International Journal of Plant Production 5 (4), October 2011

ISSN: 1735-6814 (Print), 1735-8043 (Online)

www.ijpp.info
GUASNR

Wheat and barley seed system in Syria: farmers, varietal

perceptions, seed sources and seed management
Z. Bishawa,*, P.C. Struikb, A.J. G van Gastelc
a
Seed Section, ICARDA, P.O. Box 5466, Aleppo, Syria.
b
Center for Crop Systems Analysis, Wageningen University, Wageningen, The Netherlands.
c
Harspit 10, 8493KB, Terherne, The Netherlands.
*Corresponding author. E-mail: [email protected]

Received 23 January 2011; Accepted after revision 30 April 2011; Published online 15 September 2011

Abstract

A total of 206 wheat and 200 barley farmers were interviewed in northeastern Syria to understand
farmer perceptions and practice relating to modern varieties, seed sources and seed quality. Wheat
farmers had better awareness and grew modern varieties (87%), applied fertilizers (99.5%), herbicides
(93%), seed treatment (90%) or insecticides (41%). In contrast barley growers had low awareness
(36%) and use (0.5%) of modern varieties, herbicides (4%), insecticides (3%) and fertilizers (56%).
Grain yield, grain size, food quality and tolerance to lodging, drought and frost were the agronomic
characteristics farmers sought from new wheat varieties. For barley, grain yield, grain size, grain
color, feed quality, marketability and tolerance to diseases and drought were the key traits sought. The
informal sector-seed retained from the previous harvest or obtained from neighbors or local
traders/markets-was the main source of seed for both wheat and barley. Most farmers practiced on-
farm seed selection, cleaning, treatment, separate storage or quality assessment of seed that was
obtained locally. Farmers’ perceptions and preferences of new varieties/technologies and their seed
sources and seed management practices must be taken into account in any efforts to develop or to
strengthen seed sector development.

Keywords: Syria; Seed system; Seed source; Seed management.

Introduction

Wheat and barley are strategic crops, important for food and feed security. The two
crops are widely grown but in contrasting agro-ecologies. Wheat is a major food staple
while barley is mostly used for livestock feed (Bishaw, 2004).
Syria’s plant breeding program began with the establishment of the Department of
Agricultural and Scientific Research (DASR) in the 1960s. Several varieties were released
with recommended agronomic packages for wheat (Hamblin et al., 1995; Nachit et al., 1998;
Mazid et al., 2003) and barley (Haddad et al., 1997). The General Organization for Seed
Multiplication (GOSM) was established in 1976 as the sole public sector agency supplying
324 Z. Bishaw et al. / International Journal of Plant Production (2011) 5(4): 323-348

certified seed of a range of crops. During 1994-1999, average annual seed supply was
227,869 tons; of which 74% was agricultural crop seed. Wheat and barley accounted for 94%
and 3.4% of agricultural crop seeds distributed in the country. In the past 10 years, little has
changed in the structure and performance of the formal seed sectors for wheat and barley.
Adoption studies assume that new agricultural technology is appropriate and suits
farmers’ needs. In Syria and elsewhere, little information is available on farmers’
perception and preferences of wheat and barley varieties and crop management
technologies, farmers’ seed sources and seed management practices. This study of wheat
and barley seed systems compares the two crops at different stages in the technology
adoption process. The objectives of the study were to: (i) understand the functioning of the
wheat and barley seed systems, particularly the informal sector; (ii) understand farmers’
perceptions and preferences relating to modern varieties and associated technologies; (iii)
document farmers’ seed sources and knowledge of seed management to strengthen seed
delivery systems.

Materials and Methods

Selection of study areas and administration of survey questionnaires are described in

Bishaw et al. (2010). A multistage purposive random sampling procedure was followed
from higher to lower administrative levels, comprising selection of provinces, districts,
zones, villages and individual wheat or barley farmers. In each district zones 1 and 2 were
sampled for wheat, and zones 2, 3 and 4 for barley. A semi structured questionnaire was
used to gather data on farmers' perceptions, adoption and diffusion of modern varieties,
farmers' seed sources, selection and management. This was supplemented by secondary
data from different sources.

Study Areas

Aleppo, Raqqa and Hassakeh, three major wheat and barley production provinces in
northeastern Syria, were selected based on secondary data from the Central Bureau of
Statistics (Figure 1). These provinces together account for nearly 65% of wheat and 78% of
barley area in the country and provide contrasting situations in terms of agro-ecology,
farming systems, and proximity to agricultural research and service institutions.

Data Collection

Wheat: the survey was conducted in November and December in the 1998/99 crop season.
A total of 206 wheat farmers were surveyed from three provinces, six districts and 61
villages. These farmers were located in Aleppo (36%), Raqqa (15%) and Hasakeh (49%)
provinces. In total, 33% of farmers were in zone 1 and 67% in zone 2.
Barley: the survey was conducted in October and November in the 1997/98 crop season,
with 200 farmers (not covered by he wheat survey) in three provinces, eight districts and 59
villages. These farmers were located in Aleppo (47%), Raqqa (24%) and Hasakeh (30%)
provinces. In total, 47% of farmers were in zone 2, 38% in zone 3 and 26% in zone 4.
In both surveys, a minimum of two farmers were interviewed in each village. Data were
analyzed using SPSS software where descriptive statistics were used.
Z. Bishaw et al. / International Journal of Plant Production (2011) 5(4): 323-348 325

Figure 1. Wheat and barley seed system study areas (dark shade) in Syria.

Results and Discussion

Demographic and Socio-Economic Factors

The average age of the farmers surveyed was 46 (wheat) and 47 (barley); all had many years
of farming experience. About 86% of wheat and 94% of barley farmers were married, with an
average of 7 children and 1:1 female to male ratio in both groups. The findings for wheat differ
from Issa (1991), but the barley findings are similar to those reported by Mazid (1994).
Education levels were different in wheat and barley growers: 54% of wheat farmers
could read and write, nearly 20% had formal education (elementary or secondary school).
In contrast, 47% of barley farmers were illiterate and many lacked formal education,
probably because some barley growing areas (zone 4) were remote. These differences in
education might influence the adoption of new agricultural technologies.
Almost 88 and 95% of wheat and barley growers respectively, own land. The remaining
farmed on government land. About 54% and 5% of wheat farmers (n=206) owned tractors
and combine harvesters, compared to 34% and 9% of barley farmers (n=200). Mazid (1994)
also reported that 29% of barley growers owned tractors and 10% owned combine harvesters.
Agriculture is the main source of income for all wheat farmers and they were less likely
to engage in seasonal labor migration. In contrast, barley farmers in drier zones work as
migratory seasonal labor during the off-season.

Awareness of Wheat and Barley Production Technologies

Modern agricultural technologies in each environment have been recommended for

wheat production, but less so for barley. Wheat farmers (n=206) had better awareness of
and information about modern varieties (100%), agronomic practices (100%), fertilizers
326 Z. Bishaw et al. / International Journal of Plant Production (2011) 5(4): 323-348

(99%), herbicides (97%) and seed treatment (96%) compared to barley growers (Table 1).
In both crops, fellow farmers (relatives, neighbors and other farmers) were the main sources
of information, rather than formal extension services. This has been reported in other
studies as well, e.g. Tripp and Pal (1998) for hybrid pearl millet growers in India.

Table 1. Farmers’ awareness and sources of information for technology packages.

Awareness of Modern Grain

Crop Agronomy Fertilizers Herbicides Fungicides1 Insecticides
information variety storage
Wheat
Farmers 206 206 204 200 197 124 115
(n=206)
% 100 100 99 97 96 60 56
Barley
Farmers 72 187 159 62 - 62 141
(n=200)
% 36 94 79 31 - 31 71
Note: 1Seed treatment (chemical stores).

Among wheat growers, agricultural extension was the most important source of
information for herbicides (52%), seed treatment (34%) and varieties (22%). Among barley
growers, one-third had heard about new varieties and pesticides, the extension services
being the main source of information. Although 36% of farmers were aware of new barley
varieties none had tried growing them due to farmer preferences, lack of adapted varieties
or seed availability. Farmers in zone 4 were less aware of modern varieties, herbicides and
fertilizers; in this zone use of such inputs is officially discouraged in order to limit
expansion of barley cultivation to marginal areas.
Although agricultural extension offices are well spread in Syria and many farmers are
aware of their offices and activities (Mazid, 1994), only 38% of farmers visited them and
only 23% of farmers had ever been visited by an extension agent. The existence of
extension services alone is not sufficient for transfer of technology unless regular training is
provided and research-extension-farmer linkages are created.

Adoption and Perception of Wheat and Barley Varieties

Adoption of varieties

Wheat

Since 1977, the Directorate of Agricultural and Scientific Research (DASR) has
released eight bread and seven durum wheat varieties developed in partnership with
CIMMYT/ICARDA and three varieties developed with ACSAD. In 1998/99, about 62 and
38% of sample farmers (n=206) grew durum and bread wheat varieties or landraces in
Aleppo, Raqqa and Hasaskeh provinces respectively (Table 2). Farmers grew seven modern
durum varieties (five recommended, one obsolete, one not released) and one landrace.
Among durum varieties Cham 3 was planted by 26% of farmers on 26.3% of wheat area
followed by Lahan (10% and 7%), Bohouth 5 (8.4% and 15%) and Cham 1 (6% and 2%).
Nachit et al. (1998) found that Cham 1 and Cham 3 were the most widely grown durum
varieties, both in terms of area (33 and 30% of area) and the proportion of farmers growing
Z. Bishaw et al. / International Journal of Plant Production (2011) 5(4): 323-348 327

them (22 and 24% of farmers). Van Gastel and Bishaw (1994) found that Cham 1 was
grown by 28% of farmers and Cham 3 by 11% in Aleppo province. Since then the
proportion of Cham 1 has declined whereas that of Cham 3 has increased steadily.
Similarly, bread wheat growers planted eight modern varieties (five recommended, one
obsolete, two not released). Cham 6 was planted by 23% of farmers on 26.1% of the wheat
area followed by Cham 4 (9.5% and 7.6%), Bohouth 6 (2.2% and 1.9%) and Bohouth 4
(1.1% and 0.6%). The number of farmers growing Cham 4 was almost doubled compared
to earlier reports (van Gastel and Bishaw, 1994). Bread wheat varieties from the 1970s
(Mexipak) and 1980s (Cham 2) were still grown by a smaller proportion of farmers,
showing the persistence of older varieties.

Table 2. Wheat varieties grown by farmers in different regions of Syria (n=273).

Year
Wheat type Variety Origin Aleppo Raqqa Hasakeh Totala %
released
Durum wheat
Cham 1 DASR/ICARDA 1984 16 - - 16 5.9
Cham 3 DASR/ICARDA 1987 25 - 46 71 26
Recommended Cham 5 DASR/ICARDA 1994 4 1 6 11 4
Acsad 65 ACSAD 1987 9 - 9 18 6.6
Bohouth 5 DASR 1987 1 9 13 23 8.4
Not recommended Lahan CIMMYT/ICARDA - 19 2 5 26 9.5
Obsolete Gezira 17 DASR 1975 - 2 - 2 0.7
Hamari Local 1 - - 1 0.4
Landrace
Sub-total 75 14 79 168 61.5
Bread wheat
Cham 2 CIMMYT/ICARDA 1984 1 - 1 0.4
Cham 4 CIMMYT/ICARDA 1986 3 15 8 26 9.5
Recommended Cham 6 CIMMYT/ICARDA 1991 24 3 35 62 22.7
Bohouth 4 DASR 1987 - 3 - 3 1.1
Bohouth 6 DASR 1991 3 1 2 6 2.2
Memof CIMMYT/ICARDA 1 - - 1 0.4
Not recommended
Lagous CIMMYT/ICARDA 2 - - 2 0.7
Mexipak CIMMYT 1971 - - 4 4 1.5
Obsolete Sub total 34 22 49 105 38.5
Total 109 36 128 273 100
a
214, 22 and 5 farmers, respectively planted one, two and three wheat varieties.

Earlier studies showed that Hourani was the most widely grown landrace before the
introduction of modern varieties such as Senator Cappelli and Florence Aurore followed by
semi-dwarf wheat varieties like Mexipak in the 1970s (Bailey, 1982). The area under
modern varieties increased dramatically (Table 2) with the introduction of mechanization,
irrigation and intensification of agriculture from 8% of wheat area in 1973 to 55% in 1977
(Bailey, 1982), 89% in the early 1990s (Nachit et al., 1998) and 100% in the late 1990s
(Pingali, 1999). These varieties replaced the landraces, but were grown for market, not for
home consumption. Virtually the entire production is sold to government agencies at
attractive prices (di Fonzo et al., 1995). Wheat landraces were abandoned because of low
grain yield and low economic returns. Many farmers, however, continue growing small
plots of durum landraces for home consumption because of preferences in taste and cooking
quality for traditional foods (di Fonzo et al., 1995). In a targeted survey some durum
328 Z. Bishaw et al. / International Journal of Plant Production (2011) 5(4): 323-348

landraces such as Bayadi, Hamari, Hourani, Hourani-Bayadi and Swadi were found grown
in isolated pockets in Aleppo and Idleb provinces. Some landraces were traded over long
distances from south to north Syria (Dara’a to Aleppo) by local merchants, demonstrating
that farmers' varietal choices are not based solely on grain yield.
Cham 1, Cham 2, Cham 4, Bohouth 5, Bohouth 6 and Gezira 17 were recommended for
irrigated and high rainfall areas (zone 1), whereas Cham 3 and Cham 5 were recommended
for dry areas in zone 2 (Hamblin et al., 1995; Nachit et al., 1998). An old variety ACSAD
65 and later releases like Cham 6 and Cham 7 were recommended for zones 1 and 2.
Bohouth 1 and Cham 8 were released exclusively for irrigated conditions. Almost all wheat
varieties were grown interchangeably in zones 1 and 2 and under rainfed and irrigated
conditions despite the recommendation domains based on agro-ecological zones. For
example, 33% of respondents (n=273) grew varieties outside the recommended zones.
Nachit et al. (1998) also found that Cham 3 was widely grown in zone 1 and under irrigated
conditions despite the recommendation to grow the variety in zone 2. It is critical that
varieties be matched to agro-ecological adaptation and recommendation domains, for
maximum economic returns. Extension services and seed suppliers should make concerted
efforts to ensure that farmers are aware of this.

Barley

Barley is grown in marginal environments where severe drought and thermal stress at
maturity accompanied by spatial and temporal variations in rainfall remain major
production constraints. Developing varieties with high yield and yield stability for such
risky environments is difficult. From 1981 to 1994, however, seven modern barley varieties
were released in Syria: three from ACSAD (ACSAD 60, ACSAD 68 and ACSAD176) and
four from ICARDA (Badia, Furat 1, Furat 2 and Arta) materials. However, none of these
varieties has been widely adopted; possibly because of poor adaptation or farmers’
preferences. During the survey, farmers across different provinces and zones predominantly
grew one landrace, Arabi Aswad (99%; n=200). Arabi Abiad (0.5%) and Furat 1 (0.5%)
were grown by two farmers. Mazid (1994) also found a single farmer who adopted a
modern barley variety in northwestern and northeastern Syria. Similar situations are
reported in other countries. Despite numerous variety releases, new varieties occupied <2%
of pearl millet area in Niger (Ndjeunga, 2002) and 12% of sorghum area in the eastern
lowlands of Ethiopia (Mekbib, 2007).
Arabi Aswad (black seeded) and Arabi Abiad (white seeded) are the two landraces
primarily cultivated in two distinct regions. Arabi Abiad is adapted and mostly cultivated in
the wetter areas in western and northwestern Syria. Arabi Aswad is adapted to drier areas
and popular in the northeast. It is important to note the predominance, in the entire survey
region, of a single landrace (Arabi Aswad) with low on-farm varietal diversity. This
demonstrates broad adaptation of this landrace to different production environments. In
contrast, farmers in Ethiopia grew 14 different varieties (6 modern, 14 local)
(Woldeselassie, 1999) probably owing to the diversity of agro-ecology and end-uses.
It is also interesting to note that one fourth of wheat growers who planted modern
varieties grew only landrace of barley. In contrast, from 200 barley growers, over 50% also
planted modern bread and durum wheat varieties, but still continued growing a barley
Z. Bishaw et al. / International Journal of Plant Production (2011) 5(4): 323-348 329

landrace. In Ethiopia, although 61% of barley growers surveyed adopted modern bread
wheat varieties, they continued cultivating barley landraces (Woldeselassie, 1999). Bishaw
(2004) also found that from 304 wheat farmers, two-thirds grew barley; and two thirds of
these grew landraces. Low adoption of modern barley varieties might be due to several
factors: (i) barley is grown in harsh environments and farmers are risk averse, unwilling to
try new varieties; (ii) new barley varieties might not be high yielding as claimed from on-
station and on-farm trials; (iii) quality traits of new varieties may not fully meet farmer
preferences; (iv) differential grain price incentives encouraged greater resource allocations
to wheat at the expense of barley.

Perception of varieties

Wheat

Wheat growers articulated their perceptions of modern varieties they grew on their
farms (Figure 2). Of the 206 farmers, 96% were satisfied with existing commercial wheat
varieties and believed they are adapted to local conditions and have a good combination of
agronomic traits. Cham 3 was rated high for its grain yield, grain quality, food quality, frost
tolerance, drought tolerance and better response to low moisture. Lahan was appreciated for
its high yield, grain size, non-lodging and frost tolerance. ACSAD 65 and Cham 5 were
rated high for their low water requirements. In bread wheat Cham 6 was rated highly for
grain yield, food quality, non-lodging and high yield with limited water. In contrast, Cham
4 was rated highly for yield, but low for other agronomic traits. This demonstrates that
farmers' decisions to grow a particular wheat variety are primarily based on yield and
economic returns (di Fonzo et al., 1995).

Figure 2. Farmers’ perception of commercial wheat varieties (n=273).

Non-shattering and non-lodging appeared factors farmers considered most important,

probably based on practical experience during crop production. The existing varieties were
rated as very poor for both criteria by 33% and 17% of farmers, respectively. Under irrigated
conditions farmers apply very high amounts of fertilizers to maximize productivity. This
330 Z. Bishaw et al. / International Journal of Plant Production (2011) 5(4): 323-348

could increase lodging, which is apparently more affected by management than by variety
alone. Similarly, mechanical harvesting of wheat might result in shattering particularly if
delayed. Therefore, there is strong demand for varieties with better input response but also
good agronomic characters such as tolerance to lodging and shattering.
Grain color and marketability appeared to be less important in wheat unlike for wheat in
Ethiopia and barley in Syria (Bishaw, 2004). At present, neither the government nor the flour
industry pays premium prices for grain quality. Wheat production at the time was attractive
because of price incentives as most farmers produce the grain for direct marketing to
government rather than for domestic consumption (di Fonzo et al., 1995). Farmers’
perceptions on productivity of wheat varieties were heavily influenced by the amount and
distribution of rainfall in a particular year. Farmers who were entirely dependent on rainfall
expected large fluctuations in wheat production and productivity. There was a general
perception that productivity of wheat varieties had increased over the years. More than 50%
of farmers expected wheat yield of 3 to 5 t ha-1 while one-fifth expected over 5 t ha-1 probably
due to adoption of high yielding varieties, and continued expansion of irrigation facilities.
Interestingly high yield, non-lodging, non-shattering, yield with less water, frost
tolerance, and drought tolerance appeared to be varietal characteristics farmers expected in
new wheat varieties (Figure 3). Maximizing productivity become the major criterion for
adoption of new varieties as part of agricultural intensification whereas non-lodging and
non-shattering requirements are a response to mechanized harvesting. Erratic rainfall and
declining availability of irrigation were major concerns for farmers who seek alternative
varieties with drought tolerance and low water requirements. Adoption of modern varieties
could be influenced by yield, disease resistance and particularly early maturity which are
valuable in drought prone areas (Tripp, 2000; Mekbib, 2007).

100%
90%
80%
70%
Do not know
Farmers (%)

60%
Less important
50%
Important
40%
Most important
30%
20%
10%
0%
Disease tolerance
Pest tolerance
Lodging tolerance
Shattering tolerance
Frost tolerance

Heat tolerance
Grain color

Less fertilizer
Less water
Grain size

Marketability

Drought tolerance

Yield in poor soils

Straw yield
Grain yield

Food quality

Straw quality

Early maturity

Others

Agronomic characters

Figure 3. Farmers’ criteria for adoption of new wheat varieties (n=206).

Barley

Farmers had a very positive perception of the barley landrace Arabi Aswad (one third
see no disadvantage) and continued growing it for generations (Figure 4). Grain yield, grain
size, grain color, feed quality and marketability were considered important varietal
Z. Bishaw et al. / International Journal of Plant Production (2011) 5(4): 323-348 331

characters. The majority of farmers (57%; n=198) believed this landrace gives good and
stable yield under erratic rainfall and stress conditions. Feed quality appeared the second
most important characteristic, mentioned by 41% of farmers. This is crucial in crop-
livestock systems. Farmers’ preference for Arabi Aswad is linked to their perception of
better animal productivity and milk quality compared to other landraces (Arabi Abiad) or
modern varieties. Although there is no evidence to substantiate the difference in feed
quality between black and white seeded barley landraces, adoption of modern varieties
depends ultimately on farmers’ perceptions and preferences. Grain color is associated with
the marketability of barley for feed or seed through local market channels as farmers use
barley grain for livestock feed (91%), sale surplus on market (85%) or for seed (46%).
Haddad et al. (1997) reported that farmer preferences for grain color was a major constraint to
the adoption of modern varieties. Since black-seeded barley is preferred it would be difficult
if not impossible to introduce other types.

Fa rme rs (%)
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

Gra in yie ld
Gra in s ize
Gra in c olor
Agronomic characters

Fe e d qua lity
Ma rke ta bility
S tra w yie ld
S tra w qua lity
Ve ry good
Lodging Good
S ha tte ring P oor
Fros t Ve ry poor
Do not know
Drought
Dise a se
P e s ts
Le s s fe rtilize r
Le s s wa te r
Yie ld in poor s oils
Othe rs

Figure 4. Farmers’ perception of barley landrace (Arabi Aswad) in Syria (%; n=198).

Adoption of new varieties is based on several criteria (Figure 5). About 65% of farmers
said grain yield was the most important factor, followed by grain color (44%) and grain size
(37%). Farmers appeared to link grain color with marketability and grain size with feed
quality; they valued their landrace based on these criteria. Farmers also sought varieties that
could give high and stable yield under moisture stress. Participatory plant breeding has
identified grain yield, kernel weight, spike length and plant height as most important
selection criteria in barley (Ceccarelli et al., 2000). Aw-Hassan et al. (2008) reported that
barley farmers mentioned up to 15 criteria; the most important being grain yield (53%),
lodging resistance (31%), grain size (16%), plant height (12%), feed quality (9%) and
drought tolerance (6%).
The survey results indicate that there is no single variety that embodies all the traits
demanded by wheat and barley farmers. Farmers are explicit in their demand for agronomic
or quality traits in each crop and plant breeders should target these traits in breeding
programs to provide portfolio of new varieties.
332 Z. Bishaw et al. / International Journal of Plant Production (2011) 5(4): 323-348

Fa rme rs
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

Gra in yie ld
Gra in size
Gra in c olor
Varietal characters

Fe e d qua lity
Ma rke ta bility
S tora bility
S tra w yie ld
S tra w qua lity Most importa nt
Lodging tole ra nc e Importa nt
S ha tte ring tole ra nc e Le s s importa nt
Frost re sista nc e Do not know

Drought tole ra nc e
Dise a s e tole ra nc e
P e st tole ra nc e
Le ss fe rtilize r
Le ss wa te r
Yie ld in poor soil
Othe rs

Figure 6. Farmers' criteria for adoption of new barley variety (%; n=200).

Farmers’ Seed Sources and Perceptions

Seed acquisition patterns are rather complex with no single seed source. Farmers obtain
seed from multiple sources including the formal and informal sectors. Walker and Tripp
(1997) reported that the proportion of seed obtained from a particular source differed
among farmers, crops, seasons, regions and countries. Bishaw et al. (2010) and Mekbib
(2007) identified four key factors that would influence farmers’ seed sources: ecological-
adaptation to production environments; biological-variety characteristics and farmer
preferences; economic-perceived benefits; and social-cultural values or consumptive uses.

Variety sources

For variety sources, one could distinguish between the ‘primary’ source of a newly
introduced variety, and ‘secondary’ seed source of earlier released varieties. In wheat, the
formal sector was the initial source of new varieties for nearly 60% of farmers (Table 3),
although the informal sector (farmers 28% and traders 13%) also played important roles in
varietal diffusion. On the other hand, the informal sector was reported as the major initial
source of new varieties for wheat in Pakistan (Tetlay et al., 1991) and maize in Ghana
(Tripp, 1997). In barley, almost all farmers grew Arabi Aswad; 88% initially obtained the
landrace from informal sources (Table 3). In contrast, parents were identified as the ‘sole’
initial seed source for sorghum landraces in Ethiopia (Mekbib, 2007). The wheat and barley
cases demonstrate the importance of the informal sector in acquisition and diffusion of
landraces as well as new varieties.

Seed sources

In the 1998/99 crop season, 150, 44 and 12 farmers planted one, two and three wheat
varieties respectively, using 273 seed lots from different sources. Among 273 wheat seed lots,
Z. Bishaw et al. / International Journal of Plant Production (2011) 5(4): 323-348 333

59% sourced on-farm, 24% from the formal sector, 13% from neighbors/other farmers and
less than 5% from markets/traders (Table 3). Hasan (1995) and van Gastel and Bishaw (1994)
found similar results in both Jordan and Syria, with the majority of wheat farmers using on-
farm or informal sources rather than the formal sector. In the 1997/98 crop season, no farmer
in the survey obtained barley seed from the formal sector; 83% used their own seed. This
pattern is similar in most years, with local traders or markets accounting for less than 15%.
However, a non-random barley diffusion study in Syria reported that about half of barley
farmers used their own seed saved from the previous season, and 37% of them purchased seed
from neighbors (Aw Hassan et al., 2008). The informal sector remains the major source of
seed in developing countries: barley in Ethiopia (Woldeselassie, 1999) and Syria (Aw-Hassan
et al., 2008), lentil in Jordan (Al-Faqeeh, 1997), sorghum in Ethiopia (Mekbib, 2007) and
pearl millet in Niger (Ndjeunga, 2002).

Table 3. Farmer's wheat and barley varietal and seed sources in Syria.

Variety sources Seed sources

Wheat (n=273)a
Variety sources Counts Responses (%) Seed sources in 1998/99 Counts Responses (%)
Formal sector 162 60 Formal sector 65 24
Relatives 2 1 Neighbors/other farmers 34 13
Neighbors 28 10 Local traders/markets 12 4
Other farmers 45 17 Own seed 162 59
Local traders/markets 35 13
Total 272 100 Total 273 100
Barley (n=200)
Variety sources Farmers % Seed sources in 1997/98 Farmers %
Formal sectorb 27 14 Formal sector 0 0
Neighbors/other farmers 71 36 Neighbors/other farmers 22 11
Local traders/markets 37 19 Local traders/markets 13 7
Relatives 65 33 Own saved seed 165 83
Total 200 100 Total 200 100
Note: a 214, 22 and 5 farmers respectively planted one, two and three wheat varieties; 1 One farmer get seed of one
variety from two sources; b GOSM still produces and distributes barley landraces.

Formal sector seed sources

The formal sector was the second most popular wheat seed source. One fourth of
respondents obtained their seed (Table 4) on credit from ACB (56%; n=65 responses) and
farmers’ cooperatives (27%) or by direct cash purchase from GOSM (19%). ACB and
GOSM have 45 and 14 seed sale points respectively in the surveyed provinces. More than
half of the farmers had high appreciation for the quality of certified seed (58%); and
purchase it for yield benefits (22%) or to replace an old variety (16%) or to obtain fresh
seed (24%). Most farmers indicated that it was always available, properly cleaned, properly
treated and of good quality. Similar positive perceptions for certified seed were also
observed for wheat in Jordan (Hasan, 1995) and Syria (van Gastel and Bishaw, 1994).
However, only 36% were satisfied with price, although certified seed is distributed at
nominal profit in Syria (Rohrbach et al., 1997). For wheat the relatively high use of
certified seed could be attributed to five possible factors: (a) sustained government policy
promoting use of modern varieties; (b) low seed price which is provided at production cost;
334 Z. Bishaw et al. / International Journal of Plant Production (2011) 5(4): 323-348

(c) farmers’ perception of certified seed quality and benefits; (d) adequate facilities and
rural infrastructure guaranteeing better access to inputs; (e) better grain marketing
incentives delivering directly to government depots at premium prices.
For barley none of the sample farmers obtained seed from the formal sector during the
survey year although some farmers had purchased seed in earlier years. Since the early
1990s, barley seed purchases from the formal sector have declined steadily, particularly due
to the increased seed price to grain price ratio. In 1997/98, GOSM distributed 4214 t of
barley seed, sufficient for only 3% of national barley area. In such circumstances it was not
surprising not to find a single barley grower who purchased seed from the formal sector in
less favorable environments of zones 2, 3 and 4.

Table 4. Farmers’ perceptions of formal wheat seed sources (n=65).

Seed sources Farmers % Distance traveled (km) Farmers %

ACB 36 55.6 Up to 10 22 33
Cooperatives 17 26.7 10.1 to 20 20 31
GOSM 12 17.8 20.1 to 30 4 7
Why purchase certified seed 30.1 to 40 4 7
Replace old variety 10 16 40.1 to 50 10 16
Replace old seed 16 24 Over 50 4 7
Better seed quality 38 58 Time seed purchased (months) Farmers %
Better grain yield 14 22 8/9 7 11
Cheap price 1 2 10 14 22
No own seed 3 4 11 32 49
Others (credits) 7 11 12 12 18

Off-farm local seed sources

Although only 13% and 4% of wheat farmers (n=273) sourced seed from other farmers
and local traders, respectively in the 1998/99 season (Table 5), most of them had previously
obtained seed from other local sources. From 46 farmers who sourced off-farm seed locally,
50% got their seed from other farmers; 26% and 24% got their seed from neighbors and
traders, respectively. Farmers cited several reasons for acquiring seed off-farm locally:
perception of good seed quality (57%), timely availability (13%), lack of own seed (15%),
low price (13%) or interest in changing variety or seed (11%). Similarly, timely availability
and adaptation of variety were the main reasons for acquiring seed from these sources in
Jordan (Hasan, 1995). Farmers who sourced off-farm wheat seed locally also gave several
reasons for not purchasing certified seed from the formal sector. Non-availability, quality
and price of certified seed together accounted for 59% of farmers (n=46) not sourcing seed
from the formal sector (Table 5). Lack of access to credit (e.g. because of defaulting
cooperative members) is also an impediment to formal sector purchase. This has also been
observed elsewhere (Beyene et al., 1991). Such group obligations appeared problematic
where farmers associations are dysfunctional. Moreover, bureaucratic credit procedures
oblige farmers to use alternative sources.
Barley farmers who purchased seed off-farm from other farmers and from local traders
or markets were pooled together for analysis (Table 3). During the 1997/98 crop season,
close to one-fifth of barley growers (n=200) sourced their seed from other farmers (11%)
and local traders (7%) although previously more farmers had obtained seed from these
Z. Bishaw et al. / International Journal of Plant Production (2011) 5(4): 323-348 335

sources on several occasions and for various reasons. The overriding factor for obtaining
seed from outside was lack of own seed (80%) followed by perception of good seed quality
(20%) from their neighbors, other farmers or traders. About 83% of the farmers were
satisfied with the price of seed purchased from their neighbors and/or traders. Price was the
most important reason (97%) for not acquiring seed from the formal sector. It was reported
elsewhere in this article that the removal of price incentives for barley has led to decline for
barley seed demand. Bishaw et al. (2010) indicated that in the absence of subsidized seed
the actual seed demand from the formal sector may not be necessarily high for self-
pollinated crops.

Table 5. Farmers’ perceptions of off-farm local seed sources.

Why source seed from neighbors/traders Farmers % Why not source seed from formal sector Farmers %
Wheat (n=46)
Seed quality is good 26 57 Certified seed is expensive 16 35
Seed available on time 7 15 Certified seed not available 11 24
Seed price is cheap 6 13 Certified seed is of poor quality 5 11
No own saved seed 7 15 No cash to buy certified seed 6 13
Certified seed not available 4 9 No credit to buy certified seed 3 7
Acquire variety or seed change 5 11 Not aware of certified seed 3 7
Othersa 9 20 Othersb 7 15
Barley (n=35)
No own seed 28 80 Formal seed is expensive 33 97
Seed quality is good 7 20 Formal seed not available 1 3
Seed available on time 4 11 Cash shortage 4 11
Own seed not good 1 3 Lack of awareness 1 3
Exchange old seed 1 3 Others (small quantity, etc) 1 3
Price is cheap 1 3 Certified seed is expensive 3 14
Note: a Labor exchange, combine rent; b Long process, etc.

On-farm seed sources

Although farmers obtain seed off-farm for various reasons they are more likely to use
retained seed particularly for self-pollinating crops such as wheat and barley (Table 6)
where seed quality can be easily maintained. A significant number of wheat farmers used
retained seed (62%; n=273) and their overall perception was high in terms of seed quality,
timely availability, etc. Most farmers (61%; n=127) believed that own saved seed was of
good or better quality and preferable because they considered certified seed involves extra
cost, is not readily available on the market and involves bureaucratic procedures (Table 6).
Price remained the single most important factor for farmers not purchasing certified seed
(47%). In Jordan, 34.1% of wheat farmers used retained seed (Hasan, 1995) whereas the
figure is over 85% for lentil (Al-Faqeeh, 1997) showing great variation between crops.
About 165 barley growers (82.5%; n=200) used own saved seed, and among them 144
(87%; n=165) were satisfied with the quality. A little over 50% of farmers considered the
quality of own saved seed better or equal to seed from the formal sector or other sources.
Moreover, timely seed availability (27%), seed price (6%), lack of improved variety (4%)
and small quantity of seed required (2%) were some of the reasons for sourcing seed on the
farm. The overriding issues for farmers not buying seed from the formal sector were seed
price (71%), cash shortage (15%) and lack of credits (4%). The absence of modern barley
336 Z. Bishaw et al. / International Journal of Plant Production (2011) 5(4): 323-348

varieties also contributed to farmers not sourcing seed from the formal sector. In Ethiopia,
lack of alternative seed sources (57%), better adaptation of landraces (41%) and good
quality of own seed were the main reasons for barley farmers using retained seed
(Woldeselassie, 1999). Lyon and Danquah (1998) reported that farmers who use their own
seed do not incur a transaction cost; this is an economic incentive. For most self-pollinated
cereal crops such as wheat and barley, own seed is the major source in developing (Bishaw,
2004) as well as developed countries (Stanelle et al., 1984).

Table 6. Farmers’ perception of on-farm seed sources in Syria.

Why farmers source seed on-farm Farmers % Why farmers not source certified seed Farmers %
Wheat (n=127)
Seed quality is good 77 61 Certified seed is expensive 59 47
Seed available on time 14 11 Certified seed not available 21 17
No extra seed cost 7 6 Poor certified seed quality 6 4
Certified seed not available 12 9 No cash/credit to buy certified seed 19 15
Difficult credit procedures 7 6 Not aware of certified seed 2 2
Variety is not adapted 3 2 Own saved seed is good 13 10
Others 10 8 Others 7 6
Barely (n=165)
Why farmers source seed on-farm Why farmers not source formal seed
Perception Farmers % Certified seed is expensive 117 71.3
Seed available on time 45 27.3 No cash to buy certified seed 25 15.2
Good seed quality 90 54.6 Lack of credit 7 4.2
No extra seed cost 10 6.1 No new variety 5 3.0
Small seed quantity 3 1.8 Lack of awareness 3 1.8
No improved variety 6 3.6 Poor seed quality 5 3.0
Others 10 6.1 Lack of seed 2 1.2

Seed Flow, Retention and Transaction

While the formal sector remains a source of ‘primary’ diffusion for injecting new crop
varieties, the informal sector serves as ‘secondary’ diffusion through farmer-to-farmer seed
exchange. Compared to the formal sector, local seed systems are traditional, informal and
operate at community level (Cromwell et al., 1993).

Local seed flow

Wheat (50%; n=206) and barley (58%; n=200) farmers indicated that they exchanged
seed of modern wheat varieties and barley landraces with other farmers (Table 7). Local
seed exchange for wheat was slightly lower than barley, probably because of high varietal
turnover and seed replacement rates from the formal sector. Nachit et al. (1998) reported
that wheat farmers in irrigated areas are more dependent than in rainfed areas on seed from
the formal sector. Although almost all barley farmers used a landrace the practice of seed
exchange was high (Table 7). Aw-Hassan et al. (2008) reported that 24% of barley growers
sold seed of new varieties to other farmers. However, the retention of barley seed for longer
periods might have reduced local seed turnover. For wheat and barley farmers who reported
selling seed informally, the major customers were other farmers, neighbors, relatives or
local grain traders.
Z. Bishaw et al. / International Journal of Plant Production (2011) 5(4): 323-348 337

Table 7. Wheat and barley seed flow among farmers in Syria.

Wheat (n=206) Barley (n=200)

Local seed transaction mechanisms Farmers % Local seed transaction mechanisms Farmers %
Local seed sales Local seed sales
Not sale/exchange seed 103 50 Not sale/exchange seed 85 43
Sale/exchange seed 103 50 Sale/exchange seed 115 58
Users of seed exchange Users of seed exchange
Relatives 50 49 Relatives 53 46
Neighbors 71 69 Neighbors 61 53
Other farmers 72 70 Other farmers 74 64
Others (traders) 2 2 Others (traders) - -
Mechanism of exchange Mechanism of exchange
Cash 100 97 Cash 102 89
Others (gift, seed exchange) 3 3 Others (gift, seed exchange, etc) 13 11

Local seed exchange is important for not only acquiring seed but also for introducing
new crops/varieties over long distances and increasing on-farm crop/varietal diversity.
Bajracharya (1994) reported the role of women as key players in such endeavors in Nepal.
Empirical evidence from Ethiopia, Syria (Bishaw, 2004) and elsewhere demonstrates this.
Wheat farmers had traveled over 50 km to buy certified seed; one third of farmers traveled
up to 10 km (33%) and 20 km (31%). Barley farmers traveled up to 50 km to buy seed from
local markets or traders, but about two thirds of them traveled < 25 km. Tetlay et al. (1991)
found that 80% of farmers who sourced seed of new wheat varieties from other farmers get
the seed within 5 km. Cromwell et al. (1993) reported that farmers in Malawi walk over 30
km for bean seed and undertake five day’s travel in Nepal to acquire potato seed. Mekbib
(2007) also reported the role of informal sorghum landrace seed exchange extending over
300 km in eastern Ethiopia.
Some farmers who provided seed to others were contract growers for GOSM serving as
potential nodes for diffusion of new varieties. Moreover, farmers who rented combine
harvesters and receive payment in kind, also assisted in varietal and/or seed diffusion: they
use the grain as seed or sell it to other farmers for planting. Even in situations where
modern varieties are widely adopted, farmer-to-farmer exchange still remains the main
diffusion mechanism for new varieties.

Seed retention

Several factors influence farmers’ decision to change variety and/or seed stock. These
include yield gains from new variety, and perceived decline in performance of existing
varieties (Heisey and Brennan, 1991; Mpande and Mushita, 1996). Almost all sample farmers
(n=206) who planted wheat in the 1998/99 crop season had replaced their wheat seed stock
within the previous five years: 41% obtained fresh certified seed or changed their seed
informally in survey year; 35% retained seed for one year; 14% for two years; 8% for three
years; and 2% for four years (Table 8). Among certified seed users the wheat seed
replacement rate was high: about two thirds claimed buying seed from the formal sector
annually; the rest buy seed every three years. High seed replacement rate is useful provided
new varieties are released frequently and seed is available in the market. Van Gastel and
Bishaw (1994) found high seed renewal rates among wheat growers in Syria where nearly
338 Z. Bishaw et al. / International Journal of Plant Production (2011) 5(4): 323-348

80% replace seed within three years. Byerlee and Moya (1993) and Nachit et al. (1998)
found low average age of wheat varieties in Syria, an indicator of rapid varietal turnover
probably due to such quick seed replacement. Cromwell et al. (1993) reported that over
75 and 40% of farmers growing soybean and beans, respectively, replaced their seed in less
than five years.

Table 8. On-farm seed retention among wheat and barley farmers in Syria.

Wheat (n=273) Barley (n=200)

Number of years Farmers % Number of years Farmers %
0 111 41 0 35 18
1 95 35 1 23 12
2 37 14 2 21 11
3 22 8 3 13 7
4 3 1 4 15 8
5 2 1 5 20 10
>5 3 1 6-9 19 10
10-19 43 22
>20 11 6
Total 273 100 Total 200 100

There was a relatively high turnover of barley seed although all farmers were growing a
landrace (Table 8). About two thirds of farmers replaced their seed during the previous five
years; 85% during the previous 10 years. This high turnover is driven by three factors. First, the
formal sector used to provide cleaned and treated seed of the landrace at a relatively low price.
Second, the government grain price for barley prompted farmers to sell their produce and buy
subsidized seed from the formal sector. Third, frequent droughts and crop failures particularly in
marginal areas forced farmers to seek seed from outside sources. When different grain and seed
prices for barley and wheat were introduced, farmers opted to use their own seed. Purchases
from the formal sector dropped significantly except in drought years.
Despite frequent droughts and crop failures, 25% of farmers still retain barley seed for
over 10 years. In Ethiopia, for example 30% of barley farmers who used own saved seed
retained the same seed lot for over nine years while some of them claim they inherited it from
their ancestors (Woldeselassie, 1999). Similarly, Cromwell et al. (1993) quoted data from
Nepal where farmers typically replace wheat seed every 7 years, open pollinated maize every
10 years and rice seed every 20 years. Nagarajan and Smale (2007) also reported that farmers’
cultivars of sorghum and millets have been grown for longer periods (25-32 years) than
improved open-pollinated sorghum varieties (10 years) or hybrid sorghum or pearl millet (5-7
years). Mpande and Mushita (1996) indicated that sorghum and pearl millet farmers in
Zimbabwe keep enough seed for two seasons as security against droughts. Mekbib (2007)
also emphasized the survival of three predominant sorghum landraces extending over several
regions in eastern Ethiopia. This might explain the survival of two barley landraces with
better adaptation to the extremely harsh environments in Syria.

Seed transactions

Wheat and barley farmers who sold and/or acquired seed off-farm from local sources
(relatives, neighbors, other farmers, traders or markets) used a variety of different
Z. Bishaw et al. / International Journal of Plant Production (2011) 5(4): 323-348 339

transactions including cash, gifts, seed swaps, in-kind seed loans or labor exchange (GTZ
and CGN, 2000). Most wheat (97%; n=103) and barley (89%; n=115) seed transactions
locally, however, were in cash showing how the cash economy is replacing traditional
exchange mechanisms as farmers become integrated into commercial markets. Some earlier
studies also showed that almost all local seed transactions were cash purchases: few were
gifts or exchanges for wheat in Pakistan (Tetlay et al., 1991), barley in Syria (Aw-Hassan
et al., 2008) or for beans in Rwanda (Sperling, 1998).
On the contrary, Rohrbach et al. (1997) found that about 80% of local seed exchange
among smallholder sorghum and pearl millet farmers in southern Zimbabwe was in the
form of gifts; and relatives and other farmers accounted for nearly 30% of the seed supply.
Mekbib (2007) also reported that apart from own stock, gifts were the important sorghum
seed source in eastern Ethiopia. However, this form of transaction could be possible
because only a small amount of seed (less than 2 kg) is exchanged particularly for millets
compared to wheat and barley where larger quantities of seed are required.
Empirical evidence shows that smallholder farmers are willing to pay cash: twice the
grain price in Zimbabwe (Mugedza and Musa, 1996), 32% more than the price of certified
seed for new wheat varieties in Ethiopia (Ensermu et al. (1998). However, high price and
lack of credit remain constraints for buying certified seed (Rohrbach et al., 1997) or for
adopting new varieties (Kotu et al., 2000).The high level cash transactions observed in this
study suggest opportunities for local seed businesses that would facilitate the introduction
and diffusion of new varieties. However, any local initiative requires careful consideration
of crop suitability and socio-cultural factors.

Farmers’ Seed Management

In traditional agriculture, genetic resources conservation, crop improvement, crop

production and seed supply are integrated at the farm level. Farmers-consciously or
unconsciously-manage their genetic resources by continuously selecting and adapting
landraces to their environment, and selecting seed as an integral part of crop production.
Plant/seed selection, seed cleaning, seed treatment and monitoring quality are closely
linked. A decision to use a given seed lot for planting involves continuous monitoring and
evaluation of the entire crop growth in the field, later at harvesting, threshing and during
storage to differentiate between seed for planting and grain for consumption. This approach
although it persists in many areas, is rapidly changing in commercial agriculture as farmers
are increasingly dependent on the formal seed sector. A broader understanding of farmers’
seed management practices is useful to develop strategies for strengthening the informal
seed sector.

Perception of seed quality

Farmers clearly differentiate seed from grain. All wheat growers (98%; n=200)
distinguish seed from grain and attributed the difference to cleanliness (53%), chemical
treatment (18%), freedom from weeds (31%) and diseases (9%), good germination (6%)
and seed size (13%). They clean (90%), treat (89%), store separate (64%), select (54%) and
340 Z. Bishaw et al. / International Journal of Plant Production (2011) 5(4): 323-348

check germination (4%) of wheat seed. Likewise, from 200 barley growers, 99% recognize
seed from grain and attributed the difference to purity (17%), seed size (9.5%), treatment
(2.5%), quality (2%) and freedom from weeds (1%). Most barley farmers also clean their
seed (91%), store separately (76%) and select (46%), but fewer apply seed treatment (7%)
or check germination (3%).
Germination is an important aspect of seed quality, but there is little evidence to suggest
that farmers monitor it effectively. Few wheat and barley farmers reported checking
germination before planting. Similarly, sorghum farmers in Zimbabwe did not consider
germination as important in seed quality (Mugedza and Musa, 1996). Few maize farmers in
Ghana considered poor field establishment to be associated with poor seed quality (Walker
and Tripp, 1997). Introducing simple and practical germination tests using cheap and locally
available materials could help farmers monitor germination (Mathur and Talukder, 2002).

Plant/seed selection

Seed selection is part of on-farm seed management (Walker and Tripp, 1997); and
selection time and subsequent management practices determine seed quality. Almost half of
wheat (n=206) and barley (n=200) farmers claimed to practice seed selection (Table 9). The
selection criteria are based on conditions of standing crops or grain quality at harvesting
time, on threshing floors, during storage or at planting time. Famers decide which field or
part of field could be harvested and further evaluate the product to differentiate between
seed for planting or grain for food/feed. Most wheat farmers select a field or section of a
field of standing crops (87%; n=111) and usually before (6%) or at harvesting time (87%).
Most barley growers also select a field or section of a field (79%; n=99) usually before
(20%) or at harvesting time (76%). Similar results were reported for sorghum and pearl
millet: seed is selected mostly in the field and at threshing time (Mpande and Mushita,
1996), allowing farmers to evaluate the crop for agronomic characters like lodging, pest
infestation, etc.

Table 9. Farmers’ plant and/or seed selection in Syria (n=206).

Wheat (n=206) Barley (n=200) Wheat (n=111) Barley (n=99)

Plant/seed selection Selection criteria
Farmers % Farmers % Farmers % Farmers %
Not select for seed 95 46 101 51 Early maturity 5 5 10 10
Select for seed 111 54 99 50 Shattering tolerance 3 3 1 1
Method of selection1 Lodging tolerance 5 5 - -
Select field/part of field 96 87 79 79 Disease tolerance 20 18 3 3
Select grain 14 13 23 23 Pest tolerance 3 3 - -
Select ears 8 7 1 1 Plant height 12 11 13 13
Select plants 2 2 8 8 Ear size 38 34 22 22
Others 2 2 - - Grain yield 19 17 41 41
Time of selection1 Grain size 36 32 68 69
At planting 6 5 5 5 Grain color 8 7 42 42
Before harvesting 7 6 19 20 Marketability 1 1 4 4
At harvesting 96 87 75 76 Straw yield 2 2 4 4
During storage 1 1 - - Straw quality 2 2 1 (3) 1 (3)
Responsibility for selection1 Freedom from weeds 80 72 55 56
Men 109 98 95 96 No offtype/other crop 7 6 5 5
Both (men and women) 2 2 4 4 Cleanliness of field 5 5 - -
Note: 1 Based on farmers who select seed for wheat (n=111) and barley (n=99).
Z. Bishaw et al. / International Journal of Plant Production (2011) 5(4): 323-348 341

For wheat farmers, freedom of standing crops from contaminating weed plants (72%),
ear size (34%), grain size and absence of disease infection were important criteria in
selecting seed for planting. Barley farmers considered seed size (69%; n=99) followed by
freedom from weeds (56%) and ear size (22%) of standing crop, grain yield (41%) or grain
color (42%) as important criteria. These criteria were used to decide which field or
harvested grain should be kept for seed. Other methods and criteria have been reported,
such as individual ear or head selection for maize, sorghum or pear millet; and selection
based on grain yield, grain color, grain size, early maturity, drought tolerance or their
combinations (Mpande and Mushita, 1996). Walker and Tripp (1997) reported seed
selection in the field was practiced by <4% of maize and cowpea farmers in Ghana; and by
18 to 25% of sorghum and cowpea farmers in Zambia. In maize, field selection was based
on big ear, big seed size and absence of disease and early maturity whereas post-harvest
selections focus on size of grain, cob or its conditions (cleanliness, appearance, absence of
insects). In cowpea on-farm storage (threshed or unthreshed) is a major criterion. In
Ethiopia, it was reported that the main selection criteria for sorghum landraces are
agronomic performance (adaptation, yield, stress tolerance), consumption quality (taste,
color, consistency, cooking time, processing) and animal feed value (Mekbib, 2006).
Most farmers in the study area grew both wheat and barley, particularly in zone 2 where
they have similar criteria. Seed selection started from choosing the right field to the
conditions of the standing crops to grain quality at harvesting, threshing, storage and/or
planting time based on the knowledge and perception of individual farmers. Selection of a
field could be attributed to its previous history, e.g. properly rotated, fertilized, irrigated or
clean field with good crop stand, grain yield and grain size. These selection criteria will not
significantly change the genetic characteristics of the variety as most plants are bulk
harvested and farmers may be unable to assess all the characters where selection is indirect.
For example, early maturity refers to a field which escapes drought thus the crop is good in
terms of grain yield or grain size. Biotic stress tolerance means rather an absence of pest or
disease attack rather than actual measurement. Similarly by selecting fields with no weed
infestation farmers may be indirectly selecting for plants that have some inherent weed
tolerance as evidence suggests varietal differences in wheat (Rizvi et al., 2002). However,
the key factor in on-farm selection is that farmers can relate each selection criterion with
their knowledge and experiences to differentiate between seed of good or poor quality.
Women’s role in seed selection and management appeared less visible compared to that
reported for wheat in Ethiopia (Bishaw et al., 2010) and vegetables in Bangladesh (Sillitoe,
2000) and Nepal (Bajracharya, 1994). Increased use of combine harvesters substantially
reduced female labor contribution in wheat and barley production while previously women
were involved in manual harvesting (Tully, 1990) and therefore directly contributed to
on-farm selection and maintenance of wheat and barley seed. Mazid (1994) reported that
about 64% of farmers shared their barley production decision with their immediate family
(including spouses). In another study women were responsible for most on-farm cleaning of
barley seed (Daniela Mangione, Personal communication) although men failed to
acknowledge it. Bajracharya (1994) indicated that although women’s contributions to on
farm work and decision making on the average was 57%, the agricultural development
officers (men) perceived that the contribution of women was low (11-23%) compared to the
ratings of female development officers (62%), a clear reflection of a gender bias. In
practical terms, women directly or indirectly contributed to plant and/or seed selection both
in wheat and barley crops despite a generally held view which underestimates their roles.
342 Z. Bishaw et al. / International Journal of Plant Production (2011) 5(4): 323-348

Plant and/or seed selection practices can be summarized as follows: (i) no methodological
approaches were observed in plant selection in either wheat or barley; (ii) farmers’ seed
selection was anecdotal and not systematic, largely influenced by observation in the field or at
harvesting or planting time; (iii) intensification and commercialization of production is
leading to loss of traditional plant/seed selection practices; (iv) high seed renewal rate and
varietal turnover resulting from availability of new wheat varieties has reduced farmer’s need
to ‘improve’ existing varieties; (v) plant and/or seed selection in wheat and barley is roughly
similar, although almost all barley farmers still used landraces.

Seed cleaning and treatment

Wheat and barley seed is commonly cleaned, whether sourced locally from farmers, traders
or retained on the farm. This may include all post-harvest operations like drying (removing
excess moisture), cleaning (removing impurities), grading (improving quality), treating
(protection against pests) and packaging. In certain circumstances, on-farm seed cleaning is no
more than winnowing the seed after harvesting; this does not guarantee selection of uniform
grain sizes (Mpande and Mushita, 1996). In other cases, detailed seed cleaning techniques are
employed to maintain seed quality (Mugedza and Musa, 1996) or an elaborate traditional seed
treatment techniques are used against storage pests (Monyo et al., 2000).
All wheat growers surveyed reported using cleaned and treated seed either from the
formal sector or through on-farm seed cleaning and treatment (Table 10). Forty-five
farmers (22%) used cleaned and treated certified seed sourced from the formal sector.
About 161 farmers (78%) sourced seed from other farmers, traders or used their own seed;
most of the seed was cleaned and/or treated by farmers themselves. Similarly, 91% of
farmers (n=200) cleaned their barley seed obtained from local sources or saved on farm
(Table 10). The majority of wheat/barley farmers used manual cleaning with wire mesh
sieves (78%/87%) while the remaining used locally manufactured mobile cleaners. The
main purpose of cleaning was to remove weeds, inert matter, broken seeds, other crop seeds
(e.g. barley), facilitate machine planting or removing insect infested grains. Most farmers
who used cleaning machines were in Hasakeh where the service is easily available, through
small-scale mobile cleaners fabricated in one of the nearby towns. This has emerged as an
important rural enterprise.
The striking observation in wheat seed management was the extent of chemical
treatment used by farmers in Syria. The availability of chemicals induced most wheat
growers to use seed treatment, probably influenced by the practices of the formal sector.
On-farm seed treatment was widely used. Almost all farmers treated their seed before
planting (76%) except those who purchased treated seed (24%). In Jordan, 64 and 62% of
seed sourced informally was cleaned and treated on the farm, respectively (Hasan, 1995).
Stanelle et al. (1984) reported that seed treatment was practiced by 36% of wheat farmers,
but more targeted towards areas with disease problems due to high rainfall and humidity.
Surprisingly, fewer farmers used chemical seed treatment in barley (7%) compared to
wheat. Although 56 farmers (27%; n=206) planted barley as a second crop along with
wheat none reported chemical treatment for barley seed. In Ethiopia, nearly 90% of barley
growers who retained seed on the farm or purchased seed from neighbors cleaned their seed
using traditional hand tools (Woldeselassie, 1999); but these were found inefficient in
removing weeds and inert matter. None of the farmers also used seed treatment as well.
Z. Bishaw et al. / International Journal of Plant Production (2011) 5(4): 323-348 343

Table 10. On-farm wheat and barley seed processing and management in Syria (n=161).

Wheat (n=161) Barley (n=200)

Seed cleaning and treatment Farmers % Farmers %
Not clean seed - - 12 6
Purchased clean seed 13 8 6 3
Cleaned seed 148 92 182 91
Method of cleaning
Hand sieving 125 78 174 87
Machine cleaning 23 14 8 4
Purpose of cleaning
Remove inert matter 61 38 146 73
Remove weed/crop seeds 90 61 74 37
Remove small seeds 24 15 20 10
Remove broken seeds 54 34 21 11
Facilitate planting 4 3 29 15
Remove insect infested seeds - - 8 4
Seed treatment
Not treat seed - - 169 85
Purchased treated seed 5 3 - -
Treat seed 156 97 13 7
Check germination 9 6 6 3

A number of chemicals, local or imported, were available for seed treatment in Syria.
From 156 wheat farmers who treated their seed on farm most used broad-spectrum
fungicides like Quinolate (69%) followed by Agrospor 60 (19%). The main purpose of seed
treatment was for control of smuts (73%), but only 3% applied Vitavax a systemic
fungicide. The chemical, usually in powder form, is first diluted in water and then mixed
with seed manually on tarpaulins in the open using shovels (87%) whereas mobile cleaners
were used to treat the rest. Only one-third of farmers applied the seed treatment at
recommended rates, and only 12% used the necessary safety measures while treating seed.
Most farmers did not have adequate knowledge of the chemicals (could not identify the
name) and their application and efficacy. Inadequate formulation of chemicals; lack of
knowledge on methods and rates of application; lack of adequate equipment and knowledge
in handling; and lack of safety precautions are matters for concern. Sub-standard chemicals
without proper formulation and of unknown origin were available on the market. Adequate
quality assurance and appropriate extension would help ensure safety, increase efficacy,
target pests more effectively, reduce costs and lessen environmental pollution.

Seed storage and management

Among 206 wheat and 200 barley farmers surveyed, 64% and 76% respectively, stored
seed and grain separately (Table 11). Almost all grain/seed was primarily stored in jute or
polypropylene bags, showing the disappearance of traditional storage facilities reported
elsewhere. Walker and Tripp (1997) found that farmers in Zambia tend to separate
sorghum, bean and groundnut seed whereas in Ghana they are less predisposed to such
practice for maize and cowpea seed; jute/polypropylene sacks were widely used in both
countries. Traditional storage structures much quoted elsewhere (Mpande and Mushita,
1996; Walker and Tripp, 1997) are uncommon because they are irrelevant for wheat and
barley where large quantities are stored.
344 Z. Bishaw et al. / International Journal of Plant Production (2011) 5(4): 323-348

Table 11. On-farm wheat and barley seed storage and management in Syria.

Store seed separately1 Not store seed separately1

Seed storage practices Wheat (n=132) Barley (n=152) Seed storage practices Wheat (n=74) Barley (n=48)
Farmers % Farmers % Farmers % Farmers %
Separate seed store 132 64 152 76 Not separate seed store 74 36 48 24
Polypropylene bag 52 39 - - Polypropylene bag 24 32 1 2
Jute bag 75 57 143 94 Jute bag 48 65 45 94
Poly & jute bag 5 4 - - Poly & jute bag 2 3 - -
Local bins - - 5 3 Local bins - - 1 2
Bulk - - 4 3 Bulk - - 1 2
Control of storage pests Control of storage pests
No infestation 48 36 44 29 No infestation 28 38 10 21
Sun drying 15 11 27 18 Sun drying 6 8 13 27
Cleaning 42 32 57 38 Cleaning 20 27 14 29
Change/dispose seed 15 11 13 9 Change/dispose seed 14 19 - -
Chemical spraying 28 21 2 1.3 Chemical spraying 10 14 - -
Fumigation 32 24 - - Fumigation 17 23 - -
Others - - 9 6 Others - - 2 4
Responsibility for storage Responsibility for storage
Men 111 84 134 88 Men 68 92 38 79
Women 6 5 9 6 Women 3 4 4 8
Both (men & women) 15 11 9 6 Both (men & women) 3 4 6 13
Note: 1 Percentages are calculated based on 206 and 200 wheat and barley growers, respectively; and then adjusted
accordingly based on farmers who store seed and grain separately or not.

Generally pests were reported as major constraints for on-farm grain and/or seed storage
by wheat (64%) and barley (74%) producers. Weevils, khapara beetle and rodents were
reported as serious storage pests in wheat and barley. A survey of grain and seed storage
facilities found that khapara beetle was the most widespread and destructive storage pest in
northwestern Syria (Niane, 1992). Most wheat farmers changed their seed frequently
(particularly formal sector seed), and/or sold their produce directly to government depots
with relatively less on-farm seed storage and therefore experienced less pest problems.
Wheat and barley farmers used both traditional (cleaning, sun drying, changing,
disposing) and modern (spraying, fumigation) control measures to manage storage pests.
There is an increasing trend to use contact insecticides or fumigants, even by farmers who
store seed and grain together. Similarly combination of traditional (heat treatment, sun
drying, cleaning, changing storage structures or disposing infested seed) and modern
(contact insecticides, fumigants) chemicals was commonly used to control grain storage
pests for barley in Ethiopia (Woldeselassie, 1999) and for cereals and legumes in Ghana
and Zambia (Walker and Tripp, 1997). Farmers in Ghana are more inclined to use
protectants on cowpea than on maize. Farmers in Zambia apply little or no chemical on
sorghum, although insects cause substantial damage to seed.
There was widespread use of chemicals (contact or fumigant) for wheat storage pests. The
type, rate and application method and equipment raise fundamental questions of efficacy and
safety. Inappropriate use of chemicals has led to the development of pesticide resistance
worldwide. Likewise, the strains of kahpra beetle collected from various grain/seed storage
facilities have shown different levels of pesticide resistance (Niane, 1992).

Conclusions

Awareness and adoption of modern wheat varieties and associated technologies has
increased spectacularly in Syria. Within a short period the country has become self-sufficient
Z. Bishaw et al. / International Journal of Plant Production (2011) 5(4): 323-348 345

in wheat, producing surpluses for export in good years. Availability of modern wheat varieties
with high and stable yield, generation and transfer of appropriate agronomic packages for
different agro-ecological zones and expansion of irrigation are the main driving forces behind
this achievement. The success also hinges on the existence of a strong wheat seed system
where certified seed is available, affordable and regularly used by farmers. The wheat case
demonstrates how linkages between research and the seed sector can help achieve national
food security if properly backed by government commitment and an enabling policy
environment in providing inputs, marketing and price incentives. Such a strategy is important
if the wheat success is to be replicated in other crops in Syria or elsewhere.
Barley is a typical crop of marginal environments where yield is limited by severe abiotic
and biotic stresses. Farmers in the major production areas grow landraces and depend on
informal seed sources. Several improved varieties have been released but none has been
widely adopted; possibly because of lack of adaptation or farmers’ preferences. This has led
to researchers exploring alternative strategies, for example exploring participatory plant
breeding (PPB) methods which show some promise. Greater effort is needed to combine
research with farmers’ knowledge to identify new barley varieties adapted to marginal
environments, with traits acceptable to farmers. More flexible policy options are also needed,
in addressing crop improvement, technology transfer, seed provisions, price incentives, etc
and enhance barley production and productivity in the country.
Farmers have clearly articulated perceptions of wheat and barley varieties they were
growing; and identified several technological and socio-economic criteria that determine
adoption. High yield, grain size, food quality, and tolerance to lodging, drought and frost
are the traits that farmers seek in new bread and durum wheat varieties. Barley growers
seek grain yield, grain size, grain color, feed quality, marketability, drought tolerance and
disease resistance in new varieties. Plant breeders need to combine agronomic traits desired
by famers and grain quality that matches the needs of industry.
In reality, the informal sector remains the major source of new varieties and the
‘default’ seed supplier for botyh wheat and barley. Most farmers recognize the difference
between seed for planting and grain for consumption and as a result used different
management practices to maintain seed quality such as on-farm selection, cleaning,
treatment, separate storage or seed quality assessment. The majority of farmers have shown
a considerable degree of sophistication in introducing new technology such as cleaners,
treatment, etc. The government should recognize the role of the informal sector, and
provide adequate policy, regulatory and technical support in order to integrate it with the
formal sector.

References
Al-Faqeeh, T.M., 1997. A survey of lentil seed quality in Jordan. University of Jordan, Amman, Jordan. M.Sc.
Thesis, 97p.
Aw-Hassan, A., Mazid, A., Salahieh, H., 2008. The role of informal farmer-to-farmer seed distribution in diffusion
of new barley varieties in Syria. Exp Agric. 44, 413-431.
Bajracharya, B., 1994. Gender issues in Nepali agriculture: a review. Research Report No. 25. HMG Ministry of
Agriculture/Winrock International. Kathmandu, Nepal. 78p.
Bailey, E., 1982. An examination of the role of supplementary irrigation in stabilizing wheat production in an area
of variable rainfall. University of Reading, Reading, UK. M.Sc. Thesis, 120p.
Bishaw, Z., Struik, P.C., Van Gastel, A.J.G., 2010. Wheat seed system in Ethiopia: Farmers’ varietal perception,
seed sources and seed management. J. New Seeds. 11, 1-47.
346 Z. Bishaw et al. / International Journal of Plant Production (2011) 5(4): 323-348

Bishaw, Z., 2004. Wheat and Barley Seed Systems in Ethiopia and Syria. Ph.D. Thesis, Wageningen University,
Wageningen, the Netherlands, 383p.
Byerlee, D., Moya, P., 1993. Impacts of international wheat breeding research in developing world, 1966-90.
CIMMYT, Mexico, 29p.
Ceccarelli, S., Grando, S., Tutwiler, R., Baha, J., Martini, A.M., Salahieh, H., Goodchild, A., Michael, M., 2000.
A methodological study on participatory barley breeding: I. Selection phase. Euphytica. 111, 91-104.
Central Bureau of Statistics, 2000. Statistical Abstract 2000: Fifty Three Years. Central Bureau of Statistics,
Damascus, Syria, 565p.
Cromwell, E., Wiggins, S., Wentzel, S., 1993. Sowing beyond the state: NGOs and seed supply in developing
countries. ODI, London, UK, 143p.
Di Fonzo, N., Kaan, F., Nachit, M., (eds.), 1995. Durum wheat quality in the Mediterranean region. Proceedings of
the seminar on durum wheat quality in the Mediterranean region, 17-19 September 1993, Zaragoza, Spain.
CIHEAM, Zaragoza, Spain, 284p.
Deutscher Gesellschaft für Technische Zusammenarbeit (GTZ) and Center for Genetic Resources (CGN), 2000.
Support for the informal seed sector in development cooperation-conceptual issues. GTZ, Eschborn,
Germany, 24p.
Haddad, N., Tutwiler, R., Thomson, E., (eds.), 1997. Improvement of crop-livestock integration systems in West
Asia and North Africa. Proceedings of the regional symposium on integrated crop-livestock integration
systems in the dry areas of West Asia and North Africa, 6-8 November 1995, Amman, Jordan. ICARDA,
Aleppo, Syria, 572p.
Hamblin, J., Al-Taweel, W., Yau, S.K., Walker, G., Michel, M., Kashour, G., 1995. The use of data from on-farm
trials in planning breeding strategy. ICARDA Cereals Program Discussion Paper. ICARDA, Aleppo, Syria, 24p.
Hasan, B.M.A., 1995. A survey of wheat seed quality in Jordan. University of Jordan, Amman, Jordan. M.Sc.
Thesis, 95p.
Heisey, P.W., Brennan, J.P., 1991. An analytical model of farmers' demand for replacement seed. Amer. J. Agric
Econ. 73, 1044-1052.
ICARDA, FAO, AARINENA and CIHEAM, 1999. The national agricultural research systems in the West Asia
and North Africa Region. ICARDA, Aleppo, Syria, 278p.
ICARDA, 2002. WANA catalogue of field and seed standards (cereals, legumes, oilseeds, forages and vegetables).
ICARDA, Aleppo, Syria.
ICARDA, 1992. Fertilizer use on wheat in northern Syria 1986-1990: Part 1. collaborative research project.
ICARDA, Aleppo, Syria, 138p.
Issa, M.E.S., 1991. Economics of wheat production in Syria. Faculty of Agriculture, University of Aleppo.
Aleppo, Syria. M.Sc. Thesis, 147p.
Kotu, B.H., Verkuijl, H., Mwangi, W., Tanner, D., 2000. Adoption of improved wheat technologies in Adaba and
Dodola woredas of the Bale highlands, Ethiopia. CIMMYT, Mexico, D.F and EARO, Ethiopia, 26p.
Lyon, F., Danquah, S.A., 1998. Small-scale seed provision in Ghana: social relations, contracts and institutions for
micro-enterprise development. Agricultural Research and Extension Network. Network Paper No. 84. ODI,
London, UK, 10p.
Mathur, S.B., Talukder, M.H., 2002. Jute mat, a quick method for testing germination of rice seeds by the farmers
in Bangladesh. Seed Sci. Technol. 30, 459-462.
Mazid, A., Tutwiler, R.N., Al-Ahmed, H., Martini, M.A., Maya, F., 2003. Impact of modern agricultural
technologies on durum wheat production in Syria. Integrated Natural Resource Management Research Report
Series No. 3. ICARDA, Aleppo, Syria, 34p.
Mazid, A.M.M., 1994. Factors influencing adoption of new agricultural technology in dry areas of Syria.
University of Nottingham, Nottingham, UK. Ph.D. Thesis, 570p.
Mekbib, F., 2007. Farmers' seed system of sorghum (Sorghum bicolor (L.) Moench) in the center of diversity: I.
Seed sources, distribution, and networking, J. New Seeds. 8, 63-86.
Mekbib, F., 2006. Farmer and formal breeding of sorghum (Sorghum bicolor (L.) Moench) and the implications
for integrated plant breeding, Euphytica. 152, 163-176.
Monyo, E.S., Awala, S.K., Lileka, J., Ipinge, S.A., 2000. How much trouble a farmer would it take to reserve
seed? International Sorghum and Millets Newsletter. 41, 6.
Mpande, R., Mushita, A., 1996. Enhancing a better understanding of local level seed production systems in
Zimbabwe. SADC/GTZ, Harare, Zimbabwe, 30p.
Mugedza, C., Musa, T., 1996. Brief review of smallholder seed supply systems in Zimbabwe. SADC/GTZ,
Harare, Zimbabwe, 16p.
Z. Bishaw et al. / International Journal of Plant Production (2011) 5(4): 323-348 347

Nachit, M.M., Baum, M., Porceddu, E., Monnevex, P., Picard, E., (eds.), 1998. SEWANA (South Europe, West
Asia and North Africa) Durum Research Network. Proceedings of the SEWANA Durum Network Workshop,
20-23 March 1995, ICARDA, Aleppo, Syria. ICARDA, Aleppo, Syria, 354p.
Nagarajan, L., Smale, M., 2007. Village seed systems and the biological diversity of millet crops in marginal
environments of India. Euphytica. 155, 167-182.
Ndjeunga, J., 2002. Local village seed systems and pearl millet seed quality in Niger. Exp Agric. 38, 149-162.
Niane, A.A., 1992. Phosphine resistance in field strains of Rhizopertha dominica (Fabricius (COLEOPTERA
BOSTRICHIDAE) infesting wheat and barley in different storage facilities of Northern Syria (Aleppo).
University of Cukurova, Adana, Turkey. M.Sc. Thesis, 70p.
Pingali, P.L., (ed.), 1999. CIMMYT 1998-99 world wheat facts and trends. Global wheat research in a changing
world: Challenges and achievements. CIMMYT, Mexico, 82p.
Rizvi, S.J.H., Bazzari, D., Roustii, M., Ketata, H., 2002. In search of a natural alternative to herbicides in the
highlands. Caravan. 16, 41.
Rohrbach, D.D., Bishaw, Z., Van Gastel, A.J.G., (eds.), 1997. Alternative strategies for smallholder seed supply.
Proceedings of an International Conference on Options for Strengthening National and Regional Seed
Systems in Africa and West Asia. ICRISAT, Andra Pradesh, India, 281p.
Sillitoe, P., (ed.), 2000. Indigenous knowledge development in Bangladesh: present and future. The University
Press Limited, Dhaka 1000, Bangladesh, 238p.
Sperling, L., 1998. The effects of the Rwandan war on crop production, seed security and varietal security: a
comparison of two crops, 71-91. In Restoring farmers’ seed systems in disaster situations. FAO Plant
Production and Protection Paper 150. FAO, Rome, Italy, 225p.
Stanelle, J.R., Vanderlip, R.L., Burchett, L.A., Shroyer, J.P., 1984. Kansas wheat seed: comparison of the quality
of wheat seed planted in 1973 and 1984. J. Appl. Seed Prod. 6, 62-67.
Tetlay, K.A., Heisey, P.W., Ahmed, Z., Munir, A., 1991. Farmers' sources of wheat seed and wheat seed
management in three irrigated regions of Pakistan. Seed Sci. Technol. 19, 123-138.
Tripp, R., 2000. Strategies for seed system development in Sub-Saharan Africa: a study of Kenya, Malawi,
Zambia and Zimbabwe. Working Paper Series No. 2. Socioeconomic and Policy Program, ICRISAT,
Bulawyo, Zimbabwe, 56p.
Tripp, R., Pal, S., 1998. Information exchange in commercial seed markets in Rajasthan. Agricultural Research
and Extension Network. Network Paper No. 83. ODI, London, UK, 11p.
Tripp, R., (ed.), 1997. New Seed and Old Laws. Regulatory reform and diversification of national seed systems.
ODI, London, UK, 259p.
Tully, D., (ed.), 1990. Labor and rainfed agriculture in West Asia and North Africa. ICARDA, Aleppo, Syria,
299p.
Van Gastel, A.J.G., Bishaw, Z., 1994. Seed survey in Syria, 13-16. In: Seed Unit Annual Report for 1994,
ICARDA, Aleppo, Syria, 39p.
Walker, D.J., Tripp, R., 1997. Seed management by small-scale farmers in Ghana and Zambia, 207-215. In
International Workshop on Seed Security for Food Security, 30 November-1 December 1997, Florence, Italy.
FAO, Rome, Italy, 231p.
Woldeselassie, Y.S., 1999. Evaluation of status and quality of barley seed used by the northern and central
Ethiopian farmers. University of Jordan, Amman, Jordan. M.Sc. Thesis, 65p.
348 Z. Bishaw et al. / International Journal of Plant Production (2011) 5(4): 323-348