmental Entomology 39: 1513-1522.
Duan, J.J., Bauer, L.S., Ulyshen, M.D., Gould, J.R., and Van Driesche, R. 2011. Development of
methods for the field evaluation of Oobius agrili (Hymenoptera: Encyrtidae) in North
America, a newly introduced egg parasitoid of the emerald ash borer (Coleoptera:
Buprestidae). Biol. Control, 56(2): 170–174.
Duan, J.J., L.S. Bauer, K.J. Abell and R. Van Driesche. 2012a. Population responses of
hymenopteran parasitoids to the emerald ash borer (Coleoptera: Buprestidae) in recently
invaded areas in north central United States. BioControl 57: 199-209.
Duan, J.J., Bauer, L.S., Hansen, J.A., Abell, K.J., and Van Driesched, R. 2012b. An improved
method for monitoring parasitism and establishment of Oobius agrili (Hymenoptera:
Encyrtidae), an egg parasitoid introduced for biological control of the emerald ash borer
(Coleoptera: Buprestidae) in North America. Biol. Control. 60(3): 255–261.
Dunn, J.P., T.W. Kimmerer and G.L. Nordin. 1986. The role of host tree condition in attack of
white oaks by the twolined chestnut borer Agrilus bilineatus (Weber) (Coleoptera:
Buprestidae). Oecologia 70: 596-600.
158

Dunn, J.P., D.A. Potter and T.W. Kimmerer. 1990. Carbohydrate reserves, radial growth, and
mechanisms of resistance of oak trees to phloem-boring insects. Oecologia, 83(4), 458468.
Drooz, A.T. 1985. Insects of Eastern Forests. USDA Forest Service. Misc. Publ. 1426.
Washington, D.C., USA. 608 p.
EAB.info. 2013. Emerald ash borer information website [online]. www.emeraldashborer.info
[Accessed 30 March 2013].
El-Khashab, A.A., A.F. El-Sammak, A.A. Elaidy, M.I. Salama and M. Rieger. 1997.
Paclobutrazol reduces some negative effects of salt stress in peach. J. Amer. Soc. Hort.
Sci. 122: 43-46.
Ewers, F.W. 1985. Xylem structure and water conduction in conifer trees, dicot trees and lianas.
IAWA Bulletin n.s. 6: 309-317.
Eyles, A., Jones, W., Riedl, K., Cipollini, D., Schwartz, S., Chan, K., Herms, D.A., and Bonello,
P. 2007. Comparative phloem chemistry of Manchurian (Fraxinus mandshurica) and two
North American ash species (Fraxinus americana and Fraxinus pennsylvanica). J. Chem.
Ecol. 33(7): 1430–1448.
Eyles, A., P. Bonello, R. Ganley, and C. Mohammed. 2009. Induced resistance to pests and
pathogens in trees. New Phytologis 185: 893-908.
Farrer, J.L. 1995. Trees in Canada. Fitzhenry and Whiteside and Canadian Forest Service,
Markham, Ontario, Canada.
Ferrini, F. and M. Baletto. 2006. Response to fertilization of different tree species in the urban
environment. Arboriculture and Urban Forestry 32: 93-99.
Fletcher, R.A. and A. Gilley. 2000. Triazoles as plant growth regulators and stress protectants.
Hort. Rev. 24: 55–138.
Gandhi, K.J.K., and Herms, D.A. 2010a. Direct and indirect effects of alien insect herbivores on
ecological processes and interactions in forests of eastern North America. Biol. Invasions,
12(2): 389–405.
Gandhi, K.J.K., and Herms, D.A. 2010b. Potential biodiversity loss due to impending
devastation of the North American genus Fraxinus by the exotic emerald ash borer. Biol.
Invasions, 12: 1839–1846.
Gao, R., T. Zhao, H. Liu, L.S. Bauer and T.P. Petrice. 2004. Distribution and damage survey of
Agrilus planipennis (Coleoptera: Buprestidae) in China. Trans. of China Pulp and Paper
19: 363–365.
159

George, A.P. and R.J. Nissen. 1992. Effects of water stress, nitrogen and paclobutrazol on
flowering, yield and fruit quality of the low-chill peach cultivar, ‘Flordaprince’. Scientia
Horticulturae 49: 197-209.
Gilman, E.F. 2004. Effects of amendments, soil additives, and irrigation on tree survival and
growth. Journal of Arboriculture 30: 301-310.
Gottschalk. USDA Forest Service, Northern Research Station, Newtown Square, Pennsylvania,
Gen. Tech. Rep. NRS-P-36.
Gouingené, S. P. and T.C.J. Turlings. 2002. The effects of abiotic factors on induced volatile
emissions in corn plants. Plant Physiol. 129: 1296-1307.
Haas, H.U. 2012. New aspects of plant regulators. In: W. Kramer, U. Schirmer, P. Jeschke and
M. Witschel (editors), Modern Crop Protection Compounds Wiley-VCH Verlag GmbH
and company KGaA. Pp 523-534.
Haase, D.L. and R. Rose. 1995. Vector analysis and its use for interpreting plant nutrient shift in
response to silvicultural treatments. Forest Science 41: 54-66.
Hahn, J., D.A. Herms and D.G. McCullough. 2011. Frequently asked questions regarding
potential side effects of systemic insecticides used to control emerald ash borer. Multistate extension bulletin. 4 p.
Hanley, M.E., B.B. Lamont, M.M. Fairbanks, and C.M. Rafferty. 2007. Plant structural traits
and their role in anti-herbivore defense. Perspectives in Plant Ecology, Evolution and
Systematics 8: 157-178.
Hanover, J.W. 1975. Physiology of tree resistance to insects. Annu. Rev. Biocontrol. 20: 75-95.
Harris, R.W. 1992. Arboriculture: Integrated Management of Landscape Trees, Shrubs, and
Vines, 2nd ed. Prentice Hall, Englewood Cliffs, NJ. 592 pp.
Harris, J.R., S.D. Day and B. Kane. 2008. Nitrogen fertilization during planting and
establishment of the urban forest: a collection of five studies. Urban Forestry and Urban
Greening 7: 195-206.
Hasselkus, E.R. and E.E. Sculte. 2010. Tree and Shrub Fertilization. Produced by the
Department of Agricultural Journalism, University of Wisconsin-Madison. Extension
Bulletin R-10-89-2.5M-25-E.
Heering, H. 1956. Zur Biologie, Okologie und zum Massenwechsel des Buchenpractkafers
(Agrilus vividis L.) Z Angew Entomol (38): 250-267.

160

Hendrick, R.L. and K.S. Pregitzer. 1993. The dynamics of fine root length, biomass, and
nitrogen content in two northern hardwood ecosystems. Canadian Journal of Forest
Research 23: 2507-2520.
Herms, D.A. 2002. Effects of fertilization on insect resistance of woody ornamental plants:
reassessing an entrenched paradigm. Environ. Entomol. 31: 923-933.
Herms, D.A., Gandhi, K.J.K., Smith, A., Cardina, J., Knight, K.S., Herms, C.P., Long, R.P., and
McCullough, D.G. 2009a. Ecological impacts of emerald ash borer in forests of southeast
Michigan. In Proceedings, 20th U.S. Department of Agriculture Interagency Research
Forum on Invasive Species, Annapolis, Maryland., 13– 16 January 2009. Edited by K.A.
McManus and K.W. Gottschalk. USDA Forest Service, Northern Research Station,
Newtown Square, Pennsylvania, Gen. Tech. Rep. NRS-P-51. pp. 36–37.
Herms, D.A., D.G. McCullough, D.R. Smitley, C. Sadof and P.L. Nixon. 2009b. Insecticide
options for protecting ash trees from emerald ash borer. North Central IPM Center
Bulletin. 12pp.
Hill, A.L., J.G.A. Whitehill, S.O. Opiyo, P.L. Phelan and P. Bonello. 2012. Nutritional attributes
of ash (Fraxinus spp.) outer bark and phloem and their relationships to resistance against
the emerald ash borer. Tree Physiology 32: 1522-1532.
Huett, D.O., A.P. George, J.M. Slack, and S.C. Morris. 1997. Diagnostic leaf nutrient standard
for low-chill peaches in subtropical Australia. Aust. J. Exp. Agric. 37: 119-126.
Hutzler, P., R. Fischbach, W. Heller, T.P. Jungblut, S. Reuber, R. Schmitz, M. Veit, G.
Weissenböck, and J.-P. Schnitzler. 1998. Tissue localization of phenolic compounds in
plants by confocal laser scanning microscopy. Journal of Experimental Botany 49: 953965.
Imo, M. and V.R. Timmer. 1999. Vector competition analysis of black spruce seedling responses
to nutrient loading and vegetation control. Canadian Journal of Forest Research 29: 474486.
Invitrogen.com. 2009. Amplex® red hydrogen peroxide/peroxidase assay kit manual. Online at:
www.invitrogen.com. Accessed 1 July 2012.
Kent, M., and Coker, P. 1992. Vegetation description and analysis: a practical approach. John
Wiley and Sons, New York.
Kitin, P.B., T. Fujii, H.Abe and R. Funada. 2004. Anatomy of the vessel network within and
between tree rings of Fraxinus lanuginose (Oleaceae). American Journal of Botany 91:
779-788.
Kitin, P., T. Fujii, H. Abe and K. Takata. 2009. Anatomical features that facilitate radial flow
across growth rings and from xylem to cambium in Cryptomera japonica. Annals of
Botany 103: 1145-1157.
161

Kitin, P., S.L.Voelker, F.C. Meinzer, H. Beeckman, S.H. Strauss and B. Lachenbruch. 2010.
Tyloses and phenolic deposits in xylem vessels impede water transport in low-lignin
transgenic poplars: a study by cryo-fluorescence microscopy. Plant Physiology 154:
887-898.
Klooster, W.S., B.M. Cregg, R.T. Fernandez, and P. Nzokou. 2012. Growth and physiology of
deciduous shade trees in response to controlled-release fertilizer. Scientia Horticulturae,
135: 71-79.
Knight, K.S., Herms, D.A., Cardina, J., Long, R., Rebbeck, J., Gandhi, K.J.K., Smith, A.,
Klooster, W.S., Herms, C.P., and Royo, A.A. 2010. Emerald ash borer aftermath forests:
the dynamics of ash mortality and the responses of other plant species. In Proceedings of
Symposium on Ash in North America, West Lafayette, Indiana, 9–11 March 2010. Edited
by C.H. Michler and M.D. Ginzel. USDA Forest Service, Northern Research Station,
Newtown Square, Pennsylvania, Gen. Tech. Rep. NRS-P-72.
Knight, K.S., D. Herms, R. Plumb, E. Sawyer, D. Spalink, E. Pisarczyk, E. Ziegler and K.
Menard. 2011. Dynamics of surviving ash (Fraxinus spp.) Populations in areas long
th
infested by emerald ash borer (Agrilus planipennis). In Proceedings of the 4
International Workshop on Genetics of Host-Parasite Interactions in Forestry, 31 July–5
August 2011, Eugene, OR. General Technical Report PSW-GTR-240. [online]
http://www.fs.fed.us/psw/publications/documents/psw_gtr240/psw_gtr240_143.pdf
Koike, T. and M. Sanada. 1989. Photosynthesis and leaf longevity in alder, birch and ash
seedlings grown under different nitrogen levels. In Annales des Sciences Forestieres 46:
476s-478s).
Kovacs, K.F., Mercader, R.J., Haight, R.G., Siegert, N.W., McCullough, D.G., and Liebhold,
A.M. 2011. The influence of satellite populations of emerald ash borer on projected
economic costs in U.S. communities, 2010–2020. J. Environ. Manage. 92(9): 2170–
2181.
Kozlowski, T.T. and C.H. Winget. 1963. Patterns of water movement in forest trees. Botanical
Gazette 124: 301-311.
Krishnan, N. and d. Kodrik. 2006. Antioxidant enzymes in Spodoptera littoralis (Boisdubal): are
they enhanced to protect gut tissues during oxidative stress? Journal of Insect Physiology
52: 11-20.
Kytö, M., P. Niemela, and S. Larsson. 1996. Insects on trees: population and individual response
to fertilization. Oikos 75: 148-159.
Lawson, A.B. and D.L. Dahlsten. 2003. Evaluation of systemic insecticides as a treatment option
in integrated pest management of the elm leaf beetle, Xanthogaleruca luteola (Müller)
(Coleoptera: Chrysomelidae). Journal of Economic Entomology 96: 1455-1462.
162

LeGoff, N. and J.M. Ottorini. 2001. Root biomass and biomass increment in a beech (Fagus
sylvatica L.) stand in North-East France. Annals of forest science, 58: 1-13.
Liebold, A.M., K.W. Gottschalk, R-M Muzika, M.E. Montgomery, R.Young, K. O’Day and B.
Kelley. 1995. Suitability of North American tree species to the gypsy moth: a summary
of field and laboratory tests. United States Department of Agriculture, Forest Service,
Northeastern Forest Experiment Station, General Technical Report NE-211.
Lieutier, F. 2008. Changing forest communities: role of tree resistance to insects in insect
invasions and tree introductions. In Invasive forest insects, introduced forest trees, and
altered ecosystems. Edited by T.D. Paine. Springer Science Media, Dordrecht, the
Netherlands. pp. 15–51.
Limback, C.K. 2010. Tree vigor and its relation to emerald ash borer (Agrilus planipennis
Fairmaire) adult host preference and larval development on green and white ash trees.
M.S. thesis, Department of Entomology, Michigan State University, East Lansing,
Michigan.
Liu, H., L.S. Bauer, R. Gao and T. Zhao. 2003. Exploratory survey for the emerald ash borer,
Agrilus planipennis (Coleoptera: Buprestidae), and its natural enemies in China. The
Great Lakes Entomologist 36: 191-204.
Liu, H.P., Bauer, L.S., Miller, D.L., Zhao, T.H., Gao, R.T., Song, L. W., Luan, Q.S., Jin, R.Z.,
and Gao, C.Q. 2007. Seasonal abundance of Agrilus planipennis (Coleoptera:
Buprestidae) and its natural enemies Oobius agrili (Hymenoptera: Encyrtidae) and
Tetrastichus planipennisi (Hymenoptera: Eulophidae) in China. Biol. Control, 42(1): 61–
71.
Lloyd, J.E., D.A. Herms, M.A. Rose and J. Van Wagoner. 2006. Fertilization rate and irrigation
scheduling in the nursery influence growth, insect performance, and stress tolerance of
‘Sutyzan’ crabapple in the landscape. HortScience 41: 442-445.
Loh, F.C.W., J.C. Grabosky and N.L. Bassuk. 2002. Using the SPAD-502 meter to assess
chlorophyll and nitrogen content of Benjamin fig and cottonwood leaves. Hort.Technol.
12: 682–686.
Lyons, D.B., P. de Groot, G.C. Jones and R. Scharbach. 2009. Host selection by Agrilus
planipennis (Coleoptera: Buprestidae): inferences from sticky-band trapping. The
Canadian Entomologist, 141: 40-52.
MacFarlane, D.W and S. Patterson Meyer. 2003. Characteristics and distribution of potential ash
tree hosts for emerald ash borer. Forest Ecology and Management 213: 15–24.

163

Maffei, M. and S. Bossi. 2006. Electrophysiology and plant responses to biotic stress. In: Plant
Electrophysiology – Theory and Methods (edited by: Volkov). Springer-Verlag,
Heidelberg. Pp 461-481.
Maffei, M.S., A. Mithöfer and W. Boland. 2007. Insects feeding on plants: rapid signals and
responses preceding the induction of phytochemical release. Phytochemistry 68: 29562959.
Mann, M.P., H.A. Holt, W.R. Chaney, W.L. Mills and R.L. McKenzie. 1995. Tree growth
regulators reduce line clearance trimming time. Journal of Arboriculture 21: 209–212.
Marenco, R.A., Antezana-Vera, S MacFarlane, D.W and S. Patterson Meyer. 2003.
Characteristics and distribution of potential ash tree hosts for emerald ash borer. Forest
Ecology and Management 213: 15–24.
Marion, D.F. 2003. Tree fertilization in the 21st century: where to from here? In A. Siewert, B.
Rao, and D. Marian (Editors) Tree and Shrub Fertilization. Dixon Graphics Publishing,
Champaign, IL. Pp 3-5.
Marshall, J.G., R.G. Rutledge, E. Blumwald and E.B. Dumbroff. 2000. Reduction in turgid water
volume in jack pine, white spruce and black spruce in response to drought and
paclobutrazol. Tree Physiology 20: 701–707.
Marshall, J.G., R.G. Rutledge, E. Blumwald and D.F. Marion. 2003. Tree fertilization in the 21
century: where to from here? In A. Siewert, B. Rao, and D. Marian (Editors) Tree and
Shrub Fertilization. Dixon Graphics Publishing, Champaign, IL. Pp 3-5.

st

Martinez-Trinidad, T., W.T. Watson and R.K. Book. 2011. Impacts of paclobutrazol on rootpruned live oak. HortTechnology 21: 46-50.
McCullough, D.G. and H.M. Kulman. 1991. Effects of nitrogen fertilization on young jack pine
(Pinus banksiana) and on its suitability as a host for jack pine budworm (Choristoneura
pinus pinus) (Lepidoptera: Tortricidae). Canadian Journal of Forest Research 21: 14471458.
McCullough, D.G., and Siegert, N.W. 2007a. Estimating potential emerald ash borer
(Coleoptera: Buprestidae) populations using ash inventory data. J. Econ. Entomol.
100(5): 1577–1586.
McCullough, D.G., and Siegert, N.W. 2007b. Using girdled trap trees effectively for emerald ash
borer detection, delimitation and survey. [Online.] Available at
http://www.emeraldashborer.info/files/Using_Girdled_Trap_Trees_Effectively_For_EAB
_Detection_Delimination_&_Survey.pdf [accessed 5 March 2012].
McCullough, D.G., T.M. Poland, A.C. Anulewicz, P. Lewis and D. Cappaert. 2007. Evaluation
of Agrilus planipennis (Coleoptera: Buprestidae) control provided by emamectin
164

benzoate and two neonicotinoid insecticides, one and two seasons after treatment. J Econ
Entomol. 104:1599-612.
McCullough, D.G., Poland, T.M., and Cappaert, D. 2009a. Attraction of the emerald ash borer to
ash trees stressed by girdling, herbicide treatment, or wounding. Can. J. For. Res. 39(7):
1331–1345.
McCullough, D.G., Poland, T.M., Anulewicz, A.C., and Cappaert, D. 2009b. Emerald ash borer
(Coleoptera: Buprestidae) attraction to stressed or baited ash trees. Environ. Entomol.
38(6): 1668–1679.
McCullough, D.G., T.M. Poland, A.C. Anulewicz, P. Lewis and D. Cappaert. 2011. Evaluation
of Agrilus planipennis (Coleoptera: Buprestidae) control provided by emamectin
benzoate and two neonicotinoid insecticides, one and two seasons after treatment. Journal
of Economic Entomology 104: 1599-1612.
McCullough, D.G. 2013. Will we kiss our ash goodbye? American Forests188: 16-22.
McKenzie, N., B. Helson, D. Thompson, C. Otis, J. Mcfarlane, T. Buscarini and J. Meating.
2010. Azadirachtin: An effective systemic insecticide for control of Agrilus planipennis
(Coleoptera: Buprestidae). Journal of economic entomology, 103(3), 708-717.
Mengel K. and E.A. Kirkby. 1982. Principles of Plant Nutrition. 4th edition. International Potash
Institute, Bern, Switzerland, 687 p.
Mercader, R.J., Siegert, N.W., Liebhold, A.M., and McCullough, D. G. 2011a. Influence of
foraging behavior and host spatial distribution on the localized spread of the emerald ash
borer, Agrilus planipennis. Popul. Ecol. 5 (2): 271–285.
Mercader, R.J., Siegert, N.W., Liebhold, A.M., and McCullough, D. G. 2011b. Simulating the
effectiveness of three potential management options to slow the spread of emerald ash
borer (Agrilus planipennis) populations in localized outlier sites. Can. J. For. Res. 41(2):
254–264.
Miller, R.O., Bloese, P.D., Hanover, J.W., and Haack, R.A. 1991. Paper birch and European
white birch vary in growth and resistance to bronze birch borer. J. Am. Soc. Hortic. Sci.
116: 580–584.
Miller, R.W. 1998. Tree fertilization: science, myth and ethics. Arborist News 6: 25-27.
Mithöfer, A., W. Boland and M.E. Maffei. 2009. Chemical ecology of plant-insect interactions.
Plant Disease Resistance. Wiley-Blackwell, Chichester, UK, 261-291.
Mota-Sanchez, D., B.M. Cregg, D.G. McCullough, T.M. Poland and R.M. Hollingworth. 2009.
Distribution of trunk-injected 14C-imidacloprid in ash trees and effects on emerald ash
borer (Coleoptera: Buprestidae) adults. Crop Protection 28: 655-661.
165

Nadezhdina, N. and J. Čermák. 2003. Instrumental methods for studies of structure and
function of root systems of large trees. Journal of Experimental Botany 54: 1511-1521.
Navarro, A., M.J. Sánchez-Blanco, A. Morte and S. Bañón. 2009. The influence of mycorrhizal
inoculation and paclobutrazol on water and nutritional status of Arbutus unedo L.
Environmental and Experimental Botany 66: 262-271.
Nielsen, D.G., Muilenburg, V.L., and Herms, D.A. 2011. Interspecific variation in resistance of
Asian, European, and North American birches (Betula spp.) to bronze birch borer
(Coleoptera: Buprestidae). Environ. Entomol. 40(3): 648–653.
Norberg, P.H. and H. Meier. 1966. Physical and chemical properties of the gelatinous wood
fibres of aspen (Populas tremula L.). Holzforschung 20: 174-178.
Oszako, T., M. Siewniak and M. Siewniak. 2007. The negative influence of the so called
“injection” measure on threatened horse chestnut trees. Alien Invasive Species and
International Trade 2007: 123-132.
Ott, R.L. and M. Longnecker. 2001. An Introduction to Statistical Methods and Data Analysis.
Duxbury Thomson Learning, Inc.
Ozawa R., C.M. Bertea, M. Foti, R. Narayana, G. Arimura, A. Muroi, J. Horiuchi, T. Nishioka,
M. Maffei, and J. Takabayashi. 2009. Exogenous polyamines elicit herbivore-induced
volatiles in lima bean leaves: involvement of calcium, H2O2 and jasmonic acid. Plant
Cell Physiology 50: 2183-2199.
Pallardy, S.G., Nigh, T.A., and Garrett, H.E. 1988. Changes in forest composition in central
Missouri: 1968–1982. Am. Midl. Nat. 120 (2): 380–390.
Peltonen, P.A., E. Vapaavuori and R. Julkunen‐tiitto. 2005. Accumulation of phenolic
compounds in birch leaves is changed by elevated carbon dioxide and ozone. Global
Change Biology 11: 1305-1324.
Percival, G.C., I.P. Keary and S. AL-Habsi. 2006. An assessment of the drought tolerance of
Fraxinus genotypes for urban landscape plantings. Urban Forestry and Urban Greening 5:
17-27.
Percival, G.C. and A.M.S. AlBalushi. 2007. Paclobutrazol – induced drought tolerance in
containerized English and evergreen oak. Arboriculture and Urban Forestry 33: 397-409.
Pfündel, E.E., G. Agati and Z.G. Cerovic. 2006. Optical properties of plant surfaces. In: Biology
of the Plant Cuticle (Edited by M. Riederer and Müller). Wiley-Blackwell Publishing. Pp
216-249.

166

Poland, T.M., and McCullough, D.G. 2006. Emerald ash borer: invasion of the urban forest and
the threat to North America’s ash resource. J. For. 104 (3): 118–124.
Poland, T.M., R.A. Haack, T.R. Petrice, D.L. Miller and L.S. Bauer. 2006. Laboratory evaluation
of the toxicity of systemic insecticides for control of Anoplophora glabripennis and
Plectrodera scalator (Coleoptera: Cerambycidae). J Econ Entomol 99: 85–93.
Poland, T.M., McCullough, D.G., and Anulewicz, A.C. 2011. Evaluation of double-decker traps
for emerald ash borer (Coleoptera: Buprestidae). J. Econ. Entomol. 104(2): 517–531.
Pontius, J.A., R.A. Hallett and J.C. Jenkins. 2006. Foliar chemistry linked to infestation and
susceptibility to hemlock woolly adelgid (Homoptera: Adelgidae). Environmental
Entomology, 35: 112-120.
Potters, G., T.P. Pasternak, Y. Guisez, K.J. Palme and M.A.K. Jansen. 2006. Stress-induced
morphogenic responses: growing out of trouble. TRENDS in Plant Science 12: 98-105.
Prasad, A.M., Iverson, L.R., Matthews, S., and Peters, M. 2007– ongoing. A climate change atlas
for 134 forest tree species of the eastern United States. [Online.] USDA Forest Service,
Northern Research Station, Delaware, Ohio. Available at www.nrs.fs.fed.us/atlas/tree
[accessed 5 March 2012].
Premachandra, G.S., W.R. Chaney and H.A. Holt. 1997. Gas exchange and water relations of
Fraxinus americana affected by flurprimidol. Tree Physiology 17: 97-103.
Prodhan, A.K.M.A., R. Funada, J. Ohtani, H. Abe and K. Fukazawa. 1995. Orientation of
microfibrils and microtubules in developing tension-wood fibres of Japanese ash
(Fraxinus mandshurica var. japonica). Planta 196: 577-585.
Pugh, S.A., Liebhold, A.M., and Morin, R.S. 2011. Changes in ash tree demography associated
with emerald ash borer invasion, indicated by regional forest inventory data from the
Great Lakes States. Can. J. For. Res. 41(11): 2165–2175.
Pureswaran, D.S., and Poland, T.M. 2009. Host selection and feeding preference of Agrilus
planipennis (Coleoptera: Buprestidae) on ash (Fraxinus spp.). Environ. Entomol. 38(3):
757–765.
Rademacher, W. 2000. Growth retardants: effects on gibberellin biosynthesis and other
metabolic pathways. Annual Rev. Plant Physiol. Plant Mol. Biol. 51: 501-531.

Raese, J.T. and E.C. Burts. 1983. Increased yield and suppression of shoot growth and mite
populations of ‘d'Anjou’ pear trees with nitrogen and paclobutrazol. HortScience 18:
212-214.

167

Rajarapu, S.P., P. Mamidala, D.A. Herms, P. Bonello and O. Mittapalli. 2011. Antioxidant
genes of the emerald ash borer (Agrilus planipennis): gene characterization and
expression profiles. Journal of Insect Physiology 57: 819-824.
Ramina, A., P. Tonutti and T. Tosi. 1985. The effect of paclobutrazol on strawberry growth and
fruiting. Journal of Horticultural Science 60: 501-506.
Raupp, M.J., P.M. Shrewsbury and D.A. Herms. 2010. Ecology of herbivorous arthropods in
urban landscapes. Ann. Rev. Entomol. 55: 19-38.
Rebek, E.J., Herms, D.A., and Smitley, D.R. 2008. Interspecific variation in resistance to
emerald ash borer (Coleoptera: Buprestidae) among North American and Asian ash
(Fraxinus spp.). Environ. Entomol. 37(1): 242–246.
Rebek, K.A. and D.R. Smitley. 2007. Homeowner guide to emerald ash borer treatment.
Michigan State University Extension Bulletin E-2955.
Redding K.D., P.L. Burch and K.C. Miller. 1994. Growth, biomass and trim/chip time reduction
following application of flurprimidol tree growth regulator. Journal of Arboriculture 20:
38-45.
Remphrey, W.R. and C.G. Davidson. 1996. The effect of nitrogen on growth and architecture of
Fraxinus pennsylvanica 'Patmore' (green ash) in field-grow containers at two sites in
Manitoba, Canada. Canadian Journal of Plant Science 76: 825-833.
Rieger, M. and G. Scalabrelli. 1990. Paclobutrazol, root growth, hydraulic conductivity, and
nutrient uptake of ‘Nemaguard’ peach. HortScience 25: 95-98.
Robbertse, P. J. and J.A. Duvenhage. 1999. Reaction of avocado wood on injections into the
trunk. South African Avocado Growers’ Association Yearbook 22: 39-47.
Rodriguez-Saona, C., Poland, T.M., Miller, J.R., Stelinski, L.L., Grant, G.G., de Groot, P.,
Buchan, L., and MacDonald, L. 2006. Behavioral and electrophysiological responses of
the emerald ash borer, Agrilus planipennis, to induced volatiles of Manchurian ash,
Fraxinus mandshurica. Chemoecology, 16(2): 75–86.
Roshchina, V.V. 2012. Vital autofluorescence: application to the study of plant living cells.
International Journal of Spectroscopy 2012: 1-14.
Sanchez-Zamora, M.A. and R. Fernandez-Escobar. 2004. Uptake and distribution of trunk
injections in conifers. J Arboricult 30: 73–79.
Santantonio, D. 1989. Dry-matter partitioning and fine-root production in forests – new
approaches to a difficult problem. In: Biomass Production by fast-growing trees.
Editors J.S. Pereira and J.J. Lnadsberge) pp. 55-72.

168

SAS Institute. 2003. PROC user’s manual, version 9.1. Cary, NC, SAS Institute.
Scharenbroch, B.C. and J.E. Lloyd. 2004. A literature review of nitrogen availability indices for
use in urban landscapes. Journal of Arboriculture 30:214–229.
Scharenbroch, B.C., J.E. Lloyd, and J.L. Johnson-MaynardSchoon, K.J. 1993. The Midwest
urban tree index. Journal of Arboriculture 19: 230-237.
Schmelz, E.A., H.T. Alborn and J.H. Tumlinson. 2001. The influence of intact-plant and
excised-leaf bioassay design on volicitin- and jasmonic acid-induced sesquiterpene
volatile release in Zea mays. Planta 214: 171-179.
Schoon, K.J. 1993. The Midwest urban tree index. Journal of Arboriculture 19: 230-237.
Schoonhoven, L.M., J.J.A. Loon and M. Dicke. 2005. Insect-Plant Biology. Oxford
University Press, Oxford, New York.
Schulte, J.R. and C.E. Whitcomb. 1975. Effects of soil amendments and fertilizer levels on the
establishment of silver maple. Journal of Arboriculture 1: 192-195.
Sclar, D.C. and W.S. Cranshaw. 1996. Evaluation of new systemic insecticides for elm insect
pest control. Journal of Environmental Horticulture 14: 22-26.
Scriber, J.M. and F. Slansky. 1981. The nutritional ecology of immature insects. Annu. Rev.
Ecol. 26: 183-211.
Shaltiel-Harpaz, L., R. Kedoshim, D. Openheim, R. Stern and M. Coll. 2010. Effect of host
plant makeup through nitrogen fertilization and growth regulators on the pear psylla
population. Israel Journal of Plant Sciences 58: 149-156.
Shapiro, S.S., and M.B. Wilk. 1965. An analysis of variance test for normality. Biometrika
52:591–599.
Sheppard, C.J.R. 1993. Confocal microscopy – principles, practice and options. In: Fluorescent
and luminescent probes for biological activity (Edited by: W.T. Mason). Academic Press,
New York: 229-236.
Shigo, A.L. 1977. Compartmentalization of Decay in Trees. Agricultural Information Bulletin
No. 405. USDA Forest Service.
Shigo, A.L. 1984. Compartmentalization: a conceptual framework for understanding how trees
grow and defend themselves. Ann. Rev. Phytophathol. 22: 189-214.
Shipley, B. and D. Meziane. 2002. The balanced‐growth hypothesis and the allometry of leaf and
root biomass allocation. Functional Ecology 16: 326-331.

169

Siegert, N.W., McCullough, D.G., Liebhold, A.M., and Telewski, F. W. 2009. Reconstruction of
the establishment and spread of emerald ash borer through dendrochronological analysis.
In Proceedings of the 19th U.S. Department of Agriculture Interagency Research Forum
on Invasive Species, Annapolis, Maryland., 8–11 January 2008. Edited by K. McManus
and K.W.
Siegert, N.W., McCullough, D.G., Williams, D.W., Fraser, I., Poland, T.M., and Pierce, S.J.
2010. Dispersal of Agrilus planipennis (Coleoptera: Buprestidae) from discrete epicenters
in two outlier sites. Environ. Entomol. 39(2): 253–265.
Simpson, S.J. and C.L. Simpson. 1990. The mechanisms of compensation by phytophagous
insects. pp. 111-160. In E.A. Bernays (ed.), Insect-plant interactions, vol. 2. CRC, Boca
Raton, FL.
Smiley, E.T., L. Holmes and B.R. Fraedrich. 2009. Paclobutrazol foliar sprays to suppress
growth on landscape shrubs. Journal of Arboriculture 35: 300-304.
Smith, K. T. and P.A. Lewis. 2005. Potential concerns for tree wound response from stem
injection. In Proceedings of Third Symposium on hemlock woolly adelgid in the Eastern
United States, Asheville, North Carolina, 1-3 February 2005. Edited by B. Onken and R.
Reardon. USDA Forest Service, FHTET-2005-01: P 173-178.
Smitley, D.R., J.J. Doccola and D. L. Cox. 2010. Multiple-year protection of ash trees from
emerald ash borer with a single trunk injection of emamectin benzoate, and single year
protection with an imidacloprid basal drench. Arboriculture and Urban Forestry 36: 206211.
Sterrett, J.P. and T.J. Tworkoski. 1987. Response of shade trees to root collar drenches of
inhibitors flurprimidol and paclobutrazol. J. Plant Growth Regul. 5: 163-167.
Struve, D.K. 2002. A review of shade tree nitrogen fertilization research in the United States.
Journal of Arboriculture 28: 252-263.
Suckling, D.M., Tobin, P.C., McCullough, D.G., and Herms, D.A. 2012. Combining tactics to
exploit Allee effects for eradication of alien insect populations. J. Econ. Entomol. 105(1):
1–13.
Sur, R. and A. Stork. 2003. Uptake, translocation, and metabolism of imidacloprid in plants. Bull
Insectol 56: 35–40.
Tanis, S.R. and D.G. McCullough. 2012a. Differential persistence of blue ash and white ash
following emerald ash borer invasion. Canadian Journal of Forest Research. 42: 15421550.
Tanis, S.R., and McCullough, D.G. 2012b. It’s good to be blue: evidence that Fraxinus
quadrangulata may persist in North American forests. In Proceedings of the Emerald
Ash Borer Research and Technology Development Meeting, 12–13 October 2011,
170

Wooster, Ohio. Edited by V. Mastro and R. Reardon. USDA Forest Service, Forest
Health Technology Enterprise Team, Morgantown, West Virginia. In press.
Tanis, S.R., B.M. Cregg, D. Mota-Sanchez, D.G. McCullough, T.M. Poland. 2012. Spatial and
temporal distribution of trunk-injected 14C-imidacloprid in Fraxinus trees. Pest
Management 68: 529-536.
Tatter, T.A., J.A. Dotson, M.S. Ruizzo and V.B. Steward. 1998. Translocation of imidacloprid
in three tree species when trunk- and soil-injected. J. Arboricult 24: 54–56.
Taylor, P.B., Duan, J.J., Fuester, R.W., Hoddle, M., and Van Driesche, R. 2012. Parasitoid guilds
of Agrilus woodborers (Coleoptera: Buprestidae): their diversity and potential for use in
biological control. Psyche (Stuttg.), 2012: Article ID 813929.
Tišler, T., A. Jemec, B. Mozetič and P. Trebše, P. 2009. Hazard identification of imidacloprid to
aquatic environment. Chemosphere 76: 907-914.
Tluczek, A.R., McCullough, D.G., and Poland, T.M. 2011. Influence of host stress on emerald
ash borer (Coleoptera: Buprestidae) adult density, development, and distribution in
Fraxinus pennsylvanica trees. Environ. Entomol. 40(2): 357–366.
Toole, G and S. Toole. 2004. Essential Biology. Nelson Thomes Ltd. Cheltenham, UK.
Tsagkarakis, A.E., M.E. Rogers and T.M. Spann. 2012. Applications of plant growth regulators
to container-grown citrus trees affect the biology and behavior of the Asian citrus psyllid.
Journal of the American Society for Horticultural Science 137: 3-10.
uky.edu. 2012. University of Kentucky, Kentucky tree website [online].
http://www.uky.edu/Ag/Horticulture/kytreewebsite/ [Accessed 16 November 2012].
U.S. Department of Agriculture. 2012a. PLANTS database [online]. USDA Natural Resources
Conservation Service. Available at http://plants.usda.gov/java/ [accessed 6 March 2012].
U.S. Department of Agriculture. 2012b. Web soil survey [online]. USDA Natural Resources
Conservation Service. Available at
http://websoilsurvey.nrcs.usda.gov/app/WebSoilSurvey.aspx [accessed 6 March 2012].
Vanninen, P. and A. Mäkelä. 1999. Fine root biomass of Scots pine stands differing in age and
soil fertility in southern Finland. Tree physiology 19: 823-830.
Vitousek, P.M. and R.W. Howarth. 1991. Nitrogen limitation on land and in the sea: how can it
occur? Biogeochemistry13: 87-115.
Wallander, E. 2008. Systematics of Fraxinus (Oleaceae) and evolution of dioecy. Plant Syst.
Evol. 273(1–2): 25–49.

171

Wang, S.Y. and G.L. Steffens. 1987. Effect of paclobutrazol on accumulation of organic acids
and total phenols in apple wood. J. Plant Growth Regul. 6: 209-213.
Wang, X., Yang, Z., Gould, J.R., Zhang, Y., Liu, G., and Liu, E. 2010. The biology and ecology
of the emerald ash borer, Agrilus planipennis, in China. J. Insect Sci. 10(128): 1–23.
Wang, M., S. Shi, F. Lin, Z. Hao, P. Jiang, et al. 2012. Effect of soil water and nitrogen on
growth and photosynthetic response of Manchurian ash (Fraxinus mandshurica) seedling
in Northeastern China. PLoS ONE 7(2): e30754.
Waring, G.L. and N.S. Cobbs. 1992. The impact of plant stress on herbivore population
dynamics. In: Bernays EA (ed) Plant-insect interactions, Vol 4. CRC, Boca Raton, FL, pp
167–226
Watson, G.W. 1996. Tree root system enhancement with paclobutrazol. Journal of Arboriculture
22: 211-217.
Watson, G.W. 2000. Tree root system enhancement with paclobutrazol, p. 131-135. In: A.
Stokes (editors). The supporting roots of trees and woody plants: Form, function and
physiology. Kluwer Academic Publishers, Dodrecht, The Netherlands.
Watson, G.W. 2001. Soil applied paclobutrazol affects root growth, shoot growth, and water
potential of American elm seedlings. J. Environ. Hort. 19: 199-122.
Watson, G.W. 2004. Effect of transplanting and paclobutrazol on root growth of ‘Green Column’
black maple and ‘Summit’ green ash. J. Environ. Hort. 22: 209-212.
Watson, G. 2010. The effect of broadcast nitrogen fertilization rates and placement on the
growth of green ash trees. Arboriculture and urban forestry 36: 179-182.
Werner, H. 1993. Influence of paclobutrazol on growth and leaf nutrient content of mango (CV.
Blanco). Acta Horticulturae 341: 225-231.
Whitehill, J.G.A., Popova-Butler, A., Green-Church, K.B., Koch, J. L., Herms, D.A., and
Bonello, P. 2011. Interspecific proteomic comparisons reveal ash phloem genes
potentially involved in constitutive resistance to the emerald ash borer. PLoS ONE, 6(9):
e24863.
Whitehill, J. G., Opiyo, S. O., Koch, J. L., Herms, D. A., Cipollini, D. F., & Bonello, P. (2012).
Interspecific comparison of constitutive ash phloem phenolic chemistry reveals
compounds unique to Manchurian ash, a species resistant to emerald ash borer. Journal of
Chemical Ecology: 1-13.
Wieland, W.F. and R.L. Wample. 1985. Effects of paclobutrazol on growth, photosynthesis and
carbohydrate content of ‘dDlicious’ apples. Scientia Horticulturae 26: 139-147.

172

Wiesman, Z. and S. Lavee. 1994. Vegetative growth retardation, improved rooting and viability
of olive cuttings in response to application of growth retardants. Plant Growth Regulation
14: 83-90.
Wilson, J.W., S.J. Dircks and R.I. Grange. 1983. Regeneration of sclerenchyma in wounded
dicotyledon stems. Annals of Botany 52: 295-303.
Witzell, J.W. and J.A. Martín. 2008. Phenolic metabolites in the resistance of northern forest
trees to pathogens-past experiences and future prospects. Canadian Journal of Forest
Research 38: 2711-2727.
Wymer C. L., A.F. Beven, K. Boudonck and C.W. Lloyd. 1999. Confocal microscopy of plant
cells. In: Methods in molecular biology Volume 122: Confocal Microscopy Methods
and Protocols (Edited by S. Paddock). Humana Press Inc. Totowa, NJ, USA. Pp 103130.
Yadav, R.K., N. Rai, D.S. Yadav and B.S. Asati. 2005. Use of paclobutrazol in horticultural
crops – a review. Agric. Rev 26: 124-132.
Yelenosky, G., J.C.V. Vu and H.K. Wutscher. 1995. Influence of paclobutrazol in the soil on
growth, nutrient elements in the leaves, and flood/freeze tolerance of citrus rootstock
seedlings. J. Plant Growth Regul. 14: 129-134.
Young, L.C. 2002. The efficacy of micro-injected imidacloprid and oxydemeton-methyl on red
gum eucalyptus trees (Eucalyptus camaldulensis) infested with red gum lerp psyllid
(Glycaspis brimblecombei). Journal of Arboriculture 28: 144-147.
Young Song, M., J.D. Stark and J.J. Brown. 1997. Comparative toxicity of four insecticides,
including imidacloprid and tebufenozide, to four aquatic arthropods. Environmental
Toxicology and Chemistry 16: 2494–2500.
Yu, C. 1992. Agrilus marcopoli Obenberger (Coleoptera: Buprestidae). [Online.] In Forest
insects of China. 2nd ed. Edited by G-R. Xiao. China Forestry Publishing House, Beijing,
China. pp. 400- 401. Available at http://www.ncrs.fs.fed.us/dev/4501/eab/downloads/
biologyYu1992.pdf [online translation accessed 6 March2012].
Zarnoch, S.J., Bechtold, W.A., and Stolte, K.W. 2004. Using crown condition variables as
indicators of forest health. Can. J. For. Res. 34(5): 1057–1070.
Zwieniecki, M.A. and N.M. Holbrook. 1998. Diurnal variation in xylem hydraulic conductivity
in white ash (Fraxinus americana L.) red maple (Acer rubrum L.) and red spruce (Picea
rubens Sarg.). Plant, Cell and Environment: 1173-1180.
Zwieniecki, M.A. and N.M. Holbrook. 2000. Bordered pit structure and vessel wall surface
properties. Implications for embolism repair. Plant Physiology 123: 1015-1020.
173

Zwieniecki, M.A., P. J. Melcher and N.M. Holbrook. 2001. Hydraulic properties of individual
xylem vessels of Fraxinus americana. Journal of Experimental Botany 52: 257-264.

174