By: Hintsa Gebru*, D.S. Virk, Eyasu Abraha, Tesfay Belay, and Zeray Siyum *Corresponding author: hagy516@yahoo.com

1. Searching and testing bread wheat genotypes for adaptation in
Northern Ethiopia through participatory varietal selection
Hintsa Gebru*1, DS Virk2, Eyasu Abraha1, Tesfay Belay1 and Zeray Siyum1
1
Mekelle Agricultural Research Center, P.O.Box: 258, Mekelle, Tigray, Ethiopia,
2
Centre for Advanced Research in Agriculture and Development (CARIAD), Bangor University, Bangor, Wales, UK
Introduction Result
Wheat is one of the major cereal crops grown Differences between varieties for grain yield
in the mid and high land areas of the Tigray re- and biomass were significant with REML
gion of Ethiopia. But, its productivity has been but not with GLM and the REML analysis
very low because of lack of improved varieties was more appropriate (Table 1).
(early maturing, drought tolerant and high
yielding genotypes), poor soil fertility and high
moisture stress. Many improved bread wheat Varieties HI-1418 and HUW-468 yielded
varieties have been released at national level, significantly higher than the two check vari-
but their dissemination and acceptance by farm- eties by 8-21% with earlier flowering by 5-9
ers in Tigray have been unsatisfactory. days and earlier maturity by 13-16 days.
They had significantly shorter reproductive
Materials and methods period by 7- 9 days that imparted them better
terminal drought tolerance.
A search process for the PVS was first done
within Ethiopia and identified HAR-2501 and
HAR-1868 as farmer preferred varieties. Se- They were also superior to checks for matrix
cond search led to the selection of four bread ranking, disease score, and number of seeds
wheat varieties (HUW-468, HI-1418, DL-788-2 per spike. Compared with check varieties,
and GW-273) brought from India that matched the two new varieties responded to the im-
farmers’ criteria. They were tested in PVS by provements in environments and also main-
32 farmers in six locations in 2008 and 2009 in tained their superiority in the low yielding
the Tigray region. The initial and harvestable environments for grain yield.
areas were 20 and 9 meter square, respectively.
Varieties HAR-2501 and HAR-1868 were used
as check varieties. A mother-baby trial system In general, the results showed that a precise
was used and all trials were conducted under search process increases the success rate of
farmer management. Quantitative data were PVS.
collected and analyzed using the GLM fixed ef-
fects and REML mixed effects models.
Table 1. F-test for the fixed effects varietal differences in
the REML analysis for mother trials over 2008 and 2009
Source GY BY DH DM RP TKW
Var 6.5 3.6 54 64 37 17
SS SL HI Dis PH MR
Var 5.0 6.5 4.2 21 7.6 54
All F-values were significant at p<0.001
Where,
GY=grain yield, BY=biomass yield, DH=days to heading, DM=days to maturity, RP=reproductive period,
TKW=1000 kernel weight, SS=seeds/spike, SL=spike length, HI=harvest index Dis=disease scoring,
PH=plant height, MR=matrix ranking