Content uploaded by Zewdie Bishaw
Author content
All content in this area was uploaded by Zewdie Bishaw on Oct 27, 2017
Content may be subject to copyright.
Journal of Agricultural Science and Technology B 3 (2013) 623-634
Earlier title: Journal of Agricultural Science and Technology, ISSN 1939-1250
Effects of Temperature, Relative Humidity and Moisture
Content on Seed Longevity of Shrubby Russian Thistle
(Salsola vermiculata L.)
Abdoul Aziz Niane1, Paul Christiaan Struik2 and Zewdie Bishaw1
1. International Center for Agricultural Research in the Dry Areas (ICARDA), Beirut, Lebanon
2. Centre for Crop Systems Analysis, Wageningen University, AK Wageningen NL-6700, Netherlands
Received: June 24, 2013 / Published: September 20, 2013.
Abstract: Salsola vermiculata is a highly palatable shrub and widely used in rangeland rehabilitation programs, but has short seed
longevity. To identify the most cost effective storage method for S. vermiculata, experiments were carried out to test the effects of
fruit bracts (wings), temperature regimes, seed moisture and packaging methods on storage life. Seed samples were removed from
storage at monthly intervals for testing and towards the end of the experiments samples were transferred from hermetic to ambient
storage conditions and tested for germination. Experiment 1 continued for 1,140 days, Experiment 2 for 720 days. For de-winged
seed, high moisture content increased seed longevity, suggesting that desiccation susceptibility is one of the causes of limited
longevity in this species. Most longevity regression lines of winged seeds had negative intercepts suggesting increase in germination
resulting from gradual dormancy-breaking. Drying and packaging alone increased longevity by 7.6 and 3.8 times in Experiments 1
and 2, respectively. Samples kept at lower temperature and lower moisture treatments survived longer under ambient conditions.
Increased longevity by drying and vacuum packaging alone can provide simple, cost effective and environmentally friendly options
for rangeland rehabilitation programs.
Key words: Salsola vermiculata L., seed storage, vacuum packaging, seed longevity, probit analysis.
1. Introduction
Rangeland degradation is taking place at alarming
rates in arid Mediterranean rangelands [1-3]. Severe
depletion of soil seed banks associated with rangeland
degradation limits self-regeneration, thus necessitating
reseeding [4]. The seed required for rehabilitation is
usually collected from wild plants of the target species
growing in less degraded areas of the rangelands. Due
to temporal and spatial erratic rainfall distribution in
arid rangelands [5, 6], the required quantities of high
quality seed can not be harvested every year. To
mitigate seed shortage in drought years, to maintain
seed stocks for use in range nurseries, and for
distribution to local communities, large seed stocks
Corresponding author: Abdoul Aziz Niane, Ph.D., research
field: soil seed bank dynamics in arid rangeland rehabilitation.
E-mail: a.niane@cgiar.org.
are collected in seasons with an abundant harvest.
These stocks are stored under ambient conditions for
long periods. In most rangeland rehabilitation
programs in West Asia and North Africa, shrub seed
stocks are kept in simple storage structures to
minimize cost. This makes seed storability under
ambient storage conditions critical for rangeland
rehabilitation through reseeding.
Rangeland rehabilitation programs in West Asia
and North Africa (WANA), including Syria, rely
heavily on Chenopodiaceae species, especially
Atriplex spp. and Salsolavermiculata L. [7-9].
However, research on the physical and physiological
seed quality attributes and propagation methods for
shrubs of the arid Mediterranean basin, is limited [10].
Research on saltbush (Atriplexhalimus L.) showed
significant variation in seed quality amongst seeds
D
DAVID PUBLISHING
Effects of Temperature, Relative Humidity and Moisture Content on Seed Longevity
of Shrubby Russian Thistle (Salsola vermiculata L.)
624
from different individual shrubs [11]. Three categories
of seed with different germination rates could be
differentiated.
The shrubby Russian thistle (Salsola vermiculata L.)
is a major species in the Syrian rangeland flora and
rehabilitation programs [12]. It is a native species with
high ecological and forage value found distributed
throughout the arid, semi-arid, saline and hyper-saline
ecosystems of temperate and subtropical regions [13,
14]. It has a high success rate of establishment when
self-sown, direct seeded or transplanted. It is easily
propagated from seed and produces high quality
biomass for feed and good seed yield for direct
sowing under a wide range of rangeland conditions in
WANA region [15]. However, seed of S. vermiculata
loses its viability within 6-9 months under ambient
storage conditions [16, 17].
Effects of storage time, moisture and temperature
on seed longevity have been investigated and reported
for many crops and wild plant species. Based on their
desiccation tolerance, seeds are classified as orthodox
and recalcitrant [18]. The orthodox seeds include a
wide range of annual species [19] for which
Harrington’s Rule of thumb applies [20]. The term
recalcitrant refers to those species for which
Harrington’s Rule does not apply because desiccation
results in rapid loss of seed viability.
For S. vermiculata, research findings have already
shown that by reducing storage temperature, seed
longevity is significantly extended [16, 21]. However,
the effect of controlling seed moisture on storage has
not been properly investigated. Therefore, this study
was intended to investigate the combined effects of
storage temperature and seed moisture on S.
vermiculata seed longevity in order to determine the
most cost-effective storage conditions for use in arid
rangeland rehabilitation.
2. Materials and Methods
2.1 Test Material
Mature seed from Syrian ecotypes of Salsola
vermiculata L. were collected in October 2002 and
2003 from the ICARDA rangeland nursery site
located in North of Syria at 36°01'N, 36°56'E and 284
m above sea level. The seed stocks were cleaned and
then subdivided into two parts, and dried to different
moisture contents and stored for 1,140 days (38
months) in Experiment 1 and 720 days (24 months) in
Experiment 2. One part of the seed stock was dried to
low moisture content of 7% and 6.5% in Experiments
1 and 2, respectively, by storing the seed for 6 weeks
in a dehumidification room set at 16 °C and 18%-22%
relative humidity. The other part was kept at its
harvest moisture content level of 10.7% and 9.6% in
Experiments 1 and 2. Then the two batches of seed
were subdivided into two parts. One part was
de-winged using a Westrup-La-h brushing machine
(http://westrup.com/Products-Seed-and-Grain/Laborat
ory-equipment/LA-H) and the other part was left with
wings intact in its natural condition. De-winging was
done to break dormancy by eliminating the
germination inhibitors accumulated in the wings [16].
These different batches were stored in normal paper
envelopes in non-vacuum sealed polythene bags or in
sealed vacuum packages prepared using a chamber
type vacuum packaging machine equipped with air
extraction and heat sealing facilities at ICARDA gene
bank. The steps described above resulted in eight
batches of seed, namely non-dried and dried batches
of winged and de-winged seeds in vacuum-sealed or
paper packaging in sealed polythene bags.
2.2 Statistical Design and Set-up
In both Experiments 1 and 2, a completely
randomized factorial design was carried out with four
treatments: seed type (winged vs. de-winged), three
combinations of seed moisture content (SMC) and
packaging, combined with three storage temperatures
and time of storage. The treatment levels were winged
or de-winged seeds, high SMC with vacuum
packaging, high initial SMC without vacuum
packaging and low SMC with vacuum packaging; the
Effects of Temperature, Relative Humidity and Moisture Content on Seed Longevity
of Shrubby Russian Thistle (Salsola vermiculata L.)
625
storage temperatures were -21, 4 or 24 °C. Low initial
SMC could not have been maintained without vacuum
packaging, neither high SMC could have been
maintained under non vacuum packaging. In both
Experiments 1 and 2, there were two replications and
samples were removed at regular intervals for testing.
In each of the three storage temperatures, 288 seed
samples were stored, representing two replications of
six treatment combinations and 24 sample withdrawal
with 50 seeds per experimental unit. In Experiment 2,
extra seed packages were stored for 720 days, then
transferred to non-vacuum polyethylene bags, kept at
24 °C and then tested at one month intervals for a
period of 4 months to assess the seed longevity after
hermetic storage.
In Experiment 1, the sampling interval was one
month for the first 18 months and 4 months for the last
20 months. However, the data of the 7th and 8th
months for the winged seeds were removed because of
incubator breakdown. In Experiment 2, the sampling
interval was one month throughout the period.
2.3 Seed Testing Procedures
Each month, a total of 12 envelopes, representing
two replicates of the six treatment combinations were
drawn from each of the three temperature regimes and
tested for germination according to the International
Seed Testing Association rules [22]. Petri dishes of 9
cm diameter with two layers of Whatman No. 41 filter
paper were used in the germination tests. For each
treatment combination, two replicates of 50 seeds were
planted and placed in a germination chamber set at 20 ±
2 °C and light (8 h, fluorescent light of 4.22 W/m2),
dark (16 h) regime. The samples were watered every
two days for a period of 10 days and germination was
then assessed. The Petri dishes were kept in the
germination room for up to 20 days after evaluation but
no additional germination was observed.
2.4 Statistical Analysis
Analysis of variance was carried out to evaluate the
significance of the main treatment effects and the
interactions between them. The means and their
standard errors were computed using Genstat
statistical package [23]. The restricted maximum
likelihood (REML) procedure was used to test the
significance of main effects and interactions of the
treatment factors and to estimate the standard errors of
the means. Instead of simple analysis of variance,
REML facilitated to model the unbalanced design
arising from the fact that seed could not be maintained
at low moisture content without vacuum packaging
and that zero germination percentages were recorded
in the high moisture, paper-packaged seeds.
Regression analysis was applied by plotting
germination proportions against time of storage in
days to quantify the effects of time on seed longevity.
The time in days for seed viability to drop to 50%
(known as P50) was estimated from the probit model.
3. Results
3.1 Longevity Trends
The overall analysis of variance showed that all
possible three-way interactions were statistically
significant (P < 0.01). To facilitate interpretation and
simplify presentation, the high order interaction table
of seed type, moisture and packaging combination,
storage temperature and time was disaggregated into
winged and de-winged treatments, each of which was
then sub-divided into the three temperature regimes
under which the seed was maintained during storage.
Mean germination percentages after one month
storage of winged and de-winged seed were 51% and
93%, respectively, in Experiment 1, 27% and 96% for
Experiment 2 in the same order. This shows a 42%
and 69% increase in initial germination as a result of
wing removal in Experiment 1 and Experiment 2,
respectively.
The trends of change in seed longevity for winged
and de-winged seeds separately are presented in six
graphs embedded within two figures representing
treatments grouped by temperature regime within each
Effects of Temperature, Relative Humidity and Moisture Content on Seed Longevity
of Shrubby Russian Thistle (Salsola vermiculata L.)
626
experiment (Fig. 1 for winged and Fig. 2 for
de-winged seeds). Regardless of the moisture content
and packaging, both winged and de-winged seeds
maintained their initial germination levels when stored
at -21 °C and 4 °C throughout both experiments.
However, at 24 °C, the trends showed different rates
of decline depending on seed moisture content.
The mean germination proportions (GP) over the
entire storage period are presented in Fig. 3
(Experiment 1) and Fig. 4 (Experiment 2). Within
each temperature regime, the GP of winged seeds
consistently and significantly (P < 0.01) declined
between the vacuum and non-vacuum packaged seeds
and from low to high MC treatments. For de-winged
seeds, GP was the highest (P < 0.01) under the higher
compared to the lower MC seeds. When stored at
24 °C, the mean GP for non-vacuum packaged winged
was higher compared to winged seeds in vacuum
packages in Experiment 1.
3.2 Parameter Estimates for Longevity Curves
3.2.1 Winged and De-winged Seed
Probit analyses results showed that for winged
seeds, the slopes were positive at -21 °C and 4 °C. The
intercepts expressed in probit units were either
negative or very low when positive. The decline in the
P50s expressed inseed germinability gradient and
slopes of regression with storage period did not follow
the opposite direction of change in SMC. For
de-winged seeds, all regression line intercepts were
positive and higher compared with the winged seeds.
The decline in longevity was irregular along the seed
moisture gradient under -21 °C and 4 °C for both
winged and de-winged seeds and the seed longevity
was the highest under low SMC with vacuum
packaging followed by seed with high MC with
vacuum then the high SMC with no vacuum
packaging (Tables 1 and 2). Moreover, for de-winged
seeds stored at 4 °C and 24 °C, the regression
coefficients, intercepts and correlation coefficients
were all significant (P < 0.01).
Results of the probit analyses also showed that in
the control (non-vacuum packaging with MC of 10.7%
in Exp. 1 or 9.6% in Exp. 2), lowering the storage
temperature from 24 °C to 4 °C resulted in an increase
of P50 from 156 to 1,651 days in Exp. 1 and from 161
Storage period in days
Fig. 1 Proportion of germination of winged seeds of Salsola vermiculata L. with a moisture content (MC) of 7% and vacuum
packaging (VP) (solid line with diamonds), with MC = 10.7% and VP (dashed line with circles), and with MC = 10.7% and no
VP (dotted line with triangles) stored at -21, 4 and 24 °C for 1,140 and 720 days in 2002/04 (left panels) and 2003/05 (right
panels), respectively. Note that the scales are not equidistant at the higher end of the x-axes.
0
0.2
0.4
0.6
0.8
1‐21°C 2002
/
04
2003/05
‐21°C
0
0.2
0.4
0.6
0.8
14°C 4°C
0
0.2
0.4
0.6
0.8
124°C 24°C
Germination proportion
Effects of Temperature, Relative Humidity and Moisture Content on Seed Longevity
of Shrubby Russian Thistle (Salsola vermiculata L.)
627
Storage period in days
Fig. 2 Proportion of germination of de-winged seeds of Salsola vermiculata L. with a moisture content (MC) of 7% and
vacuum packaging (VP) (solid line with diamonds), with MC = 10.7% and VP (dashed line with circles), and with MC = 10.7%
and no VP (dotted line with triangles) stored at -21, 4 and 24 °C for 1,140 and 720 days in 2002/04 (left panels) and 2003/05
(right panels), respectively. Note that the scales are not equidistant at the higher end of the x-axes.
Fig. 3 Mean predicted germination proportions with standard error bars for winged and de-winged seed of Salsola
vermiculata L. with low (7 MC) and high (10.7 MC) moisture content and vacuum and no-vacuum packaging stored at three
temperature regimes for 1,140 days (Experiment 1).
to 929 days in Exp. 2. The increases in P50 in the
Experiments 1 and 2, respectively, were 10.6 and 5.6
fold. When stored in vacuum packaging at 24 °C,
lowering the MC from 10.7% to 7% in Exp. 1 resulted
in an increase of P50 from 322 to 1,179 while MC
reduction from 9.6% to 6.5% in Exp. 2 increased the
P50 from 274 to 610. The increases in P50 in the
Experiments 1 and 2 were in the same order, and of
0
0.2
0.4
0.6
0.8
1
‐21°C ‐21
°
C
0
0.2
0.4
0.6
0.8
1
4°C 4°C
0
0.2
0.4
0.6
0.8
1
24°C 24°C
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
-21C 4C 24C -21C 4C 24C
Winged De-winged
Germination proportion
Seed type, moisture and packaging
Vacuum with 7MC Vacuum with 10.7MC No vacuum with 10.7MC
Germination proportion
Effects of Temperature, Relative Humidity and Moisture Content on Seed Longevity
of Shrubby Russian Thistle (Salsola vermiculata L.)
628
Fig. 4 Mean predicted germination proportions with standard error bars for winged and de-winged seed of Salsola
vermiculata L. with low (6.5 MC) and high (9.6 MC) moisture content and vacuum and no vacuum packaging stored at three
temperature regimes for 720 days (Experiment 2).
Table 1 Parameter estimates for winged seed longevity curves of two seed lots of Salsola vermiculata L. under different
storage conditions in 2002/04 and 2003/05.
Years T (°C) MC% Packaging Gradient (1/δ) P50 (days) Intercept (probit) Slope
1
-21
7.0 Vacuum 0.00061 1,665* 0.5 ± 0.1 0.0003 ± 0.0002ns
10.7 Vacuum 0.00244 410 ±7 -0.2 ± 0.05 0.001 ± 0.0001***
No vacuum 0.00131 766 ± 774 -0.2 ± 0.05 0.0003 ± 0.0001**
4
7.0 Vacuum -0.00108 -930 ± 1464 0.3 ± 0.05 0.0003 ± 0.0001**
10.7 Vacuum 0.00277 361 ± 62.2 -0.2 ± 0.05 0.0006 ± 0.0001***
No vacuum -0.00080 -1,246* -0.1 ± 0.05 -0.0001 ± 0.0001***
24
7.0 Vacuum 0.00083 1,203 ± 1,170 0.3 ± 0.05 -0.0003 ± 0.0001***
10.7 Vacuum 0.08621 11.9 ± 21.6 0.04 ± 0.07 -0.003 ± 0.0002***
No vacuum 0.02849 35.1 ± 7.9 0.4 ± 0.1 -0.01 ± 0.001***
2
-21
6.5 Vacuum 0.00137 730 ± 236 -0.3 ± 0.05 0.0004 ± 0.0001***
9.6 Vacuum 0.00068 1,481 ± 1,160 -0.5 ± 0.05 0.0003 ± 0.0001***
No vacuum 0.00053 1,877 ± 5,365 -0.5 ± 0.06 0.0003 ± 0.0001**
4
6.5 Vacuum 0.00089 1,120 ± 8,802 -0.3 ± 0.05 0.0003 ± 0.0001**
9.6 Vacuum 0.00025 4,031* -0.3 ± 0.05 0.0001 ± 0.0001ns
No vacuum 0.00061 1,640 ± 776.9 -0.6 ± 0.06 0.0004 ± 0.0001***
24
6.5 Vacuum 0.00045 2211 ±* -0.3 ± 0.05 0.0002 ± 0.0001ns
9.6 Vacuum -0.02915 -34.3 ± 22.1 -0.1 ± 0.06ns -0.003 ± 0.0002***
No vacuum -0.05244 -19.1 ± 14.9 -0.2 ± 0.1 -0.009 ± 0.001***
The gradient is the inverse of the longevity curve variance (δ) calculated here at P50 in days; ***significant at P > 0.01; Letters in front
of figures indicate significant differences between treatments; ns: non-significant; *sign for SE shows non converging iteration due to
deviation from the standard life curve; T = temperature; MC is moisture content.
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
-21C 4C 24C -21C 4C 24C
Winged De-winged
Germination proportion
Seed type, moisture and packaging
Vacuum with 6.5MC Vacuum with 9.6MC No vacuum with 9.6MC
Effects of Temperature, Relative Humidity and Moisture Content on Seed Longevity
of Shrubby Russian Thistle (Salsola vermiculata L.)
629
Table 2 Parameter estimates for de-winged seed longevity curves of two seed lots of Salsola vermiculata L. stored under
different storage conditions in 2002/04 and 2003/05.
Years T (°C) MC% Packaging Gradient
(1/δ) P50 (days) Intercept (probit) Slope
1
-21
7 Vacuum 0.00033 3,007 ± 1,065 1.5 ± 0.07 -0.0005 ± 0.0001***
10.7 Vacuum 0.00011 8,740* 1.6 ± 0.1 -0.0002 ± 0.0002ns
No vacuum -0.00007 18,621* 1.4 ± 0.07 -0.0001 ± 0.0002ns
4
7 Vacuum 0.00011 8747* 1.4 ± 0.07 -0.0002 ± 0.0002ns
10.7 Vacuum 0.00002 44,069* 1.7 ± 0.09 -0.00004 ± 0.0002ns
No vacuum 0.00058 1,738 ± 222.5 1.5 ± 0.06 -0.0009 ± 0.0001***
24
7 Vacuum 0.00096 1,040 ± 66.2 1.4 ± 0.06 -0.001 ± 0.0001***
10.7 Vacuum 0.00311 322 ± 5.7 1.9 ± 0.08 -0.006 ± 0.0002***
No vacuum 0.00650 154 ± 2.9 3.2 ± 0.2 -0.02 ± 0.001***
2
-21
6.5 Vacuum 0.00020 5,075* 1.7 ± 0.09 -0.0003 ± 0.0002ns
9.6 Vacuum 0.00014 7,190* 2.0 ± 0.1 -0.0003 ± 0.0003ns
No vacuum 0.0004 2,491 ± 2 1.9 ± 0.1 -0.0008 ± 0.0002***
4
6.5 Vacuum 0.00028 3,562 ± 3,419 1.7 ± 0.1 -0.0005 ± 0.0002***
9.6 Vacuum 0.00033 3,041 ± 1,616 1.9 ± 0.1 -0.0006 ± 0.0002***
No vacuum 0.00108 929 ± 43.9 2.0 ± 0.1 -0.002 ± 0.0002***
24
6.5 Vacuum 0.00164 610 ± 12.5 2.0 ± 0.08 -0.003 ± 0.0002***
9.6 Vacuum 0.00365 274 ± 4.9 2.0 ± 0.09 -0.007 ± 0.0003***
No vacuum 0.00620 161 ± 3.4 2.6 ± 0.2 -0.02 ± 0.001***
The gradient is the inverse of the longevity curve variance (δ) calculated here at P50 in days; ***significant at P > 0.01; Letters in front
of figures indicate significant differences between treatments; ns: non-significant; *sign for SE shows non converging iteration due to
deviation from the standard life curve; T = temperature; MC = moisture content.
the magnitude of 3.7 and 2.2 fold, respectively. The
increases in P50 resulting from drying and packaging
alone were from 156 to 1,179 in Exp. 1 and from 161 to
610 in Exp. 2. These results correspond to 7.6 and 3.8
fold increases in P50 (Fig. 5). When stored at high
temperatures and high MC, P50 was consistently higher
for vacuum compared to non-vacuum packaged seed
but this was only significant in Exp. 2. The P50 values
were significantly (P < 0.01) higher in Exp. 1 than in
Exp. 2. Moreover, all intercepts of the regression lines
were positive, and significantly different from zero.
In addition to the trends of change in seed longevity
shown in Figs. 1-4, the results of the probit analyses
performed on the individual sets of data representing
regression of germination proportions against days of
storage at 24 °C are presented in Fig. 5. Fig. 5
components showed significant decline in seed
viability to zero within a 6, 10 and 23 months period
of storage under 24 °C with high MC% with and
without vacuum packaging and low MC% with
vacuum packaging, respectively.
3.3 Post-hermetic Storage Longevity
In Experiment 2, the seeds transferred from vacuum
and non-vacuum packages stored at -21, 4 and 24 °C
to ambient conditions showed slopes, intercepts and
coefficients of determination for the regression lines
which were all significant at P < 0.01 (Table 3). Probit
analysis on germination proportions with days of
storage for seed transferred from hermetic to
non-hermetic storage conditions showed that
differences in P50 between treatments were significant
for winged but not for de-winged seed. For de-winged
seeds, differences in P50s were significantly different
between vacuum and non-vacuum and among the
temperature regimes. For non-vacuum packaging with
fixed stable MC, the lower the temperature the higher
the P50 became (Table 3).
Effects of Temperature, Relative Humidity and Moisture Content on Seed Longevity
of Shrubby Russian Thistle (Salsola vermiculata L.)
630
Fig. 5 1:Seed longevity curves for germination proportion with days of storage for Salsola vermiculata L. (a1&a2 = 24 °C,
without vacuum packaging (WVP) and 10.7% moisture content (MC) for Exp. 1 and 2; b1&2= 24 °C, VP and 10.7%MC for
Exp. 1&2; c1&2 = 24 °C, (VP) and 7% MC for Exp. 1&2.
The estimated number of days for a 50% drop in
seed germination (P50) was consistently higher for the
seed transferred from lower compared to those from
higher temperature storage.
10.7% MC
No vacuum
24 °C
P50:156 days
r = -0.956***
2002/04 2003/05
9.6% MC
No vacuum
24 °C
P50:161 days
r = -0.936***
10.7% MC
Vacuum
24 °C
P50:321 days
r = -0.914***
9.6% MC
Vacuum
24 °C
P50:274 days
r = -0.909***
7% MC
Vacuum
24 °C
P50:1179 days
r = -0.840***
6.5% MC
Vacuum
24 °C
P50:610 days
r = -0.925***
0 200 400 600 800 300 700 100 500
Da
y
s
0.0
0.4
0.8
0.2
0.6
1.0
Germination proportion
1100 0 1300 900 800 700 500
0.0
300
0.2
100
0.4
0.6
0.8
1000
1.0
400 1200 200 600
Da
y
s
Germination proportion
0.0
0.4
0.8
0.2
0.6
1.0
Germination proportion
Effects of Temperature, Relative Humidity and Moisture Content on Seed Longevity
of Shrubby Russian Thistle (Salsola vermiculata L.)
631
Table 3 Parameter estimates for winged and de-winged seed longevity curves of S. vermiculata L. transferred to ambient
conditions after 720 days of hermetic and non-hermetic storage.
Seed type T (°C) MC% Packaging Intercept (probit) Slope (degree) P50 (days)*
Winged
-21
6.5 Vacuum -0.1 ± 0.2ns -0.01 ± 0.002*** -5.6 ± 23.7a
9.6 Vacuum -0.3 ± 0.2ns -0.01 ± 0.002*** -47.7 ± 69.0a
No vacuum -0.7 ± 0.2*** -0.01 ± 0.003*** -88.8 ± 95.7a
4
6.5 Vacuum -0.2 ± 0.2ns -0.01 ± 0.002*** -19.5 ± 29.5a
9.6 Vacuum 0.1 ± 0.2ns -0.01 ± 0.002*** 4.2 ± 13.7a
No vacuum -0.8 ± 0.2*** -0.01 ± 0.003*** -96.2 ± 103.1a
24 6.5 Vacuum -0.1 ± 0.2ns -0.01 ± 0.003*** -6.5 ± 16.5a
De-winged
-21
6.5 Vacuum 1.5 ± 0.2*** -0.02 ± 0.002*** 71.0 ± 3.4a
9.6 Vacuum 1.8 ± 0.3*** -0.03 ± 0.002*** 71.0 ± 2.9a
No vacuum 0.8 ± 0.2*** -0.02 ± 0.002*** 48.0 ± 5.4c
4
6.5 Vacuum 1.3 ± 0.2*** -0.02 ± 0.002**** 65.0 ± 3.7ab
9.6 Vacuum 1.4 ± 0.2*** -0.02 ± 0.002***** 59.0 ± 3.2
b
No vacuum 0.9 ± 0.2*** -0.02 ± 0.002*** 36.0 ± 4.5
d
24 6.5 Vacuum 0.1 ± 0.2*** -0.02 ± 0.003*** 6.0 ± 10.2e
***significant at P ≤ 0.001; ns: non-significant; *Different letters within seed type indicate significant difference in P50; T =
temperature; MC is moisture content.
4. Discussion
4.1 Longevity Trends
Short seed longevity under high temperature and
high SMC and extended seed longevity under low
temperature and low seed SMC found from the two
experiments on both winged and de-winged seeds is
consistent with the well-established and
widely-reported storage behavior for S. vermiculata
[13, 16, 21] and for other orthodox seed species [18].
Nonetheless, the previous studies on Salsola seed
longevity focused on the effect of temperature,
whereas, the present study introduced the control of
moisture content as a more cost-effective storage
approach for rangeland management and
rehabilitation.
4.2 Parameter Estimates for Longevity Curves
4.2.1 Winged and De-winged Seeds
The low and negative intercepts and P50 values
with high standard errors recorded in the winged seeds
are attributable to the confounded effects of
simultaneous seed dormancy breaking and natural
deterioration in viability due to aging. Loss of
dormancy continuously generates new germinable
seeds while natural deterioration in seed viability
takes place within the seed populations harvested from
the wild with high inherent variability clearly
expressed in flower and fruit color. The cyclic trend of
germination disappeared when fruit bract was
removed as shown in the de-winged seed longevity
trend graphs.
The large differences in germination percentages
between winged and de-winged seeds recorded after
one month of storage indicate a high level of
dormancy in the winged seeds. The presence of
germination inhibitors in S. vermiculata wings and
their effects on germination have been reported [16].
It seems that the rates of dormancy breaking and
natural deterioration among the dormant and
non-dormant seeds seem to be canceling out each
other and maintaining the overall germination at its
initial low level. Nonetheless, the negative intercepts
of regressions of the germination proportions against
days of storage indicate that the rate at which
dormancy breaking is taking place seems to be
slightly greater than the rate of deterioration in seed
Effects of Temperature, Relative Humidity and Moisture Content on Seed Longevity
of Shrubby Russian Thistle (Salsola vermiculata L.)
632
viability. The negative intercepts suggest an increase
in seed longevity instead of decline. This increase
implies that the regression line will intersect with the
x-axis before its zero level. In other words, the model
is estimating the number of days required for the seed
viability to increase instead of dropping to 50%.
Dormancy breaking and deterioration processes have
been reported to be controlled by temperature ranging
from -10 °C to 70 °C [24]. The experiments were
conducted within this range of temperatures.
The higher seed longevity of vacuum packaged
seeds with higher MC compared to lower and similar
seed moisture content with and without vacuum
packing is probably due to desiccation damage in the
low MC treatment and higher respiration rate in the
non-vacuum packaged seeds (Figs. 3 and 4). Reduced
seed longevity under lower moisture content found in
this study is not in line with the reported negative
logarithmic relationship of seed longevity with
moisture content [24]. Upper and lower limits for this
negative relationship have been reported [18],
although these limits vary among the orthodox species.
Beyond the limits, further reduction in seed moisture
does not increase or decrease seed longevity.
Nevertheless, the moisture and temperature treatments
under which change in seed longevity did not match
the expected negative relationship falls within the
operational boundaries of the negative relationship
which is -20 °C to 75 °C [18]. Desiccation below the
optimum moisture content greatly increases seed
storage deterioration [24]. Low germination in drier
compared to more moist seed has been reported in
sorghum [25].
The reduced longevity under lower moisture
content could also be attributed to interdependence of
temperature and moisture effects on longevity or to
desiccation damage. It has been found [26] that
optimum storage moisture content can not be
considered independently of temperature. It seems that
the temperature at which longevity was reduced was
not optimum for the level of seed moisture content to
which the seed was dried. In a study on
desiccation-induced damage in orthodox seeds,
Leprince et al. [27] concluded that the expression of
desiccation damage depends on the drying history and
that factors that limit metabolism also reduce the
incidence of desiccation injury. The improved seed
longevity in the seed with higher moisture content
suggests that the short storage life in S. vermiculata
seed is due to its sensitivity to desiccation. Seed
moisture content above 6% is not considered too low
for the desert environments in which S. vermiculata is
widespread and endemic. In addition, the P50 value of
1,651 days predicted for non-vacuum packaged seeds
with 10.7% MC stored at 4 °C compared with a
predicted P50 value of 1,179 days for vacuum
packaged seed with 7% MC stored at 24 °C indicates
that S. vermiculata seed longevity is more dependent
on temperature than moisture. Nevertheless, the actual
and the theoretical longevity results from the present
study suggest that S. vermiculata can be truly
classified as an orthodox species. Seeds of some
tropical crops show an intermediate category of seed
storage behavior [18] and it is not yet clear how many
species belong to this category. The findings from the
present study suggest that S. vermiculata could be a
candidate for that intermediate category.
The significantly greater values for P50 in Exp. 1
compared to those in Exp. 2 are probably due to the
fact that Exp. 1 continued for longer than Exp. 2. The
7.6 and 3.8 fold increase in P50 achieved in Exps. 1
and 2, respectively through drying and packaging has
important practical and cost implications for rangeland
rehabilitation. Increases in P50 from 156 to 1,179 in
Exp. 1 and from 161 to 610 in Exp. 2 are equivalent to
an increase in storage life from less than six months to
3 years in Exp. 1 and to about 2 years in Exp. 2.
4.3 Post-hermetic Storage Longevity
For the seeds which were transferred from vacuum
and non-vacuum packages to ambient conditions, the
slopes, intercepts and correlation coefficients of
Effects of Temperature, Relative Humidity and Moisture Content on Seed Longevity
of Shrubby Russian Thistle (Salsola vermiculata L.)
633
determination for the regression lines were all
significant at P < 0.01 (Table 3). Probit analysis on
germination proportion with days of storage for the
seed without wings transferred from the vacuum and
non-vacuum packaged seeds showed that differences
in P50 were not significant between high and low
moisture contents within the three storage temperature
regimes. For non-vacuum packaging with fixed MC,
the lower the storage temperature, the higher the P50
became (Table 3).
The estimated number of days for 50% decline in
seed germination (P50) was consistently higher (P <
0.01) under high MC and low temperatures compared
to low MC and high temperatures. For seed stored at
high temperatures and high MC, the P50 was higher
for vacuum-packaged seed, but only significantly so
in Exp. 2. For the de-winged seeds, the intercepts of
the regression lines were all positive and significantly
(P < 0.001) different from zero, indicating a consistent
decline in seed longevity with time.
The slower decline in germination of seed
transferred from the lower temperature and moisture
content treatment is probably due to the fact that
deterioration was minimal in hermetic storage
conditions. Seeds continuously deteriorate at a rate
that is mainly dependent on moisture content and
temperature and, unless germinated, will ultimately
die [28]. The most important reasons for Ellis and
Roberts to develop an improved equation for
predicting longevity was to reflect variations in lot
history among crop species and genotypes [29].
5. Conclusions
The improved seed storage in the higher moisture
content treatments found in the present study suggests
that the short storage life in S. vermiculata observed in
the field is attributable to sensitivity to desiccation.
Special attention should be given to desiccation
control in medium and long term storing seed of this
species.
The negative intercepts of the regression lines of the
winged seeds indicate that seed dormancy resulting
from germination inhibitors in the wings increases
seed longevity.
The study clearly showed that drying and vacuum
packaging alone resulted in a substantial increase in
seed longevity. This finding is a significant step
towards more cost-effective and environmentally
friendly rangeland rehabilitation.
The reduced storage life of seed transferred from
vacuum packaging under ambient conditions
treatments to non-vacuum packaging needs to be
taken into consideration. Such seed should be sown
late when the probability of rainfall is high or used in
the rangeland nurseries under irrigation.
Acknowledgments
This research was carried out with full support from
the International Center for Agriculture Research in
the Dry Areas (ICARDA); the Wageningen University
of the Netherlands and the United Nations’ Food and
Agriculture Organization (FAO).
References
[1] D. Gallacher, J. Hill, Effects of camel grazing on the
ecology of small perennial plants in the Dubai (UAE)
inland desert, Journal of Arid Environments 66 (4) (2006)
738-750.
[2] G. Gintzburger, H.N. Le-Houerou, K.N. Toderich, The
steppes of Middle Asia: Post-1991 agricultural and
rangeland adjustment, Arid Land Research and
Management 19 (3) (2005) 215-239.
[3] H.N. Le Houerou, Impact of man and his animals on
Mediterranean vegetation, ecosystems of the world,
Mediterranean-Type Ecosystems, In Elsevier Scientific
Co., Amsterdam 7 (1981) 479-517.
[4] Z.K. Tessema, W.F.D. Boer, R.M.T. Baars, H.H.T. Prins,
Influence of grazing on soil seed banks determines the
restoration potential of aboveground vegetation in a
semi‐arid savanna of Ethiopia, Biotropica 44 (2) (2012)
211-219.
[5] A.E. Sidahmed, The rangelands of the arid/semi-arid
areas: Challenges and hopes for the 2000s, International
Conference on Desert Development in the Arab Gulf
Countries, Kuwait 1996, pp. 1-16.
[6] H.N. Le Houerou, Climate change, drought and
desertification, Journal of Arid Environments 34 (1996)
Effects of Temperature, Relative Humidity and Moisture Content on Seed Longevity
of Shrubby Russian Thistle (Salsola vermiculata L.)
634
133-185.
[7] M. Louhaichi, ICARDA’s rangeland ecology and
management research strategy for nontropicaldry areas,
Rangelands 33 (4) (2011) 64-70.
[8] A. Nefzaoui, H.B. Salem, M. El-Mourid, Innovations in
small ruminants feeding systems in arid Mediterranean
areas, in: R. Bouche, A. Derkimba, F. Casablanca (Eds.)
New Trends for Innovation in the Mediterranean Animal
Production, 2011, pp. 99-116.
[9] J. Rae, G. Arab, T. Nordblom, K. Jani, G. Gintzburger,
Tribes, state and technology adoption in arid land
management, Syria, CAPRI Working Paper 15, 2001, p. 34.
[10] J.M. Baskin, Seed propagation of Mediterranean trees
and shrubs, in: B. Piotto, A.D. Noi (Eds.), Seed Science
Research, Rome, Italy, Agency for the Protection of the
Environment and for Technical Services (APAT), Vol. 14,
2003, p. 108.
[11] R.V. Holdtz, Saltbush propagation: Seed handling from
picking to planting, in: G. Gintzberger, M. Bounjmate, A.
Nefzaoui (Eds.), Fodder Shrub Development in Arid and
Semi-arid Zones Workshop Proceedings, Vol. 1, 1996, p.
679.
[12] N. Mourad, A study on the syrian steppe and forage
shrubs, in: G. Gintzberger, M. Bounjmate, A. Nefzaoui
(Eds.), Fodder Shrub Development in Arid and Semi-arid
Zones, Vol. 1, 2000, pp. 109-121.
[13] I.R. Guma, M.A. Padrón-Mederos, A. Santos-Guerra, J.A.
Reyes-Betancort, Effect of temperature and salinity on
germination of Salsola vermiculata L. (Chenopodiaceae)
from Canary Islands, Journal of Arid Environments 74 (6)
(2010) 708-711.
[14] Z.B. Wen, M.L. Zhang, G.L. Zhu, S.C. Sanderson,
Phylogeny of Salsoleaes L. (Chenopodiaceae) based on
DNA sequence data from ITS, psbB–psbH, and rbcL,
with emphasis on taxa of northwestern China, Plant
Systematics and Evolution 288 (1-2) (2010) 25-42.
[15] A.E. Osman, F. Bahhady, N. Hassan, F. Ghassali, T. Al
Ibrahim, Livestock production and economic implications
from augmenting degraded rangeland with Atriplex
halimus and Salsola vermiculata in Northwest Syria,
Journal of Arid Environments 65 (3) (2006) 474-490.
[16] A.E. Osman, F. Ghassali, Effects of storage conditions
and presence of fruiting bracts on the germination of
Atriplex halimus and Salsola vermiculata, Experimental
Agriculture 33 (2) (1997) 149-155.
[17] K. Tadros, Fodder shrubs in Jordan, in: G. Gintzburger,
M. Bounejmate, A. Nefzaoui (Eds.), Fodder Shrub
Development in Arid and Semi-arid Zones, 2000, p. 289.
[18] R.H. Ellis, The longevity of seeds, Horticultural Science
26 (9) (1991) 1119-1125.
[19] M.W. King, E.H. Roberts, The storage of recalcitrant
seeds-achievements and possible approaches, Rome, Italy,
IBPGR, 1979.
[20] J.F. Harrington, Seed storage and longevity, Seed
Biology, in: T.T. Kozlowski (Ed.), Academic, New York,
Vol. 3, 1972, pp. 145-245.
[21] S. Zaman, S. Padmesh, H. Tawfiq, Seed germination and
viability of Salsola imbricate Forssk, International
Journal of Biodiversity and Conservation 2 (12) (2010)
388-394.
[22] ISTA, International rules for seed testing, International
Seed Testing Association, Zurich, Switzerland, 2002.
[23] R.W. Payne, S.A. Harding, D.A. Murray, D.M. Soutar,
D.B. Baird, A.I. Glaser, The Guide to GenStat Release 14,
Part 2: Statistics, VSN International, Hemel Hempstead,
UK, 2011.
[24] R.H. Ellis, T.D. Hong, Temperature sensitivity of the
low-moisture-content limit to negative seed
longevity-moisture content relationships in hermetic
storage, Annals of Botany 97 (5) (2006) 785-791.
[25] J.C. Phillips, V.E. Youngman, Effect of initial seed
moisture content on emergence and yield of grain
sorghum, Crop Science 11 (3) (1971) 354-357.
[26] C.W. Vertucci, E. Roos, J. Crane, Theoretical basis of
protocols for seed storage III, optimum moisture contents
for pea seeds stored at different temperatures, Annals of
Botany 74 (1994) 531-540.
[27] O. Leprince, C.W. Vertucci, G.A.F. Hendry, N.M.
Atherton, The expression of desiccation-induced damage
in orthodox seeds is a function of oxygen and temperature,
Physiologia Plantarum 94 (2) (1995) 233-240.
[28] E.H. Roberts, Temperature and seed germination, in: S.P.
Long, F.I. Woodward (Eds.), Symposium on Plant and
Temperature, Colchester, England: Society of
Experimental Biology, British Ecological Society, Vol.
42, 1988, pp. 109-132.
[29] R.H. Ellis, E.H. Roberts, Improved equations for the
prediction of seed longevity, Annals of Botany 45 (1980)
13-30.