Assessment of bread wheat (Triticum aestivum L.) seed quality accessed through different seed sources in northwest Ethiopia

2.1 Data collection

This study was conducted in Baso Liben district, Amhara region, Ethiopia, during the 2017/18 main cropping season (Figure 1). The data were collected in two steps. First, data were collected from sample farmers to understand farmers’ bread wheat seed production and management experiences. Multistage sampling procedures were used to select the sample villages (peasant associations) and individual respondents. Six villages were purposely selected based on their potential for wheat production and their representation in the district regarding production potential. A random sampling technique was used to select the respondents from the villages. A total of 108 farmers (92 males and 16 females) were interviewed. There were more males than females represented in this study. This is primarily due to the small number of female-headed households in the study area. A semi-structured questionnaire was used to gather the required information from respondents. Besides, two focus group discussions (FGDs) were carried out with local experts and farmers to collect further information. Second, data were gathered for laboratory tests on the seed samples collected from 58 farmers (48 males and 10 females). Of the 58 farmers, 30 accessed seeds from informal sources and 28 from formal sources. The seed samples were drawn from the seed lots prepared for the following planting season. About 400 g seeds was collected from each of the 58 farmers and immediately taken to the laboratory for seed quality analysis. The samples were subjected to testing for various seed quality parameters, including physical purity, physiological quality, and seed health. The analysis was conducted at the Bahir Dar and Debre Markos seed testing laboratories. The detailed procedures are described below.

Figure 1

Map of the study area.

Physical purity: The seeds collected from farmers were separated based on their sources. The samples collected from each farmer were reduced to a working sample of 120 g using a mechanical seed divider. The seed samples were thereafter purified, and the different components were separated (pure seed, other crop seed, weed seed, and inert matter). Each component was weighed using an analytical scale and analyzed based on the weight of each component, following the International Seed Testing Agency (ISTA) standard [19].

Thousand seed weight (TSW): TSW was determined by weighting 1,000 seeds from the pure seed component of each of the samples collected according to the ISTA standard [19].

Moisture content: Moisture content was determined using the oven-drying method. About 5 g seeds was taken from each sample, finely ground, and put in containers of less than 8 cm in diameter and covered by aluminum foil for pre-dry measurement. The samples were then dried in an oven adjusted at 130°C ± 1°C for 2 hours. At the end of 2 h, the containers were placed in desiccators to cool off for 30 min. Then, the containers were weighed with their covers and contents, and the moisture content was determined following the standard formula [19]:

where, M ₁ is the weight of the container and its cover in grams; M ₂ is the weight of the container, its cover, and its contents before drying in grams; and M ₃ is the weight of the container, its cover, and its contents after drying in grams.

The physiological seed quality test is the vigor of the seed, which determines the germination and subsequent seedling emergence and crop establishment in the field as well as the storage potential of the seed lot.

Germination: A germination test was conducted according to the ISTA rules by taking 100 seeds from the pure seed components of each seed sample collected [8]. The seeds were planted on a sterilized sand substratum and placed at room temperature using light transparent germination dish for 8 days. On the eighth day, the germinated seeds were categorized into normal and abnormal seedlings, while ungerminated seeds were classified as dead seeds as specified by Fairey and Hampton [20]. The percentage of each component was calculated following the ISTA procedures [19]:

Seedling shoot and root lengths: Seedling shoot length and seedling root length were assessed after the final count in the standard germination tests. Ten normal seedlings were randomly selected from each of the samples. The shoot length was measured from the point of attachment to the cotyledon to the tip of the seedling, while the root length was measured (i.e., in cm) from the point of attachment to the cotyledon to the tip of the root. The average shoot or root length was calculated by dividing the total shoot length or root length by the total number of normal seedlings measured [21].

Seedling dry weight: Seedling dry weight was measured after the final count in the standard germination test. Ten seedlings were randomly selected from each sample, cut free from their cotyledons, placed in the envelopes, and dried in an oven at 80 ± 1°C for 24 h. The dried seedlings were weighed in grams, and the average seedling dry weight was calculated [22].

Vigor index: For each of the two seed sources (formal and informal), two vigor indices (VIG1 and VIG2) were calculated. VIG1 was calculated by multiplying the standard germination percentage with the average sum of seedling shoot and root lengths at the final day of the germination test, and VIG2 was calculated by multiplying the standard germination percentage with the mean seedling dry weight [22].

VIG1 = (G%) × (L), where G% is the standard germination percentage and L is the seedling length in (cm).

VIG2 = (G%) × (DW), where G% is the standard germination percentage and DW is the seedling dry weight in (g).

Speed of germination: The standard germination tests only distinguish between normal and abnormal seedlings. However, variation in seedling size and vigor is likely to occur within the category of “normal seedling.” To distinguish the variation among normal seedlings, the speed of germination was measured by taking 100 seeds from each sample and planting them independently on a germination paper. The seeds were kept at room temperature until no further germination took place. Starting from the fourth day up to the eighth day of planting, normal seedlings were counted and removed each day, and the speed of germination was calculated by dividing the number of normal seedlings counted each day by the number of days during which they were counted [23].

where n is the number of normal seedlings counted each day and Dn is the number of days during which seedlings were counted.

Seed health: Seed samples were tested to check for the presence of seed-borne pathogens associated with bread wheat seed sources. Identification of different seed-borne pathogens was conducted by the agar plate method using potato dextrose agar according to the ISTA standard [24]. The seeds were washed with tap water, surface-sterilized by soaking them in 1% sodium hypochlorite (NaClO) for 2 minutes, and then washed in distilled water. Twenty-five seeds from each sample were planted at equal distance on the Petri dishes and incubated at 25°C for 8 days with an agar plate that had an alternating light and dark period of 12 h near ultraviolet light. At the end of the incubation period, slides were prepared, and colonies of fungi were identified based on the morphological characteristics of the pathogens (colony characteristics, shape, and septation of conidia) in comparison with appropriate reference materials (laboratory manuals, colored images, and literature). Finally, the percentage of infected seeds produced by fungi was recorded.

2.2 Data analysis

The data collected through individual interviews were analyzed using descriptive statistics, such as minimum, maximum, and percentage values. For laboratory analysis, the samples were divided into two groups, representing the formal and informal sources. The percentage of seed quality parameters related to physical quality, physiological quality, and seed health was computed using a t-test (pairwise comparison) between formal and informal sources. SPSS v23.0 was used to analyze the collected data.

3.1 Farmers’ experience in bread wheat seed production

About 32.4% of the respondents have experience in quality bread wheat seed production. The remaining 67.6% of the respondents said that they do not have experience in bread wheat seed production. The experience of farmers is mainly based on contractual seed production schemes with public seed enterprises. In other words, their role is limited as outgrowers for enterprises. They have 1 to 3 years (on average, 2 years) of experience in such seed production arrangements.

A contractual seed production arrangement is usually based on cluster farmlands, in which farmers who have land in the selected cluster have a chance to participate in seed production. Farmers who do not have their own land in selected clusters cannot participate in seed production unless they rent land from other farmers within the cluster. During FGDs, participants said that contractual seed production started in the district a few years ago in some villages, and the coverage has increased to most of the villages of the district. As mentioned by the respondents, most farmers do want or are willing to participate in contractual seed production for several reasons, including market security (market guarantee) and access to certified seed of improved varieties from the contracting party to overcome the problem of seed shortage (seed security) [6]. This arrangement, as explained by local experts during FGDs, is a legally binding agreement between the seed enterprise (contracting party) and the farmers (outgrowers). FGD participants said that the additional premium price motivates farmers to engage in contractual seed production, though they were not happy sometimes due to the presence of better alternative markets in the areas. However, compared with farmers engaging in contractual hybrid maize seed production, the premium price is not attractive.

3.2 Farmers’ practices of post-harvest seed management

The various farmers’ post-harvest seed management practices are presented in Table 1. All farmers use the manual or traditional threshing method on a cemented floor plastered with cow dung. About 77% of the farmers used the same threshing floors for all crops, while 19% prepared separate threshing floor for different varieties, and only 1% had experience preparing a new floor for each of the crops they threshed. Using the same floor for all crops or varieties during threshing could cause the mixing up of crops and varieties, which eventually leads to low purity.

Regarding farmers’ experiences with various seed management practices, nearly half (51.8%) of the respondents clean the seeds manually by hand winnowing (Table 1). In most parts of Ethiopia, seed cleaning is usually carried out before the commencement of sowing, as farmers traditionally stored seeds until planting time. About 42.6% of the respondents explained their experience in separating the seed during threshing and storing the cleaned seeds in advance for the following cropping season. Only 5.6% of respondents practiced seed treatment before planting, particularly using cow dung, ash, and malathion, and then stored the seeds in separate bags to protect the seeds from insects such as weevils. In some wheat-producing areas of Ethiopia, farmers do have experience treating the seeds with chemicals to protect them from insect infestation [25]. Regarding storage structure, 60% of the respondents used traditional storage structure, namely “Gotera or Gota,” which is found to be the most common storage structure in Ethiopia. Farmers usually plaster the “Gotera or Gota” with cow dung to prevent storage pests from attacking the stored seeds. About 39.8% of the respondents used sacks for storing seeds. Although “Gotera or Gota” is the common storage structure for most farmers in Ethiopia, as cited by the FGDs’ participants, there is a shift toward replacing it with sacks and other modern storage structures. Farmers used local seed management practices, such as seed selection, cleaning, treatment, and better storage facilities, to maintain the seed quality [18].

3.3 Farmers’ perceptions about seed quality

The perception of the respondents toward their understanding of seed quality is displayed in Figure 2. Interviewed farmers were asked to mention the most important criterion for determining quality seeds, though farmers have various insights toward seed quality. Most of the respondents (61%) perceived that quality seeds should be free from weeds as an important criterion, whereas about 23% of the respondents believed that quality seeds must be free from any deformities (shriveled). The remaining 13 and 3% of the respondents pointed out that quality seeds should be free from diseases and insect damage and have good uniformity in the field. Most farmers consider physical purity to be an important characteristic of quality seed. The possible reason may be that the seed, which is physically pure, has a good physical appearance in the field and may attract the attention of farmers. This may not necessarily be the only important perception to be considered. There are other attributes such as being free from diseases or insect damage (seed health) and having uniformity in the field (genetic quality), as was explained by experts during FGDs. These attributes mainly determine the germination, establishment, and true-to-type of a given variety. Physical purity, uniformity, and freedom from insects, diseases, and weeds are the most important seed quality-measuring parameters perceived by farmers in the previous studies conducted in Ethiopia [26,27].

Figure 2

Farmers’ perception on quality seed criteria.

3.4 Physical quality

The differences between the seed accessed through formal and informal seed sources regarding the various physical quality parameters are presented in Table 2. Highly significant (P = 0.000) differences were observed between formal and informal seed sources for pure seed, other crop seed, weed seed, and inert materials. The mean percentages of pure seed were 98 and 94.3% for the seed obtained through formal and informal seed sources, respectively. Earlier studies show that certified seed of improved wheat varieties from the formal sector had the highest analytical purity but was not significantly different from other seed sources [18]. Participants during FGDs explained that seeds collected from informal seed sources are usually subject to seed mixtures due to poor handling mechanisms and other abiotic factors.

The presence of weed seeds and other crop seeds was not observed in the samples accessed from the formal seed source, indicating good care during seed production in the field, processing, storage, and transportation. On the other hand, weed seeds and other crop seeds were observed in 43 and 90%, respectively, of the samples accessed from the informal seed sources. Previous studies found significant differences in the purity of bread wheat seed samples collected from different sources in Ethiopia [28]. This might be attributed to the different ways that farmers used to produce, select, save, and acquire wheat seeds. Poor-quality seeds have a significant impact on market value [30].

3.5 TSW

There was a highly significant (P = 0.000) difference between the formal and informal seed sources for TSW (Table 2). The mean percentages of TSW were 38.55 and 36.56 g for samples collected from the formal and informal seed sources, respectively. Seed with a higher TSW has better milling quality and ensures better emergence [30,31]. Previous studies in Ethiopia showed significant differences in TSW among wheat varieties collected from formal seed sources and farmers [32]. A grand mean TSW value of 35.34 g, with minimum (25.08 g) and maximum (49.87 g) values, has already been reported from wheat seed samples collected from different wheat-producing areas of Ethiopia [28].

3.6 Seed moisture content

The moisture content (MC) of seed samples collected from formal and informal seed sources differed significantly (P = 0.001) (Table 2). On average, 12.74% of the MC was recorded from formal sources and 12.24% from informal sources. The values obtained from the formal and informal seed sources were in the acceptable range of MC (13%) for the wheat seed quality specification set by ethiopian standards authority (ESA) [28]. The high MC could result from poor pre-harvest seed management practices and post-harvest handling during processing and particularly during storage conditions. It has been reported that changes in seed MC of common bean are associated with the management practice [33].

3.7 Standard germination

Table 3 presents the average values of germination percentage, abnormal seedlings, and dead seeds. Highly significant (P = 0.000) differences were observed between the two sources in the proportion of normal, abnormal seedlings, and dead seedlings. The mean percentages of germination (normal seedlings) were 90.07 and 82.13% for seed obtained from formal and informal seed sources, respectively. This indicates that the formal sources contain higher-quality seed than the informal sources. High-quality seed germinates earlier and thus produces larger roots and shoots than low-quality seed [34].

In terms of abnormal seedlings, the highest mean percentage (11.47%) was recorded from informal sources and lowest (7.39%) from the formal sources. Similarly, the highest mean percentage (6.43%) of dead seed was recorded from an informal source, while the lowest mean (2.53%) was recorded from a formal source. These are important variables that greatly contribute to the germination capacity of a given seed [33].

3.8 Seedling vigor test

To assess the vigor of bread wheat seed collected from the two seed sources, physiological tests such as speed of germination, seedling shoot length, seedling root length, seedling dry weight, VIG1, and VIG2 were performed (Table 4). Highly significant (P = 0.000) differences were observed between the two groups for seedling shoot length, speed of germination, VIG1, and VIG2, and significant differences (P < 0.05) for seedling root length and seedling dry weight. The highest seedling dry weight (0.27 g) was observed on the seed samples collected from the formal seed sources.

Speed of germination was conducted to distinguish the variation in seedling size and vigor within the category of “normal seedlings.” The presence of highly significant (P = 0.000) variation for speed of germination between the two groups indicates that the seed samples accessed from the formal seed sources have higher germination rates and more vigor than the seed samples obtained from the informal sources. This is supported by the fact that “speed of germination” is the rate at which the seed germinates, and seedlings with a higher index or higher on first count are expected to show faster germination and seedling emergence, as well as to escape adverse field conditions than seedlings with a lower index [28]. The samples collected from the informal seed sources have shown low vigor, which is in line with the findings of the previous studies [29]. Seed with a low germination rate can have a negative effect on the performance of the crop as it is apparent that the seed will not germinate properly [29]. Both VIG1 and VIG2 also showed a highly significant (P = 0.000) difference between the formal and informal seed sources. Similar results were reported from sorghum. The difference in physiological quality of sorghum seed from different sources may be attributed to differences in storage age and temperatures, seed moisture content in storage, and growers’ level of awareness of seed production and handling [35].

3.9 Phytosanitary test

The study identified the presence of seed-borne fungi pathogens in wheat seed samples collected from both formal and informal sources (Table 5). These include Aspergillus flavus, Aspergillus niger, Chaetomium spp. (Chaetomium funicola, Chaetomium globosum), Alternaria alternata, Penicillium spp. (Penicillium chrysogenum, Penicillium verrucosum), and Fusarium spp. (Fusarium culmorum, Fusarium graminearum). The differences between the formal and informal seed sources were significant for Aspergillus flavus, Aspergillus niger, Chaetomium spp. and Fusarium spp., and the differences were non-significant for Alternaria alternata and Penicillium spp. Seed samples collected from formal sources were highly affected by Aspergillus flavus, Aspergillus niger, Chaetomium spp., and Alternaria alternata, whereas seed samples collected from informal sources were highly affected by Penicillium and Fusarium. In terms of magnitude of effect, Aspergillus flavus has the highest effect, followed by Fusarium and Aspergillus niger. Previous studies identified different Fusarium spp. and several saprophytic fungi (Alternaria spp., Aspergillus spp., Cladosporium spp., Curvularia spp., Mucor spp., Nigrospora spp., Penicillium spp., and Trichothecium spp.) from wheat seed samples collected from farmers in the major bread wheat-growing areas of Ethiopia [36]. The higher incidence and association of fungal species with bread wheat seed samples agree with the results of recent studies on common bean and other grain crops [27]. For example, a study conducted on the quality assessment of common bean seed from formal and informal sources in the Central Rift Valley of Ethiopia indicated that Fusarium oxysporum and Aspergillus spp. were the two most common fungal diseases associated with seed retained by farmers from local markets and cooperative unions [37]. The presence of these important fungi species is primarily due to the management conditions in both seed sources. These conditions include the proper management of critical agronomic practices and post-harvest management such as seed drying mechanisms, threshing, and storage facilities.