Directory UMM :Data Elmu:jurnal:T:Tree Physiology:vol17.1997:
Tree Physiology 17, 211--219
© 1997 Heron Publishing----Victoria, Canada
Time and method of floral initiation and effect of paclobutrazol on
flower and fruit development in Shorea stenoptera (Dipterocarpaceae)
DIDA SYAMSUWIDA1 and JOHN N. OWENS2
1
Forest Tree Improvement Research and Development Institute, Jl. A.M. Sangaji Km. 15, Purwobinangun, Pakem, Sleman, DI Yogyakarta 55582,
Indonesia
2
Centre for Forest Biology, University of Victoria, Victoria, B.C. V8W 2Y2, Canada
Received January 11, 1996
Summary Small Shorea stenoptera Burck. (Dipterocarpaceae) trees of reproductive age growing in an arboretum in
west Java were studied to determine the pattern of vegetative
shoot development, the time and method of floral initiation and
the effect of paclobutrazol on floral enhancement. Vegetative
buds were enclosed by two stipules between which was a leaf
primordium, a small axillary vegetative bud and another pair
of stipules. This sequence was reiterated five to seven times
before the vegetative apex was visible. At the time of floral
initiation, axillary buds developed into floral spikes and compound inflorescences formed at the end of drooping branches.
A compound inflorescence might bear many floral spikes and
each floral spike bore many flowers. The compound inflorescence was a modification of the reiterative developmental
pattern observed in vegetative shoots. The time of floral initiation began in late June or early July and continued until about
November. Floral enhancement using paclobutrazol as a soil
drench was attempted in mid-July, but this was later found to
be after the onset of floral initiation, and the treatment failed
to enhance flowering; however, it appeared to enhance the rate
of floral and fruit development. The similarity in vegetative bud
development among dipterocarp genera suggests that the time
of floral initiation may be easily determined in many species
based on simple dissection techniques.
Keywords: dipterocarp, floral enhancement, shoot development.
Introduction
The family Dipterocarpaceae may account for 10% of all tree
species and 80% of all emergent trees in tropical rain forests
(Ashton 1982). Dipterocarps occur in both aseasonal lowland
rain forests where they are evergreen, and in seasonal upland
or more xerophytic forests where they are deciduous (Smitinand et al. 1980).
A striking feature of many dipterocarp forests is the phenomenon of mass flowering followed by mass fruiting (Ashton
et al. 1988). Mass flowering occurs at intervals of two to 10
years, over a period of a few weeks to a few months in either
small areas or large regions such as the Malaysian Peninsula.
Large dipterocarps may produce up to four million flowers and
120,000 fruits (Ashton 1988). Although the fruits are poorly
protected and are eaten in vast numbers by wild pigs and
heavily parasitized by weevils (Chan 1980, Ashton 1988), the
forest floor following a mass fruiting is often blanketed by
dipterocarp seedlings. In some regions, reforestation is
achieved by lifting and transplanting the wild seedlings.
Managing dipterocarps for seed production is difficult. First
flowering commonly does not occur until trees are 20 to 30
years old (Ng 1966), by which time most are too large for easy
management or seed harvest. Fruit production is episodic, and
because seeds of most species are recalcitrant and store poorly
(Sasaki 1980), there can be long intervals during which seed
for reforestation is unavailable. Although flowers have three
locules and two ovules per locule, normally only one indehiscent seed develops per fruit. To overcome these obstacles,
methods are needed to induce flowering at regular intervals in
relatively small trees.
Effective control of flowering, requires knowledge of the
time of floral initiation and the development of methods of
floral induction or enhancement (Owens and Blake 1985). The
time of natural floral initiation in dipterocarps has never been
determined developmentally or experimentally. Dipterocarps
show episodic shoot growth with distinct periodic vegetative
flushes, followed by periods of shoot elongation. Flushes commonly precede or coincide with flowering, and inflorescences
tend to be borne near the ends of shoots (Burgess 1972).
Stimuli that induce flushing and flowering in dipterocarps are
not known. Ashton et al. (1988) proposed that flowering is
controlled by an endogenous rhythm, perhaps related to episodic shoot flushes and triggered by an external cue. Wycherley (1973) suggested that temperature may be responsible for
floral induction in dipterocarps. Ashton et al. (1988) found no
close correlation between flowering intensity of Shorea species and photoperiod or water availability, and proposed that
exposure to a 2 °C decrease in minimum nighttime temperature for several nights may be sufficient to induce flowering in
some dipterocarps.
Increased temperature and drought, usually in conjunction
with gibberellin A 4/7 application, are common cultural treatments applied to induce cones in Pinaceae (Owens and Blake
212
SYAMSUWIDA AND OWENS
1985). The mode of action of these treatments, however, has
yet to be demonstrated. One hypothesis is that these cultural
treatments reduce vegetative growth, and the resulting excess
gibberellins promote flowering (Pharis 1976). Gibberellins
have not been found to enhance flowering in hardwood forest
trees. However, flowering in several temperate fruit trees
(Lever 1986, Steffens et al. 1993), clove (Martin and Dabek
1988) and Eucalyptus (Griffin et al. 1993) is enhanced when
the growth retardant paclobutrazol is applied as a foliar spray,
stem injection or soil drench.
The objectives of this study were to determine the time and
method of floral initiation and to investigate the use of
paclobutrazol as a method of enhancing flowering in young,
reproductively mature Shorea stenoptera Burck. trees.
Materials and methods
In June 1994, 27 reproductively mature S. stenoptera trees that
had previously flowered were selected from a stand of over 100
trees at the Kebun Percorbaan Arboretum in west Java, Indonesia (lat. 6° N, long. 106° E). Trees were grouped into three
categories, with nine trees in each group, on the basis of
circumference at breast height: Group I, 51--55 cm; Group II,
46--50 cm; Group III, 41--45 cm. Three trees of each size class
were randomly assigned to each of three paclobutrazol treatments: A, high concentration; B, low concentration; and C,
control.
Floral initiation
From mid-July through October, branches were sampled from
several trees every month. The terminal bud on each branch
was removed and placed in a vial with formalin/acetic acid/alcohol (FAA) fixative (Johansen 1950). Buds were stored in
FAA and taken to the University of Victoria where a central
longitudinal section 1--2 mm was embedded in ‘‘Tissuemat’’
(Fisher Scientific, Ottawa, Canada) using the tertiary-butyl
alcohol dehydration series of Johansen (1950). Embedded
specimens were softened in Gifford’s softening reagent (Gifford 1950) for 2 to 4 weeks and then serially sectioned longitudinally in increments of 6 µm. The median sections were
mounted on microscope slides and stained with safranin and
fast green and examined to determine whether they were vegetative or reproductive.
Other buds were dissected by removing successive stipules
until the terminal bud or apex and the axillary vegetative buds
or inflorescence primordia were visible. The shoot tips, or
portions of them, were dehydrated in an ethanol series, critical
point dried, mounted on aluminum stubs, sputter-coated with
gold and observed with a JOEL JSM-35 scanning electron
microscope (SEM) (JOEL, Boston, MA) at 10 kV. Other buds
were dissected, sketched and photographed with a Wild M 400
Photomacroscope (Leica Canada Inc., Willowdale, Canada) to
show bud morphology. In all three types of specimen observation, vegetative buds and buds that had undergone floral initiation and various stages of floral development were observed
and photographed.
Floral enhancement
In mid-July, sample trees were treated once with one of three
paclobutrazol treatments. A shallow circular trench (about
10-cm-deep) was dug at a radius of 1 m from the base of each
tree. Paclobutrazol in the water soluble form ‘‘Cultar’’ was
obtained from ICI (Indonesian Operations, Jakarta). Concentrations and application methods were based on successful
experiments with Eucalyptus (Griffin et al. 1993, Moncur et al.
1994). Paclobutrazol treatments were either high concentration (1.0 g of paclobutrazol per 15 cm of trunk circumference),
low concentration (0.5 g of paclobutrazol per 15 cm of trunk
circumference) or a control (water only). The appropriate
amount of paclobutrazol was added to 4 liters of water, which
was then mixed and poured into the trench around each tree.
Once per month in September, November, December and
January 1994, trees were climbed and the numbers of inflorescences with flowers and with developing fruits counted. The
number of flowering compound inflorescences and number of
fruiting compound inflorescences were counted on each observation date. Numbers of flowers or fruits per compound inflorescence or per spike were not counted. Because trees varied
in size, numbers of branches and sites in which inflorescences
formed and the number of primary branches per tree were
counted to determine the percentage of branches bearing inflorescences.
To observe any treatment carry-over effect, trees were assessed for frequency of flowering in November 1995, i.e., the
year following paclobutrazol treatment.
Average rainfall per month was obtained from the weather
station at the study site.
Figures 1--8. Vegetative and reproductive shoots and buds and fruits of
Shorea stenoptera.
Figure 1. Vegetative branch of mature tree showing red stipules (S).
Figure 2. Close-up of vegetative branch showing stipules and emerging leaf (L).
Figure 3. Partially dissected vegetative bud with outer stipule removed
to reveal the leaf primordium, axillary buds (A) and small green stipule
inside.
Figure 4a. Vegetative bud with one stipule removed to show leaf
primordium and stipule within and axillary bud at the onset of floral
initiation. Figure 4b. Compound inflorescence with outer stipule removed to show enlarged axillary inflorescence (AI).
Figure 5. Later stage of a compound inflorescence with two stipules
removed to show two stages of axillary inflorescence development.
Figure 6. Compound inflorescence during flowering showing red
stipules at the base of each spike (SP), and developing fruits ( ).
Figure 7. Distal portion of a compound inflorescence showing several
spikes. Most flowers (F) have been shed and one is aborting ( )
leaving the subtending bracts (B). Most stipules have been shed.
Figure 8. Three nearly mature globose fruits showing leaf-like wings.
All other flowers abscised.
FLOWER AND FRUIT DEVELOPMENT IN SHOREA STENOPTERA
213
214
SYAMSUWIDA AND OWENS
Statistical analysis
Numbers of inflorescences bearing flowers and fruits were
analyzed separately, or summed for each date and analyzed by
a balanced two-factor analysis of variance (ANOVA) design
with size class and paclobutrazol dosage as fixed effects.
Because size class did not influence the number of branches
bearing flower or fruit inflorescences (α = 0.05), all size
classes were pooled and subjected to a single-factor ANOVA
for each date. Significant effects were further analyzed with
the Duncan’s multiple range test (Zar 1984).
Observations and results
Shorea stenoptera formed pendant compound infloresences at
the ends of long, often drooping branches in younger trees
(Figure 1), and on shorter secondary and tertiary branches in
older trees. The inflorescences, which were initiated and underwent early development within the terminal, partially
fused, protective, stipules on the branch (Figures 2 and 3),
commonly had seven to 15 floral spikes (Figure 6). Spikes
were spirally arranged but appeared alternate, and were separated by long internodes (Figure 7). Spikes matured acropetally on the branch and flowers matured acropetally on each
spike (Figure 6). Each spike was enclosed by two stipules
during early floral development (Figures 4 and 5), and bore one
to usually many spirally arranged flowers (Figure 7), each on
a short pedicel. Each flower had two opposite subtending
bracts. Most flowers abscised at the base of the pedicel, leaving
the two bracts that usually abscised a few days later (Figure 7).
Rarely did more than one fruit develop per spike. Fruits rapidly
enlarged, and the five sepals elongated, enclosing the globose
fruit at their base, and forming three large and two small, broad
flat wings. Spikes were pendant and the wings hung downward
(Figure 8).
Vegetative bud development
Development within the vegetative bud provided a remarkable
example of reiteration. Each bud was enclosed by two large
spoon-shaped red stipules, each of which attached around half
of the stem and the base of a leaf. The two stipules adhered
along their margins. Between the pair of stipules was a leaf, an
axillary bud and another pair of smaller green stipules (Figures
2, 3 and 4a). Within these stipules was, again, a leaf, an axillary
bud and a yet smaller pair of stipules. These structures were
often reiterated five to seven times (Figure 9) as revealed by
dissection (Figures 3, 4 and 10) and SEM, before the shoot
apical meristem was visible (Figure 11). The inner and, to a
lesser extent, the outer surfaces of the young stipules were
covered by long epidermal hairs (Figure 11) that stuck to the
leaf primordia and other stipules.
The apical meristem was a low dome with an oval cross
section, from which successive leaf primordia arose at opposite ends (Figure 12). Soon after a leaf primordium was initiated, two ridges of meristematic tissue arose at the primordium
base along the sides of the oval (Figure 12). These ridges
extended around the apical meristem and base of the leaf
primordium, and elongated upward to form two opposite
stipule primordia (Figure 13). As the stipule primordia elongated, they became spoon-shaped and contacted one another
along their margins. Epidermal hairs formed on the stipules,
causing them to stick together along most of their margins,
completely enclosing the leaf primordium and apical meristem
within (Figure 11). As the leaf primordium elongated and a leaf
blade began to form, a lateral bud primordium arose in the axil
of the leaf primordium (Figure 12). By this time, the next leaf
primordium and its stipules had been initiated (Figures 12-14).
Axillary buds on most vegetative shoots remained small.
The axillary apex initiated two scales (Figure 12) that elongated and enclosed the apex (Figure 11). Development may be
arrested at this stage (Figure 15) or the apical meristem may
initiate leaf primordia and stipules. In the smaller trees, the
primary branches generally lacked secondary branches so axillary buds appeared to become latent (Figure 15). In larger
trees, secondary and tertiary branches formed from the axillary
buds.
Figure 9. Diagram of a Shorea
stenoptera vegetative bud with
successive stipules (S) removed
to show the enclosed leaf primordia (L) and axillary buds (A). The
sequence may be reiterated five
to seven times as shown in a--e.
FLOWER AND FRUIT DEVELOPMENT IN SHOREA STENOPTERA
215
Figures 10--15. Vegetative Shorea
stenoptera buds using light (LM)
and scanning electron microscopy
(SEM).
Figure 10. Fixed and partially dissected bud viewed with a dissecting LM. Outer stipules were
removed to show an inner stipule
(S3), its leaf primordium (L3) and
a bud (A 4) in the axil of each leaf.
Different subscripts indicate
stages of development.
Figure 11. Enlarged SEM of a bud
as in Figure 10. The stipules,
leaves and axillary buds having
the same subscript as in Figures
10 and 11 are at the same stage.
The stipule (S2) has been removed
to reveal its leaf (L 2), the apical
meristem (arrow) and the youngest leaf primordium (L1).
Figure 12. Enlarged SEM of the
apical meristem ( ) and leaf primordia (L1, L2, L3) and the axillary bud (A3) as shown in Figure
11. Two stipules (S1) are those
that belong to L1. Outer stipules
(S2) have been removed. An axillary bud (A2) has just begun to be
initiated in the axil of L 2. Axillary
bud (A3) has two lateral scales (arrowhead).
Figure 13. A later stage of stipule
development to that shown in Figure 12. Stipules (S1) for L1 have
enlarged covering the apex. The
next leaf primordium (L 0) is being
initiated.
Figure 14. Median longitudinal
section of a vegetative bud as in
Figure 13 showing the apical meristem ( ) leaf primordia (L1, L2)
and the axillary bud (A3) for leaf
three.
Figure 15. Median longitudinal
section of an axillary vegetative
bud that has become latent soon
after being initiated.
Floral initiation and development
Inflorescence initiation was observed in some buds collected
in July and in all subsequent collections through November. At
inflorescence initiation, the vegetative apex underwent transition to an inflorescence apex. The terminal inflorescence apex
was dome-shaped and initiated a spiral sequence of stem units
consisting of a leaf primordium, two stipules and an axillary
apex (Figure 16). Development of the components of the stem
units differed depending on the time of inflorescence initiation, with leaf primordia being smaller or absent in stem units
initiated later, whereas stipules and internodes were longer and
axillary buds enlarged more rapidly (Figures 16--18). All axillary buds initiated in the inflorescence developed into floral
spikes (Figure 18).
An axillary bud that developed into a floral spike was enclosed by two opposite stipules (Figures 4, 5 and 19). A leaf
primordium was present in the first few stem units, which were
reproductive (Figures 4 and 5), but was absent in subsequent
more distal stem units (Figures 6 and 7). Spike primordia
elongated and initiated a series of lateral apices. Each lateral
apex initiated two large bract primordia (Figures 19--21), then
five sepals that remain fused at the base (Figures 20--22), five
216
SYAMSUWIDA AND OWENS
Figures 16--23. Inflorescence development in Shorea stenoptera
using light (LM) and scanning
electron microscopy (SEM).
Figure 16. Fixed and partially dissected compound inflorescence
bud viewed with a dissecting LM
after outer stipules were removed
to reveal inner stipule (S), leaf primordia (L1--L4), large axillary floral spike buds (SP) and terminal
inflorescence bud (arrowhead).
Figure 17. LM of median longitudinal section of a young floral
spike in the axil of a leaf (L)
showing bract (B) primordia. Axillary flower primordia are out of
the plane of view or not yet initiated.
Figure 18. LM of median longitudinal section of an inflorescence
bud showing floral spike primordia at various stages of development.
Figure 19. SEM of a floral spike
primordium elongating beyond
basal stipules showing bract primordia with axillary flower apices (arrows) and the distal spike
apex (arrowhead) starting to initiate more bracts.
Figure 20. SEM of a developing
spike after most bracts were removed. Bracts on lower side of
photograph cover spike apex and
youngest flower primordia.
Shown clockwise, three stages of
flower development in above
flower primordia. Five sepals ( )
have been initiated, three large
and two small, and inside these
are five petal primordia.
Figure 21. SEM of newly initiated floral spike showing distal spike apex (arrowhead), paired bract primordia (one removed) and flower primordium
showing five sepals and onset of petal initiation.
Figure 22. SEM of flower primordium at slightly later stage than in Figure 21, showing five sepal primordia ( ) and five petal primordia on
star-shaped flower primordium.
Figure 23. SEM of a flower bud at latest stage shown in Figure 20 but with bracts, sepals and petals removed to show the 15 stamen primordia.
petals fused along most of their length forming a coronate
flower bud (Figures 20--22), fifteen stamens (Figure 23) fused
to the base of the petals, and finally, a central pistil with three
locules and a single style. Proximal spikes were short with only
one or few flowers, central spikes bore many flowers and distal
spikes decreased in length and number of flowers acropetally
until the most distal spikes usually bore only one flower (Figure 7). The inflorescence usually terminated in a single or pair
of flowers (Figure 7). Occasionally, the inflorescence apex
reverted back to a vegetative apex and resumed leaf initiation.
This resulted in one or more floral spikes along an otherwise
vegetative branch. Floral spikes had no arrested period of
FLOWER AND FRUIT DEVELOPMENT IN SHOREA STENOPTERA
development. Within an inflorescence, they developed, elongated and burst from the stipules in an acropetal manner over
several weeks (Figure 6).
Floral enhancement
Dissections of buds collected in July showed that floral initiation had begun in some branches of some trees before the
paclobutrazol treatments were applied in mid-July. In September, nonfruiting inflorescences were visible, but there were no
significant differences among treatments. No observations
were made in October. In November, there were abundant
flowering and fruiting inflorescences per tree. Statistical
analyses showed that paclobutrazol concentration had a significant effect (P = 0.0463) on the number of branches bearing
flowering inflorescences in November only. The Duncan’s
multiple range test (α = 0.05) verified that paclobutrazol appeared to depress the number of branches bearing flowering
inflorescences relative to the control treatment. However, the
numbers of fruiting and flowering inflorescences combined
were not influenced by paclobutrazol on any observation date.
The treatment differences in November suggest that, although
paclobutrazol did not influence the total number of flowering
and fruiting inflorescences, it did hasten fruit development so
that control trees had significantly more flowering inflorescences but fewer fruiting inflorescences than paclobutrazoltreated trees. In January, there was a marked decrease in
flowering and fruiting inflorescences because flowering had
ceased and many inflorescences had shed their fruits (Figure 24).
There was no carry-over effect of paclobutrazol in the 1995
flowering--fruiting season. Flowering was sparse and random
among the control and treated trees.
Figure 24. Histogram showing percent of branches bearing flowering
and fruiting inflorescences for control (C) and low (L) and high (H)
concentrations of paclobutrazol over four months.
217
Discussion
The development of vegetative buds and the initiation or development of inflorescences have not been described for any
dipterocarp. Based on our observations of species of Dipterocarpus, Hopea, and Shorea, at least these major genera have
very similar vegetative bud and shoot development. Shorea
stenoptera vegetative buds were enclosed by two large, red,
opposite, obtuse stipules, the bases of which encircled the stem
just below the leaf axil. These stipules adhered along their
margins, and opened as the leaf, shoot axis and reiterated
terminal bud inside elongated. Five to seven reiterative buds
were enclosed within the outermost stipules (Figure 9). Reiteration is a well documented developmental pattern in many
woody perennials, and is well represented in the above genera
and easily observed in S. stenoptera. The predictability of this
pattern of development means that the position of potential
inflorescence primordia as axillary buds is easily seen and the
time of floral initiation can be easily determined by dissecting
terminal buds.
We found that shoot elongation and flowering in S. stenoptera began during the dry season, June through September, but
continued through the rainy season, December and January
(Figure 25). This agrees with the results of van Schaik et al.
(1993) who found that peaks of flushing and flowering in
tropical forests occur preceding and during the early rainy
period. Water availability is thought to be important in controlling the phenologies of many tropical forest trees (Reich and
Borchert 1984). Vegetative phenology, in turn, may affect
floral initiation as it does in temperate trees (Owens and Blake
1985).
Floral initiation began in June or July and continued at least
into December. Inflorescences were at all stages of development on a tree, and many stages of development were apparent
within an inflorescence. Therefore, a single time of floral
initiation, as found in most temperate trees (Owens and Blake
1985), does not exist in S. stenoptera. As a result, it is difficult
to determine particular climatic factors causing flowering.
Several factors may be involved. Even in temperate trees, in
which climate is seasonal and time of initiation precise, many
climatic factors, including rainfall, temperature and light intensity, affect and interact to stimulate floral initiation. This
commonly occurs during or after lateral branch elongation, the
branches on which flowers are initiated (Owens and Blake
1995). Because floral initiation in S. stenoptera begins and
predominantly occurs during the early rainy season, the preceding drier period could be a factor. Vegetative growth, as
shoot elongation, may be reduced during the dry season. During the rainy season, floral initiation stops but flowering continues, though less abundantly. van Schaik et al. (1993)
demonstrated major roles for irradiance and water stress in the
control of phenology of tropical woody plants. Chaikiattiyos
et al. (1994) describe how floral induction in tropical trees
follows a check in vegetative growth, which normally occurs
during the dry season. Ashton et al. (1988) considered drought
to be an unlikely factor in floral initiation in dipterocarps; but,
until shoot water potentials are measured, as has been done in
some tropical fruit trees (Chaikiattiyos et al. 1994), drought
218
SYAMSUWIDA AND OWENS
Figure 25. Total monthly rainfall at
the study site in 1994. Horizontal
line shows average monthly rainfall.
cannot be ruled out as a factor in floral initiation in dipterocarps. The dry season is also probably accompanied by increased periods of sunshine because of less cloudy conditions.
Ng (1977) suggested that an increase of 2 h per day of sunshine
could induce flowering in dipterocarps. Less cloud may also
result in slight increases and decreases in day and night temperature, respectively.
Although we were unable to identify the mechanism by
which environmental cues affect floral initiation in dipterocarps, we demonstrated that the time of floral initiation can be
determined by bud dissection. Floral spikes arose from the
axillary buds and the latter were always present within the
reiterated buds. Also, the early axillary inflorescence buds
were easily distinguished from axillary vegetative buds. Although bud dissection will reveal only after the fact that floral
initiation has begun, this information is useful because floral
initiation likely occurs at about the same time each year and
over several months. Therefore, floral enhancement may be
possible before or during the time when inflorescence buds
become visible.
Paclobutrazol affected flowering and fruiting, but not in the
manner we expected. In September, there was a trend toward
increased flowering with increasing concentration of paclobutrazol (Figure 24); however, this trend was not statistically
significant. In November, control trees had more flowers than
trees in either paclobutrazol treatment, but no significant treatment effects were found when numbers of flowering and
fruiting inflorescences were combined. This, together with our
developmental data, leads us to conclude that the application
of paclobutrazol was either too late or at too low a concentration to affect floral initiation, but that it hastened floral and fruit
development. This is consistent with effects of paclobutrazol
on fruit trees where it acts as a growth retardant by suppressing
gibberellin production, thereby reducing cell division and cell
expansion (Dalziel and Lawrence 1984). Reduction in vegetative growth allows greater partitioning of assimilates to reproductive growth, flower and fruit formation and fruit growth
(Lever 1986). In apple (Quinlan and Richardson 1986), peach
(Erez 1986), clove (Martin and Dabek 1988) and Eucalyptus
(Griffin et al. 1993, Moncur et al. 1994), paclobutrazol reduced
vegetative growth and enhanced fruit production, but in all
cases, the time of treatment was important. Our results together
with those listed above indicate that paclobutrazol may be
useful for enhancing fruit production in some dipterocarps. It
may be especially beneficial if applied before floral initiation
to young dipterocarps thereby reducing vegetative growth over
several years and thus making the trees a more manageable
size for seed collection (Erez 1986).
Acknowledgments
This research was supported by the ASEAN Forest Tree Seed Centre
Project (Muak Lek, Saraburi, Thailand), the Forest Tree Improvement
Research and Development Institute (Yogyakarta, Indonesia) and
through a Canadian NSERC Operating Grant (No. 1982) to J.N.O.
Appreciation is extended to Parlin Tambunan for collection of specimens and flowering data, to the staff at The Kebun Percorbaan Arboretum, west Java, Indonesia for use of the trees and assistance in
collection of specimens, Glenda Catalano for technical assistance, Dr.
Derek Harrison for statistical assistance and Diane Gray for typing the
manuscript.
References
Ashton, P.S. 1982. Dipterocarpaceae. In Flora Malesiana Ser. I. Spermatophyta. Ed. C.G.G.J. van Steenis. 9:237--552.
Ashton, P.S. 1988. Dipterocarp biology as a window to the understanding of tropical forest structure. Annu. Rev. Ecol. Syst. 19:347-370.
Ashton, P.S., T.J. Givnish and S. Appanah. 1988. Staggered flowering
in the Dipterocarpaceae: new insights into floral induction and
evolution of mass fruiting in the aseasonal tropics. Am. Nat.
132:44--66.
Burgess, P.F. 1972. Studies of the regeneration of the hill forests of the
Malay Peninsula. Malay. For. 35:103--123.
Chan, H.T. 1980. Reproductive biology of some Malaysian dipterocarps. II. Fruiting biology and seedling studies. For. Ecol. 43:438-451.
Chaikiattiyos, S., C.M. Menzel and T.S. Rasmussen. 1994. Floral
induction in tropical fruit trees: effects of temperature and water
supply. J. Hortic. Sci. 69:397--415.
FLOWER AND FRUIT DEVELOPMENT IN SHOREA STENOPTERA
Dalziel, J. and D.K. Lawrence. 1984. Biochemical and biological
effects of kaurene oxidase inhibitors such as paclobutrazol. In
Biochemical Aspects of Synthetic and Natural Occurring Plant
Growth Regulators. British Plant Growth Regulator Group Monogr.
11:43--57.
Erez, A. 1986. Effect of soil applied paclobutrazol in drip irrigated
peach orchards. Acta Hortic. 179:513--520.
Gifford, E.M. 1950. Softening refractory plant material embedded in
paraffin. Stain Technol. 25:161--162.
Griffin, A.R., P. Whiteman, T. Rudge, I.P. Burgess and M. Moncur.
1993. Effect of paclobutrazol on flower-bud production and vegetative growth in two species of Eucalyptus. Can. J. For. Res. 23:640-647.
Johansen, D.A. 1950. Plant embryology. Chronica Botanica Co.,
Waltham, MA, 305 p.
Lever, B.G. 1986. ‘‘Cultar’’----A technical overview. Acta Hortic.
179:459--466.
Martin, P.J. and A.J. Dabek. 1988. Effects of paclobutrazol on the
vegetative growth and flowering of young clove trees. Trop. Agric.
(Trinidad) 65:25--28.
Moncur, M.W., G.F. Rasmussen and O. Hasan. 1994. Effect of
paclobutrazol on flower-bud production in Eucalyptus nitens espalier seed orchards. Can. J. For. Res. 24:46--49.
Ng, F.S.P. 1966. Age at first flowering in Dipterocarps. Malays. For.
19:290--295.
Ng, F.S.P. 1977. Gregarious flowering of dipterocarps in Kepong,
1976. Malay. For. 40:126--137.
219
Owens, J.N. and M.D. Blake. 1985. Forest tree seed production. A
review of literature and recommendations for future research. Can.
For. Serv. Inf. Rep. PI-X-53, 161 p.
Pharis, R.P. 1976. Manipulation of flowering in conifers through use
of plant hormones. In Modern Methods in Forest Genetics. Ed. M.P.
Miksche. Springer-Verlag, New York, pp 265 − 282.
Quinlan, J.D. and P.J. Richardson. 1986. Uptake and translocation of
paclobutrazol and implications for orchard use. Acta Hortic.
179:443--451.
Reich, P.B. and R. Borchert. 1984. Water stress and tree phenology in
a tropical dry forest in the low lands of Costa Rica. J. Ecol.
72:61--74.
Sasaki, S. 1980. Storage and germination of dipterocarp seeds. Malay.
For. 43:290--308.
Smitinand, T., T. Santisuk and C. Phengklai. 1980. The manual of
Dipterocarpaceae of mainland south-east Asia. The Forest Herbarium, Royal Forestry Dept., Bangkok, Thailand, 133 p.
Steffens, G.L., F.W. Jacobs and M.E. Engelhaupt. 1993. Size, flowering and fruiting of maturing own-rooted ‘‘Gala’’ apple trees treated
with paclobutrazol sprays and trunk drenches. Sci. Hortic. 56:13-21.
van Schaik, C.P., J.W. Terborgh and S.J. Wright. 1993. The phenology
of tropical forests: adaptive significance and consequences for primary consumers. Annu. Rev. Ecol. Syst. 24:353--377.
Wycherley, P.R. 1973. The phenology of plants in the humid tropics.
Micronesia 9:75--96.
Zar, J.H. 1984. Biostatistical analysis. 2nd Edn. Prentice-Hall, Englewood Cliffs, NJ, 620 p.
© 1997 Heron Publishing----Victoria, Canada
Time and method of floral initiation and effect of paclobutrazol on
flower and fruit development in Shorea stenoptera (Dipterocarpaceae)
DIDA SYAMSUWIDA1 and JOHN N. OWENS2
1
Forest Tree Improvement Research and Development Institute, Jl. A.M. Sangaji Km. 15, Purwobinangun, Pakem, Sleman, DI Yogyakarta 55582,
Indonesia
2
Centre for Forest Biology, University of Victoria, Victoria, B.C. V8W 2Y2, Canada
Received January 11, 1996
Summary Small Shorea stenoptera Burck. (Dipterocarpaceae) trees of reproductive age growing in an arboretum in
west Java were studied to determine the pattern of vegetative
shoot development, the time and method of floral initiation and
the effect of paclobutrazol on floral enhancement. Vegetative
buds were enclosed by two stipules between which was a leaf
primordium, a small axillary vegetative bud and another pair
of stipules. This sequence was reiterated five to seven times
before the vegetative apex was visible. At the time of floral
initiation, axillary buds developed into floral spikes and compound inflorescences formed at the end of drooping branches.
A compound inflorescence might bear many floral spikes and
each floral spike bore many flowers. The compound inflorescence was a modification of the reiterative developmental
pattern observed in vegetative shoots. The time of floral initiation began in late June or early July and continued until about
November. Floral enhancement using paclobutrazol as a soil
drench was attempted in mid-July, but this was later found to
be after the onset of floral initiation, and the treatment failed
to enhance flowering; however, it appeared to enhance the rate
of floral and fruit development. The similarity in vegetative bud
development among dipterocarp genera suggests that the time
of floral initiation may be easily determined in many species
based on simple dissection techniques.
Keywords: dipterocarp, floral enhancement, shoot development.
Introduction
The family Dipterocarpaceae may account for 10% of all tree
species and 80% of all emergent trees in tropical rain forests
(Ashton 1982). Dipterocarps occur in both aseasonal lowland
rain forests where they are evergreen, and in seasonal upland
or more xerophytic forests where they are deciduous (Smitinand et al. 1980).
A striking feature of many dipterocarp forests is the phenomenon of mass flowering followed by mass fruiting (Ashton
et al. 1988). Mass flowering occurs at intervals of two to 10
years, over a period of a few weeks to a few months in either
small areas or large regions such as the Malaysian Peninsula.
Large dipterocarps may produce up to four million flowers and
120,000 fruits (Ashton 1988). Although the fruits are poorly
protected and are eaten in vast numbers by wild pigs and
heavily parasitized by weevils (Chan 1980, Ashton 1988), the
forest floor following a mass fruiting is often blanketed by
dipterocarp seedlings. In some regions, reforestation is
achieved by lifting and transplanting the wild seedlings.
Managing dipterocarps for seed production is difficult. First
flowering commonly does not occur until trees are 20 to 30
years old (Ng 1966), by which time most are too large for easy
management or seed harvest. Fruit production is episodic, and
because seeds of most species are recalcitrant and store poorly
(Sasaki 1980), there can be long intervals during which seed
for reforestation is unavailable. Although flowers have three
locules and two ovules per locule, normally only one indehiscent seed develops per fruit. To overcome these obstacles,
methods are needed to induce flowering at regular intervals in
relatively small trees.
Effective control of flowering, requires knowledge of the
time of floral initiation and the development of methods of
floral induction or enhancement (Owens and Blake 1985). The
time of natural floral initiation in dipterocarps has never been
determined developmentally or experimentally. Dipterocarps
show episodic shoot growth with distinct periodic vegetative
flushes, followed by periods of shoot elongation. Flushes commonly precede or coincide with flowering, and inflorescences
tend to be borne near the ends of shoots (Burgess 1972).
Stimuli that induce flushing and flowering in dipterocarps are
not known. Ashton et al. (1988) proposed that flowering is
controlled by an endogenous rhythm, perhaps related to episodic shoot flushes and triggered by an external cue. Wycherley (1973) suggested that temperature may be responsible for
floral induction in dipterocarps. Ashton et al. (1988) found no
close correlation between flowering intensity of Shorea species and photoperiod or water availability, and proposed that
exposure to a 2 °C decrease in minimum nighttime temperature for several nights may be sufficient to induce flowering in
some dipterocarps.
Increased temperature and drought, usually in conjunction
with gibberellin A 4/7 application, are common cultural treatments applied to induce cones in Pinaceae (Owens and Blake
212
SYAMSUWIDA AND OWENS
1985). The mode of action of these treatments, however, has
yet to be demonstrated. One hypothesis is that these cultural
treatments reduce vegetative growth, and the resulting excess
gibberellins promote flowering (Pharis 1976). Gibberellins
have not been found to enhance flowering in hardwood forest
trees. However, flowering in several temperate fruit trees
(Lever 1986, Steffens et al. 1993), clove (Martin and Dabek
1988) and Eucalyptus (Griffin et al. 1993) is enhanced when
the growth retardant paclobutrazol is applied as a foliar spray,
stem injection or soil drench.
The objectives of this study were to determine the time and
method of floral initiation and to investigate the use of
paclobutrazol as a method of enhancing flowering in young,
reproductively mature Shorea stenoptera Burck. trees.
Materials and methods
In June 1994, 27 reproductively mature S. stenoptera trees that
had previously flowered were selected from a stand of over 100
trees at the Kebun Percorbaan Arboretum in west Java, Indonesia (lat. 6° N, long. 106° E). Trees were grouped into three
categories, with nine trees in each group, on the basis of
circumference at breast height: Group I, 51--55 cm; Group II,
46--50 cm; Group III, 41--45 cm. Three trees of each size class
were randomly assigned to each of three paclobutrazol treatments: A, high concentration; B, low concentration; and C,
control.
Floral initiation
From mid-July through October, branches were sampled from
several trees every month. The terminal bud on each branch
was removed and placed in a vial with formalin/acetic acid/alcohol (FAA) fixative (Johansen 1950). Buds were stored in
FAA and taken to the University of Victoria where a central
longitudinal section 1--2 mm was embedded in ‘‘Tissuemat’’
(Fisher Scientific, Ottawa, Canada) using the tertiary-butyl
alcohol dehydration series of Johansen (1950). Embedded
specimens were softened in Gifford’s softening reagent (Gifford 1950) for 2 to 4 weeks and then serially sectioned longitudinally in increments of 6 µm. The median sections were
mounted on microscope slides and stained with safranin and
fast green and examined to determine whether they were vegetative or reproductive.
Other buds were dissected by removing successive stipules
until the terminal bud or apex and the axillary vegetative buds
or inflorescence primordia were visible. The shoot tips, or
portions of them, were dehydrated in an ethanol series, critical
point dried, mounted on aluminum stubs, sputter-coated with
gold and observed with a JOEL JSM-35 scanning electron
microscope (SEM) (JOEL, Boston, MA) at 10 kV. Other buds
were dissected, sketched and photographed with a Wild M 400
Photomacroscope (Leica Canada Inc., Willowdale, Canada) to
show bud morphology. In all three types of specimen observation, vegetative buds and buds that had undergone floral initiation and various stages of floral development were observed
and photographed.
Floral enhancement
In mid-July, sample trees were treated once with one of three
paclobutrazol treatments. A shallow circular trench (about
10-cm-deep) was dug at a radius of 1 m from the base of each
tree. Paclobutrazol in the water soluble form ‘‘Cultar’’ was
obtained from ICI (Indonesian Operations, Jakarta). Concentrations and application methods were based on successful
experiments with Eucalyptus (Griffin et al. 1993, Moncur et al.
1994). Paclobutrazol treatments were either high concentration (1.0 g of paclobutrazol per 15 cm of trunk circumference),
low concentration (0.5 g of paclobutrazol per 15 cm of trunk
circumference) or a control (water only). The appropriate
amount of paclobutrazol was added to 4 liters of water, which
was then mixed and poured into the trench around each tree.
Once per month in September, November, December and
January 1994, trees were climbed and the numbers of inflorescences with flowers and with developing fruits counted. The
number of flowering compound inflorescences and number of
fruiting compound inflorescences were counted on each observation date. Numbers of flowers or fruits per compound inflorescence or per spike were not counted. Because trees varied
in size, numbers of branches and sites in which inflorescences
formed and the number of primary branches per tree were
counted to determine the percentage of branches bearing inflorescences.
To observe any treatment carry-over effect, trees were assessed for frequency of flowering in November 1995, i.e., the
year following paclobutrazol treatment.
Average rainfall per month was obtained from the weather
station at the study site.
Figures 1--8. Vegetative and reproductive shoots and buds and fruits of
Shorea stenoptera.
Figure 1. Vegetative branch of mature tree showing red stipules (S).
Figure 2. Close-up of vegetative branch showing stipules and emerging leaf (L).
Figure 3. Partially dissected vegetative bud with outer stipule removed
to reveal the leaf primordium, axillary buds (A) and small green stipule
inside.
Figure 4a. Vegetative bud with one stipule removed to show leaf
primordium and stipule within and axillary bud at the onset of floral
initiation. Figure 4b. Compound inflorescence with outer stipule removed to show enlarged axillary inflorescence (AI).
Figure 5. Later stage of a compound inflorescence with two stipules
removed to show two stages of axillary inflorescence development.
Figure 6. Compound inflorescence during flowering showing red
stipules at the base of each spike (SP), and developing fruits ( ).
Figure 7. Distal portion of a compound inflorescence showing several
spikes. Most flowers (F) have been shed and one is aborting ( )
leaving the subtending bracts (B). Most stipules have been shed.
Figure 8. Three nearly mature globose fruits showing leaf-like wings.
All other flowers abscised.
FLOWER AND FRUIT DEVELOPMENT IN SHOREA STENOPTERA
213
214
SYAMSUWIDA AND OWENS
Statistical analysis
Numbers of inflorescences bearing flowers and fruits were
analyzed separately, or summed for each date and analyzed by
a balanced two-factor analysis of variance (ANOVA) design
with size class and paclobutrazol dosage as fixed effects.
Because size class did not influence the number of branches
bearing flower or fruit inflorescences (α = 0.05), all size
classes were pooled and subjected to a single-factor ANOVA
for each date. Significant effects were further analyzed with
the Duncan’s multiple range test (Zar 1984).
Observations and results
Shorea stenoptera formed pendant compound infloresences at
the ends of long, often drooping branches in younger trees
(Figure 1), and on shorter secondary and tertiary branches in
older trees. The inflorescences, which were initiated and underwent early development within the terminal, partially
fused, protective, stipules on the branch (Figures 2 and 3),
commonly had seven to 15 floral spikes (Figure 6). Spikes
were spirally arranged but appeared alternate, and were separated by long internodes (Figure 7). Spikes matured acropetally on the branch and flowers matured acropetally on each
spike (Figure 6). Each spike was enclosed by two stipules
during early floral development (Figures 4 and 5), and bore one
to usually many spirally arranged flowers (Figure 7), each on
a short pedicel. Each flower had two opposite subtending
bracts. Most flowers abscised at the base of the pedicel, leaving
the two bracts that usually abscised a few days later (Figure 7).
Rarely did more than one fruit develop per spike. Fruits rapidly
enlarged, and the five sepals elongated, enclosing the globose
fruit at their base, and forming three large and two small, broad
flat wings. Spikes were pendant and the wings hung downward
(Figure 8).
Vegetative bud development
Development within the vegetative bud provided a remarkable
example of reiteration. Each bud was enclosed by two large
spoon-shaped red stipules, each of which attached around half
of the stem and the base of a leaf. The two stipules adhered
along their margins. Between the pair of stipules was a leaf, an
axillary bud and another pair of smaller green stipules (Figures
2, 3 and 4a). Within these stipules was, again, a leaf, an axillary
bud and a yet smaller pair of stipules. These structures were
often reiterated five to seven times (Figure 9) as revealed by
dissection (Figures 3, 4 and 10) and SEM, before the shoot
apical meristem was visible (Figure 11). The inner and, to a
lesser extent, the outer surfaces of the young stipules were
covered by long epidermal hairs (Figure 11) that stuck to the
leaf primordia and other stipules.
The apical meristem was a low dome with an oval cross
section, from which successive leaf primordia arose at opposite ends (Figure 12). Soon after a leaf primordium was initiated, two ridges of meristematic tissue arose at the primordium
base along the sides of the oval (Figure 12). These ridges
extended around the apical meristem and base of the leaf
primordium, and elongated upward to form two opposite
stipule primordia (Figure 13). As the stipule primordia elongated, they became spoon-shaped and contacted one another
along their margins. Epidermal hairs formed on the stipules,
causing them to stick together along most of their margins,
completely enclosing the leaf primordium and apical meristem
within (Figure 11). As the leaf primordium elongated and a leaf
blade began to form, a lateral bud primordium arose in the axil
of the leaf primordium (Figure 12). By this time, the next leaf
primordium and its stipules had been initiated (Figures 12-14).
Axillary buds on most vegetative shoots remained small.
The axillary apex initiated two scales (Figure 12) that elongated and enclosed the apex (Figure 11). Development may be
arrested at this stage (Figure 15) or the apical meristem may
initiate leaf primordia and stipules. In the smaller trees, the
primary branches generally lacked secondary branches so axillary buds appeared to become latent (Figure 15). In larger
trees, secondary and tertiary branches formed from the axillary
buds.
Figure 9. Diagram of a Shorea
stenoptera vegetative bud with
successive stipules (S) removed
to show the enclosed leaf primordia (L) and axillary buds (A). The
sequence may be reiterated five
to seven times as shown in a--e.
FLOWER AND FRUIT DEVELOPMENT IN SHOREA STENOPTERA
215
Figures 10--15. Vegetative Shorea
stenoptera buds using light (LM)
and scanning electron microscopy
(SEM).
Figure 10. Fixed and partially dissected bud viewed with a dissecting LM. Outer stipules were
removed to show an inner stipule
(S3), its leaf primordium (L3) and
a bud (A 4) in the axil of each leaf.
Different subscripts indicate
stages of development.
Figure 11. Enlarged SEM of a bud
as in Figure 10. The stipules,
leaves and axillary buds having
the same subscript as in Figures
10 and 11 are at the same stage.
The stipule (S2) has been removed
to reveal its leaf (L 2), the apical
meristem (arrow) and the youngest leaf primordium (L1).
Figure 12. Enlarged SEM of the
apical meristem ( ) and leaf primordia (L1, L2, L3) and the axillary bud (A3) as shown in Figure
11. Two stipules (S1) are those
that belong to L1. Outer stipules
(S2) have been removed. An axillary bud (A2) has just begun to be
initiated in the axil of L 2. Axillary
bud (A3) has two lateral scales (arrowhead).
Figure 13. A later stage of stipule
development to that shown in Figure 12. Stipules (S1) for L1 have
enlarged covering the apex. The
next leaf primordium (L 0) is being
initiated.
Figure 14. Median longitudinal
section of a vegetative bud as in
Figure 13 showing the apical meristem ( ) leaf primordia (L1, L2)
and the axillary bud (A3) for leaf
three.
Figure 15. Median longitudinal
section of an axillary vegetative
bud that has become latent soon
after being initiated.
Floral initiation and development
Inflorescence initiation was observed in some buds collected
in July and in all subsequent collections through November. At
inflorescence initiation, the vegetative apex underwent transition to an inflorescence apex. The terminal inflorescence apex
was dome-shaped and initiated a spiral sequence of stem units
consisting of a leaf primordium, two stipules and an axillary
apex (Figure 16). Development of the components of the stem
units differed depending on the time of inflorescence initiation, with leaf primordia being smaller or absent in stem units
initiated later, whereas stipules and internodes were longer and
axillary buds enlarged more rapidly (Figures 16--18). All axillary buds initiated in the inflorescence developed into floral
spikes (Figure 18).
An axillary bud that developed into a floral spike was enclosed by two opposite stipules (Figures 4, 5 and 19). A leaf
primordium was present in the first few stem units, which were
reproductive (Figures 4 and 5), but was absent in subsequent
more distal stem units (Figures 6 and 7). Spike primordia
elongated and initiated a series of lateral apices. Each lateral
apex initiated two large bract primordia (Figures 19--21), then
five sepals that remain fused at the base (Figures 20--22), five
216
SYAMSUWIDA AND OWENS
Figures 16--23. Inflorescence development in Shorea stenoptera
using light (LM) and scanning
electron microscopy (SEM).
Figure 16. Fixed and partially dissected compound inflorescence
bud viewed with a dissecting LM
after outer stipules were removed
to reveal inner stipule (S), leaf primordia (L1--L4), large axillary floral spike buds (SP) and terminal
inflorescence bud (arrowhead).
Figure 17. LM of median longitudinal section of a young floral
spike in the axil of a leaf (L)
showing bract (B) primordia. Axillary flower primordia are out of
the plane of view or not yet initiated.
Figure 18. LM of median longitudinal section of an inflorescence
bud showing floral spike primordia at various stages of development.
Figure 19. SEM of a floral spike
primordium elongating beyond
basal stipules showing bract primordia with axillary flower apices (arrows) and the distal spike
apex (arrowhead) starting to initiate more bracts.
Figure 20. SEM of a developing
spike after most bracts were removed. Bracts on lower side of
photograph cover spike apex and
youngest flower primordia.
Shown clockwise, three stages of
flower development in above
flower primordia. Five sepals ( )
have been initiated, three large
and two small, and inside these
are five petal primordia.
Figure 21. SEM of newly initiated floral spike showing distal spike apex (arrowhead), paired bract primordia (one removed) and flower primordium
showing five sepals and onset of petal initiation.
Figure 22. SEM of flower primordium at slightly later stage than in Figure 21, showing five sepal primordia ( ) and five petal primordia on
star-shaped flower primordium.
Figure 23. SEM of a flower bud at latest stage shown in Figure 20 but with bracts, sepals and petals removed to show the 15 stamen primordia.
petals fused along most of their length forming a coronate
flower bud (Figures 20--22), fifteen stamens (Figure 23) fused
to the base of the petals, and finally, a central pistil with three
locules and a single style. Proximal spikes were short with only
one or few flowers, central spikes bore many flowers and distal
spikes decreased in length and number of flowers acropetally
until the most distal spikes usually bore only one flower (Figure 7). The inflorescence usually terminated in a single or pair
of flowers (Figure 7). Occasionally, the inflorescence apex
reverted back to a vegetative apex and resumed leaf initiation.
This resulted in one or more floral spikes along an otherwise
vegetative branch. Floral spikes had no arrested period of
FLOWER AND FRUIT DEVELOPMENT IN SHOREA STENOPTERA
development. Within an inflorescence, they developed, elongated and burst from the stipules in an acropetal manner over
several weeks (Figure 6).
Floral enhancement
Dissections of buds collected in July showed that floral initiation had begun in some branches of some trees before the
paclobutrazol treatments were applied in mid-July. In September, nonfruiting inflorescences were visible, but there were no
significant differences among treatments. No observations
were made in October. In November, there were abundant
flowering and fruiting inflorescences per tree. Statistical
analyses showed that paclobutrazol concentration had a significant effect (P = 0.0463) on the number of branches bearing
flowering inflorescences in November only. The Duncan’s
multiple range test (α = 0.05) verified that paclobutrazol appeared to depress the number of branches bearing flowering
inflorescences relative to the control treatment. However, the
numbers of fruiting and flowering inflorescences combined
were not influenced by paclobutrazol on any observation date.
The treatment differences in November suggest that, although
paclobutrazol did not influence the total number of flowering
and fruiting inflorescences, it did hasten fruit development so
that control trees had significantly more flowering inflorescences but fewer fruiting inflorescences than paclobutrazoltreated trees. In January, there was a marked decrease in
flowering and fruiting inflorescences because flowering had
ceased and many inflorescences had shed their fruits (Figure 24).
There was no carry-over effect of paclobutrazol in the 1995
flowering--fruiting season. Flowering was sparse and random
among the control and treated trees.
Figure 24. Histogram showing percent of branches bearing flowering
and fruiting inflorescences for control (C) and low (L) and high (H)
concentrations of paclobutrazol over four months.
217
Discussion
The development of vegetative buds and the initiation or development of inflorescences have not been described for any
dipterocarp. Based on our observations of species of Dipterocarpus, Hopea, and Shorea, at least these major genera have
very similar vegetative bud and shoot development. Shorea
stenoptera vegetative buds were enclosed by two large, red,
opposite, obtuse stipules, the bases of which encircled the stem
just below the leaf axil. These stipules adhered along their
margins, and opened as the leaf, shoot axis and reiterated
terminal bud inside elongated. Five to seven reiterative buds
were enclosed within the outermost stipules (Figure 9). Reiteration is a well documented developmental pattern in many
woody perennials, and is well represented in the above genera
and easily observed in S. stenoptera. The predictability of this
pattern of development means that the position of potential
inflorescence primordia as axillary buds is easily seen and the
time of floral initiation can be easily determined by dissecting
terminal buds.
We found that shoot elongation and flowering in S. stenoptera began during the dry season, June through September, but
continued through the rainy season, December and January
(Figure 25). This agrees with the results of van Schaik et al.
(1993) who found that peaks of flushing and flowering in
tropical forests occur preceding and during the early rainy
period. Water availability is thought to be important in controlling the phenologies of many tropical forest trees (Reich and
Borchert 1984). Vegetative phenology, in turn, may affect
floral initiation as it does in temperate trees (Owens and Blake
1985).
Floral initiation began in June or July and continued at least
into December. Inflorescences were at all stages of development on a tree, and many stages of development were apparent
within an inflorescence. Therefore, a single time of floral
initiation, as found in most temperate trees (Owens and Blake
1985), does not exist in S. stenoptera. As a result, it is difficult
to determine particular climatic factors causing flowering.
Several factors may be involved. Even in temperate trees, in
which climate is seasonal and time of initiation precise, many
climatic factors, including rainfall, temperature and light intensity, affect and interact to stimulate floral initiation. This
commonly occurs during or after lateral branch elongation, the
branches on which flowers are initiated (Owens and Blake
1995). Because floral initiation in S. stenoptera begins and
predominantly occurs during the early rainy season, the preceding drier period could be a factor. Vegetative growth, as
shoot elongation, may be reduced during the dry season. During the rainy season, floral initiation stops but flowering continues, though less abundantly. van Schaik et al. (1993)
demonstrated major roles for irradiance and water stress in the
control of phenology of tropical woody plants. Chaikiattiyos
et al. (1994) describe how floral induction in tropical trees
follows a check in vegetative growth, which normally occurs
during the dry season. Ashton et al. (1988) considered drought
to be an unlikely factor in floral initiation in dipterocarps; but,
until shoot water potentials are measured, as has been done in
some tropical fruit trees (Chaikiattiyos et al. 1994), drought
218
SYAMSUWIDA AND OWENS
Figure 25. Total monthly rainfall at
the study site in 1994. Horizontal
line shows average monthly rainfall.
cannot be ruled out as a factor in floral initiation in dipterocarps. The dry season is also probably accompanied by increased periods of sunshine because of less cloudy conditions.
Ng (1977) suggested that an increase of 2 h per day of sunshine
could induce flowering in dipterocarps. Less cloud may also
result in slight increases and decreases in day and night temperature, respectively.
Although we were unable to identify the mechanism by
which environmental cues affect floral initiation in dipterocarps, we demonstrated that the time of floral initiation can be
determined by bud dissection. Floral spikes arose from the
axillary buds and the latter were always present within the
reiterated buds. Also, the early axillary inflorescence buds
were easily distinguished from axillary vegetative buds. Although bud dissection will reveal only after the fact that floral
initiation has begun, this information is useful because floral
initiation likely occurs at about the same time each year and
over several months. Therefore, floral enhancement may be
possible before or during the time when inflorescence buds
become visible.
Paclobutrazol affected flowering and fruiting, but not in the
manner we expected. In September, there was a trend toward
increased flowering with increasing concentration of paclobutrazol (Figure 24); however, this trend was not statistically
significant. In November, control trees had more flowers than
trees in either paclobutrazol treatment, but no significant treatment effects were found when numbers of flowering and
fruiting inflorescences were combined. This, together with our
developmental data, leads us to conclude that the application
of paclobutrazol was either too late or at too low a concentration to affect floral initiation, but that it hastened floral and fruit
development. This is consistent with effects of paclobutrazol
on fruit trees where it acts as a growth retardant by suppressing
gibberellin production, thereby reducing cell division and cell
expansion (Dalziel and Lawrence 1984). Reduction in vegetative growth allows greater partitioning of assimilates to reproductive growth, flower and fruit formation and fruit growth
(Lever 1986). In apple (Quinlan and Richardson 1986), peach
(Erez 1986), clove (Martin and Dabek 1988) and Eucalyptus
(Griffin et al. 1993, Moncur et al. 1994), paclobutrazol reduced
vegetative growth and enhanced fruit production, but in all
cases, the time of treatment was important. Our results together
with those listed above indicate that paclobutrazol may be
useful for enhancing fruit production in some dipterocarps. It
may be especially beneficial if applied before floral initiation
to young dipterocarps thereby reducing vegetative growth over
several years and thus making the trees a more manageable
size for seed collection (Erez 1986).
Acknowledgments
This research was supported by the ASEAN Forest Tree Seed Centre
Project (Muak Lek, Saraburi, Thailand), the Forest Tree Improvement
Research and Development Institute (Yogyakarta, Indonesia) and
through a Canadian NSERC Operating Grant (No. 1982) to J.N.O.
Appreciation is extended to Parlin Tambunan for collection of specimens and flowering data, to the staff at The Kebun Percorbaan Arboretum, west Java, Indonesia for use of the trees and assistance in
collection of specimens, Glenda Catalano for technical assistance, Dr.
Derek Harrison for statistical assistance and Diane Gray for typing the
manuscript.
References
Ashton, P.S. 1982. Dipterocarpaceae. In Flora Malesiana Ser. I. Spermatophyta. Ed. C.G.G.J. van Steenis. 9:237--552.
Ashton, P.S. 1988. Dipterocarp biology as a window to the understanding of tropical forest structure. Annu. Rev. Ecol. Syst. 19:347-370.
Ashton, P.S., T.J. Givnish and S. Appanah. 1988. Staggered flowering
in the Dipterocarpaceae: new insights into floral induction and
evolution of mass fruiting in the aseasonal tropics. Am. Nat.
132:44--66.
Burgess, P.F. 1972. Studies of the regeneration of the hill forests of the
Malay Peninsula. Malay. For. 35:103--123.
Chan, H.T. 1980. Reproductive biology of some Malaysian dipterocarps. II. Fruiting biology and seedling studies. For. Ecol. 43:438-451.
Chaikiattiyos, S., C.M. Menzel and T.S. Rasmussen. 1994. Floral
induction in tropical fruit trees: effects of temperature and water
supply. J. Hortic. Sci. 69:397--415.
FLOWER AND FRUIT DEVELOPMENT IN SHOREA STENOPTERA
Dalziel, J. and D.K. Lawrence. 1984. Biochemical and biological
effects of kaurene oxidase inhibitors such as paclobutrazol. In
Biochemical Aspects of Synthetic and Natural Occurring Plant
Growth Regulators. British Plant Growth Regulator Group Monogr.
11:43--57.
Erez, A. 1986. Effect of soil applied paclobutrazol in drip irrigated
peach orchards. Acta Hortic. 179:513--520.
Gifford, E.M. 1950. Softening refractory plant material embedded in
paraffin. Stain Technol. 25:161--162.
Griffin, A.R., P. Whiteman, T. Rudge, I.P. Burgess and M. Moncur.
1993. Effect of paclobutrazol on flower-bud production and vegetative growth in two species of Eucalyptus. Can. J. For. Res. 23:640-647.
Johansen, D.A. 1950. Plant embryology. Chronica Botanica Co.,
Waltham, MA, 305 p.
Lever, B.G. 1986. ‘‘Cultar’’----A technical overview. Acta Hortic.
179:459--466.
Martin, P.J. and A.J. Dabek. 1988. Effects of paclobutrazol on the
vegetative growth and flowering of young clove trees. Trop. Agric.
(Trinidad) 65:25--28.
Moncur, M.W., G.F. Rasmussen and O. Hasan. 1994. Effect of
paclobutrazol on flower-bud production in Eucalyptus nitens espalier seed orchards. Can. J. For. Res. 24:46--49.
Ng, F.S.P. 1966. Age at first flowering in Dipterocarps. Malays. For.
19:290--295.
Ng, F.S.P. 1977. Gregarious flowering of dipterocarps in Kepong,
1976. Malay. For. 40:126--137.
219
Owens, J.N. and M.D. Blake. 1985. Forest tree seed production. A
review of literature and recommendations for future research. Can.
For. Serv. Inf. Rep. PI-X-53, 161 p.
Pharis, R.P. 1976. Manipulation of flowering in conifers through use
of plant hormones. In Modern Methods in Forest Genetics. Ed. M.P.
Miksche. Springer-Verlag, New York, pp 265 − 282.
Quinlan, J.D. and P.J. Richardson. 1986. Uptake and translocation of
paclobutrazol and implications for orchard use. Acta Hortic.
179:443--451.
Reich, P.B. and R. Borchert. 1984. Water stress and tree phenology in
a tropical dry forest in the low lands of Costa Rica. J. Ecol.
72:61--74.
Sasaki, S. 1980. Storage and germination of dipterocarp seeds. Malay.
For. 43:290--308.
Smitinand, T., T. Santisuk and C. Phengklai. 1980. The manual of
Dipterocarpaceae of mainland south-east Asia. The Forest Herbarium, Royal Forestry Dept., Bangkok, Thailand, 133 p.
Steffens, G.L., F.W. Jacobs and M.E. Engelhaupt. 1993. Size, flowering and fruiting of maturing own-rooted ‘‘Gala’’ apple trees treated
with paclobutrazol sprays and trunk drenches. Sci. Hortic. 56:13-21.
van Schaik, C.P., J.W. Terborgh and S.J. Wright. 1993. The phenology
of tropical forests: adaptive significance and consequences for primary consumers. Annu. Rev. Ecol. Syst. 24:353--377.
Wycherley, P.R. 1973. The phenology of plants in the humid tropics.
Micronesia 9:75--96.
Zar, J.H. 1984. Biostatistical analysis. 2nd Edn. Prentice-Hall, Englewood Cliffs, NJ, 620 p.