Plant diversity and soil characteristics of the Ussangoda serpentine site

Abstract

Serpentine soils are derived from the weathering of serpentine and ultramafic rocks, which have a high content of ferromagnesian minerals. The high content of heavy metals in serpentine soils alter their physical and chemical properties making them unsuitable for plant growth. There are six serpentine sites in Sri Lanka and the Ussangoda site is on the southern coast in Hambantota. The moisture content, organic matter and cation exchange capacity (CEC) are low in serpentine soils. The available calcium (Ca) content is low and the magnesium (Mg) content is relatively high. The Ca to Mg ratio is 0.60, which is typical for serpentine soils. Two distinct forms of vegetation grow on the Ussangoda serpentine soil. The large plain is covered by stunted, prostrate species with an extensive root system. Patches of shrubs and trees occur on the plains as small islands. The serpentine flora is sharply demarcated from the surrounding non-serpentine flora by their growth habit. The number of plant families and species is lower in the serpentine soil than in the adjacent non-serpentine areas. Four families comprising six species grew only on the serpentine soil. Five species growing in the serpentine soil contained 560 - 830 ppm of nickel (Ni) in their tissues. Hybanthus enneaspermus had 1800 ppm of nickel. Two species, Vernonia zeylanica and Scolopia acuminata, are endemic to Sri Lanka.

Full text

RESEARCH ARTICLE
Plant diversity and soil characteristics of the Ussangoda
serpentine site
J.Natn.Sci.Foundation Sri Lanka 2011 39 (4): 355-363
H.A.S. Weerasinghe and M.C.M. Iqbal*
Plant Biology, Institute of Fundamental Studies, Hantana Road, Kandy.
Revised: 03 May 2011 ; Accepted: 20 May 2011
* Corresponding author (mcmif2003@yahoo.com)
Abstract: Serpentine soils are derived from the weathering of
serpentine and ultramafic rocks, which have a high content of
ferromagnesian minerals. The high content of heavy metals in
serpentine soils alter their physical and chemical properties
making them unsuitable for plant growth. There are six
serpentine sites in Sri Lanka and the Ussangoda site is on the
southern coast in Hambantota. The moisture content, organic
matter and cation exchange capacity (CEC) are low in serpentine
soils. The available calcium (Ca) content is low and the
magnesium (Mg) content is relatively high. The Ca to Mg ratio
is 0.60, which is typical for serpentine soils. Two distinct forms
of vegetation grow on the Ussangoda serpentine soil. The large
plain is covered by stunted, prostrate species with an extensive
root system. Patches of shrubs and trees occur on the plains as
small islands. The serpentine flora is sharply demarcated from
the surrounding non-serpentine flora by their growth habit. The
number of plant families and species is lower in the serpentine
soil than in the adjacent non-serpentine areas. Four families
comprising six species grew only on the serpentine soil. Five
species growing in the serpentine soil contained 560 – 830
ppm of nickel (Ni) in their tissues. Hybanthus enneaspermus
had 1800 ppm of nickel. Two species, Vernonia zeylanica and
Scolopia acuminata, are endemic to Sri Lanka.
Keywords: Biodiversity, Evolvulus alsinoides, heavy metals,
Hybanthus enneaspermus, serpentine, Ussangoda.
INTRODUCTION
Serpentine soils occur in isolated patches all over the
world, particularly along continental margins. They are
derived from serpentine rocks composed of serpentine
minerals and ultramafic rocks, which have a high
content of ferromagnesian minerals (Brooks, 1987). The
weathering of these rock types enriches the soil with
Mg, Fe and other heavy metals such as Ni and Co. These
metals change the physical and chemical properties of
the soil to form an unfavourable and hostile environment
for plant growth. Over time, plant species have adapted
to these unusual soil conditions producing unique
associations between the soil and plants. They can be
regarded as living laboratories for studying evolutionary
biology and adaptations by the plant species to the
extreme environments peculiar to serpentine soils.
Very little is known of the geology and flora of
serpentine sites in Southeast Asia. In India, an ultramafic
site was reported in the State of Orissa (Brooks, 1987).
The serpentine areas in the Malay Archipelago are
well documented and many endemic species have
been identified (Brooks, 1987; Proctor 2003). The
soil characteristics of the ultramafic sites in Malaysia,
Indonesia and Philippines have been compared by
Brearley (2005). Geological studies in Sri Lanka have
shown six serpentinite sites at Ussangoda, Indikolapelessa,
Ginigalpelessa, Katupota, Yodhagannawa and Rupaha.
These sites lie close to a Precambrian suture zone
between the lithotectonic units, the Vijayan and Highland
series (Dissanayake & Van Riel, 1978; Munasinghe &
Dissanayake, 1979; Munasinghe & Dissanayake, 1980).
Of these sites the geochemistry of the Uda-Walawe
serpentine site has been described (Dissanayake & Van
Riel, 1978; Dissanayake, 1982).
Ultramafic soils are generally shallow with a high
Mg to Ca ratio, deficient in essential plant nutrients and
contain potentially toxic concentrations of heavy metals
(Proctor & Woodell, 1975; Brooks 1987). The rocky
nature of the soil and low clay and organic matter content
provide a highly permeable soil with low water holding
capacity. The exchangeable Mg concentrations are high
and exchangeable Ca concentrations are low in the soil,
which is deficient in essential plant nutrients N, P and
K (Kruckeberg,1984). The vegetation on serpentine soils

December 2011 Journal of the National Science Foundation of Sri Lanka 39 (4)
356
H.A.S. Weerasinghe & M.C.M.Iqbal
have adapted morphologically and physiologically to this
environment. Morphologically plants are dwarfed with
narrow, glaucescent thick leaves, strong sclerenchyma
development and enlarged root system (Kruckeberg,
1984). Physiologically they are able to tolerate the heavy
metal presence in the soil by absorption and sequestration
or exclusion. Thus serpentine soils are a unique ecological
niche, which is poor in diversity of plant species.
The flora of Ussangoda was first described by
Brooks (1987), who on a visit to the island referred to
the Ussangoda serpentine site: “At Welipatanwila, the
heavily laterised ultramafics are almost completely
devoid of vegetation except for a scattering of a small
blue-flowered herb, Evolvulus alsinoides. The boundary
of the serpentine is marked by low thorn bushes
interspersed with Opuntia species”. In a subsequent
study (Seneviratne et al., 2000), 14 plant species were
identified all of which accumulated over 100 ppm of Ni.
Rajakaruna and Bohm (2002) studied nine plant species
growing on the Ussangoda serpentine soil and the soil
chemistry from their root zones. Their study found three
species, which hyperaccumulated over 1000 ppm of Ni
from the soil. In a recent study (Iqbal et al., 2006) 29
species of flowering plants, which included trees, shrubs,
vines and prostate plants growing within the Ussangoda
serpentine site were identified. Cassia kleinii was
identified as a Ni hyper-accumulator. In the last 20 years,
few studies were devoted to study this unique habitat and
much work remains to be done.
For a comprehensive study of the Ussangoda
serpentine site, it is necessary to determine the diversity
of plant species, their physiological adaptations to the
adverse soil environment and the soil characteristics
contributing to their adaptations. This would subsequently
assist in the study of the other serpentine sites in Sri Lanka
and potentially contribute towards identifying heavy
metal hyper-accumulating species. The objective of this
study is to determine the important physical and chemical
characteristics of the Ussangoda soil, its consequence
on the flora and diversity of the flora in relation to the
surrounding non-serpentine areas.
METHODS AND MATERIALS
Site description: The serpentine study site at Ussangoda
(Figure 1) is a flat plain (approximately 1 km2) located
close to the Nonagama junction on the Matara –
Hambantota highway (latitude 06º 05’ N, longitude 80º
59’ E). The southern boundary of the plain is a cliff,
approximately 30 m amsl, overlooking the Indian Ocean.
The highest point on the plain is 34.5 m amsl.
Climate: The mean annual temperature is 27.9 ºC with
a mean range of 31.1 – 24.6 ºC. Daytime temperature
during field visits reached 35.8 ºC. The annual rainfall is
less than 1250 mm (mainly from October – January) with
more than 5 dry months with less than 50 mm rainfall
(IUCN, 2004).
Soil analysis: Soil samples were collected according to
the stratified random sampling technique to represent the
10 , 20 and 30 m contours. Each sample was collected
from below the soil surface to a depth of 10 cm and
stored in labelled polythene bags. All the samples were
analysed separately. The samples were air-dried for
1 wk at room temperature and passed through a 2 mm
sieve. Soil moisture was determined by drying 5 g of soil
to a constant mass in an oven at 105 °C for 24 h and
cooling it in a desiccator for 30 min. To determine the
pH, 20 mL of de-ionized water or calcium chloride was
added to 10 g of soil and stirred 4-5 times within 30 min.
The suspension was allowed to settle for 30 min and
the pH of the supernatant determined. Soil conductivity
was measured in a suspension from mixing 40 g of soil
and 80 mL de-ionized water. The above methods are
described in Kalra & Maynard (1991).
Cation exchange capacity (CEC) of the soil was
determined according to Anderson & Ingram (1993). Air-
dried soil was shaken with 1M potassium chloride for 5
min and repeatedly centrifuged until the supernatant was
clear. Ethanol was added to the filtrate and centrifuged
until the supernatant to be discarded was clear. To this
1M ammonium acetate was added, and the mixture
shaken and centrifuged until the supernatant was clear.
The supernatant volume was made up to 100 mL with
ammonium acetate. The potassium concentration was
measured by the atomic absorption spectrophotometer
(AAS). Soil organic matter was determined by percentage
loss-on-ignition (LOI) after heating 5 g of oven-dried soil
in a crucible to 375 0C in a muffle furnace for 16 h (Kalra
& Maynard, 1991). Soil exchangeable cations (Ca, Mg,
Na & K) were extracted by shaking 5 g of air-dried soil
with 25 mL of neutral 1N ammonium acetate for 30 min,
filtered (Whatman no.42) and the filtrate analysed by
AAS (AESL, 1999). Soil micronutrients were extracted
by adding 50 mL of diethylene triamine penta acetic acid
(DTPA) to 2.5 g of air dried soil in 250 mL open conical
flasks and filtered. The micronutrients in the filtrate were
measured by AAS (Amacher, 1996).
Plant material: Plants with their reproductive structures
were identified after comparison with specimens in the
National Herbarium of the Royal Botanic Gardens,
Peradeniya. Herbarium sheets were prepared of the

Journal of the National Science Foundation of Sri Lanka 39 (4) December 2011
Plant diversity and soil of Ussangoda serpentine site
357
collected species. Plant samples for analysis were bagged
and labelled in polythene bags.
Plant analysis: Plant samples were cleaned with a nylon
brush under running tap water to remove adhered soil
and then washed with ethylene diamine tetra-acetic acid
(EDTA) for 1min and rinsed thrice with de-ionized water
to remove any surface bound metals. The samples were
dried at 80 °C for 24 h in a forced air oven and ground
in a plant grinder and stored in airtight bags. Samples
of 500 mg were ashed in a muffle furnace at 550 °C for
8 h and the ash dissolved in 5mL of concentrated HCl
and volume made up to 50 mL with distilled water. The
macro and micro nutrients were determined by AAS.
Statistical analysis: Statistical analysis was carried out
using the Minitab 14 statistical software package. One-
way ANOVA along with multiple comparison of means
using Tukey’s test was performed.
RESULTS
The serpentine plain has isolated patches of dense shrubs
and trees, and rocky outcrops. The plants on the plains
are prostrate, stunted with an extensive root system. The
plain is clearly demarcated from the surrounding non-
serpentine area by large shrubs and trees. The surface
soil is a fine reddish dust with small lateritic stones.
Soil characteristics
The serpentine and non-serpentine soils showed
significant differences in their physical and chemical
properties. Within the serpentine site there are two
contrasting vegetation types, the prostrate stunted
species and the patches of shrubs and trees. The available
moisture was low in the serpentine soil while the shrub
patches had significantly higher moisture approaching
that of the non-serpentine soil. This was not associated
Figure 1: Map of Sri Lanka showing the serpentine sites and an enlarged view of the Ussangoda serpentine
site. (S = Serpentine, NS = Non-serpentine, W = Water hole).

December 2011 Journal of the National Science Foundation of Sri Lanka 39 (4)
358
H.A.S. Weerasinghe & M.C.M.Iqbal
with the organic matter content of the soils (Table 1). The
CEC of the serpentine soils was also significantly lower.
The pH of the serpentine soil tended towards acidity
and the non-serpentine soil was significantly neutral
(Table 1). The conductivity of the non-serpentine soil
was significantly higher than that of the serpentine soil
and the soil of the shrub patches.
The chemical properties of the serpentine soil
showed highly significant differences from the non-
serpentine soil. The Ca content of the non-serpentine soil
was 3 to 15 times higher than that in the shrub patches and
the serpentine soil respectively. Although Mg content of
the different soils were comparably similar, the difference
in Ca was reflected in the Ca:Mg ratio (Table 1). The
serpentine soil has a characteristic value of 0.60. The
shrub patches with a higher Ca content of 3 to 4 times
that of the serpentine soil have a Ca:Mg ratio higher than
that of the serpentine soils. Similarly the Ni content was
much higher than in the non-serpentine soils. Fe does not
show a significant difference between the soil types and
Mn was high in the serpentine soil (Table 1).
Flora
In this study the diversity of plant families and
species in Ussangoda is based on a field survey of
the flora on the serpentine soil and a previous survey
conducted by the International Union for Conservation
of Nature (IUCN, 2004) on all the species in the
Ussangoda region.
This study identified 26 plant families on the
serpentine soil. Of these, nine families with 12 species
occur on the serpentine plain (Table 2). These species
are prostrate, dwarfed or stunted, with small leaves, thick
sclerenchymatous stems and extensive root systems. The
prostrate species E. alsinoides, had three colour morphs:
white, pale purple and dark blue. They occurred either
in close proximity or spread apart on the plains. The
dominant family on the serpentine plains was Fabaceae
with three species. The other families occur in patches of
shrubs and trees on the serpentine plain.
Table 1: Physical and chemical properties of the serpentine soil in Ussangoda and the adjacent non-serpentine soil
(mean values ± SE)
Soil parameter Serpentine Island1 1 Island1 2 Non-serpentine
(Area = 1km2) (Area = 2141m2) (Area = 764 m2)
n 14 9 7 6
Moisture (%) 4.4 ± 0.26 a 7.01 ± 0.64 b 8.92 ± 0.42 b,c 10.61 ± 1.6 c
Organic matter (%) 2.96 ± 0.26 a 5.46 ± 0.42 b 5.84 ± 0.23 b,c 4.31 ± 0.24 a,b
CEC 6.76 ± 0.85 a 6.29 ± 1.59 a 9.72 ± 1.57 a,b 13.44 ± 1.23 b
(cmol(+)/kg of
dry soil)
n 6 4 4 5
pH 5.32 ± 0.21 a 5.95 ± 0.13 a,b 6.23 ± 0.21 b 7.3 ± 0.19 c
Conductivity 61 ± 6.54 a 55.8 ± 9.54 a 130 ± 28 b 162.4 ± 10.9 b
(micro-siemens)
Exchangeable cations
n 29 9 7 8
Ca (µg/g) 187 ± 27.5 a 645 ± 84.8 b 905 ± 135 b 2911 ± 650 c
Mg (µg/g) 311 ± 48.3 a 352 ± 50.3 a 456 ± 35.7 a 519 ± 53.2 a
Ca/Mg 0.601 1.832 1.984 5.608
K (µg/g) 140 ± 12.2 a 321 ± 38.3 b 314 ± 21.7 b 338 ± 93.9 b
Na (µg/g) 67 ± 16.8 a 50 ± 9.44 a 67 ± 10.1 a 636 ± 207 b
Available micro-elements
n 31 9 7 8
Ni (µg/g) 101 ± 12.1 a 137 ± 12.14 a 151 ± 7.8 a 4 ± 1.13 b
Fe (µg/g) 65 ± 4.61 a 56 ± 3.19 a 46 ± 3.35 a 62 ± 15.2 a
Mn (µg/g) 42 ± 6.96 a 15 ± 1.73 b 10 ± 0.94 b 19 ± 4.18 a,b
1 Island refers to the patches of shrubs and trees on the serpentine plains. n = number of samples.
Figures followed by different letters in the same row are significantly different at p<0.05.

Journal of the National Science Foundation of Sri Lanka 39 (4) December 2011
Plant diversity and soil of Ussangoda serpentine site
359
The species recorded in the IUCN plant survey were
compared with the species identified in this study. Five
families with seven species growing only on the serpentine
soil were recorded, which were not recorded by the IUCN
study (IUCN 2004) in the Ussangoda region (Table 3).
Of these Violaceae occurs on the plains and species of
the other four families occur as shrubs or trees on the
serpentine site. The diversity of families and their species
was low on the serpentine soil. The survey of species has
identified 16 species growing only on the serpentine soil
and 44 species growing on serpentine and non-serpentine
soil in Ussangoda (Table 3). The habits of these species
were climbers, herbs, shrubs and trees. Monocotyledons
were confined to Cyperaceae and Poaceae with four
species. Of the 31 families on serpentine soil, Fabaceae
included six species, Capparaceae and Malvaceae five
species each and the rest had three species or less. In
contrast, the diversity of species within the families
was greater on the non-serpentine region of Ussangoda.
Fabaceae had the most with 14 species while 10 families
had three or more species.
Five species growing on the serpentine soil showed
Ni accumulation. The species E. alsinoides, Euphorbia
thymifolia, C. kleinii, and Vernonia cinerea accumulated
560 – 830 µg/g (dry weight) of nickel in their tissues
while Hybanthus enneaspermus was a hyperaccumulator
(Table 2). Although Fimbristylis ovata showed low
nickel accumulation on the serpentine soil, under
experimental hydroponic conditions with Ni, levels of
2000 µg/g (dry weight) were detected (unpublished data,
M.C.M. Iqbal).
Name of the plant Family Plant Nickel (µg/g
part dry weight)
Hybanthus enneaspermus Violaceae Lvs+sh 1828 ± 106
Evolvulus alsinoides Convolvulaceae Lvs+sh 828 ± 36
Euphorbia thymifolia Euphorbiaceae Lvs+sh 745 ± 67
Cassia kleinii Fabaceae Lvs+sh 652 ± 23
Desmodium triflorum Fabaceae not determined
Crotolaria sp. Fabaceae not determined
Vernonia cinerea Asteraceae Lvs+sh 561 ± 14
Eragrostis sp. Poaceae Lvs 233 ± 8
Frimbristylis ovata Cyperaceae Lvs 220 ± 2
Frimbristylis falcata Cyperaceae Lvs 73 ± 19
Lvs = Leaves, sh = shoots
Table 2: Nickel (Ni) content (mean ± SE) in some prostrate plants on the Ussangoda
serpentine plain
Families Habit Species Only on On Source
serpentine serpentinite (Ref no.)
soil and non- P =
serpentinite present
soils survey
Apocynaceae H Carissa spinarum X P
Asclepiadaceae S Calotropis gigantea X P
Asparagaceae C Asparagus racemosus X P
Asteraceae H Eupatorium odoratum X P
H Vernonia cinerea X P
Boraginaceae S Ehretia laevis X P
Cactaceae H Opuntia sp. X 1, P
Continued on page 362...
Table 3: Plant species and their families found on serpentine soil and non-serpentine regions in Ussangoda (habit: H – herb,
S – shrub, C – climber, T – tree).

December 2011 Journal of the National Science Foundation of Sri Lanka 39 (4)
360
H.A.S. Weerasinghe & M.C.M.Iqbal
Capparaceae S Capparis pedunculosa X P
S Capparis rotundifolia X P
S Capparis sepiaria X P
S Capparis zeylanica X P
S Maerua arenaria X P
Celastraceae S Maytenus emarginata X P
Convolvulaceae H Evolvulus alsinoides X 1,10,11, P
H Ipomoea pes-caprae X P
Cyperaceae H Fimbristylis falcata X 10, 11, P
H Fimbristylis monticola X P
H Frimbistylis ovata X P
Euphorbiaceae H Euphorbia hirta X P
H Euphorbia thymifolia X P
S Flueggea leucopyrus X P
Fabaceae C Acacia caesia X P
T Cassia auriculata X 11, P
H Cassia kleinii X P
H Crotolaria tecta X P
H Desmodium sp. X 11, P
S Dichrostachys cinerea X P
Flacourtiaceae Casearia zeylanica X P
S Flacourtia indica X P
T Scolopia acuminata X P
Linaceae T Hugonia mystax X P
Loranthaceae S Dendrophthoe falcata X P
Malvaceae H Abutilon indicum X P
H Pavonia odorata X P
H Sida acuta X P
H Sida cordifolia X P
H Sida rhombifolia X P
Melastomataceae S Memecylon umbellatum X P
Melliaceae T Azadirachta indica X P
Menispermaceae C Pachygone ovata X P
Olacaceae S Olax imbricata X P
Oleaceae C Jusminum angustifolium X P
Poaceae H Eragrostis tenella X P
Rhamnaceae S Ziziphus oenoplia X P
Rhizophoraceae H Cassiopourea ceylanica X P
Rubiaceae T Canthium dicoccum X P
T Morinda tinctoria X P
T Tarenna asiatica X 11, P
Rutaceae S Glycosmis mauritiana X P
T Limonia acidissima X P
C Toddalia asiatica X 11, P
Salvadoraceae T Azima tetracantha X P
T Salvadora persica X P
Sapindaceae T Allophylus cobbe X P
T Lepisanthes tetraphylla X P
T Sapindus emarginatus X P
Verbenaceae S Lantana camara X P
H Stachytarpheta jamaicensis X P
Violaceae H Hybanthus enneaspermus X 10, 11, P
Vitaceae H Cissus quadrangularis X P
The plant species on the Ussangoda serpentine site include those identified from this and other cited studies. The species on the non-
serpentine regions in Ussangoda are based on the IUCN (2004) survey.

Journal of the National Science Foundation of Sri Lanka 39 (4) December 2011
Plant diversity and soil of Ussangoda serpentine site
361
DISCUSSION
A defining characteristic of a serpentine soil is a Ca: Mg
ratio of less than 0.6 (Brooks, 1987). The Ussangoda
serpentine soil on the plains showed a ratio of 0.6, and soil
under the shrubs and trees a ratio of 1.8 to 1.9. This was
significantly below than that of the non-serpentine soil.
Although the Mg content was not significantly different
between the serpentine and non-serpentine soils, the Ca
content was almost 15 times less on the serpentine plains.
In a previous study on the soil chemistry of Ussangoda
(Rajakaruna & Bohm, 2002), the Ca concentration
determined in the root zones of six species ranged from
112 – 210 µg/g (dry weight), which compares well with
a mean value of 187 µg/g (dry weight) in this study.
However, the Mg content determined in this study was
3–5 times of that was determined by Rajakaruna and
Bohm (2002). In their study of four serpentine soils in
Sri Lanka, Ussangoda showed the highest Ca:Mg ratio
of 1.3 – 2.4 while the other three sites had Ca:Mg as low
as 0.07 (in Ginigalpelessa) due to a high Mg content.
A serpentine site in Malaysia has shown a similar ratio
of 0.6 (Brearley, 2005) and a serpentine site in Portugal
(Lázaro et al., 2006) showed similar high values of
exchangeable Mg over Ca in the soil to give a Ca:Mg
ratio of 0.7 whereas ratios of the non-serpentine sites
were 2.4 to 6.9. In a serpentine site in Italy similarly low
values of 0.3 were found (Chiarucci et al., 1998) due to
the high Mg content in the soil. The Ussangoda site falls
within the basic definition for a serpentine soil based on
Ca:Mg ratio.
The CEC determined in this study is similar to that
obtained previously on the same site (Rajakaruna &
Bohm, 2002). The other serpentine sites in Sri Lanka in
Indikolapelessa and Ginigalpelessa had a higher CEC
(Rajakaruna & Bohm, 2002) of 25-50 cmol(+)/kg of soil.
Higher CEC values of 16 cmol(+)/kg was also shown by a
temperate serpentine soil (Lázaro et al., 2006). The soil pH
values in Ussangoda determined previously (Rajakaruna
& Bohm 2002) were more acidic (4.3 – 4.9) than that
was determined in this study, which could be due to time
of sampling and determination of the pH. In Malaysia,
a pH value of 5.3 similar to Ussangoda was determined
(Brearley, 2005). Acidic pH values were also obtained in
serpentine soils in temperate conditions (Lázaro et al.,
2006). The loss-of-ignition determined in a serpentine
site in Italy of 10.8 – 23.7% (Chiarucci et al., 1998) was
much higher than the 2.9 recorded in Ussangoda. This
could be accounted for by the cooler temperate climate
while Ussangoda is exposed to continuously warm
temperatures. However, high loss-of- ignition values of
12.6% were determined in a Malaysian serpentine soil
(Brearley 2005). The physical and chemical properties
of the Ussangoda serpentine soil are similar to that are
found elsewhere. However, there is a wide range of
values depending on the time and location of sampling
within the site.
Brooks (1987) first identified three species on
the Ussangoda (E. alsinoides) and Uda-Walawe
(Cymbopogon flexuosus and Morinda tinctoria)
serpentine sites in Sri Lanka. He concluded that the
serpentine flora was impoverished in species number
and endemism without further potential for botanical
research. Further studies (Seneviratne et al., 2000) on
the flora of Ussangoda identified 14 species confined
to specific areas on the serpentine plain and of limited
distribution. In a preliminary survey of four serpentine
sites in Sri Lanka Rajakaruna and Bohm (2002) reported
on the flora, soil characteristics and heavy metal uptake
by the plants.
Flora on serpentine soils are generally poor in
species diversity and show a high degree of endemism.
While the flora on the Ussangoda soil is poor in species
and families compared to the surrounding non-serpentine
region, attention has been drawn to the fact that although
Sri Lanka has a greater degree of biodiversity per unit
area than other Asian countries, endemic species were
not found on the serpentine soils (Rajakaruna & Bohm,
2002).
From a previous study of the flora of four serpentine
sites in Sri Lanka, Rajakaruna & Bohm (2002) concluded
that the floristic composition was typical of tropical
serpentine habitats. Their study added at least 23 new
genera from Sri Lanka to the list of plants known to
occur on serpentine soils. Of the 51 species identified
in this study, upto now, growing on the plains and as
patches of shrubs and trees on the Ussangoda serpentine
soil, two species, Vernonia zeylanica (Grierson, 1980)
and Scolopia acuminate (Verdcourt, 1996) were found
endemic to Sri Lanka while C. kleinii (Rudd, 1991) is
apparently endemic to Sri Lanka and Southwest India.
Colour polymorphism of the flowers of E. alsinoides
growing on the serpentine soil was previously reported
(Seneviratne et al., 2000; Rajakaruna & Bohm, 2002).
Both studies refer to two colours, while this study
identified three colour morphs. The non-serpentine
species is also known to have two colours (Austin, 1980).
Rajakaruna and Baker (2004) speculate if the colour
morphs could be due to edaphic differences in their
micro-habitat. However, in this study, the three flower
types were observed in plants growing very close to

December 2011 Journal of the National Science Foundation of Sri Lanka 39 (4)
362
H.A.S. Weerasinghe & M.C.M.Iqbal
each other as well as far apart. Seneviratne et al. (2000)
found distinct flavonoid profiles and suggested they are
“flavonoid races”.
The plant tissue concentration of heavy metals (Ni,
Fe, Mn) growing on the serpentine soil were above the
amount considered to be normal for non-serpentine plants
(Greger, 2004). Such accumulation of heavy metals is
characteristic of serpentine adapted plants (Brooks, 1987).
Plants capable of accumulating over 1000 µg metal/g dry
leaf tissue are considered as hyperaccumulators (Baker &
Brooks, 1989). H. enneaspermus accumulated over 1000
ppm Ni and three other species accumulated 500 – 800
ppm Ni (Table 2). The different Ni contents in the different
species suggest that the plants have varied mechanisms
of overcoming Ni toxicity in the soil by increased uptake
and sequestering of the Ni (hyperaccumulators) or by
excluding from uptake (species with low Ni). Previous
studies (Seneviratne et al., 2000) found four other
species in addition to H. enneaspermus, that were able
to hyperaccumulate Ni. Rajakaruna and Bohm (2002)
had found H. enneaspermus as a Ni hyperaccumulator
besides E. alsinoides and Crotolaria biflora. While
the status of H. enneaspermus is unequivocal as a
Ni hyperaccumulator, the differences in the other
species could be due to time and location of sampling
or analytical methods. Since surface contamination of
the plant tissue with the metal rich soil is a problem,
in this study plant tissues were washed with EDTA to
remove surface bound metals, which may account for
the lower values determined. According to Rajakaruna
and Bohm (2002), H. enneaspermus and E. alsinoides
occur in serpentine sites in Queensland, Australia but
are not Ni hyperaccumulators. They suggest that since
many varieties of both species are known, the Sri Lankan
species may represent a physiologically distinct form,
perhaps an edaphic race.
After the primary mineralization of the serpentine
rock, the elemental composition of the soil is also altered
by the colonization of vegetation, which along with
climate determines the present mineral composition
(Chiarucci et al., 1998). There is a dynamic interaction
between the soil and the vegetation determining the
physical and chemical characters of the soil at a given
time. Tropical serpentine soils are different from
temperate soils, primarily due to the cooler climate and
vegetation. However, more studies on tropical serpentine
sites are necessary to better understand the tropical
serpentine soils and their interaction with the flora that is
established on these soils. The Ussangoda serpentine site
has the basic soil attributes of a typical serpentine soil
and has a particularly low Ca content. The floral diversity
is restricted and shows adaptation to the soil conditions
morphologically and physiologically. Amongst the
species are hyperaccumulators as well as excluders of
Ni.
Acknowledgement
We thank the Director, Department. of Archaeology for
permission to collect field samples from the Ussangoda
serpentine site, Dr. D.S.A. Wijesundera, Director
General of the National Botanic Gardens, Peradeniya
for permitting the use of the Herbarium for identification
of the species. Dr. G.W.A.R. Fernando, Department of
Physics, The Open University of Sri Lanka, Ms. Y.A.S.
Samithri in the preparation of this manuscript and
Mr. R.B. Hapukotuwa for field assistance. Financial
assistance provided by the British Ecological Society,
the Institute of Fundamental Studies, and the National
Science Foundation of Sri Lanka (RG/2006/EB/08) is
gratefully acknowledged.
References
AESL (1999). 1. Methods for the analysis of soil, plant,
water and environmental samples. Agricultural and
Environmental Services Laboratories, College of
Agriculture and Environmental Sciences, University of
Georgia. Available at (http://aesl.ces.uga.edu/protected/
methods/stl-soil.html). Accessed on 1 May 2011.
Amacher M.C. (1996). Nickel, Cadmium and Lead. 2.
Methods of Soil Analysis: Chemical Methods,
part 3. pp. 739–768. Soil Science Society of America and
American Society of Agronomy, Madison, Wisconsin,
USA.
Anderson J.M. & Ingram J.S.I. (1993). 3. Tropical Soil
Biology and Fertility: A Handbook of Methods. p. 221.
CABI Publishing, Wallingford, UK.
Austin D.F. (1980). Convolvulaceae. 4. A Revised Handbook
to the Flora of Ceylon, vol. I. (eds. M.D. Dassanayake
& F.R. Fosberg), pp. 288–363. Amerind Publishing Co.,
New Delhi, India.
Baker A.J.M. & Brooks R.R. (1989). Terrestrial higher 5.
plants which accumulate metallic elements – a review
of their distribution, ecology and phytochemistry.
Biorecovery 1: 81–126.
Brearley F.Q. (2005). Nutrient limitation in a Malaysian 6.
ultramafic soil. Journal of Tropical Forest Science 17 (4):
596–609.
Brooks R.R. (1987). 7. Serpentine and its Vegetation: a
Multidisciplinary Approach, p. 224. Dioscorides Press,
Portland Oregon, USA.
Chiarucci A., Riccucci M., Celesti C. & de Dominicis 8.
V. (1998). Vegetation-environment relationships in the
ultramafic area of Monte Ferrato, Italy. Israel Journal of
Plant Sciences 46: 213–221.
Dissanayake C.B. & Van Riel B.J. (1978). Petrology 9.
and geochemistry of a recently discovered nickeliferous

Journal of the National Science Foundation of Sri Lanka 39 (4) December 2011
Plant diversity and soil of Ussangoda serpentine site
363
serpentinite from Sri Lanka. Journal of the Geological
Society of India 19: 464–471.
Dissanayake C.B. (1982). The geology and geochemistry 10.
of the Uda Walawe serpentinite, Sri Lanka. Journal of the
National Science Council of Sri Lanka 10(1): 13–34.
Greger M. (2004). Metal availability, uptake, transport 11.
and accumulation in plants. Heavy Metal Stress in Plants
(ed. M.N.V. Prasad), pp. 462. Narosa Publishing House,
New Delhi, India.
Grierson A.J.C. (1980). Compositae. 12. A Revised Handbook
to the Flora of Ceylon, vol. I (eds. M.D. Dassanayake
& F.R. Fosberg), pp. 111–278. Amerind Publishing Co.,
New Delhi, India.
Iqbal M.C.M., Kulasekara L., Rajakaruna N. & Iqbal 13.
S.S. (2006). Plant-soil relations of a serpentine site in the
southern coast of Sri Lanka. Fifth International Conference
on Serpentine Ecology, Siena, Italy, 9-13 May. University
of Siena, Siena, Italy. p. 42.
IUCN (2004). 14. The Environmental Profile of Rekawa,
Ussangoda and Kalametiya (RUK) Coastal Ecosystems. pp.
7–10, 111–117. International Union for the Conservation
of Nature, Sri Lanka Country Office, 53, Horton Place,
Colombo 7.
Kalra Y.P. & Maynard D.G. (1991). Methods manual for 15.
forest soil and plant analysis. Information report NOR-
X-319. Forestry Canada, Northwest region, Edmonton,
Alberta, Canada.
Kruckeberg A.R. (1984). 16. California Serpentines: Flora,
Vegetation, Geology of Soils, and Management Problems,
p. 180. University of California press, Berkeley, California,
USA.
Lázaro J.D., Kidd P.S. & Martínez C.M. (2006). A 17.
phytogeochemical study of the Trás-os-Montes region
(NE Portugal): Possible species for plant-based soil
remediation technologies. Science of the Total Environment
354: 265–277.
Munasinghe T. & Dissanayake C.B. (1979). Is the 18.
highland-eastern Vijayan boundary in Sri Lanka a possible
mineralized belt? Economic Geology 74: 1495–1496.
Munasinghe T. & Dissanayake C.B. (1980). Is the 19.
highland-eastern Vijayan boundary in Sri Lanka a
possible mineralized belt? – a reply. Economic Geology
75: 775–777.
Proctor J. & Woodell S.R.J. (1975). The ecology of 20.
serpentine soils. Advances in Ecological Research 9:
255–366.
Proctor J. (2003). Vegetation and soil and plant chemistry 21.
on ultramafic rocks in the tropical Far East. Perspectives
in Plant Ecology, Evolution and Systematics 6: 105–124.
Rajakaruna N. & Baker A.J.M. (2004). Serpentine: a 22.
model habitat for botanical research in Sri Lanka. Ceylon
Journal of Science 32: 1–19.
Rajakaruna N. & Bohm B.A. (2002). Serpentine and its 23.
vegetation: a preliminary study from Sri Lanka. Journal of
Applied Botany 76: 20–28.
Rudd V.E. (1991). Fabaceae. 24. A Revised Handbook to
the Flora of Ceylon, vol.VII (eds. M.D. Dassanayake
& F.R. Fosberg), pp. 34–381. Amerind Publishing Co.,
New Delhi, India.
Seneviratne A.S., Nandadasa H.G., Fernando W.S., 25.
Sanjeevani H.H.V.M. & Rajapakse R.L.H.R. (2000). The
serpentine vegetation of Ussangoda (Hambantota District)
and nickel accumulating plant species. Proceedings of the
Sixth Annual Forestry and Environmental Symposium,
Kandy, Sri Lanka, 29-30 December. University of
Sri Jayawardenepura, Nugegoda. p.1.
Verdcourt B. (1996). Flacourtiaceae. 26. A Revised Handbook
to the Flora of Ceylon, vol.X (eds. M.D. Dassanayake
& W.D. Clayton), pp. 199-235. Oxford and IBH Co.,
New Delhi, India.