 

4;»:

4C

.
. 5.11.131! I
.204.» hr. ‘. ”in
I I: o v.
uggr

. oLv‘ an"
t 0 in) I. i
. k 1'! vxl. lyblnlllu
.t I .nlufv ‘i‘tlr‘lo‘l-
.ol‘lotgli‘lll!
v0 l,‘!l?\llut.0 [-37
. IKI'.J.

 

Hint . u... .. all“: - .‘. . $5M llq..-!1, .
.1v .1. ,... : ... . , millahéﬁf ...1.~4!-.K.l.3ou.l

o. . 1.;
..,..\ . 51.41:

Mtgr at... .. i». an mum

 

THE-3‘8

Q

MIC CGHI STA TEUN

WWII"! HI “"1“”IIIHIIHHHIHI

31293 01712 8996

it

 

 

 

 

 

 

 

 

 

                         

This is to certify that the

dissertation entitled

EFFECTS OF FRUIT, SEED AND SHOOT DEVELOPMENT
ON FLOWER INDUCTION IN APPLE

presented by

JOHN CALVIN NEILSEN

has been accepted towards fulﬁllment
of the requirements for

PhoDo degree in Horticulture

 

 

 

Major professor

Date May 8, 1998

 

MSU is an Afﬁrmative Action/Equal Opportunity Institution 0-12771

 

LIBRARY

Michigan State
' UnlversIty

 

 

 

PLACE IN RETURN BOX
to remove this checkout from your record.
TO AVOID FINES return on or before date due.

 

DATE DUE . DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1/98 animus-m4

EFFECTS OF FRUIT, SEED AND SHOOT DEVELOPMENT
ON FLOWER INDUCTION IN APPLE

BY

JOHN CALVIN NEILSEN

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

DOCTOR OF PHILOSOPHY

Department of Horticulture

1998

ABSTRACT

EFFECTS OF FRUIT, SEED AND SHOOT DEVELOPMENT
ON FLOWER INDUCTION IN APPLE

By

John Calvin Neilsen

Studies were conducted to determine the effects of seeds, shoot length, crop
density, and gibberellin application on ﬂower induction in apple. Spurs of ‘Spencer
Seedless’ were defruited at intervals during fruit development in 1988-1991. Fruit mass
and seed number were recorded for each spur. Flower induction was inhibited in spurs
bearing fruits containing more than ﬁve seeds, the effect increasing as time of deﬁ'uiting
was delayed. When fruits were left until harvest in 1994 and 1996, ﬂowering decreased
linearly with an increase in seed number per spur, but response varied with bourse shoot
length. All shoots less than 2 mm long failed to ﬂower. In spurs bearing seeded fruits,
ﬂowering increased with shoot length, and most shoots > 10 mm long ﬂowered.
Neither ﬁ'uit mass per spur nor crop density affected ﬂowering in the absence of seeds.
In spurs bearing fruits containing ﬁve or more seeds, high crop density appeared to
inhibit ﬂowering, but fruit mass per spur had little effect. In ‘Paulared’ ﬂowering was

not inhibited if trees were entirely deﬁ'uited within 60 days aﬁer full bloom, but

declined rapidly as fruit removal was delayed further. In 1996, fruits were removed 24
days after full bloom, leaving either 1 or 2 fruits per bearing terminal; half of the trees
were girdled to stimulate ﬂowering. Flowering in 1997 was variable, with more
ﬂowering than expected for trees bearing moderate to heavy crOps. No effect of bourse
shoot length on ﬂowering was detected, and non-bearing shoots ﬂowered only slightly
more than did bourse shoots on terminals that bore one or two fruits. In the absence of
girdling, percentage ﬂowering decreased as crop density increased. To determine if
substitution of gibberellins for seeds would inhibit ﬂowering, fruits of several seeded
and seedless cultivars were cut transversely 30 days after full bloom, and the seeds
removed and replaced with lanolin pastes containing gibberellins A3 or A4”. Percentage
ﬂowering in the controls (fruits intact) was relatively high, and neither seed removal nor
gibberellin treatment had appreciable effects, regardless of cultivar. Sprays of GA,”
were applied to whole trees of ‘Gala’ apple to determine if ﬂowering could be inhibited
in lateral buds. Treatments applied 3 July inhibited ﬂowering more than did later
applications, but the variability was high, resulting in nonsigniﬁcant effects of
concentration. These experiments demonstrate that, in ‘Spencer Seedless’, seeds rather
than fruits are responsible for inhibition of ﬂowering and that the effects of seeds
declines as bourse shoot length increases. These relationships could not be conﬁrmed in
the seeded cultivar ‘Paulared’ as response was too variable. Substitution of lanolin
pastes containing gibberellins for seeds did not inhibit ﬂowering, suggesting that the
role that seeds play in biennial bearing interacts with other growth factors, including

shoot growth and cropload.

To my parents, Calvin E. Neilsen and Sarah C. Neilsen,
champions of agriculture, who planted the seed,
and my eternal companion Kelly A. Neilsen,

whose faith in me nurtured it to fruition.

iv

ACKNOWLEDGMENTS

I’m grateﬁrl to Dr. Frank G. Dennis Jr. for serving as my major professor and
chair of my advisory committee. He persevered beyond my expectations. I also
appreciate the service of Drs. Martin J. Bukovac, James A. Flore, Kenneth L. Poff and
Jan A. D. Zeevaart as members of my advisory committee for their advice and
encouragement.

I gratefully acknowledge the assistance of Drs. Martin J. Bukovac and Jerome
Hull Jr. with experiments reported in Section 4 and Abbott Laboratories for the gift of
Provideo and GAM.

I gratefully acknowledge the assistance of Dr. Jun Ban, Mario Mandujano, Dr.
John Everard, Adam J. Neilsen and many other colleagues for assistance in data
collection and discussion of the results. I also acknowledge my friends and family who
provided ﬁnancial and moral support to make it possible for me to pursue this goal.

Finally, I express deep love and appreciation to my dear wife Kelly A. Neilsen
and our children Adam, Maren, Anna, Camille, and Sarah who have endured many

deprivations during the many years I’ve pursued education.

TABLE OF CONTENTS

LIST OF TABLES ..................................................... x
LIST OF FIGURES ................................................... xiii
LITERATURE REVIEW ............................................... 1
Economic importance of ﬂower formation ............................ 1
Growth and ﬂowering habit ........................................ 2

Time of induction ................................................ 4
Factors inﬂuencing ﬂower induction ................................. 5
Genotype and juvenility ..................................... 5
Environmental and cultural factors ............................ 5

Shoot length and leaf area ............................. 6

Leaf/fruit ratio ............................................ 8

Seeds. ................................................... 9

Possible reasons for effects of seeds ................................ 11
Competition for carbohydrate ............................... 11

vi

Competition for water ..................................... 12

Competition for “ﬂorigen” .................................. 12
Seed-produced hormones ................................... 12
Summary ..................................................... 15
Literature Cited ................................................ 15

SECTION 1. EFFECTS OF SEED NUMBER, FRUIT REMOVAL, BOURSE

SHOOT LENGTH AND CROP DENSITY ON F LOWERIN G IN

‘SPENCER SEEDLESS’ APPLE ............................. 21
Abstract ...................................................... 21
Introduction ................................................... 22
Materials and Methods ........................................... 23

Plant material ............................................ 23

Pollination and fruit removal ................................ 23

Effects of ﬁ'uit removal, seed number, bourse shoot length and fruit mass

on ﬂower induction ................................. 24
Effects of seed number, crop density and bourse shoot length ...... 24 .
Results ...................... ' ................................. 25

Effects of fruit removal, seeds, bourse shoot length and ﬁ'uit mass on

ﬂower induction .................................... 25
Effects of seed number, crop density and bourse shoot length ...... 28
Discussion .................................................... 37

vii

Literature Cited ................................................ 42

SECTION 2. EFFECTS OF FRUIT REMOVAL, CROP DENSITY, AND TRUNK

GIRDLING ON FLOWER FORMATION IN 'PAULARED' APPLE:

GENERAL VS. LOCAL EFFECTS OF FRUITS. ............... 45
Abstract ...................................................... 45
Introduction ................................................... 47
Materials and Methods ........................................... 48

Expt. 1. Effects of time of ﬁrm removal on ﬂower initiation ....... 48

Expt. 2. Effects of crop density, fruit number per spur, and girdling on

ﬂower initiation .................................... 48
Results ....................................................... 50
Expt. 1. Effects of time of fruit removal on ﬂower initiation ....... 50

Expt. 2. Effects of crop density, fruit number per spur, and girdling on

ﬂower initiation .................................... 50
Discussion .................................................... 57
Literature Cited ................................................ 61

SECTION 3. EFFECTS OF DEFRUITING, SEED REMOVAL AND

REPLACEMENT OF SEEDS WITH GIBBERELLINS ON FLOWER

FORMATION IN APPLE. ................................. 63

Abstract ...................................................... 63

viii

Introduction ................................................... 64

Materials and Methods ........................................... 65
Expt. 1. Seeded cultivars ................................... 65
Expt. 2. Seedless cultivars ................................. 66
Results ....................................................... 66
Expt. 1. Seeded cultivars ................................... 66
Expt. 2. Seedless cultivars .................................. 69
Discussion .................................................... 69
Literature Cited ................................................ 74

SECTION 4. EFFECT OF GIBBERELLIN SPRAYS ON FLOWER FORMATION

IN LATERAL BUDS OF ‘GALA’ APPLE. .................... 76

Abstract ...................................................... 76
Introduction ................................................... 77
Materials and Methods ........................................... 78
Results ....................................................... 79
Discussion .................................................... 83
Literature Cited ................................................ 91

LIST OF REFERENCES .............................................. 100

ix

LIST OF TABLES

SECTION 2 .......................................................... 45

Table 1. Effect of time of deﬁ'uiting entire ‘Paulared’ apple trees in 1991 and 1992 on
mean shoot length ( 3: standard error) in the same growing season and number of
lateral ﬂowers per shoot the following year. .......................... 52

Table 2. Main effects of girdling, crop density and number of fruits per spur in 1996 on
ﬂowering of ‘Paulared’ apple in 1997. Percentages calculated from 21 to 46
bourse shoots per tree for each category. Three replicate trees were used for
each treatment in a split-plot design. ................................ 53

Table 3. Effect of medium (M) and high (H) crop density on ﬂower induction of non-
bearing shoots and spurs bearing 1 or 2 fruits on individual, non-girdled

‘Paulared’ apple trees. ........................................... 57

SECTION 3 .......................................................... 63
Table 1. Harvest data for ﬁve ‘Golden Delicious’ apple trees used in Expt. 1.
Clarksville Horticulture Experiment Station, 1996 - 1997. ............... 67

Table 2. Effects of deﬁ'uiting, ﬁ'uit cutting, seed removal, and replacement of seeds

with lanolin pastes on ﬂowering of ‘Golden Delicious’ apple. Clarksville
Horticulture Experiment Station, 1996 - 1997. Treatments were applied to 5
trees on 26 June 1996. ........................................... 67
Table 3. Effects of defruiting, ﬁ'uit cutting, seed removal, and replacement of seeds
with lanolin pastes on ﬂowering of ‘Jonagold’ apple. Horticulture Teaching and
Research Center, 1996 - 1997. Treatments were applied to 6 trees on 29 June
1996 .......................................................... 68
Table 4. Effects of defruiting and fruit cutting on ﬂowering of ‘Delicious’ apple.
Clarksville Horticulture Experiment Station, 1996 - 1997. Treatments were
applied to 7 trees on 30 June 1996. ................................. 69
Table 5. Effects of seed number and of applications of lanolin pastes to locules of cut
fruits of ‘Spencer Seedless’ at the Horticulture Teaching and Research Center,

1996 - 1997. Treatments were applied to fruits on one tree on 29 June 1996.

Table 6. Effects of deﬁ'uiting and applications of lanolin pastes to locules of cut fruits
of ‘Ohio No. 3’ apple at the Horticulture Teaching and Research Center, 1996 -

1997. Treatments were applied to fruits on 2 trees on 25 June 1996. ....... 71

SECTION 4 .......................................................... 76
Table 1. Effect of GA,” sprays applied 3 July 1996 to trees with varying crop density
on bloom rating and percentage of lateral buds ﬂowering in 1997 in ‘Gala’ apple

at Clarksville Horticulture Experiment Station, Clarksville, Mich. All values are

xi

means for 5 trees. ............................................... 80
Table 2. Effect of timing and concentration of ProVide" treatments in 1996 on crop
density and bloom rating in 1997 in ‘Gala’ apple at Clarksville Horticulture
Experiment Station, Clarksville, Michigan. Five trees per treatment. ...... 81
Table 3. Effect of timing and concentration of ProVideQ treatments in 1996 on
ﬂowering of terminal and of lateral buds in 1997 in ‘Gala’ apple at Clarksville
Horticulture Experiment Station, Clarksville, Michigan. Analysis of variance

based on transformed (arcsin square-root) data. ....................... 82

xii

LIST OF FIGURES

SECTION 1 .......................................................... 21

Figure 1. Effects of seeds and time of ﬁ'uit removal the previous year on ﬂowering of
terminal buds of ‘ Spencer Seedless’ apple bourse shoots. Each point represents
ﬁom 6 to 45 bourse shoots. Year indicates time of fruit removal. ......... 26

Figure 2. Effects of shoot length and seed number the previous year on ﬂowering of
terminal buds of bourse shoots of ‘Spencer Seedless’ apple. All fruits were
harvested at maturity. Each point represents one bourse shoot. ........... 27

Figure 3. Relationship between total seed number and total weight of fruit per spur at
harvest in 1989, 1990, 1991, and 1992 for 'Spencer Seedless' apple spurs that
were ﬂowering or vegetative the following season. All fruits were harvested at
maturity. Only fruits from hand-pollinated branches were weighed in 1990.
Each point represents one bourse shoot. ............................. 29

Figure 4. The relationship between seed number per spur and percentage of 'Spencer
Seedless' apple bourse buds that ﬂowered the following growing season in 1994
(y = 96.1 - 15.2x, R2 = 0.97) and 1996 (y = 88.2 - 19.6x, R2 = 0.96). Seed

number classes are based on observation of apparent break points and sample

xiii

size. ......................................................... 30
Figure 5. The relationship between seed number and shoot length of bourse shoots of
'Spencer Seedless' apple in 1994 and 1996 vs. percentage of bourse buds that
ﬂowered the following growing season. Seed number and shoot length classes
were based on observation of apparent break points and sample size. ...... 31
Figure 6. Distribution of bearing spurs of ‘Spencer Seedless’ apple by total fruit mass
borne per spur in 1994 and 1996. No spurs bore less than 40 g. of fruit. 40 g =
40 to 59 g; 60 g = 60 to 79 g, etc. .................................. 33
Figure 7. Effect of total ‘Spencer Seedless’ apple fruit mass (g) borne per spur on
percentage of spurs that ﬂowered the following year, in 1994 and 1996. Spurs
bearing seedless and seeded fruits were ranked by total fruit mass (20 g
increments). Classes were combined when necessary to give sufﬁcient numbers
to determine P = 0.95 conﬁdence limits. Number of spurs per class varied from
49 to 337, as required to calculate conﬁdence limits for binomial data. Classes
represent spurs with fruit mass ﬁom the mass given to the next class above for a
given treatment, except for the lowest class, which represents all ﬁ'uits less than
the value given. Insufﬁcient spurs bore seedless fruit in 1996 to classify, so
mean fruit mass of all seedless spurs is reported. ...................... 34
Figure 8. The relationship between seed number per spur in 'Spencer Seedless' apple
and percentage of bourse buds that ﬂowered the following growing season on
spurs bearing less than or greater than the median fruit mass. Seed number

classes are based on observation of apparent break points and sample size. . . 35

xiv

Figure 9. The relationship between seed number and crop density per cm2 branch cross-
sectional area in 'Spencer Seedless' apple in 1994 vs. percentage of bourse buds
that ﬂowered the following growing season. Seed number and crop density

classes are based on observation of apparent break points and sample size. . . 36

SECTION 2 .......................................................... 45
Figure 1. Effect of time of defruiting entire ‘Paulared’ apple trees in 1991 and 1992 on
percentage of terminal buds of fruiting spurs ﬂowering the following year.
Shoots with dead terminal buds were excluded. Bars represent the standard
errors of the means for three replicate trees. .......................... 51
Figure 2. Effects of crop density and number of fruits borne on a spur on the average
number of lateral ﬂowers per bourse shoot ............................ 54
Figure 3. Effect of crop density on percentage of terminals ﬂowering (90 shoots per
tree) on girdled (y = 98.9 - 1.5x, R2 = 0.093, P = 0.34) and nongirdled (y = 105.7

- 7.0x, R2 = 0.56, P = 0.000056)‘Paulared’ trees. ...................... 56

SECTION 4 .......................................................... 76

Figure 1. Crop density on ‘Gala’ apple trees used (A) and effects of concentration and
time of application of gibberellin in the form of ProVide" in 1996 on bloom
rating in 1997 (B). Each point is the mean for ﬁve replicate trees. ........ 84

Figure 2. Effects of concentration and time of application of gibberellin in the form of

ProVide° on percentage terminal (A) and lateral (B) buds ﬂowering the

XV

following year on ‘Gala’ apple. Each point is the mean for ﬁve replicate trees.

Figure 3. Relationship (y = 8.32 - 0.38x, R2 = 0.38, P < 0.0001 ) between return bloom
rating in 1997 and crop density in 1996 in ‘Gala’ apple over three treatment
dates and three concentrations of ProVide" at Clarksville Horticulture
Experiment Station. ............................................. 86

Figure 4. Relationship between crop density in 1996 in ‘Gala’ apple and percentage of
lateral nodes ﬂowering in 1997 over three treatment dates for ProVide"
concentrations from 150 to 600 mg-L‘l at Clarksville Horticulture Experiment

Station. ....................................................... 87

xvi

LITERATURE REVIEW

Economic importance of ﬂower formation

Many fruit and nut trees exhibit cyclic yield variation known as alternate or
biennial bearing. This is characterized by a growing season where an abundance of
ﬁ'uits are produced, followed by a growing season where few if any ﬂowers and fruits
are produced. This alternate habit can occur among branches of a single tree, among
different trees within an orchard, or be synchronized by adverse weather conditions,
such as spring freezes, to cause entire orchards and regions to ﬂuctuate widely from
year to year. In the “on” years with an excessive crop, fruits are small and may not
mature with the quality consumers desire. In the “off" year, what few fruit that are
produced are often too large and lack the internal characteristics that allow them to store
and maintain good quality through the marketing channels. Thus, alternate bearing
poses considerable ﬁnancial risk for apple growers and creates unstable supply and
prices for consumers (Singh, 1948). After decades of work to understand and overcome
this problem, irregular cropping is still a major impediment to apple production. While
considerable progress has been made through the use of precocious rootstocks and by
partial defruiting with chemical thinning agents, understanding of the basic cause(s) of
alternate bearing is still limited. Better understanding of the physiology of ﬂowering
and fruiting could lead to novel ways to overcome this persistent problem.

Despite the relative complexity of woody plant genomes, Weigel and Nilsson
(1995) have successfully transferred the LF Y gene responsible for precocious ﬂowering

from Arabidopsis to the woody plant Populus, resulting in solitary axillary ﬂowers and

an abnormal terminal ﬂower on ﬁve-month-old shoots grown in tissue culture. Tobutt
(1994) has determined that the apetelous parthenocarpy trait in apple is controlled by a
single recessive gene and has combined this with columnar growth habit. The gene for
apetelous parthenocarpy may also inﬂuence ﬂowering in apple, as seedless apples do
not inhibit ﬂowering (Chan and Cain, 1967). Detailed understanding of a plant and its
physiology is required to apply the tools of molecular biology in solving production
problems. This certainly applies to apple.
Growth and ﬂowering habit

Apple (Malus domestica Borkh.) is a woody perennial plant adapted to growth
in temperate climates. Thus, in addition to buds formed in the axils of leaves, temrinal
buds also form as episodic shoot growth ceases, beginning a dormant phase. The shoots
grow to various lengths before setting terminal buds. A lateral bud that produces a
shoot with less than about six leaves and elongates only 1 to 20 mm is called a spur.
These frequently form ﬂower buds and constitute the best form of growth for fruit
production. Their growth ceases early in the season, and terminal buds usually begin
forming a few days following full bloom. Some lateral and terminal buds produce
shoots that have greater than six leaves and lengths from several centimeters to a meter
or more. Their growth period can last from just a few days longer than that of spurs to
the entire growing season. Thus, terminal buds can form at any time, ﬁom a few days
after growth starts in spring to late autumn.

The ﬁrst visible evidence that terminal growth is ceasing is the abscission of the

lamina and petiole of the newest emerging leaf; the leaf base forms the ﬁrst bud scale

(Abbott, 1970). Once terminal buds begin to form, shoot elongation ceases until
dormancy is broken, usually by chilling temperatures in temperate climates, but
occasionally by summer precipitation, which causes a second growth ﬂush. In spring,
buds begin to grow and produce another ﬂush of growth. Much of the new tissue is
already differentiated within the bud and the new growth is simply the result of cell
expansion, thus the content of these buds is determined during the previous growing
season. Apple produces both vegetative buds (containing only shoot primordia) and
reproductive buds (containing both an inﬂorescence and shoot primordia) known as
ﬂower, ﬂoral or fruit buds.

All terminal and lateral buds have the potential to ﬂower (Zeller, 1960), but the
probability of ﬂower buds developing varies greatly with cultivar and growing
conditions. This variation leads to cultivars being classiﬁed as “spur bearing”, “terminal
bearing”, and “lateral ﬁ'uiting”. These types of fruiting are not exclusive and occur
simultaneously to varying degrees in most cultivars. Apple ﬂower buds are mixed buds
that produce a terminal inﬂorescence in the form of a cyme with shoots arising from the
axils of leaves attached to the swollen peduncle (bourse). These bourse shoots become
the new branch or spur terminals, eventually with their own terminal buds. Usually
only one bourse shoot forms, but weak spurs may have none, and two bourse shoots are
common on vigorous spurs. Since temrinal growth generally ceases after full bloom,
and ﬂower induction occurs some time near cessation of terminal growth, ﬂower
induction and differentiation are concurrent with fruit development. Although induction

can occur after harvest, few if any terminals are still growing at that time. Lateral

ﬂowers form from August through March, after those in the terminal buds of spurs, and
seldom in a uniform manner (Zeller, 1960). Thus growth conditions during the season
inﬂuence reproductive growth and development in both the developing fruit and the
buds being formed for the next growing season.
Time of induction

Some time after terminal growth ceases, a physiological “switch” determines
whether the bud will be vegetative or generative (ﬂower induction). There is no visible
evidence that induction has occurred, but cytochemical and histological differences
between induced and non-induced buds have been observed prior to morphological
changes. The level of DNA rises in both the apical zone and in the central meristem.
This is followed by increased mitotic activity throughout the apex, and the central
meristem can be seen immediately surrounded by the subderrnatogen (Buban, 1981).
Existing leaf primordia are also modiﬁed and develop into eight more scales. These
form an enclosure as the meristem continues to initiate primordia of transition leaves,
true leaves, and, if ﬂower induction has occurred, bracts, a terminal (king) ﬂower, and
lateral ﬂowers in the axils of the bracts and most apical true leaves. The ﬁrst
microscopic evidence that ﬂower induction has occurred (ﬂower initiation) is a
ﬂattening of the terminal dome of the apical meristem (Goff, 1899, as cited by Davis,
1957) 17-21 nodes (and several weeks) after the ﬁrst bud scale (McLaughlin and
Greene, 1991). A bourse shoot(s) is differentiated at a node proximal to those with

ﬂower primordia. This microscopic differentiation takes place over the growing season.

Understanding the biological switch that determines whether a given bud is
vegetative or ﬂoral is an important key to understanding and controlling alternate
bearing. Once the switch has been thrown, the fate of that bud for the following
growing season is ﬁxed. However, the vitality of these buds will largely depend on
growing conditions during differentiation and maturation - after induction takes place.
Factors inﬂuencing ﬂower induction

W J uvenility is the ﬁrst impediment to ﬂower and fruit
production in apple; however, in commercial apple production this problem is overcome
by the use of mature clones. This imparts uniformity in growth habit of the trees and
characteristics of the fruit, and all cuttings and bud sticks used for propagation are
obtained from stock already in the adult phase; thus the new shoots are adult tissue and
capable of ﬂower bud formation. The scope of this study will be conﬁned to the control
of ﬂowering of adult tissue. Genetic variation within the scion and rootstock, which are
usually different genotypes, includes variation in precocity. A primary criterion for
selecting a suitable rootstock is the inﬂuence it has on the scion in hastening and
enhancing ﬂower production. Although Hirst and F erree (1995a, 1995b and 1996)
reported that rootstocks affect vegetative growth and precocity, but not time or pattern
of ﬂower bud development, Nesterov et al. (1972, as cited by Buban, 1996 ) reported
that dwarﬁng rootstocks caused induction to occur earlier in the season.

WW3. Many environmental and cultural factors
inﬂuence ﬂower formation, including shading (Auchter et al., 1926; Gourley, 1920),

temperature (T romp, 1992, 1993), water supply (Sritharan and Lenz, 1988), nutrition

(Denker and Hansen, 1994; Hipps, 1992) and growth regulators, including
naphthaleneacetic acid (Harley et al., 1958), benzyladenine (W ertheirn and Estabrooks,
1994), carbaryl (McArtney et al., 1995), growth retardants (Ramirez and Hoad, 1981;
Tromp, 1987), and gibberellins (Bangerth and Schroder, 1994; Greene, 1993; Looney et
al., 1992).

Foliar application of GA inhibits ﬂowering in apple, depending on time of
application, concentration and the gibberellin used (Guttridge, 1962; Marcelle and
Sironval, 1963). McArtney (1994) reduced the severity of the alternate bearing cycle in
‘Braebum’ apple by applying GA, and GA4.7 at full bloom. Although there was a linear
response to concentration, even the highest rate of GA3 (330 ppm) did not entirely
eliminate alternate bearing.

Environmental effects and chemical treatments often inﬂuence the current crop,
making it difficult to determine whether effects on ﬂowering are direct or indirect.
Some investigators have proposed a direct effect of chemical thinning agents in
stimulating ﬂowering (Harley et al., 195 8), but most consider the effect of thinning to
be indirect via fruit removal. Branch orientation also affects ﬂowering (Abbott, l 960;
Edwards, 1987). Emphasis in this thesis will be conﬁned to the interactions within a
tree between shoot, fruit and seed development in affecting ﬂower induction.

WW Flower induction is associated with the cessation of
shoot growth and is dependent on adequate leaf area (Davis, 195 7). In temperate
regions, bloom is synchronized so that most ﬂowers open within 2 - 5 days. These

dates vary from year to year, and subsequent fruit development varies in a parallel

manner. Thus full bloom is a convenient developmental event used to relate subsequent
events in fruit development, usually expressed in days after full bloom (DAF B). How
cessation of shoot growth or ﬂower induction within buds is related temporally to
ﬂowering and fruit development is not known, but DAF B is the common reference used
for lack of a better alternative.
Cessation of bourse shoot grth may be rather soon after full bloom in the “on”
year of an alternate bearing tree or relatively later in an annual bearing one (Davis,
1957). Variations in shoot length affect other characteristics; longer shoots can have
higher leaf numbers, more leaf area, and a later date of cessation of growth. Manual
reduction of spur leaf area reduces ﬂowering of “off" trees the following year depending
on the time of defoliation (Harley et al., 1958; Li et al., 1995). Struckmeyer and
Roberts (1942) found that removal of the two terminal leaves of ‘Wealthy’ spurs had an
inhibitory effect on ﬂower induction similar to that of removing all leaves except the
two terminal ones, but less effect than if all leaves were removed. Caustic compounds
used for fruit thinning reduced ﬂowering when injury to foliage was pronounced.
Clearly the presence of leaves on spurs is critical for ﬂower induction, but what role
they play is not well deﬁned.

Roberts (1920), studying spurs (terminals with not more than 65 mm growth) in
“off-year trees, found that vegetative spurs were either long or short, averaging 2.6 mm
vs. 18.1 mm of growth the previous year, respectively. Bourse shoots that ﬂowered

were intermediate in length; those that set fruit averaged 12.1 mm vs. 4.3 mm for those

that did not. While length varied with cultivar and location, the relationship between

length and ﬂowering/fruiting was conﬁrmed by several other investigators. Auchter and
Schrader (1923) classiﬁed 1321 spurs (shoots 76 mm or less) of ‘York Imperial’ into
ﬁve classes, and found that most of those that grew only 1 - 3 mm were continuously
vegetative, whereas most of those that grew more than 5 mm were ﬂoral the following
year, with percentage ﬂowering increasing with shoot length. This was only true of trees
with an annual bearing habit. When a heavy crop threw a tree into an “off year, no
spurs formed fruit buds, regardless of their growth the previous year. The next year
nearly all spurs blossomed, including those that previously had been continuously
vegetative. Growth of such spurs was stimulated in the “off” year, and the few that
remained vegetative in the subsequent “on” year were the shortest ones, with few
exceptions. “Off" year trees had average spur growth greater than “on” year trees (10.6
mm vs. 6.3 mm, respectively), but general vigor was good in these trees in all years.
Thus ﬂowering is correlated with shoot growth, but fruiting can affect ﬂowering
independently of shoot growth.

W9. The dominant effect of fruiting on ﬂowering is well established.
The removal of the fruit from a tree or large branch allows ﬂower buds to form in
alternate-bearing cultivars, whereas on comparable fruiting units few or none will
differentiate (Davis, 195 7). Two alternative hypotheses have been proposed to explain
this dominant effect: (a) overall crop acts collectively on all developing buds (crop
effect) (Palmer, 1992) vs. (b) the fruits have a local effect only on bourse shoots arising
from their own cluster bases (local effect) (Palmer et al., 1991). Auchter (1919),

Roberts (1920) and Crow (1920) reported that spurs that bore fruit until ‘June’ drop

rarely produced ﬂower buds. Auchter and Schrader (1923) also reported that 50 percent
of the non-blossoming spurs ﬂowered the following year in a ‘Stayman Winesap’ tree
bearing a light to medium crop. Response of ﬂowering spurs declined as
deﬂoration/defruiting was delayed. Blossom/fruit removal at pink, full bloom, ‘June’
drop, and harvest resulted in 37, 32, 5, and 1 percent bloom the following year,
respectively. Blossom removal increased secondary growth of spurs, and ﬂowering was
correlated with this growth. In reviewing studies on leafzfruit ratio, Davis (1957)
concluded that two to three times as many leaves were required to induce ﬂower
induction on non-ringed branches compared with ringed branches. Apparently some
promoter(s) moved out of the non-ringed branch and more leaves were therefore
required to permit induction.

Seeds, In his study characterizing ﬂower initiation, Goff (l 899, as quoted by
Davis, 1957) stated "...We can now easily understand why a bountiful fruit crop so often
causes reduced ﬁ'uitage the following year. The nourishment that might otherwise
contribute to the formation of ﬂowers is absorbed by the numerous seeds that are
maturing.” Likewise, Heinicke (1917) noted, “It is conceivable that a tree bearing a
heavy crop of many-seeded fruits is being devitalized to a far greater extent than another
tree of the same variety bearing a crop of fruits equally heavy but having relatively few
seeds.” In listing reasons for early thinning (about the time of ‘June’ drop), Gourley
(1922) stated, “The development of seeds and “pits” drains the energies of the tree.”
Gourley’s inference was based on the data of Bigelow and Gore (1905, as cited by

Gourley, 1922) for peach, which indicated that a large percentage of total solids

occurred in the stone and kernel at maturity and that most pit development occurred
early in fruit development, before pit hardening.

'Spencer Seedless' is a homeotic ﬂoral mutant apple cultivar that is homozygous
recessive ape ape (Acquaah et al. 1992; Coen, 1991; Tobutt, 1994). This gene causes a
second whorl of sepals to be substituted for the petals (Brase, 193 7), and an additional
10 carpels are substituted in place of the stamens, making the ﬂowers pistillate and
capable of producing over 30 seeds per fruit. Although the apetalous ﬂowers are not
attractive to bees and thus not normally insect-pollinated, they set seedless ﬁ'uits readily
in most years, and form numerous ﬂower clusters annually both on spurs and shoots.
By hand pollinating the apetalous ﬂowers, seeded and seedless ﬁ'uits can be produced
on the same tree. The fruits are of no commercial value.

Using a ’Spencer Seedless' apple tree at Geneva, New York, Chan and Cain
(1967) discovered that when fruits were left until maturity, seeded fruits inhibited
ﬂower formation but seedless fruits did not. Further, by removing fruits at intervals
dming development, they were able to determine the time when developing seeds
became active in preventing ﬂower induction. When similar experiments were
performed in a greenhouse, seeds failed to inhibit ﬂowering to the same degree and this
failure was associated with bourse shoots longer than 2 cm. Huet, (1972, 1973)
conﬁrmed the effects of seeds on ﬂowering of 'Bartlett' pear, and showed that seeds
were more inhibitory at low than at high leaf area. This supports the conclusion that a
substance other than carbohydrate is responsible for the fruit’s effect on ﬂower

induction. The data of Griggs et al. (1970) in California were more ambiguous; over 50

10

percent of the bourse shoots on spurs bearing seeded fruit still ﬂowered in ‘Bartlett’
pear, and most of those were shorter than 25 mm.
Possible reasons for effects of seeds

Flower induction is controlled by genes that are activated under certain
conditions. The unknown factors are the precise conditions required and the nature of
the signal. From the above review, the time period during which a developing bud can
be induced to become ﬂoral is ﬁxed, and the signal must be sent at the appropriate time.

In apple, leaves promote and seeds inhibit ﬂowering. Several hypotheses have
been suggested to explain these facts. Both Goff (1 899, as cited by Davis, 1957) and
Heinicke (1917) suggested that seeds require “nutrients” over and above those needed
by the fruit and therefore “drain the tree of nutrients”. Heinicke (1917) demonstrated
that seeds, “...supplement the forces that bring sap to the fruit.” The compounds
attracted by seeds could be carbohydrates and minerals, on the one hand, or hormones
on the other. Their movement must be explainable in terms of the known transport
mechanisms in plants (xylem and phloem).

MW One alternative hypothesis is that fruits are
superior competitive sinks for carbohydrate, thus buds are “starved” into being
vegetative. Flowers and fruits accumulate dry matter slowly during the cell-division
phase of development, but use large amounts of carbohydrate in respiration. This
occurs when shoot growth is also rapid and leaf area is limited, creating strong
competition for stored carbohydrates. Seeds enhance the sink effect of fruits for

carbohydrate, thus reducing the ﬂow of carbohydrates to the developing buds to a level

11

that causes them to remain vegetative. This hypothesis assumes that ﬂoral-gene
activation is mediated by carbohydrate supply or concentration. Little evidence exists
for this mechanism.

Wm, Since developing fruits, leaves and buds all transpire,
the competition could be driven by water relations rather than carbohydrate. F ruitlets
can lose water and shrink during daily periods of high water demand, followed by
recovery at night when transpiration demands are reduced. This suggests that it is
unlikely that fruits have a lower water potential than the developing bud.

A major discrepancy that needs resolution with the “competing sinks” hypothesis
is that both seeded and seedless apple fruits are sinks for carbohydrate and water, but
the former inhibit ﬂowering while the latter do not. The need for carbohydrate may be a
quantitative one, since there is a correlation between seed number and individual fruit
size. This correlation is mediated by spur size, ﬂower size, ﬂower position within the
inﬂorescence, fruits borne on a spur and cropload (Heinicke, 1917).

Wen: The positive effect of leaves on ﬂowering of
herbaceous plants led to coinage of the term “ﬂorigen” by Chailykyan (193 7); however,
this compound(s) has never been identiﬁed. Seedless fruits use carbohydrates and
water, but conceivably do not attract “ﬂorigen” while seeded fruits do. Thus buds
adjacent to seedless fruits can ﬂower. This would require a mechanism in which
developing seeds control loading or unloading of “ﬂorigen” from sap, thus altering the
transport of “ﬂorigen” to the developing bud(s).

Wane; Seeds could have a direct effect on ﬂowering by

12

exporting an inhibitor of ﬂowering to the buds. Such export would require movement
against the general ﬂow of metabolites. Transient or continuous ﬂow of water out of
fruitlets through the xylem could carry inhibitors to the bud.

Seeds are sources of several known hormones. Applications of auxin, ethylene
and cytokinin can promote ﬂowering, so there is little support for their roles as
inhibitors of ﬂowering; however, seed-produced gibberellins could inhibit ﬂower
induction. Gibberellins A4 and A-, were ﬁrst extracted from apple seeds (Dennis and
Nitsch, 1966), and many other gibberellins have been identiﬁed subsequently in seed
extracts using gas chromatography-mass spectrometry (Hedden, et al., 1993; Hoad,
1980; Hoad and Ramirez, 1980; Lin et al. 1991, Luckwill et al., 1969).

Grochowska (1968) replaced seeds with cotton swabs soaked in growth
regulators. Indoleacetic acid (1AA) at 20 mgoL“ and gibberellic acid (GA,) at 500
mgoL" had no effect, but GA, at 1000 mg-L'l reduced ﬂowering 33 percent and
naphthaleneacetic acid (NAA) at 10 mg-L'l promoted ﬂowering. To my knowledge this
experiment has never been repeated. Hoad and Ramirez (1980) and Hoad (1980)
reported more gibberellins moving out of fruitlets of the strongly biennial ‘Laxton’s
Superb’ than the more annual ‘Cox’s Orange Pippin’, as measured in diffusates
collected from the pedicels in vitro. Stephan et al. (1997) reported export of GA,, GA,,
GA“ GA,, GA20, and GA,, by examining developing apple fruit exudates. One can
question whether exudates from fruits represents a natural transport, since removing the
fruits would disrupt natural transport mechanisms. They did not report the details of

how these exudates were collected and how this process would compare to natural

13

transport. Further, they collected over thirteen times as much total gibberellins from
‘Spencer Seedless’ fruits (without seeds) than ﬁ'om ‘Elstar’, a biennial cultivar. Most of
the exudate gibberellin from ‘Spencer Seedless’ was GA4, which is reported to even
promote ﬂowering (Looney, et al., 1985), but excluding GA4, ‘Spencer Seedless’ still
exuded 2.5 times as much inhibitory gibberellins from fruits. They did not report
ﬂowering response from spurs growing in similar conditions to those that were sampled
for analysis.

Neither Green (1987), Stephan et al. (1997), nor Ban (1996) found signiﬁcant
transport of labeled GA from seeds to buds of apple. When Ban (1996) injected l“C-
GA” into apple seeds in viva, , no radioactivity occurred in the apex, with one
exception. One unidentiﬁed polar metabolite was found in the cluster base and two
metabolites were found in the fruit ﬂesh. Less than 0.5 percent of the radioactivity was
found outside the fruit. Green (1987) injected 3H-GA4 into apple seeds and found only
2-3 percent movement out of the seeds at most, and much less (0.0003-0.07 percent)
was found in the bud when treated 49 - 77 days after full bloom. The cultivar that
exhibited more movement in one year showed less movement the next, thus casting
doubt on any correlation between alternate bearing and GA export from fruits. This
contrasts with the reports of Hoad and Ramirez (1980) and Hoad (1980). They found
movement of 3H to buds following injection of 3H-GA4 into seeds, and this appeared to
be greater in ‘Laxton’ than in ‘Cox’, but the values were small and the differences
nonsigniﬁcant. Both Ban (1996) and Green (1987) found even less movement of GAs

in vitro than in viva. Movement of gibberellins may vary, not only with cultivar, but

14

with growing conditions, or time and method of measurement. Also puzzling is the fact
that while applied GAs inhibit ﬂowering, they also reduce seed number of ‘McIntosh’
when applied at pink, full bloom and petal fall (Dennis and Edgerton, 1966).
Summary

Apple is an important fruit crop in an industry that depends on abundant annual
ﬂowering of each tree to produce economic yields of high value fruit. Alternate bearing
continues to be a major problem in apple and the speciﬁc causes are still unknown.
Shoots, seeds within developing ﬁ'uits, crop load and applied gibberellins all have been
shown to inﬂuence ﬂower induction in some way. Further data is needed to better

understand the interactions among these parameters that inﬂuence ﬂower induction.

Literature Cited

Abbott, UL. 1960. The bourse shoot as a factor in the growth of apple fruits. Ann.
App. Biol. 48:434-438.

Abbott, BL. 1970. The role of budscales in the morphogenesis and dormancy of the
apple fruit bud, p. 65-82. In: L.C. Luckwill and CV. Cutting (eds.). Physiology
of Tree Crops. Academic Press, New York.

Acquaah, G., J .W. Saunders and L.C.Ewert. 1992. Homeotic ﬂoral mutations. Plant
Breeding Rev. 9:63-99.

Auchter, EC. 1919. Some inﬂuences of thinning, pollination, and fruit spur growth on
the yearly performance record of ﬁ'uit spurs and on the size of fruit produced.
Proc. Amer. Soc. Hort. Sci. 16:118-131.

Auchter, EC. and AL. Schrader. 1923. Fruit spur growth and fruit bud production.
Proc. Amer. Soc. Hort. Sci. 20:127-144.

Auchter, E.C., A.L. Schrader, F.S. Lagasse and W.W. Aldrich. 1926. The effect of

shade on the growth, fruit bud formation and chemical composition of apple
trees. Proc. Amer. Soc. Hort. Sci. 23:368-382.

15

Ban, J. 1996. Gibberelin metabolism and apple ﬂowering. PhD Dissertation, Michigan
State Univ.

Bangerth, F. and M. Schrader. 1994. Strong synergistic effects of gibberellins with the
synthetic cytokinin N-(2-chloro-4-pyridyl)-N-phenylurea on parthenocarpic
fruit-set and some other fruit characteristics of apple. Plant Growth Regul.

l 5 :293 -3 02.

Bigelow, W.D. and HG Gore. 1905. US. Dept. Agr. Bur. Chem. Bul. 97. (Cited by
Gourley, 1922.)

Brase, K.D. 1937. The vascular anatomy of the ﬂower of M domestiea Borkh. f.
Ma van Eseltine. MS. Thesis, Cornell University.

Buban, T. 1981. Histological and cytological aspects of ﬂower bud formation in apple.
Acta Hort. 120:113-117.

Buban, T. 1996. Flower development and formation of sexual organs, p. 3-54. In: I.
Nyéki and M. Soltész (eds.). Floral biology of temperate zone fruit trees and
small fruits. Akadéniai Kiado, Budapest.

Chailakhyan, M.Kh. 1937. Hormone theory of plant development (Russian) Moscow,
Leningrad. Akad. Nauk. SSR.

Chan, B.G. and J.C. Cain. 1967. The effect of seed formation on subsequent ﬂowering.
Proc. Amer. Soc. Hort. Sci. 91:63-68.

Coen, ES, 1991. The role of homeotic genes in ﬂower development and evolution.
Annu. Rev. Plant Physiol. Plant Mol. Biol. 42:241-279.

Crow, J .W. 1920. Biennial fruit bearing in the apple. Proc. Amer. Soc. Hort. Sci.
17:52-54.

Davis, L.D. 1957. Flowering and alternate bearing. Proc. Amer. Soc. Hort. Sci.
70:545-556.

Denker, I. and P. Hansen. 1994. Flower initiation and node development of axillary
buds, growth partitioning, and bleeding sap production of young apple trees as
affected by fertigation. J. Hort. Sci. 69:869-876.

Dennis, F.G., Jr. and L.J. Edgerton. 1966. Effects of gibberellins and ringing upon

apple fruit development and ﬂower bud formation. Proc. Amer. Soc. Hort. Sci.
88: 14-24.

16

Dennis, F.G., Jr. and JP. Nitsch. 1966. Identiﬁcation of gibberellins A, and A7 in
immature apple seeds. Nature 211:781-782.

Edwards, GR. 1987. Producing temperate-zone fruit at low latitudes: Avoiding rest and
the chilling requirement. HortScience 22: 1236-1240.

Goff, ES. 1899. The origin and early development of the ﬂower in the cherry, plum,
apple, and pear. Wisconsin Agr. Expt. Sta. Annu. Rept. p. 289-303. (Cited by
Davis, 1957.)

Gourley, J.H. 1920. The effect of shading some horticultural plants. Proc. Amer. Soc.
Hort. Sci. 17:256-260.

Gourley, J .H. 1922. Text book of pomology. Macmillan, New York.

Green, JR. 1987. The hormonal control of biennial bearing in cider apples. PhD
Dissertation, Univ. of Bristol.

Greene, D.W. 1993. Effects of GA, and GA-, on ﬂower bud formation and russet
development on apple. J. Hort. Sci. 68: 1 71-176.

Griggs, W.H., G.C. Martin and ET Iwakiri. 1970. The effect of seedless versus seeded
fruit development on ﬂower bud formation in pear. J. Amer. Soc. Hort. Sci.
95:243-248.

Grochowska, M.J. 1968. The inﬂuence of growth regulators inserted into apple ﬁ'uitlets
on ﬂower bud initiation. Bul. Acad. Polon. Sci., Ser. Biol. 16:581-584.

Guttridge, C.G.1962. Inhibition of fruit-bud formation in apple with gibberellic acid.
Nature 196:1008.

Harley, C.P., H.H. Moon and LO. Regeimbal. 1958. Evidence that post-bloom apple-
thinning sprays of naphthaleneacetic acid increase blossom-bud formation.
Proc. Amer. Soc. Hort. Sci. 72:52-56.

Hedden, P., G.V. Hoad, P. Gaskin, M.J. Lewis, J.R. Green, M. Furber and LN. Mander.
1993. Kaurenoids and gibberellins, including the newly characterized
gibberellin A3,, in developing apple seeds. Phytochemistry 32:231-237.

Heinicke, A.J. 1917. Factors inﬂuencing the abscission of ﬂowers and partially
developed fruits of the apple (Pyrus malus L.). Cornell Univ. Agr. Expt. Sta.
Bul. 393.

Hipps, NA. 1992. Fertigation of newly planted ‘Queen Cox’/M9 apple trees-

17

establishment, early growth and precocity of cropping. J. Hort. Sci. 67:25-32.

Hirst, PM. and DC. Ferree. 1995a. Rootstock effects on shoot morphology and spur
quality of ‘Delicious’ apple and relationships with precocity and productivity. J.
Amer. Soc. Hort. Sci. 120:622-634.

Hirst, PM. and DC. Ferree. 1995b. Rootstock effects on the ﬂowering of ‘Delicious’
apple. 1. Bud development. J. Amer. Soc. Hort. Sci. 120:1010-1017.

Hirst, PM. and DC. Ferree. 1996. Effects of rootstock on bud development and ﬂower
formation of ‘Starkspur Supreme Delicious’ apple. Fruit Var. J. 50:25-34.

Hoad, G.V. 1980. The role of seed derived hormones in the control of ﬂowering in
apple. Acta Hort. 80:93-103.

Hoad, G.V. and H. Ramirez. 1980. La ftmcion de las giberelinas sintetizadas en las
sernillas del fruto para el control de la ﬂoracion en manzanos. Turrialba 30:284-
288.

Huet, J. 1972. Etude des effe‘ts des feuilles et des fruits sur l'induction ﬂorale des
brachyblastes du Poirier. Physiol. Vég. 10:529-545.

Huet, J. 1973. Floral initiation in pear trees. Acta Hort. 34: 193-198.

Li, S.H., S.Q. Meng, T.H. Li, H.Z. Liu and Y.C. Tu. 1995. Critical period of ﬂower bud
induction in ‘Red Fuji’ and ‘Ralls Janet’ apple trees. Gartenbauwissenschaﬁ
60:240-245.

Lin, J.-T., A.E. Stafford and G.L. Steffens. 1991. Identiﬁcation of endogenous
gibberellins in immature apple seeds. Agr. Biol. Chem. 55:2183-2185.

Looney, N.E., R.L. Granger, C.L. Chu, S.J. McArtney, L.N. Mander and RP. Pharis.
1992. Inﬂuences of gibberellins A4, A,.-,, and A,+iso-A7 on apple fruit-quality
and tree productivity. 1. Effects on ﬁ'uit russet and tree yield components. J.
Hort. Sci. 67:613-618.

Looney, N.E., R.P. Pharis, and M. Noma. 1985. Promotion of ﬂowering in apple trees
with gibberellin A4 and C-3 epi-gibberellin A4. Planta 165:292-294.

Luckwill, L.C., P. Weaver and J. MacMillan. 1969. Gibberellins and other growth
hormones in apple seeds. J. Hort. Sci. 44:413-424.

Marcelle, R. and C. Sironval. 1963. Effect of gibberellic acid on ﬂowering of apple
trees. Nature 197:405.

18

McArtney, SJ. 1994. Exogenous gibberellin affects biennial bearing and the fruit shape
of ‘Braebum’ apple. New Zealand J. Crop and Hort. Sci. 22:343-346.

McArtney, S.J., D.S. Tustin, S. Seymour, W. Cashmore and NE. Looney. 1995.
Benzyladenine and carbaryl effects on fruit thinning and the enhancement of
return ﬂowering of 3 apple cultivars. J. Hort. Sci. 70:287-296.

McLaughlin, J .M. and D.W. Greene. 1991. Fruit and hormones inﬂuence ﬂowering of
apple. 1. Effect of cultivar. J. Amer. Soc. Hort. Sci. 116:446-449.

Nesterov, Y.S., G.G. Anisimova and NJ. Savelev. 1972. Vestnik Sel’skokhov. Nauki
10:114-118. (Cited by Buban, 1996.)

Palmer, J .W. 1992. Effects of varying crop load on photosynthesis, dry-matter
production and partitioning of ‘Crispin’lM.27 apple trees. Tree Physiol. 11:19-
33.

Palmer, J.W., Y.L. Cai and Y. Edjamo. 1991. Effect of part-tree ﬂower thinning on
fruiting, vegetative growth and leaf photosynthesis in ‘Cox Orange Pippin’
apple. J. Hort. Sci. 66:319-325.

Ramirez, H. and G.V. Hoad. 1981. Effects of growth substances on fruit-bud initiation
in apple. Acta Hort. 120:131-136.

Roberts, RH. 1920. Off year bearing and apple spur growth. Wisconsin Agr. Expt.
Sta. Bul. 317.

Singh, LB. 1948. Studies in biennial bearing. 11. A review of the literature. J. Hort.
Sci. 24:45-65.

Sritharan, R. and F. Lenz. 1988. Effects of water-stress in apple. 1. Effect of water-
supply on ﬂower formation and fruit-set of the apple cultivar ‘Golden
Delicious’. Gartenbauwissenschaft 53:223-226.

Stephan, M., F. Bangerth and G. Schneider. 1997. Transport and metabolism of the
gibberellins A,, A,, and A4 after application to developing apple fruits of Malus
domestica cv. Jonagold. Acta Hort. 463:113-119.

Struckmeyer, BE. and RH. Roberts. 1942. Investigations on the time of blossom
induction in Wealthy apple trees. Proc. Amer. Soc. Hort. Sci. 40:113-119.

Tobutt, KR. 1994. Combining apetalous parthenocarpy with columnar growth habit in
apple. Euphytica 77:51-54.

19

Tromp, J. 1987. Growth and ﬂower-bud formation in apple as affected by
paclobutrazol, darninozide, and tree orientation in combination with various
gibberellins. J. Hort. Sci. 62:433-440.

Tromp, J. 1992. The effect of soil-temperature on lateral shoot formation and ﬂower-
bud formation in apple in the 1st year after budding. J. Hort. Sci. 67:787-793.

Tromp, J. 1993. Lateral shoot formation and ﬂower-bud formation in apple in the lst
year after budding as affected by air-temperature and exposure to red-light. J.
Hort. Sci. 68:255-260.

Weigel, D. and O. Nilsson. 1995. A developmental switch sufﬁcient for ﬂower
initiation in diverse plants. Nature 377:495-500.

Wertheim, SJ. and EN. Estabrooks. 1994. Effect of repeated sprays of 6-
benzyladenine on the formation of sylleptic shoots in apple in the fruit tree
nursery. Scientia Hort. 60:31-39.

Zeller, O. 1960. Entwicklungsgeschichte der Bltitenknospen und F ruchtanlangen an
einjahrigen Langtrieben von Apfelbﬁschen. I. Entwicklungsverlauf und
Entwicklungsmorphologie der Bltiten am einjahrigen Langtrieb. Z. ﬁir
Pﬂanzenz. 44:175-214.

20

SECTION 1. EFFECTS OF SEED NUMBER, FRUIT REMOVAL, BOURSE
SHOOT LENGTH AND CROP DENSITY ON FLOWERING IN
‘SPENCER SEEDLESS’ APPLE.

Abstract

The objective of this study was to determine the effects of seed number, bourse
shoot length, and crop density on ﬂowering in ‘Spencer Seedless’ apple. ‘Spencer
Seedless’ is an apetalous cultivar that is facultatively parthenocarpic; pollination results
in seeded fruits. Individual spurs bearing either seedless or seeded fruits were tagged
and the ﬁ'uits were removed at various times between bloom and harvest in 1989-1992.
The fruits were weighed, seeds counted and bourse shoots measured for each spur.
Seventy to 100 percent of the spurs bearing seedless fruits ﬂowered the following year.
The percentage ﬂowering in spurs bearing seeded fruits decreased as defruiting was
delayed; this reduction was variable among years, with 10 to 70 percent of the spurs
ﬂowering when fruits were left until harvest. In 1994 and 1996 spurs were tagged at
harvest, and grouped into classes based on fruit weight, seed number, shoot length, and
crop density. Flowering of each bourse shoot was recorded the following year.
Percentage ﬂowering decreased from 94 to 38 (1995) or 15 (1997) as seed number
increased from 0 to >15. Flowering increased with bourse shoot length; no bourse
shoots less than 2 mm long ﬂowered, whereas almost all bourse shoots longer than 16
mm ﬂowered regardless of seed number. Seeds had the greatest inhibitory effect in
shoots 2-4 mm long or when crop density was greater than four fruits per cm2 branch

cross-sectional area. There was little inhibitory effect of fruit mass independent of

seeds.

21

Introduction

Critical to studying ﬂower induction is having a model plant system that can be
manipulated to produce ﬂoral or vegetative buds with a high degree of predictability.
Most research on the control of ﬂowering has been conducted with plants in which
ﬂowering is controlled by photoperiod (Salisbury, 1982), but this work has not been of
use in solving the alternate bearing problem found in tree fruits, particularly apple, in
which photoperiod has little effect on ﬂower formation, as evidenced by the production
of apple year-round at low latitudes (Edwards, 1987; Gorter, 195 5).

Excessive cropping reduces or eliminates ﬂowering in apple. When fruits are
removed within 30 days of bloom, ﬂower formation is enhanced for the following year
(Harley, et al. 1942). 'Spencer Seedless' is a ﬂower mutant apple cultivar that is
homozygous recessive ape ape (Acquaahet al. 1992; Coen, 1991; Tobutt, 1994). This
gene causes a second whorl of sepals to be substituted for the petals (Brase, 193 7), and
an additional 10 carpels are substituted for the stamens, making the ﬂowers pistillate
and capable of producing over 30 seeds per fruit. Although the apetalous ﬂowers are
not attractive to bees and thus not normally insect-pollinated, they set seedless fruits
readily in most years, and form numerous ﬂower clusters annually both on spurs and
shoots. By hand pollinating the apetalous ﬂowers, seeded and seedless fruits can be
produced on the same tree. The fruits are of no commercial value.

Using a 'Spencer Seedless' apple tree at Geneva, New York, Chan and Cain
(1967) discovered that when fruits were left until maturity, seeded fruits inhibited

ﬂower formation but seedless fruits did not. Further, by removing fruits at intervals

22

during development, they were able to determine the time when developing seeds
became active in preventing ﬂower induction. When similar experiments were
performed in a greenhouse, seeds failed to inhibit ﬂowering to the same degree and this
failure was associated with bourse shoots longer than 2 cm. Huet (1972, 1973)
conﬁrmed the effects of seeds on ﬂowering of 'Bartlett' pear, and showed that seeds
were more inhibitory at low than at high leaf area.

We repeated Chan and Cain's work as a means to determine the time of ﬂower
induction for sampling of spurs for a related study of the role of endogenous
gibberellins in ﬂowering. In this paper we report on the effects of several factors that

modify the inhibitory effects of seeds on ﬂowering.

Materials and Methods

1219913919991, Two 'Spencer Seedless' apple trees on M.7 rootstock, planted in
1969 at the Horticultural Research and Teaching Center, East Lansing, Michigan, were
used.

W931, By hand pollinating ﬂowers of selected branches
of each tree with mixtures of 'Red Delicious', 'Golden Delicious', 'Rome Beauty' or
'McIntosh' pollen (collected locally or purchased ﬁ'om Antles Pollen Supplies Inc.,
Wenatchee, Wash, or F irman Pollen Co., Inc., Yakima, Wash), both seeded and
seedless fruits were produced on the same trees in 1989, 1990, 1991, 1992, 1994 and
1996. One or two ﬂowers per spur were hand pollinated and the remaining ﬂowers

removed.

23

 

WWII. To establish when ﬂowering was inhibited by seeds, ﬁ'uits from
pollinated and non-pollinated ﬂowers were removed 28, 35, 56 and 147 days after full
bloom (DAFB) in 1988; 10, 27, 84 and 149 DAFB in 1989; 32, 68 and 138 DAFB in
1990 and 15, 21, 28, 35, 49, 70,110,138,168 DAFB in 1991, and the fruits and spurs
labeled. The fruits were weighed and their seeds counted. Bourse shoot length and
ﬂowering status were recorded the following spring. When more than one fruit was
borne on a spur, the fruit mass and seed number were totaled for each spur. Because
'Spencer Seedless' frequently produces two bourse shoots per spur, one spur can
produce two ﬂowering or non-ﬂowering terminal buds or one of each. These shoots
were counted independently when tabulating data.

In late October,

 

1994, a detailed observation of spur characteristics relative to return ﬂowering was
undertaken to identify sources of variation in ﬂowering response. Portions of the same
trees were hand pollinated in May 1994 as described previously. Twenty-ﬁve branches
with varying crop loads were selected and their fruit number and base diameters
recorded. Individual bearing spurs were then tagged and labeled. The fruit-bearing
spurs on the remainder of the trees were also labeled. The fruits were removed, labeled
as to source, weighed, and their seeds counted. In the spring of 1995 the length of the
bourse shoots was measured and the type of terminal bud (ﬂowering or vegetative)
recorded. Spurs were then grouped into classes based on various parameters [seed

number, shoot length, total fruit mass borne, crop density (no. of ﬁ'uits per cm2 branch

24

cross-sectional area = BCSA)] and the percentages of the terminal buds within these
classes that were ﬂoral were determined. Seed number classes were selected by
observation of numbers of bourse shoots with a given number of seeds, and apparent
break points in percentage ﬂowering as seed number increased from 0 to over 15.
Regression analysis was performed on seed number and percentage ﬂowering where the
relationship appeared linear. In 1996-97 this experiment was repeated except that the
crop density was not determined.
Results

ta. 0 u. ‘un -_. “r. on ..t~' no ‘I‘II ... u. n... u to .‘
199991199, In all years seedless ﬁ'uits of 'Spencer Seedless' did not inhibit ﬂowering
even when left attached to the spurs until harvest (Figure 1). In contrast, seeded fruits
consistently inhibited ﬂowering. The degree of inhibition varied with year, seeds being
most inhibitory in 1990. Little inhibition occurred if fruits were removed 0-20 DAF B,
but inhibition increased as defruiting was delayed. Late ﬁ'uit removal in 1992 resulted
in less inhibition than mid-season removal. Overall among the labeled spurs, seeded
fruits always reduced percentage ﬂowering but frequently 40 to 50 % still ﬂowered,
demonstrating great heterogeneity in the spur population.

Following treatment in 1989 and 1990, bourse shoots on spurs bearing fruits
containing more than six seeds were mostly vegetative, those on spurs bearing fruits
with less than six seeds mostly ﬂoral, regardless of shoot length (Figure 2). In 1991 and
1992 the exceptions were more numerous, and ﬂowering was not strongly correlated

with shoot length alone. Spurs with few seeds ﬂowered regardless of shoot length,

25

 

100

 

——o
3 o Seedless
°v 50 o Seeded
1989 1990

 

 

o——-o~ A m

1991

SPURS FLQWERING
8o

0'!
O

1 992

l l l 1 14 l j 14 A l I l

 

 

 

.1 131.11.].. .1.
0 50 100 1 50 0 50 1 00 1 50
TIME OF DEFRUITING (DAFB)
Figure 1. Effects of seeds and time of Mt removal the previous year on ﬂowering of
terminal buds of ‘Spencer Seedless’ apple bomse shoots. Each point represents from 6
to 45 bourse shoots. Year indicates time of fruit removal.

26

 

 

 

 

 

50 o Flowering 1939 1990
oVegetative ,
4O 5 .
.o

A 30 o :
g 20 ° 3 o
v 0 O ‘ O .
as w ' : 0 0
g 0 QMCDB 6 8° 06808 00 °
LU
-‘ 50 - 1991 . 1992
'5 4O 00
CI> 30 ' .
(D o . . 0 °

20 . 3' 0

10 r, ’0 '

0 o
0 8080 00 0 §' 00' 8
a l 4 l a I n l a l 1 J L l n a l 1 l n l a
O 6 12 18 24 3O 36 42 48 0 6 12 18 24 3O 36 42 48
SEED NUMBER PER SPUR

Figure 2. Effects of shoot length and seed number the previous year on ﬂowering of
terminal buds of bourse shoots of ‘Spencer Seedless’ apple. All ﬁ'uits were harvested at
maturity. Each point represents one bourse shoot.

27

while those with many seeds ﬂowered less frequently. In 1989 - 1991, few of the spurs
bearing seeded fruits had bourse shoots longer than 10 cm and very few ﬂowered. In
contrast, the 1992 sample of spurs bearing seeded fruits had few shoots shorter than 10
cm and ﬂowering was more frequent (Figure 2).

At harvest fruit mass per spur varied from 0.05 to 0.5 kg for both ﬂowering and
vegetative shoots and generally increased with seed number. Seeded ﬁ'uits were
generally larger than seedless fruits, but with multiple fruits per spur the total fruit mass
per spur was comparable for both ﬂoral and vegetative shoots (Figure 3). Spurs bearing
fruits with less than ﬁve seeds were ﬂoral and those bearing ﬁ'uits with ﬁve or more
seeds were vegetative with few exceptions, regardless of fruit mass.

When seed

 

number alone was plotted against ﬂowering, percent ﬂowering decreased from near 90
to 40 (1994) or 10 (1996) percent as seed number increased from 0 to more than 15 per
spur (Figure 4). Nearly 40 % of spurs that bore fruits with more than 15 seeds still
ﬂowered in 1994, similar to observations from previous years. The fact that near 10
percent of the spurs that bore no seeds did not ﬂower is also noteworthy. When the
spurs in seed-number classes were further separated by shoot length, large differences
were observed (Figure 5). Shoots less than 2 mm long showed little evidence of leaf
scars and were essentially buds that formed directly on the bourse. None of these
ﬂowered. Shoots longer than 10 mm almost always ﬂowered (11 of 333 were
vegetative in 1994) regardless of seed number. For shoots from 2 to 4 mm long percent

ﬂowering decreased sharply as seed number increased. In shoots 5 to 9 mm long

28

ﬂowering

 

. Flowering. n=50
O Vegetative, n=36

0.51.

010.4; 0

{gr - 00.

_ O
£1303... O
10.2

8 9

‘0 I

O

0
Do
000
o

O

1 989

Flowering, n=2
Vegetative, n=1 9

 

a: .

1.1.1 0 1

O- .

(D 05. Flowenng, n=40
. Vegetative. n=17

g 0'4?

 

1991

. 00
0. o
‘0
a" .
. O

 

Flowering, n=24
Vegetative, n=10

111111111L

 

.1.
18

24

.11.1.1.1
30 0 61218

SEED NUMBER PER SPUR

2430364248

Figure 3. Relationship between total seed number and total weight of fruit per spur at
harvest in 1989, 1990, 1991, and 1992 for 'Spencer Seedless' apple spurs that were
ﬂowering or vegetative the following season. All fruits were harvested at maturity.
Only ﬁ'uits from hand-pollinated branches were weighed in 1990. Each point represents

one bourse shoot.

29

 

 

"100 n = 482

on
O

O)
O

.5
O

N
O

 

n=12

BOURSE SHOOTS FLOWERING (%

O

 

 

l l l I
0 1-4 5-9 10-15 >15
SEED NUMBER PER SPUR

Figure 4. The relationship between seed number per spur and percentage of 'Spencer
Seedless' apple bourse buds that ﬂowered the following growing season in 1994 (y =

96.1 - 15.2x, R2 = 0.97) and 1996 (y = 88.2 — 19.6x, R2 = 0.96). Seed number classes

are based on observation of apparent break points and sample size.

30

 

100

FLOWERING (%)
-b O) on
O O O

N
O

100

FLOWERING (%)
-h 0) on
O O O

N
0

Figure 5. The relationship between seed number and shoot length of bourse shoots of
'Spencer Seedless' apple in 1994 and 1996 vs. percentage of bourse buds that ﬂowered

 

 

   
 

0 <2 mm (11 =18)
o 2-4 mm (11 = 306)
I 5-9 mm (n = 289)
1: >10 mm (n = 333)

 

 

 

  

 

 
 
  

1 996

0 <2 mm (n=38)

o 2-4 mm (n= 182)
I 5-9 mm (n=92)
13 >10 mm (n=48)

 

 

f C O 3

J 1 l l

0 1-4 5-9 >10
SEED NUMBER PER SPUR

the following growing season. Seed number and shoot length classes were based on
observation of apparent break points and sample size.

31

 

decreased as seed number increased, but response was less than in shorter shoots.

When bearing spurs were sorted by total fruit mass per spur, the number of spurs
in classes less than 80 g or more than 240 g were insufﬁcient to determine percent
ﬂowering with conﬁdence (Figure 6). Large sample sizes are required for binomial
data, such as ﬂowering status of buds, to calculate conﬁdence limits at P = 0.95. For
calculations to be valid, based on an approximation of a normal distribution, a minimum
of 30 spurs per class is required for 50 % ﬂowering, whereas the sample size required
increases to n = 1400 at 95% ﬂowering. If tables based directly on the binomial
distribution are used, 11 = 300 for each class is required for a percentage of 96% (Steel
and Torrie, 1980). Still larger values for n will narrow the conﬁdence limit. Data were
available for over 1200 spurs in 1994-95, providing a much larger population than the
360 spurs studied in 1996-97. The similar distributions provide conﬁdence in the
results. When spurs were classiﬁed based on the total ﬁ'uit mass borne per spur, fruit
mass had little effect on percentage ﬂowering, whereas seeds clearly inhibited ﬂowering
(Figure 7). When classiﬁed by total seed number borne per spur, less than ﬁve seeds
were not inhibitory regardless of whether the spurs bore fruit mass above the median or
below (Figure 8). When ﬁve or more seeds were present, higher fruit mass was
associated with increased ﬂowering in 1994 but the reverse was true in 1996.

As crop density increased, ﬂowering decreased (Figure 9), with maximum effect
when crop density exceeded four fruits/cm2 BC SA. Crop density greater than four
(maximum was 6.0) alone was not sufficient to prevent ﬂowering by seedless fruits, but

the inhibitory effect of seeded fruits was greatest when the crop density was high.

32

 

20° POPULATION DISTRIBUTION

w 160 Spencer Seedless

m I - _

a ‘ Midlan - 160 D 1996
l

w I 1994

'63 120

n:

3 so

2

D

Z

&
o

    

 

 

 

W.HII

I80100120140160180200220240260280300320340360380400420

 

FRUIT MASS PER SPUR (20 9 classes)

Figure 6. Distribution of bearing spurs of ‘Spencer Seedless’ apple by total ﬁ'uit mass
borne per spur in 1994 and 1996. No spurs bore less than 40 g. offruit. 40 g = 40 to 59
g;60g=60to79g, etc.

33

 

100 'Spencer Seedless'

 

 

.
O
0:
u.1 00
C3)
.1 70
: Seedless P u
E 60 o 1994
o o 1996 .
E 50 Seeded
“- n 1994 0 .
4° :1 1996 _ '3
«00 100+ 120+ 140+ 160+ 100+
<140
FRUIT MASS PER SPUR (9)

Figure 7. Effect of total ‘Spencer Seedless’ apple fruit mass (g) borne per spur on
percentage of spurs that ﬂowered the following year, in 1994 and 1996. Spurs beeﬁng
seedless and seeded fruits were ranked by total fruit mass (20 g increments). Classes
were combined when necessary to give sufﬁcient numbers to determine P = 0.95
conﬁdence limits. Number of spurs per class varied from 49 to 337, as required to
calculate conﬁdence limits for binomial data. Classes represent spurs with fruit mass
from the mass given to the next class above for a given treatment, except for the lowest
class, which represents all ﬁ'uits less than the value given. Insufﬁcient spurs bore
seedless ﬁ'uit in 1996 to classify, so mean fruit mass of all seedless spurs is reported.

34

 

100

 

 

80 T \ \.
g \\ .
o 60
.2.
a: .
UJ
; 40 _
9 o 1994, <160.3 g (n = 608)
“- 20 o 1994, >160.3 g (n = 608) -
n 1996, <156.0 g (n = 180)
n 1996, >156.0 g (n = 180)
0
l l I I
o 14 5-9 >10
SEED NUMBER PER SPUR

Figure 8. The relationship between seed number per spur in 'Spencer Seedless' apple
and percentage of bourse buds that ﬂowered the following growing season on spurs
bearing less than or greater than the median fruit mass. Seed number classes are based
on observation of apparent break points and sample size.

35

 

 

 

 

 

80
$5
(.9 60
Z
a:
E 40
8
U- 20 0 <2.5 Fruits I BCSA (n = 303)
o 2.5-4.0 Fruits / BCSA (n =294)
- >4.0 Fruits / BCSA (n = 341)
0
1 l ' l
o 1-4 5-9 >10
SEED NUMBER PER SPUR

Figure 9. The relationship between seed number and crop density per cm2 branch
cross-sectional area in 'Spencer Seedless' apple in 1994 vs. percentage of bourse buds
that ﬂowered the following growing season. Seed number and crop density classes are
based on observation of apparent break points and sample size.

36

Discussion

In this study, seedless fruits of 'Spencer Seedless' did not inhibit ﬂowering,
regardless of total fruit mass per spur or shoot length. Among spurs bearing seeded
fruit, as the aggregate number of seeds per spur increased, ﬂowering decreased in the
terminal buds of the subtending bourse shoots less than 10 mm long. However,
unexpectedly, seeds had little inhibitory effect in bourse shoots longer than 10 mm.
These results only partially agree with those of Chan and Cain (1967), who found that
one to ﬁve seeds inhibited ﬂowering 65 to 98% under orchard conditions. Their work
was done in New York State under climatic conditions similar to those in E. Lansing,
but they did not provide data on shoot length, and the tree that they used may have been
less vigorous than those in E. Lansing. However, Chan and Cain (1967) did report that
the effects of seeds were less pronounced in greenhouse experiments, where, among
spurs bearing seeded fruits, all bourse shoots longer than 50 mm, and 33 to 35% of
those shorter than 50 mm, ﬂowered. Only 72-85% of spurs bearing seedless fruits
ﬂowered under these conditions. The authors mentioned that “the axillary buds of the
young spurs were often forced out and grew into axillary shoots in the same season,
contrary to the ﬁeld experiments in which the axillary bud remained in bud form. The
longer the axillary shoots grew the more of them ﬂowered.” Whether they are referring
to axillary or terminal buds of bourse shoots, which are axillary to the ﬂower cluster, is
not clear.

Similar results to these have been found with seeded and seedless ‘Bartlett’ pear.

Huet (1972), working in France, reported a positive relationship between leaf area and

37

ﬂowering. Shoots on spurs bearing seeded fruits required a larger leaf surface to ﬂower
than did those on spurs bearing seedless fruits. The data of Griggs, et al. (1970) in
California are more difﬁcult to interpret. Flower buds were produced on 69 % of spurs
bearing seeded fruits, compared with 94 % of spurs bearing seedless ﬁ'uits, if fruits
were left attached 31 DAF B. However, seeded and seedless fruits were equally
inhibitory if they remained attached longer, and the inhibition was similar to that in
spurs bearing seeded fruits at 31 DAF B. They found no relationship between ﬂowering
and axillary (bourse) shoot length, but presented data for only 32 spurs per category
(seeded/seedless x fruit no. per spur). These were divided among shoot length classes,
and most shoots were less than 25 mm long. How they calculated percentages or
analyzed the data is not apparent.

Presumably extension of the shoot axis must cease before a terminal ﬂower bud
can form. The ﬁrst indication of growth cessation is the abscission of the lamina and
petiole of the eighth leaf primordium from the apical meristem (Abbott, 1970). One
way in which long shoots differ from shorter ones may be the time of cessation of
growth. The time of growth cessation relative to the timing of defoliation and deﬁ'uiting
treatments and ﬂower induction has seldom been reported. This is understandable,
given that apple ﬂower induction is not sensitive to photoperiod (Gorter, 1955). In
temperate climates with adequate chilling, all terminals begin growth and development
simultaneously. In many cultivars, including ‘Spencer Seedless’, bourse shoots on
spurs typically form two to four leaves with very short intemodes and cease growth

relatively synchronously within a few days after anthesis. Crow (1920) observed that

38

most ‘Oldenberg’ and ‘Wealthy’ apple spurs grew for only four to ten days before
setting terminal buds, which is about the time of fruit set. Terminal shoots with
elongated intemodes can bear 3 to 20 or more leaves, and therefore cessation of growth
is later and ﬂower induction may occur over an extended period of time. Hipps (1992)
reported that fertigation and irrigation increased shoot length by accelerating, rather
than prolonging, growth in newly planted ‘Queen Cox’ trees on M.9 rootstock, so that
shoots of various lengths set terminal buds at about the same time. He did not report
how many days after bloom this occurred. A better understanding of the relationships
between shoot length and the period of shoot elongation might help in predicting when
ﬂower induction occurs and which factors are causal vs. which are parallel responses to
environmental or internal signals.

Luckwill (1970) proposed that gibberellins in developing seeds inhibited
ﬂowering. This hypothesis was based on the inhibitory effects of applied gibberellins
on ﬂowering of apple, and the high concentrations of endogenous gibberellins in the
seeds. However, he reported peak production of seed gibberellins in ‘Emneth Early’ at
9 weeks after full bloom , which coincides with rapid embryo development rather than
with the period of ﬂower induction 3 weeks earlier. Similar discrepancies with regard
to timing of these processes have been reported by others (e.g., Dennis, 1976).
However, timing and rate of transport would be more important than relative
concentrations in determining ﬂowering response (Bangerth, 1997). Grochowska
(1968), for example, showed that the rate of diﬁusion of auxin from immature apple

seeds following excision from the Suﬁ differed markedly from the quantities that could

39

be extracted from similar seeds. Longer shoots may provide greater spatial separation
of the terminal bud from the developing seeds, with greater opportunity for dilution or
metabolism of the gibberellins along the way. To date, however, studies of transport of
gibberellins from seed to bourse shoot have provided little support for their role in
inhibiting ﬂowering (Ban, 1996; Green, 1987; Stephan, et al., 1997).

Several hypotheses, not involving gibberellins, could explain the relationship
between seeds and shoot length in affecting ﬂowering. Two of these relate to a
hypothetical ﬂowering-promoting hormone (“ﬂorigen”), and to the promotive effect of
carbohydrates. The ﬁrst assumes that seeds have priority over buds for “ﬂorigen”.
Another explanation involving “ﬂorigen”is that the greater leaf area on long shoots may
produce an abundance of “ﬂorigen” that overrides the inhibitory effect of seeds.
Molecular biology using model species may be helpful in identifying the reason for the
effects of leaves on ﬂower induction. This research should be performed on adult tissue
of a polycarpic plant to separate ﬂower induction responses from other maturation and
senescence processes and to increase the likelihood that the results will be applicable to
ﬁ'uit trees. If “ﬂorigen” could be identiﬁed, preferential transport to seeded ﬁ'uits, or
increasing abundance with increasing leaf number, could be evaluated to test these
hypotheses. Ethylene stimulates ﬂowering in bromeliads. If it plays a role in ﬂowering
of apple, differences between seeded and seedless ﬁ'uit in ACC content or transport to
bourse buds could be determined. Gibberellins may inhibit this process.

Long bourse shoots have a higher leaf number and more leaf area than short

ones, and therefore may furnish more carbohydrate to enhance ﬂower induction.

40

However, assuming that ﬁ'uit mass is an indicator of carbohydrate supply to a spur,
carbohydrate cannot be the limiting factor for ﬂower induction in ‘Spencer Seedless’,
since spurs with seedless fruits ﬂower over a wide range of fruit mass per spur whereas
spurs bearing seeded fruit of similar mass fail to ﬂower. Stutte and Martin (1986b) used
light intensity and CO, enrichment to create varying carbohydrate levels in bearing and
nonbearing olive trees, and found that fruits were inhibitory to ﬂowering regardless of
the carbohydrate concentration. They also reported that killing the seed prior to
endocarp scleriﬁcation overcame alternate bearing without altering dry matter
accumulation by the fruit (Stutte and Martin, 1986a). A similar experiment using apple
would conﬁrm the independence of ﬂower induction from carbohydrate production.
While seeds have an inhibitory effect on ﬂower induction of ‘Spencer Seedless’
apple, there appears to be an interacting effect of bourse shoot length and possibly
cropload. Because of lack of sufﬁcient numbers of trees, no replication was possible in
these experiments and therefore statistical analysis was limited to regression based on
spur data classiﬁed by various criteria. The main source of conﬁdence in these results is
the repeatability over several years. Large samples (> 100) increase one’s conﬁdence in
discrete data where proportions (ﬂowering vs. nonﬂowering) are the data of interest.
The relationship of seeds, shoots, and cr0p to ﬂower induction should be explored
further in commercial cultivars, and with seeded and seedless fruits where adequate
numbers of small uniform trees are available to allow better replication and analysis of
variance. Cultivars should be selected to represent the diversity of morphological and

physiological traits that inﬂuence ﬂowering in apple. Regardless of the mechanism

41

whereby shoot length inﬂuences ﬂowering, it plays a more dominant role than seeds for
shoots less than 2 mm or longer than 10 mm in ‘Spencer Seedless’. For spurs of
intermediate length seeds play a dominant role. Study of the mechanism whereby seeds
inhibit ﬂowering in 'Spencer Seedless' may need to be conﬁned to spurs producing

shoots 2 to 10 mm long.

Literature Cited

Abbott, D.L. 1970. The role of budscales in the morphogenesis and dormancy of the
apple fruit bud. p. 65-82 In: L.C. Luckwill and C.V.Cutting (eds.). Physiology
of Tree Crops. Academic Press, New York.

Acquaah, G., J .W. Saunders and L.C.Ewert. 1992. Homeotic ﬂoral mutations. Plant
Breeding Rev. 9:63-99.

Ban, J. 1996. Gibberelin metabolism and apple ﬂowering. PhD Dissertation, Michigan
State Univ.

Bangerth, PK. 1997. Can regulatory mechanism in fruit growth and development be
elucidated through the study of endogenous hormone concentrations? Acta
Hort. 463:77-87.

Brase, K.D. 1937. The vascular anatomy of the ﬂower of Mama 9991951199 Borkh. f.
Ma van Eseltine. MS. Thesis, Cornell University.

Chan, B.G. and J .C. Cain. 1967. The effect of seed formation on subsequent ﬂowering.
Proc. Amer. Soc. Hort. Sci. 91:63-68.

Coen, ES, 1991. The role of homeotic genes in ﬂower development and evolution.
Annu. Rev. Plant Physiol. Plant Mol. Biol. 42:241-279.

Crow, J .W. 1920. Biennial fruit bearing in the apple. Proc. Amer. Soc. Hort. Sci.
17:52-54.

Dennis, F.G. Jr. 1976. Gibberellin-like substances in apple seeds and fruit ﬂesh. J.
Aner. Soc. Hort. Sci. 101 :629—633.

42

Edwards, GR. 1987. Producing temperate-zone fruit at low latitudes: Avoiding rest and
the chilling requirement. HortScience 22: 1236-1240.

Gorter, C.J., 1955. Photoperiodism of ﬂowering in apple trees. Rep. XIVth Int. Hort.
Cong., p 351-354.

Green, JR. 1987. The hormonal control of biennial bearing in cider apples. PhD
Dissertation, Univ. of Bristol.

Griggs, W.H., G.C. Martin, and B.T. Iwakiri. 1970. The effect of seedless versus seeded
ﬁ'uit development on ﬂower bud formation in pear. J. Amer. Soc. Hort. Sci.
95:243-248.

Grochowska, M.J. 1968. Translocation of indole-3-acetic acid-2-“C injected into seeds
of ﬁve-week-old apple fruits. Bul. Acad. Polon. Sci., Ser. Biol. 16:577-580.

Harley, C.P., J .R. Magness, M.P. Masure, L.A. Fletcher and ES. Degrnan. 1942.
Investigations on the cause and control of biennial bearing in apple trees.
U.S.D.A. Tech. Bul. 792.

Hipps, NA. 1992. Fertigation of newly planted ‘Queen Cox’/M9 apple trees:
Establishment, early growth and precocity of cropping. J. Hort. Sci. 67:25-32.

Huet, J. 1972. Etude des effe‘ts des feuilles et des fruits sur l'induction ﬂorale des
brachyblastes du Poirier. Physiol. Vég. 10:529-545.

Huet, J. 1973. Floral initiation in pear trees. Acta Hort. 34:193-198.
Luckwill, LC. 1970. The control of growth and fruitfulness of apple trees. p. 237-254
In: L.C. Luckwill and C.V.Cutting (eds.). Physiology of Tree Crops. Academic

Press, New York.

Salisbury, RB. 1982. Photoperiodism. Hort. Rev. 4:66-105.

Steel, R.G.D. and J.H. Torrie. 1980. Principles and Procedures of Statistics. 2nd
Edition. McGraw-Hill Book Company, Inc., New York.

Stephan, M., F. Bangerth and G. Schneider. 1997. Transport and metabolism of the
gibberellins A,, A,, and A, after application to developing apple fruits of MaIus
domestica cv. Jonagold. Acta Hort. 463:113-119.

Stutte, G.W. and GO Martin. 1986a. Effect of killing the seed on return bloom of
olive. Scientia Hort. 29: 107-1 13.

43

Stutte, G.W. and GO. Martin. 1986b. Effect of light intensity and carbohydrate
reserves on ﬂowering of olive. J. Amer. Soc. Hort. Sci. 111:27-31.

Tobutt, KR. 1994. Combining apetalous parthenocarpy with columnar growth habit in
apple. Euphytica 77:51-54.

SECTION 2. EFFECTS OF FRUIT REMOVAL, CROP DENSITY, AND
TRUNK GIRDLING ON FLOWER FORMATION IN
'PAULARED' APPLE: GENERAL VS. LOCAL EFFECTS OF
FRUITS.

Abstract

The purpose of this study was to determine the general vs. local effect of
developing apple fruits on ﬂower induction. Whole ‘Paulared’ were defruited at several
times from 20 to 104 days after full bloom (DAF B) in 1991 and 1992 to determine the
critical time at which fruit removal could no longer stimulate ﬂowering. Twenty
fruiting terminals were tagged on each tree to determine ﬂowering response the
following season. Nearly 100 percent of tagged terminals ﬂowered if fruits were
removed before 60 DAF B, but ﬂowering declined sharply thereafter. To determine the
effect of crop density, fruit number per spur and girdling, the crop on whole trees was

estimated as light, medium or heavy, and then adjusted by hand on 14 June 1996 (24

DAF B) to target crop densities of three, six or nine fruits per cm2 trunk cross-sectional

area (T CSA). On each tree, half of the scaffold limbs were thinned to two fruits per

spur and half to one fruit per spur. Half of the trees with each of the target crop
densities were girdled by removing a strip of bark 5 mm wide 30 cm above the soil level

and applying grafting compound to the wound. Thirty spurs bearing one huh and 30

spurs bearing two fruits around the periphery of each tree were marked and the fruit

mass, seed number, bourse shoot length and ﬂowering response recorded. Thirty new
vegetative shoots per tree were also randomly selected, measured and ﬂowering

response recorded. Flowering was moderate to heavy in all cases, and girdling did not

increase ﬂowering signiﬁcantly. Nearly all spurs ﬂowered on trees with light crops, and

45

approximately 70 percent of the spurs on trees with medium or heavy crops. Flowering
was signiﬁcantly greater in non-bearing shoots than in terminals bearing one or two
fruits. More lateral ﬂowers formed on non-fruiting shoots on trees with light or medium
crops than on trees with heavy crops, or on spurs bearing ﬁ'uit at any cropping level. As
crop density increased, ﬂowering declined in non-girdled, but not in girdled, trees. The
limited effect of fruit thinning on ﬂowering in this biennial cultivar was unexpected, and

suggests that weather conditions or other uncontrolled factors favored ﬂowering over

treatment effects.

46

Introduction

Developing fruits can inhibit the formation of ﬂoral buds in apple, leading to
alternate hearing, but this effect varies greatly with cultivar. Some cultivars, e.g.,
‘McIntosh’, are annual bearers, while others, e.g., ‘Paulared’, are biennial. Auchter
(1919) and Crow (1920) suggested that alternate bearing results from too few “resting”
(non-bearing) spurs in a given year, implying that the inhibitory effect of fruits is local
and does not affect buds beyond the closest bourse shoot. Auchter and Schrader (1923)
observed that a high percentage of spurs were continuously vegetative on ‘York
Imperial’ trees. McKee and Forshey (1966) found that 22.8 percent of ﬁ'uiting spurs on
'Mclntosh' trees ﬂowered the following season (repeat bloom) when cropping was
moderate, but only 7.3 percent did so following a heavy crop. This suggests that
alternate bearing is more complex than simply biennial bearing by individual spurs.
Chemical thinning is standard practice to promote annual bearing, but results vary from
year to year. Ideal crop density for annual bearing is not well deﬁned and varies with
cultivar and growing conditions.

Why developing fruits inhibit ﬂowering is unknown. However, Chan and Cain
(1967) demonstrated that the seeds were responsible for much of the inhibitory effect of
the fruits on ﬂower induction within individual spurs of 'Spencer Seedless', a
facultatively parthenocarpic apple cultivar. Huet (1972, 1973) conﬁrmed these results
with seedless fruits of ‘Bartlett’ pear, and observed that ﬂowering of spurs bearing
seeded fruit increased as leaf area increased. Neilsen and Dennis (1997) found that

bearing spurs of 'Spencer Seedless' may ﬂower repeatedly if the number of seeds in the

47

developing fruits is less than 5 or bourse shoot length is greater than 10 mm.
Furthermore, the inhibitory effect of seeds declines as bourse shoot length increases. A
given seed number was more inhibitory at high crop density, suggesting that there is a
general inhibitory effect of a crop in addition to the local effect of developing seeds.
'Paulared' is an apple cultivar that produces long bourse shoots with terminal
ﬂower buds and forms few spurs, thus it is characterized as a "terminal bearer" with a
"weeping growth habit". It also has a strong tendency toward alternate bearing. The
fruits mature in late August to early September in Michigan, which makes it one of the
earliest of the major commercial cultivars. These experiments were conducted to
determine the effects of bourse shoot length and crop load on ﬂower induction in a
cultivar that bears seeded ﬁ'uit. Because ﬂowering was expected to be very limited

when crop load was heavy, half of the trees were girdled to encourage ﬂowering.

Materials and Methods

WWW To determine the
inﬂuence of the time of fruit removal on ﬂower induction, three replicate 'Paulared' trees
planted at the Clarksville Horticultural Experiment Station, Clarksville, Mich., in 1980
were entirely defruited on each of four or ﬁve dates from 20 to 104 days after full bloom
(DAFB) in 1991 and 1992. On each tree twenty terminal bourse shoots were tagged and
evaluated for shoot length and return bloom.

.o i' . o u- 1'1' 11.. 1.111; or 1. =11 '0.“ n the

19199999, To determine the inﬂuence of crop density, fruit number per spur and

48

girdling on ﬂower induction, bloom density on 21 May 1996 was visually graded on a
scale from 0 (no ﬂower clusters) to 10 (no visible vegetative terminals). Trunk
circumferences were measured , and their cross-sectional areas (T CSA) calculated.
Trees with light bloom were left untreated. Six replicate trees (18 total) with moderate
to heavy bloom were assigned to one of three target crop density treatments (3, 6 or 9
ﬁ'uits per cm2 TCSA) and the cropload adjusted by selecting scaffold limbs representing
approximately half of the bearing surface on each tree and on 14 June removing fruits
from the fruiting terminals to leave two fruit per fruiting terminal (ﬁ'uiting branch with
bourse shoots up to 570 mm); the remaining scaffolds were thinned to one fruit per
fruiting terminal. Fruits that were removed were collected, counted, and the total mass
and average seed number determined. Three of the six replicate trees in each crop
density group were girdled on 14 June by removing a 5 mm wide band of bark 30 cm
above the soil level, after which the wound was covered with grafting compound.

Total fruit remaining per tree was estimated by counting selected branches and if
too many fruits remained to reach the target crop density, additional fruiting terminals
were deﬁ'uited on 26 June (36 DAF B) to result in a uniformly distributed crop
throughout the tree; half of which was borne two fruits per spur, half was one ﬁlm per
spur. Final mean crop densities were 3.4, 6.1 and 8.2 for the respective target crop
density classes. At harvest, fruits and spurs (30 per tree bearing 2 or 1 fruits,
respectively) were labeled, and the fruits were weighed and their seeds counted.
Nonﬂowering shoots were also labeled (30 per tree) to represent zero fruits. The entire

crop per tree was also counted and weighed and the crop density and yield efficiency

49

calculated. Data from the treated trees were analyzed using a split-plot design with
girdling and crop load as main plots and fruit number per spur as the split plot. Fruiting
spurs on additional trees with light crops were also labeled, their fruits weighed and
seeds counted to allow regression analysis of responses to crop density from very low to
high levels.
Results

WWW When whole trees
were defruited within 60 DAF B in 1991, no inhibition of ﬂowering occurred, but
delaying fruit removal to 100 DAF B reduced ﬂowering by 60 percent (Figure 1).
Similar data were obtained in 1992, except that ﬂowering was also severely inhibited
when defruiting was delayed to 73 or 103 DAF B (Figure 1). In 1992 several of the
terminal buds had died and a greater percentage of those that survived were vegetative
at the earliest defruiting date. In addition to terminal ﬂowers, ﬂowers developed from
lateral buds on the bourse shoots when fruits were removed early. This rarely occurs in
'Paulared' trees bearing moderate crops. Only terminals were selected for recording
ﬂowering, and ﬂowering did not appear to be related to shoot length, which ranged from
60 to 300 mm per tree at the time of Suﬁ removal (Table l).

.11 i'.o 011:1; 11-1-11»; or 1.. 211'.01'11,io.'
19199999, Flowering in 1997 was unexpectedly heavy regardless of treatment, with 70
to 98 percent of the terminals ﬂowering (Table 2). Girdling appeared to increase
ﬂowering, but the difference was statistically signiﬁcant only at P = 0.17. Although

values for trees with lowest crop density were greater than those for trees with heavier

50

 

 

FLOWERING (%)
S
T v

N
0|

U T U l I U U

 

 

 

O
T'

I l l l I I

50 100
TIME (DAFB)

Figure 1. Effect of time of defruiting entire ‘Paulared’ apple trees in 1991 and 1992 on
percentage of terminal buds of ﬁ'uiting spurs ﬂowering the following year. Shoots with
dead terminal buds were excluded. Bars represent the standard errors of the means for
three replicate trees.

51

Table 1. Effect of time of defruiting entire ‘Paulared’ apple trees in 1991 and 1992 on
mean shoot length ( 1 standard error) in the same growing season and number of lateral
ﬂowers per shoot the following year.

 

 

 

 

 

   

Time of deﬁ'uiting (DAF B) Shoot length (mm) No. of lateral ﬂowersz

_ ___ 1992 7 __ _ 1991 _ 1992 _ _ 1992 - ., 993
20 24 298 i 38 109 :h 24 5.28 a 1.80
35 37 190 3: 37 138 :t 50 4.57 a 2.21
58 60 212 :t 24 62 :1: 04 4.03 a 0.20

- 74 - 164 :1: 33 - 0.67

97 104 238 i 16 173 :1: 13 0.42 b 0.00
P= 0.24 0.20 0.019 0.063

 

’ Count data transformed by./ y + 0.5 before analysis.

crops, the differences were again non-signiﬁcant. Among trees with moderate and
heavy crop densities (6.1 and 8.2 average fruits per TCSA, respectively) ﬂowering of
bourse shoots on spurs bearing no fruits was significantly greater than that on those
bearing one or two fruits (Table 2), but the difference was small. Further, there were no
signiﬁcant interactions between crop density, girdling or fruit number per spur and
ﬂowering of terminal buds. However, the interaction between crop density and number
of fruits borne per spur on the number of lateral ﬂower buds per bourse shoot was
signiﬁcant (Figure 2); lateral ﬂowering was greater on non-bearing shoots, but only at
low crop densities. As would be expected, seed number and fruit mass per spur
increased with crop density and with number of fruits per spur, but were not affected by

girdling (Table 2).

52

Table 2. Main effects of girdling, crop density and number of fruits per spur in 1996 on
ﬂowering of ‘Paulared’ apple in 1997. Percentages calculated from 21 to 46 bourse
shoots per tree for each category. Three replicate trees were used for each treatment in a

 

 

split-plot design.
Main effect Percent Mean seed Mean fruit
terminals number per mass per
_ ﬂowering spur spur (g)
Girdling “
Girdled 89 6.9 128
Control 70 6.6 129
P = 0.17
Crop density class V
(ﬁ'uits/cm2 TCSA)
3.45:1:028 98 6.1 153
6.142t0.58 70 7.0 117
8.19:1:0.27 70 7.0 l 16
P = 0.12
Fruits per spur "
0 85 a‘“ - -
l 79 b 6.4 a 133 a
2 74 b 13.7 b 252 b

 

‘ Means for 9 trees x 3 treatments (fruits per spur) per tree.
Y Means for 6 trees x 3 treatments (ﬁuits per spur) per tree, mean :1: SE.

" Means for 18 trees.

" Mean separation within columns and factors by DMRT, P = 0.05.

53

 

.0 9 .o
«b O) on
I I I

LATERAL FLOWERS PER SHOOT
.0
N

 

l l l l l l A L l l A l l

O OFruits
O 1Fruit
I 2 Fruits

 

 

4 6 7

5
CROP DENSITY (Fruit no. / cmztcspi)

Figure 2. Effects of crop density and number of fruits borne on a spur on the average

number of lateral ﬂowers per bourse shoot.

54

 

Using regression analysis, all tagged spurs from all trees, including those with
very light crop density, were ranked based on crop density, fruit number per spur, and
shoot length. Only crop density signiﬁcantly affected the percentage of terminals that
ﬂowered (R2 = 0.32). The correlation was improved by excluding the trees that were
girdled (Figure 3). The three trees with exceptionally low ﬂowering (one girdled and
two non-girdled) were the smaller trees, suggesting less vigor than others.
Unexpectedly, ﬂowering was not correlated with number of fruits (or seeds) per spur, or
with bourse shoot length. The unusually high amount of return bloom in non-girdled
trees with medium to high crop densities warrants speciﬁc examination of individual
trees. Flowering response varied greatly among these six trees (Table 3), suggesting
that factors other than crop density or fruits (seeds) per spur had a strong effect on
ﬂowering. In two trees with nearly identical crop density (4.3-4.9 fruits/cm2 TCSA),
almost all of the buds ﬂowered in one tree regardless of the number of fruits borne on
the spurs, while in the other ﬂowering decreased as ﬁ'uit number per spur increased. An
even more dramatic difference was observed with three trees with crop density near 7.6
fruits/cm2 TC SA. Two were unaffected by fruit number per spur; one with nearly all
spurs ﬂowering and the other with none. The remaining tree exhibited reduced
ﬂowering as fruit number increased. This variation suggests unknown inﬂuences on

ﬂowering within this experiment.

55

 

.a
O
o

 

 

 

A
N
V 80
(I)
.1
<
Z 60
2
ES
1— 40
_.I
§ 0 Girdled trees
3 20 O Nongirdled trees 0
L1.
0 o

IRELLJIJIIIIIIIJALIIllllllllll'lllllllllllllll

O 1 2 3 4 5 6 7 8 9 1O
CROP DENSITY (fruit no./ cm’TCSA)

Figure 3. Effect of crop density on percentage of terminals ﬂowering (90 shoots per
tree) on girdled (y = 98.9 - 1.5x, R2 = 0.093, P = 0.34) and nongirdled (y = 105.7 - 7.0x,
R2 = 0.56, P = 0.000056)‘Paulared’ trees.

56

 

Table 3. Effect of medium (M) and high (H) crop density on ﬂower induction of non-
bearing shoots and spurs bearing 1 or 2 fruits on individual, non-girdled ‘Paulared’
apple trees.

 

 

Crop density Crop density Mean Flowering (%)z
class and (fruit no. / shoot length of shoots on spurs bearing:
replicate (tree) cm2 TC SA) (mm)

 

 

 

M-l 4.34 75 100.0 100.0 97.1

 

M-2 4.91 79 93.1 70.0 59.1
M-3 7.58 99 96.7 90.5 96.9
H-1 7.62 74 42.9 0.0 17.2
H-2 7.63 25 0.0 0.0 0.0
H-3 9.41 109 60.0 48.3 24.0
zn=30
Discussion

Based upon published data, 60 DAF B should be too late for ﬂower induction to
be affected appreciably by fruit removal, especially for a biennial cultivar such as
‘Paulared’. Hand thinning is considered to be effective in enhancing return bloom only
if performed before “June” drop (around 40 DAF B). Flowering response declines much
more rapidly in biennial cultivars, such as ‘Yellow Newtown’, than in annual ones, such
as ‘Jonathan’ (e.g., Harley, et al., 1942), and some early work, (e.g., Auchter, et al.,
1926; McCormick, 1933) suggested that in biennial trees even excessive ﬂowers were

capable of inhibiting ﬂowering in “on” years. However, I am aware of no data similar to

57

that obtained for ‘Paulared’ in 1992, when percentage ﬂowering in response to fruit
removal dropped from nearly 100 % to nearly 0 % within 2 weeks.

The response of ‘Spencer Seedless’ (Section 1) resembled that of ‘Jonathan’,
rather than that of ‘Paulared’. The time of ﬂower induction in ‘Paulared’ was well
synchronized within trees and across years, suggesting that this was not a chance
occurrence.

The earliest Chan and Cain (1967) removed seeded fruit of ‘Spencer Seedless’
was at 20 DAF B, when ﬂowering was already reduced to 50 percent. Further delay in
fruit removal decreased ﬂowering to 20-30 percent by 60 DAF B, with minor further
declines when seeded fruits remained until harvest. Harley et al., (1942) found steep
declines in percentage ﬂowering on ‘Yellow Newtown’ as defruiting was delayed, but
not with ‘Jonathan’, which exhibited a gradual decline with increasing delay. In the
results from Section 1 on ‘ Spencer Seedless’ both patterns are apparent, depending on
the year.

This variation among cultivars and experiments may be explained by variation
in the time the bourse shoots stopped growing. Spurs have been the primary focus of
other experiments to determine the time of ﬂower induction. All labeled branch
terminals of ‘Paulared’ in 1991 and 1992 had one or two long bourse shoots, as is
typical for this cultivar. Most of these had stopped growing by 60 DAF B, but the
precise date of cessation of growth was not recorded. Since ﬂower induction cannot
occur until shoot growth ceases, induction in these shoots may have been delayed

relative to that in spurs, which cease growth earlier. This assumes that seeds are

58

uniformly inhibitory over the entire time, which may not be true. The dramatic decrease
in percent ﬂowering that occurred when entire ‘Paulared’ trees were defruited later than
60 DAF B could be because overall crop density was heavier than on the 'Spencer
Seedless' trees, particularly if only seeded fruits are considered to have an inhibitory
effect. Alternatively, the difference in bearing habit (terminal bearing in ‘Paulared’ vs.
spur-bearing in ‘Spencer Seedless’) could be important.

Overall, ﬂowering was much greater than expected, given the crops produced;
even in nongirdled trees with high crop density nearly 50 percent of the spurs ﬂowered.
One variable that may have inﬂuenced ﬂowering was that the trees were not irrigated
until mid-August, and the soil surface was noticeably dry. How much drought stress
existed and what inﬂuence this had on ﬂower formation is unknown. Drought
conditions are considered by many to favor ﬂowering, but actual data are limited.

In contrast with the results with ‘Spencer Seedless’, seeded fruits of ‘Paulared’
did not exhibit a consistent local effect on ﬂowering, nor did shoot length appear to be
related to ﬂower formation; rather, crop density had the greatest inﬂuence, at least in
non-girdled trees. The few bearing spurs on trees with very light crops still ﬂowered,
including those with 3-5 fruits. Conversely, ﬂowering was frequently inhibited in non-
fruiting shoots on heavily cropping, non-girdled trees, indicating an overall inhibitory
effect of crop. Lakso (personal communication) found an interaction between cropload
and fruit number per spur on ‘Empire’ apple in New York. One ﬁ'uit per spur inhibited
ﬂowering only when cropload was light, whereas two fruits per spur were inhibitory

regardless of cropload.

59

Auchter (1919) recorded data for large numbers of individual spurs for several
cultivars to determine the relationships between growth, ﬂowering and ﬁ'uiting. In
‘Baldwin’, ‘Northern Spy’, ‘Wagner’, ‘Ben Davis’ and ‘Tompkin’s King’ few or no
spurs ﬂowered two years in succession, whereas in ‘Rome Beauty’, ‘Grimes Golden’,
‘Smokehouse’ and ‘Delicious’ 20 to 50 % of the spurs ﬂowered in two successive years;
however, the spurs that did so did not set fruit. When spurs of these “annual cultivars”
bore fruit, less than one percent ﬂowered the following year. The difference between
biennial and annual cultivars may be the degree to which overall crop inhibits ﬂower
induction, even on nonbearing spurs or shoots. Auchter (1919) further observed that
‘Rome Beauty’ and ‘Smokehouse’ bear terminally, and the side growth (long bourse
shoot) often produces a ﬂower bud and fruits the following year; thus these cultivars are
considered annual bearers. This condition also occurs with ‘Yellow Transparent’ and
‘Ben Davis’ when the trees are young. This is not the case with ‘Paulared’, which
exhibits similar terminal bearing habit, but is still biennial. Auchter (1919) also found
that fruit removal at June drop reduced ﬂowering the following year, except for 10
unusually long spurs of one ‘Delicious’ tree that produced terminal ﬂower buds. ’ This is '
similar to the pattern observed with ‘Spencer Seedless’ reported earlier.

Girdling generally overcame the inhibitory effect of heavy cropping, with 90
percent of the terminals ﬂowering. The one exceptional tree was notably low in vigor.
A local inhibitory effect of developing fruits on ﬂowering was apparent in some of the
‘Paulared’ trees used, but this was not related to crop density or other known factors.

‘Paulared’ appears to differ from many other biennial cultivars in that fruits

6O

appear to become inhibitory to ﬂowering very late in their development. This, plus its
terminal bearing habit, make it an interesting subject for further investigation. Further
work with ‘Paulared’ should focus on determining how the ﬂower induction period
varies within a given tree or cultivar and how overall cropping interacts with local
distribution of fruits to inﬂuence bourse shoots. Methods of altering seed number
mechanically or chemically may help determine the speciﬁc role of the fruit. The time
when shoots cease growth relative to treatments also should be noted. Comparison of
the biennial ‘Paulared’ with another terminal bearing cultivar with an annual bearing
habit, such as ‘Rome Beauty’ may help explain the difference in ﬂowering response.
Care should be taken to use uniform trees with ample replication and a minimum of

environmental variables.

Literature Cited
Auchter, EC. 1919. Some inﬂurences of thinning, pollination and fruit spur growth on
the yearly performance record of fruit spurs and on the size of fruit produced.
Proc. Amer. Soc. Hort. Sci. 16:118-131.

Auchter, EC. and AL. Schrader. 1923. Fruit spur growth and fruit bud production.
Proc. Amer. Soc. Hort. Sci. 20:127-144.

Auchter, E.C., A.L. Schrader, F .S. Lagasse and W.W. Aldrich. 1926. The effect of
shade on the growth, fruit bud formation and chemical composition of apple
trees. Proc. Amer. Soc. Hort. Sci. 23:368-382.

Chan, B.G. and J .C. Cain. 1967. The effect of seed formation on subsequent ﬂowering.
Proc. Amer. Soc. Hort. Sci. 91 :63-68.

Crow, J .W. 1920. Biennial fruit bearing in the apple tree. Proc. Amer. Soc. Hort. Sci.
17:52-54.

Harley, C.P., J .R. Magness, M.P. Masure, L.A. Fletcher and ES. Degrnan. 1942.

61

Investigations on the cause and control of biennial bearing of apple trees. US.
Department of Agriculture Tech. Bul. 792, 57 p.

Huet, J. 1972. Etude des effets des feuilles et des fruits sur l’induction ﬂorale des
brachyblastes du Poirier. Physiol. Vég. 10:529-545.

Huet, J. 1973. Floral initiation in pear trees. Acta Hort. 34:193-198.

McCormick, AC. 1933. Control of biennial bearing in apples. Proc. Amer. Soc. Hort.
Sci. 30:326-329.

McKee, M.W. and CG. Forshey. 1966. Effects of chemical thinning on repeat bloom
of ‘McIntosh’ apple trees. Proc. Amer. Soc. Hort. Sci. 88:25-32.

Neilsen, J .C. and F.G. Dennis, Jr. 1997. Effects of seed number, ﬁ'uit removal, bourse

shoot length and crop density on ﬂowering in ‘Spencer Seedless’ apple. Acta
Hort. (In press)

62

 

 

SECTION 3. EFFECTS OF DEFRUITING, SEED REMOVAL AND
REPLACEMENT OF SEEDS WITH GIBBERELLINS ON
FLOWER FORMATION IN APPLE.
Abstract
Experiments were conducted to determine the effects of defruiting, adjusting
seed number and replacement of seeds with gibberellin on ﬂowering of ‘Golden
Delicious’, ‘Jonagold’, ‘Red Delicious’, ‘Spencer Seedless’ and ‘Ohio No. 3' (another
seedless cultivar) apple. Seed number was altered by cutting fruits to remove or injure
seeds in late June, approximately 30 days after full bloom when fruits were about 30
mm in diameter. Gibberellin A3 or A,,,7 was mixed with lanolin (1000 mg-L“) and 0.02
mL applied to each empty locule (seeds removed or naturally seedless). Flowering the
following spring was greater than expected in both controls and gibberellin-treated
spurs. Neither cutting fruits nor seed removal increased ﬂowering significantly. The
treatments were applied late in the period of ﬂower induction when deﬁ'uiting had little

effect upon response. Nevertheless the results do not support the hypothesis that seed-

produced gibberellins are responsible for inhibiting ﬂowering in apple.

63

 

Introduction

The inhibition of ﬂowering in apple by heavy crops is a serious limiting factor to
proﬁtable fruit production. Chan and Cain (1967) provided evidence that seedless ﬁ'uits
do not inhibit ﬂowering. Developing apple seeds produce hormones, including
gibberellins (Dennis and Nitsch, 1966; Hedden, et al., 1993; Lin et al., 1991), and
spraying whole trees or foliage with gibberellins will inhibit ﬂowering, suggesting that
seed-produced gibberellins may be responsible for inhibition.

One can test the hypothesis that seed-produced gibberellins inhibit ﬂower
induction in apple by substituting gibberellins for seeds. In a study on fruit set, Abbott
(1959) described techniques for reducing seed numbers in fruit by injecting ethanol into
locules or cutting ﬁ'uits transversely and replacing seeds with lanolin paste containing
growth regulators. Grochowska (1968) replaced seeds with cotton swabs soaked in
growth regulators. Seed removal alone was performed too late to affect ﬂowering.
Indoleacetic acid (1AA) at 20 mg-L‘I and gibberellic acid (GA3) at 500 mg-L" had no
effect, but GA, at 1000 mg-L'l reduced ﬂowering 33 percent and naphthaleneacetic acid
(NAA) at 10 mg-L'l increased ﬂowering. To my knowledge this experiment has never
been repeated. Hoad (1980) and Hoad and Ramirez (1980) reported more gibberellins
moving out fruitlets of the strongly biennial ‘Laxton’s Superb’ than the more annual
‘Cox’s Orange Pippin’, as measured in diffusates collected ﬁ'om the pedicels in vitro.
The purpose of my work was to determine if replacing seeds with gibberellin can inhibit

ﬂowering in seeded and/or seedless apple fruits.

Materials and Methods

WW On 25 June 1996 (32 DAF B) ﬁve ‘Golden

Delicious’lM 26 trees planted in 1991 at the Clarksville Horticulture Experiment

Station, Clarksville, Mich., were selected. On each tree spurs with bourse shoots 2-20

mm long bearing two or more ﬁ'uits were treated in the following manner:

A.

B.

Fruiting control.

Fruits removed.

Fruits cut transversely through locules and seeds, but apex of fruit not
severed.

Fruits cut transversely, all but three seeds removed, and top of fruit
replaced and ﬁxed in place with Paraﬁlm".

Fruits cut as in D, all but one seed removed.

Fruits cut as in D, all seeds removed and a total of approximately 0.1 mL
of lanolin paste placed in the empty locules (0.02 mL in each) with a
one-mL syringe.

Same as F except that the lanolin paste contained 1000 mg-L" GA3.

Same as F except that the lanolin paste contained 1000 mg-L" GAM.

Two ﬁ'uits were treated on each spur, and any additional fruits were removed.

Each tree bore about ﬁve spurs per treatment (25 total) except for treatment A with a

total of 100 spurs being tagged throughout the ﬁve trees. At harvest, fruits were

weighed and all seeds counted. Trunk diameter was also measured and total crop was

65

counted and weighed (Table l). The following spring (1997) tagged spurs were
evaluated for ﬂower buds, and entire trees were rated visually from 0 to 10 based on
estimated bloom density.

A nearly identical experiment was conducted on ‘Jonagold’fM 9 trees planted in
1990 at the Horticulture Teaching and Research Center, E. Lansing, Mich., except that
the 25 spurs per treatment were distributed among six trees and not all seed number and
cropping data were collected. Treatments A - C were also applied to spurs on seven
Redchief ‘Delicious’ trees planted in 1981 at the Clarksville Horticulture Experiment
Station, Clarksville, Mich. No data were available on seed number per fruit because the
crop was harvested before the treated fruits could be collected.

W Similar experiments were also conducted in 1996
on ‘Spencer Seedless’ and ‘Ohio No. 3' (both facultatively parthenocarpic) trees planted
at the Horticulture Teaching and Research Center, E. Lansing, Mich., in 1969. Hand
pollination during bloom provided some spurs bearing seeded fruits for treatment A.
The 25 spurs used per treatment were all on one ‘Spencer Seedless’ tree, but were
distributed between two ‘Ohio No. 3' trees. Treatments C - B were omitted and
treatments P - H applied to seedless fruits.

Results

W In ‘Golden Delicious’ return ﬂowering in bearing
spurs was greater than expected, given the biennial bearing habit of this cultivar, with
67.5% of the buds on control spurs ﬂowering (Table 1,2). Because of lack of

replication, only large differences are of interest. Defruiting enhanced ﬂowering only

66

Table 1. Harvest data for ﬁve ‘Golden Delicious’ apple trees used in Expt. 1.
Clarksville Horticulture Experiment Station, 1996 - 1997.

 

 

Tree Fruit Fruit Trunk Crop density Yield Relative

no. (no./tree) mass dia. (fruit no./ em2 efﬁciency return
(kg/tree) (cm) TC SA) (g-cm'2 bloom

TCSA) (0-10)

1 71 10.32 6.2 2.35 342 10.0

2 88 12.48 5.8 3.33 473 9.8

3 68 11.28 4.9 6.61 598 6.5

4 71 11.71 6.3 2.28 376 8.9

5 283 35.55 6.2 9.37 1177 5.2

 

Table 2. Effects of defruiting, fruit cutting, seed removal, and replacement of seeds
with lanolin pastes on ﬂowering of ‘Golden Delicious’ apple. Clarksville Horticulture
Experiment Station, 1996 - 1997. Treatments were applied to 5 trees on 26 June 1996.

 

 

 

 

Treatment Total Flower Flowering
buds buds (°/o) Number of mature seeds
M‘_H_ (no.) (no.) Mean SE. C.V.
Defruited control 22 17 77 7.22 0.2 3.4
Fruiting control 114 77 68 7.7 0.4 46.1
Locules cut 25 21 84 1.5 0.4 123.4
3 seeds left 19 17 90 0.9 0.3 151.8
1 seed left 21 20 95 0.0

0 seeds left + lanolin paste

 

 

Control 24 23 96

GA3 25 21 84

GA,+7 25 23 92
z Immature seeds

67

slightly. Cutting locules reduced average seed number per spur, but up to ﬁve seeds
still developed to maturity. Both seed removal treatments also reduced seed number,
although the number of seeds remaining varied widely, and appeared to increase
ﬂowering relative to the fruiting control. Replacement of seeds with gibberellins had no
appreciable effect on ﬂowering. Results with ‘Jonagold’ were similar to those with
‘Golden Delicious’ (Table 3), except that almost all buds on fruiting spurs ﬂowered.
Although no seed number data were collected for ‘Delicious’, both defruiting and

cutting the locules of developing fruits appeared to increase ﬂowering (Table 4).

Table 3. Effects of defruiting, fruit cutting, seed removal, and replacement of seeds
with lanolin pastes on ﬂowering of ‘Jonagold’ apple. Horticulture Teaching and
Research Center, 1996 - 1997. Treatments were applied to 6 trees on 29 June 1996.

 

 

 

Treatment Total Flower Flowering
buds buds Number of mature seeds
_ (no.) r (no.) {(0/0) Mian S.E. C.V.
Defruited control 22 19 86 4.32 0.4 37.0
Fruiting control 17 16 94 4.4 0.3 29.9
Locules cut 25 20 80
3 seeds left 19 18 95
1 seed left 20 20 100
0 seeds left + lanolin paste
Control 19 19 100
GA, 19 18 95
GA“, 22 20 91

 

' Immatureseeds

68

Table 4. Effects of defruiting and fruit cutting on ﬂowering of ‘Delicious’ apple.
Clarksville Horticulture Experiment Station, 1996 - 1997. Treatments were applied to
7 trees on 30 June 1996.

 

Treatment Total buds Flower buds Flowering
(no.) (no.) (%)
Fruiting control 37 23 62
Defruited 36 3 1 86
Locules cut 31 30 97

 

W In ‘Spencer Seedless’, as the seed number borne per

spur increased, percentage ﬂowering decreased (Table 5). Defruiting, or cutting and
treating seedless fruits with the control lanolin paste did not affect response. Treating
locules with gibberellin appeared to decrease the percentage ﬂowering slightly in
‘Spencer Seedless’, but not in ‘Ohio No. 3' (Table 6) where ﬂowering was 90 to 100
percent in all treatments.
Discussion

The data for ‘ Spencer Seedless’ demonstrated the quantitative inhibitory effect
of seeds, but substituting gibberellin for the seeds was not as inhibitory as expected, and
failed to inhibit ﬂowering in ‘Ohio No. 3’. Flowering of ﬁ'uiting control spurs of seeded
cultivars was greater than expected (62 - 94 %), which can best be attributed to the
moderate crop most of these trees carried. In ‘Golden Delicious’ return bloom rating
was proportional to crop density. Taken together, this suggests an effect of cropload
that is stronger than the local effect of seeds within ﬁ'uits borne on the spur; similar

gibberellin treatments might have been more inhibitory had cropload been heavier.

69

Table 5. Effects of seed number and of applications of lanolin pastes to locules of cut
fruits of ‘Spencer Seedless’ at the Horticulture Teaching and Research Center, 1996 -
1997. Treatments were applied to fruits on one tree on 29 June 1996.

 

Treatment Mean seed Total buds Flower buds Flowering

   

Defruited control 24 23 96

Fruiting control 0 126 119 94
I -42 66 49 74

5-9 73 30 41

10-15 63 23 36

>1 5 13 20 1 5

Locules cut + lanolin paste

Control 19 19 100
GA, 22 1 8 82
GA“, 20 16 80

 

‘ Data in seed number classes taken from Section I for comparison.

70

Table 6. Effects of defruiting and applications of lanolin pastes to locules of cut fruits
of ‘Ohio No. 3’ apple at the Horticulture Teaching and Research Center, 1996 - 1997.
Treatments were applied to fruits on 2 trees on 25 June 1996.

 

Treatment Total buds Flower buds Flowering
(no.) (no.) (%)
Defruited control 47 46 98
Fruiting control 65 58 89
Locules cut + lanolin paste
Control 41 39 95
GA, 39 39 100
GA.” 46 46 100

 

This effect of crop was also observed in ‘Paulared’ as discussed in Section 2. Other
possible explanations for the small and inconsistent differences between defruited and
ﬁ'uiting controls may be the small numbers of spurs used to calculate percentages (about
20 per cultivar in most cases), and the need to use several trees of each cultivar without
being able to factor out crop density and other tree differences. Had more spurs been
available per tree, trees could have been used as blocks.

In ‘Spencer Seedless’ 5 to 9 seeds were sufﬁcient to limit ﬂowering to 41 %,
whereas in seeded cultivars similar numbers of seeds has less effect (68 % ﬂowering
with 7.7 seeds per spur in ‘Golden Delicious’; 94 % with 4.4 seeds per spur in
‘Jonagold’). Note that return bloom was greater in ‘Jonagold’, with fewer seeds per
spur, than in the biennial cultivar ‘Golden Delicious’.

The results of this study fail to support the hypothesis that seed-produced

71

 

———_l

gibberellins inhibit ﬂowering. Gibberellins applied to foliage are inhibitory, even when
applied to trees bearing few or no fruits (Dennis and Edgerton, 1966; Li et al., 1995;
McArtney, 1994), so the inhibitory effect of gibberellins is well established. Whether or
not seed gibberellins are involved in the control of ﬂowering is less clear. Several
experiments in which GAs have been substituted for seeds have had little or no effect on
ﬂowering in apple.

Grochowska (1968) applied gibberellins to locules in cotton swabs, which may
make limited contact with the surface of the locule, dry out, or result in rapid absorption

by the fruit tissue, providing a nearly instantaneous dose for metabolism or transport,

 

and therefore be ineffective. I assumed that lanolin paste would provide better contact
with the locular surface and would not dry out, and that diffusion of GA into the fruit
tissue would be slower, but more prolonged, better simulating the effects of seeds as
sources of GAs. Applying GAs directly to the bourse bud in lanolin might verify that
they can diffuse out of lanolin paste into the tissue and inhibit ﬂowering, but would not
test whether they can diffuse from seeds in sufﬁcient quantities to do so.

Although application of GAs in lanolin may be preferable to application in
cotton swabs, neither method is ideal, for seeds are attached to the fruits by specialized
placental tissue, and transport of substances through this tissue is metabolically
regulated. Furthermore, this tissue provides a very small “target" relative to the entire
suface of the locule.

Application may also have been after ﬂower induction had occured. The time of

ﬂower induction varies considerably with cultivar and bearing condition, as noted in

72

Sections 1 and 2 above. Selected spurs had bourse shoots 5 to 20 mm. long that should
have had high potential to ﬂower (Auchter and Schrader, 1923; Bobb and Blake, 193 8)
except that they were bearing fruit. At 30 DAF B they had ceased growth, but the
precise time of ﬂower induction was not determined.

Another possibility is that the amount of GA applied was insufﬁcient. This is
unlikely, however, for approximately 0.1 mg of GA was applied per fruit, which is more
by several orders of magnitude than the levels found in seeds. Green (1987) injected
3H-GA4 into apple seeds and found that only 2-3 percent moved out of the seeds at
most; much less (0.0003-0.07 percent) was found in the bud. More was transported
later than earlier in the growing season and the cultivar that exhibited more movement
in one year showed less movement the next, eliminating any correlation between
alternate bearing and GA export from fruits. Stephan et al. (1997) injected 3H-GA1, 3H-
GA, and 3H-GA4 into the core of apple fruits at 49 DAF B and found that much less than
ﬁve percent was transported out of the fruit and much less than one percent occurred in
the bourse shoot. Of these limited amounts, GAl and GA, was transported
preferentially to GA.. In contrast, Hoad (1980) and Hoad and Ramirez (1980) reported
movement of 3H to bourse shoots following injection of 3H-GA4 into seeds, with more
transport in ‘Laxton’ (biennial) than in ‘Cox’ (annual), though the values were small
and the differences nonsigniﬁcant.

Movement of gibberellins may vary, not only with cultivar, but with growing
conditions, or time and method of measurement. However, in general, this mechanism

is not well supported by the data at hand and must be viewed as only one of several

73

alternative hypotheses (e.g., “ﬂorigen”, auxin transport, cytokinin/gibberellin balance,

ethylene production or sensitivity) to explain the role of seeds in alternate bearing.

Literature Cited

Abbott, D.L. 1959. The effects of seed removal on the growth of apple fruitlets. Annu.
Rep. Long Ashton Agr. Hort. Res. Sta. 1958. p 52-56.

Auchter, EC. and AL. Schrader. 1923. Fruit spur growth and fruit bud production.
Proc. Amer. Soc. Hort. Sci. 20:127-144.

Bobb, A.C. and MA. Blake. 1938. Annual bearing in the ‘Wealthy’ apple was induced
by blossom thinning. Proc. Amer. Soc. Hort. Sci. 36:321-327.

Chan, B.G. and J .C. Cain. 1967. The effect of seed formation on subsequent ﬂowering.
Proc. Amer. Soc. Hort. Sci. 91 :63-68.

Dennis, F.G. Jr., and L.J. Edgerton. 1966. Effect of gibberellins and ringing upon apple
fruit development and ﬂower bud formation. Proc. Amer. Soc. Hort. Sci. 88:14-
24.

Dennis, F.G., Jr. and J .P. Nitsch. 1966. Identiﬁcation of gibberellins A4 and A7 in
immature apple seeds. Nature 211:781-782.

Green, J.R. 1987. The hormonal control of biennial bearing in cider apples. PhD
dissertation, Univ. of Bristol.

Grochowska, M.J. 1968. The inﬂuence of growth regulators inserted into apple
fruitlets on ﬂower bud initiation. Bul. Acad. Polan. Sci. Ser. 16:581-584.

Hedden, P., G.V. Hoad, P. Gaskin, M.J. Lewis, J .R. Green, M. Furber and LN. Mander.
1993. Kaurenoids and gibberellins, including the newly characterized
gibberellin A,,, in developing apple seeds. Phytochemistry 32:231-237.

Hoad, G.V. 1980. The role of seed derived hormones in the control of ﬂowering in
apple. Acta Hort. 80:93-103.

Hoad, G.V. and H. Ramirez. 1980. La funcién de las giberelinas sintetizadas en las

sernillas del ﬁ'uto para el control de la ﬂoracién en manzanos. Turrialba 30:284-
288.

74

 

Li, S.H., Z.Q. Meng, T.H. Li, H.Z. Liu and Y.C. Tu. 1995. Critical period of ﬂower
bud induction in ‘Red Fuji’ and ‘Ralls Janet’ apple trees. Gartenbauwissenschaft
60:240-245.

Lin, J .-T., A.E. Stafford and G.L. Steffens. 1991. Identiﬁcation of endogenous
gibberellins in immature apple seeds. Agr. Biol. Chem. 55:2183-2185.

McArtney, SJ. 1994. Exogenous gibberellin affects biennial bearing and the fruit shape
of ‘Braebum’ apple. New Zealand J. Crop and Hort. Sci. 22:343-346.

Stephan, M., F. Bangerth and G. Schneider. 1997. Transport and metabolism of the

gibberellins A,, A,, and A, after application to developing apple fruits of Malus
domestica cv. Jonagold. Acta Hort. 463:113-119.

75

 

+—

SECTION 4. EFFECT OF GIBBERELLIN SPRAYS ON FLOWER
FORMATION IN LATERAL BUDS OF ‘GALA’ APPLE.

Abstract

The objective of this experiment was to determine if treatment of bearing ‘Gala’
apple trees with gibberellins could prevent ﬂowering in lateral buds without affecting
ﬂowering in terminal buds. A mixture of gibberellins A4 and A, (GA,,-,, ratio not stated)
in the commercial formulation ProVide" was sprayed on trees at 150, 300 and 600 E
mg-L" on each of three dates (3 July, 19 July and 5 August 1996) in a factorial .

arrangement. On 3 July two unforrnulated mixtures of GA.,,7 with differing ratios (64/27 F

 

and 48/42 percent) were also applied at 150 mg-L". Non-treated controls were not
sprayed. Response was measured by subjectively rating bloom on the entire tree, and
by counting ﬂower buds on 10 shoots on each tree, distinguishing between terminal and
lateral ﬂower buds. There was no signiﬁcant difference (P = 0.05) in response to
treatments applied on 3 July 1997, when compared alone, nor was interaction between
time of application and concentration signiﬁcant. However, the effects of time of
application on subjective bloom rating, and on percentages of terminal and lateral buds
ﬂowering, were signiﬁcant. The early application (3 July 1996) reduced bloom
compared to later applications. Values for the latter two dates were similar to those of
the non-treated controls. Interpretation of the data was complicated by variability in
crop density among the trees selected for treatment. Using crop density as a covariant

did not affect the analysis of variance for ﬂowering parameters, so unadjusted means are

reported.

76

 

————_|

Introduction
Bobb and Blake (1938) pointed out that apple can produce many more ﬂowers
on spurs alone than are required to produce a full crop, and thatlateral ﬂowers just add
to the excess. They also reported the inhibition of spur leaf area caused by this
overabundance of ﬂowers. ‘Wealty’, ‘Wagner’, ‘Oldenburg’ and ‘Jonathan’ often

produce enough fruit from lateral ﬂowers to be considered annual bearing (Auchter,

 

1919), but fruits developing from these ﬂowers are smaller than those from terminal 5
ﬂower buds (V 012 et al., 1994). ‘Gala’ is a new commercial cultivar in many regions of

the United States, and size is often smaller than optimum. It produces many lateral

ﬂower buds on current season’s growth, thus reducing the average size. Fruits from ’

lateral ﬂowers also mature later, requiring multiple harvests or resulting in large
variations in maturity of fruit going into storage. Zeller (1960) reported that lateral
ﬂowers on terminal shoots are initiated later than are spur buds. Gibberellin sprays
inhibit ﬂower formation in apple (Dennis and Edgerton, 1966; Guttridge, 1962).
Greenhalgh and Edgerton (1967) found that potassium gibberellate (KGA,) treatments
of 100, 200 and 400 mg-L‘l applied 2 days after full bloom (DAF B) inhibited ﬂower
induction of ‘McIntosh’ apple. By 25 DAF B the same concentrations were much less
inhibitory. Marino and Greene (1981) found that GA, at 30 mg-L" selectively reduced
ﬂowering of lateral buds when applied 10 DAF B to mature ‘Early McIntosh’ trees,
while GA,” at the same rate was ineffective. At 300 mg-L" both treatments reduced
ﬂowering of both spurs and laterals and a second identical treatment 20 DAF B did not

signiﬁcantly increase the effect. McArtney (1994) reduced the severity of the alternate

77

 

ﬁﬁ

bearing cycle of ‘Braeburn’ apple by applying GA, and GA,” at full bloom. Although
there was a linear response to concentration, even the highest rate of GA, (330 ppm) did
not eliminate alternate bearing. None of the treatments inﬂuenced ﬂower bud formation
on one-year wood, which accounted for a high proportion of the total ﬂower clusters.
When applied 6, 9 or 12 weeks after full bloom, GA, or GA, inhibited ﬂowering on
one-year wood but not on spurs of vigorously growing ‘Braeburn’ shoots that had been
top-worked on ‘Royal Gala’lMM. 106 apple trees (McArtney and Li, 1998). GA, was
more effective than GA, and response increased as treatment was delayed. My purpose

was to determine the optimum concentration and timing for application of gibberellin

 

sprays for inhibiting ﬂowering in lateral buds of ‘Gala’ apple.

Materials and Methods

‘Gala’ apple trees propagated on ‘Mark’ rootstock and planted in 1991 at the
Clarksville Horticulture Experiment Station, Clarksville, Mich., were used in 1996.
Prior to treatment trees were visually rated for crop load and blocked accordingly. On 3
July, 19 July and 5 August 1996 (40, 56 and 73 DAF B, respectively) whole trees were
sprayed with 150, 300 or 600 mg-L'l GA4,,, formulated as ‘ProVide"’ (courtesy of
Abbott Laboratories, North Chicago, 111.). In addition, two mixtures of GA“, (64/27
and 48/42 percent GA4 and GA, , also ﬁ'om Abbott Laboratories) in powder form were
applied on the earliest date for comparison with the commercial product. All sprays
were applied with a hand gun to drip. Nontreated trees were included as controls. At

harvest, fruit number, yield and trunk diameter were recorded for each tree to better

78

 

control variation due to crop density during data analysis. In spring 1997, 10 current-
season shoots with at least 10 nodes (approximately 20 to 30 cm long) were selected on
each tree and the number of lateral ﬂowers were counted. In addition, four persons
visually rated each entire tree for bloom density on a scale from 0 to 10, and the four
scores were averaged. All treatments applied on 3 July 1996 and the nontreated control
were compared statistically to determine the effects of GA concentration, formulation
and ratio. A second factorial analysis was performed on the data for three
concentrations of GA”, repeated over three dates to determine the effects of both timing
and concentration. Further, bloom rating and number of lateral ﬂowers were correlated
with crop density to determine how closely these parameters were related across
treatments. All proportional values were transformed (W) to ensure near normal

distribution of values for analysis of variance.

Results

Though most of the trees in this experiment carried moderate to heavy crops,
return bloom was still adequate for a full crop the following year (Table 1). When
considered by themselves, none of the treatments applied 3 July inhibited ﬂowering
signiﬁcantly at P s 0.05, as evaluated by subjective rating of the whole trees or by
sampling shoots (Table 1). Although mean values for all treatments except one were
less than the control value, variability was also great. Considering the time by
concentration factorial treatments, bloom rating and lateral ﬂowers per shoot increased

as time of application was delayed (Tables 2 and 3), but no interaction between

79

 

Table 1. Effect of GA,” sprays applied 3 July 1996 to trees with varying crop density
on bloom rating and percentage of lateral buds ﬂowering in 1997 in ‘Gala’ apple at
Clarksville Horticulture Experiment Station, Clarksville, Mich. All values are means
for 5 trees.

 

 

 

 

Treatment Concn. Crop density Bloom ratingz Percentage
(mg-L'l a.i.) (Fruit no./ lateral buds
cm2 TC SA) ﬂoweringy
Nontreated 8.61 5.20 8.3
ProVide 150 10.40 4.00 4.2
300 8.99 3.50 3.2
600 9.35 3.73 3.1
GA,” (64/27 %) 150 8.04 4.65 0.9
GA.” (48/42 %) 150 8.20 5.30 2.4
P" 0.156 0.166 0.488
1 Mean of four evaluators’ visual bloom rating (0-10) at full bloom.
Transformation (J y + 0.5) performed before analysis for bloom rating.
Y Based on node counts of 10 shoots per tree. Percentages transformed (arcsin
square-root) before analysis.
" Probability (P) of differences being due to random variability based on analysis
of variance.

80

 

 

Table 2. Effect of timing and concentration of ProVideO treatments in 1996 on crop
density and bloom rating in 1997 in ‘Gala’ apple at Clarksville Horticulture Experiment
Station, Clarksville, Michigan. Five trees per treatment.

 

 

 

 

 

 

 

 

 

 

 

 

Crop density Bloom rating‘
(Fruit no./cm2 TCSA)

ProVide concn.(mgoL") ProVide concn.(mg-L")
ﬂing 150 300 600 Mean 150 300 600 Mean
3 Jul 96 10.4 9.0 9.4 9.6 4.0 3.5 3.7 3.7
19 Jul 96 6.6 11.0 9.2 8.9 5.5 5.0 5.6 5.4
5 Aug 96 7.0 9.9 8.5 8.5 6.3 4.8 6.0 5.7

P’ 0.014
Mean 8.0 10.0 9.0 0.0005" 5.3 4.4 5.1
‘ Mean of four evaluators’ visual bloom rating (0-10) at full bloom.

Transformation (W) performed before analysis.
Y Probability (P) of differences being due to random variability based on analysis
of variance. P values near 1 not shown.
Signiﬁcance of interaction of timing x concentration.

81

' -ﬂ‘:

 

 

Table 3. Effect of timing and concentration of ProVide“ treatments in 1996 on
ﬂowering of terminal and of lateral buds in 1997 in ‘Gala’ apple at Clarksville Ir

Horticulture Experiment Station, Clarksville, Michigan. Analysis of variance based on
transformed (arcsin square-root) data.

 

 

 

 

 

 

 

 

 

 

Terminal buds ﬂowering (%)z Lateral buds ﬂowering (%)y
Provide concn.(mg-L") Provide concn.(mg-L") P
Timing 150 300 600 Mean 150 J 300 600 Mean
3 Jul 96 28 20 18 22.0 4.3 3.2 3.1 3.5
19 Jul 96 60 54 50 54.7 13.1 5.0 13.1 10.4
5 Aug 96 66 56 52 58.0 14.2 12.6 14.1 13.6
P" 0.003 0.042
Mean 51.3 43.3 40.0 10.5 6.9 10.1
2 Based on 10 selected shoots per tree.
Y Based on total node no. of 10 selected shoots per tree.

Probability (P) of differences being due to random variability based on analysis
of variance. P values near 1 not shown.

82

concentration and time of treatment was detected (Figures 1 and 2). As crop density
increased, bloom rating declined signiﬁcantly (P < 0.0001) (Figure 3). Crop density
also appeared to have a stronger effect on lateral ﬂowering than did gibberellin
treatments. Many trees with more than eight fruits per cm2 TCSA produced few or no
lateral ﬂowers (Figure 4). Within the range of 5 - 13 fruits per cm2 TCSA, percentage

of lateral buds that were ﬂoral varied from 0 to 45.

Discussion

A speciﬁc objective of this experiment was to determine the effect of GA sprays
on lateral ﬂower induction in bearing ‘Gala’ apple trees. When young dwarf trees bear
widely varying crops, controlling alternate bearing with chemical thinning can be very
difﬁcult. If the undesirable lateral ﬂowers could be prevented from developing without
affecting ﬂowering of spurs, more uniform ﬂowering and cropping could be achieved.

Lateral ﬂowering is especially abundant in New Zealand, where photosynthetic
activity and development can continue for some time after harvest (R. Giuliani, personal
communication). McArtney and Li (1998) applied GA to ‘Braebum’ apple topworked
on ‘Royal Gala’lMM 106. The trees were extremely biennial and all were totally
vegetative in the year of treatment, resulting in control trees having 83 percent of the
lateral buds ﬂowering the following year. Treatment with surfactant alone (Regulaid)
reduced lateral ﬂowering to 70 percent, but treatment with GA, or GA, (100 mg-L“)
reduced this to 59 percent. Flowering decreased to 38 percent as the concentration was

raised to 400 mg-L“. In contrast, I found that ﬂowering of lateral buds of ‘Gala’ varied

83

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3‘1 1 . 'Gala' A
o 3 Jul 96
£7
E 10
Q
40
o \o
c 9 Nontreated
:2'.‘
3
t 8
V - 5 Aug 96
c: /
9
M 7 '
0
§ 6 19 Jul 96
o l- l a a l A L l a A l
5 Aug 96 B
8 6
u.- 19 Jul 96
8 o\
g 5r- Nontreated
S
2 3 Jul 96
O 4 -
8
m .\ a
3 b l a a l a l a l l
150 600

300 450
GIBBERELLIN CONCENTRATION (mg ' L“)

Figure 1. Crop density on ‘Gala’ apple trees used (A) and effects of concentration and
time of application of gibberellin in the form of ProVide“ in 1996 on bloom rating in
1997 (B). Each point is the mean for ﬁve replicate trees.

84

 

 

 

'Gala'

 

 

 

 

 

 

 

 

 

 

5 Aug 96 A
3?
v
g 601- \
E 19“
3 g
3
u.
8
3 4O Nontreated
m
2' _ 3 Jul 96
Z
2
El \
l- 20'-
l . n l r l A
15 . 5 Aug 96
S?
v
0 v.
E
9.3 19 Jul 96
g 105.
_l
I...
m Nontreated
O
D
m
'5' 5 - 3 Jul 96 B
'u—J .\
_.
S
l l l J

 

 

 

300 450 600
GIBBERELLIN CONCENTRATION (mg 'L")

Figure 2. Effects of concentration and time of application of gibberellin in the form of
ProVideO on percentage terminal (A) and lateral (B) buds ﬂowering the following year
on ‘Gala’ apple. Each point is the mean for ﬁve replicate trees.

85

 

 

10f 'Gala', all treatments

 

BLOOM RATING (0-10)

 

0
I- I a I a I a L 1 I a I a I a a
0 2 4 6 8 10 12 14 16
CROP DENSITY (Fruit no./ cm’TCSA)

Figure 3. Relationship (y = 8.32 - 0.38x, R2 = 0.38, P < 0.0001 ) between return bloom
rating in 1997 and crop density in 1996 in ‘Gala’ apple over three treatment dates and
three concentrations of ProVide" at Clarksville Horticulture Experiment Station.

86

 

 

 

 

'Gala'

407— o

301-

201

lol-

PERCENT LATERAL BUDS FLOWERING

 

L I a

0': v we ﬁacmmAﬂ- v u

GA4 +7(64127 %)

G + (48/42 %)
N:r4ltr6’ated

ProVide, 3 Jul, 150 ppm
Pronde, 3 Jul, 300 ppm
Pronde, 3 Jul, 600 ppm

ProVide, 19 Jul, 150 ppm
ProVide, 19 Jul, 300 ppm

ProVide, 19 Jul, 600 ppm I

ProVide, 5 Aug, 150 ppm
ProVide, 5 Aug, 300 ppm

ProVide, 5 Aug, 600 ppm

¥+004d>>DIOC

O
O

 

0

' “a
ﬁ

. +

a t

 

I L I

 

. 1
0 2 4

a I a I 1
6 8 1O 12 14 16

CROP DENSITY (Fruit no. / cm’TCSA)

Figure 4. Relationship between crop density in 1996 in ‘Gala’ apple and percentage of
lateral nodes ﬂowering in 1997 over three treatment dates for ProVide“ concentrations
from 150 to 600 mg-L" at Clarksville Horticulture Experiment Station.

87

 

from 0 to 44.3 percent; of the 60 trees used, 24 formed no lateral ﬂower buds. Although
the range in percentage ﬂowering was approximately the same in both experiments, the
absolute values in ‘Gala’ were much lower than in ‘Braebum’.

GA did not inhibit ﬂowering of spurs in ‘Braeburn’ (McArtney and Li, 1998),
presumably because the earliest treatment (42 DAF B) was applied after ﬂower
induction had occurred. A linear reduction in ﬂowering of lateral buds occurred as time
of GA application was delayed; maximum sensitivity could have occurred even later
than 84 DAF B - their latest treatment. In contrast, although application times were
similar, my earliest GA treatments were applied too late to establish the time of
maximum sensitivity. ‘Gala’ is an early-maturing cultivar, so these trees had
considerable time for bud development between harvest and leaf fall, but terminal
growth had ceased prior to harvest. If early crop removal alone favored ﬂower
induction, summer apple varieties as a group would be expected to ﬂower annually, yet
they are notoriously altemate-bearing in the United States. Other important differences
may be the effects of climate and cultivar on cessation of growth. The moderate
temperatures in New Zealand allow shoot growth to occur later than in continental
climates, such as that of Michigan (McArtney and Li, 1988). McArtney and Li (1998)
used top-worked trees with a few vigorous branches developing per trunk, and growth
may have been more uniform. This is supported by the fact that differences of 10
percent were highly signiﬁcant in their study, but nonsigniﬁcant in mine.

Marino and Greene (1981) also obtained selective inhibition of ﬂowering in

lateral buds by applying GA, (30 mg-L") at 10 DAF B to mature ‘Early McIntosh’ trees

88

 

in Massachusetts; GA4+7 was ineffective at this concentration. At 300 mg-L'l both
GAs reduced ﬂowering of both spurs and laterals; a second identical treatment 20

DAF B had no additional effect. This is much earlier than my treatments (40-70 DAF B),
but differences in response could also reﬂect differences in growth habit among
cultivars. Although the effects on ‘Early McIntosh’ were statistically signiﬁcant, they
were small, in terms of the potential for lateral ﬂower buds to develop, based on node
number. Marino and Greene (1981) expressed the results as number of blossom clusters
per meter of shoot length, rather than as percentage of total buds, and 300 mg'L‘I
reduced lateral ﬂowering from 17.0 (control) to 6.7 buds per meter. If expressed
similarly, the data for the 60 ‘Gala’ trees I used varied from 0 to 33.3 lateral clusters per
meter of one-year-old wood, but this difference was related more to crop density than to
GA treatment. The shoots selected (average length 245 mm) had an average of 61
nodes per meter.

In contrast with the trees used by McArtney and Li (1998) and Marino and
Greene (1981), which bore no crop the year of treatment, these ‘Gala’ trees all were
cropping. Even though the cropload varied widely, ﬂowering generally, including
lateral ﬂowering, was unexpectedly abundant, and cropload inﬂuenced ﬂowering more
than did the applied gibberellins. This was particularly true for some trees with heavy
crops, where the combination of cropload and GA treatment was expected to be very
inhibitory to ﬂowering.

My selection of GAM was primarily because ProVide” is already registered for

use on apple in early summer as a means of reducing fruit russeting and cracking. No

89

 

F—ﬁ

differences were found in response to the two samples of pure chemical with differing
ratios of GA4 and GA, vs. the ProVide” formulation of GA.” (unknown ratio). GA4 is
reported to be less inhibitory to ﬂowering than is GA, (Looney et al., 1985; Tromp,
1982; Wertheim, 1982). Greene (1993) applied GA4 or GAM, each at 10-40 mg-L", to
‘Golden Delicious’ at weekly intervals for 3-4 weeks beginning at petal fall. GA4
promoted ﬂowering in 2 years and inhbited it in one, whereas GA,” inhibited ﬂowering
in 3 out of 4 years, although response varied with year and concentration.

Li et al. (1995) found that GA, treatment could be used to determine the time of

 

ﬂower induction. They also reported that ﬂower induction in ‘Red Fuji’ occurred
earlier in spurs than in terminal buds of new shoots. The different types of buds in
‘Ralls Janet’ were more uniform within a given year, but the period of high sensitivity
to GA, was shorter (10 days) in an off-year than in an on-year (40 days), indicating an
interaction between sensitivity to GA and cropload. This was not the case in my work
with ‘Gala’, probably because the treatments were applied too late. Marino and Greene
(1981) conﬁrmed earlier reports that ﬂower induction in spurs begins much later in the
“off-year” (42 DAF B) than in the “on-year” (0 DAF B). Assuming that induction in trees
with moderate crops occurs between these two extremes, the time of induction in both
spurs and lateral buds of the ‘Gala’ trees used in this study could vary depending on
crop load and type of bud (spur, shoot terminal and shoot lateral).

A better understanding of the timing of induction is needed to be able to apply
treatments to prevent ﬂowering in lateral buds. Denker and Hansen (1994) found that

initiation in lateral buds does not occur until the shoots stop elongating. Terminal

90

 

 

ﬂower induction also appears to be related to time of cessation of growth, but greater
speciﬁcity would allow better timing of treatments. Two critical questions remain to be
answered: how far in advance of growth cessation does ﬂower induction occur, and
does induction occur simultaneously in terminal and lateral buds or does it progress
basipetally, starting with the apex? A related question is how long is GA effective in
inhibiting induction?
In this study, only those shoots that formed temrinal ﬂowers initiated lateral '

ﬂower buds, with very few exceptions, and abortion of terminal ﬂowers might account

‘ —.-—- r—dT-L A L

for these. Zeller (1960), working in Germany, reported that lateral ﬂowers can be

 

initiated very late in the growing season and even the following spring. Repeated
application of gibberellins at low concentrations may be necessary to prevent such late
induction. This would be similar to the protocol used for russet control in ‘Golden
Delicious’ (Greene, 1993), except that treatment would begin after spur buds have been
initiated, or about 40 DAF B for light-cropping trees.

Most of the shoots sampled in my work were longer than 20 cm. To better
determine the timing of initiation in terminal buds of spurs and in terminal and lateral
buds of shoots, samples of each should be collected for dissection. The fruiting status
of the spurs and the cropload for the tree should be recorded, as a clearer understanding

of ﬂower induction can be obtained only if all interacting factors are known.

Literature Cited

Auchter, EC. 1919. Some inﬂuences of thinning, pollination, and fruit spur growth on
the yearly performance record of fruit spurs and on the size of fruit produced.

91

 

 

Proc. Amer. Soc. Hort. Sci. 16:118-131.

Bobb, A.C. and MA. Blake. 1938. Annual bearing in the ‘Wealthy’ apple was induced
by blossom thinning. Proc. Amer. Soc. Hort. Sci. 36:321-327.

Denker, I. and P. Hansen. 1994. Flower initiation and node development of axillary
buds, growth partitioning, and bleeding sap production of young apple trees as
affected by fertigation. J. Hort. Sci. 69:869-876.

Dennis, Jr., F .G. and L.J. Edgerton. 1966. Effect of gibberellins and ringing upon apple
ﬁ'uit development and ﬂower bud formation. Proc. Amer. Soc. Hort. Sci. 88:14-
24.

Greene, D.W. 1993. Effects of GA4 and GA, on ﬂower bud formation and russet
development on apple. J. Hort. Sci. 68:171-176.

Greenhalgh, W.J. and L.J. Edgerton. 1967. Interaction of Alar and gibberellin on growth
and ﬂowering of the apple. Proc. Amer. Soc. Hort. Sci. 91 :9-17.

Guttridge, C.G.1962. Inhibition of fruit-bud formation in apple with gibberellic acid.
Nature 196:1008.

Li, S.H., Z.Q. Meng, T.H. Li, H.Z. Liu and Y.C. Tu. 1995. Critical period of ﬂower bud
induction in ‘Red Fuji’ and ‘Ralls Janet’ apple trees. Gartenbauwissenschaft
60:240-245.

Looney, N.E., R.L. Granger, C.L. Chu, S.J. McArtney, L.N. Mander and RP. Pharis.
1992. Inﬂuences of gibberellins A,, A,,,, and A,+iso-A, on apple fruit-quality
and tree productivity. 1. Effects on fruit russet and tree yield components. J.
Hort. Sci. 67:613-618.

Marino, F. and D.W. Greene. 1981. Involvement of gibberellins in the biennial bearing
of ‘Early McIntosh’ apples.

McArtney, SJ. 1994. Exogenous gibberellin affects biennial bearing and the ﬁ'uit shape
of ‘Braeburn’ apple. New Zealand J. Crop and Hort. Sci. 22:343-346.

McArtney, SJ. and SH. Li, 1998. Selective inhibition of ﬂowering on ‘Braeburn’
apple trees with gibberellins. Hortscience (in press).

Tromp, J. 1982. Flower-bud formation in apple as affected by various gibberellins. J.
Hort. Sci. 57:277-282.

Volz, R.K., I.B. Ferguson, E.W. Hewett and D.J.Wooley. 1994. Wood age and leaf

92

 

 

area inﬂuence fruit size and mineral composition of apple ﬁ'uit. J. Hort. Sci.
69:385-395.

Wertheim, SJ. 1982. Fruit russeting in apple as affected by various gibberellins. J.
Hort. Sci. 57:283-288.

Zeller, O. 1960. Entwicklungsgeschichte der Blﬁtenknospen und Fruchtanlangen an
einjahrigen Langtrieben von Apfelbl'lschen I. Entwicklungsverlauf und
entwicklungsmorphologie der Blﬁten am einjﬁhrigen Langtrieb. Z. ﬁlr
Pﬂanzenzilchtung 44: 1 75-214.

 

93

 

 

 

SUMMARY

Seedless fruits of ‘Spencer Seedless’ did not inhibit ﬂowering regardless of total
ﬁ'uit mass per spur or shoot length. Flowering was not inhibited unless the number of
seeds exceeded ﬁve. This contrasts with the data of Chan and Cain (1967), who found
that one to ﬁve seeds were sufficient to inhibit ﬂowering 65 to 85 %. When bourse
shoot length was less than 2 or greater than 10 mm, seeds had no effect; none of the
very short shoots formed ﬂowers, whereas almost all of the long ones did. This latter
observation agrees with the data of Huet (1972, 1973) for ‘Bartlett’ pear, which
indicated that the inhibitory effects of seeds could be overcome provided leaf surface L
per spur were sufﬁcient. My attempts to reduce seed number per fruit in commercial
cultivars were successful in reducing seed number and appeared to increase ﬂowering,
but the effects were too small to be signiﬁcant. Further experiments using this
technique may help elucidate the interaction between seeds and cropload.

For ‘Spencer Seedless’ spurs of intermediate length, seeds played a dominant
role in controlling ﬂowering. This contrasts with the results I obtained with ‘Paulared’,
for most shoots were much longer than 10 mm, yet a heavy crop severely inhibited
ﬂowering if left until harvest. Crop density appeared to affect ﬂowering both in
‘Spencer Seedless’ in the one year (1994) that it was recorded, and in ‘Paulared’, in
which bourse shoots of spurs bearing up to ﬁve seeded fruits ﬂowered when crops were
light. When cropload was moderate to heavy in ‘Paulared’, the number of fruit (and

seeds) per ﬁ'uiting terminal had little effect on ﬂower induction in the subtending bourse

bud. Responses of individual trees, however, varied widely, some being sensitive to

94

 

 

fruit number per spur and others either ﬂowering or failing to ﬂower regardless of fruit
number borne on the terminal. This suggests that in ‘Paulared’, at least, the aggregate
number of seeds per branch or tree is more important than the presence or absence of
seeds per se.

The high return bloom in 1997 on ‘Paulared’, ‘Golden Delicious’, ‘Jonagold’,
and ‘Gala’ trees bearing good crops in 1996 suggests that the effect of cropload can be
modiﬁed by environmental factors. This may explain why some cultivars are annual in

some locations and biennial in others. The profound effects of extreme alternate

 

bearing on shoot growth, spur leaf area, and time of ﬂower induction indicate that heavy
bloom and cropping have a number of important effects. The effect on ﬂowering may
be direct, via GAs emanating from the seeds (Luckwill, 1970; Marino and Greene,
1981), or indirect, either by reducing spur leaf area and therefore the supply of
“ﬂorigen”, or by excessive use of carbohydrate reserves by fruits. However, Stutte and
Martin (1986b) demonstrated that reduced carbohydrate content was not responsible for
the inhibitory effect of fruiting on ﬂowering of olive. This should be conﬁrmed in
apple. Hoad (1980) and Hoad and Ramirez (1980) reported the movement of 3H to buds
following injection of 3H-GA, into seeds. Somewhat more radioactivity was recovered
from buds of ‘Laxton’ (biennial) than from those of ‘Cox’ (annual); however, the values
were small and the difference nonsigniﬁcant. These studies suggest that if endogenous
GAs are responsible for alternate bearing, the mechanism is likely to be considerably
more complex than has been assumed heretofore. My attempt to substitute GAs for

seeds resulted in no detectable inhibition of ﬂowering. In further trials, application

95

 

directly to the buds would be an important control to verify that GAs are indeed
inhibitory at the time of treatment, and thus separate the question of timing from that of
transport. Defruited and ﬁ'uiting controls are also needed for comparison of gibberellin
treatment vs. ﬁ'uit effect. The time required for seed removal from individual fruits
limits the number of spurs that can be treated. Seedless fruits can simply be injected,
but precise placement of the injected GAs is difﬁcult, and their uptake, metabolism and
transport is not likely to mimic that of seed-produced GAS.

When GAs were sprayed on whole ‘Gala’ trees, the earliest treatment ( 3 July)

appeared to inhibit lateral ﬂowering of both entire trees and lateral buds, but the effect

 

was not statistically signiﬁcant when compared with nontreated controls. Later sprays
were without effect, regardless of concentration. Flowering was more closely
associated with crop density than with GA treatment. In contrast with the trees used by
McArtney and Li (1998) and Marino and Greene (1981), cropload varied considerably
in the ‘Gala’ trees. Li et al. (1996) reported that bearing trees of ‘Ralls Janet’ were
more sensitive to GA, application, and at an earlier date than were non-bearing trees.
Marino and Greene (1981) clearly demonstrated that ﬂower induction occurred earlier
(0 DAF B) in “on-year” than in “off-year” trees (42 DAF B) of ‘Early McIntosh’. In my
study ﬂowering was unexpectedly abundant on several heavily cropping ‘Gala’ trees
that were sprayed with GA. The best explanation for this discrepancy is that the
treatments were applied too late to affect ﬂowering. Flowering of spurs was not
recorded. In subsequent studies, ﬂowering of spur buds, in addition to that of terminal

and lateral buds on shoots, should be recorded. Also, repeated treatment with GA may

96

 

 

be necessary to prevent lateral ﬂower induction, which occurs over a long time period
(Zeller, 1960). Though the role of endogenous GAs remains to be determined, GA
application is a valuable tool to both investigate the mechanisms of ﬂower induction
and to control ﬂower induction commerically.

Greenhalgh and Edgerton (1967) measured or derived 123 variables that could
affect ﬂowering of apple, including treatment with daminozide (Alar ) and GA. Nine of
the variables were signiﬁcantly associated (P = 0.05) with return bloom, and accounted I ._...
for 78 % of the total variance. The variable with the largest positive coefﬁcient was log

mean fruit mass, for which the authors had no physiological explanation; this was

 

probably an indicator of reduced fruit set following daminozide application. The
variable with the largest negative coefﬁcient was log GA concentration, conﬁrming
other reports (Dennis and Edgerton, 1966; Guttridge, 1962; Marcelle and Sironval,
1963). Less important but still signiﬁcant positive associations were found with both
insoluble nitrogen content of spur leaves in June and time from full bloom until
cessation of shoot growth. Although ﬂowering was not associated with shoot length per
se, it increased with duration of shoot growth. This parallels, in some respects, at least,
my observations with regard to ﬂowering in spurs bearing seeded fruits of ‘Spencer
Seedless’. They also reported that although GA treatment increased both the duration of
shoot growth and shoot length, yet it inhibited ﬂowering, indicating that GAs may have
several effects, some positive and some negative, with respect to ﬂower induction.

A comprehensive model to describe control of ﬂowering in apple is still needed,

including a more detailed study of the interaction of cropload, shoot and bud

97

 

development, and fruit (seed) number per spur in several commercial cultivars. Care
should be taken to allow for ample replication and to consider previous cropping history
in order to eliminate as much variability as possible. Gibberellin and other hormones,
both applied or endogenous, should be researched to determine whether they inﬂuence

ﬂower induction directly or indirectly.

98

 

 

 

 

LIST OF REFERENCES

LIST OF REFERENCES

Abbott, D.L. 1959. The effects of seed removal on the growth of apple ﬁ'uitlets. Annu.
Rept. Long Ashton Agr. Hort. Res. Sta. 1958. p 52-56.

Abbott, D.L. 1960. The bourse shoot as a factor in the growth of apple fruits. Ann.
App. Biol. 48:434-438.

Abbott, D.L. 1970. The role of budscales in the morphogenesis and dormancy of the
apple fruit bud, p. 65-82. In: L.C. Luckwill and CV. Cutting (eds.). Physiology
of Tree Crops. Academic Press, New York.

Acquaah, G., J .W. Saunders and L.C.Ewert. 1992. Homeotic ﬂoral mutations. Plant
Breeding Rev. 9:63-99.

Auchter, EC. 1919. Some inﬂuences of thinning, pollination, and fruit spur growth on
the yearly performance record of fruit spurs and on the size of fruit produced.
Proc. Amer. Soc. Hort. Sci. 16:118-131.

Auchter, EC. and AL. Schrader. 1923. Fruit spur growth and fruit bud production.
Proc. Amer. Soc. Hort. Sci. 20:127-144.

Auchter, E.C., A.L. Schrader, F.S. Lagasse and W.W. Aldrich. 1926. The effect of
shade on the growth, fruit bud formation and chemical composition of apple
trees. Proc. Amer. Soc. Hort. Sci. 23:368-382.

Ban, J. 1996. Gibberelin metabolism and apple ﬂowering. PhD Dissertation, Michigan
State Univ.

Bangerth, F.K. 1997. Can regulatory mechanism in fruit growth and development be
elucidated through the study of endogenous hormone concentrations? Acta
Hort. 463 :77-87.

Bangerth, F. and M. Schroder. 1994. Strong synergistic effects of gibberellins with the
synthetic cytokinin N-(2-chloro-4-pyridyl)-N-phenylurea on parthenocarpic
fruit-set and some other fruit characteristics of apple. Plant Growth Regul.
15:293-302.

100

 

Bigelow, W.D. and H.C. Gore. 1905. US. Dept. Agr. Bur. Chem. Bul. 97. (Cited by
Gourley, 1922.)

Bobb, A.C. and MA. Blake. 1938. Annual bearing in the ‘Wealthy’ apple was induced
by blossom thinning. Proc. Amer. Soc. Hort. Sci. 36:321-327.

Brase, K.D. 1937. The vascular anatomy of the ﬂower of Malus domesﬁca Borkh. f.
M van Eseltine. MS. Thesis, Cornell University.

Buban, T. 1981. Histological and cytological aspects of ﬂower bud formation in apple.
Acta Hort. 120:113-117.

Buban, T. 1996. Flower development and formation of sexual organs. p. 3-54. In: J.
Nyéki and M. Soltész (eds.). Floral biology of temperate zone fruit trees and
small fruits. Akadéniai Kiado, Budapest.

 

Chailakhyan, M.Kh. 193 7. Hormone theory of plant development (Russian) Moscow,
Leningrad. Akad. Nauk. SSR.

Chan, B.G. and J .C. Cain. 1967. The effect of seed formation on subsequent ﬂowering.
Proc. Amer. Soc. Hort. Sci. 91:63-68.

Coen, E.S., 1991. The role of homeotic genes in ﬂower development and evolution.
Annu. Rev. Plant Physiol. Plant Mol. Biol. 42:241-279.

Crow, J.W. 1920. Biennial fruit bearing in the apple. Proc. Amer. Soc. Hort. Sci.
17:52-54.

Davis, L.D. 1957. Flowering and alternate bearing. Proc. Amer. Soc. Hort. Sci.
70:545-556.

Denker, I. and P. Hansen. 1994. Flower initiation and node development of axillary
buds, growth partitioning, and bleeding sap production of young apple trees as
affected by fertigation. J. Hort. Sci. 69:869-876.

Dennis, F .G. Jr. 1976. Gibberellin-like substances in apple seeds and fruit ﬂesh. J.
Amer. Soc. Hort. Sci. 101:629-633.

Dennis, F.G., Jr. and L.J. Edgerton. 1966. Effects of gibberellins and ringing upon
apple fruit development and ﬂower bud formation. Proc. Amer. Soc. Hort. Sci.
88:14-24.

Dennis, F.G., Jr. and J .P. Nitsch. 1966. Identiﬁcation of gibberellins A, and A, in

101

 

 

immature apple seeds. Nature 211:781-782.

Edwards, GR. 1987. Producing temperate-zone fruit at low latitudes: Avoiding rest and
the chilling requirement. Hortsci. 22:1236-1240.

Goff, ES. 1899. The origin and early development of the ﬂower in the cherry, plum,
apple, and pear. Wisconsin Agr. Expt. Sta 16th Annu. Rept. p. 289-303. (Cited
by Davis, 1957.)

Gorter, C.J., 1955. Photoperiodism of ﬂowering in apple trees. Rep. XIVth Intern.
Hort. Cong., p 351-354.

Gourley, J .H. 1920. The effect of shading some horticultural plants. Proc. Amer. Soc.
Hort. Sci. 17:256-260.

Gourley, J .H. 1922. Text book of pomology. p. 115. Macmillan Co., New York.

 

Green, J .R. 1987. The hormonal control of biennial bearing in cider apples. PhD
Dissertation, Univ. of Bristol.

Greene, D.W. 1993. Effects of GA, and GA, on ﬂower bud formation and russet
development on apple. J. Hort. Sci. 68:171-176.

Greenhalgh, W.J. and L.J. Edgerton. 1967. Interaction of Alar and gibberellin on grth
and ﬂowering of the apple. Proc. Amer. Soc. Hort. Sci. 91 :9-17.

Griggs, W.H., G.C. Martin, and B.T. Iwakiri. 1970. The effect of seedless versus seeded
ﬁ'uit development on ﬂower bud formation in pear. J. Amer. Soc. Hort. Sci.
95:243-248.

Grochowska, M.J. 1968. The inﬂuence of growth regulators inserted into apple ﬁ'uitlets
on ﬂower bud initiation. Bul. Acad. Polon. Sci., Ser. Biol. 16:581-584.

Grochowska, M.J. 1968. Translocation of indole-3-acetic acid-2-"C injected into seeds
of ﬁve-week-old apple ﬁ'uits. Bul. Acad. Polon. Sci., Ser. Biol. 16:577-580.

Guttridge, C.G.1962. Inhibition of fruit-bud formation in apple with gibberellic acid.
Nature 196:1008.

Harley, C.P., J .R. Magness, M.P. Masure, L.A. Fletcher and ES. Degrnan. 1942.
Investigations on the cause and control of biennial bearing in apple trees.
U.S.D.A. Tech. Bul. 792.

Harley, C.P., H.H. Moon and L.O. Regeimbal. 1958. Evidence that post-bloom apple-

 

102

 

thinning sprays of naphthaleneacetic acid increase blossom-bud formation.
Proc. Amer. Soc. Hort. Sci. 72:52-56.

Hedden, P., G.V. Hoad, P. Gaskin, M.J. Lewis, J .R. Green, M. Furber and L.N. Mander.
1993. Kaurenoids and gibberellins, including the newly characterized
gibberellin A,,, in developing apple seeds. Phytochemistry 32:231-237.

Heinicke, A.J. 1917. Factors inﬂuencing the abscission of ﬂowers and partially
developed fruits of the apple (Pyrus malus L.). Cornell Univ. Agr. Expt. Sta.
Bul. 393.

Hipps, N.A. 1992. F ertigation of newly planted ‘Queen Cox’/M9 apple trees-
establishment, early growth and precocity of cropping. J. Hort. Sci. 67:25-32.

Hirst, RM. and DC. F erree. 19953. Rootstock effects on shoot morphology and spur
quality of ‘Delicious’ apple and relationships with precocity and productivity. J.
Amer. Soc. Hort. Sci. 120:622-634.

Hirst, RM. and DC. Ferree. 1995b. Rootstock effects on the ﬂowering of ‘Delicious’
apple. 1. Bud development. J. Amer. Soc. Hort. Sci. 120:1010-1017.

Hirst, RM. and DC. Ferree. 1996. Effects of rootstock on bud development and ﬂower
formation of ‘Starkspur Supreme Delicious’ apple. Fruit Var. J. 50:25-34.

Hoad, G.V. 1980. The role of seed derived hormones in the control of ﬂowering in
apple. Acta Hort. 80:93-103.

Hoad, G.V. and H. Ramirez. 1980. La funcién de las giberelinas sintetizadas en las
sernillas del ﬁ'uto para 61 control de la ﬂoracién en manzanos. Turrialba 30:284-
288.

Huet, J. 1972. Etude des efféts des feuilles et des fruits sur l'induction ﬂorale des
brachyblastes du Poirier. Physiol. Vég. 10:529-545.

Huet, J. 1973. Floral initiation in pear trees. Acta Hort. 34:193-198.

Li, S.H., S.Q. Meng, T.H. Li, H.Z. Liu and Y.C. Tu. 1995. Critical period of ﬂower bud
induction in ‘Red Fuji’ and ‘Ralls Janet’ apple trees. Gartenbauwissenschaft
60:240-245.

Lin, J .-T., A.E. Stafford and G.L. Steffens. 1991. Identiﬁcation of endogenous
gibberellins in immature apple seeds. Agr. Biol. Chem. 55:2183-2185.

Looney, N.E., R.L. Granger, C.L. Chu, S.J. McArtney, L.N. Mander and RF. Pharis.

103

 

 

 

‘T—__———I

1992. Inﬂuences of gibberellins A,, A,,,, and A,+iso-A, on apple fruit-quality
and tree productivity. 1. Effects on fruit russet and tree yield components. J.
Hort. Sci. 67:613-618.

Looney, N.E., R.P. Pharis, and M. Noma. 1985. Promotion of ﬂowering in apple trees
with gibberellin A, and C-3 epi-gibberellin A,. Planta 165:292-294.

Luckwill, LC. 1970. The control of growth and fruitfulness of apple trees. p. 237-254
In: L.C. Luckwill and C.V.Cutting (eds.). Physiology of Tree Crops. Academic
Press, New York.

Luckwill, L.C., P. Weaver and J. MacMillan. 1969. Gibberellins and other growth
hormones in apple seeds. J. Hort. Sci. 44:413-424.

Marcelle, R. and C. Sironval. 1963. Effect of gibberellic acid on ﬂowering of apple
trees. Nature 197:405.

 

Marino, F. and D.W. Greene. 1981. Involvement of gibberellins in the biennial bearing
of ‘Early McIntosh’ apples. J. Amer. Soc. Hort. Sci. 106:593-596.

McArtney, SJ. 1994. Exogenous gibberellin affects biennial bearing and the Mt shape
of ‘Braeburn’ apple. New Zealand J. Crop and Hort. Sci. 22:343-346.

McArtney, SJ. and SH. Li, 1998. Selective inhibition of ﬂowering on ‘Braebum’
apple trees with gibberellins. Hortscience (in press).

McArtney, S.J., D.S. Tustin, S. Seymour, W. Cashmore and NE. Looney. 1995.
Benzyladenine and carbaryl effects on ﬁ'uit thinning and the enhancement of
return ﬂowering of 3 apple cultivars. J. Hort. Sci. 70:287-296.

McCormick, A.C. 1933. Control of biennial bearing in apples. Proc. Amer. Soc. Hort.
Sci. 30:326-329.

 

McKee, M.W. and CG. F orshey. 1966. Effects of chemical thinning on repeat bloom
of ‘McIntosh’ apple trees. Proc. Amer. Soc. Hort. Sci. 88:25-32.

McLaughlin, J .M. and D.W. Greene. 1991. Fruit and hormones inﬂuence ﬂowering of
apple. 1. Effect of cultivar. J. Amer. Soc. Hort. Sci. 116:446-449.

Neilsen, J .C. and F .G. Dennis Jr. 1997. Effects of seed number, fruit removal, bourse
shoot length and crop density on ﬂowering in ‘ Spencer Seedless’ apple. Acta
Hort. (In press).

Nesterov, Y.S., G.G. Anisimova and NJ. Savelev. 1972. Vestnik Sel’skokhov. Nauki

104

 

 

10:114-118. (Cited by Buban, 1996.)

Palmer, J .W. 1992. Effects of varying crop load on photosynthesis, dry-matter
production and partitioning of ‘Crispin’/M.27 apple trees. Tree Physiol. 11:19-
33.

Palmer, J.W., Y.L. Cai and Y. Edjamo. 1991. Effect of part-tree ﬂower thinning on
fruiting, vegetative growth and leaf photosynthesis in ‘Cox Orange Pippin’
apple. J. Hort. Sci. 66:319-325.

Roberts, RH. 1920. Off year bearing and apple spur growth. Wisconsin Agr. Expt.
Sta. Bul. 317.

Ramirez, H. and G.V. Hoad. 1981. Effects of growth substances on fruit-bud initiation
in apple. Acta Hort. 120:131-136.

Salisbury, RB. 1982. Photoperiodism. Hort. Rev. 4:66-105.

Singh, LB. 1948. Studies in biennial bearing II. A review of the literature. J. Hort. Sci.
24:45-65.

Sritharan, R. and F. Lenz. 1988. Effects of water-stress in apple. I. Effect of water-
supply on ﬂower formation and fruit-set of the apple cultivar ‘Golden
Delicious’. Gartenbauwissenschaft 53:223-226.

Steel, R.G.D. and J.H. Torrie. 1980. Principles and Procedures of Statistics, 2nd
Edition. McGraw-Hill Book Company, Inc., New York.

Stephan, M., F. Bangerth and G. Schneider. 1997. Transport and metabolism of the
gibberellins A,, A,, and A, after application to developing apple fruits of Malus
domestica cv. Jonagold. Acta Hort. 463:113-119.

Stutte, G.W. and G.C. Martin. 1986a. Effect of killing the seed on return bloom of
olive. Scientia Hort. 29: 107-1 13.

Stutte, G.W. and G.C. Martin. 1986b. Effect of light intensity and carbohydrate
reserves on ﬂowering of olive. J. Amer. Soc. Hort. Sci. 111:27-31.
Struckmeyer, BE. and RH. Roberts. 1942. Investigations on the time of blossom

induction in Wealthy apple trees. Proc. Amer. Soc. Hort. Sci. 40:113-119.

Tobutt, KR. 1994. Combining apetalous parthenocarpy with columnar growth habit in
apple. Euphytica 77:51-54.

105

 

 

Tromp, J. 1982. Flower-bud formation in apple as affected by various gibberellins. J.
Hort. Sci. 57:277-282.

Tromp, J. 1987. Growth and ﬂower-bud formation in apple as affected by
paclobutrazol, daminozide, and tree orientation in combination with various
gibberellins. J. Hort. Sci. 62:433-440.

Tromp, J. 1992. The effect of soil temperature on lateral shoot formation and ﬂower-
bud formation in apple in the lst year after budding. J. Hort. Sci. 67:787-793.

Tromp, J. 1993. Lateral shoot formation and ﬂower-bud formation in apple in the lst
year alter budding as affected by air-temperature and exposure to red-light. J.
Hort. Sci. 68:255-260.

Volz, R.K., I.B. Ferguson, E.W. Hewett and D.J.Wooley. 1994. Wood age and leaf
area inﬂuence fruit size and mineral composition of apple fruit. J. Hort. Sci.
69:3 85-395.

 

Weigel, D. and O. Nilsson. 1995. A developmental switch sufficient for ﬂower
initiation in diverse plants. Nature 377:495-500.

Wertheim, SJ. 1982. Fruit russeting in apple as affected by various gibberellins. J.
Hort. Sci. 57:283-288.

Wertheim, SJ. and EN. Estabrooks. 1994. Effect of repeated sprays of 6-
benzyladenine on the formation of sylleptic shoots in apple in the fruit tree
nursery. Scientia Hort. 60:31-39.

Zeller, O. 1960. Entwicklungsgeschichte der Blﬁtenknospen und Fruchtanlangen an
einjahrigen Langtrieben von Apfelbttschen I. Entwicklungsverlauf und
entwicklungsmorphologie der Blttten am einjahrigen Langtrieb. Z. ﬁtr
Pﬂanzenzﬁchtung 44:175-214.

 

106

"‘ﬁllTJJlllluluTll‘5