141, 365–377. With 3 figures 2003 The Linnean Society of London, Botanical Journal of the Linnean Society, 2003

Yüklə 239.84 Kb.
PDF просмотр
ölçüsü239.84 Kb.
  1   2   3


Botanical Journal of the Linnean Society


, 2003, 




, 365–377. With 3 figures

© 2003 The Linnean Society of London, 


Botanical Journal of the Linnean Society, 






, 365–377




Blackwell Science, LtdOxford, UKBOJBotanical Journal of the Linnean Society0024-4074The Linnean Society of London, 2003


Original Article




*Corresponding author: E-mail: mark.ashton@yale.edu


Leaf structure of 




 spp. (Myrtaceae) in relation to 

site affinity within a tropical rain forest
















Department of Forestry and Environmental Science, University of Sri Jayawardenapura,

Nugegoda, Sri Lanka




School of Forestry and Environmental Studies, Yale University, New Haven, CT 06511, USA


Received July 2002; accepted for publication October 2002


This study examined four species of 






S. firmum



S. makul



S. operculatum



S. rubicundum


) Myrtaceae, a

tree genus that dominates the canopy of rain forests of south-west Sri Lanka. 




 spp. occupy differing hab-

itats with relation to succession and forest topography. We examined differences in leaf morphology and physiology

in response to amount of shade, an important environmental variable affecting 




 distribution within the for-

est. To study change in leaf structure and physiology, environmental shelters were constructed simulating forest

shade that differed in quality, quantity and duration. Seedlings were exposed to: (i) 0% shade (full sun, FS), red : far

red (R : FR) ratio 1.27; (ii) 65% shade (large opening, LO) with direct sunlight similar to the centre of a large canopy

opening, R : FR ratio 1.27; (iii) 82% shade (small opening, SO) with direct sunlight similar to the centre of a small

canopy opening, R : FR ratio 1.27; (iv) 58% uniform light shade (LS) with a quality similar to the outside edge of a

large canopy opening, R : FR ratio 1.05; (v) 85% uniform medium shade (MS) with a quality similar to the inside for-

est edge of a large canopy opening, R : FR ratio 0.97; (vi) 99% uniform deep shade (DS) similar to that of the forest

understorey, R : FR ratio 0.23. The shelters were constructed in a large open area at the field station of the Sinharaja

World Heritage site, Sri Lanka. Seedlings of each species were grown for two years in their respective shade treat-

ments before physiological, morphological and anatomical measurements were made on leaves. Variation in leaf

structure and physiology between the species was associated with differences in shade-tolerance and water-use. All

species increased in photosynthesis rates and dimensions in leaf structure (leaf blade and cuticle thickness, stomatal

density, thickness of upper and lower epidermis, and thickness of palisade mesophyll) with decrease in shade. In con-

trast, stomatal conductivity was highest in the DS (99% shade) treatment. Leaves of 


Syzygium firmum


 were thickest

and largest in area. 


S. firmum


 also had highest photosynthesis in the SO (82% shade) treatment. 


S. firmum


 was the

most shade-tolerant of all species: it grows well in low shade and its leaf structure suggests it to be the most con-

servative in water-use of the 




 spp. In the forest 


S. firmum


 can persist in the forest shade as established

seedlings, but grows best within canopy openings of late-seral rain forest. Leaves of 


S. operculatum


 were thinnest but

had highest stomatal densities of the four species. 


S. operculatum


 is considered shade-intolerant, with a leaf struc-

ture suggesting it to be prone to desiccation, and by implication susceptible to drought. 


S. operculatum


 is found along

streams within early seral rain forest habitat, often originating on stream banks after land clearance for cultivation.

In the FS (0% shade) treatment, 


S. rubicundun


 had highest photosynthesis rates and greatest number of leaves but

smallest leaf area of the 






S. rubicundum


 is more shade-intolerant but more efficient in water-use



S. operculatum



S. rubicundum


 is a mid-seral canopy tree of the midslope stands that are thought to have orig-

inated after catastrophic windthrows or swidden cultivation. The leaf physiology and structure of 


S. makul



it to be both moderately shade-tolerant and conservative in water-use. It is the most widely distributed 




species across the topography of late-seral rain forest. We suggest forest disturbance and hydrology are important

environmental factors that influence distribution of 




 species across the topography. Results from this study

contribute to a body of knowledge suggesting that canopy tree species of rain forests in south-west Sri Lanka have

discrete affinities to topography and differences in successional status, and that adaptations in leaf structure and

physiology are indicative of such phenomena.

© 2003 The Linnean Society of London, 


Botanical Journal of the Lin-

nean Society


, 2003, 




, 365–377.





leaf morphology – mixed-dipterocarp forest – photosynthesis – physiology –


plasticity – shade – Sinharaja – Sri Lanka










© 2003 The Linnean Society of London, 


Botanical Journal of the Linnean Society, 






, 365–377




Light, a primary determinant of seedling growth, is

one of the major limiting factors in tropical rain forest

understories (Lee 


et al


., 1996; Zipperlen & Press,

1996). In the understorey of tropical rain forests, less

than 2% of photosynthetic photon flux (PPF) is trans-

mitted through the canopy and the R : FR ratio can be

decreased by a factor of six (Bazzaz & Picket, 1980;

Whitmore, 1990; Ashton, 1992; Tinoco-Ojanguren &

Pearcy, 1995). Canopy openings or tree fall gaps allow

more sunlight to penetrate to the forest understorey

and are an important environmental variable for

establishment and release of tree seedlings (Pearcy,

1987; Popma & Bongers, 1988; Whitmore, 1988;

Runckle, 1989; Raich, 1990; Ashton, Gunatilleke &

Gunatilleke, 1995; Dai, 1996; Ashton 


et al


., 1998). In

different sizes of canopy gaps and in different loca-

tions within the same gap light quality and quantity

differ and affect seedling establishment and growth



et al


., 1995; Montgomery & Chazdon, 2002).

In order to maximize growth performances in different

light environments tree species adapt phenotypically



et al


., 1996). Leaf morphology, anatomy

and physiology vary between sun and shade grown

seedlings (Jackson, 1967; Chabot & Chabot, 1977;

Carpenter & Smith, 1975; Fetcher, Strain & Ober-

bauer, 1983; Lee 


et al


., 1990; Ashton & Berlyn, 1992,

1994; Strauss-Debenedetti & Berlyn, 1994; Ashton


et al


., 1998). Sun leaves are smaller in size and have

thicker palisade mesophyll layers and greater volume

to surface area ratios that promote high photosyn-

thetic capacity and higher maximum leaf conductance

(Carpenter & Smith, 1975; Jurik, Chabot & Chabot,

1982; Ashton & Berlyn, 1992, 1994; Ashton 


et al



1998). Changes in seedling leaf structure in response

to shade between and within related tree species may

help explain why species are well suited to occupy par-

ticular sites within the same forest (Ashton & Berlyn,

1994; Richardson 


et al


., 2000).

The rich endemic tropical rain forest of south-west

Sri Lanka is dominated by tree species belonging to

the families Dipterocarpaceae, Clusiaceae, Myrtaceae

and Sapotaceae (Gunatilleke 


et al


., 1996). Change in

leaf structure and physiology in relation to different

qualities and quantities of shade has been investi-

gated for canopy tree species in the Dipterocarpaceae

(Ashton & Berlyn, 1992). Ashton & Berlyn (1992)

showed that dipterocarp leaf structure is indicative of

species difference in shade tolerance and other inter-

acting factors such as proneness to desiccation. Fol-

lowing this line of inquiry we selected four 




spp. (Myrtaceae) for this study. All are common can-

opy and subcanopy rain forest trees. Our long-term

goal is to build a comprehensive autecological data-

base that can determine the degree to which species

have discrete affinities to topography and differences

in successional status, and to demonstrate that

adaptations in leaf structure and physiology are

indicative of  such  phenomena.  This  information  can

be used for  the  purpose  of  better  management

and restoration of Sri Lankan rain forests.

We hypothesize that shade tolerance (as a proxy for

disturbance regime) is a primary factor determining

topographic affinities among 




  spp. To test

this we selected 




 spp. known to occur on dif-

ferent sites within the forest. Seedlings of 




spp. were grown in shelters that created a range of

shade environments. Experiments were designed to

determine: (i) photosynthesis and stomatal conductiv-

ity, (ii) leaf number and leaf area, and (iii) leaf anat-

omy. Specific hypotheses tested in this study were






 species differ in leaf anatomy and phys-

iology with more shade-intolerant species with pio-

neer traits (




  Swaine & Whitmore, 1988) having

thinner leaf blades, thinner cuticles and palisade layer

thickness; higher numbers of stomata, and higher net

photosynthesis rates under shade conditions than spe-

cies that are more shade-tolerant and late-seral.





 species differ in leaf morphology with the

more late-seral species having larger leaf areas per

seedling dry mass, and lower numbers of leaves, and

the more pioneer-like species having lower leaf areas

but higher numbers of leaves under brighter light con-


(3) Adaptive traits in leaf anatomy, physiology and

morphology match the known site affinities and suc-

cessional status of the 























We conducted this study at the field station of the Sin-

haraja World Heritage Site, a 20 000 ha forest located

in the south-west of Sri Lanka. The forest is an ever-

wet mixed-dipterocarp forest type comprising canopy

trees of the genera 









paceae) and 




  (Clusiaceae) (De Rosayro, 1942;

Gunatilleke & Ashton, 1987).

The region receives 4000–6000 mm of rainfall per

year. Most rain falls during the south-west (May –

July) and north-east monsoons (October –

January). Seasonal temperatures range between 25

and 27



C with a greater diurnal variation of 








Soils are classified as ultisols following the USDA

(1975) terminology, or red yellow latosols using the

classification system of the Food and Agriculture

Organization (Moorman & Panabokke, 1961).










© 2003 The Linnean Society of London, 


Botanical Journal of the Linnean Society, 






, 365–377


The four native 




  spp. (Myrtaceae) com-



S. firmum


  Thw., and 


S. makul




S. operculatum


  (Roxb.) Niedz., 


S. rubicundum



and Arn (Ashton, 1981). Two species, 


S. firmum




S. makul


,  are endemic to south-west Sri Lanka. All


Figure 1.


Demographic stem maps of a 25 ha plot in Sin-

haraja Forest for juveniles (all individuals 



1 cm  and

<10 cm d.b.h.; small black dots), intermediate-size trees

(all individuals 



10 cm d.b.h and <30 cm d.b.h.; black

squares), and mature trees (all individuals 



30 cm  d.b.h.;

large black circles) of 




  species. (A) 


S. makul


;  (B)


S. operculatum


;  (C) 


S. rubicundum



Syzygium firmum



found at lower elevations and is therefore not present in

the plot. The plot is alligned N–S and ranges in elevation

between 400–560 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 north-east toward

the south-west. Contour intervals are given at 15 m inter-

vals with horizontal and vertical axes indicating metre

dimensions of the plot (C.V.S. Gunatilleke and I.A.U.N.

Gunatilleke, unpublished data).

four species are either canopy or subcanopy dominants

of the mixed-dipterocarp forests.

Each of the four species has been documented to

occupy different parts of the forest topography (Fig. 1).


Syzygium operculatum


  occurs within early succes-

sional or disturbed forest along small rivers and

perennial streams (Ashton, 1981). 


Syzygium firmum




S. makul


  occur on the deep soils of valleys and

midslopes as canopy and subcanopy trees of late-suc-

cessional forest (Ashton, 1981). However, 


S. firmum


mainly occurs in the adjacent Kanneliya Forest

Reserve, a lowland rain forest (50–300 m above sea

level, a.s.l.), while 


S. makul


 occurs mainly in the Sin-

haraja hill rain forest (300–700 m a.s.l.). 





  occurs on midslope sites but separate



S. firmum




S. makul


, and in close association

with other tree species that have autecological char-

acteristics suggesting the forest association is more

shade-intolerant and mid-seral, perhaps originating

after catastrophic windthrow or old swidden cul-

tivation (Ashton, 1981; De Zoysa, Gunatilleke &

Gunatilleke, 1991).






















Twenty-four well-ventilated shelters (6 








2.2 m,







 h) were constructed in the full open at the

Sinharaja field station in January 1996. They were

designed to create shade treatments that represented

a range of photosynthetic photon flux densities (PFD)

and red : far red ratios (R : FR) found beneath the

Sinharaja forest canopy (Ashton, 1992).

Six treatment combinations of irradiance and spec-

tral quality were created. Each treatment was blocked

into four separate shelters (6 







24). Seedlings were

exposed to:










© 2003 The Linnean Society of London, 


Botanical Journal of the Linnean Society, 






, 365–377


(i) 0% shade (full sun, FS) (maximum measured PFD

of 1600 




mols m







;  maximum daily PFD of

38.1 mols m


; R : FR ratio 1.27); (ii) 65% shade (large

opening, LO) with direct light similar to the centre of

a large (300–400 m


) canopy opening (maximum mea-

sured PFD of 1600 mmols m




; maximum daily PFD

of 13.3 mols m


;  periods of direct light for 6 h of the

day; R : FR ratio 1.27);

(iii) 82% shade (small opening, SO) with direct light

similar to the centre of a small (150–200 m


) canopy

opening (maximum measured PFD of 1600 mmols





; maximum daily PFD of 7.4 mols m


; periods of

direct light for 2 h of the day; R : FR ratio 1.27);

(iv) 58% shade (uniform light shade, LS) with a quality

similar to the outside edge of a large canopy opening

(maximum daily PFD of 800 mmols m




; maximum

daily PFD of 16.3 mols m


; R : FR ratio 1.05);

(v) 85% shade (uniform medium shade, MS) with a

quality similar to the inside edge of a large canopy

opening (maximum measured PFD of 350 






;  maximum daily PFD of 6.0 mols m


; R : FR

ratio 0.97);

(vi) 99% shade (uniform deep shade, DS) with a qual-

ity similar to the forest understorey (maximum mea-

sured PFD of 50 mmols m




; maximum daily PFD of

1.2 mols m


; R : FR ratio 0.23).

Shade treatments that altered the duration of direct

PFD (LO, SO), were created by construction a series of

parallel vertical slats aligned north–south, placed hor-

izontally 2 m above the ground and across the com-

plete interior of a shelter. As the sun rose from the east

and set in the west the duration and number of direct

light periods was controlled by slat orientation (N–S),

slat height, and the proximity of slats to one another.

For shelters that altered the quality and intensity of

PFD (DS, MS, LS) a mix of pigments was sprayed onto

UV durable plastic film in a concentration of 10% with

clear varnish following the protocol of Lee (1985) and

Ashton (1995). The amount sprayed regulated the

R : FR ratio and the intensity of PFD.

Seed of each species was collected from several par-

ent trees in Sinharaja except for S. firmum, which was

collected from the nearby Kanneliya Forest Reserve.

For each species the seeds of the parent trees were

mixed together and germinated on a nursery bed

beneath 50% shade cloth. One-and-a-half month old

seedlings were then taken bare-rooted from the nurs-

ery in January 1996, and individually planted in black

plastic bags that were perforated at the bottom to

allow drainage. Bags were 20 cm in diameter and

30 cm depth with a 20 000 cm


 volume. Forest topsoil

was taken from a single valley location and mixed

with an equal part of river sand to improve drainage.

Nutrient contents of this potting mixture have been

reported elsewhere by Gunatilleke et al.  (1996). Six

seedlings of each species were used in each of four

blocks for a shade treatment. The total number of

seedlings in this experiment was 576 (4 species ¥6

seedlings/block  ¥4 blocks/treatment ¥6 shade treat-

ments). Seedlings were grown in soil that was main-

tained at field capacity over the two-year study period.

Potting bags were large enough to allow free root

development and growth over the duration of the

experiment. No root balling was observed at the end of

the experiment in any of the treatments.





Physiology was measured in April and May 1997 after

one-and-a-quarter years of growth within the shade

treatments. Seedlings of each species were taken from

the full sun (FS, 0% shade), uniform light shade (LS,

58%), medium shade (MS, 85%) and deep shade (DS,

99%) shelters for measurement of photosynthesis and

stomatal conductance. Measurements were made on

at least 12 seedlings for each species and for each

shade treatment (three per block for a treatment).

Seedlings were sampled in the morning to insure max-

imum physiological rates of response. 

Soil was

watered to field capacity the day before sampling.

Net photosynthesis and stomatal conductivity were

measured using a portable closed-system infra-red gas

analyser (Li-Cor 6200, Lincoln, NE) with a one-litre

leaf chamber that had been adjusted to measure a con-

stant leaf area (Welles, 1986). The instrument was

assembled in a well-ventilated place, which attained

an ambient atmospheric CO


  level of 340 


Attached leaves of seedlings were sampled using the

appropriate ambient light quality and quantity of

their respective shade treatments in the chamber.

The chamber was kept open until atmospheric CO


was reached. It was then sealed and measurements

recorded after a steady CO


 decline was noticed on the

monitor screen. Chamber temperature was 27 ∞C and

relative humidity was kept between 60 and 70% dur-

ing measurements. Each measurement of a leaf con-

sisted of a set of three sequential readings.











Leaf number was recorded for the 24 seedlings per

species–shade treatment combination after each six-

month period for the two-year duration of the experi-

ment. To determine leaf area, three mature leaves

were randomly selected from each of 12 seedlings per

species and treatment (three per block). Measure-

ments of leaf area were made using a CI-202 leaf area

meter (CID Inc., Vancouver, Washington). Total plant

leaf area was calculated using leaf number and the

mean leaf area of single leaves for a given species–

shade treatment combination.



© 2003 The Linnean Society of London, Botanical Journal of the Linnean Society, 2003, 141, 365–377

Individual leaves from each of eight individuals per

species and treatment (two per block) were selected for

anatomy measurements. All leaves were fully

expanded, undamaged with no signs of herbivory. Leaf

strips (0.5 ¥ 1.0 cm) were taken from the middle por-

tion of the lamina across the mid-rib. The leaf strips

were immediately fixed in FAA (formalin : acetic acid :

ethanol), and then dehydrated in an ethanol series,

immersed in xylene and embedded in wax. Cross sec-

tions of 10 mm thickness were cut using a rotary

microtome, mounted on slides, stained with safranin

and fast green following a modified procedure of Ber-

lyn & Miksche (1976) and mounted in Canada balsam.

One slide from each leaf strip was prepared and mea-

sured for thickness of leaf, cuticle, palisade mesophyll,

upper epidermis and lower epidermis at ten separate

points, avoiding the region around the midrib. These

structural attributes were measured because of their

known dimensional variability in relation to changes

in environment and their importance in the leaf-level

processes of photosynthesis and light capture, and for

the regulation of water (Ashton & Berlyn, 1992). A

light microscope with a 5 ¥ filar micrometer eyepiece

was used to measure cell dimensions at 400 ¥ magni-

fication, leaf thickness at 100 ¥  magnification and

cuticle thickness at 1000 ¥ magnification.

Stomatal frequencies were observed from cellulose

acetate positives of silicone-rubber impressions

(Heichel, 1971). Each impression was taken midway

between the base and tip on both adaxial and abaxial

surfaces of each leaf. Five impressions were taken

from each leaf surface and from each peel two separate

points were viewed.





Analyses of variance (general linear model) were per-

formed on each physiological, anatomical and morpho-

logical measure using Statistica Version 5. Analyses

tested for differences and interactions between shade

treatment and species. Data were log-transformed

prior to analysis. All F statistics that were significant

were evaluated for differences between species using

Tukey’s Studentized range at the 5% significance level.

For a measure of the variation in anatomical, mor-

phological and physiological attributes we used an

index of phenotypic plasticity [P =(X/x)] that included

comparisons between mean values for the darkest (x)

and brightest (X) shade treatments.






Shade treatments and species differed significantly

(P < 0.0001). Overall, net photosynthesis (P


) and sto-

matal conductivity (g) were less significantly different

between species than morphological measures of leaf

structure. However, physiological measures were dif-

ferent between shade treatments and between species

(for P


 only), and in interactions between treatments

and species (Table 1). Shade treatments showed much

greater differences as compared to differences among

species. Block effects within shade treatments showed

no difference.

All species increased P


  with increase in light


2a). Comparisons between species revealed

S. rubicundum  to have highest P


  among species in

the full sun treatment (FS, 0%), but lowest P



species for the deep shade treatment (DS, 99%)

(Fig. 2a).  Syzygium firmum  had highest P



species in the medium shade (MS, 85%) treatment.

Syzygium operculatum had highest P


 among species

for the DS (99%) treatment. Syzygium rubicundum

exhibited greatest differences in P


 amongst the shade

treatments followed in order by S. firmum,  S. makul

and S. operculatum (Table 2).

Unlike P


,  conductivity (g) was highest in the DS

(99%) treatment (Fig. 2b). The other treatments had

low g with no apparent trends or differences among

treatments. In the DS (99%) treatment S. operculatum

had highest g among species, S. firmum had lowest g

(Fig. 2b). No other differences were shown between

species for shade treatments. Syzygium operculatum

differed most in g among shade treatments, followed

in order by S. makul,  S. rubicundum  and  S. firmum

(Table 2).





Two-way analyses of variance for leaf number, individ-

ual leaf area, and total leaf area showed significant

differences between species, between shade treat-

ments, and in interactions between both. In general,

species differences were greater than the effect of

shade treatment for leaf number and leaf size mea-

sures, but the product of both (total leaf areas) showed

the reverse – greater differences among shade treat-

ments than among species. Block effects within shade

treatments did not differ (Table 1).

Species trends across treatments

In general, leaf numbers per seedling were greatest in

the FS (0%), LO (65%), LS (58%) and SO (82%) shade

treatments (Fig. 3). The lowest leaf numbers per seed-

ling were recorded in the DS (99%) shade treatment.

Syzygium rubicundum,  S. operculatum  and  S. makul

had the highest leaf numbers in all the brighter shade

treatments (SO, LS, LO, FS), while S. firmum had the

highest numbers in SO (82%) treatment (Fig. 


Table 3).



© 2003 The Linnean Society of London, Botanical Journal of the Linnean Society, 2003, 141, 365–377

  1   2   3

Verilənlər bazası müəlliflik hüququ ilə müdafiə olunur ©azkurs.org 2016
rəhbərliyinə müraciət

    Ana səhifə