Comparative growth of four Syzygium species within simulated shade environments of a Sri Lankan rain forest

Abstract

In this study we tested the hypothesis that related tree species within the timber tree genus Syzygium differ in their shade-tolerance. We propose that difference in tolerance relate to the successional status and site affinities of each species found within the rain forests of southwest Sri Lanka. Seedlings of each of the four Syzygium species were grown for 24 months in replicated environmental treatments that simulated six different shade quality and quantities recorded from a Sri Lankan rain forest. Treatments were: (i) a deep uniform shade (DS) environment that comprised only 1% of photosynthetic photon flux density (PFD) as compared to that of the full open; (ii) a medium uniform shade (MS) environment receiving 14% of PFD as compared to the full open; (iii) a light uniform shade (LS) environment receiving 50% of PFD; (iv) the center environment of a small 200 m2 opening (SD), receiving 18% of PFD; (v) the center environment of a large 400 m2 canopy opening (LD), receiving 54% of PFD; and (vi) full sun (FS) receiving 100% of PFD. All species increased both above- and below-ground growth with increasing amounts of PFD. Seedling height, root collar diameter and dry mass gain were greatest in the brighter shade treatments with little discrimination shown among LD, LS, and FS. Significant differences in growth also occurred among the four species. Comparisons among species in the full sun (FS) treatment revealed S. rubicundum and S. operculatum to have greater height increments than S. makul and S. firmum. The low leaf mass ratio of S. operculatum, in particular, and S. rubicundum, suggests both to be prone to wilt during periods of desiccation. S. rubicundum also had greatest leaf and branch numbers and smallest leaves compared to the other three species. S. firmum in particular, but also S. makul, had larger, thicker leaves, with greater total dry mass in the FS treatment compared to the other two species. In the deep shade treatment (DS) S. firmum had greatest total dry mass and S. operculatum had the least. Taken together these findings reveal S. rubicundum and S. operculatum to be the most shade-intolerant of the four Syzygium species. Both appear prone to desiccation and water loss, though we speculate the small, numerous leaves and fine branches of S. rubicundum (characteristics of more drought-tolerant species) make this species less so. Both S. firmum and S. makul do best in the brighter shade treatments. Compared to the other two Syzygium spp., both are less susceptible to desiccation in high light environments because of their larger, thicker leaves and greater bulk. S. firmum appears to be the most shade-tolerant of the four Syzygium species. Findings have direct implications for forest management. To secure regeneration establishment and release or to create suitable planting environments Syzygium spp. require silvicultural treatments that account for species specific limitations of site (water availability) and shade (canopy opening size).

Full text

Page 1

Comparative growth of four Syzygium species within simulated
shade environments of a Sri Lankan rain forest
B.M.P. Singhakumaraa, Harshi K. Gamagea,1, Mark S. Ashtonb,*
aDepartment of Forestry and Environmental Science, University of Sri Jayawardenapura, Nugegoda, Sri Lanka
bSchool of Forestry and Environmental Studies, Yale University, New Haven, CT 06511, USA
Received 10 September 2001; accepted 18 February 2002
Abstract
In this study we tested the hypothesis that related tree species within the timber tree genus Syzygium differ in their shade-
tolerance. We propose that difference intolerance relate tothe successional status and site affinities of each species found within
the rain forests of southwest Sri Lanka. Seedlings of each of the four Syzygium species were grown for 24 months in replicated
environmental treatments that simulated six different shade quality and quantities recorded from a Sri Lankan rain forest.
Treatments were: (i) a deep uniform shade (DS) environment that comprised only 1% of photosynthetic photon flux density
(PFD) as compared to that of the full open; (ii) a medium uniform shade (MS) environment receiving 14% of PFD as compared
to the full open; (iii) a light uniform shade (LS) environment receiving 50% of PFD; (iv) the center environment of a small
200 m2opening(SD), receiving 18%of PFD; (v) the center environment of a large 400 m2canopy opening (LD), receiving 54%
of PFD; and (vi) full sun (FS) receiving 100% of PFD. All species increased both above- and below-ground growth with
increasing amounts of PFD. Seedling height, root collar diameter and dry mass gain were greatest in the brighter shade
treatments with little discrimination shown among LD, LS, and FS. Significant differences in growth also occurred among the
four species. Comparisons among species in the full sun (FS) treatment revealed S. rubicundum and S. operculatum to have
greater height increments than S. makul and S. firmum. The low leaf mass ratio of S. operculatum, in particular, and S.
rubicundum, suggests both to be prone to wilt during periods of desiccation. S. rubicundum also had greatest leaf and branch
numbers and smallest leaves compared to the other three species. S. firmum in particular, but also S. makul, had larger, thicker
leaves, with greater total dry mass in the FS treatment compared to the other two species. In the deep shade treatment (DS) S.
firmum had greatest total dry mass and S. operculatum had the least. Taken together these findings reveal S. rubicundum and S.
operculatum to be the most shade-intolerant of the four Syzygium species. Both appear prone to desiccation and water loss,
though we speculate the small, numerous leaves and fine branches of S. rubicundum (characteristics of more drought-tolerant
species)makethisspecieslessso.BothS.firmumandS.makuldobestinthebrightershadetreatments.Comparedtotheothertwo
Syzygiumspp.,botharelesssusceptibletodesiccationinhighlightenvironmentsbecauseoftheirlarger,thickerleavesandgreater
bulk. S. firmum appears to be the most shade-tolerant of the four Syzygium species. Findings have direct implications for forest
management. To secure regeneration establishment and release or to create suitable planting environments Syzygium spp. require
silvicultural treatments that account for species specific limitations of site (water availability) and shade (canopy opening size).
# 2002 Elsevier Science B.V. All rights reserved.
Keywords: Biomass allocation; Desiccation; Leaf morphology; Red:far red ratio; Seedling growth; Shade-tolerance; Syzygium spp.
Forest Ecology and Management 174 (2003) 511–520
*Corresponding author.
E-mail addresses: singha@eureka.lk (B.M.P. Singhakumara), harshi.gamage@vuw.ac.nz (H.K. Gamage), mark.ashton@yale.edu
(M.S. Ashton).
1Present address: Department of Biological Sciences, Victoria University of Wellington, P.O. Box 600, Wellington, New Zealand.
0378-1127/02/$ – see front matter # 2002 Elsevier Science B.V. All rights reserved.
PII: S0378-1127(02)00071-3

Page 2

1. Introduction
Tree seedling response to shade has long been the
subject of forest research because of importance in
predicting establishment of natural regeneration for
forestmanagement.Plantsthatareabletosurviveand
grow in the shade of a forest understory are categor-
ized as shade-tolerant (Lee et al., 1996). Shade-
intolerant species require high irradiance regimes
for establishment and growth (Whitmore, 1989;
Veneklaas and Poorter, 1998). Above- and below-
ground response of seedlings to variation in light
availability is a central area of study in plant ecology
(McConnaughay and Coleman, 1999). Understand-
ing tree species shade-tolerance in a forest is one of
the underlying principles to the development of
silvicultural regeneration techniques for forest man-
agement.
Shade-tolerant plants have evolved various adapta-
tions to utilize low levels of light in the rain forest
understory. The high leaf area per unit leaf mass, low
shoot to root ratio and smaller leaf mass per unit area,
are some of the measures that have been used to infer
shade-tolerance(DeLuciaetal.,1998;Osunkoyaetal.,
1994; Kitajima, 1994). Shade-tolerant plants allocate
most of their photosynthate to leaves in order to
increase the area available for interception of light.
Studies have therefore revealed that the shade-toler-
ance of a species is influenced by the way biomass is
partitioned between plant parts (King, 1994). Such
abilities to partition and re-allocate structural attri-
butesofroot,stemandleaftissuestomaximizegrowth
and survival in light-limited habitats is therefore a
good indicator of shade-tolerance (DeLucia et al.,
1998).
This study assessed the performance of four Syzy-
gium spp. grown over a 2-year period in shade
treatments that provided different light qualities
and quantities. The objective of this study was to
gain an understanding of how related species differ
in shade-tolerance with one another and to contri-
bute toward an autoecological information base that
can be used to develop techniques for the sustainable
management of Sri Lanka’s rain forests. We hypo-
thesized that the more shade-intolerant Syzygium
spp. would have higher growth, leaf number and
number of branches, higher dry mass and lower
leaf area than the more shade-tolerant species in
the brighter shade treatments; while the more
shade-tolerant Syzygium species would exhibit more
water-use efficient growth morphologies and would
have higher growth and leaf area, higher dry mass
and lower numbers of branches and leaf numbers
in the shade as compared to the shade-intolerant
species.
2. Site and species descriptions
The experiment was done at the Sinharaja World
Heritage site. The region is classified as a mixed-
dipterocarp forest (Gunatilleke and Ashton, 1987).
The study site comprises upland hills that range in
altitude between 300 and 700 m with a mean annual
rain fall of 5000 mm and a mean daily temperature of
27 8C (Ashton et al., 1995).
The four species selected for this study were Syzy-
gium firmum Thw., S. makul Gaertn., S. rubicundum
Wight and Arn. and S. operculatum (Roxb.) Niedz.
All belong to the family Myrtaceae and are very
common in the southwest rain forest region of Sri
Lanka. S. firmum and S. makul are endemic species to
the island. All four species are dominants of the forest
canopy (S. firmum and S. rubicundum) and sub-
canopy (S. makul and S. operculatum) of late-succes-
sional forest. They are important timber sources and
have edible fruits. Demographic data obtained from a
20 ha plot shows that at least three of the species
appear to have different site affinities across the rain
forest topography. S. operculatum occurs in lower
slopes and valleys of the rain forest topography,
where small rivers and perennial streams are found.
S. makul occupies deep soils of valleys and mid-
slopes. S. rubicundum occurs on mid-slopes that have
different aspects to those upon which S. makul is
found (see Fig. 1) (Ashton, 1981). The remaining
species, S. firmum, is found at lower elevations within
the rain forest and has therefore not been recorded in
the demographic plot.
Allfourofthesespecieshavewildlifevaluebecause
their fruit provides a considerable resource to avariety
of forest birds. Fruit from S. operculatum also pro-
vides an important source of vitamin C for children in
rural areas. S. rubicundum, S. firmum and S. makul are
important plywood and framing timbers for house
construction.
512
B.M.P. Singhakumara et al./Forest Ecology and Management 174 (2003) 511–520

Page 3

Fig. 1. Demographic stem maps of a 25 ha plot in Sinharaja for juveniles (all individuals ? 1 and <10 cm dbh; small black squares)
intermediate size trees (all individuals ? 10 and <30 cm dbh; large black square squares) and mature trees (all individuals ? 30 cm dbh; large
black circles): (A) S. makul; (B) S. operculatum; (C) S. rubicundum. The plot is aligned N–S and ranges in elevation between 400 and 560 m.
The topography of the plot comprises two slopes; one with a generally eastern and southern aspect and the other with a western and northern
aspect. The slopes are bisected approximately down the middle of the plot by a stream that runs from the northeast toward the southwest.

Page 4

3. Methods and materials
3.1. Experimental design
Twenty-four well-ventilated environmental shelters
were constructed in a large opening at the Sinharaja
field station. The shelters were designed to create
shade treatments that represented a range of photo-
synthetic photon flux densities (PFDs) and red:far
red (R:FR) ratios found within the Sinharaja forest
(Ashton, 1992).
Six combinations of irradiance and spectral quality
were created, each represented by four environmental
shelters as replicates ð4 ? 6 ¼ 24Þ (see Table 1 for
details). Treatments comprised: (i) a deep uniform
shade (DS) environment that comprised only 1% of
photosynthetic PFD as compared to that of the full
open; (ii) a medium shade (MS) environment that
simulated the inside forest edge adjacent to a
400 m2canopy opening, receiving 14% of PFD as
compared to the full open; (iii) a light uniform shade
(LS) environment that simulated an outside edge of a
400 m2canopy opening, receiving 50% of PFD; (iv)
the center environment of a small 200 m2opening
(SD), receiving 18% of PFD most of which occurred
daily as direct sunlight over a short period (2 h) of
Fig. 1. (Continued).
Table 1
The various measures of light quantity and quality in the six shade treatmentsa
DSMSLSSD LDFS
Maximum PFD recorded on clear sunny days (mmol m?2s?1)
Total daily PFD recorded on clear sunny days (mol m?2day)
R:FR ratio
Periods of direct sunlight on clear sunny days (h/day)
50
1.2
0.23
–
350 700
16.3
1.05
–
1600 1600
13.2
1600
38.16.0
0.97
–
7.4
1.27
2.5
1.27
5.5
1.27
12
aDS: deep uniform shade; MS: medium uniform shade; LS: light uniform shade; SD: small opening (short duration of direct light); LD:
large opening (long duration of direct light); FS: full sun.
514
B.M.P. Singhakumara et al./Forest Ecology and Management 174 (2003) 511–520

Page 5

time; (v) the center environment of a large 400 m2
canopy opening (LD), receiving 54% of PFD most of
which occurred daily as direct sunlight over a long
period (6 h) of time; and (vi) FS receiving 100% of
PFD. For the uniform shade treatments the quality and
quantity of irradiance were altered by spraying a parti-
cular ratio of paint pigments within a clear varnish base
onto a clear plastic film (Lee, 1985; Ashton, 1995). A
photo-spectro-radiometer (LI-1800, Li-Cor, Lincoln,
NB) verified the shade quality treatments. Light treat-
ments altering the duration of direct PFD (small open-
ing (SD) and large (LD) opening), were created by
constructing a series of parallel vertical slats aligned
north–south, horizontally placed 2 m above the ground
andacrossthecompleteinteriorofashelter.Allshelters
allowed for adequate ventilation through a system of
louvers without using electrical power.
Seed was collected from different parent trees
located in Sinharaja region. Seeds were mixed
together for each species and germinated on a nursery
bed in 50% shade. Two-month-old seedlings were
taken bare-rooted from the nursery and individually
planted in plastic bags in January 1996. Twenty-four
seedlings of each species were used in four replicates
(six seedlings per each replicate) for each shade
treatment. The total number of seedlings that were
established for the experiment was 576 (6 shade
treatments ? 4 species ? 24 seedlings/species).
Each seedling was planted in forest topsoil obtained
from one valley bottom location and mixed with sand
to improve drainage. Details on the soil nutrition of
this soil mixture have been reported by Gunatilleke
et al. (1996). The soil (3 kg) was packed into black
circular plastic bags 30 cm deep and 15 cm in dia-
meter. Bags were placed at regular intervals at
30 ? 30 cm spacing between bag centers. Watering
was done on a daily basis to insure soils approached
saturation. A black plastic lining beneath the bags
preventedroot contact with theground.Anywater that
collected on the floor was allowed to drain away
through strategically placed holes in the plastic.
3.2. Seedling growth measurements
Seedling height (from the top of apical shoot to the
ground), number of leaves, and number of branches
on the dominant apical shoot were measured every
6 months over a 2-year period from January 1996 to
1998. At the end of the 2-year period seedlings were
destructively sampled and measured for dry weight
and weight allocation to fine root, tap root, leaf and
stem tissue. Twelve seedlings per species were
selected (three per shelter) for each shade treatment.
Sample seedlingswereoven dried for48 hat80 8Cfor
dry mass measurements. Dry mass allocation to leaves
(LMR) was calculated by dividing the dry leaf mass
with the total dry mass of the whole seedling. Stem
mass ratio (SMR: dry mass of both main shoot and the
branches),androotmassratio(RMR:drymassofboth
taproot and fine roots) were calculated similarly.
Before oven drying leaf areas were estimated for
eachseedlingbyrandomlyselectingthreematureleaves
per seedling. This was done using a CID-202 leaf area
meter(CID,Vancouver,WA).Areameasurementswere
used to calculate mean area for single leaves.
3.3. Data analysis
Height increment was calculated by subtracting the
measurements taken at the initiation of the experiment
from the measurements taken at the end of the experi-
ment. A two-way ANOVA test (general linear model)
Table 2
Variance ratios following two-way ANOVA on height increment (cm), leaf number and branches, leaf size (cm2), root length (cm), root collar
diameter (cm), LMR, SMR and RMR using data from the six shade treatments (DS, MS, LS, FS, SD, LD) (*: P < 0:05;**: P < 0:01)
Height
increment (cm)
Number
of leaves
Number
of branches
Leaf size
(cm2)
Root
length (cm)
Root collar
diameter (cm)
Total dry
mass (g)
LMR SMR RMR
Shade
Species
Block
Shade ? species
***P < 0:001.
****P < 0:0001.
198.12****
8.53****
ns
3.38***
92.93****
871.53****
ns
31.27****
59.47****
171.40****
ns
7.60****
728.23****
3358.36****50.14****
ns
83.01****
1016.13****849.76****
2666.99****110.33****
180.90****
ns
26.09****
152.59****
53.59****
ns
6.20****
111.57****
277.65****
ns
5.18***
106.71****
ns
11.64****
228.50****
ns
12.05****
ns
24.24****
B.M.P. Singhakumara et al./Forest Ecology and Management 174 (2003) 511–520
515

Page 6

Table 3
(a) Means of growth performance measures: height increment (cm), root collar diameter (cm) and total dry mass (g). (b) Means of morphology
measures: leaf number/seedling, branch number/seedling, leaf size (cm2). (c) Mass ratios: leaves (LMR), stems (SMR) and roots (RMR)a
DSMS SDLSLDFS
(a) Shade treatments: growth performance measures
Height increment (cm)
S. firmum
2.75 (0.53) b
S. makul
1.00 (0.84) c
S. operculatum
2.71 (0.43) b
S. rubicundum
4.00 (1.12) a
14.79 (1.50) b
18.00 (2.68) ab
24.65 (1.65) a
25.11 (2.16) a
38.13 (2.85) ab
34.10 (2.91) b
40.43 (2.61) a
43.04 (2.29) a
26.54 (3.95) b
32.00 (1.89) ab
41.79 (3.09) a
40.45 (3.12) a
33.39 (3.49) b
33.77 (3.60) b
41.25 (2.67) a
44.86 (3.45) a
27.33 (2.41) b
28.87 (2.47) b
36.68 (3.66) ab
42.35 (3.40) a
Root collar diameter (cm)
S. firmum
S. makul
S. operculatum
S. rubicundum
0.19 (0.014) a
0.11 (0.017) b
0.11 (0.022) b
0.12 (0.013) b
0.35 (0.014) b
0.26 (0.015) c
0.45 (0.021) a
0.30 (0.018) bc
0.60 (0.023) ab
0.52 (0.029) b
0.64 (0.024) a
0.53 (0.021) b
0.58 (0.035) ab
0.55 (0.019) ab
0.68 (0.029) a
0.45 (0.032) b
0.56 (0.027) a
0.59 (0.033) a
0.65 (0.036) a
0.58 (0.026) a
0.58 (0.032) ab
0.62 (0.038) ab
0.67 (0.029) a
0.53 (0.029) b
Total dry mass (g)
S. firmum
S. makul
S. operculatum
S. rubicundum
0.84 (0.10) a
0.14 (0.02) b
0.26 (0.07) c
0.12 (0.06) b
6.66 (0.49) a
3.17 (0.28) b
8.19 (0.95) a
3.50 (0.38) b
23.70 (2.24) a
14.90 (1.75) b
15.55 (1.18) b
13.82 (1.33) b
20.24 (2.42) a
23.01 (1.70) a
18.47 (1.22) ab
11.60 (1.88) b
18.36 (1.75) a
18.59 (1.78) a
16.31 (1.68) a
17.43 (1.39) a
21.93 (1.92) ab
25.32 (2.90) a
19.70 (0.94) ab
15.81 (1.38) b
(b) Shade treatments: morphological measures
Leaf number
S. firmum
S. makul
S. operculatum
S. rubicundum
4.43 (0.50) a
4.83 (0.70) a
8.17 (1.15) a
8.20 (4.26) a
11.58 (0.77) b
15.46 (1.18) b
24.50 (2.07) b
96.63 (9.50) a
18.54 (1.22) b
25.24 (1.58) b
28.74 (1.77) b
172.74 (12.50) a
16.67 (1.56) b
33.73 (2.30) b
32.21 (2.09) b
164.00 (16.85) a
16.78 (1.06) b
26.23 (2.06) b
29.08 (1.80) b
161.73 (10.43) a
14.42 (0.89) b
26.78 (2.01) b
31.78 (1.98) b
136.84 (14.85) a
Branch number
S. firmum
S. makul
S. operculatum
S. rubicundum
0.08 (0)
0
0
0
0.54 (0.17) b
0.41 (0.14) b
0.90 (0.31) b
4.53 (0.63) a
1.21 (0.24) b
1.71 (0.27) b
0.78 (0.21) b
9.13 (1.01) a
1.13 (0.22) b
2.82 (0.35) b
1.33 (0.25) b
9.09 (0.95) a
1.00 (0.18) b
1.73 (0.19) b
0.58 (0.20) b
9.05 (0.79) a
0.79 (0.17) b
2.48 (0.32) b
1.13 (0.29) b
11.05 (1.00) a
Leaf size (cm2)
S. firmum
S. makul
S. operculatum
S. rubicundum
13.06 (1.13) a
1.34 (0.08) b
2.39 (0.46) b
0.19 (0.04) b
44.26 (2.31) a
26.68 (1.77) ab
37.67 (1.96) b
4.05 (0.20) c
62.90 (2.46) a
42.03 (1.82) b
43.53 (1.58) b
7.04 (0.20) c
53.14 (1.77) a
41.66 (1.38) b
35.31 (1.53) c
6.03 (0.19) d
63.75 (3.44) a
35.88 (1.90) b
37.41 (1.71) b
6.58 (0.25) c
54.21 (3.54) a
36.57 (1.31) b
27.03 (1.50) c
5.82 (0.21) d
(c) Shade treatments: mass ratios
LMR
S. firmum
S. makul
S. operculatum
S. rubicundum
0.55 (0.031) a
0.35 (0.025) c
0.46 (0.067) b
0.25 (0.087) d
0.59 (0.025) a
0.53 (0.037) b
0.42 (0.026) c
0.53 (0.013) b
0.57 (0.015) a
0.47 (0.021) b
0.40 (0.019) c
0.43 (0.021) bc
0.50 (0.023) a
0.45 (0.024) b
0.33 (0.023) c
0.46 (0.026) ab
0.58 (0.020) a
0.46 (0.018) b
0.39 (0.010) c
0.43 (0.020) bc
0.51 (0.022) a
0.42 (0.028) b
0.32 (0.016) c
0.38 (0.025) bc
Shoot mass ratio
S. firmum
S. makul
S. operculatum
S. rubicundum
0.27 (0.022) b
0.45 (0.037) a
0.20 (0.042) b
0.46 (0.073) a
0.19 (0.020) b
0.15 (0.019) b
0.17 (0.013) b
0.24 (0.016) a
0.23 (0.014) b
0.22 (0.013) b
0.23 (0.007) b
0.30 (0.018) a
0.21 (0.016) ab
0.19 (0.010) b
0.22 (0.013) ab
0.25 (0.011) a
0.21 (0.009) b
0.25 (0.016) b
0.24 (0.015) b
0.31 (0.010) a
0.20 (0.015) ab
0.18 (0.019) b
0.24 (0.018) a
0.24 (0.015) a
RMR
S. firmum
S. makul
S. operculatum
S. rubicundum
0.18 (0.012) c
0.20 (0.016) c
0.33 (0.078) a
0.29 (0.040) b
0.21 (0.027) c
0.32 (0.048) b
0.41 (0.037) a
0.24 (0.023) c
0.20 (0.018) c
0.30 (0.024) b
0.37 (0.023) a
0.27 (0.033) ab
0.29 (0.018) c
0.36 (0.031) b
0.45 (0.031) a
0.30 (0.029) c
0.21 (0.021) c
0.29 (0.026) b
0.37 (0.023) a
0.26 (0.023) b
0.30 (0.017) b
0.40 (0.032) a
0.44 (0.026) a
0.38 (0.037) ab
aMeans are given for the four species across six different shade treatments (DS: deep shade; MS: medium shade; SD: small opening; LS:
light shade; LD: large opening; FS: full sun). Means across species for each shade treatment sharing the same letter are not significantly
different at P < 0:05% level. Data in parentheses are standard errors of the mean.

Page 7

was performed for all measures using Statistica, Ver-
sion 5. One-way ANOVA tests were performed for
each variable to test replication effect. All data were
logtransformedpriortoanalysis.Analysestested each
variable for differences among species, among shade
treatments and in their interactions. The differences
among species and among shade treatments that were
significant were evaluated at the 5% level of signifi-
cance using Tukey’s Studentized range.
4. Results
Variance ratios of all measurements for two-way
ANOVAs showed significant differences among
shade treatments, among species, and in interactions
between shade treatments and species (Table 2).
Growth performance measures (height increment,root
collar diameter, total dry mass) all exhibited greater
levelsofsignificanceacrossshadetreatmentsirrespec-
tive of species. Alternatively, morphological measures
(leaf number, leaf size, branch number) all showed
greater levels of significance among species irrespec-
tive of shade treatment.
4.1. Growth performance measurements
Treatment differences: All species except S. firmum
exhibited greatest height increments across all the
brighter shade treatments (SD, LD, LS, FS) (Table 3a
and Table 4). S. firmum was the only species that
appeared to show some discrimination by growing
better in SD and LD shade treatments as compared to
FS and LS. Root collar diameters showed similar trends
as for height among the brighter shade treatments with
largest dimensions for all brighter environments (SD,
LD, LS, FS) irrespective of shade treatment. Except
for S. firmum species showed greater amounts of dis-
crimination in regard to the brighter shade treatments
for total dry mass. S. rubucundum had greatest total
dry mass in the LD treatment; S. operculataum and
S. makul had greatest total dry mass in theFS treatment.
Speciesdifferences:Differencesamongspecieswere
also evident with S. rubicundum and S. operculatum
exhibiting greater height increments across the treat-
ments than S. firmum and S. makul (Table 3a). For dry
mass trends were reversed with S. firmum exhibit-
ing greater dry mass gains than S. operculatum and
S. rubicundum. In the DS treatment S. firmum had
greatestdrymassgainand S.operculatumhadtheleast.
4.2. Morphological measurements
Treatment differences: Overall there were greater
numbers of branches and leaves, and larger leaf sizes
in the brighter shade treatments (SD, LD, LS, FS) as
compared to medium uniform shade (MS) (Table 3b
and Table 4). Measurements of leaf number and size
Table 4
Differences in the various growth performance and morphological
measures between shade treatments for each speciesa
DS MS SDLS LD FS
Height increment
S. firmum
S. makul
S. operculatum
S. rubicundum
d
c
c
c
c
b
b
b
a
a
a
a
b
a
a
a
ab
a
a
a
b
ab
ab
a
Root collar diameter
S. firmum
S. makul
S. operculatum
S. rubicundum
c
d
c
d
b
c
b
c
a
b
a
ab
a
ab
a
b
a
a
a
a
a
a
a
ab
Total dry mass
S. firmum
S. makul
S. operculatum
S. rubicundum
c
d
d
d
b
c
c
c
a
b
b
b
a
ab
ab
b
a
ab
b
a
a
a
a
ab
Leaf number
S. firmum
S. makul
S. operculatum
S. rubicundum
d
c
c
c
c
b
b
b
a
a
a
a
b
a
a
a
ab
a
a
a
b
ab
ab
a
Branch number
S. firmum
S. makul
S. operculatum
S. rubicundum
c
d
c
d
b
c
ab
c
a
b
b
b
a
a
a
b
ab
b
b
b
b
ab
ab
a
Leaf size
S. firmum
S. makul
S. operculatum
S. rubicundum
c
c
d
d
b
b
ab
c
a
a
a
a
ab
a
b
ab
a
a
ab
a
ab
a
c
b
aThe six different shade treatments are: DS: deep shade; MS:
medium shade; SD: small opening; LS: light shade; LD: large
opening; FS: full sun. Letters qualitatively indicate significant
differences among treatments ða > b > cÞ according to Tukey’s
Studentized range test ðP < 0:05%Þ.
B.M.P. Singhakumara et al./Forest Ecology and Management 174 (2003) 511–520
517

Page 8

were least in the DS treatment. For S. firmum, S.
operculatum and S. rubicundum greatest leaf size
was attained in the SD shade treatment. S. firmum
also attained the highest leaf numbers in the SD
treatment.
Species differences: S. rubicundum had the greatest
number of leaves and branches across all treatments
except the DS treatment (Table 3b). Differences in
branch and leaf number among the other species
were negligible. S. firmum exhibited the largest leaf
size followed by S. makul and S. operculatum, with
S. rubicundum having the smallest leaves.
4.3. Mass ratios
Trends among species remained consistent across
the shade treatments. S. operculatum had greatest
RMR, followed by S. makul and S. rubicundum, with
S. firmum having the least (Table 3c). Leaf mass ratio
(LMR) was reverse with S. firmum having greatest
proportion allocated to leaf tissue and S. operculatum
having the least. S. rubicundum had greatest SMR
as compared to the other species especially in the
brighter shade treatments.
5. Discussion
Results clearly define the breadth of shade treat-
ments under which each of the four Syzygium
spp. attained their greatest height growth, leaf area,
and dry mass gain. Dry mass gain as a measure of
performance would suggest that S. makul and
S. firmum were superior competitors as compared to
the other two species. Both S. makul and S. firmum are
similar, putting on greater structural bulk (as demon-
strated by their larger root collar diameters, larger leaf
size, low numbers of fine branches, and high dry mass
gains) as compared to S. operculatum and S. rubicun-
dum. Their similarity in growth and morphology may
beonereasonforwhyS.makulappearstoberestricted
to higher elevations in the forest than S. firmum—the
two are rarely found to co-exist within the same forest
landscape. Both may be considered slow-growing
in height increment, tolerant of understory shade,
and presumably with their thick, large leaves—less
desiccation prone in bright, hot environments than
S. operculatum and S. rubicundum.
Of the two we suggest S. firmum to have leaves that
are more adapted to the rain forest climates of lower
elevations. S. firmum’s larger leaves and greater LMR
in the brighter environments would make it better
adapted to the higher levels of heat and humidity of
wet tropical climates at lower elevations. This is
corroborated by leafanatomicalmeasuresofS. firmum
which indicate it also has thicker leaves and cuticle
(Gamage et al., unpublished data).
Growthinheightisoftenausefulindicatoroffitness
because it is usually correlated with increases in
biomass. In this study this is not the case: height
increment and dry mass gain change rank among
species. Height increment, however, is also a good
measure of seedling response to competition for light
(Fetcher et al., 1983). Clearly, S. rubicundum and
S. operculatum are superior in height growth as com-
pared to S. makul and S. firmum. This, together with
greater branching, smaller and more numerous leaves
(especiallyS.rubicundum)andmoreslenderstructure,
make S. operculatum and S. rubicundum fast-growing
and pioneer-like. This is corroborated by their known
distribution within the rain forest. Both are associated
with forest disturbance. With a larger leaf size than
S. rubicundum, we speculate S. operculatum to be the
most prone to desiccation—this fits well with its
restricted distribution along disturbed areas of rivers
and streams.
S. rubicundum can be found on mid-slope sites,
often with other fast-growing, but long-lived tree
species, such as Shorea trapezifolia (Ashton et al.,
1995, 2001). The dramatically smaller leaf size of
S. rubicundum could be related to a trade-off between
higher photosynthetic rates per unit leaf area and
increased evaporative demands arising in high light
(Nobel, 1977; Givinish, 1988), making this species
shade-intolerant adapted to drier or more stressed
environments. This kind of pioneer-like growth mor-
phology supports our contention that the sites where
S. rubicundum occurs with S. trapezifolia are actually
legacies of disturbance perhaps caused by forest
clearance for swidden agriculture or from multiple
blowdowns by wind (De Zoysa et al., 1991; Ashton
et al., 2001).
In this study there is some evidence to suggest that
some of the Syzygium spp. are able to maintain high
rates of height growth and dry mass gains in partial
shade (i.e. equal to or better than in the FS treatment)
518
B.M.P. Singhakumara et al./Forest Ecology and Management 174 (2003) 511–520

Page 9

by increasing leaf size (S. operculatum, S. rubicun-
dum) or leaf number (S. firmum). Interestingly, the
SD treatment which allowed only 18% PFD of FS,
still had equal or better growth than the LS treat-
ment with 50% PFD of FS. This may demonstrate
the poorer light quality effect (lower R:FR ratio) of the
LS treatment though the evidence for this is only
suggestive.
All this adds to a growing body of literature
that suggests many seedlings of rain forest canopy
trees actually do better under partial shade condi-
tions than in the FS. Most of this work has been done
on canopy tree species in mixed-dipterocarp forest
of the Asian tropics (Sasaki and Mori, 1981; Turner,
1989, 1990; Ashton and Berlyn, 1992; Ashton,
1995). These results are contrary to work done
particularly in the neotropics where seedlings of
canopy tree species do best in FS conditions (Popma
and Bongers, 1988, 1991; Kitajima, 1994; Pattison
et al., 1998).
In summary this study demonstrates the differential
degrees of shade-tolerance and growth of the four
Syzygium spp. examined. The study also demonstrates
that differences among species are best elucidated by
using several measures of growth performance with
Syzygium spp. changing rank when measures of
height growth and dry mass gain are compared with
each other. This has implications for forest manage-
ment and rain forest restoration planting. Syzygium
species are site specific with some that are shade-
tolerant and others that are clearly not. Care must be
taken to plant Syzygium spp. on appropriate site and
shade environments. To apply the findings in this
study to forest management applications further
experimental plantings under field conditions needs
to be done within the forest and on abandoned
agricultural lands as a next step in testing the findings
in this study.
Acknowledgements
We would like to thank Chaminda Kumarasingha
and B.W. Gunasoma for helping to set the experiment
up and for measuring the seedlings. We also acknowl-
edge logistic support and help from the Forest Depart-
ment of Sri Lanka, and financial support from the
MacArthur Foundation, USA.
References
Ashton, P.S., 1981. Myrtaceae. In: Dassanayake, M.D., Fosberg,
F.R. (Eds.), Revised Handbook to the Flora of Ceylon, Vol. II.
Amerind Publishers, New Delhi, India, pp. 403–472.
Ashton, P.M.S., 1992. Some measurements of the microclimate
within a Sri Lankan tropical rainforest. Agric. For. Meteorol.
59, 217–235.
Ashton, P.M.S., Berlyn, G.P., 1992. Leaf adaptations of some Shorea
species to sun and shade. New Phytologist 121, 587–596.
Ashton, P.M.S., 1995. Seedling growth of co-occurring Shorea
species in the simulated light environments of a rain forest. For.
Ecol. Manage. 72, 1–12.
Ashton, P.M.S., Gunatilleke, C.V.S., Gunatilleke, I.A.U.N., 1995.
Seedling survival and growth of four Shorea species in a Sri
Lankan rain forest. J. Trop. Ecol. 11, 263–279.
Ashton, M.S., Gunatilleke, C.V.S., Singhakumara, B.M.P., Guna-
tillke, I.A.U.N., 2001. Restoration pathways for rain forest in
southwest Sri Lanka: a review of concepts and models. For.
Ecol. Manage. 154, 409–430.
DeLucia, E.H., Sipe, T.W., Herrick, J., Maherali, H., 1998. Sapling
biomass allocation and growth in the understorey of a
deciduous hardwood forest. Am. J. Bot. 85, 955–963.
De Zoysa, N.D., Gunatilleke, C.V.S., Gunatilleke, I.A.U.N., 1991.
Comparative phytosociology of natural and modified rain forest
sites in Sinharaja, MAB Reserve, Sri Lanka. In: Gomez-Pmpa,
A., Whitmore, T.C., Hadley, M. (Eds.), Rain Forest Regenera-
tion and Management. UNESCO/Parthenon Press, Paris,
France, pp. 154–168.
Fetcher, N., Strain, B.R., Oberbauer, S.F., 1983. Effects of light
regime on the growth leaf morphology and water relations of
seedlings of two species of tropical trees. Oecologia 58, 314–319.
Givinish, T.J., 1988. Adaptation to sun and shade: a whole plant
perspective. Aust. J. Plant Physiol. 15, 63–92.
Gunatilleke, C.V.S., Ashton, P.S., 1987. New light on the plant
geography of Ceylon. II. The ecology biogeography of the
lowland endemic tree flora. J. Biogeogr. 14, 295–327.
Gunatilleke, C.V.S., Perera, G.A.D., Ashton, P.M.S., Ashton, P.S.,
Gunatilleke, I.A.U.N., 1996. Seedling growth of Shorea section
Doona (Dipterocarpaceae) in soils from topographically
different sites of Sinharaja rain forest in Sri Lanka. In: Swaine,
M.D. (Ed.), The Ecology of Tropical Forest Seedlings.
UNESCO/Parthenon Press, Paris, France, pp. 245–266.
King, D.A., 1994. Influence of light level on the growth and
morphology of saplings in a Panamanian forest. Am. J. Bot. 81,
948–957.
Kitajima, K., 1994. Relative importance of photosynthetic traits
and allocation patterns as correlates of seedling shade tolerance
of 13 tropical trees. Oecologia 98, 419–428.
Lee, D.W., 1985. Duplicating foliage shade for research and plant
development. HortScience 20, 116–118.
Lee, D.W., Baskaran, K., Mansor, M., Mohamad, H., Yap, S.K.,
1996. Irradiance and spectral quality affect Asian tropical rain
forest tree seedling development. Ecology 77, 568–580.
McConnaughay, K.D.M., Coleman, J.S., 1999. Biomass allocation
in plants: ontogeny or optimality? A test along three resource
gradients. Ecology 80, 2581–2593.
B.M.P. Singhakumara et al./Forest Ecology and Management 174 (2003) 511–520
519

Page 10

Nobel, P.S., 1977. Internal leaf area and cellular CO2resistance:
photosynthetic implications of variations with growth condi-
tions and plant species. Physiol. Plant. 40, 137–144.
Osunkoya, O.O., Ash, J.E., Hopkins, M.S., Graham, A.W., 1994.
Influence of seed size and seedling ecological attributes on
shade-tolerance of rainforest tree species in northern Queens-
land. J. Ecol. 82, 149–163.
Pattison, R.R., Goldstein, G., Ares, A., 1998. Growth, biomass
allocation and photosynthesis of invasive and native Hawaiian
rainforest species. Oecologia 117, 449–459.
Popma, J., Bongers, F., 1988. The effect of canopy gaps on growth
and morphology of seedlings of rain forest species. Oecologia
75, 625–632.
Popma, J., Bongers, F., 1991. Acclimation of seedlings of three
Mexican tropical rainforest tree species to a change in light
availability. J. Trop. Ecol. 7, 85–97.
Sasaki, S., Mori, T., 1981. Growth responses of dipterocarp
seedlings to light. Malaysian For. 44, 319–345.
Turner, I.M., 1989. A shading experiment on some tropical rain
forest tree seedlings. J. Trop. For. Sci. 1, 383–389.
Turner, I.M., 1990. Tree seedling growth and survival in a
Malaysian rain forest. Biotropica 22, 146–154.
Veneklaas, E.J., Poorter, L., 1998. Growth and carbon partitioning
of tropical tree seedlings in contrasting light environments. In:
Lambers, H., Poorter, H., Van Buren, M.M.I. (Eds.), Inherent
Variation in Plant Growth: Physiological Mechanisms and
Ecological Consequences. Blackhuys Publishers, Leiden, The
Netherlands, pp. 337–361.
Whitmore, T.C., 1989. Canopy gaps and the two major groups of
forest trees. Ecology 70, 536–538.
520
B.M.P. Singhakumara et al./Forest Ecology and Management 174 (2003) 511–520