South African Journal of Botany 72 (2006) 284 286

www.elsevier.com/locate/sajb

Short communication

Response of South African indigenous grass species to drought stress

induced by polyethylene glycol (PEG) 6000
L. van den Berg a,*, Y.J. Zeng b
a

School of Environmental Sciences and Development, North West University (Potchefstroom Campus), Potchefstroom 2520, South Africa
b
Gansu Grassland Ecological Research Institute, P.O. Box 61, Lanzhou 730020, Gansu Province, China
Received 14 April 2005; accepted 12 July 2005

Abstract
Degradation in arid and semi-arid regions can be ascribed primarily to environmental conditions such as extreme changes in temperature and
erratic rainfall patterns. Seed germination and seedling growth of three grass species (Anthephora pubescens, Heteropogon contortus and
Themeda triandra) indigenous to arid and semi-arid regions of southern Africa were assessed under controlled conditions that simulated drought
stress using polyethylene glycol (PEG) 6000. Exposure to water potentials below 0.3 MPa significantly decreased germination rates and shoot
lengths for all three species, despite slight differences in their sensitivities to osmotic stress. More dramatic differences between the species were
evident when the effects of water stress on root growth were evaluated. These findings form the basis for future trials involving the use of
indigenous grasses in the restoration of rangelands.
D 2005 SAAB. Published by Elsevier B.V. All rights reserved.
Keywords: Drought stress; Polyethylene glycol (PEG); Restoration; Rangelands

It has been estimated that about nine million square

kilometers of the worlds arid rangelands have been turned into
man-made deserts over the past half century (Le Houerou and
Gillet, 1986). Rangelands are areas unsuitable for cultivation,
but provide forage for animals (Miller, 1997). Rangeland
degradation is often manifested by decreases in plant species
diversity, low grass height and vegetation cover, and increase in
unpalatable grass species and sharp reductions in plant yields.
The dispersal of seeds through oversowing is an important
strategy to actively restore vegetation in degraded areas as their
seed banks are usually depleted of viable seed. One of the
greatest challenges in restoration ecology is to sow a seed type
or cultivar that has the capacity to produce abundant biomass
and cover in a short period of time (Van den Berg, 2002).
In addition to grazing by domestic livestock, rainfall
patterns are the most important factors influencing rangeland
condition (Hoffman and Ashwell, 2001). Drought, extreme
temperatures, increased soil salinity, soil crusting or sand
covering, as well as pathogens and herbivores all adversely
affect the germination and growth of seedlings. Grasses, with
* Corresponding author.
E-mail address: plblvdb@puknet.puk.ac.za (L. van den Berg).

strong development of underground organs, tend to have

efficient adaptive mechanisms to cope with drought, fire and
herbivory and provide superior protection against soil erosion
than most woody shrub and tree species (Imevbore, 2003).
Grasses with network root systems bind soil better than woody
species with taproot systems (Cowling et al., 1986).
Osmotic solutions are used to impose water stress reproducibly
under in vitro conditions (Pandey and Agarwal, 1998). Polyethylene glycol molecules with a M r  6000 (PEG 6000) are inert,
non-ionic and virtually impermeable chains that have frequently
been used to induce water stress and maintain a uniform water
potential throughout the experimental period (Hohl and Peter,
1991; Lu and Neumann, 1998). Molecules of PEG 6000 are small
enough to influence the osmotic potential, but large enough to not
be absorbed by plants (Carpita et al., 1979). Because PEG does
not enter the apoplast, water is withdrawn from the cell and the
cell wall. Therefore, PEG solutions mimic dry soil more closely
than solutions of low M r osmotica, which infiltrate the cell wall
with solutes (Verslues et al., 1998).
The objective of this study was to determine the germination
and morphological responses of Anthephora pubescens Nees,
Heteropogon contortus L. and Themeda triandra Forssk.
(Gibbs Russell et al., 1990) to drought stress induced by

0254-6299/$ - see front matter D 2005 SAAB. Published by Elsevier B.V. All rights reserved.
doi:10.1016/j.sajb.2005.07.006

L. van den Berg, Y.J. Zeng / South African Journal of Botany 72 (2006) 284 286

Table 1
Germination capacity (%) of Anthephora pubescens, Heteropogon contortus
and Themeda triandra after incubation with different osmotic potentials
Treatment (MPa)

Anthephora
pubescens

Heteropogon
contortus

Themeda
triandra

Control
0.3
0.6
0.9
1.2
1.5
1.8
2.1
Species mean
(treatments)

40.75 T 2.06a
41.00 T 2.16a
39.25 T 1.50a
25.25 T 2.99a
9.50 T 3.11a
7.50 T 1.73a
3.25 T 2.63a
0a
20.81 T17.00a

38.00 T 1.15b
38.75 T 4.99a
24.00 T 3.83a
17.00 T 5.72a
10.00 T 5.48a
3.50 T 0.57a
5.25 T 2.99a
0a
17.16 T 14.72b

38.50 T 3.11c
31.00 T 4.16a
16.75 T 1.71a
13.50 T 3.00a
9.25 T 3.30a
7.00 T 3.56a
3.75 T 4.19a
0a
15.06 T 12.95c

Different letters indicate statistically significant differences for germination

capacity ( P < 0.05).

least significant difference (LSD) at the 5% level of significance (Snedecor and Cochran, 1980).
Germination capacity decreases for all three of the species
with a decrease in osmotic potential (Table 1). The final
germination capacity for the control treatment differed significantly for all three species. There is also a significant

(a)

A. pubescens
H. contortus
T. triandra

Shoot length (cm)

PEG 6000. Results from initial germination capacity tests

conducted according to the rules of the International Seed
Testing Association (ISTA, 1976) indicated that 30 -C is the
optimal temperature for seed germination of A. pubescens, H.
contortus and T. triandra. Four replicates of 100 seeds per
species were used to determine germination rates in the
absence of osmotic stress (Fig. 1). Both A. pubescens and H.
contortus show a high germination percentage (75% and 48%)
within the first two days, after which germination seems to
reach a plateau. T. triandra germinates poorly during the first
two days, but after the fourth day, germination rates increased
to 67% and reached a plateau of 77%. Therefore, at 30 -C and
in the absence of osmotic stress, seed from all three species
tested showed a high germination capacity within the first four
days. This could be of significance when planning oversowing
treatments with the specific species for restoration purposes.
Water stress was applied through incubation in seven
different concentrations of PEG 6000 that provide solutions
with water potentials ranging from 0.3 to 2.1 MPa (Michel
and Kaufmann, 1973). PEG 6000 was dissolved in water and
placed in a shaker bed (25 -C) for 16 h. Four replicates of 50
pure seeds per species were used for each of the different
osmotic potential experiments. Seeds of the different species
were germinated at 30 -C on filter paper discs moistened with
8 ml of the different solutions. After the addition of the
solution, as well as the seeds, each petri dish was weighed and
the weight recorded. Petri dishes were weighed daily and
supplemented with distilled water to the original weight. The
seeds were investigated for germination every second day.
Germination data were analysed using the statistical program
GenStat (2000). The experiment was designed as a completely
randomised design with two factors. The first factor was the
species and the second the drought stress treatments. Differences in germination between species and between treatments,
as well as the interaction between these variables, were tested
for an analysis of variance. The arcsine transformation was
used to stabilise variances. After the transformation, the data
was acceptably normal with homogeneous treatment variances.
Treatment means were separated using Fishers protected t-test

285

4
3
2
1
0

Control -0.3

-0.6

-0.9

-1.2

-1.5

A. pubescens
H. contortus
T. triandra

Root length (cm)

Percentage (%)

60

40

20

-2.1

Treatments (MPa)

(a)

100

80

-1.8

4
3
2
1
0

Control -0.3
0

10

12

-0.6

-0.9

-1.2

-1.5

-1.8

-2.1

Treatments (MPa)

Days
Fig. 1. Germination rate of Anthephora pubescens (), Heteropogon contortus
(>) and Themeda triandra (- - -) at 30 -C over a period of 12 days.

Fig. 2. Growth rates of emerging shoots (a) and roots (b) of Anthephora
pubescens, Heteropogon contortus and Themeda triandra seedlings exposed to
various treatments of decreasing osmotic potentials.

286

L. van den Berg, Y.J. Zeng / South African Journal of Botany 72 (2006) 284 286

difference in the germination of the three species when

comparing the total germination percentage of all treatments
(species mean). Both A. pubescens and H. contortus show
better germination in response to the 0.3 MPa osmotic
treatment than the control treatment. Although seed germination occurred at all of the different osmotic potentials tested,
with the exception of 2.1 MPa, none of the three species
tested showed a very high tolerance of water stress.
Although the germination capacity decreased with a
decrease in osmotic potential, seeds from all species germinated at an osmotic potential of 1.8 MPa. Although the seeds
did not germinate at an osmotic potential less than 1.8 MPa,
this might indicate an important survival mechanism that
ensures dormancy within the soil seed bank until sufficient
moisture is available for optimal seed germination and seedling
establishment.
Drought plays an important role not only in determining
germination rates, but also influences seedling development.
Even when enough moisture is available to promote germination, in the absence of subsequent rain, growth of newly
germinated seedlings will be compromised with potentially
fatal consequences. Five seedlings of each treatment were
selected and root and shoot lengths were measured on the fifth
day of incubation to determine whether a decrease in osmotic
potential would affect seedling growth. Analysis of variance
(ANOVA) was used to test for differences between the three
species. The data was acceptably normal but with some
heterogeneous treatment variances. Glass et al. (1972) indicated that the consequences on inference after ANOVA are not
serious, as no transformation could be found to rectify the
problem. Species means were separated using Fishers protected t-test least significant difference (LSD) at the 1% level of
significance (Snedecor and Cochran, 1980).
For all three the species, the shoot lengths of seedlings
decreased with an increase in water stress (Fig. 2a). Root
lengths of A. pubescens and H. contortus were greater under
osmotic potentials of 0.3 MPa than under control conditions
and A. pubescens seedlings displayed even more dramatic root
elongation at 0.6 MPa (Fig. 2b). In contrast, root elongation
of T. triandra decreased progressively with any decrease in
osmotic potential (Fig. 2b). All three species displayed
decreases in root length as osmotic potential was decreased
from 0.6 to 2.1 MPa (Fig. 2b).
Decreases in the external osmotic potential generally caused
decreased morphological development of organs from the
young grass seedlings of the three species tested. Whereas
shoot growth was always decreased by exposure to all of the
stress levels tested, there was a slight increase in root length
associated with the 0.3 MPa treatments for A. pubescens and
H. contortus. Most likely, this reflects an adaptive response
involving an increase in root length to reach water deeper in the
soil.

This study has shown that specific environmental conditions

are needed for optimal seed germination of A. pubescens, H.
contortus as well as T. triandra and that water stress
differentially affects germination capacity as well as root and
shoot lengths in these three species. All three species seem
suitable for restoration trials in areas that are not subjected to
short periods of drought, but field studies are needed to
determine their suitability for areas with specific environmental
conditions.
References
Carpita, N., Sabularse, D., Monfezinos, D., Delmer, D.P., 1979. Determination
of the pore size of cell walls of living plant cells. Science 205, 1144 1147.
Cowling, R.M., Roux, P.W., Pieterse, A.J.H., 1986. The Karoo biome: a
preliminary synthesis: Part 1. Physical environment. South African National
Scientific Programmes Report 124, 1 100.
GenStat for Windows, 2000. Release 4.2, fifth edition. Oxford: VSN
International.
Gibbs Russell, G.E., Watson, L., Koekemoer, M., Smook, L., Barker, N.P.,
Anderson, H.M., Dallwitz, M.J., 1990. Grasses of southern Africa.
Memoirs of the Botanical Survey of South Africa 58, 1 437.
Glass, G.V., Peckham, P.D., Sanders, J.R., 1972. Consequences of failure to
meet assumptions underlying the fixed effects analyses of variance and
covariance. Review of Educational Research 42, 237 288.
Hoffman, T., Ashwell, A., 2001. Nature Divided: Land Degradation in
South Africa. University of Cape Town Press, Cape Town, pp. 99 100.
ISBN: 1-9197-1354-9.
Hohl, M., Peter, S., 1991. Water relations of growing maize coleoptiles.
Comparison between mannitol and polyethylene glycol 6000 as external
osmotica for adjusting turgor pressure. Plant Physiology 95, 716 722.
Imevbore, A.M., 2003. Desertification and desiccation as a threat to the conservation and utilization of biodiversity. United Nations Convention to
Combat Desertification. http://www.unccd.int/knowledge/INCDinfoSeg/parti.
php.
International Seed Testing Association (ISTA), 1976. International rules for
seed testing. Seed Science and Technology 4, 5 177.
Le Houerou, H.N., Gillet, H., 1986. Conservation versus desertification in
African arid lands. In: Soule, M.E. (Ed.), Conservation Biology: the Science
of Scarcity and Diversity. Sinauer, Sunderland, Massachusetts, pp. 1 20.
ISBN: 0-8789-3795-1.
Lu, Z., Neumann, P.M., 1998. Water-stressed maize, barley and rice seedlings
show species diversity in mechanisms of leaf growth inhibition. Journal of
Experimental Botany 49, 1945 1952.
Michel, B.E., Kaufmann, K., 1973. The osmotic potential of polyethylene
glycol 6000. Plant Physiology 51, 914 916.
Miller, D., 1997. Rangelands and range management. International Center for
Integrated Mountain Development Newsletter 27, 1 3.
Pandey, R., Agarwal, R.M., 1998. Water stress-induced changes in praline
contents and nitrate reductase activity in rice under light and dark
conditions. Physiological and Molecular Biology of Plants 4, 53 57.
Snedecor, G.W., Cochran, W.G., 1980. Statistical Methods, 7th ednR Ames,
Iowa State University Press, p. 507.
Van den Berg, L., 2002. The evaluation of a number of technologies for the
restoration of degraded rangelands in selected arid and semi-arid regions of
South Africa. MSc Thesis, Potchefstroom University for Christian Higher
Education, Potchefstroom, South Africa.
Verslues, P.E., Ober, E.S., Sharp, R.E., 1998. Root growth and oxygen relations
at low water potentials. Impact of oxygen availability in polyethylene
glycol solutions. Plant Physiology 116, 1403 1412.