Farmers’ varieties are important crop genetic resources and are valued by plant breeders and farmers because of diversity (heterogeneous population), rarity (embodying unique traits) and adaptability (exhibiting wide ecological and socio-cultural adaptation) (Brush and Meng, 1998; FAO, 1998; Smale, 2006). Farmers throughout the world continue to maintain and manage farmers’ varieties within their production systems (Hawkes, 1971; Duvick, 1984;
Brush et al., 1995; Brush, 2004; Jarvis et al., 2008; FAO, 2010). Yet the value they contain for the farming communities that maintain them has not been fully capitalized on.
Not all landraces and improved varieties are equally valued by farmers for yield and yield related components and seed quality attributes. Some landraces are adapted to marginal ecosystems (Vandermeer, 1995; Bezancon et al., 2009; Barry et al., 2007; Rana et al., 2008) or have cultural, religious, or nutritional values (Rana et al., 2007; Sthapit et al., 2008; Johns and Sthapit, 2004). Some landraces maybe highly valued but their use is constrained by poor access to quality and quantity of seeds for planting (Tripp, 2001; Almekinders et al., 2006; Sperling, 2008; Hodgkin et al., 2007). Landrace populations may, themselves, not be uniform in their adaptive or quality traits, having significant variation both within and amount populations (Teshome et al., 2001; Harlan, 1975; Mariac et al., 2006; Barry et al., 2007).
One way of distinguishing farmer’s varieties that provide high public value is to classify those in terms of their immediate and future plant breeding benefits (Smale et al., 2004). This required consultation with farmers and breeders but also other concerned including consumers, wholesalers and retailers (Sthapit et al., 2001; Sperling et al., 2001; Bellon et al., 2003; Witcombe et al., 2005).
The recent approach in participatory and decentralized plant breeding over the last decade has shown that improving varietal performance in low input systems can help improve local livelihoods (Ceccarelli et al., 2000; Smith et al., 2001; Almekinders et al., 2006; Zeven, 2000; Dawson et al., 2008). Varieties can be improved by selection of preferred traits from the heterogeneous populations, collected locally before any crop improvement programme is initiated. However, insufficient attention has been given to the potential use of the existing landrace variability in production systems to provide direct benefits to local communities (Sthapit and Rao, 2009).
Damoy is considered nutritious and is preferred and selected for medicinal value believed for healing broken bone and for women during childbirth. Garbuu Adii is preferred for porridge during Meskel festival and for kincheatetee (facasa) for the blessing of God for an Ethiopian new year. Garbuu Gurracha preferred for socio-cultural (for labor sharing) values and for its high yield. Damoy and G/gurracha selected for their early maturity and malt for local beverage.
Variability in grain quality of landrace is the main concern for traders and consumers in marketing it. Consumers are willing to pay a high price for its purchase, but the landrace has a problem with quality variation. Therefore, this study was to assess the one year data for performance between farmers’ and improved
varieties of barley for some yield and yield related traits and seed quality attributes.
Description of the study sites
The experiment was conducted at Degem Wereda, North Shewa Zone of Oromiya (Figure 1) selected for the following reasons: (i) Barley was the dominant crop in the area; (ii) Some FV’s were still being grown; (iii) Less government attention to conservation of FV’s of barley, and (iv) There was no similar study that could be used as a baseline in the area. This assumption emanates from the fact that the prevailing networks in the farmers’ seed system had been highly influenced by the formal seed system.
Field experiment
Experimental design and treatments
Three representative peasant associations (PAs), namely; Anno Degem, Anno Qarree and Tumno Abdii were selected for the field experiment. The experimental design used was randomized complete block design (RCBD) with three replications. Ten food barley varieties were included in the field experiment, that is, five FV’s currently under production (Damoy, G/adii, Tolasee, G/gurracha and Magee), two lost FV’s (Hadho and Karfee) and three improved varieties (HB42, HB1307 and Shagee) (Table 1). Seeds were planted at a seed rate of 30 g plot-1 in a plot size of 3 m2 (six rows of 2.5 m long spaced 0.2 m apart between rows).
The experiment was planted on May 23 and 24 /2010. The first weeding was carried out thirty days after crop emergence and the second weeding was performed thirty days after the first weeding. Data were collected on 11 developmental and yield related traits (Table 2).
Seed quality analysis
Barley seed obtained from experimental plots was used to make seed quality analysis. The experiment was conducted in a Complete Randomized Design (CRD) with four replications as per ISTA methods (1996).
Determination of physical seed quality (Experiment-1)
Analytical purity test: From submitted sample of 120 g two replicates of 60 g was analyzed (ISTA, 1996). The samples were divided into three; (I) pure seed, (II) other crop seed, and (III), inert matter. The components were weighed on precision balance to the nearest two decimal places and the percentage of each component was determined (ISTA, 1996).
The percentage by weight of each component fraction was calculated by the following formula (ISTA, 1996):
Thousand kernel weight (TKW): Eight replicates of 100 seeds each were weighed from pure seed fraction (ISTA, 1996).
The coefficient of variation was calculated to assess the acceptability of the test and the thousand kernel weight was calculated (Bishaw, 2004).
Determination of physiological seed quality (Experiment-2)
Several physiological tests such as standard germination, speed of germination, seedling shoot length, and seedling root length and seedling dry weight were measured to assess the vigour of barley seed from field experiment.
(a) Standard germination (SG) test: SG test was done for all seed samples obtained from all treatments. Four hundred (400) seeds of the pure seeds component were divided into four replicates of 100 seeds each in germination box size of 18, 9 and 13 cm (length, height and width, respectively), which were then sown in sterilized sand media. The planted seeds were incubated at a temperature of 20°C for 7 days as specified by International Seed Testing Agency (ISTA, 1996). At the end of the incubation period the germination boxes were removed and the seedlings were evaluated. Germinated seedlings were divided into (I) normal seedlings, (II) abnormal seedlings, and (III) dead seeds to determine the percentage of different seedlings.
(b) Seed vigour test: Seed vigour is an important quality parameter which needs to be assessed to supplement germination and viability tests to gain insight into the performance of a seed lot in the field or in storage.
Seedling shoots and root length: The seedling shoot length and seedling root length were assessed after the final count in the standard germination test. Ten normal seedlings were randomly selected from each replicate after 7 days of seed sowing. The shoot length was measured from the point of attachment to the cotyledon to the tip of the seedling. Similarly, the root length was measured from the point of attachment to the cotyledon to the tip of the root. The averages shoot and root lengths were computed by dividing the total shoot or root lengths by the total number of normal seedlings measured (ISTA, 1996). Varieties producing the taller seedlings were considered more vigorous than the varieties producing shorter seedlings.
Seedling dry weight: Ten randomly selected seedlings from each replicate were cut from the cotyledons and placed in envelops to be dried in an oven at 80°C for 24 h. The dried seedlings were weighed to the nearest milligram using sensitive balance and the average seedling dry weight was calculated. The seedling dry weight provides additional information for assessing seed vigour.
Vigour Index-I and Vigour Index-II: For each sample, two vigour indexes were calculated. Seedling vigour index-I was calculated by multiplying the normal germination percentage with the average sum of shoot and root length after seven days of germination and vigour index-II were calculated by multiplying the standard germination with mean seedling dry weight.
Vigor Index I=Standard germination (%) × Seedling length (cm)
Vigor Index II=Standard germination (%) × Seedling dry weight (mg)
Speed of germination: 100 seeds were replicated into four from each sample and kept at 20°C for maximum of 7 days in an incubator. Each day, normal seedlings were removed. Then speed of germination (GS) was calculated as follows (Maguire, 1962):
The varieties having greater germination index were considered more vigorous.
Seed health testing (Experiment-3)
The agar plate method: About 15 ml of sterilized medium was poured in each Petri dish (9 cm, Pyrex, USA) under aseptic conditions under micro flow. Four hundred seeds from each sample were surface sterilized in 1% NaOCl for ten minutes, rinsed with sterilized water, and then 20 seeds were plated in each Petri dish. The set was incubated at 22 ± 2°C for 12 h of alternating cycles of day and night under fluorescent light (Anon, 1996). Colonies and fruiting bodies of the fungi were identified using stereo and compound microscopes with aids of appropriate reference materials.
Mycological evaluation: Mycoflora associated with barley seed were detected by standard methods (Anonymous, 1996). The presence and type of fungi were determined according to their development on the seed, which were incubated on Potato Dextrose Agar (PDA) medium. Seeds were examined by Binocular microscope, Compound microscope and CIMMYT Manual for Detection of Seed-borne Microorganisms and Descriptions (1998). Fungi appearing on Petri plates were directly identified up to the species level with the help of a compound microscope and relevant literature (Booth, 1971; Ellis, 1976; Sutton, 1980; Nelson et al., 1983; CIMMYT, 1998). Percent incidence of fungi was recorded. Data collected on incidence of fungi were analyzed statistically.
Data coding and entry
Semi structured and structured questions were properly coded and entered into the computer using Microsoft Excel Application Program. Quantitative data were organized to suit the different statistical packages used in the analysis. For better interpretation of results, some of the data sets were transformed into standard units. Qualitative data were organized in such a way that cumulative of the respondent’s information was presented.
Data analysis
Data collected from the field and laboratory experiments were subjected to analysis of variance using software SAS version 9 and Genstat discovery edition. Treatment means were separated using Least Significance Difference (LSD) test and Duncan’s Multiple Range Test (DMRT).
Performance of improved and farmers’ varieties for yield and yield components
The performance of ten food barley varieties for various phenological and agronomic traits is presented subsequently.
Days to heading and maturity
Both days to heading and days to maturity differed significantly among varieties. Days to flowering ranged from 91 to 109 days with a mean value of 102 days. Days to maturity ranged from 115 to 137 days with a mean value of 132.15 days. Damoy was the earliest variety to heading (91 days) and maturity (115 days) while Magee took longer days to heading (109) and maturity (137 days). Hadho, G/gurracha and Karfee were medium maturing varieties. Variety Damoy was the dominant variety grown during the belg season where rainfall variability is high. The variation for both phenological traits was high among the test varieties and this was more apparent on the FV’s compared to improved varieties which all were late maturing (Table 3).
Grain filling period and plant height
Grain filling period ranged from 23.88 to 34.22 days with mean value of 30.11. The shortest grain filling period was observed on Damoy and the longest on G/adii (Table 3).
Varieties with the shortest grain-filling period had the advantage to escape terminal moisture stress and good character to cope up with the rainfall variability in the highlands of Degem.
Variation for plant height among the varieties ranged from 94.9 to 111.1cm with mean site value of 104.75. Karfee was the shortest variety (94.9 cm) whereas G/gurracha and Magee (111.5 cm) were the tallest varieties. Most of the FV’s were tall with weak stem, which was the common character of the local FV’s. The improved varieties were relatively short with strong stem. Fekadu (2010) also reported that modern varieties are shorter in plant height compared to the FV’s.
In contrast to our findings, Martiniello et al. (1987) reported that modern genotypes showed trends towards earliness in both six and two row barley genotypes compared to landraces. It was reported that most modern barley varieties were relatively earlier than the landraces whereas maturity time was similar among all varieties (Wych and Rasmusson, 1983). Furthermore, similar observation was also reported in hard red winter wheat in USA (Cox et al., 1988), in spring wheat in Australia (Perry and D’Antuono, 1989), in wheat in UK (Austin, 1999) and in winter wheat genotypes in Great Plains (Donmez et al., 2001) that modern varieties reached flowering and maturity earlier than the older ones. Metzger et al. (1984) indicated that selection for grain filling duration was not promising to improve yield of barley in Minnesota, suggesting that grain-filling duration is not yield limiting factor in barley.
Spike length and number of spikletes per spike
The spike length ranged from 4.6 to 6.7 cm with mean value of 6.2 cm (Table 4). The total variation among the varieties was small and almost most of the varieties had spike length of 6.1 to 6.7 cm. Among the varieties, Karfee had the shortest spike length (4.6 cm) while G/adii and Tolasee were the longest (6.67 cm). The mean difference between the number of spikletes per spike varied from 40.8 for HB-1307 to 51.3 for Tolasee with the mean value of 47. The number of spiklets per spike for most of the varieties was in the range of 45 to 51 cm. Although, no significant difference was observed among varieties on number of spikelets per spike, kernel weight, spikelets per spike and spike length are observed as the main components of yield (Table 4). The morphology of spike is a major concern in crop improvement. Similarly, Bensemane (2011) reported that the yield components, spikes per square meter followed by kernels per spike exerted the greatest effect on grain yield. Similarly, Sinebo (2002) reported that barley spike is a source and sink of assimilates that ultimately determines grain yield.
Biomass yield, grain yield, thousand kernel weight and harvest index
Mean differences were observed among the barley varieties for biomass and grain yield, but not for thousand kernel weight and harvest index (Table 4). Biomass yield ranged from 3711.4 for Karfee to 8058.6 kg ha-1 for HB-1307 with mean value of 5883.3 kg ha-1 (Table 4).Varieties such as Magee, Tolasee and G/adii was on par in biomass yield. The variation for grain yield per hectare ranged from 1489.2 for Karfee to 3573.6 kg ha-1 for G/gurracha 2603.6 kg ha-1 (Table 4).
The site mean value of grain yield was 2554.6 kg ha-1 at Anno Degem, 2282.4 kg ha-1 at Anno Qarree and 2973.8 kg ha-1 at Tummano Abdii. Grain yield potential of Anno Degem was intermediate while that of Tummano Abdii was relatively higher than both locations.
The low yield obtained at Anno Qarree and high yield at Tummano Abdi could be attributed to low and high soil fertility respectively (Table 5).
Garbuu gurracha gave the highest grain yield followed by the recently released stiff straw variety HB-1307. Grain yield performance of HB-42 and Shagee varieties was lower compared to other varieties except Damoy, which was the least in grain yield (Table 4). Harvest index ranged from 39 to 48% ha-1 with mean value of 44%. Table 4 indicates that majority of the varieties had harvest
index, between 40 and 48%.
No significant variation was observed among varieties for spike length, harvest index and thousand kernel weights. Number of spikelet per spike ranged from 42.2 to 51.1 with site mean
value of 47.0 (Table 4).
Performance of improved and farmers’ varieties of barley for yield and yield components at each location
At Anno Degem, the analysis of variance revealed highly significant differences (P<0.001 or P<0.01 or P<0.05) among the varieties for all the traits measured, except for TKW. Plant height was significantly different at P<0.05, while grain filling period was significantly different at P<0.01. Moreover, most yield and yield related traits showed highly significant at P<0.0001. However, number of spikelet per spike and thousand kernel weights were not significant.
The analysis of variance carried out at Anno Qarree on the eleven quantitative traits revealed highly significant differences among the varieties for most of the traits measured. Days to heading, grain filling period, biomass production rate, and biomass yield were significantly different at P≤0.01. Days to maturity and grain yield were highly significant different at P<0.0001. However, spike length, plant height number of spikelet per spike, harvest index and thousand kernel weights did not show any significant differences among the varieties.
At Tumano, the analysis of variance revealed significant differences among the varieties for most of the traits, except for spike length, number of spikelts per spike and thousand kernel weights. Plant height was significantly different at P≤0.01.
The other morphological and yield related components showed highly significant differences at P<0.0001.
Genotypic performance for phenological traits at each site
Mean phenotypic variation for phenological/ developmental/, of the 10 food barley varieties are shown in (Table 5). The amount of variation among the varieties for most of phenological traits was relatively high between the testing sites. At Anno degem, days to heading ranged from 87 to 106 days while at Anno qarree from 96 to 110 days and at Tumano it ranged from 91 to 111 days. The overall mean difference in DF among varieties were small (100, 103 and 103 days for Anno degem, Anno qarree and Tumano, respectively).
At Anno degem, plant height ranged from 90.7 to 108.7 cm with site mean of 101.4 cm while at, Anno qarree from 87 to 111 cm with site mean of 100.9 cm, and at Tumano from 100.7 to 121 cm with site mean value of 112 cm (Table 5). The variation in plant height across the three sites ranged from 100.9 to 112 cm, at Anno qarree and Tumano.
Mean days to grain filling period across the sites was 31.6 days at Anno degem, 29.4 days at Anno qarree and Tumano. The highest grain-filling period was observed by variety HB 42 (37 days) at Anno degem followed by variety Shagee (37 and 35 days) at Anno qarree and Tumano, respectively. Generally, FV’s showed relatively short days to flowering and days to grain filling than the improved varieties (Table 5).
Agronomic / yield component traits / of food barley
Mean performance for agronomic/ yield component/ trait
of 10 food barley varieties are shown in Table 6. The amount of variation among the varieties for most traits was relatively high among the testing sites. At Anno Degem biomass production rate (BMPR) ranged from 38.7 to 79.6 days while at, Anno qarree from 24.0 to 80.2 days and at Tumano from 39.2 to 90.8 days. There were no significant difference in overall mean among varieties in BMPR at Anno degem and Anno qarree, however, significant different was observed at Tumano site which was (65.3).
At Anno degem, spike length ranged from 2.7 to 8 cm with mean of 6.3 cm at Anno qarree from 5 to 6.7 cm with mean of 5.6 cm, at Tumano from 5 to 7.7 cm with site mean value of 6.6 cm (Table 6).
At Anno degem, number of spikelet per spike ranged from 42.7 to 55 with mean of 50.9 while at Anno qarree from 38 to 49 with mean of 44 and at Tumano from 35.3 to 52.3 with site mean of 46.1 (Table 6). The variation in number of spike per spikelet was relatively high across the three sites, it ranged from 44 to 50.9 at Anno qarree, and Anno degem.
Grain yield, biomass yield, thousand kernel weight and harvest index at each location
Mean values for grain yield, biomass yield, thousand kernel weigh and harvest index of each genotype and testing sites are shown in Table 7. At Anno degem, the grain yield ranged from 1722.2 kg ha-1 for variety Shagee to 3437.5 kg ha-1 for variety G/adii followed by 3395.8 kg ha-1 for variety Hadho while at, Anno qarree it ranged from 1037 kg ha-1 for Karfee to 3777.8 kg ha-1 for G/gurracha followed by 3596.3 kg ha-1 for HB1307. At Tumano grain yield varied from 1666.7 kg ha-1 for variety Karfee to 4244.4 kg ha-1 for variety G/gurracha followed by 3895.8 kg ha-1 for variety HB1307.
Location mean grain yield ranged from 2282.4 kg ha-1 at Anno qarree to 2554.6 kg ha-1 for Anno degem (Table 7). Variation for grain yield across the three sites was relatively high and ranged from 1037 kg ha-1 produced by Karfee at Anno qarree to 4244.4 kg ha-1 produced by G/gurracha at Tumano. Except HB 42, Damoy and Karfee, all varieties showed similar yield performance with HB-1307 (Table 7).
At Anno qarree mean grain yield ranged from 1037 kg ha-1 for variety Karfee to 3777.8 kg ha-1 for variety G/gurracha. Varieties Hadhoo, HB42, Damoy, Shage and Tolasee showed similar performance with an improved variety HB1307 and significantly higher than varieties G/adii and Karfee (Table 7). Moreover, the lowest and highest grain yield at Anno qarree was 1037 and 3777.8 kg ha-1 for varieties Karfee and G/gurracha, respectively. G/gurracha variety had significantly higher grain yield than the other varieties except at Anno degem, G/adii had significantly higher grain yield than the other local and improved varieties. This study showed that FV’s of food barley had good yield performance than improved varieties on low input farmer’s condition (Table 7). Similarly, Jalata (2011) reported that there was a differential yield performance among genotypes across testing environment mainly due to genotypic, genotypic and environmental interaction. Similarly, Sinebo (2002) reported that high straw yield was essential to high grain yield production.
Significant mean biomass yield differences among the varieties were observed at each site (Table 7). At Anno degem, mean biomass yield ranged from 3958 kg ha-1 for variety Damoy to 7968.1 kg ha-1 for variety Hadho followed by varieties G/adii and HB1307 (7958 kg ha-1). Likewise, at Anno qarree mean biomass yield ranged from 2870 kg ha-1 for Hadho and karfee to 7777.8 kg ha-1 for G/gurracha followed by HB1307 (7759 kg ha-1).
At Tumano, mean biomass yield ranged from 3958.3 kg ha-1 for Karfee to 8951.4 kg ha-1 for G/gurracha followed by variety HB1307 ( 8458 kg ha-1). Location mean biomass yield for Anno degem Anno, qarree and Tumano were 6155.8, 4743.5 and 6750.7 kg ha-1, respectively. Varietal difference for biomass yield was less marked at Anno qarree than the two sites, which might be due to better adaptation of all varieties to the growing conditions at Tumano, and Anno degem. Variety G/gurracha which gave the highest mean grain yield at Anno qarree and Tumano was also characterized by high biomass yield, 7777.8 and 8951 kg ha-1 at the two sites respectively. Moreover, the lowest mean biomass yield was obtained by varieties Hadho and Karfee (2870 kg ha-1) at Anno qarree and the highest biomass yield was obtained by variety G/gurracha (8951.4 kg ha-1) at Tumano. This result revealed that biomass yield was relatively related to grain yield for FV’s except for HB-1307.
Similarly, Ortiz et al. (2002) reported that there was significant trend in increasing straw yield in Nordic spring barley germplasm whereas the biological yield of recent varieties and old varieties was almost the same. Fekadu (2010) also reported that the recent varieties (Dimtu and HB-1307) which gave high grain yield were also characterized by high biomass yield at all locations except HB-1307 at Holetta. Similarly, Sinebo (2002) reported that high biomass yield is essential to high grain yield production.
At Anno degem, the mean harvest index ranged from 0.36 for HB-42 and HB-1307 to 0.46 for Damoy while at Anno qarree, it was 0.35 for Karfee to 0.54 for Hadho and G/gurracha and that of Tumano was 0.37 for Hadho and HB-42 to 0.49 for Damoy. The highest and lowest varietal difference for harvest index was observed at Anno qarree than Anno degem and Tumano. The overall mean harvest index in this study varied from 0.35 for karee to 0.54 for Shagee and G/gurracha indicating FV’s had shown significant variation for harvest index (Table 7).
There was no significant variation in thousand-kernel weight at Anno degem, however, at Anno qarree ranged from 39 to 48.2 gm with mean of 44.3 gm and at Tumanno ranged from 37.5 to 48.4 gm with mean of 44.24 gm. In this study, the effect of thousand-kernel weight is small for performance evaluation among the varieties tested. Similarly, Sinebo (2002) reported that the effect of kernel weight on grain yield was small.
Similarly, Hockett (2000) reported that mean values of the varieties were recorded within the range of barley yield potential performance study as described for yield potential of barley varies from state to state in America and ranged from 1451.0 to 5499.0 kg ha-1 and barley yield of ten top leading producing countries was within the range of 1730.0 to 5470.0 kg ha-1.
Comparative yield potential of farmers’ varieties and improved varieties over three sites
There were significant difference in performance of local against improved varieties in yield and yield related traits on low input and marginal environment. Accordingly, G/gurracha was the highest in grain yield followed by Tolasee and Magee among the FV’s. Ceccarelli and Grando, 2001 cited in Brush (2000) reported that black-seeded FVs’ was better adapted to dry areas and less vigorous in early growth, more cold resistant and more productive under stress than improved cultivars. However, HB-1307 gave high yield (3446.5 kg ha-1) than other improved varieties and FV’s except G/gurrach that gave 3573.6 kg ha-1. High biological yield was obtained on HB-1307 (8058.6 kg ha-1) and followed by FV’s G/gurrach, Tolasee, Magee and G/adii. On the contrary, high TKW was observed by variety Magee followed by HB-1307. Karfee was the lowest in harvest index and Damoy was the highest in harvest index. In this study, Karfee was the lowest performing FV’s. In contrast, HB-1307 gave relatively better yield and yield related traits over the other two improved varieties and the FV’s except G/gurrach, which was superior in some traits (Table 8).
Correlation among the quantitative traits
The correlation coefficient among the quantitative traits was computed on the mean trait values of the two sites, that is, Anno degem and Tumano (Table 9). Anno qarree was excluded because of high CV value (>30%). Significant and strong positive correlation coefficients were found for DH with DM (0.88), BMPR with BY and GY (0.97 and 0.98, respectively) and BY with GY (0.97). Days to heading was positively correlated to grain filling period. On the other hand, days to grain filling period was negatively correlated to number of spikelet per spike at (r=-0.64) (Table 9). Similarly, (Ahmad, 2004; Tarekegn, 2009) reported that days to heading were positively and significantly correlated with days to maturity. Moreover, biomass yield showed significant correlation with grain yield. Sinebo (2002) reported that grain yield was correlated positively with mature heights, and grain-filling duration. On the contrary, grain yield was not correlated with kernel weight.
This finding suggests that characters showing positive correlation could effectively be utilized in improving FV’s. The tendency of positive correlation among developmental traits, in spite of wide range of genetic diversity in FV’s could effectively be utilized to improve barley FV’s.
Seed quality analysis
Seed testing evaluates seed lot quality and is essential for both seed production and commercial seed transactions (AOSA, 1981). Seed testing is done to assess seed lot attributes to determine overall quality and value for production and storage. Seed testing standards provide set of procedures to conduct tests in a uniform manner to ensure comparable results that are within acceptable ranges (ISTA, 2005).
Analytical purity analysis
The analytical purity results of seed samples collected from field experiment were ≥ 98.17% which was greater than 85% of the national standard for commercial seed in Ethiopia (ICARDA, FAO, AAR INENA and CIHEAM, 1999). Physical purity, however, ranged from 98.17 to 99.76% with an average of 99.38%. There was no difference in physical purity level between seed sources. However, varieties showed significant difference for other crop seed, which ranged from 0.14 for G/adii to 1.12% for Karfee with mean of 0.41%. This variety contains more other crop seed
than other varieties. Thousand seed weight also showed significant difference which ranged from 30.7 to 46.25 g with mean of 40.58 g. Damoy was the least in thousand seed weight followed by Karfee, which was 34.8 g. On the contrary, the highest seed weight was observed on Magee 46.8 gm followed by HB-1307 and G/adii (Table 10).
The presence of different “other seeds” species in the seed samples was identified. Major other crops seed observed were Avena spp., Lolium, Triticum and Bromous spp. Avena spp. and Bromous spp. were the major problematic weeds in Degem wereda. Significant (p<0.0001) mean differences were observed on other crop seeds between varieties. However, all local and improved varieties maintained the minimum purity standards. Moreover, all samples had less inert matter contamination than the standard (Table 10).
Thousand kernel weight
The overall average of thousand-kernel weight was 40.5 g ranging from 30.7 g for Damoy to 46.8 g Magee. Significant difference in seed weight was observed among different varieties at (p<0.0001) (Table 10). Seed sample of Magee were on average heavier than the grand mean and ranked first followed by HB-1307. This study revealed that except HB-1307, improved varieties had low 1000 kernel weight than the local ones. These results agreed with Sahilu (1999) and with Briggs (1978) in which they recorded that Ethiopian barleys were characterized by very high seed weights.
Physiological seed quality
Standard germination: Physiological seed quality analysis was conducted at different time. This was because of the appearance of seed dormancy on different varieties. The first laboratory analysis was conducted one month after harvest and the second was four months after harvest.
During the first germination test, there were high standard germination mean differences, which ranged from 5.25% for Magee to 98.25% for Damoy. Four varieties Damoy, HB-1307, G/gurracha and Karfee had germination percentage of 98.25, 97.75, 93.5 and 88.0%, respectively. Although these varieties were in the range of barley germination standards, the rest of the varieties were out of the standards due to seed dormancy (Table 11). Hence, further evaluation was conducted to capture this quality trait.
Accordingly, the second germination testing was made after four months of harvest. During the second germination test, mean differences for the trait ranged from 98.5% for Tolese and HB-42 to 99.75% Shagee. The result indicated that all varieties were above nationally recommended standard germination range after breaking dormancy following four months of storage.
Significant in seed germination differences were observed between the first and the second time of analysis (Table 11). This study revealed that seed dormancy was broken after-ripening (dry storage) for four months although varieties had different period of seed dormancy (Figure 2).
Hull thickness has an effect on barley seed germination. Those varieties showed seed dormancy during germination test after one month of harvest have thick hull. At harvest, intact barley grains exhibited dormancy and germinate poorly. Bradrord et al. (2008) reported similar result that restriction to germination is attributable to the presence of the glumellae or hulls, as their removal or excision of the embryos greatly improves germination capacity. The restriction of germination by the hull has been attributed largely to its ability to reduce the availability of O2 to the embryo (Lenoir et al., 1986). Bradford et al. (2008) reported that germination rates and percentages improved following after breaking dormancy for both seeds and embryos. This finding was supported by five months of after dormancy break allowed almost complete germination of seeds in 21% O2, but germination remained highly sensitive to O2 percentage as (Bradrord et al., 2008).
Speed of germination: Mean difference in speed of germination ranged from 18.47 for HB-1307 and Hadho to 22.86 for G/gurracha. The others were in par for this trait. Least mean shoot length was observed on Magee while, Hadho was the highest in shoot length. Likewise, G/gurraha was the shortest in root length but G/adii was the longest (Table 11). The speed of germination measures the rate at which the seeds were germinating and those seedlings with higher index or highest germination were expected to show rapid germination and seedling emergence and to escape adverse field conditions.
Seedling shoots and roots length: Mean difference in seedling shoots length ranged from 13.15 cm for Tolasee and 16.69 cm for Hadhoo with grand mean of 15.19 cm. Moreover, Hadhoo, HB-1307 and Damoy were in par for this trait. Likewise, mean difference in root length ranged from 10.22 cm for G/gurracha to 13.59 cm for G/adii with grand mean of 12.25 cm. Hadhoo, HB42 Damoy and Karfee were in par for the trait considered (Table 11). HB1307, G/gurracha, Hadho, Damoy and Karfee showed longer in their shoot length and G/adii, Hadho and Karfee showed longer in their root length. Significant variations were observed among varieties for seedling shoots and root length at (p<0.0001). Seedlings with well-developed shoot and root systems would withstand any adverse conditions and provide better seedling emergence and seedling establishment in the field.
Seed vigour test (Viguor Index I and Viguor Index II):
Mean difference in vigour index I ranged 2462.35 for Tolasee to 2972.28 for Hadhoo with grand mean of 2117. G/adii, HB1307, HB 42 and Karfee were in par for this trait. Hadho, Damoy and Kerfee showed higher vigor index I while Tolasee, Magee and G/gurracha were lower in vigor index I (Table 11). Significant variations were observed among varieties for vigour index I at (p<0.0001). On the other hand, no significant difference observed among varieties for vigour index II at (p<0.05).
Varieties that had higher speed of germination were generally considered more vigorous. Moreover, vigorous varieties could be stored for longer periods without loss of germination. Vigor tests measure the potential for rapid, uniform emergence of seeds under a wide range of field conditions (Elias et al., 2010). Its results may be more closely associated with field emergence than the standard germination test (Elias et al., 2006).
Mycological evaluation
The presence and type of fungi were determined according to their development on seed, which had been incubated on Potato Dextrose Agar (PDA) medium. Accordingly, sixteen different fungi genera were identified. Their importance, in relation to their occurrence were Helmithosporium sativum, Helminthosporum teres, Fusarium graminarum, Fusarium oxysporum, Fusarium avenacerum, Cladosporium spp., Alternaria spp., Botryodiplodia spp., Phoma sp., and Stemphylium sp.. In addition, storage fungi associated to barley seed were also identified, which include Penicillium spp., Trichoderma sp., Aspergillus spp. and Chaetomium funicola (Table 12).
The result indicated that high incidence of facultative fungal such as, Fusarium avenacerum and Phoma spp. and low incidence of field fungi like Botryodiplodia, Aspergillus, and Trichoderma. Fusarium spp were commonly occurring as seedborne fungi and responsible for causing seed decay and seedling mortality. Likewise, some diseases were also observed in large number on different varieties. This showed that there was differential occurrence of some fungi on varieties tested. The mean difference between varieties showed that karfee (3.52%) was highly infected followed by Damoy (2.44%) while G/adii (0.98%) was the least affected (Table 12). HB-1307, HB-42, Damoy and G/gurracha were more infected by H. sativum, while Shagee, G/gurracha and Magee were more infected by F. oxysporum. Damoy, Hadhoo, Tolasee and Magee were more infected by F. avenacerum (Table 12). In general significant different was observed between varieties for different seed-borne fungi. Similarly, Bekele et al. (2005) reported that most of the FV’s were found susceptible to scald (Rhynchosporium secalis Oud.), net blotch (Helminthosporium teres Sacc.), spot blotch (H. sativum Pum.), leaf rust (Puccinia hordei Otth.) and lodging. Novak et al. (2001) also reported that difference in disease incidence were due to the ability of saprophytes to colonize, rapid germination of spores, quick hyphal invasion, high competitive nature, their ability to utilize a wide variety of substrates and their nutrient composition.
Among sixteen seed fungi observed, eight of them were known to be seed transmitted (Table 13). Seed transmitted pathogens are pathogen that can affect the yield in next production season and transmit pathogen to disease free environment. Therefore, to increase the planting values of the seed, treatment is important to produce healthy seed for the next planting season.
Significant different in seed infection was observed on varieties. High H. sativum was observed on G/gurracha, HB-1307, HB-42, and Damoy respectively.
H. teres was more observed on Damoy followed by HB-42. More F. graminarum seed infection was observed on Shagee followed by Magee and HB-1307and F. oxysporum was observed on Sahgee, Magee and Tolesee. Fusarium avenacerum was more observed on Hadhoo, HB- 42 and Magee (Table 13).
On the other hand, seven of the fungi were seed deteriorating pathogen (Table 14). These seed deteriorating fungi were affecting the planting value of the seed due to their deteriorating action. Seed infection was significant at (p<0.001) except for Penicillium, and
Hence, appropriate seed handling should be made during pre-harvest and post-harvest operations to ensure quality seed system.
Heleminthosporium spp, H. teres, Rhynchosporium, Ustilago nuda, U. hordi, H. graminea are common barley seedborne disease in Ethiopia. Semeane et al. (1996) reported that net blotch, scald and leaf rust are the most three important barley diseases targeted for control. Of those, net blotch was an endemic disease in most highlands of the country where barley is important. Similarly, Hundie et al. (2001) also reported that on farm average yield loss of 28 to 29% was accounted for net blotch and leaf rust infection.
