Genetic Variability and Correlation Analysis of Ethiopian Linseed (Linum usitatissimum L.) Accessions

This study was executed in 2012/13 cropping season to determine variation of 202 accessions and correlation coefficients of some agronomic characters. The experiment was carried using augmented design with no replication. Accessions showed variation for days to flowering ranged between 77 and 135 days with a mean value of 99.65 and coefficient of variation 13%. Accessions such as acce # 10 077, 10 078, 1 0079, 10 084, 10 086, 10 087, 13 578, 13 598, 13 620 and 13 649 showed earliness, which ranged from 170-172 days to mature, while accessions such as acce # 212 854, 212 857, 215 716, 10 103, 10 104, and 10 105 took 180-182 days to mature. A range of 14.0 to 38.8 was recorded for number of capsule per plant with a mean value of 23.12 and coefficient of variation 19%. Accessions such as acce # 230 822, 10 070, 10 105, 10 086, 10 069, 10 079, 10 052 and 219 969 had number of capsule ranged from 32 -41. A significant amount of variation was also observed for seed yield, which was between 104 kg/ha and 1982 kg/ha with a mean value of 984.37 kg/ha and coefficient of variation 39%. Accessions such as acce # 10 067, 10 061, 10 084, 10 069, 13 755, 10 070, 10 054, 10 068, 241 826, 10 064 and 10 150 were high yielding, which varied between 1 592 kg/ha and 1 990 kg/ha. Correlation coefficient analysis indicated that seed yield had significant positive association with plant height (0.215 9), number of secondary branches (0.204 8), and number of capsule (0.384 6), while it recorded significant negative correlation with days to maturity (-0.142 6) and number of primary branches (-0.298 2). This study has also suggested that number of capsule per plant was found to be the most important yield component for the improvement of seed yield of linseed.


Background
Linseed (Linum usitatissimum L.; 2n=30) is one of the most important oil crops in the highlands of Ethiopia. It is cultivated for seed production, which is used for extracting edible oil. Ethiopia is the seventh world linseed producing countries after Canada, Russia, China, Kazkhstan, United States and India (FAO, 2012). According to CSA (2014) data, it covered an area of 82, 325.78 ha of cultivated land, and produced 83, 130.5 tones with productivity of 1.0 t ha-1. It is commonly grown on marginal lands with less or no input (Teklewolde et al., 1992). Increasing seed yield of linseed is the main breeding objective along with others such as breeding for quality, disease resistance, frost resistance etc.

Number of primary branches
The magnitude of variation for this character was between 3 and 6.7 with a mean value of 4.25 and coefficient of variation 16% (Table 1). 86.1% of the accessions had 3-5 primary branches per plant (Figure 4).

Number of secondary branches
The variability of this character ranged from 3.4 to 9.5 with a mean value of 6.61 and coefficient of variation 18% (Table 1). 78.7% of the accessions had 5.4 -8.4 secondary branches per plant ( Figure 5).

Number of capsules
A range of 14.0 to 38.8 was recorded for this character with a mean value of 23.12 and coefficient of variation 19% ( Figure 6). 49% of the accessions recorded above the average ( Figure 6). Eight accessions such as ace # 230 822, 10 070, 10 105, 10 086, 10 069, 10 079, 10 052, and 219 969 had number of capsule ranged from 32-41. Variation in number of capsules was also reported by Wakjira (2011).

Correlation analysis
Correlation coefficient analysis is presented (Table 2). This analysis depicted that seed yield had highly significant postive association with plant height (0.215 9), number of secondary branches (0.204 8), and number of capsule per plant (0.384 6), while it recorded significant negative relationship with days to maturity ( -0.142 6) and number of primary branches (-0.298 2). Similar results were recorded on days to maturity, number of secondary branches, and number of capsule, but the opposite for number of primary branches (Savita, 2006). Number of capsule had also significant positive association with seed yield (Popescu et al. 1999;Belete and Wolde Yohanes, 2013).

Materials and Methods
The present study was carried out at Holetta Agricultural Research Center in 2012 cropping season. The experiment was laid out using augmented design with no replication, and 202 accessions including checks were grown in two rows of 3 m length and 30 cm between rows. All the management practices were followed as per the recommendation. Ten plants from each accession were selected to record data on days to flowering, days to maturity, plant height (cm), number of primary branches, number of secondary branches and number of capsule per plant. Seed yield (kg/ha) was recorded on plot basis. Descriptive statistics and Pearson correlation coefficients were determined using AGROBASETM software (Agronomix Software Inc., Canada).