2.6 Factors affecting seed longevity
Seed longevity refers to the period seeds will remain viable in store and determined by the
genetic and physiological storage potential of seeds. Any deteriorating events or damage
during storage affects storability of a given seed. The most important trait for seed
conservation is the time that seeds will remain alive under a given storage condition (seed
longevity) (Probert et al., 2007). Seeds should be stored under optimum conditions in order to
maximize their longevity. Longevities
in the desiccated state vary from a few days in some
and spore types to many decades in some seeds and moss spores and
tissues (Hoakstra, 2005).
Seed longevity is mainly influenced by the environmental conditions such as storage
temperature and moisture content (Spara et al., 2003) within broad limits, reducing moisture
content and temperature
during storage increases the longevity of desiccation-tolerant
(orthodox) seeds (Girma Balcha et al., 2000). To survive long storage in the dehydration state,
seed have to be able to withstand desiccation to low water contents (Scande et al., 2000).
However, there is low moisture content limit below which further desiccation had no further
effect on longevity. This is because of the type of water present in seeds. Thus, removal of
weakly bound water improves longevity
but, once this is all removed, the withdrawal of
water by further desiccation
has no additional effect because strongly bound
water has negligible
chemical potential (Ellis and Hong, 2006).
Seedbanks have a long term conservation objective. Because they need to maintain high levels
of seed viability in their collections, effective
tools to predict longevity under given storage
to minimize damage to the germplasm are required. Buitikin et al. (2000)
suggested that the limiting
factor of the longevity of dry germplasm is the availability of water
for chemical reactions, and that therefore the water potential
would be a better way of
predicting optimal storage conditions.
Longevity of seeds varies from species to species even if they are provided with identical
storage conditions. Species and sometimes genera show an inherited storage behavior which
may be either orthodox or recalcitrant. Each species is likely to respond identically to a given
set of storage condition (Phartyal et al., 2002).
Immature seeds generally have shorter storability than fully matured seeds unless early
collected seeds attain full maturity including normal storability. Neya et al
investigated that the preservation of viability of stored Neem seeds improved with maturity.
Thus development of seeds and possible ripening of fruits can determine storability of seeds.
Neya et al. (2003) concluded that the effects of storage moisture content and temperature on
Neem seed longevity are highly dependent on the maturity of seeds, with the most mature seed
exhibiting orthodox-type responses. According to Probert et al. (2007), slow or delayed drying
of immature seeds has shown to increase both desiccation tolerance and storability in several
wild plant species.
Development stage is especially evident and important in recalcitrant seeds. Because the
process of maturation and germination are more or less continuous, deteriorations proceed
rapidly if germination does not occur. Dehydration is an essential part of the maturation
process. As water lost from the seed cell membranes, physiological processes such as
respiration diminish to very low levels. Simple compounds are changed to starches, fats and
proteins. These complex compounds can remain stable over many years for long periods
Seed deterioration may start already in the field and be influenced by handling from collection
and transport through processing. Generally, seeds with high initial viability have a higher
longevity. Loss of viability is initially slow, followed by a period of rapid decline. The higher
the viability when the seed lot enters into storage the longer the seed will keep viable under a
given storage environment (Schmidt, 2000). As orthodox seeds, recalcitrant seeds of poor
quality will have reduced storage life spans (Berjak and Pammenter, 2003).
2.7 Seed ageing
Ageing is a process of deteriorating events that take place within the seed which lead to the
death of the seed. Orthodox seeds are characterized by their ability to tolerate
to retain their viability for a long time in
the dry state. However, these seeds age during storage
lose their ability to germinate (Murthy et al., 2003).
Causes of physiological ageing may be grouped into extrinsic factors which are external factors
influencing viability and intrinsic factors, where the ageing is a result of events within the seed
only (Hoakstra, 2005). In relation to the intrinsic longevity properties of anhydrobiotes,
may be several ways by which aging can be held up. One
such way is curtailing the production
of free radicals. This
may be achieved by down regulation of the metabolism before
desiccation, because the mitochondrial system is a major source
of free radicals, even at low
water contents (Leprince et al.
2000). Another way to control free radical production may be
slowing down chemical reaction rates and molecular diffusion
in the cytoplasm by a
glassy state (Hoakstra, 2005).
Under the long-term storage conditions, seeds are likely to
be in the glassy state because of the
cool storage environment
and low seed water content. The high intracellular viscosity of the
could retard molecular mobility and thus slow down seed deterioration
storage. With increasing temperature or seed water content,
the solid-like glassy state may
soften into the rubbery state
or even ‘melt’ into the liquid. The low viscosity and enhanced
molecular mobility in the rubbery
or liquid state would permit certain deteriorative reactions
proceed rapidly, which are otherwise retarded in the glassy
state. This implies that the major
primary process that initiates seed ageing
could be different under different storage conditions
(Murthy et al
., 2002; Pritchard and Dicke, 2003).
Temperature and moisture content are the two major factors determining the rate of ageing.
Biochemical deterioration during seed ageing has been studied
mostly under accelerated ageing
conditions using high temperature
and high seed water content (McDonald, 1999). Under such
conditions, seeds typically lose their viability within a few
days or weeks. Loss of
viability is also associated with an increase in the accumulation of chromosomal aberrations.
Sivritepe and Dourado (1998) reported that both in improved and landrace pea seeds, loss of
viability and the induction of chromosomal aberrations occurred more rapidly in seeds of high
moisture content than low moisture content.
In desiccation sensitive recalcitrant seeds moisture content is always high, hence seeds are
currently metabolically active. Metabolically active seeds continue to accumulate dry weight.
If germination does not occur, deterioration proceeds rapidly. Molecular mobility is
increasingly considered a key factor influencing storage stability of bimolecular substances
(Buitink et al., 2000).
Metabolism is strongly related to temperature. At low temperature and moisture content
biochemical and cytological deterioration can be slowed down. At low moisture contents,
enzymatic reactions are expected to
play little role in seed ageing. Within broad limits,
reducing moisture content and temperature
during storage increases the longevity of
(orthodox) seeds (Schmidt, 2000; Murthy et al., 2003).
2.8 Description of Syzygium guineense
(Willd.) DC. ssp. Afromontanum F. White is a member of the family
Myrtaceae. The species is a large forest tree up to 35 m high. Young branches are cylindrical or
triangular (Friis, 1995). Flowers are white in dense heads and the sweet smell attracts many
insects and frits are edible. The species grows at altitudes of 1500- 2600m, found in montane
forest and also secondary forest growth occasionally seen as an isolated tree left in farmlands
(Fichl and Admasu Adi, 1994).
According to Azene Bekele (1993), the species is widely distributed in Africa, prefers moist
soils with high water table and grows well in moist and wet kolla and woyena dega
agroclimatic zones.Geographically, this species is distributed in Gahana, Ethiopia, Somalia,
Afghanistan, Zaire, Rwanda, Burundi, Sudan, Uganda, Kenya, Tanzania, Angola, Zambia,
Malawi and Zimbabwe (Demel Teketay and Assefa Tigineh, 1991; Fichtl and Admasu Adi,
is one of multipurpose tree species found in different forest areas of Ethiopia.
Zerihun Woldu (1999) reported the presence of S. guineense in dry evergreen montane forests,
in moist evergreen forests and also in the forests in transition between dry evergreen and moist
evergreen montane forests, while Kumelachew Yeshitela and Simon Shibru (2004) reported
the distribution of S. guineense in eight moist montane forests of southwest Ethiopia.
is known for its uses like hardwood, timber, construction material, shade tree and
the like in different parts of Ethiopia. In a study of coffee shade tree in Harargie, Demel
Teketay and Assefa Tegineh (1991) pointed out that S. guineense used as shade, house
building, various utensils, and fuel wood. It is also mentioned that the tree is important honey
bee forage. The plant is an edible fruit bearing species which adds more importance for it as
wild edible plants are important for family diet and household food security (Azene Bekele,
1993; Kebu Balemie and Fassil Kebebew, 2006).
As most authors reported, S. guineense is a plant with several uses in most areas of Africa. In a
study undertaken on desiccation sensitivity of economically important trees in Kenya, Omondi
(2004) reported that the species has several uses such as timber, charcoal, tool handles,
construction and food /fruit. According to Maunda et al. (1999), S. guineense is a tree with
sweet edible fruits which is made into drinks in some areas of Kenya.
3. Materials and methods
This study was based on both field study and laboratory experiments. While the field study
component included reconnaissance survey for identification of the tree with seeds, seed
collection and ethnobotanical data collection; the work in the laboratory included moisture
content determination and viability test of the seeds of S. guineense.
3.1 Study site and species identification
3.1.1 Study site
Seed collection was carried out in Lephis, which is found in Reji site in Bonbaso Regi peasant
association of Arsi Negelie Woreda, West Arsi Zone of Oromia Region. The sampling site was
geographically located at 09° 00'N, 037° 07.5'E about 25 km South of Addis Ababa. The
collection site was a school compound and the site collection primarily based on the
availability of mature seeds by time of the study.
3.1.2 Species identification
Herbarium specimens of S. guineense were collected, pressed and identified in the National
Herbarium of the Addis Ababa University Ethiopia to identify the subspecies. Species
nomenclature follows flora of Ethiopia and Eritrea volume 2 part 2 (Friis, 1995).
3.2 Field and laboratory methods
3.2.1 Seed collection and processing
Ripe fruits of S. guineense (Wild.) DC. spp. afromontanum. F. White at the point of natural
dispersal was collected from four neighboring trees at the same stage of maturity on 8 June
2007. The fruits were harvested by climbing the trees and cutting down large inflorescences.
The mature fruits were collected before they had fallen to the ground to avoid the risk of
contamination and to collect fruits with healthy seeds.
During transport the fruits were kept in small quantity (about 10kg) of small bags to be well
ventilated to avoid seed suffocation and decay. The fruits were transported to the Institute of
Biodiversity Conservation, Addis Ababa, on a vehicle in an ambient condition within two days
of collection. Two sub-samples were sealed (to prevent moisture loss or absorption) by plastic
bags one for initial germination and the other for initial moisture content test.
After arrival at IBC, the fleshy fruits were soaked in water for three hours to facilitate softening
of the pulp (Fig. 1). The fleshy pulp was removed by rubbing between hands and continuous
washing with water. The bulk sample was made to dry under shade of open air for monitoring
of regular moisture content. The initial moisture content of the sub-sample was determined
both with and without pulp by standard oven dry method. The initial germination and moisture
content tests were conducted by depulping the fruits by rubbing between hands in the
Figure 1. Processing of S. guineense seeds at IBC
3.2.2 Seed desiccation and moisture content determination
The cleaned seeds were slowly dried down under ambient conditions to five levels of target
moisture contents for germination test. During drying, the seeds were allowed to be in a single
layer to make the drying condition uniform for both seed groups.
For moisture content determination 5g seeds were used by grinding into pieces and three
replicates were used for each measurement. Each replicates were weighed before and after
drying for 17 hrs in an oven at 103ºc following the method recommended by Hanson (1985)
and Rao et al
. (2006). The moisture content was determined by taking the average for each
replicate and was conducted for each subsequent germination test.
The germination tests were done at six moisture content levels: 51, 44, 37, 32, 27, and 24%.
Coming down in the moisture level, the range in the first two gradients was 7 each while the
range in the next two gradients was 5 each. The range in lower gradient was only 3%. The
reason to increase the frequency of germination test was that as the seeds dry to a lower level,
seed death increases. This was, therefore, to determine the points in the gradient where severe
In this study sampling for germination and moisture content determination were applied by
randomly taking out eight samples from different parts of the seed lot which are called primary
samples and then mixing them into a large sample which is called a composite sample
following ISTA (2007). The composite sample was then reduced to a working sample by
dividing into the required quantity. The seed population was thoroughly mixed before any
division to increase the homogeneity of the sample.
3.4 Germination experiments
Germination tests were employed under ambient temperature (15 to 21 ºC) in a germinator
room. Germination of seeds was conducted on plastic boxes of 6x18x26 cm with 13 randomly
equidistant holes at the bottom. The holes are used for air circulation on sand medium
moistened with 20ml of water for 1kg of sand. The sand was sterilized and the plastic boxes
were half filled. The germination was in sand in which the seeds covered with the sand as
recommended by Rao et al. (2006).
There were 25 seeds per plastic box for each germination test employed at the given moisture
level with four replicates for each test gives a total of 100 seeds. The samples were watered
during the whole germination period to maintain the moisture of the sand.
Each plastic box was put in a polyethylene bag (38cmx45cm) with the open end folded and
clipped at one point to keep the sand moist and were randomly placed in the germination
room. The plastic boxes were labeled with the name of the plant, sowing date and moisture
content of sowed seeds (Fig. 2).
Figure 2. An illustration of the set up used for the germination experiment
Germination assessments were made on weekly basis. Germination was based on emerging
of seedlings with visual observation which have the potential to be grown to a healthy plant.
Germinated seeds were recorded and removed (seedlings evaluated counted and removed)
from the trays in each count.
3.5 Ethnobotanical data collection
Ethnobotanical data were collected through interaction with community members in Regi area
following Martin (1995). Data collection was conducted in two phases. In the first phase,
some preliminary information was collected during seed collection on June 2007. The local
name and flowering and fruiting time of the plant was recorded when the informants reached to
consensus. In the second phase, open ended and semi-structured interviews were employed on
September 2007 using pre-formulated questions (Appendix 2). The informants were 135
members of the local community with 60 female participants.
3.6 Data analysis
Germination results under different moisture content were analyzed using SPSS software
program (13.0) (ASF, 2003). One-way analysis of variance (ANOVA) was used to test the
significant effect of moisture content on the germination (Table 2). The ethnobotanical data
were analyzed using description and percentages. The data were interred into Microsoft Office
Excel data sheet and used to analyze results of germination and responses of informants.
4. Results and discussion
4.1 Impacts of desiccation on the germination of seeds of S.
The germination tests were conducted at five levels of moisture content following drying of
seeds at room temperature. S. guineense
initially germinated with 99% germination percentage.
Seeds showed no dormancy; they germinated easily at harvest, the mean moisture content
(fresh weight basis) were 51% without pulp and 68% with pulp. The germination percentage
went on decreasing in-line with lowering moisture content and further drying reduced the
germination level to zero at 24% moisture content (Appendix 1). The germination percentage
declined significantly with reduction of moisture content (p< 0.01, Table 2).
It is known that recalcitrant seeds are sensitive to desiccation. Pritchard et al. (2004) pointed
out that desiccation sensitive seeds are killed by drying to water contents as high as 20-30%.
Berjak and Pammenter (1994) characterized recalcitrant seeds as desiccation sensitive through
a decreasing ability to withstand dehydration stress, to that of extreme sensitivity to even the
slightest loss of water.
This research finding shows that S. guineense can be categorized as extremely desiccation
sensitive. The seeds have shown abrupt loss of viability at high moisture content level (37%).
This finding is in a similar pattern to that of Normah et al. (1997) in which Garicinia
with germination percentage 10% at 15% moisture content reported as extremely
desiccation sensitive. In addition to loss of viability with desiccation moisture content at the
time of harvest also can indicate the desiccation sensitivity of Seeds. This character is also
observed in this species which was 51% moisture content at the time of harvest. Seeds of
recalcitrant species have high moisture content and lack dormancy at the time of dispersal
(Garwood and Linghton, 1990).
When fresh recalcitrant seeds begin to dry, viability is first slightly reduced as moisture is lost
but then begins to be reduced considerably at a certain critical moisture content some times
termed lowest safe moisture content (Hong and Ellis, 1996). The present study finding is in
agreement with this justification (Fig. 3). Therefore, the lowest safe moisture content of S.
can be between 37% and 32% moisture content. And no seeds germinated below
24% moisture content.
Although the critical moisture level for survival varies among species, there is a threshold of
minimal value of water content below which there is damage to germination of recalcitrant
seeds (Hong and Ellis, 1996; Barbedo and Bilia, 1998).
Desiccation sensitivity has been used by other authors for identifying seeds of species.
Pritchard et al
. (2004) reported results for 10 species on the basis of desiccation tolerances
alone. In their finding seeds were categorized as desiccation-tolerant if they survived drying to
5% moisture content and desiccation-intolerant if the seeds had lost most of their viability after
dehydration to 20% moisture content.
Most authors who studied seed behaviors of species within the family Myrtaceae reported
results showing desiccation sensitivity of recalcitrant seeds have high initial viability which
gradually decreased during desiccation and the seeds are sensitive to higher moisture contents
compared to seeds from trees of other families. Wang et al. (2004) reported that seeds of
which belongs to the Myrtaceae family are sensitive to desiccation
below 40% moisture content. This result compares well with this research finding as it is
shown that seeds of S. guineense are sensitive to desiccation below 37%. Baxter et al. (2004)
assessed the responses of Syzygium cuminii seeds to desiccation. They reported that S. cuminii
seeds are desiccation sensitive as none of the seeds germinated below 16% which was 24% for
the present study.