Ethiopian Institute of Agricultural Research

ETHIOPIAN INSTITUTE OF AGRICULTURAL RESEARCH

JIMMA AGRICULTURAL RESEARCH CENTER

SOYBEAN BREEDING PROGRAM

REPORT ON SOYBEAN INNOVATION LAB PROJECT ACTIVITIES FOR THE PERIOD 2015

MARCH 2016

JIMMA, ETHIOPIA

1. Title: Evaluation of selected plant to rows/recombinant inbred lines (RILs)

One hundred sixty five recombinant inbred lines (RILs) that were selected from S6 segregating populations were planted in 16 blocks in augmented design with five check varieties replicated in each block at Jimma to identify superior genotypes for further evaluation. Two rows of 4m length plots were used with 5 X 60cm intra and inter row spacing, respectively. Recommended rate of 100kg/ha DAP was applied at planting; and the trial were hand weeded three times. Data was analysed on-line using Federer (1956) method. Mean grain yield and other parameters are presented in table 3. The statistical analysis showedhighly significant difference among controls, test entries and test Vs control for Days to Maturity (DTM), plant height (PHT), blight score and grain yield, respectively (Table 4). Warm climate during main growing period and prolonged rain as a result of ElNion weather change might have favoured high production of soybean very different from what was exhibited at Jimma before. Mean grain yield of the top ten high yielder recombinant inbred lines ranged from 5.5-5.9 t ha-1, whereas mean of checks was 4.46t ha-1. Grain yield and disease resistance/tolerance of the lines is quite impressive. As a result, promising lines will be advanced to National, Pre-National and advanced observation nursery variety trials for further evaluation. It was very interesting to note that a good number of early maturing lines with better yield than released early set varieties were identified. Therefore, early set variety trial will be composed from RILs developed at Jimma, and materials introduced from University of Illinois for multi-location variety trial in moisture stress areas of the country, Tirafeta, Dimtu, Areka and Hawassa.

Table1 Adjusted mean yield and other parameters of ten high yielder recombinant inbred lines and checks in the year 2015 at Jimma

Remarks-: DTF=date of flowering; DTM=date of maturity; PH =Plant height; NOP= NO of pods/plant; NOS = NO of seeds/plant;NOB=NO of branches, LG=Logging score HSW =Hundred seed weight, CBB-Common bacterial blight; BP=bacterial postule, HSW=hundred seeds weight; AD.GY=Adjusted grain yield; UN.AD GY=UN adjusted grain yield

1. Title: Soybean multi-location variety trial (MSc thesis research)

About 20 soybean genotypes selected from germplasm introduced from theNational Soybean Genetic Resource Center of USA in the year 2014 were planted this year along with four locally released varieties in six major soybean growing areas of Ethiopia (Jimma, Mettu, Pawe, Tiroafeta, Assosa and Bako). The 20 genotypes were selected based on their relative performance to the locally released varieties. These selected materials were sown in a RCBD with three replications at six locations across the country. The experiment was planted in a 3mX1.8m plot in3 rows. All data was taken from the centre two rows, except destructive data which will be taken from the border row. Spacing of 60cm between rows and 5cm within row was used, and recommended rate of 100kg DAP/ha was applied at planting.

In general, the trials were very well managed and good in locations,Jimma and Assosa. At Mettu the trial was very good up to flowering and early pod filling stage, and then was damaged by hail. The trial at Dimtu was well managed; however the performance of the genotypes was very low which might be as a result of the confounding effect of moisture stress and soil acidity problem. Damage by termites and weak trial management at early growth stage was reported at Pawe.

In the trial at Jimma, SCS-1 produced the highest yield of 5.0 t ha-1 followed by Afgat, and AGS-7-1, with respective yield of 4.58 and 4.46t ha-1. The highest yielder among the US introduced materials was KS 4895, Deasha and UA4885, with respective yield of 4.39, 4.32 and 4.31 t ha-1. At Dimtu, SCS-1 produced the highest mean yield of 2.2 t ha-1 followed by KS 4895, which produced 1.92t ha-1. At Metu,Delsoy 4710 produced the maximum yield of 1.95t ha-1 followed by Hang dou No. 1, which produced 1.87t ha-1. At Assosa, the top yielding genotype was Hs93-4118 which produced 3.5t ha-1, followed by Motte that produced 3.44 t ha-1. At Bako, LDOO-3309 and UA4885 gave the highest mean grain yield of 1.99 and 1.91 respectively. The mean grain yield performance of the check varieties Clark 63 K and AGS-7-1 gave the highest mean yield of 3.93 and 3.88t ha-1, respectively at Pawe (data is not reliable because of termite damage). The across locations mean grain yield showed that the standard check varieties Clarck 63 K, SCS-1 and AGS-7-1 produced the highest yield of 2.69, 2.65 and 2.60 t ha-1, respectively.

Table 3 Performance of selected introduced soybean genotypes

1. Evaluation of Variety X Fungicides spraying against Soybean Rust in Southwestern Ethiopia

Summary of results

Six fungicide spray treatments namely: No spray or Control (F0); Luna sensation 0.15l/ha(F1), Nativo 0.5 l/ha(F2), Opera Max 1l/ha (F3), Opera Max 0.75 litter per hectare(F4) and Opera Max 0.50 litter per hectare(F5) in a factorial combination with three soybean varieties viz, Clark 63k (V1), Afgat (V2) and SCS-1(V3) ) were tested in RCBD with three replications at Jimma and Metu to evaluate the efficacy of selected fungicides against soybean rust and to determine the interaction of soybean varieties with fungicides for soybean rust control. The experiment was planted in a 4m X 2.4 m plot in 4 rows at 60cm between rows and 5cm within row spacing, respectively. DAP fertilizer was applied at planting at the rate of 100kg/ha.The experimental field was hand weeded three times and all data were recorded from the middle two rows. The trial at Metu was damaged by hail storm at flowering and early pod filling stage; but later on recovered, though poor grain yield obtained as a result of the damage. On the other hand, disease incidence at Metu was better than Jimma; as a result,Metu data is presented for this season. The soybean rust was assessed at five different times between vegetative growth (R1) and full grainfilling stage (R5) at every two weeks interval. In addition,grain yield(GY) and yield component data were recorded. Area underdiseases progressive curve (AUDPC) and disease severity (DS) visual score were calculated. All parameters were analyzed by SAS software using general linear model (GLM).The statistical analysis showed highly significant difference among varieties(V) and fungicide spray (F)treatments at Metu, whereas the V X F interaction was non-significant. AFGAT responded better to fungicide spray than the other two varieties for GY, whereas disease severity score was better for SCS-1 than others.Among fungicide spray treatments,Opera Max 0.75 litter/ha produced better GY (2.88t/ha) than others. Moreover, disease severity percentage score was also less than 10.1% for the same treatment(Table 5, 6 and 7).Low record for DS,AUDPC and GY were observed for non-fungicide spray plots compared to others indicating fungicide spray could control rust disease and increase GY. The trial should be repeated in the coming season to reach at valid conclusion.

Table4 Mean grain yield and other parameters the soybean varieties

Table 5 Mean grain yield and other agronomic parameters of the different fungicide spray treatments

Remark: DS (%) = NOP=No. of pods/plant, NOS=No of seeds/plant; DS (%) =disease severity (%); AUDPC =area under disease progress curve; HSW=Hundred seeds weight

Table 6 Mean square of grain yield and other parameters of soybean fungicide spray Vs variety trial at Metu in the year 2015

1. Title: Effect of different strains of rhizobium bacteria on soybean varieties

Summary of results

Thirty treatment combinations, which consisted of five rhizobium strains, and six elite and released soybean varieties were planted in RCBD in a factorial combinations to identify the best rhizobium strain and variety performance. The experiment was planted at Jimma and Metu in a RCBD with three replications.Plotsize of 4 X 2.4 m2 consisting of 4 rows, and inter and intra row spacing of 60cm and 5cm, respectively was used to plant the experiment. Recommended rate of 100kg ha-1 DAP was applied at planting; and the trial was hand weeded three times. The trial at Metu was damaged by hail storm at flowering and grain filling period; and hence data only from Jimma is considered for this report.Yield and yield component data were recorded, while nodule count was recorded from five sample plants of the border rows. The statistical analysis showed highly significant differences among varieties (V), rhizobium races (R) and their interaction (V XR) for lodging visual score(LG)and GY (Table 3). For other parameters, variety and rhizobium strains responded independently. Variety SCS-1 in combination with rhizobium strains SB-MAR-1495 (R1) and race SB-6-1-B2 produced GY of 5.91 t/ha and 5.22 t/ha, respectively, which is significantly better than other treatment combinations. The trial should be repeated for one more additional season.

Table 7 Mean yield and other parameters of varieties at Jimma

Table8Mean yield and other parameters of rhizobium races at Jimma

Table9MS value of Gy and other parameters of rhizobium races against soybean varieties evaluation at Jimma in the year 2015

Table 10 Mean yield and other parameters of variety and rhizobium race combination at Jimma in the year 2015

Remarks-: PH =Plant height; NOP= NO of pods/plant;NOS = NO of seeds/plant; HSW =Hundred seed weight;TNN=Total nodule No

1. Evaluation of introduced Brazilian soybean materials

A total of sixteen Brazilian materials introduced from USAwere planted along with 9 released and elite soybean varietiesin a 5X5 simple lattice design at Jimma and Metu. Two rows of two meter length plot was used with intra and inter row spacingof 6cms and 60cms;respectively. 100kg DAP/ha rate was applied at planting and three times hand weeding were made to make the field free from weeds. Grain yield (GY) at Metu was greatly reduced as a result of heavy hail storm at flowering and early pod filling stage.Therefore, only data from Jimma were considered to make selection for further test. The F-testshowed highly significant difference among genotypes for bacterial pustule and rust severity visual score, number of pods (NOP), and adjusted grain yield (ADGY).Mean GYof both adjusted and unadjusted and other parameters is shown in Table2. As shown in thetable, the top yielder genotype was BRS 283 with 5.8 t/ha followed by two pipe line varieties from JimmaG9945 (5.2tha-1) and PR-143(5.1tha-1) in that order. The GY exhibited in this season is very high and unique to our location which might be due to climate change (the country encountered sever Eli nino manifested weather change) in the season, which is characterized around Jimma by warm temperature during the growing period and prolonged rainy season which might have favoured the crop. It is very interesting to note that most Brazilian genotypesshowed high to moderate resistance to the major diseases found at Jimma, though one year data is not conclusive enough.Promising genotypes selected from this trial will be advanced to either pre-national variety trial (PNVT) or National variety trial (NVT) across the country in the coming season. The former will have at least three locations; while the later will be conducted in at least six locations.

Table 11Mean grain yield (t/ha) and other parameters of15 high yielding Brazilian soybean genotypes and checks evaluated at Jimma in the year 2015

Remarks-: DTM=date of maturity; PH =Plant height; NOP= NO of pods/plant; NOS = NO of seeds/plant; HSW =Hundred seeds weight,CBB-Common bacterial blight;BP=bacterial postule,HSW=hundred seeds weight;ADGY=Adjusted grain yield;UNAD GY=UN adjusted grain yield

1. Title: Hybridization among selected soybean parental lines and selection from segregating generations

Summary of results

In February 2015,twelve selected parents were planted in a pot in the screen house in six different sowing dates staggered at 7 days interval, and 77 successful F1 hybrids were developed. The resulting F1 hybrids were again planted in a pot in the screen house and F2 seeds were harvested in October, 2016.The resulting F2 seedsare planted in the offseason nursery inFeb., 2016 and the young seedlings are under good condition. In addition, 79 F2 segregating populations were received from University of Illinois were planted in July 2015 main season, and F3 seeds were prepared using modified single seed descent method and the harvested seeds are now planted in the offseason nursery. The F4 seeds will be harvested and planted in the main season of 2016 (June-October, 2016) for plant to rowsselection. A total of fourteen parental lines were selected based on their performance for yield, disease resistance, and earliness from the various variety trials of 2015. The selected parental lines are now planted in the screen house in pots for crossing to start as of end of March.

Table 12 Soybean lines selected for crossing off-season 2016

1. Title: Screening soybean accessions for rust resistance

Summary of results

About 116 soybean accessions, some of which were reported to have rust resistance genes were introduced from the National Soybean Genetic Resource center, and tested at Jimma and Metuto evaluate their performance mainly for their resistance to rust disease and identify rust resistant lines for further release and use as parental lines in crossing programs. The materials were planted in 11 X 11 simple lattice designs along with five check varieties in a 1.2 X 2mplot intwo rowswith 60cms and 6 cms inter and intra row spacing, respectively. A recommended rate of 100kg ha-1 DAP was applied at planting. Statistical analyses were made using GLM procedure of SAS software. Unlike other trials at Metu, this trial specific site was not hit by hail storm. The F test showed highly significant difference among varieties for all parameters at Metu and for all, except bacterial pustule and rust visual score at Jimma.Mean grain yield (t/ha) and other parameters of 20high yielding genotypes of each location ispresented in Tables 12 and 13.As shown in the table below,the mean GY of the top 20 genotypes that were introduced for rust resistance screening ranged from 4.8-6.8tha-1 at Jimma and 3.04-4.00tha-1atMetu showing that there are promising genotypes for grain yield performance, besides possibility of identifying rust resistant line.However, the materials were not well evaluated for rust resistance, as there was no rust incidence in the season. The trial should be repeated for one more season to make good evaluation of the genotypes for rust resistance.It was very interesting to note that some of these genotypes were high yielders; as a result some of the best performing materials were selected as parental lines for crossing and will be advanced to PNVT and NVT for further for yield performance evaluation.