Sorghum (Sorghum bicolor (L.) Moench) is a crop of world-wide importance, ranking fifth among the important cereal crops (Chantereau and Nicou, 1994). In the sub-Saharan Africa, it is arguably the most important cereal crop. The world production of grain sorghum amounted to 63.4 million tonne resulting from growing an area of about 47 million ha 63% of which in the African continent (FAO, 2009). The major uses of sorghum have been for food, feed, starch, and for fuel alcohol.  Initially, sorghum grain is used primarily for food; however, its use as a feed now exceeds its use as food especially in developed countries. In view of the pressing demand for fodder coupled with the fact that grain sorghum is the stable  diet for millions of people in the sub-Saharan Africa and India, it is imperative to reconsider the present mono-commodity breeding strategy of sorghum. Kelly et al (1991) questioned the current strategy of strictly adopting grain-yield criteria in evaluating sorghum genotypes arguing that fodder’s contribution to the total value of sorghum production has increased considerably.

 

Being a possible centre of origin, Sudan is endowed with a wealth of genetic variability in sorghum (Yasin, 1978) enabling selection for most economic traits. The sorghum germplasm of Sudan has been utilized extensively all over the world especially in the USA to improve yield of both grain and fodder (Mahmoud et al,1996). In contrast, local efforts to exploit such variability to develop dual sorghum types have been very limited and mostly directed towards developing improved grain types. Simultaneous improvement of sorghum for both fodder and grain attributes will help in meeting the demand for feed and food and allow maximum utilization of the limited farmer’s resource. Research efforts of such kind were very few or lacking in the Sudan. The objectives of this investigation were: To assess the magnitude of variability among some local and exotic sorghum for some traits that aid in developing dual forage/grain sorghum cultivars and to investigate associations between the major forage and grain attributes contributing to developing of dual purpose sorghum cultivars.

The experimental site

 

The study was conducted in Shambat (lat.15° 39 N; Long.32° 31 E) in the Experimental Farm of Agricultural Research Corporation (ARC) during 2010- 2011. The soil is heavy clay, non-saline, non-sodic with pH 7.8. The climate is hot and dry. Average Min-Max temperature was 14 and 40°C. The rainy season is short extending from July to September with scant and fluctuating precipitation.

 

Plant materials

 

The breeding nursery

 

The source population of materials used in this study was based on a breeding nursery established in 25/11/2010 in the Experimental Farm of Shambat Research Station (Sh. R.S.). The material grown consisted of 122 entries comprising 34 Sudan grass, 33 Abu Sabein, 29 grain sorghum, 17 sweet sorghum and 9 Ankolib genotypes. All materials other than grain sorghum were developed or kept by the Forage Improvement Program (FIP) at Sh. R.S. The grain sorghum genotypes were collected from different parts of the country, or donated by local research programs. Each entry was represented by one 5 m - ridge replicated twice. Sowing was done manually on the eastern side of the ridge by placing 3 to 5 seeds in holes spaced at 20 cm. Nitrogen fertilizer (urea) was added at the second irrigation at rate of 55 kgN/ha..Irrigation water was applied at 10 to 12 days interval.  Weed population was kept at minimum by hand weeding.

 

Selection criteria

 

The major criteria used to select for dual purpose (fodder/grain) genotypes included but not limited to: Early to semi early flowering time,  high  regrowth and tillering capacity,  medium to tall plant stature, large panicle size (thick diameter), leafiness, stay green, juicy sweet stems, bold white grains with no testa color. Panicle and grain characteristics were evaluated in the laboratory from five heads randomly chosen from each genotype. Based on the above attributes, 21 genotypes comprising 7 Sudan Grass, 5 each Abu Sabin and grain sorghum and 2 each Ankolib and sweet sorghum were selected. Selection was firstly based on high dual grain / forage yield then on related attributes with more emphasis given to earliness, regrowth and leaf to stem ratio (Table 1).

 

 

The trial

 

The 21 selected genotypes plus 3 standard checks (totaling 24) were arranged in α lattice design (Patterson and Williams, 1976) with 12 incomplete blocks (iblock) and 4 complete blocks. The iblock composed of two plots each having two 5 m ridges. Planting date was effected on 11/10/201. Sowing method and planting density were similar to those of the breeding nursery. In each incomplete iblock, forage and grain attributes were evaluated from the two outer and the two inner rows, respectively. Data collected for forage attributes included: Green (GMY) and dry (DMY) matter yields (t/ha), days to 50% flowering, plant height and stem diameter. Leaf to stem ratio: Measured on dry weight basis by dividing the weight of the leaves by the total weight of leaves and stems taken from five randomly chosen plants from each plot. Regrowth (g): Measured two weeks following the date of cutting of each entry, new immerging shoots from 5 competitive plants randomly chosen from each harvested plot were collected, air dried and the dry weight was determined in grams. Data collected for grain attributes included seed yield /plant (g), head circumference (cm), head length (cm), 1000 seeds weight (gm) and seed number/head.

 

Statistical analysis

 

The data was subjected to analysis of variance (ANOVA) following the procedure of alpha lattice design (Patterson and Williams, 1976). Correlation between different characters was worked out. The statistical software packages Agrobase Gen II (2008) was used to run alpha lattice whereas GenStat (2011) was used to run correlation analysis.

Agronomic performance

 

 

 

The analysis of variance (Table 2) revealed highly significant differences among genotypes for forage and grain yields and all related traits.

 

Forage yield

 

The overall mean for DMY was 7.34 t/ha (Table 3). SG33 gave the highest DMY (11.35 t/ha) followed by Abjaro (9.85 t/ha), S.25Abu70 (9.80 t/ha), SG8 (9.74 t/ha) and S.03Abu70 (9.74 t/ha). The lowest DMY was shown by Abnaffain (3.78 t/ha). The grain checks (WadAhmed, and ArfaaGadamak) and Ankolib types gave below average DMY. The overall mean for GMY was 33.8 t/ha. Generally, the genotypes kept similar trend as in DMY. The highest GMY was shown by S.25Abu70 (47.97 t/ha) and SG33 (47.72 t/ha) whereas the lowest GMY was shown by Abnaffain (17.95 t/ha).

 

Seed yield per plant

 

The overall mean for seed  yield  per  plant  was  31.61 g (Table 3). Abjaro showed the highest seed yield (72.50 g) followed by S.134Abu70 (57.54 g), S.26Abu70 (49.10 g), Hemasi (45.58 g) and S.25Abu70 (44.13 g). S.03Abu70 and E-35-1 averaged 41.85 and 36.92 g, respectively. SG53-1 and SG51 gave the best seed yield among Sudan grass group averaging 30.1 and 28.83 g, respectively. Abnaffain, ArfaaGadamak and WadAhmed gave below average seed yield amounting to 25.65, 23.60 and 21.72 g, respectively. The lowest seed yield was shown by the Sudan grass genotypes SG33 (18.39 g), SG32-2A (13.69 g) and SG32-1(12.89 g).

 

Growth traits

 

The genotypes SG12-1 and Abnaffain were the earliest with respective flowering time of 52.3 and 53.2 days (Table 4). In contrast, Abjaro and E-35-1 were the latest taking 71.3 and 66.0 days to flower. The flowering time for Abu Sabein genotypes ranged from 55.9 to 58.7 days. The average performance for plant height was 182 cm. Abjaro was the tallest (217 cm) whereas ArfaaGadamak and WadAhmed showed the shortest stature (122 cm). Abu Sabein genotypes  showed  above  average  plant height. The average performance for stem diameter was 0.95 cm. Abjaro was the thickest (1.57 cm) whereas SG32-2A was the thinnest (0.65 cm) in stem diameter. Abu Sabein genotypes, apart from S.24Abu70 showed above average stem diameter.

 

 

The highest leaf to stem ratio was shown by ArfaaGadamak (0.45), WadAhmed (0.44), FakiMustahi (0.44), Abjaro (0.43) and Atlas (0.43). The highest regrowth weight was given by the Sudan grass genotypes SG32-2A (33.8 g/plant) and SG8 (30.6 g/plant). Abnaffain Fakimustahi and All of the Abu Sabein genotypes gave below average regrowth weight whereas Abjaro gave above average regrowth.

 

Grain yield components

 

Mean performance for  number  of  seeds  per  head  was 1098 seeds (Table 4). Abjaro gave the greater number of seed per head (1879). Abu Sabein genotypes gave above average number of seed per head whereas the opposite is true for most of Sudan grass genotypes. Mean performance for 1000 seed weight was 28.5 g. The greater seed weight was expressed by the genotypes: Hemasi, S.26Abu70, S.134Abu70, S.25Abu70, Abjaro and Abnaffain, showing seed weight ranging from 36.8 (for Abnaffain) to 40 g (for Hemasi).

 

Mean performance for panicle length was 20.5 cm. The largest value for panicle length was shown by ArfaaGadamak (26.9 cm) and FakiMustahi (25.0 cm) whereas the smallest value was shown by Abjaro (15.7 cm) and E-35-1(15.8 cm). Mean performance for panicle circumference was 14.9 cm. Abjaro gave the largest panicle circumference (24.2 cm) whereas the smallest value was shown by SG32-2A (10.9 cm). All of the Abu Sabein genotypes and Abnaffain showed above average panicle circumference ranging from 15.8 to 17.3 cm.

 

Association study

 

 

Table 5 shows correlation among different grain and forage sorghum attributes. Positive highly significant correlations were observed between green matter yield (GMY) and each of seed yield (R = 0.402), 1000 seed weight, head circumference, plant height and stem diameter. Negative highly significant correlations were detected between GMY and leaf to stem ratio. GMY has week and insignificant correlation with days to flowering and regrowth. Seed yield has positive highly significant correlation with each of plant height, stem diameter. Correlations were weak and insignificant between seed yield and each of days to flowering and leaf to stem ratio. Plant height was positively and significantly correlated with 1000 seed weight and head circumference but has weak and insignificant correlation with days to flower, regrowth and number of seeds/ head. Plant height has negative significant correlation with leaf to stem ratio. Days to flowering has significant positive  correlation  with number of seeds/ head, stem diameter, head circumference and leaf to stem ratio, but has negative significant correlations with 1000 seed weight and head length. Weak and insignificant correlation was observed between days to flowering and regrowth.

Agronomic performance

 

Abjaro and the Abu Sabein selections: S.25Abu70, S.134Abu70, S.26Abu70, S.03Abu70 were the best genotypes ranking top in dual forage and grain yields (Table 3). However, of the 4 Abu Sabein genotypes, only S.25Abu70 and S.03Abu70 would be advanced due to their good performance in one or more of other attributes including regrowth, earliness and leaf to stem ratio (Table 4). Among the Sudan grass genotypes, SG08 , SG54 and SG51 averaged the best score for dual grain forage yield, however, the former has got brown seed color which is undesirable for food consumption whereas the latter showed the lowest regrowth value among Sudan grass genotypes (Table 4). SG51 has desirable seed color and can be considered for advanced testing. SG51 has been reported to have stable agronomic performance with high forage yield (Mohammed, 2010). E-35-1 was the best among sweet sorghum group showing desirable performance in dual purpose attributes.

 

Abnaffain is a traditional cultivar grown for dual production of grain and fodder. The literal translation of the Arabic name “Abnaffain” (Father of two benefits) implies the dual utility of this cultivar. It was included here as a check for dual fodder / grain production. In this study, all of the advanced materials outperformed the check Abnaffain in dual forage and grain yields. Its low fodder and grain yields could be attributed to its reduced plant height and seed number / head (Table 4). However, none of the studied  materials was earlier in flowering than Abnaffain. Earliness is a key factor in dual cultivations systems with limited moisture and other resources of production. ArfaaGadamak and WadAhmed exhibited below average performance in both grain and fodder yields. Their low forage yield could be attributed to their short stature. Both cultivars are the outcome of the national breeding program that emphasizes developing high grain yielding types at the expense of fodder yield, that is, dwarf combinable types suitable for mechanized harvesting. Abjaro belongs to the landraces of Northern Sudan. It was the best genotype combining the highest fodder and grain yields (Table 3). Its high fodder yield could be explained by its unique tallness and stem thickness (Table 4) whereas the high grain yield is attributable to the high number of seeds per head coupled with high weight of the seed (Table 4). Unexpectedly, Abjaro was among the best genotypes in leaf to stem ratio (Table 3). Usually leaf to stem ratio correlates unfavorably with forage yield and plant height as proved by this study (Table 5) or reported by other workers (Rashida and Mohammed, 2012). This trait represents a good measure for fodder quality as the greater part of the nutritive value is stored in leaves which also have better intake potential and digestibility (Mohammed and Zakaria, 2014). Unfortunately, Abjaro was the latest to flower (Table 4). However, being highly productive along with leafiness justify its advancement as a candidate for dual fodder/grain production. Some of Abu Sabein genotypes, that is, S.25Abu70 and S.03Abu70 were comparable to Abjaro in high fodder/grain productivity. Though less leafy, they were however, excelling Abjaro in earliness. Abu Sabein is basically a grain cultivar more probably derived from the palatable ‘Dibekri’ (Kambal 2003). Dibekri is a land race variety widely grown in Northern Sudan (Nile State). Driven by the need for fodder around cities, the farmers opted to Abu Sabein as a dual grain/fodder cultivar. However, Abu Sabein had been gradually devoted to forage production in response to the increasing demand for fodder.

 

Association study

 

The highly significant positive correlation revealed by this study between green matter yield (GMY) and grain yield (Table 5) points to the possibility of simultaneously combining high levels of grain and fodder yields in one cultivar. Similar results with grain and stover were reported by a number of workers (Ross et al., 1983; Blümmel et al., 2009; Reddy et al., 2005). Ross et al (1983) reported that grain yield had no extremely strong negative phenotypic correlations with any forage residue trait. They concluded that, the correlations obtained do not suggest any formidable barriers to simultaneous improvement of agronomic, grain, and forage traits. Their results go well with our finding that GMY was positively or favorably correlated with and 1000 seed weight, head circumference and seed number per head. Furthermore, this was strongly supported by the positive and highly significant correlation shown in this study between plant height and each of grain and fodder yields. Positive significant association between grain yield and plant height was also reported by Kumar et al. (2012).

 

Correlations of days to flower with each of forage and grain yield in this study were weak and insignificant permitting development of early and improved dual grain fodder cultivars. Disagreeing results for week correlation between grain yield and days to flower were reported by Patil et al. (1995).

 

The study revealed the possibility of selecting sorghum cultivars with high capacity for dual grain/fodder production. Six genotypes were identified as having the best dual grain/fodder excelling the standard checks, these included Abjaro (a land race cultivar), S.25Abu70 and S.03Abu70 (Selections from the land race cultivar Abu Sabein), SG08 and SG51 (Selections from Garawi, traditional Sudan grass cultivar) and E-35-1 an introduced sorghum cultivar.

 

High levels of grain and fodder yields coupled with some desirable traits could be incorporated in one sorghum cultivar as suggested by the favorable associations shown in the study.

The authors have not declared any conflict of interests.