WHOLE TREE RENEWAL REGENERATES FRUITNG STRUCTURES QUICKLY IN
MATURE ORCHARDS
By
James Edward Larson Jr.

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Horticulture—Master of Science
2017

ABSTRACT
WHOLE TREE RENEWAL REGENERATES FRUITING STRUCTURES QUICKLY IN
MATURE ORCHARDS
By
James Edward Larson Jr.
Renewal of fruiting wood to maintain young reproductive meristems with optimal canopy light
interception and distribution is key for high productivity and fruit quality throughout the life of a
sweet cherry (Prunus avium L.) orchard. Typical renewal involves replacement of 10 to 20% of
the tree canopy annually by removing one to several of the largest branches. In a mature orchard,
this renewal process is subject to competition between sun-exposed fruiting sites and interior
canopy renewal sites that intercept less light and compete poorly for translocated
photoassimilates, often resulting in poor renewal growth. This is particularly problematic for
high density orchards that utilize rootstocks selected for reduced vigor and high productivity.
Renewal of canopy fruiting sites on a whole tree basis is an alternative renewal method that
eliminates the competitive inhibition of shoot regrowth. This study explores the initial response
of sweet cherry trees on various training systems and size-controlling rootstocks to whole tree
renewal. Four training systems were studied: Tall Spindle Axe, Super Slender Axe, Upright
Fruiting Offshoots, and Kym Green Bush. In 2016, whole tree renewal of the four systems was
studied with ‘Benton’ cultivar on three rootstocks of varying vigor: Gisela 3, Gisela 5, and
Gisela 6. During bloom, all fruit-bearing components of the canopy were pruned back to stubs
close to the permanent structure. TSA resulted in the higher number of shoots, while KGB and
UFO had the longest average shoot length. The results indicate that each canopy systemrootstock combination refilled canopy space, except for KGB on each rootstock, to quickly
regenerate fruiting sites.

For my parents who have always believed in me. And to Becca for always making me laugh

iii

ACKNOWLEDGEMENTS

I would first like to thank my major advisor, Dr. Greg Lang for providing me the
opportunity to conduct this research; constantly providing help and knowledge. Also, to my
graduate committee: Dr. Frank Telewski and Paolo Sabbatini for being available for questions
and providing insight. I owe a huge amount of gratitude to my lab mates: Tammy Wilkinson,
Feiran Li, Rebecca Selby, and Carly Daiek; who all helped me collect the data within this thesis.
They have made the work much more fun.

iv

TABLE OF CONTENTS

LIST OF TABLES……………………………………………………………………………….vii
LIST OF FIGURES……………………………………………………………………………..viii
THESIS INTRODUCTION…………………………………………………………………….…1
Background………………………………………………………………………………..1
Thesis Contents……………………………………………………………………………1
CHAPTER 1: A REVIEW OF THE LITERATURE...…...………………………………...…….3
Sweet Cherry Production in Michigan…………………………………………………….3
Growth and Fruiting Habit………………………………………………………………...4
Role of Auxin and Cytokinin in Branching……………………………………………….5
Shoot Renewal and Epicormic Branching………………………………………………...7
Epicormic Vegetative Meristems………………………………………………………….8
Sylleptic Branching………………………………………………………………………11
Conclusions………………………………………………………………………………12
LITERATURE CITED…………………………………………………………………..17
CHAPTER 2: WHOLE TREE RENEWAL REGENERATES FRUITING STRUCTURES
QUICKLY IN MATURE ORCHARDS…………………………………………………………22
Introduction………………………………………………………………………………22
Materials and Methods…………………………………………………………………...24
Site and Plant Material…………………………………………………………...24
Whole Tree Renewal Pruning……………………………………………………25
Initial Regrowth Measurements………………………………………………….25
2017 Maintenance Pruning……………………………………………………....26
Canopy Volume Measurements………………………………………………….26
Covering System…………………………………………………………………27
Data Analysis and Plot Design…………………………………………………..28
Results……………………………………………………………………………………29
Canopy Volume………………………………………………………………….29
Initial Growth Data………………………………………………………………29
Discussion………………………………………………………………………………..31
Conclusions……………………………...……………………………………………….35
APPENDIX……………………………………………………………………………....52
LITERATURE CITED ………………………………………………………………….55
CHAPTER 3: MAPPING EPICORMIC VEGETATIVE MERISTEMS IN SWEET CHERRY
USING X-RAY COMPUTER TOMOGRAPHY……………………………………….…….....59
Introduction……………………………………………………………………………....59
Materials and Methods…………………………………………………………………...61
Plant Material…………………………………………………………………….61

v

X-Ray Computer Tomography Scan…………………………………………….61
3-Dimensional Reconstruction…………………………………………………...61
Results…………………………………………………………………………………....62
Discussion…………………………………………………………………………...…...63
Future Directions………………………………………………………………………...64
LITERATURE CITED…………………………………………………………………..70

vi

LIST OF TABLES

Table 2.1- Analysis of variance (ANOVA) for number of renewal shoots following whole tree
renewal pruning of eight-year-old ‘Benton’ sweet cherry trees, grown in the VOEN at MSU
Clarksville Research Center. SSA data left out………………………………………………….53
Table 2.2- Analysis of variance (ANOVA) for number of sylleptic branches following whole
tree renewal pruning of eight-year-old ‘Benton’ sweet cherry trees, grown in the CRAVO at
MSU Clarksville Research Center. SSA data eliminated to compare rootstocks………………..53
Table 2.3- Analysis of variance (ANOVA) for number of sylleptic branches following whole
tree renewal pruning of eight-year-old ‘Benton’ sweet cherry trees, grown in the VOEN at MSU
Clarksville Research Center. SSA data eliminated to compare rootstocks……………………...54

vii

LIST OF FIGURES

Figure 1.1- First year of growth: branch with single leaves at each node (Long et al., 2015).....14
Figure 1.2- Second year of growth: basal fruit, vegetative spurs along one year old growth, and
annual extension of new growth (Long et al., 2015)…………………………………………….14
Figure 1.3- Third year of growth: fruiting spurs along two-year-old growth, basal fruit on one
year old growth, followed by vegetative spurs, and finally another section of new growth (Long
et al., 2015)…………..……………………………………………………………………….….14
Figure 1.4- Scanning electronic microscope image of small preventitious epicormic buds
without secondary bud primordia (Fontaine et al., 1999)………………….…………..………...15
Figure 1.5- Large preventitious epicormic bud: SEM of large primary bud in the center,
surrounded by secondary bud primordia (Fontaine et al., 1999)…………....…….……………..16
Figure 2.1- Diagram of each training system used in the study. Kym Green Bush (KGB), Tall
Spindle Axe (TSA), Super Slender Axe (SSA), and Upright Fruiting Offshoots (UFO)…...…...36
Figure 2.2- SSA on Gi3 and Gi6 whole tree renewal (WTR) canopy average spread versus
spread of annual partial renewal (APR) trees at 0.75, 1.5, and 2.25 meters in the canopy. Canopy
spread averaged over CRAVO and VOEN at Michigan State University’s Clarksville Research
Center, except where noted as CRAVO or VOEN. Lowercase letters separate means between
WTR and APR at each height tested using Tukey LSD were calculated at significance of
p≤0.05……………………………………………………………………………………………36
Figure 2.3- TSA on Gi3, Gi5, and Gi6 whole tree renewal (WTR) canopy average spread versus
spread of annual partial renewal (APR) trees at 0.75, 1.5, and 2.25 meters in the canopy. Canopy
spread averaged over CRAVO and VOEN at Michigan State University’s Clarksville Research
Center, except where noted as CRAVO or VOEN. Lowercase letters separate means between
WTR and APR at each height tested using Tukey LSD were calculated at significance of
p≤0.05…………………………………………………………………………………………....37
Figure 2.4- UFO on Gi3, Gi5, and Gi6 whole tree renewal (WTR) canopy average spread versus
spread of annual partial renewal (APR) trees at 0.75, 1.5, and 2.25 meters in the canopy. Canopy
spread averaged over CRAVO and VOEN at Michigan State University’s Clarksville Research
Center, except where noted as CRAVO or VOEN. Lowercase letters separate means between
WTR and APR at each height tested using Tukey LSD were calculated at significance of
p≤0.05………………………………………………….………………………………………...37
Figure 2.5- KGB on Gi3, Gi5, and Gi6 whole tree renewal (WTR) canopy average spread versus
spread of annual partial renewal (APR) trees at 0.75, 1.5, and 2.25 meters in the canopy. Canopy
spread averaged over CRAVO and VOEN at Michigan State University’s Clarksville Research

viii

Center, except where noted as CRAVO or VOEN. Lowercase letters separate means between
WTR and APR at each height tested using Tukey LSD were calculated at significance of
p≤0.05……………………………………………………………………………………………38
Figure 2.6- Initial response number of renewal branches initiated per permanent structure
section, of TSA and UFO grown on Gi3, Gi5, and Gi6 after whole tree renewal was imposed.
5A- Bars are average of 18 whole tree renewal trees grown in the CRAVO plus or minus
standard error of the mean. 5B- Bars are average of 12 Whole tree renewal trees grown in the
VOEN plus or minus standard of the mean at Michigan State University’s Clarksville Research
Center. Lowercase letters separate means comparing canopy sections, determined using Tukey’s
LSD, significance at p≤0.05………………………………………………………………...……39
Figure 2.7- Initial response number of renewal branch per permanent structure section of TSA,
SSA and UFO grown on Gi3 and Gi6 after whole tree renewal was imposed. 5A- Bars are
average of 18 whole tree renewal trees grown in the CRAVO plus or minus standard error of the
mean. 5B- Bars are average of 12 Whole tree renewal trees grown in the VOEN plus or minus
standard of the mean at Michigan State University’s Clarksville Research Center. Lowercase
letters separate means comparing canopy sections, determined using Tukey’s LSD, significance
at p≤0.05. ………………………………………………………………………………………..40
Figure 2.8- Initial response number of renewal branch by rootstock averaged over TSA, KGB
and UFO after whole tree renewal was imposed. Bars are an average of 27 whole tree renewal
trees grown in the CRAVO plus or minus standard error of the mean at Michigan State
University’s Clarksville Research Center. Lowercase letters separate means comparing
rootstocks, determined using Tukey’s LSD, significance at p≤0.05…………………..………...41
Figure 2.9- Initial response number of renewal branch by training system, averaged over Gi3
and Gi6. 8A- Bars are average of 21 whole tree renewal trees grown in the CRAVO plus or
minus standard error of the mean. 5B- Bars are average of 14 Whole tree renewal trees grown in
the VOEN plus or minus standard of the mean at Michigan State University’s Clarksville
Research Center. Lowercase letters separate means comparing training systems, determined
using Tukey’s LSD, significance at p≤0.05. …………………………………………………….42
Figure 2.10- Average length per renewal branch, one season of growth after whole tree renewal
was imposed. Three-way interaction: permanent structure section by rootstock by training
system. TSA and UFO both grown on Gi3, Gi5, and Gi6. 9A-Bars are an average per section of 3
whole tree renewal trees grown in the CRAVO plus or minus standard error of the mean. 9BBars are an average per canopy section of 4 whole tree renewal trees grown in the VOEN plus or
minus standard error of the mean at Michigan State University’s Clarksville Research Center.
Lowercase letters separate means comparing the system by rootstock by section interaction,
determined using Tukey’s LSD, significance at p≤0.05…………………………………………43
Figure 2.11- Average length per renewal branch, one season of growth after whole tree renewal
was imposed. Three-way interaction: permanent structure section by rootstock by training
system. TSA, SSA and UFO both grown on Gi3 and Gi6.. Three-way interaction: permanent
structure section by rootstock by training system. Without Gi5 data. 10A-Bars are an average per

ix

section of 3 whole tree renewal trees grown in the CRAVO plus or minus standard error of the
mean. 10B- Bars are an average per canopy section of 4 whole tree renewal trees grown in the
VOEN plus or minus standard error of the mean at Michigan State University’s Clarksville
Research Center. Lowercase letters separate means comparing the system by rootstock by
section interaction, determined using Tukey’s LSD, significance at p≤0.05……………………44
Figure 2.12- Length per renewal branch on a whole tree basis averaged over Gi3 and Gi6. 11ABars are average of 24 whole tree renewal trees grown in the CRAVO plus or minus standard
error of the mean. 11B- Bars are average of 16 Whole tree renewal trees grown in the VOEN
plus or minus standard of the mean at Michigan State University’s Clarksville Research Center.
Lowercase letters separate means comparing the training systems, determined using Tukey’s
LSD, significance at p≤0.05……………………………………………………………………...45
Figure 2.13- Length per renewal branch on a whole tree basis averaged over KGB, TSA, and
UFO. 12A-Bars are average of 27 whole tree renewal trees grown in the CRAVO plus or minus
standard error of the mean. 12B- Bars are average of 18 Whole tree renewal trees grown in the
VOEN plus or minus standard of the mean at Michigan State University’s Clarksville Research
Center. Lowercase letters separate means comparing rootstocks, determined using Tukey’s LSD,
significance at p≤0.05……………………………………………………………………………46
Figure 2.14- Number of sylleptic branches on each renewal parent branch, per permanent
structure section of TSA and UFO on Gi3, Gi5, and Gi6. Bars are average of 18 whole tree
renewal trees grown in the CRAVO plus or minus standard error of the mean at Michigan State
University’s Clarksville Research Center. Lowercase letters separate means comparing the
canopy sections, determined using Tukey’s LSD, significance at p≤0.05………………………46
Figure 2.15- Number of sylleptic branches on each renewal parent branch, per permanent
structure section of TSA and UFO on Gi3, Gi5, and Gi6. Three way interaction: training system
by rootstock by permanent structure section. Bars are an average per section of 12 whole tree
renewal trees grown in the VOEN plus or minus standard error of the mean at Michigan State
University’s Clarksville Research Center. Lowercase letters separate means comparing the
system by rootstock by section interaction, determined using Tukey’s LSD, significance at
p≤0.05………………………………………………………………………………………....…47
Figure 2.16- Number of sylleptic branches on each renewal parent branch, per permanent
structure section of Gi3 and Gi6 averaged over TSA, SSA, and UFO. Two-way interaction:
rootstock by permanent structure section. Bars are an average per section of 18 whole tree
renewal trees grown in the CRAVO plus or minus standard error of the mean at Michigan State
University’s Clarksville Research Center. Lowercase letters separate means comparing the
rootstock by section interaction, determined using Tukey’s LSD, significance at p≤0.05...……47
Figure 2.17- Number of sylleptic branches on each renewal parent branch, per permanent
structure section of TSA, SSA, and UFO averaged over Gi3 and Gi6. Two-way interaction:
training system by permanent structure section. Bars are an average per section of 12 whole tree
renewal trees grown in the VOEN plus or minus standard error of the mean at Michigan State

x

University’s Clarksville Research Center. Lowercase letters separate means comparing the twoway interaction, determined using Tukey’s LSD, significance at p≤0.05……………………….48
Figure 2.18- Number of sylleptic branches on a whole tree basis by training system. Averaged
over Gi3 and Gi6. 17A- Bars are and average of 24 whole tree renewal trees grown in the
CRAVO plus or minus standard error of the mean. 17B- Bars are and average of 16 whole tree
renewal trees grown in the VOEN plus or minus standard error of the mean at Michigan State
University’s Clarksville Research Center. Lowercase letters separate means comparing training
systems, determined using Tukey’s LSD, significance at p≤0.05……………………………….49
Figure 2.19- Trunk cross sectional area in 2016 of all location by training system by rootstock
combinations in the NC-140 sweet cherry rootstock trial at Michigan State University’s
Clarksville Research Center with ‘Benton’ as the scion. Bars are an average of 6 trees for each
KGB, TSA, and UFO and 12 trees for SSA on each rootstock under the CRAVO; under the
VOEN, 4 trees for KGB, TSA, and UFO and 8 trees for SSA on each rootstock……………….49
Figure 2.20- Density scatter plot of length per primary renewal branch versus number of
sylleptic branches per renewal branch of trees that have been whole tree renewed. All data is
included, being comprised of four different training systems: KGB, TSA, SSA, and UFO; three
different rootstocks: Gi3, Gi5, and Gi6 at Michigan State University’s Clarksville Research
Center. Darker hexagon shade indicates higher number of observations. A total of 1362 primary
branches are plotted……………………………………………………………………………...50
Figure 2.21- Violin plot of distribution of number of sylleptic branches per primary renewal
branch of trees that have been whole tree renewed. All data is included, being comprised of four
different training systems: KGB, TSA, SSA, and UFO; three different rootstocks: Gi3, Gi5, and
Gi6 at Michigan State University’s Clarksville Research Center. Data is broken up into
percentile by renewal branch length. 1-25th and 26-50th is the distribution of sylleptic branches on
340 primary renewal branches. 51-75th and 76-100th percentile is the distribution of sylleptic
branches on 341 primary renewal branches. 1-25th percentile represents the shortest 25% of
branches, branch length increases with percentile……………………………………………….51
Figure 3.1- Scanning Electronic microscope images of a large preventitious epicormic bud (1A)
with secondary bud primordia surrounding the bud in the center; small preventitious epicormic
buds (1B) without secondary bud primordia (Fontaine et al., 1999)…………………………….66
Figure 3.2- Images in cross-section from X-ray CT scan of a 8-year-old section of sweet cherry
of the variety NY 119 grafted onto Gisela 12 rootstock trunk: 2A) complex epicormic trace, 2B)
simple epicormic trace, 2C) V trace with a central complex trace, flanked by a simple trace on
either side, 2D) sequential branch………………………………………………………………..67
Figure 3.3 - Epicormic branches in cross-section: 3A) epicormic branch 1 and 2; 3B) epicormic
branch 3; 3C) epicormic branch 4 and 5; 3D) epicormic branch 6………………………………68
Figure 3.4- Epicormic branch number 6 with reaction wood on the underside of the branch…..69

xi

THESIS INTRODUCTION
Background
Annual sweet cherry (Prunus avium L.) pruning includes the removal of one to several of
the largest branches to maintain light distribution throughout the canopy to decrease shading.
Optimal light distribution is key for high productivity and fruit quality throughout the life of an
orchard. Pruning of the largest branches usually initiates new shoots that emerge and refill the
canopy, resulting in renewal of 10 to 20% of the canopy each year and fruit that are consistently
produced on relatively young branches (Long et al., 2015).
However, this renewal process creates competition in the tree for photoassimilates
between fruiting meristems and renewal of vegetative meristems (G.A. Lang, personal
communication). This competition is magnified in high density orchards that utilize dwarfing
rootstocks selected for high productivity but reduced vigor; the result, typically, is poor or
inadequate regrowth to refill the canopy (Lang, 2000; Perry et al., 1997). Renewal of the whole
tree at once eliminates this competition, shifting the tree’s physiology entirely to vegetative
growth instead of balancing reproductive growth and vegetative regeneration. Whole tree
renewal removes the majority of the previously active vegetative and reproductive meristems on
the tree, and promotes the activation of previously dormant vegetative meristems to form new
branches and the rapid refilling of the canopy.
Thesis Contents
This thesis looks at two aspects of whole tree renewal. First, an applied study of training
systems and rootstock vigor and how these factors affected canopy refill after whole tree renewal
was imposed. Four training systems: Tall Spindle Axe (TSA), Super Slender Axe (SSA), Kym
Green Bush (KGB), and Upright Fruiting Offshoots (UFO) were studied; each on three

1

rootstocks: semi-vigorous Gisela 6 (Gi6), semi-dwarfing Gisela 5 (Gi5), and dwarfing Gisela 3
(Gi3). The data presented shows the initial regrowth that occurred in the first year after whole
tree renewal was imposed at bloom of 2016 and the canopy volume of the trees after a second
season of growth, 2017. The second component of this thesis takes a basic approach to see how
whole tree renewal can be improved by mapping latent buds using X-ray Computer
Tomography. This is a preliminary study to determine first if it is possible to identify latent buds
from a CT scan and then look at how to map latent buds for the potential to pinpoint where a new
branch will arise after pruning.

2

CHAPTER 1: A REVIEW OF THE LITERATURE
Sweet Cherry Production in Michigan
Traverse City, Michigan, is the self-proclaimed cherry capital of the world, with
Michigan producing 75% of the domestic tart cherry (Prunus cerasus L.) supply and 20% of the
sweet cherries (Prunus avium L.) grown nationally (Michigan Ag Council, 2016). Lake
Michigan moderates the climate along the shoreline by raising the average temperature in the
winter while also delaying bloom in spring, reducing the risk of frost damage; therefore, a vast
majority of the tree fruit acreage in Michigan is within 50 miles of Lake Michigan (Schaetzal,
n.d.).
Dwarfing rootstocks became available later to sweet cherries than for apples (Malus
domestica Borkh.), therefore, many sweet cherry orchards are still grown at traditional low
planting densities, 250-400 trees per ha-1, on vigorous rootstocks, e.g., Mazzard (Prunus avium
L.) and Mahaleb (Prunus mahaleb L.). These traditional orchards produce large, complex tree
canopies that are inefficient to maintain and harvest. First fruiting typically occurs in the fourth
or fifth year after planting and the trees don’t come into full production until 8-12 years.
However, the recent profits that have been realized in high density apple production have
sparked a transition to apply those practices to sweet cherries (Robinson et al., 2004). Dwarfing
rootstocks impart precocity and high productivity (Lang, 2000; Perry et al., 1997), allowing
cherry growers to plant high density (1200-4000 trees per ha-1) orchards that realize full
production quickly (5-7 years). With high density systems, growers obtain a faster return on
investment, pruning and harvest is more labor efficient, and the opportunity to incorporate
mechanization with orchard platforms and/or hedging is increased.

3

Transitioning to high density systems has raised questions on canopy architecture, which
training systems to use, how to balance crop load, and how to modify pruning. High density
canopy architectures emphasize simplicity, with an effort to create a “fruiting wall” that
maximizes light interception (Long et al., 2015). Sweet cherry training systems take many
different forms, with some having a central leader (e.g., Tall Spindle Axe, TSA; Super Slender
Axe, SSA) and others having multiple leaders (Kym Green Bush, KGB; Steep Leader, SL;
Spanish Bush, SB). Some are 3-dimensional canopies (TSA, KGB, SL, SB), while some are
narrow planar canopies (SSA; Upright Fruiting Offshoots, UFO). Some are free standing trees
(TSA, KGB) and others utilize a trellis (SSA; UFO; Y-UFO).
Growth and Fruiting Habit
As a forest tree, sweet cherry’s natural growth habit is to produce vigorous, upright
branches to compete for light interception. The tree is therefore establishing it’s “footprint”, it’s
vegetative canopy that is large enough to compete for light with neighboring trees, before
transitioning to fruit production (Long et al., 2015). This is counterproductive to commercial
production, since growers want to get orchards into production as quickly as possible, and
develop compact trees with weak-to-moderately vigorous branches. Queue dwarfing, precocious
rootstocks and simplified training systems.
The sweet cherry fruiting habit is such that a single branch takes three years to obtain all
of its fruiting components: single node leaves, vegetative spur leaves, reproductive spur leaves,
basal flowers, and spur flowers. The first year of growth is characterized by large single leaves at
each node along the branch (fig. 1.1). These simple buds at each node can develop further in one
of a few different ways in the following years. The axillary meristems in the bud may remain
primordial in the year the bud develops; the bud will therefore remain in paradormancy (Lang,

4

1987) and become engulfed by radial growth of the branch, persisting as latent (preventitious
epicormic) buds beneath the bark. Buds at the base of the shoot can become fully reproductive
meristems, i.e., single flower buds with no accompanying vegetative meristem; after fruiting
occurs, these nodes become “blind”. The rest of the buds will either develop into a new lateral
branch or a spur, a modified branch with extremely short internodes and a rosette of five to eight
leaves (Ayala and Lang, 2017).
One-year-old branches usually have a few basal flower buds and vegetative spurs. Flower
buds begin forming in the leaf axils of vegetative spurs, giving the potential for fruiting spurs in
year 3. A new section of annual growth develops from the terminal bud (fig. 1.2). By year three,
the branch has all of its fruiting components: “blind” wood at the base, followed by fruiting (and
perhaps some continuing non-fruiting) spurs, one-year-old growth with basal flower buds and
vegetative spurs, and finally, another section of annual growth (fig. 1.3) (Ayala and Lang, 2017).
Maguylo et al. (2002) showed that spur flower density increases from the basal to the distal end
of each mature annual growth segment of a branch.
Role of Auxin and Cytokinin in Branching
The axillary meristem, giving rise to axillary buds and eventually lateral branches, is
thought to be initiated through two contrasting hypotheses: detached meristem and de novo
initiation. The detached meristem hypothesis states that the apical meristem cells that never lost
their meristematic activity is the origin of the axillary meristem; the alternative hypothesis states
that the axillary meristem forms de novo from the leaf axil (Leyser, 2003).
Auxin (indole-3-acetic acid) represses bud break and branch development; this effect is
believed to occur after axillary meristem initiation (Leyser, 2003). Auxin is produced in the
shoot apex and newly formed leaves (Ljung et al., 2001) and moves down the stem through polar

5

transport (Wignall et al., 1987; Sundberg and Uggla, 1998). Auxin production increases in buds
as they activate (Hillman et al., 1977), and applying auxin directly to a bud does not inhibit
branch formation after removal of apical meristem (Cline, 1996). These two findings lead to the
conclusion that auxin acts indirectly to inhibit branching, mainly by down-regulation of
cytokinin synthesis and export from roots (Leyser, 2003). Cytokinins (isopentenyladenine,
zeatin, and dihydrozeatin) act directly on buds to promote branching; applications of cytokinins
to a bud increase shoot outgrowth (Sachs and Thimann, 1967).
The RMS and MAX family of genes, discovered in Pea (Pisum sativum) and in
Arabidopsis thaliana, respectively (Sorefan et al., 2003), are required for apical dominance,
leading to the belief that a mobile signal serves as another indirect mode of action through which
auxin inhibits branching (Leyser, 2003). Rms and max mutants are bushier and show a reduced
response to auxin applications after removal of the shoot apex (Sorefan et al., 2003). Grafting
and reciprocal grafting experiments with rms and max mutants and wildtypes have shown that
this long-range signal is synthesized throughout the plant: apical dominance was restored with:
wildtype roots grafted on mutant shoots, mutant roots grafted onto wildtype shoots, and even
wildtype interstock between mutant shoot tissue. Wildtype tissue anywhere in the plant is able to
produce the long-range signal and maintain apical dominance (Leyser, 2003). The product of the
reaction that the MAX/RMS-dependent signal catalyzes to inhibit branch development is
unknown. However, auxin is known to up-regulate transcription of RMS1 in pea; increased
production of RMS1 and its product could be the indirect mechanism through which auxin
inhibits branching. The MAX pathway seems to work different in Arabidopsis, since auxininduced MAX4 expression in the root tip was seen 24 h after auxin treatment. This points to
branch inhibition being post-transcriptional in Arabidopsis (Leyser, 2003).

6

Shoot Renewal and Epicormic Branching
The highest quality cherries are borne on basal flower buds and young spurs. Renewal of
fruiting sites is therefore required to continually produce high quality fruit throughout the life of
an orchard. Renewal is accomplished by annual removal of a few of the largest branches in the
canopy. Ideally, pruning of these branches back to short stubs removes apical dominance and
epicormic buds are released from paradormancy, developing into an epicormic branch that
subsequently refills the canopy space. Therefore, a portion of the canopy (~10-20%) is
constantly being renewed while fruiting occurs throughout the rest of the canopy.
Epicormic branching is a regenerative measure of trees to increase leaf area following
damage or stresses. Fire (Burrows, 2008), insect defoliation (Piene and Eveleigh, 1996), wind
damage (Cooper-Ellis et al., 1999), competition (Nicolini et al., 2001), and vascular embolism
(Nicolini et al., 2001) have all been shown to initiate epicormic branching in trees. These stresses
cease or limit auxin production from the apical meristem, removing the correlative inhibition of
cytokinin biosynthesis and the auxin-induced signaling pathway, as described above. In a
complex tree canopy, removal of a single branch, as with typical annual renewal pruning in
sweet cherry, only removes a portion of the auxin that is being produced. Therefore, epicormic
buds still have the potential to remain dormant. Combining this situation with the decreased
vigor of trees on dwarfing rootstocks, which limits photoassimilate supply for fruiting sites and
vegetative growth, can result in inconsistent annual renewal with typically either no new shoot
growth or insufficient growth to refill the space in high density sweet cherry orchards.
Pruning severity increases not only the number of epicormic branches that sprout, but
also the length of those epicormic branches (O’Hara and Berrill, 2009). In coastal redwood
(Sequoia sempervirens [D. Don.] Endl.), O’Hara and Berrill showed no change in epicormic

7

branching from crown removal treatments ranging from an untreated control up to 60% removal,
but with 85% removal, epicormics branch number, length, and diameter of branch increased.
Gordon et al. (2006) also showed light and time of topping as significant factors affecting
initiation of epicormic branches in peach (Prunus persica Batsch.). Two flushes of epicormic
branches were seen; one beginning in March and dwindling in May, followed by an increase in
June, with the highest sprouting occurring in August. Epicormic branches during this second
flush had a higher dry weight than early sprouting branches. Light exposure was positively
correlated with epicormic sprouting; trees receiving no light had a 25% reduction in epicormic
branches initiated (Gordon et al., 2006). It was concluded that pruning affects epicormic
branching by removing apical dominance and increasing light exposure to epicormic buds.
Epicormic Vegetative Meristems
Epicormic buds are classified as adventitious or preventitious (Fontaine et al., 1998).
Development is the key differentiating factor, with preventitious epicormics buds developing
from the apical meristem and adventitious developing from previously non-meristematic tissue
(Brown, 1971; Fink, 1999). Preventitious epicormics buds were once axillary buds that did not
develop into a branch, remained dormant and were subsequently engulfed by the radial growth of
the branch, persisting beneath the bark (Büsgen and Münch, 1929; Stone and Stone, 1943).
Preventitious epicormics bud formation follows the phyllotaxy of the tree; sweet cherry has a
spiral phyllotaxy with 5 nodes per double revolution, ~every 144˚ (Lang et al., 2004; G.A. Lang,
personal communication).
Adventitious epicormics develop exclusive of the normal phyllotaxy, typically following
a wounding event (Fink, 1983, 1999; Kauppi et al., 1987). Formation occurs in mature tissue or
callus (Stone and Stone, 1943), in small groupings of parenchyma cells that regain meristematic

8

activity by subdivisions. Differentiation of an apex and prophylls follows. Leaf primordia then
arise from adjoining parenchyma cells. This completes the formation of the bud and vascular
tissue forms, initiating an epicormic trace (Fink, 1983). Adventitious epicormics buds are
thought to contribute little to the overall epicormic potential in unwounded trees.
Preventitious epicormics buds maintain a vascular connection to the pith by a 2-5 mm
thick dense concentration of parenchyma cells; this is known as an epicormic trace or bud trace
(Fontaine et al., 1998). This trace usually occurs perpendicular to the pith of the main stem
(Büsgen and Münch, 1929; Colin et al., 2010) and lengthens each year with annual growth rings,
with the vegetative meristem persisting just beneath the bark. Preventitious epicormics buds are
separated further into small and large buds. Small preventitious epicormic buds measure less
than 2 mm long (fig. 1.4) and are typically associated with rings of bud scale scars. Large
preventitious epicormics buds measure 3-4 mm long (fig. 1.5) and are located along the annual
shoot (Fontaine et al., 1998). The upper third of large preventitious epicormics is characterized
by meristematic areas measuring 100 µm long and 30 µm wide. Secondary bud primordia are
present in the lower two thirds, within their primary bud. Small preventitious epicormics do not
have these secondary buds, and are composed only of a terminal meristem surrounded by scales
(Fontaine et al., 1998).
Large preventitious epicormic buds are more likely to sprout than are small preventitious
epicormics buds (Braham and Kellison, 1987). Younger epicormic buds are also more likely to
sprout into an epicormic branch than are buds on older wood (Kormanick and Brown, 1969;
Colin et al., 2010). However, smaller buds tend to persist for many years compared to large buds
(Gruber, 1994) because large epicormic buds that do not form a branch are more likely to abscise
than are small buds (Harmer, 1991; Fontaine et al., 2001).

9

Epicormic traces differ in composition between large and small diameter stems in
Eucalyptus cladocalyx (Burrows, 2000). Strands of small stems (0.5-4.5 cm diameter) consisted
of meristematic tissue in 3-13 strips just behind the bud; these strips were embedded in a strand
of parenchyma cells 0.7-1.4 mm wide and 2.0 mm high. The cells in the strand became more
lignified near the pith. Strands were more complicated in large diameter (7-30 cm) stems. These
strands were most complex in the vascular cambium, with an elliptical shape in transverse
section containing16-40 meristematic domes. Cambia at the edges of domes differentiated into
parenchyma. Cambia produced all the features of secondary xylem: fibers, fiber tracheids, xylem
parenchyma and ray parenchyma. Strands in stems greater than 20 cm were significantly thicker
than thin stems, measuring 3-5 mm high; some stems actually consisted of two smaller strands.
Strand cross-sectional area increased with increasing stem diameter (Burrows, 2000).
Burrows (1989) examined the development of axillary meristems following removal of
the apical meristem on newly formed shoots in hoop pine (Araucaria cuninghamii Aiton ex D.
Don). At decapitation, apical meristem, leaf primordia or vascular connections were not present
in the axillary meristem. A shell zone of thick-walled cells in a crescent shape was on the adaxial
side of the axillary meristem; periderm, the bark patch, was on the abaxial side. There was an
increase in nucleus size and cytoplasmic density of the apical meristem on day 3. Anticlinal cell
division began at the back of the axillary meristem on day 6 and continued through day 9.
Domed bud primordia formed between days 9 and 12 from periclinal divisions that began in the
corpus of the axillary meristem. Leaf primordia initiation quickly followed on the apical dome.
Two procambial strands formed across the cortex from cortical dedifferentiation from anticlinal
divisions. These two strands followed parallel to each other through most of the cortex, with the
upper strand moving perpendicular to or slightly downward from the stem axis; the upper strand

10

took a strong downturn at the central vascular cylinder. The lower strand turned sharply before
the upper, at the inner cortex, joining the axial vascular system.
By day 15, 3-5 leaf primordia had been initiated, anticlinal division continued, enlarging
procambial strands (Burrows, 1989). Swelling of the axil occurred beginning at day 6, and by
day 18, it had swelled large enough that the bark patch had split and the older leaf primordia
were visible in the leaf axil. No new structures developed after 21 days, and further growth was a
continuation of ongoing processes. A stem had begun to form at day 35, and this stem had
developing buds. Burrows also examined bud development in old branches and found more
sclerenchyma rings in mature versus newly formed shoots, meaning a greater barrier to form
vascular connections for initiated buds in mature shoots.
Sylleptic Branching
Proleptic branches develop from axillary buds in the year after the bud forms. However,
sylleptic branches grow in the same year that their parent branch originated (Spath, 1912).
Proleptic branches with sylleptic branching have a higher translocation efficiency and growth
rate than proleptic branches without sylleptic branching (Scarascia-Mugnozza et al., 1999).
Comparing three poplar (Populus tichocarpa) hybrids, Cline and Don-Il (2002) showed that
sylleptic branching was highest in the least vigorous hybrid. The investigators did note that all
three clones had relatively high growth rates. Decapitation of the apical meristem followed by
applications of synthetic auxin (1% naphthaleneacetic acid) to maintain apical dominance was
most effective in the fastest growing clone, resulting in decreased sylleptic branching of the
shoot arising from the top bud. It was therefore concluded that greater insensitivity to auxin is
the cause for increased sylleptic branching. Cytokinin applications of 1 mM benzyladenine to an
inhibited bud were least effective in the clone that produced the fewest sylleptic branches;

11

applications had the greatest response in the clone that formed an intermediate level of sylleptic
branches. These results partially supported the investigators’ hypothesis that higher sensitivity to
cytokinin is responsible for increased sylleptic production (Cline and Dong-Il, 2002).
Other studies have examined the growth habit of two peach varieties, Pillar and Standard.
Standard trees had a more spreading orientation and lower auxin:cytokinin ratio had increased
sylleptic branching over the upright growth of Pillar (Tworkoski et al., 2006). Tworkoski et al.
hypothesized that sylleptic branching occurred further from the shoot apex in Pillar because
greater distance was needed for the auxin:cytokinin ratio to become sufficiently low. Roots are
believed to be a significant source of cytokinin. In peach, removal of 50% of the rooting area
inhibited lateral shoot production; however, this was overcome by applications of benzyladenine,
a synthetic cytokinin (Richards and Rowe, 1977). Pillar trees were shown to have a decreased
rooting area compared to standard trees (Tworkoski and Scorza, 2001), which therefore was
hypothesized to be the cause of a lower cytokinin content in Pillar trees and, in turn, less
sylleptic branching compared to standard trees (Tworkoski et al., 2006).
Pillar shoots were more vigorous in the upper third of the canopy, with twice the shoot
length and lateral branches in the upper compared to the bottom third. Auxin:cytokinin ratios did
not change throughout the canopy of Pillar trees, but sylleptic branching was greater in the upper
third (Tworkoski et al., 2006). Tworkoski et al. concluded that auxin:cytokinin ratios exert
limited control over sylleptic branching when there isn’t strong competition between sinks or a
limited resource availability.
Conclusions
Renewal of fruiting sites is necessary for production of high quality sweet cherries
throughout the life of an orchard. Traditional annual renewal is done by removing a few of the

12

largest branches in the canopy. This creates growth resource competition in the tree between
fruiting sites and vegetative renewal sites; in high density orchards utilizing dwarfing rootstocks,
this competition decreases the growth of renewal sites. Whole tree renewal eliminates this
competition, transitioning the tree to only vegetative growth and development.
Apical dominance controls branching through the production of auxin, which plays an
indirect role in maintaining axillary bud paradormancy (Leyser, 2003). Removal of the shoot
apex halts the production of auxin, and latent or axillary dormant buds have the potential to
become activated and develop into a branch. Epicormic branching is a regenerative response of
woody plants to refill canopy space from epicormic buds after the bud has been released from
paradormancy. Removal of a greater portion of the canopy results in increased epicormic
branching (O’Hara and Berrill, 2009). Sylleptic branches develop in the year that their parent
branch forms. Sylleptic branching is also under the influence of apical dominance and occurs
when auxin:cytokinin ratios are low (Tworkoski et al., 2006).

13

Year 1
Figure 1.1- First year of growth: branch with single leaves at each node (Long et al., 2015).

Year 1

Year 2

Figure 1.2- Second year of growth: basal fruit, vegetative spurs along one year old growth, and
annual extension of new growth (Long et al., 2015).

Year 3

Year 2

Year 1

Figure 1.3- Third year of growth: fruiting spurs along two-year-old growth, basal fruit on one
year old growth, followed by vegetative spurs, and finally another section of new growth (Long
et al., 2015).

14

Figure 1.4- Scanning electronic microscope image of small preventitious epicormic buds
without secondary bud primordia (Fontaine et al., 1999).

15

Figure 1.5- Large preventitious epicormic bud: SEM of large primary bud in the center,
surrounded by secondary bud primordia (Fontaine et al., 1999).

16

LITERATURE CITED

17

LITERATURE CITED

Ayala, M. and Lang, G. A. (2017). Morphology, Cropping Physiology, and Canopy Training, p.
269-304. In: Quero-Garcia, J., Iezzoni, A., Pulawska, J., and Lang, G. A. (Eds.). Cherries:
Botany, Production and Uses. CABI.
Braham, R. R., and Kellison, R. C. (1987). Suppressed buds in yellow-poplar. Journal of the
Elisha Mitchell Scientific Society, 47-55.
Brown, C.L. 1971. Primary Growth. In: Zimmerman, M. H., and Brown, C. L. (1971). Trees:
Structure and Function. New York, USA, Springer-Verlag..
Büsgen, M. and Münch, E. (1929). The Structure and Life of Forest Trees. John Wiley and Sons
Inc., New York, NY, USA. 436 pp.
Burrows, G. E. (1989). Developmental anatomy of axillary meristems of Araucaria
cunninghamii released from apical dominance following shoot apex decapitation in vitro and in
vivo. Botanical Gazette, 369-377.
Burrows, G. E. (2000). An anatomical study of epicormic bud strand structure in Eucalyptus
cladocalyx (Myrtaceae). Australian Journal of Botany, 48(2), 233-245.
Burrows, G. E. (2008). Syncarpia and Tristaniopsis (Myrtaceae) possess specialised fire-resistant
epicormic structures. Australian Journal of Botany, 56(3), 254-264.
Cline, M. G. (1996). Exogenous auxin effects on lateral bud outgrowth in decapitated
shoots. Annals of Botany, 78(2), 255-266.
Cline, M. G. (2000). Execution of the auxin replacement apical dominance experiment in
temperate woody species. American Journal of Botany, 87(2), 182-190.
Cline, M. G., and Dong-Il, K. I. M. (2002). A preliminary investigation of the role of auxin and
cytokinin in sylleptic branching of three hybrid poplar clones exhibiting contrasting degrees of
sylleptic branching. Annals of Botany, 90(3), 417-421.
Colin, F., Ducousso, A., and Fontaine, F. (2010). Epicormics in 13-year-old Quercus petraea:
small effect of provenance and large influence of branches and growth unit limits. Annals of
Forest Science, 67(3), 312.
Cooper-Ellis, S., Foster, D. R., Carlton, G., and Lezberg, A. (1999). Forest response to
catastrophic wind: results from an experimental hurricane. Ecology, 80(8), 2683-2696.
Fink, S. (1983). The occurrence of adventitious and preventitious buds within the bark of some
temperate and tropical trees. American Journal of Botany, 532-542.

18

Fink, S. (1999). Pathological and regenerative plant anatomy. Gebruder Borntraeger
Verlagsbuchhandlung.
Fontaine, F., Druelle, J. L., Clément, C., Burrus, M., and Audran, J. C. (1998). Ontogeny of
proventitious epicormic buds in Quercus petraea. I. In the 5 years following initiation. Trees,
13(1), 54-62.
Fontaine, F., Kiefer, E., Clément, C., Burrus, M., and Druelle, J. L. (1999). Ontogeny of the
proventitious epicormic buds in Quercus petraea. Trees-Structure and Function, 14(2), 83-90.
Gordon, D., Rosati, A., Damiano, C., and DeJong, T. M. (2006). Seasonal effects of light
exposure, temperature, trunk growth and plant carbohydrate status on the initiation and growth of
epicormic shoots in Prunus persica. Journal of Horticultural Science and Biotechnology, 81(3),
421-428.
Gruber, F. (1994). Morphology of coniferous trees: possible effects of soil acidification on the
morphology of Norway spruce and Silver fir. Effects of Acid Rain on Forest Processes, 265-324.
Harmer, R. (1991). The Effect of Bud Position on Branch Growth and Bud Abscission in
Quercus petraea (Matt.) Liebl. Annals of Botany, 67(5), 463-468.
Hillman, J. R., Math, V. B., and Medlow, G. C. (1977). Apical dominance and the levels of
indole acetic acid in Phaseolus lateral buds. Planta, 134(2), 191-193.
Kauppi, A., Rinne, P., and Ferm, A. (1987). Initiation, structure and sprouting of dormant basal
buds in Betula pubescens. Flora, 179(1), 55-83.
Kormanik, P. P., and Brown, C. L. (1969). Origin and development of epicormic branches in
sweetgum. US For. Serv. Res. Pap. Stheast For. Exp. Sta., No. SE-54, 17pp. [For. Abstr. 31
(1970) No. 6103.] Buds, Hamamel (PMBD, 185403152).
Lang, G. A. (2000). Precocious, dwarfing, and productive—how will new cherry rootstocks
impact the sweet cherry industry? HortTechnology, 10(4), 719-725.
Lang, G. A., Olmstead, J. W., and Whiting, M. D. (2004). Sweet cherry fruit distribution and leaf
populations: modeling canopy dynamics and management strategies. Acta Horticulturae, 591600.
Leyser, O. (2003). Regulation of shoot branching by auxin. Trends in Plant Science, 8(11), 541545.
Ljung, K., Bhalerao, R. P., and Sandberg, G. (2001). Sites and homeostatic control of auxin
biosynthesis in Arabidopsis during vegetative growth. The Plant Journal, 28(4), 465-474.

19

Long, L. E., Lang, G. A., Whiting, M. D., and Musacchi, S. (2015). Cherry training systems.
Pacific Northwest Extension. Oregon State University. University of Idaho. Washington State
University.
Maguylo, K., Lang, G. A., and Perry, R. L. (2002, August). Rootstocks genotype affects flower
distribution and density of 'Hedelfinger'sweet cherry and'Montmorency'sour cherry. Acta
Horticulturae, 636 (pp. 259-266).
Michigan Ag Council. (2016). Michigan Cherries - Michigan Agriculture. Retrieved October 17,
2017, from https://michiganagriculture.com/foods/michigan-cherries/
Nicolini, E., Chanson, B., and Bonne, F. (2001). Stem growth and epicormic branch formation in
understorey beech trees (Fagus sylvatica L.). Annals of Botany, 87(6), 737-750.
O'Hara, K. L., and Berrill, J. P. (2009). Epicormic sprout development in pruned coast redwood:
pruning severity, genotype, and sprouting characteristics. Annals of Forest Science, 66(4), 1-9.
Piene, H., and Eveleigh, E. S. (1996). Spruce budworm defoliation in young balsam fir: the
‘green’tree phenomenon. The Canadian Entomologist, 128(06), 1101-1107.
Perry, R., Lang, G., Andersen, R., Anderson, L., Azarenko, A., Facteau, T., ... and Rom, C.
(1997, July). Performance of the NC-140 cherry rootstock trials in North America Acta
Horticulturae, 468 (pp. 291-296).
Richards, D., and Rowe, R. N. (1977). Effects of root restriction, root pruning and 6benzylaminopurine on the growth of peach seedlings. Annals of Botany, 41(4), 729-740.
Robinson, T. L., DeMarree, A. M., and Hoying, S. A. (2004, June). An economic comparison of
five high density apple planting systems. In Acta Horticulturae, 732 (pp. 481-489).
Sachs, T., and Thimann, K. V. (1967). The role of auxins and cytokinins in the release of buds
from dominance. American Journal of Botany, 136-144.
Scarascia-Mugnozza, G. E., Hinckley, T. M., Stettler, R. F., Heilman, P. E., and Isebrands, J. G.
(1999). Production physiology and morphology of Populus species and their hybrids grown
under short rotation. III. Seasonal carbon allocation patterns from branches. Canadian Journal of
Forest Research, 29(9), 1419-1432.
Schaetzl, R. (n.d.). Fruit Production. Retrieved December 03, 2017, from
http://geo.msu.edu/extra/geomich/fruit.html
Sorefan, K., Booker, J., Haurogné, K., Goussot, M., Bainbridge, K., Foo, E., ... and Leyser, O.
(2003). MAX4 and RMS1 are orthologous dioxygenase-like genes that regulate shoot branching
in Arabidopsis and pea. Genes & Development, 17(12), 1469-1474.
Spath H. (1912). Der Johannistriebe. Berlin: Parey.

20

Stone, E. L., and Stone, M. H. (1943). "Dormant" versus "Adventitious" buds. Science,
98(2533), 62-62.
Sundberg, B., and Uggla, C. (1998). Origin and dynamics of indoleacetic acid under polar
transport in Pinus sylvestris. Physiologia Plantarum, 104(1), 22-29.
Tworkoski, T., Miller, S., and Scorza, R. (2006). Relationship of pruning and growth
morphology with hormone ratios in shoots of pillar and standard peach trees. Journal of Plant
Growth Regulation, 25(2), 145-155.
Wignall, T. A., and Browning, G. (1988). Epicormic Bud Development in Quercus robur L.
Studies of Endogenous IAA, ABA, IAA Polar Transport and Water Potential in Cambial
Tissues10. Journal of Experimental Botany, 39(12), 1667-1678.
Wignall, T. A., Browning, G., and Mackenzie, K. A. D. (1987). The physiology of epicormic
bud emergence in Pedunculate Oak (Quercus robur L.) Responses to partial notch girdling in
thinned and unthinned stands. Forestry: An International Journal of Forest Research, 60(1), 4556.

21

CHAPTER 2: WHOLE TREE RENEWAL REGENERATES FRUITING
STRUCTURES QUICKLY IN MATURE ORCHARDS
Introduction
The largest and highest quality sweet cherries are borne on basal flower buds of one-yearold branches and on young spurs. Annual partial renewal (APR) of ~10-20% of the canopy, by
removing the largest one to several branches during dormancy and allowing for spring regrowth
from latent or epicormic buds buried beneath the bark, maintains young fruiting sites (Long et
al., 2015). APR pruning creates competition in the tree for photoassimilates between sinks, i.e.,
fruiting sites and vegetative renewal sites (G.A. Lang, personal communication). In high density
(1200-4000 trees per ha-1) sweet cherry orchards, precocious, dwarfing rootstocks are utilized to
control tree vigor and obtain high productivity (Lang, 2000; Perry et al., 1997). Compared to
vigorous rootstocks, dwarfing rootstocks direct a greater percentage of their photoassimilates to
fruit production rather than vegetative growth (Atkinson and Else, 2001). This can result in
inconsistent regrowth following APR pruning, making it difficult to maintain an optimal balance
of branch ages in the canopy.
Modern orchards have moved away from traditional tree spacing (250-400 trees per ha-1)
with large, complex trees and wide rows. High density orchards utilizing dwarfing rootstocks
come into full production quickly (5-7 years) compared to traditional systems (8-12 years) and
often utilize simplified training systems with narrow “fruiting walls” that increase light
penetration and labor efficiency, leading to greater returns on investment for growers. Small,
simplified canopies have made it possible to utilize protective plastic covers in sweet cherry
production, such as high tunnels and row covers. High rates of water uptake after rain through
either the fruit cuticle or the tree root system can induce cracking of sweet cherries (Measham et

22

al., 2009); covering systems can reduce or eliminate this loss, as well as limit disease and
improve frost protection.
Auxin (indole-3-aceetic acid) produced in the shoot apex and newly formed leaves
(Ljung et al., 2001) moves basipetally through polar transport (Wignall et al., 1987; Sundberg
and Uggla, 1998) and inhibits outgrowth of axillary meristems through an indirect mechanism
(Leyser, 2003). In woody plants, axillary meristems that maintain paradormancy under apical
dominance eventually become engulfed by radial growth of the branch, persisting just beneath
the bark (Büsgen and Münch, 1929; Stone and Stone, 1943). Removal of the shoot apex stops the
inhibitory effects of auxin and cytokinin transport into buds is increased, promoting bud
outgrowth (Sachs and Thimann, 1967). In the case of epicormic buds, this outgrowth creates an
epicormic branch.
Epicormic branching refills the canopy after damage or stress. Fire (Burrows, 2008),
insect defoliation (Piene and Eveleigh, 1996), wind damage (Cooper-Ellis et al., 1999),
competition (Nicolini et al., 2001), and pruning (O’Hara and Berrill, 2009) have all been shown
to initiate epicormic branching. Increasing the proportion of the canopy subjected to pruning
results in a higher number and more vigorous epicormic branches produced (O’Hara and Berrill,
2009). Trees grown under no light were shown to have a 25% reduction in epicormic branching
(Gordon et al., 2006). Pruning initiates epicormic branching by removing the correlative
inhibition of auxin being synthesized as well as by increasing light exposure.
Sylleptic branches sprout in the same year that their parent branch originates (Spath,
1912). The propensity to produce sylleptic branches is related to auxin sensitivity; greater
insensitivity to auxin increases sylleptic branching (Cline and Dong-Il, 2002). Lower auxin-tocytokinin ratios resulted in more sylleptic branching in peach; ‘Pillar’ trees exhibited stronger

23

apical dominance with a more upright growth habit than ‘Standard’, which had a lower auxin to
cytokinin ratio and more sylleptic branching (Tworkoski et al., 2006). Auxin-to-cytokinin ratios
had less of an effect on sylleptic branching when competition between sinks was less. In ‘Pillar’,
shoots in the upper third were more vigorous and produced more sylleptic branches, although
auxin-to-cytokinin ratios were constant throughout the tree (Tworkoski et al., 2006).
Whole tree renewal (WTR; the removal of all fruiting portions of the canopy, leaving
only the primary vegetative structure) eliminates competition in the canopy between fruiting and
vegetative renewal sites. In contrast to APR pruning of every tree in the orchard every year, the
concept of WTR pruning could be applied to the entire orchard every 6-7 years or to only about
15% of the trees in the orchard every year. The objective of this study was to document and
analyze the initial WTR regrowth response of various high density sweet cherry training systems
on rootstocks that differ in vigor, and determine how quickly canopy structure and fruiting sites
are regenerated after WTR.
Materials and Methods
Site and Plant Material
This study was conducted at the Michigan State University Clarksville Research Center
(Clarksville, MI, lat. 42.8ºN, long. 85.2ºW) within the NC-140 sweet cherry training systems x
rootstock trial, planted in 2010. WTR pruning was imposed on irrigated eight-year-old trees of
the cultivar ‘Benton’ grafted onto three rootstocks of differing vigor: Gisela 3, dwarfing; Gisela
5, semi-dwarfing; and Gisela 6, semi-vigorous, growing in a coarse-loamy, mixed, mesic Typic
Hapludalf soil of the Lapeer series (Soil Survey Staff, Natural Resources Conservation Service,
United States Department of Agriculture, 2017). Four canopy training systems were studied on
each of these rootstocks: a standard central leader tree, Tall Spindle Axe (TSA); a planar central

24

leader tree Super Slender Axe (SSA); a multiple leader bush, Kym Green Bush (KGB); and a
trellised narrow planar fruiting wall, Upright Fruiting Offshoots (UFO), which has multiple
leaders arising from a horizontal “cordon-like” trunk planted at a 45° angle (fig. 2.1). The only
exception was the lack of an SSA x Gi5 combination in the trial; therefore, there were 11 training
system by rootstock combinations. Between row spacing is 3.5 m. Within row spacing is 1.5 m
between trees for TSA, UFO and KGB trees and 0.75 m for SSA trees. Each of these 11 training
system by rootstock treatment combinations make up a replication, occupying two rows in the
orchard. TSA, UFO, and KGB each have four trees per treatment group, SSA at twice the density
has eight trees per treatment group. One tree in each of these treatment groups was used for
whole tree renewal, the other trees are annually renewed each year. A guard tree, either the
cultivar ‘Attika’ on Gi6 or ‘NY119’ on Gisela 12, trained to TSA separates treatment groups.
Whole Tree Renewal Pruning
WTR pruning was imposed at full bloom, April 27, 2016. Preliminary research (G.A.
Lang, personal communication) found the best regrowth response when WTR pruning was
imposed at full bloom, compared to dormant, green tip bud swell or petal fall. WTR pruning
removed all lateral fruiting branches (TSA, SSA) or upright fruiting leaders (UFO, KGB),
leaving ~10 cm stubs from the central trunk (TSA, SSA), horizontal “cordon” (UFO), or base of
the bush (KGB). After renewal cuts were made, epicormic branches were allowed to develop
uninhibited throughout the 2016 growing season.
Initial Regrowth Measurements
Initial regrowth data was collected in early winter (December 15-January 5), once leaves
had abscised and the trees had gone dormant (December 2016). Total number of primary renewal
branches, length of each primary renewal branch, and number of sylleptic branches per primary

25

branch was all measured once in the field. Length measurements were done using a tape
measure, from the base of the branch to the terminal bud. The TSA and SSA central leader and
the UFO “cordon leader” were marked into thirds (distal, middle, and proximal) to quantify
renewal shoot distribution and uniformity. The KGB was not able to be split into thirds because
all of the leaders were cut back to relatively similar points of origin in the base of the bush.
2017 Maintenance Pruning
In spring 2017, the WTR trees were pruned following standard guidelines for
establishment and maintenance of each training system as described in Long et al. (2015). For
the TSA trees, renewal shoots that were too vigorous, too upright, too pendent, or overlapping
were removed, and lateral shoots that were retained were head-pruned to remove 15-25% of their
length and stimulate secondary lateral branching. All sylleptic shoots were retained unless they
were too upright, pendent, or overlapping. For the SSA trees, all renewal shoots were “shortpruned” (head-pruned to remove all but the most basal single flower buds plus 1-3 basal
vegetative buds). For the UFO trees, vertically-oriented renewal shoots were thinned and tied to
the trellis about every 8 inches, and horizontal or pendent renewal shoots were removed, as were
any sylleptic branches. For KGB trees, vertically-oriented renewal shoots were retained unless
they were excessively vigorous, and any sylleptic branches were removed. Horizontal or pendent
renewal shoots were removed as well.
Canopy Volume Measurements
In August 2017, following the second season of regrowth and standard summer pruning,
the canopy volumes of WTR trees were compared to those of adjacent APR pruned trees (those
that have the largest one to several branches removed every year). Summer pruning composed of
postharvest hedging, leaving ~25 cm of new east-west lateral growth on UFO and KGB upright

26

leaders, or leaving ~60 to ~90 cm (measured from the central leader) of horizontal east-west
lateral branch growth on SSA and TSA central leaders, respectively. After this mid-summer
hedging, little vegetative regrowth was expected. Therefore, August canopy measurements
allowed comparison of the canopy volumes as significant spur fruiting is expected to begin the
third year after WTR (2018). Within row (N-S) and across row (E-W) canopy spread
measurements were taken at 0.75 m, 1.5 m, and 2.25 m from the ground canopy; with a
measuring tape on August 8, 2017. Canopy spread was measured at each height by measuring
from the end of the furthest reaching branch on one side of the canopy to the furthest reaching
branch on the opposite side. For example, at the 0.75-meter height the within row spread was the
distance between the northern most branch and the southernmost; across row was distance
between eastern and westernmost branch at 0.75 meters from the ground.
Covering System
Each training system by rootstock combination was replicated under two different
covering systems: CRAVO (Brantford, Ontario) and VOEN (Berg, Germany). The CRAVO is a
plastic-covered structure with a programmable retractable roof and north-south sides that
generally remain opens during the summer except when it rains, at which time the roof closes
automatically. In late winter and early spring, the sides and roof of the CRAVO are closed to
capture solar radiation that promotes earlier bloom, and propane convection heaters (80,000 BTU
Mr. Heater, Enerco, Cleveland, OH) are placed inside to protect from spring frosts. The VOEN is
a plastic-and-net rain cover that opens over the tree row and clips together in the alleyway.
VOEN covers were opened after fruit set in May.

27

Data Analysis and Plot Design
Being part of the NC-140 sweet cherry rootstock trial, the greater trial is laid out in a
completely randomized block design, blocking based on CRAVO and VOEN; with two rows
representing a replication, having all 11 training system by rootstock combinations. The NC-140
trial utilizes four trees per training system by rootstock treatment for TSA, UFO, and KGB on all
three rootstocks; SSA being planted at twice the density, has eight trees per treatment. A single
guard tree, either the cultivar ‘Attika’ on Gi6 or ‘NY 119’ on Gi12 rootstock trained to a TSA
separates treatments. Whole tree renewal was imposed on one tree per treatment, the outside tree
in the group of four or eight. Three replications were under the CRAVO and two were under the
VOEN covers, just west of the CRAVO plots. Each replication consisted of a single tree in each
location, therefore, n=33 for CRAVO and n=22 for VOEN.
For initial regrowth data, two-way ANOVAs were determined using Tukey’s LSD,
significance at p≤0.05 (PROC MIXED, SAS version 9.3, SAS Institute, Cary, NC). Due to this
study being unbalanced, missing the SSA x Gi5 combination, analyses between rootstocks were
made without any SSA data, and comparisons between training systems were made without any
data from Gi5. Furthermore, because the KGB permanent tree structure could not be split up in
thirds, comparisons were made on a whole tree basis with KGB and KGB data were left out
when comparing canopy thirds. CRAVO and VOEN data were analyzed separately and therefore
have their own ANOVA table and graphs for each initial response measurement.
Due to the varied canopy architectures, canopy spread was calculated by averaging inand between-row measurements at each height (0.75, 1.5, and 2.25 m). Canopy spread was
assessed at each height using one-way ANOVAs between WTR and APR trees for each training
system by rootstock combination. Location was included in ANOVAs, analyzing CRAVO and

28

VOEN together. ANOVAs and Tukey LSD were calculated at significance of p≤0.05 in RStudio,
version 1.1.383 (Vienna, Austria).
Results
Canopy Volume
After two seasons of regrowth and standard canopy management following WTR
pruning, all training system x rootstock combinations had completely refilled their allotted
orchard space (as measured against respective APR tree canopy spreads) except for the 0.75 m
height of canopy spread for SSA x Gi3 and the 2.25 m canopy height for KGB x all rootstocks
(figs. 2.2-5). WTR-pruned TSA and UFO trees completely refilled their space. Canopy spread
varied a bit by location, but the comparative trend between WTR versus APR pruning remained
the same in both the CRAVO- and VOEN-covered orchards (figs. 2.2, 2.3, and 2.5).
Initial Growth Data
Regarding the initial regrowth data, the number of renewal shoots was greater in the
distal section of the canopy across all rootstocks and training systems analyzed by canopy
proportion (TSA, SSA, UFO); there were no differences in renewal shoot number between
middle and proximal sections. This trend was the same between CRAVO and VOEN (figs. 2.6
and 2.7). On a whole tree basis, Gi6 responded with more replacement shoots than Gi5 and Gi3
under the CRAVO (fig. 2.8), but there were no rootstock differences for trees grown under the
VOEN (table 2.1). Under the CRAVO, TSA and SSA trees had a greater number of new shoots
compared to KGB and UFO trees (fig. 2.9A). TSA trees had more new shoots than any of the
other systems under the VOEN (fig. 2.9B).
The three-way interaction term was significant for canopy section by rootstock by system
for average shoot length across all rootstocks and systems in the CRAVO (p=0.0003) and the

29

VOEN (p=0.0361) orchards. The only discernable trend was that UFO generally had longer
shoots, regardless of rootstock or canopy section than did TSA and SSA in both locations (figs.
2.10 and 2.11). This trend remained when analyzing the whole tree over all rootstocks: UFO and
KGB trees had the longest average new shoot length, followed by TSA, and SSA trees had the
least in the CRAVO; under the VOEN, the KGB trees had greater average shoot lengths than did
UFO trees, which were greater than TSA, and SSA once again had the shortest shoot lengths
(figs. 2.12A and 2.12B). In the CRAVO, trees on Gi6 and Gi5 had longer renewal shoot lengths
than trees on Gi3 (fig. 2.13A). However, under the VOEN, new shoot length for trees on Gi6 was
greater than those on Gi5, and trees on Gi3 were not statistically different from either Gi6 or Gi5
(fig. 2.13B).
The variability of sylleptic branching was wide (fig. 2.20). Comparing the canopy
sections, the distal third had more sylleptic branching than the middle and proximal sections in
the CRAVO when SSA data was omitted (fig. 2.14). However, in the VOEN, the three-way
interaction term for system by rootstock by canopy section was significant; there was no
observable trend in this data (fig. 2.15). Removing the Gi5 data, rootstock by canopy section was
significant in the CRAVO and system by canopy section was significant in the VOEN. In the
CRAVO, the distal canopy section of trees on Gi6 had more sylleptic branching than the middle
and proximal sections, as well as all canopy sections of trees on Gi3 (fig. 2.16). The VOEN data
showed the middle canopy section of UFO trees had greater sylleptic branching than the distal
and proximal sections, as well as all canopy sections of TSA and SSA trees (fig. 2.17). When
comparing sylleptic branching across rootstocks, there were no differences in either the CRAVO
or the VOEN when the SSA data was omitted (tables 2.2 and 2.3). When comparing sylleptic
branching across all four training systems (requiring the omission of the Gi5 data), results in the

30

CRAVO and VOEN differed slightly: the KGB and UFO trees had more sylleptic branching than
TSA and SSA trees in the CRAVO (fig. 2.17A), while KGB tree sylleptic branching was greater
than that for TSA, SSA, and UFO trees in the VOEN (fig. 2.17B).
While sylleptic branching was quite variable, there was a general trend that the longest
primary renewal shoots produced the most sylleptic branches (fig. 2.20). Examining the data as
quartiles reveals that the longest 25% of renewal branches had the greatest sylleptic branch
variability and the widest distribution (fig. 2.21).
Discussion
The canopy spread data show that WTR pruning of KGB trees did not completely refill
their upper canopy volume within two growing seasons (fig. 2.5). This represents a significant
fruiting area. While KGB and UFO trees had the longest average new shoot length on a whole
tree basis (figs. 2.12A and 2.12B), KGB renewal shoots started relatively close to the ground and
had the greatest vertical and horizontal canopy volume to be refilled. UFO trees had a similar
vertical canopy space to refill, but the structured planar nature of UFO canopy provides a much
smaller horizontal canopy volume per tree, which was readily refilled within two seasons. The
inability for KGB trees to adequately refill their canopy volume within two seasons indicates that
recovery of full fruiting potential will be delayed compared to the other training systems.
The lowest canopy spread height (0.75 m) was the only part of the SSA canopy on Gi3
that was not completely refilled within two seasons following WTR pruning (fig. 2.2). These are
the weakest trees in the trial, with the greatest root competition from close spacing, the most
dwarfing rootstock, the smallest trunk cross-sectional area (fig. 2.19), and consequently the
smallest storage potential for growth reserves. These factors associated with decreased vigor
likely resulted in the proximal canopy section of SSA on Gi3 having some of the shortest

31

renewal shoot lengths (fig. 2.11A and 2.11B) and, like the KGB, a delayed recovery of full
fruiting potential. This scenario is contrary to a previous study that showed trees of decreased
vigor produced more epicormic branches (O’Hara and Valappil, 2000). However, this study did
not prune trees as severely as WTR, and pruning severity has been positively correlated with
epicormic branching (O’Hara and Berrill, 2009). Therefore, under an extreme imposed stress
such as WTR pruning, vigor might play a more important role in epicormic development than
when decreased vigor is the main stress, as in O’Hara and Valappil (2000).
Like KGB trees, TSA trees also have a complex, 3-dimensional canopy. However,
maintaining the length of the TSA central leader provides a great potential number of epicormic
buds or “epicormic potential” (Fontaine et al., 2001), and the canopy volume to be refilled after
WTR pruning is primarily horizontal rather than vertical. The greater epicormic potential of TSA
trees resulted in more renewal branches initiated. Likewise, the central leader SSA trees have a
similar epicormic potential and therefore comparable renewal branch numbers initiated in the
CRAVO. However, SSA trees had fewer renewal branches sprout under the VOEN (figs. 2.9A
and 2.9B). It is unclear why this inconsistency exists; it may be a data artifact due to only two
replications in the VOEN, or possible outlier SSA data related to an extensive bacterial canker
infection when the trees were four years old (2012) that was more severe in the VOEN plot
compared to the CRAVO plot.
While the number of sylleptic branches per renewal shoot varied greatly by primary
renewal branch length, there was a general trend of longer renewal branches producing more
syllepetic branches (fig. 2.20). The longest 24% of primary renewal branches resulted in the
widest distribution of sylleptic branches (fig. 2.21). The most vigorous renewal branches
producing more syllepetic branches is contrary to the results of Cline and Don-Il (2002), who

32

found that the most vigorous of three poplar hybrids had the greatest suppression of sylleptic
branches after synthetic auxin was applied following shoot apex decapitation.
Significant differences between training systems for number of sylleptic branches almost
mirrored those of length per renewal branch. In the CRAVO, KGB and UFO trees had the
longest length and most sylleptic branching. In the VOEN, UFO shoot length was less than
KGB, as was the number of sylleptic branches produced (figs. 2.12A, 2.12B, 2.18A, 2.18B). This
trend supports the conclusion that the most vigorous renewal shoots resulted in the highest
number of sylleptic branches. This is counter to the report of Tworkoski et al. (2006), who found
that between two architecturally distinct peach genotypes, the one with more upright growth
displayed greater apical dominance and less sylleptic branching. In our current study, the
renewal shoots of the UFO and KGB trees are trained to fill vertical space, yet these had more
sylleptic branching than the lateral renewal shoot growth of TSA and SSA trees.
Sylleptic branching may be a positive or negative response, depending on sweet cherry
training system. KGB and UFO trees crop primarily on spurs on upright leaders; a sylleptic
branch occupies a node that eventually would have become a fruiting spur, ultimately resulting
instead in blind wood that decreases fruiting potential. Conversely, some sylleptic branches on
TSA and SSA trees can be used to replace horizontal fruiting structure, refilling canopy volume
more quickly. However, this study has showed that sylleptic branches are characteristic of the
more vigorous renewal branches, which typically are unfruitful. It is therefore questionable
whether sylleptic branches are a positive or negative response for TSA and SSA trees; further
flowering data from spring 2018 will address this question.
The extreme shock imposed by WTR pruning must be considered when comparing our
results on sylleptic branching to past studies. Removing all branches removes the sites of auxin

33

synthesis (Ljung et al., 2001). Cytokinin production is increased after eliminating the inhibitory
effects of auxin on cytokinin synthesis and export from roots (Leyser, 2003). WTR trees are very
likely in a state where auxin:cytokinin ratios are low, which has been correlated with sylleptic
branching (Tworkoski et al., 2006). TSA and SSA trees had more renewal shoots than UFO and
KGB trees (fig. 2.9A and 2.9B), and therefore it’s likely that a higher auxin content and higher
auxin-to-cytokinin ratios cause less sylleptic branching in the less vigorous, laterally growing
TSA and SSA tree training systems.
CRAVO and VOEN data was analyzed separately for simplicity of graphs, there were
many instances when the location effect was significant. This then made instances where the
three way interaction was significant, system by rootstock by permanent structure section,
making 18 different treatment groups; more complicated when it was a four way interaction,
giving 36 different treatment groups in the same graph. Since we weren’t particularly interested
in differences in covering system, these data were analyzed and plotted separately.
This interaction could simply be due to there being only two replications in the VOEN,
three in the CRAVO, and having an outlier that is influencing the data more. However, there are
also environmental differences that exist between these two covering systems. Looking just at
sylleptic branching for all four training systems averaged over Gi3 and Gi6 there was a 0.5-1
sylleptic branch per renewal branch increase on KGB, TSA, and SSA trees grown in the VOEN
(fig. 2.18B) than CRAVO (fig. 2.18A); UFO remained similar in both locations at ~1.7 sylleptic
branches per primary renewal branch. This increase in sylleptic branching in the VOEN might be
explained by those covers remaining over the trees throughout the growing season, while the
CRAVO covers only close when it begins to rain. These VOEN covers produce shade, leading to
lower light environments in the VOEN than CRAVO. Phototropins, specifically Cryptochromes

34

CRY1, CRY2, and CRY3 have been shown to be upregulated in response to low light in
Arabidopsis thaliana (Takemiya et al., 2005). It is believed that bud burst is promoted by CRY1
and CRY2 protoreceptors promoting expression of ELONGATED HYPOCOTOL5 (HY5)
(Signorelli et al., 2017). HY5 genes are involed in chlorophyll biosynthesis and light harvesting
(Eberhard et al., 2008). The lower light conditions in the VOEN could be resulting in an
upregulation of CRY genes and in turn HY5 leading to greater sylleptic branching in VOEN
trees compared to trees in the CRAVO.
Conclusions
WTR pruning has been shown to quickly refill sweet cherry canopies across multiple
canopy architectures. The resulting replacement canopy fruiting structure is more uniform than
that of APR-pruned trees, which is comprised of a mix of fruiting structures of various ages.
KGB trees have the greatest canopy volume per tree and was the only training system, across all
rootstocks, that did not completely refill its canopy volume within two seasons. Therefore,
recovery of KGB fruiting potential may require an extra year compared to TSA, SSA, and UFO
trees. The number of renewal branches initiated and the average length per renewal shoot were
inversely related—TSA and SSA initiated more branches, but had shorter average renewal shoot
length. Sylleptic branching was highly variable, but tended to be greater in more vigorous
renewal shoots. Further research is needed to confirm the timing of complete recovery of fruiting
potential, as well as to understand where new renewal shoots are most likely to arise.

35

Average	Canopy	Spread	(m)

Figure 2.1- Diagram of each training system used in the study. Kym Green Bush (KGB), Tall
Spindle Axe (TSA), Super Slender Axe (SSA), and Upright Fruiting Offshoots (UFO).

1.5
1

a

a

a
a

a

1.5

2.25

b

a

a

a

a

a

a

a

a

a
a

0.5
0
0.75

0.75

overall
Gi3

1.5

1.5

2.25

2.25

CRAVO

VOEN

CRAVO

VOEN

WTR

APR

Gi6

Figure 2.2- SSA on Gi3 and Gi6 whole tree renewal (WTR) canopy average spread versus
spread of annual partial renewal (APR) trees at 0.75, 1.5, and 2.25 meters in the canopy. Canopy
spread averaged over CRAVO and VOEN at Michigan State University’s Clarksville Research
Center, except where noted as CRAVO or VOEN. Lowercase letters separate means between
WTR and APR at each height tested using Tukey LSD were calculated at significance of p≤0.05.

36

Average	Canopy	Spread	(m)

2

a

a

a

a

a

a

a

a

a

a

a

a

a

1.5

2.25

a

a

a

1

a

a

a

a

0
0.75

1.5

2.25

0.75

1.5

2.25

overall

overall

Gi3

Gi5

0.75

0.75

CRAVO VOEN

overall

Gi6
WTR

APR

Average	Canopy	Spread	(m)

Figure 2.3- TSA on Gi3, Gi5, and Gi6 whole tree renewal (WTR) canopy average spread versus
spread of annual partial renewal (APR) trees at 0.75, 1.5, and 2.25 meters in the canopy. Canopy
spread averaged over CRAVO and VOEN at Michigan State University’s Clarksville Research
Center, except where noted as CRAVO or VOEN. Lowercase letters separate means between
WTR and APR at each height tested using Tukey LSD were calculated at significance of p≤0.05.

2
1

a

a

a

a

a

a

a

a

0.75

1.5

2.25

0.75

a

a

a

a
a

a

a

a
a

a

1.5

2.25

0
1.5

2.25

0.75

overall
Gi3

Gi5
WTR

Gi6
APR

Figure 2.4- UFO on Gi3, Gi5, and Gi6 whole tree renewal (WTR) canopy average spread versus
spread of annual partial renewal (APR) trees at 0.75, 1.5, and 2.25 meters in the canopy. Canopy
spread averaged over CRAVO and VOEN at Michigan State University’s Clarksville Research
Center, except where noted as CRAVO or VOEN. Lowercase letters separate means between
WTR and APR at each height tested using Tukey LSD were calculated at significance of p≤0.05.

37

Average	Canopy	Spread	(m)

2
1

a

a

a

a

a

a
b

a

a
b

a

b

b

a a

a

a

a

a
b

a

a

0
0.75

1.5

1.5

2.25

0.75

1.5

2.25

overall CRAVO VOEN overall
Gi3

0.75

0.75

CRAVO VOEN
Gi5
WTR

1.5

2.25

overall

Gi6
APR

Figure 2.5- KGB on Gi3, Gi5, and Gi6 whole tree renewal (WTR) canopy average spread versus
spread of annual partial renewal (APR) trees at 0.75, 1.5, and 2.25 meters in the canopy. Canopy
spread averaged over CRAVO and VOEN at Michigan State University’s Clarksville Research
Center, except where noted as CRAVO or VOEN. Lowercase letters separate means between
WTR and APR at each height tested using Tukey LSD were calculated at significance of p≤0.05.

38

number	of	renewal	branches

16
14

a

12
10

b

8

b

6
4
2
0
Distal

Middle

Proximal

Figure 2.6A

number	of	renewal	branches

16
14

a

12
10

b

b

8
6
4
2
0
Distal

Middle

Proximal

Figure 2.6B
Figure 2.6- Initial response number of renewal branches initiated per permanent structure
section, of TSA and UFO grown on Gi3, Gi5, and Gi6 after whole tree renewal was imposed.
5A- Bars are average of 18 whole tree renewal trees grown in the CRAVO plus or minus
standard error of the mean. 5B- Bars are average of 12 Whole tree renewal trees grown in the
VOEN plus or minus standard of the mean at Michigan State University’s Clarksville Research
Center. Lowercase letters separate means comparing canopy sections, determined using Tukey’s
LSD, significance at p≤0.05.

39

number	of	renewal	branches

16

a

14
12

b

10
8

b

6
4
2
0
Distal

Middle

Proximal

number	of	renewal	branches

Figure 2.7A
16
14

a

12
10
8

b

b

6
4
2
0
Distal

Middle

Proximal

Figure 2.7B
Figure 2.7- Initial response number of renewal branch per permanent structure section of TSA,
SSA and UFO grown on Gi3 and Gi6 after whole tree renewal was imposed. 5A- Bars are
average of 18 whole tree renewal trees grown in the CRAVO plus or minus standard error of the
mean. 5B- Bars are average of 12 Whole tree renewal trees grown in the VOEN plus or minus
standard of the mean at Michigan State University’s Clarksville Research Center. Lowercase
letters separate means comparing canopy sections, determined using Tukey’s LSD, significance
at p≤0.05.

40

number	of	renewal	branches

35

a

30
b

25

b

20
15
10
5
0
Gi3

Gi5

Gi6

number	of	renewal	branches

Figure 2.8- Initial response number of renewal branch by rootstock averaged over TSA, KGB
and UFO after whole tree renewal was imposed. Bars are an average of 27 whole tree renewal
trees grown in the CRAVO plus or minus standard error of the mean at Michigan State
University’s Clarksville Research Center. Lowercase letters separate means comparing
rootstocks, determined using Tukey’s LSD, significance at p≤0.05.
45
40
35
30
25
20
15
10
5
0

a
a
b

KGB

b

TSA

SSA

UFO

Figure 2.9A

41

number	of	renewal	branhes

Figure 2.9 (cont’d)
45
40
35
30
25
20
15
10
5
0

a

b

b

KGB

TSA

b

SSA

UFO

Figure 2.9B
Figure 2.9- Initial response number of renewal branch by training system, averaged over Gi3
and Gi6. 8A- Bars are average of 21 whole tree renewal trees grown in the CRAVO plus or
minus standard error of the mean. 5B- Bars are average of 14 Whole tree renewal trees grown in
the VOEN plus or minus standard of the mean at Michigan State University’s Clarksville
Research Center. Lowercase letters separate means comparing training systems, determined
using Tukey’s LSD, significance at p≤0.05.

length	per	renewal	branch	(m)

1.8
1.6
					bcd

1.2
1
0.8

		ab

			abc

1.4

f

				f

f

ef def

		def

ef

de

					def			def

abc
					cd

				a

ef 			ef

0.6
0.4
0.2

Gi3

Gi5

Gi6

Gi3

TSA

Gi5
UFO

Figure 2.10A
42

Gi6

Proximal

Middle

Distal

Proximal

Middle

Distal

Proximal

Middle

Distal

Proximal

Middle

Distal

Proximal

Middle

Distal

Proximal

Middle

Distal

0

			abc
de

				a

		ab
				abcd

		abcd		abcd
			abcd			abcd
abcd
cd 	cde bcd
cd

			a

			cd

bcd

Gi3

Gi5

Gi6

Gi3

Gi5

TSA

Proximal

Middle

Distal

Proximal

Middle

Distal

Proximal

Middle

Distal

Proximal

Middle

Distal

Proximal

Middle

Distal

Proximal

e

Middle

1.8
1.6
1.4
1.2
1
0.8
0.6
0.4
0.2
0

Distal

length	per	renewal	branch	(m)

Figure 2.10 (cont’d)

Gi6

UFO

Figure 2.10B

Gi3

Gi6

Gi3

TSA

Gi6
SSA

Figure 2.11A
43

bc

ab

a
bc

Gi3

Gi6
UFO

Proximal

Middle

Distal

Proximal

Middle

Distal

def def

Proximal

fg

Middle

ef
fg efg

Distal

Distal

Proximal

g

Proximal

de

Middle

cd

Middle

ef

Distal

ef

ef de

Proximal

ab

Middle

1.8
1.6
1.4
1.2
1
0.8
0.6
0.4
0.2
0

Distal

length	per	renewal	branch	(m)

Figure 2.10- Average length per renewal branch, one season of growth after whole tree renewal
was imposed. Three-way interaction: permanent structure section by rootstock by training
system. TSA and UFO both grown on Gi3, Gi5, and Gi6. 9A-Bars are an average per section of 3
whole tree renewal trees grown in the CRAVO plus or minus standard error of the mean. 9BBars are an average per canopy section of 4 whole tree renewal trees grown in the VOEN plus or
minus standard error of the mean at Michigan State University’s Clarksville Research Center.
Lowercase letters separate means comparing the system by rootstock by section interaction,
determined using Tukey’s LSD, significance at p≤0.05.

Gi3

Gi6

Gi3

TSA

Gi6
SSA

Gi3

Proximal

Middle

Distal

Proximal

Middle

Distal

Proximal

Middle

Distal

Proximal

Middle

Distal

Proximal

Middle

Distal

Proximal

Middle

Distal

length	per	renewal	branch	(m)

Figure 2.11 (cont’d)
1.8
a
			a
1.6
	ab
	ab
abcd
1.4
abc
abcd
					bcde
abcde
bcdef
bcde
1.2
					bcdef bcdef
cdef
def ef
1 cdef
e
0.8
0.6
0.4
0.2
0

Gi6
UFO

Figure 2.11B

length	per	renewal	branch	(m)

Figure 2.11- Average length per renewal branch, one season of growth after whole tree renewal
was imposed. Three-way interaction: permanent structure section by rootstock by training
system. TSA, SSA and UFO both grown on Gi3 and Gi6.. Three-way interaction: permanent
structure section by rootstock by training system. Without Gi5 data. 10A-Bars are an average per
section of 3 whole tree renewal trees grown in the CRAVO plus or minus standard error of the
mean. 10B- Bars are an average per canopy section of 4 whole tree renewal trees grown in the
VOEN plus or minus standard error of the mean at Michigan State University’s Clarksville
Research Center. Lowercase letters separate means comparing the system by rootstock by
section interaction, determined using Tukey’s LSD, significance at p≤0.05.
1.6
1.4
1.2

a

1

a
b

0.8

c

0.6
0.4
0.2
0
KGB

TSA

SSA

UFO

Figure 2.12A

44

length	per	renewal	branch	(m)

Figure 2.12 (cont’d)
1.6
1.4

a
b

1.2

c

1

d

0.8
0.6
0.4
0.2
0
KGB

TSA

SSA

UFO

Figure 2.12B
Figure 2.12- Length per renewal branch on a whole tree basis averaged over Gi3 and Gi6. 11ABars are average of 24 whole tree renewal trees grown in the CRAVO plus or minus standard
error of the mean. 11B- Bars are average of 16 Whole tree renewal trees grown in the VOEN
plus or minus standard of the mean at Michigan State University’s Clarksville Research Center.
Lowercase letters separate means comparing the training systems, determined using Tukey’s
LSD, significance at p≤0.05.
length	per	renewal	branch	(m)

1.4
1.2
1

a

a

Gi5

Gi6

b

0.8
0.6
0.4
0.2
0
Gi3

Figure 2.13A

45

length	per	renewal	branch	(m)

Figure 2.13 (cont’d)
1.4
1.2

a

ab

b

1
0.8
0.6
0.4
0.2
0
Gi3

Gi5

Gi6

Figure 2.13B
Figure 2.13- Length per renewal branch on a whole tree basis averaged over KGB, TSA, and
UFO. 12A-Bars are average of 27 whole tree renewal trees grown in the CRAVO plus or minus
standard error of the mean. 12B- Bars are average of 18 Whole tree renewal trees grown in the
VOEN plus or minus standard of the mean at Michigan State University’s Clarksville Research
Center. Lowercase letters separate means comparing rootstocks, determined using Tukey’s LSD,
significance at p≤0.05.

number	of	syllpetic	branches

4.5
4
3.5
3
2.5
2

a

1.5

b

				b

1
0.5
0
Distal

Middle

Proximal

Figure 2.14- Number of sylleptic branches on each renewal parent branch, per permanent
structure section of TSA and UFO on Gi3, Gi5, and Gi6. Bars are average of 18 whole tree
renewal trees grown in the CRAVO plus or minus standard error of the mean at Michigan State
University’s Clarksville Research Center. Lowercase letters separate means comparing the
canopy sections, determined using Tukey’s LSD, significance at p≤0.05.

46

a					

Gi3

a

Gi5

Gi6

				abc
					abcd
				bcd

Gi3

Distal

Proximal

Distal

Proximal

Middle
Gi5

Proximal

					cd

					d
Middle

Distal

Proximal

Middle

			abcd
bcd
			cd
			cd

Middle

		abc

Distal

Proximal

Middle

Distal

Proximal

Middle

				abc			abc
ab
			abc 					abcd
					bcd

Distal

number	of	sylleptic	branches

5
4.5
4
3.5
3
2.5
2
1.5
1
0.5
0

Gi6

Gi3

b

b

Proximal

		b

Distal

b

Proximal

b

Middle

a

Middle

4.5
4
3.5
3
2.5
2
1.5
1
0.5
0

Distal

number	of	sylleptic	branches

TSA
UFO
Figure 2.15- Number of sylleptic branches on each renewal parent branch, per permanent
structure section of TSA and UFO on Gi3, Gi5, and Gi6. Three way interaction: training system
by rootstock by permanent structure section. Bars are an average per section of 12 whole tree
renewal trees grown in the VOEN plus or minus standard error of the mean at Michigan State
University’s Clarksville Research Center. Lowercase letters separate means comparing the
system by rootstock by section interaction, determined using Tukey’s LSD, significance at
p≤0.05.

Gi6

Figure 2.16- Number of sylleptic branches on each renewal parent branch, per permanent
structure section of Gi3 and Gi6 averaged over TSA, SSA, and UFO. Two-way interaction:
rootstock by permanent structure section. Bars are an average per section of 18 whole tree
renewal trees grown in the CRAVO plus or minus standard error of the mean at Michigan State
University’s Clarksville Research Center. Lowercase letters separate means comparing the
rootstock by section interaction, determined using Tukey’s LSD, significance at p≤0.05.

47

			a

b

b

	b

	b

	b

				b
b

TSA

SSA

Proximal

Middle

Distal

Proximal

Middle

Distal

Proximal

Middle

	b

Distal

number	of	sylleptic	branches

4.5
4
3.5
3
2.5
2
1.5
1
0.5
0

UFO

Figure 2.17- Number of sylleptic branches on each renewal parent branch, per permanent
structure section of TSA, SSA, and UFO averaged over Gi3 and Gi6. Two-way interaction:
training system by permanent structure section. Bars are an average per section of 12 whole tree
renewal trees grown in the VOEN plus or minus standard error of the mean at Michigan State
University’s Clarksville Research Center. Lowercase letters separate means comparing the twoway interaction, determined using Tukey’s LSD, significance at p≤0.05.

number	of	sylleptic	branches

3.5
3
2.5
2

a

a

1.5
1

b

b

TSA

SSA

0.5
0
KGB

UFO

Figure 2.18A

48

Figure 2.18 (cont’d)

number	of	sylleptic	branches

3.5
3

		a	

2.5
2

b

		b	

1.5

				b	

1
0.5
0
KGB

TSA

SSA

UFO

Figure 2.18B
Figure 2.18- Number of sylleptic branches on a whole tree basis by training system. Averaged
over Gi3 and Gi6. 17A- Bars are and average of 24 whole tree renewal trees grown in the
CRAVO plus or minus standard error of the mean. 17B- Bars are and average of 16 whole tree
renewal trees grown in the VOEN plus or minus standard error of the mean at Michigan State
University’s Clarksville Research Center. Lowercase letters separate means comparing training
systems, determined using Tukey’s LSD, significance at p≤0.05.

Figure 2.19- Trunk cross sectional area in 2016 of all location by training system by rootstock
combinations in the NC-140 sweet cherry rootstock trial at Michigan State University’s
Clarksville Research Center with ‘Benton’ as the scion. Bars are an average of 6 trees for each
KGB, TSA, and UFO and 12 trees for SSA on each rootstock under the CRAVO; under the
VOEN, 4 trees for KGB, TSA, and UFO and 8 trees for SSA on each rootstock.

49

Number of sylleptic branches
Length	of	primary	renewal	branch	(m)	
	
Figure 2.20- Density scatter plot of length per primary renewal branch versus number of
sylleptic branches per renewal branch of trees that have been whole tree renewed. All data is
included, being comprised of four different training systems: KGB, TSA, SSA, and UFO; three
different rootstocks: Gi3, Gi5, and Gi6 at Michigan State University’s Clarksville Research
Center. Darker hexagon shade indicates higher number of observations. A total of 1362 primary
branches are plotted.

50

Number of sylleptic branches
Percentile	by	length	of	renewal	branch		
	
Increasing	renewal	branch	length		
	
Figure 2.21- Violin plot of distribution of number of sylleptic branches per primary renewal
branch of trees that have been whole tree renewed. All data is included, being comprised of four
different training systems: KGB, TSA, SSA, and UFO; three different rootstocks: Gi3, Gi5, and
Gi6 at Michigan State University’s Clarksville Research Center. Data is broken up into
percentile by renewal branch length. 1-25th and 26-50th is the distribution of sylleptic branches on
340 primary renewal branches. 51-75th and 76-100th percentile is the distribution of sylleptic
branches on 341 primary renewal branches. 1-25th percentile represents the shortest 25% of
branches, branch length increases with percentile.

51

APPENDIX

52

Table 2.1- Analysis of variance (ANOVA) for number of renewal shoots following whole tree
renewal pruning of eight-year-old ‘Benton’ sweet cherry trees, grown in the VOEN at MSU
Clarksville Research Center. SSA data left out.
Type 3 Tests of Fixed Effects
Numerator Degrees
Effect
of Freedom
System
2
rootstock
2
system* rootstock 4
Type 3 Tests of Fixed Effects
Numerator Degrees
Effect
of Freedom
system
2

Denominator
Degrees of Freedom
8
8
8

F Value
9.43
1.84
0.52

Denominator
Degrees of Freedom
14

F Value Pr > F
9.44
0.0025

Pr > F
0.0079
0.2207
0.7252

Table 2.2- Analysis of variance (ANOVA) for number of sylleptic branches following whole
tree renewal pruning of eight-year-old ‘Benton’ sweet cherry trees, grown in the CRAVO at
MSU Clarksville Research Center. SSA data eliminated to compare rootstocks.
Type 3 Tests of Fixed Effects
Numerator Degrees
Effect
of Freedom
system
2
rootstock
2
system* rootstock 4
Type 3 Tests of Fixed Effects
Effect
Numerator Degrees
of Freedom
system
2

Denominator
Degrees of Freedom
650
650
650
Denominator
Degrees of Freedom
656

53

F Value
6.65
2.34
0.71

Pr > F
0.0014
0.097
0.586

F Value Pr > F
6.07

0.0024

Table 2.3- Analysis of variance (ANOVA) for number of sylleptic branches following whole
tree renewal pruning of eight-year-old ‘Benton’ sweet cherry trees, grown in the VOEN at MSU
Clarksville Research Center. SSA data eliminated to compare rootstocks.
Type 3 Tests of Fixed Effects
Numerator Degrees
Effect
of Freedom
system
2
rootstock
2
system* rootstock 4
Type 3 Tests of Fixed Effects
Numerator Degrees
Effect
of Freedom
system
2

Denominator
Degrees of Freedom
410
410
410

F Value
3.41
0.58
1.42

Denominator
Degrees of Freedom
416

F Value Pr > F
4.72
0.0094

54

Pr > F
0.0338
0.5621
0.2278

LITERATURE CITED

55

LITERATURE CITED

Ayala, M. and Lang, G.A. (2017). Morphology, cropping physiology, and canopy training, p.
269-304. In: Quero-Garcia, J., Iezzoni, A., Pulawska, J., & Lang, G. A. (Eds.). Cherries: Botany,
Production and Uses. CABI.
Atkinson, C., and Else, M. (2001). Understanding how rootstocks dwarf fruit trees. Compact
Fruit Tree, 34(2), 46-49.
Büsgen, M. and Münch, E. (1929). The Structure and Life of Forest Trees. John Wiley and Sons
Inc., New York, NY, USA. 436 pp.
Burrows, G. E. (2008). Syncarpia and Tristaniopsis (Myrtaceae) possess specialised fire-resistant
epicormic structures. Australian Journal of Botany, 56(3), 254-264.
Cline, M. G., and Dong-Il, K. I. M. (2002). A preliminary investigation of the role of auxin and
cytokinin in sylleptic branching of three hybrid poplar clones exhibiting contrasting degrees of
sylleptic branching. Annals of Botany, 90(3), 417-421.
Cooper-Ellis, S., Foster, D. R., Carlton, G., and Lezberg, A. (1999). Forest response to
catastrophic wind: results from an experimental hurricane. Ecology, 80(8), 2683-2696.
Eberhard, S., Finazzi, G., & Wollman, F. A. (2008). The dynamics of photosynthesis. Annual
review of genetics, 42, 463-515.
Fontaine, F., Colin, F., Jarret, P., and Druelle, J. L. (2001). Evolution of the epicormic potential
on 17-year-old Quercus petraea trees: first results. Annals of Forest Science, 58(5), 583-592.
Fontaine, F., Kiefer, E., Clément, C., Burrus, M., and Druelle, J. L. (1999). Ontogeny of the
proventitious epicormic buds in Quercus petraea. Trees-Structure and Function, 14(2), 83-90.
Gordon, D., Rosati, A., Damiano, C., and DeJong, T. M. (2006). Seasonal effects of light
exposure, temperature, trunk growth and plant carbohydrate status on the initiation and growth of
epicormic shoots in Prunus persica. The Journal of Horticultural Science and
Biotechnology, 81(3), 421-428.
Lang, G. A. (2000). Precocious, dwarfing, and productive—how will new cherry rootstocks
impact the sweet cherry industry? HortTechnology, 10(4), 719-725.
Leyser, O. (2003). Regulation of shoot branching by auxin. Trends in Plant Science, 8(11), 541545.
Ljung, K., Bhalerao, R. P., and Sandberg, G. (2001). Sites and homeostatic control of auxin
biosynthesis in Arabidopsis during vegetative growth. The Plant Journal, 28(4), 465-474.

56

Long, L. E., Lang, G. A., Whiting, M. D., and Musacchi, S. (2015). Cherry training systems.
Pacific Northwest Extension. Oregon State University. University of Idaho. Washington State
University.
Measham, P. F., Bound, S. A., Gracie, A. J., and Wilson, S. J. (2009). Incidence and type of
cracking in sweet cherry (Prunus avium L.) are affected by genotype and season. Crop and
Pasture Science, 60(10), 1002-1008.
Nicolini, E., Chanson, B., and Bonne, F. (2001). Stem growth and epicormic branch formation in
understorey beech trees (Fagus sylvatica L.). Annals of Botany, 87(6), 737-750.
O'Hara, K. L., and Berrill, J. P. (2009). Epicormic sprout development in pruned coast redwood:
pruning severity, genotype, and sprouting characteristics. Annals of Forest Science, 66(4), 1-9.
O'hara, K. L., and Valappil, N. I. (2000). Epicormic sprouting of pruned western larch. Canadian
Journal of Forest Research, 30(2), 324-328.
Piene, H., and Eveleigh, E. S. (1996). Spruce budworm defoliation in young balsam fir: the
‘green’tree phenomenon. The Canadian Entomologist, 128(06), 1101-1107.
Perry, R., Lang, G., Andersen, R., Anderson, L., Azarenko, A., Facteau, T., ... and Rom, C.
(1997, July). Performance of the NC-140 cherry rootstock trials in North America. In III
International Cherry Symposium 468 (pp. 291-296).
R Core Team (2017). R: A language and environment for statistical computing. R
Foundation for Statistical Computing, Vienna, Austria. URL
https://www.R-project.org/.
Sachs, T., and Thimann, K. V. (1967). The role of auxins and cytokinins in the release of buds
from dominance. American Journal of Botany, 136-144.
Signorelli, S., Agudelo-Romero, P., Foyer, C. H., & Considine, M. J. (2017). Roles for Light,
Energy and Oxygen in the Fate of Quiescent Axillary Buds. Plant physiology, pp-01479.
Soil Survey Staff, Natural Resources Conservation Service, United States Department of
Agriculture. Web Soil Survey. Available online at the following
link: https://websoilsurvey.sc.egov.usda.gov/. Accessed [December/3/2017].
Spath H. 1912. Der Johannistriebe. Berlin: Parey.
Stone, E. L., and Stone, M. H. (1943). "Dormant" versus "Adventitious" buds. Science,
98(2533), 62-62.
Sundberg, B., and Uggla, C. (1998). Origin and dynamics of indoleacetic acid under polar
transport in Pinus sylvestris. Physiologia Plantarum, 104(1), 22-29.

57

Takemiya, A., Inoue, S., Doi, M., Kinoshita, T., & Shimazaki, K. (2005). Phototropins promote
plant growth in response to blue light in low light environments(W). Plant Cell, 17(4), 1120-7.
Tworkoski, T., Miller, S., and Scorza, R. (2006). Relationship of pruning and growth
morphology with hormone ratios in shoots of pillar and standard peach trees. Journal of Plant
Growth Regulation, 25(2), 145-155.
Wignall, T. A., Browning, G., and Mackenzie, K. A. D. (1987). The physiology of epicormic
bud emergence in Pedunculate Oak (Quercus robur L.) Responses to partial notch girdling in
thinned and unthinned stands. Forestry: An International Journal of Forest Research, 60(1), 4556.

58

CHAPTER 3: MAPPING EPICORMIC VEGETATIVE MERISTEMS IN SWEET
CHERRY USING X-RAY COMPUTER TOMOGRAPHY
Introduction
Axillary vegetative meristems (buds) that remain do not form a branch, remaining
dormant, eventually become engulfed by radial growth of the stem; these are considered
preventitious epicormic buds, also termed latent buds (Büsgen and Münch, 1929; Stone and
Stone, 1943). Preventitious epicormics buds follow the phyllotaxy of the tree; sweet cherry
generally has a spiral phyllotaxy with 5 nodes per complete phyllotaxic revolution, ~every 72˚
(Lang et al., 2004), or more accurately, 5 nodes per two revolutions, every 144˚ (G.A. Lang,
personal communciation). Adventitious epicormic buds form independently, typically after a
wounding event (Fink, 1983, 1999; Kauppi et al., 1987). Epicormic buds constitute the
“epicormic potential” of a woody plant to develop epicormic branches and therefore are
important for orchardists studying pruning and forest managers evaluating timber quality due to
knots produced by epicormic branches or forest regeneration following a fire or insect
defoliation.
Preventitious epicormic buds maintain a vascular connection with the pith of the parent
branch from which they formed, an epicormic trace (Fontaine et al., 1998). An epicormic trace
consists of densely packed parenchyma cells, measuring 2-5 mm thick. Cells in the trace become
more lignified towards the pith (Burrows, 2000). Epicormic traces occur perpendicular to the
pith (Büsgen and Münch, 1929). Burrows (2000) found simple epicormic traces, more
commonly found on small stems, measured 2.0 mm high; complex traces measured 3-5 mm high
and consisted of two smaller strands. Traces elongate with annual growth, maintaining the
epicormic bud’s persistence beneath the bark (Stone and Stone, 1943).

59

Preventitious epicormic buds can be separated into large and small buds. Buds that
measure less than 2 mm long are considered small, and are usually found at the base of a shoot.
Large preventitious epicormic buds, located along the shoot (Fontaine et al., 1998), are 3-4 mm
long, and are more developmentally complex than small preventitious epicormic buds.
Meristematic areas 100 µm long and 30 µm wide are present in the upper third of large
preventitious epicormic buds. Secondary bud primordia characterize the lower two thirds (fig.
3.1A). Small buds are composed of only a terminal meristem surrounded by scales, and do not
have secondary bud primordia (Fontaine et al., 1998) (fig. 3.1B). Large preventitious epicormic
buds that do not sprout are more likely to abscise than small epicormic buds (Harmer, 1991;
Fontaine et al., 2001); small buds persist longer than large buds (Gruber, 1994). Large buds are
more likely to elongate into new shoots than are small buds (Braham and Kellison, 1987).
Colin et al. (2010) tracked epicormic traces in sessile oak (Quercus petraea) trunks using
X-ray computer tomography (CT). Both simple epicormic buds and bud clusters were found in
the scanned sections. Epicormic branches developed horizontally in the trunk, while sequential
branches arose at an oblique angle to the pith (Colin et al., 2010).
This preliminary study aimed, first and foremost, to determine if epicormic traces (and
therefore epicormic buds) in sweet cherry can be visualized using x-ray computer tomography.
Likewise, can x-ray computer tomography distinguish between simple and complex epicormic
traces and be used to examine the development of epicormic branches. Mapping epicormic buds
has the potential to improve whole tree renewal by determining where epicormic branches are
most likely to arise.

60

Materials and Methods
Plant Material
A 1.14 m long section of 8-year-old trunk of ‘NY 119’ on Gisela 12 sweet cherry that had
been subjected to whole tree renewal was used for the scan. Eight sequential branches of the tree
had been removed at bloom 2016, and the trunk section for scanning was collected in August
2016 after the top of the tree has been removed at 1.5 m from the ground. The CT scan was done
in December 2016 after drying the trunk section. The trunk circumference measured 32.5 cm at
the base, 27.6 cm at the middle, and 24.7 cm at the top of the section.
X-Ray Computer Tomography Scan
X-ray scanning was performed at Michigan State University’s Department of Radiology
(East Lansing, MI) using a whole body magnetic resonance scanner (GE Signa HDX 3.0T,
Chicago, Il). The CT scan was done in December 2016 after drying the trunk section. The high
density of parenchyma cells within the epicormic trace means that this trace shows up brighter
the rest of the trunk. Contrasts in the scan is influenced by moisture content of the trunk, with
starkness being greater in air-dried than fresh wood (Freyburger et al., 2009). Greater contrasts
allow for epicormic traces to be more easily distinguished from the rest of the trunk. Scanning
image slice thickness was set at ~0.625 mm per slice, the entire scan took about a minute.
3-Dimensional Reconstruction
The 3-D trunk image was reconstructed using 3D Slicer software version 4.8 (open
source, Cambridge, MA; http://www.slicer.org; Federov et al., 2012). The paint brush in
segment editor was used to first label the end of each epicormic trace as either a complex or
simple trace. Complex epicormic traces were those that had more than one conjoining epicormic
trace (fig. 3.1A). Simple epicormic traces were those with only a single trace (fig. 3.1B).

61

Complex or simple epicormic traces were marked with a large or small sphere, respectively. A
second segment was then created using the threshold tool in segment editor that selected the
entire trunk.
Results
1,830 images were produced from the scan. Epicormic traces were visible throughout the
trunk section; these traces showed up brighter than the surrounding trunk. There were 104 total
epicormic traces; 22 were complex traces (fig. 3.2A) and 82 were simple (fig. 3.2B), what
appeared to be two conjoining traces. These traces were spread equally throughout the trunk and
appear to follow the phyllotaxy of the tree. Many of the complex traces were flanked with a
simple epicormic trace on either side; these three traces originate at the pith: the simple traces
were oriented outward at an angle forming a V, with the complex trace splitting the middle (fig.
3.2C). These V-shaped traces occurred at sites with burls on the trunk.
Between whole tree renewal pruning at bloom and collection of the trunk section in
August, six epicormic branches had developed in the scanned section of trunk during the spring
and summer. Four of these epicormic branches (1 and 2, 4 and 5) occurred at the same spot (fig.
3.3A and 3.3C. In cross section, five of the six epicormic branches (epicormic branches 1-5) had
clear epicormic traces back to the central pith, which had left the annual growth rings
undisturbed (figs. 3.3A-C). However, the location of epicormic branch 6 showed a break in the
annual growth ring and no clear epicormic trace to the central pith in cross section (fig. 3.3D).
Epicormic branches 1-5 were from complex epicormic traces, while epicormics branch 6 appears
to be from either a simple epicormic trace at the base of a sequential shoot or an adventitious
epicormic bud. Epicormic branches 1 and 2; 4 and 5 were separate branches arising from the
same position (fig 3.3A and 3.3C).

62

Discussion
Epicormic traces are densely packed parenchyma cells (Fontaine et al., 1998); since
higher density regions of CT scans show up brighter (Freyburger et al., 2009), epicormic traces
are easily visualized using CT scanning. Traces were located equally throughout the trunk, and
there were only a few instances where the trace did not extend to the outside, suggesting that
environmental factors did not affect persistence of preventitious epicormic buds during the eightyear life of this tree. It must be noted that this section constitutes the bottom half of the tree
trunk, so it is possible that the concentration of epicormic buds may have differed in the upper
half of the trunk, if initial growth rate or increased light exposure over time might influence
epicormic bud density or persistence.
Burl wood is an outgrowth of the trunk, typically caused by insect damage, fungi or
another stress that creates a distortion in the wood grain. A V-shaped trace—two simple traces
flanking a complex trace - was present at burls in cross-section; the annual growth rings appear
wavy around these traces (fig. 3.2C). Such a V trace was reported in a scan done on sessile oak
(Colin et al., 2010). In our study, four of the six epicormic branches (1, 2, 4, and 5) produced on
the scanned trunk section following whole tree renewal arose from V traces (figs. 3.3A and
3.3C). Each of these locations produced two epicormic branches; epicormic buds present in burls
may, therefore, be an important source of epicormic branches in sweet cherry.
Epicormic branches 1-5 had a connection back to the central pith and developed outside
of the annual growth ring (fig. 3.3A-C). Epicormic branch 6 was an anomaly in this regard. The
cross section at this epicormic branch showed a break through all of the growth rings (fig. 3.3D),
in the same way that a sequential branch develops (fig. 3.2D). A possible explanation for this is
that the epicormic branch developed at the base of a sequential branch. However, there is no

63

sequential branch visible on the outside of the trunk, but there are a few signs that the sequential
branch could have become engulfed by the epicormic branch (fig. 3.3D). There appears to be two
knots in cross section that this epicormic branch has engulfed (fig. 3.3D), the one on the right is
an epicormic knot (a) developing at the base of the original sequential branch, while (b) has
annual growth rings and is therefore a sequential knot. Secondly, there is reaction wood on the
underside of the epicormic branch, which is evidence of a sequential branch (fig. 3.4).
Future Directions
These results indicate that preventitious epicormic buds can be mapped throughout a
mature section of tree trunk. While preventitious epicormic buds don’t constitute the entire
epicormic potential of a tree, mapping allows estimation of the ratio of adventitious to
preventitious meristems in epicormic branch initiation and consequently the degree to which
epicormic branch potential can be quantified. In the context of whole tree renewal, this has the
potential to provide new insights into how removal of active vegetative meristems affects the
potential sources of epicormic branching. Future studies may be valuable to test whether cultural
treatments, such as plant growth regulators or synthetic hormones, could directly activate
epicormic buds to initiate renewal shoots at specific sites along the trunk.
Preventitious epicormic buds arise in a predictable pattern, following the phyllotaxy of
the tree. Sweet cherry has a 2/5 phyllotaxy, meaning that a node will arise directly above another
one after two spirals of five successive nodes around the branch. Mapping preventitious
epicormic buds can provide data on how growth affects the distance and angle between
preventitious epicormic buds; allowing for a model to be developed that can predict where
preventitious epicormic buds persist. Ultimately this model could be applied to a field setting,

64

where a grower could utilize a sequential branch as a landmark to base where a preventitious
epicormic bud persists.

65

Figure 3.1A

Figure 3.1B
Figure 3.1- Scanning Electronic microscope images of a large preventitious epicormic bud (1A)
with secondary bud primordia surrounding the bud in the center; small preventitious epicormic
buds (1B) without secondary bud primordia (Fontaine et al., 1999).

66

Figure 3.2A

Figure 3.2B

Figure 3.2C

Figure 3.2D

Figure 3.2- Images in cross-section from X-ray CT scan of a 8-year-old section of sweet cherry
of the variety NY 119 grafted onto Gisela 12 rootstock trunk: 2A) complex epicormic trace, 2B)
simple epicormic trace, 2C) V trace with a central complex trace, flanked by a simple trace on
either side, 2D) sequential branch.

67

Figure 3.3A

Figure 3.3B

a
b

Figure 3.3D

Figure 3.3C

Figure 3.3 - Epicormic branches in cross-section: 3A) epicormic branch 1 and 2; 3B) epicormic
branch 3; 3C) epicormic branch 4 and 5; 3D) epicormic branch 6.

68

Figure 3.4- Epicormic branch number 6 with reaction wood on the underside of the branch.

69

LITERATURE CITED

70

LITERATURE CITED

Braham, R. R., and Kellison, R. C. (1987). Suppressed buds in yellow-poplar. Journal of the
Elisha Mitchell Scientific Society, 47-55.
Büsgen, M. and Münch, E. (1929). The Structure and Life of Forest Trees. John Wiley and Sons
Inc., New York, NY, USA. 436 pp.
Burrows, G. E. (1989). Developmental anatomy of axillary meristems of Araucaria
cunninghamii released from apical dominance following shoot apex decapitation in vitro and in
vivo. Botanical Gazette, 369-377.
Colin, F., Mothe, F., Freyburger, C., Morisset, J. B., Leban, J. M., and Fontaine, F. (2010).
Tracking rameal traces in sessile oak trunks with X-ray computer tomography: biological bases,
preliminary results and perspectives. Trees, 24(5), 953-967.
Fedorov A., Beichel R., Kalpathy-Cramer J., Finet J., Fillion-Robin J-C., Pujol S., Bauer C.,
Jennings D., Fennessy F., Sonka M., Buatti J., Aylward S.R., Miller J.V., Pieper S., Kikinis
R. 3D Slicer as an image computing platform for the quantitative imaging network. Magnetic
Resonance Imaging. 2012 Nov; 30(9):1323-41. PMID: 22770690.
Fink, S. (1983). The occurrence of adventitious and preventitious buds within the bark of some
temperate and tropical trees. American Journal of Botany, 532-542.
Fink, S. (1999). Pathological and regenerative plant anatomy. Gebruder Borntraeger
Verlagsbuchhandlung.
Fontaine, F., Druelle, J. L., Clément, C., Burrus, M., and Audran, J. C. (1998). Ontogeny of
proventitious epicormic buds in Quercus petraea. I. In the 5 years following initiation. Trees,
13(1), 54-62.
Fontaine, F., Colin, F., Jarret, P., and Druelle, J. L. (2001). Evolution of the epicormic potential
on 17-year-old Quercus petraea trees: first results. Annals of Forest Science, 58(5), 583-592.
Freyburger, C., Longuetaud, F., Mothe, F., Constant, T., and Leban, J. M. (2009). Measuring
wood density by means of X-ray computer tomography. Annals of Forest Science, 66(8), 804.
Gruber, F. (1994). Morphology of coniferous trees: possible effects of soil acidification on the
morphology of Norway spruce and Silver fir. Effects of acid rain on forest processes. John Wiley
and Sons Inc., New York, NY, USA. 265-324.
Harmer, R. (1991). The effect of bud position on branch growth and bud abscission in Quercus
petraea (Matt.) Liebl. Annals of botany, 67(5), 463-468.

71

Kauppi, A., Rinne, P., and Ferm, A. (1987). Initiation, structure and sprouting of dormant basal
buds in Betula pubescens. Flora, 179(1), 55-83.
Lang, G. A., Olmstead, J. W., and Whiting, M. D. (2004). Sweet cherry fruit distribution and leaf
populations: modeling canopy dynamics and management strategies. Acta Horticulturae., 591600.
Stone, E. L., and Stone, M. H. (1943). "Dormant" versus "Adventitious" buds. Science,
98(2533), 62-62.

72