Agronomic Performances of Hybrid Sorghum bicolor [ L ] Moench , Guinea Races Cultivated under Limited and High-Soil Phosphorus Conditions

Sorghum bicolor [L] Moench is among the main starchy staples in Mali. However, its yield is low. Development of Guinea sorghum hybrids adapted to the Malian environment is one of the solutions to increase sorghum production and productivity. This study was carried out to: identify the combining ability of Guinea sorghum and intermediate races for grain yield, to assess the agronomic performance of hybrids in relation to local cultivars currently grown by the farmers under low and high soil phosphorus conditions; and develop the methods of hybrid seed production. Study was conducted in research stations of Samanko during the rainy season. Results revealed that the response of yield and plant growth performance to fertilization depended on the genotype. The Guinea-Caudatum/Guinea-Caudatum hybrids were the most productive in the plot with Di-Ammonium Phosphate (DAP). On the other hand, in the poor soil without DAP the GuineaEcotype/Guinea-Ecotype and Guinea-Ecotype/Guinea-Caudatum hybrids were the most productive. The best hybrids in terms of high yield stability in both trials were evaluated in farmers’ fields in 15 villages. The hybrids: 150A/02-SB-F4DT-298, 150A/Grinkan, 12A/Lata-3Bala, PR3009A/Lata-3-Bala and FambeA/Lata-3-Bala were identified as the farmers’ preferred hybrids due to their high yield and good grain quality, and they were listed in the National Catalog. A total of 608 seed production farmers and peoples from private seed companies, farmers’ associations and extension services were trained in hybrid seed production techniques. Through this study, the introduction of sorghum hybrids farming in Mali has become a reality.


Introduction
Sorghum (Sorghum bicolor [L] Moench) is the fifth most important cereal crop in the world after wheat, maize, rice, and barley [1].It is the dietary staple of more than 500 million people in more than 30 countries and is grown on 42 million hectares in 98 countries of Africa, Asia, Oceania and America [1].Nigeria, India, USA, Mexico, Sudan, China and Argentina are its major producers.In Africa, sorghum ranks the second after maize (Zea mays) in terms of production and consumption with significant producing countries being Nigeria, Mauritania, Gambia, Mali, Burkina Faso, Ghana, Niger, Somalia, Chad, Tanzania and Mozambique [1].There are 20-30 species of sorghum.Harlan JR et al. proposed a simplified classification scheme of cultivated sorghums based on morphological characteristics that most present-day breeders have come to recognize and utilize and partitioned the primary gene pool, Sorghum bicolor (L.) Moench into five major races: Bicolor, Guinea, Caudatum, Kafir, and Durra and ten intermediate races or all combinations of the basic races [2].
The five basic races are: • Bicolor: The most primitive cultivated sorghum, characterized by open inflorescences and long clasping glumes that enclose the usually small grain at maturity.
• Caudatum: Characterized by turtle-backed grains that is flat on one side and curved on the other; the panicle shape is variable and the glumes are usually much shorter than the grain.
• Durra: Characterized by compact inflorescences, characteristically flattened sessile spikelets, and creased lower glumes; the grain is often spherical.
• Guinea: Characterized by usually large, open inflorescences with branches often pendulous at maturity; the grain is typically flattened and twisted obliquely between long gaping glumes at maturity.
• Kafir: Characterized by relatively compact panicles that are often cylindrical in shape, elliptical sessile spikelets and tightly clasping glumes that are usually much shorter than the grain.Many subgroups can be distinguished with cultivars especially adapted to high or low rainfall regimes.
In Mali, sorghum is among the major crops along with pearl millet, rice, maize, cotton, peanut, and cowpea.Production of sorghum, however, does not meet demand of the growing population in the country.Grain sorghum yields are very low (0.8 to 1.3 tons/ ha) and these yields are well below the genetic potential of sorghum, which, under favorable conditions, can yield up to 13 tons/ha [3].Toure S indicated that three major sorghum races exist in Mali: Guinea, Durra and Caudatum [4].The Guinea sorghums constitute the dominant race grown by farmers and represent about 70% of sorghum races in Mali.They are well known for their high grain quality, good weathering, pest resistance, and storage qualities [5].However, the yield potential of most cultivars is relatively low.
The sorghum production is decreasing due to several reasons including low soil fertility, high cost of input, drought, low adoption of improved cultivars and lack of sorghum hybrids.Phosphorus (P) deficiency ranks second in importance after nitrogen (N) deficiency in sorghum-growing regions of the semiarid tropics Africa [6].Amelioration of phosphorus (P) deficiency by application of P fertilizers is costly for the small subsistence farmers with limited financial resources.They need to determine efficient fertilizer practices in conjunction with the development of P-efficient cultivars.
The adoption of improved cultivars from sorghum breeding programs is very low, less than 5% of the cultivated area [7,8].This is due the poor adaptation of the available cultivars to the farmers' environments, the poor food quality, poor storability of grain and the lack of information on the existence and utility of new cultivars [9].
The adoption of hybrid sorghum is also very low by Malian sorghum producers.In Mali the exploitation of hybrid sorghum began in 1979 and between 1986 and 1990 the initial hybrid materials were lost from the Malian sorghum; hence it needs to develop new parents and high yielding hybrids [10].

General objective
Increase production and productivity of sorghum in Mali.

Specific objectives
To identify the combining ability of parents of different groups of Guinea sorghum races and Guinea intermediate races for grain yield and others agronomic traits under limited and high-phosphorus conditions.
To evaluate the agronomic performance of hybrids with check in trials without phosphorus (-P) and trials with phosphorus fertilization (+P) To develop and promote parents and hybrids of sorghum Guinea races and Guinea intermediate races in limited and high phosphorus conditions.

Experimental site
The Agricultural Research Station of Samanko, IER Mali is located at the West of the town of Bamako on the road of Kangaba.The climate is of Sudano Sahelian type.Coordinates are 12 °54 ' North, 8 °4' West on altitude 329 m and isohyet between 800 and 1000 mm.The soil is clay loam or sandy clay with very poor nutrients.

Plant material
Five female parents (cms) and eleven fertile male parents developed in Mali were selected for study (Table 1).The combination of female and male parents generated 55 hybrids, including 12 Guinea-ecotype×Guinea-ecotype, 15 Guinea-Caudatum×Guinea-Caudatum and 28 Guinea-Caudatum×Guinea-ecotype (Table 2).These hybrids were planted with four sorghum lines used as checks (Table 2).One cultivar of maize (Sotubaka) was used to supplement the total number of material at 60 genotypes.The performance of this maize cultivar was not measured in the study.

Methods
The research was conducted in 2007 and 2008 respectively at Samanko Research Station in one trial with Di-ammonium phosphate (DAP) applied and the other without phosphorus applied each year.A total of four trials were conducted (Table 3).

Experimental design
The experimental design was alpha-lattice with three replicates [(10×6)×3] for each trial.Each genotype was sown in four rows of three meters length.The spacing between rows was 0.75 m, and between hills within a row was 0.40 m (eight hills per row).Five to six seeds of genotypes were planted in each hole and thinned to two plants per hill at three weeks after emergence to give a population density of 16 plants per row either 71,111 plants per hectare.

The field management
• Soil preparation: For plowing and ridging the tractor was used, the plow was used for the butting.
• Seed treatment: The seeds were treated with Apron star (against seedlings and insect damage) at a dose of 10 g/4 kg of seed.
• Fertilization: The Di-ammonium phosphate (DAP 18-46-0) was applied to the dose of 100 kg/ha at the sowing time in the phosphorus trials in 2007 and 2008.Other two trials (without phosphorus) did not receive phosphorus in (2007 and 2008) but to supply nitrogen, these trials received 37 kg/ha nitrogen at the sowing time.And then nitrogen was also applied to all trials at 50 kg/ha at first weeding for the two years.

Data collection
• Plant height: The plant height was recorded by measuring the distance from ground level to the tip of the panicle at physiological maturity and expressed in centimeters.
• Days to 50% heading: It was recorded as number of days from the date of sowing to the date when half (50%) of the plants in the plot have visible heads.

Data analysis
The analysis of the various parameters was carried out using the GenStat twelfth edition (12.1.0.3278).Analysis of the variance was performed using the regression method.

Combining Ability Analysis of hybrids' parents
The ANOVA of General Combining Ability (GCA) and Specific Combining Ability (SCA) were estimated for each environment using the regression method.
• Model for each environment: In order to assess the effect of the Genotype*Environment interaction on combining ability, a combined analysis was performed using the regression method.

• Model for across environments
Where, X f and X m = Mean of female and male respectively GCA f and GCA m = General combining ability of female and male respectively.µ = Overall mean of crosses in the trial.
SCA was computed as: Where, X i = Observed mean value of the cross E j = Expected mean value of the cross based on the 2 GCAs of its parents ij = crosses, ij= 1…n

Combining ability analysis for grain yield of parents and their F 1 hybrids in different environments
Analyses of variance (ANOVA) for different environments showed significant or highly significant differences (at P=0.05 and P=0.01) of GCA female and male parents in all environments except for SK-DAP-07 environment for female parents (P=0.47)(Table 4).
Significant to highly significant differences (at P=0.05 and P=0.01) were observed between hybrids for SCA in two environments (SK-DAP-07 and SK-SDAP-07) (Table 4).However, there were not significant differences among hybrids for SCA in the two environments: SK-DAP-08 and SK-SDAP-08 (Table 4).
The ratio GCA/SCA values were greater and positive in all environments except SK-DAP-07 for the female parents (Table 4).
Under both low and high-soil phosphorus conditions in 2007 and 2008, the parents LATA 3 Bala and Grinkan had positive GCA effects and produced high yielding hybrids in hybrid combination (Table 5).

GCA effects of female parents for grain yield under low and highsoil phosphorus conditions in Samanko station (2007 and 2008)
Under P-limited environment in 2007 and 2008, the female parent FambeA (+210 and +290) had positive GCA and produced hybrids with high yield (Table 6).
Under high-P conditions in 2007, the parent FambeA exhibited positive GCA (+127), whereas in 2008 the parent 150A had high and positive GCA (+550) and produced best combinations for grain yield under high-soil phosphorus conditions.

SCA effects of male crossed with female parents for grain yield under low and high-soil phosphorus conditions in 2007 and 2008
Estimates of specific combining ability effects (SCA) and mean data of genotypes for grain yield (kg/ha), are presented in Appendix 1. Lsd: Least significant difference; ** : Significant at the 0.01 probability level; * : Significant at the 0.05 probability level; ns: Not significant Under P-limited environment in 2007, the genotypes FambeA/00-KO-F5DT-19 (1520 kg/ha), and 12A/LATA 3 Bala (1370 kg/ha) were the highest yielding hybrids.These two hybrids also had high and positive values for SCA with +450 and +249, respectively (Appendix 1).The best check in this environment was CSM 388 with grain yield 770 kg/ha (Appendix 1).
The hybrids 150A/CSM-388 and GPNA/Grinkan had high and positive value of SCA with +367 and +293, respectively but they did not produce high grain yield.
The group Guinea-Caudatum/Guinea-Caudatum had the highest average yield in this environment (Figure 1).

The performance of genotypes for plant height and 50% heading across low and high-soil phosphorus conditions (2007-2008)
Mean squares of plant height and 50% heading in the four environments: The analyses of variance were highly significantly different or significantly different at P=0.01 at P=0.05 and for plant height and 50% heading at all four environments, suggesting that, the performance of genotypes was different in the environments for these traits (Table 7).

Performance of genotypes for 50% heading under low and highsoil phosphorus conditions
The 50% heading under low and high-soil phosphorus conditions are presented in Appendix 2.
Under low soil phosphorus environment in 2007, the mean of number of days from the date of sowing to the 50% heading was 81; the minimum was noted with FambeA/CSM-63E (74 days).In the same condition in 2008, the mean of number of days was 85 days and the minimum number was by the same genotype FambeA/CSM-63E (79 days).
Under high soil phosphorus environment in 2007, the mean of number of days from the date of sowing to the 50% heading was 71 days; the minimum days was seen with PR3009A/02-SB-F5DT-189 (63 days), whereas, in the same environment in 2008, the mean was 81 days and the earliest genotype was PR3009A/02-SB-F5DT-189 (63 days).The high soil phosphorus conditions reached 50% heading before that of those in low soil phosphorus conditions, but the difference observed was not highly significant.The group Guinea-Caudatum/Guinea-Caudatum was later maturing than other two groups in low-P conditions but, in high P conditions, it was earlier maturing than other two groups (Figure 2).

Performance of genotypes for plant height under low and high-soil phosphorus conditions
The plant height under low and high-soil phosphorus conditions are presented in Appendix 2.
Under high soil Phosphorus conditions, the mean plant height was 338 cm and 358 cm respectively, in SK-DAP-07 and SK-DAP-08 environments.The genotype 12A/02-SB-F4DT-298 showed minimum plant height with 209 cm in SK-DAP-07 environment and in SK-DAP-08 environment, the genotype 150A/02-SB-F4DT-298 showed minimum plant height (203 cm).The plant height was higher under high soil phosphorus conditions that under low P conditions.The group of ecotype Guinea-Caudatum/Guinea-Caudatum had the shortest plant height at all four study environments followed by the group the Guinea-Caudatum/Guinea-Ecotype (Figure 3).The highest plant height was seen for the group Guinea-Ecotype /Guinea-Ecotype at all study environments (Figure 3).

Discussion
The objective of the present study was to determine the combining abilities of parents of different groups of guinea sorghum races and guinea intermediate races for grain yield and other agronomic traits under limited and high-phosphorus conditions.Analyses of variance for different environments showed significant or highly significant differences (at P=0.05 and P=0.01) of GCA for female and male parents at all environments except for SK-DAP-07 for female parents (P=0.47).
There were high significant differences between hybrids (SCA) for yield and others agronomic traits in trials without phosphorus (-P) and trials with phosphorus fertilization (+P) in 2007 and 2008.The lowest yields were observed under phosphorus limited environments, however, under high phosphorus environments, the yields of genotypes were higher in 2007 and 2008 trials.Yield reduction in trials without phosphorus (-P) compared to the trials with phosphorus (+P) has been reported by several authors [11][12][13][14][15]. Sandaña PA et al. also reported that the Low P trials generally exhibited reduced grain yields, relative to the corresponding high P trials [16].
In trials without phosphorus (-P), the maximum for Guinea-Caudatum/Guinea-Caudatum hybrids had low yield compared to the Guinea-Caudatum/Guinea-Ecotype and Guinea-Ecotype/ Guinea-Ecotype.Whereas under +P environments, the latter were the most productive.This result agrees with those obtained by other authors [17,18].The Guinea races (type local) respond poorly to intensification in contrast to improved Guinea-Caudatum races, which explains the low level of productivity of Guinea-Caudatum/ Guinea-Ecotype hybrids on poor soil.Toure A et al. reported that the Guinea sorghums are well known or for their high grain quality, good weathering, pest resistance, and storage qualities.However, the yield potential of most cultivars is relatively low and they do not respond well to mineral fertilizer [5].
The study noted early heading of genotypes under highphosphorus environments in 2007 and 2008, however, under P-limited environments, the heading was delayed.This result agrees with those obtained by Lamond RE et al. who reported that the effects of phosphorus deficiency on plant growth delayed maturity, reduced quality of forage, fruit, vegetable, and grain crops, and decreased disease resistance [19].In addition, Sahrawat KL et al. reported that the deficiency of phosphorus is known to reduce plant height and delay plant growth in sorghum [12].
The Guinea-Caudatum/Guinea-Caudatum hybrids were later maturing than Guinea-Caudatum/Guinea-Ecotype and Guinea-Ecotype/Guinea-Ecotype hybrids combinations in low-P conditions but, in high P conditions, they were earlier maturing than other two combinations.This is in line with the research of several authors who reported that the Guinea-Caudatum respond well to mineral fertilizer.
A difference in plant height in trials without phosphorus and in trials with phosphorus in 2007 and 2008 was seen.The tallest genotypes were obtained under +P environments while those under -P conditions were the shortest.Camacho R et al. reported similar results and reported that lack of phosphorus substantially reduced plant height [20].It should be noted that the yield reduction, heading delay, and reduced height in -P compared to +P have been reported for several crops : Fageria NK et al. on rice cultivar [21]; Atlin GN et al., on oat [22]; Sahrawat KL et al. on sorghum [13]; Wissuwa M et al. on rice [23] and Manske GG et al. on wheat [24].This suggests that our environmental selection based on yield differences, height reduction, and heading delay was appropriate.

Promotion of Best Hybrids Identified in Limited and High Phosphorus Environments
Experimentation in farmers' fields: From 2009 to 2012, around twenty hybrids (among the three combinations) identified with high yield stability in both trials without and with phosphorus fertilization were evaluated in the farmer's fields in 15 villages of Bougouni and Kati prefectures (Table 8).The local cultivars grow by farmers were used as checks.
The objective of the experimentation was to identify the best and the farmers' preference hybrids in different sites.
After experimentations in 15 villages of Bougouni and Kati prefectures, five hybrids were identified (two Guinea-Caudatum/ Guinea-Caudatum, two Guinea-Caudatum /Guinea-Ecotypes and one Guinea-Ecotype/Guinea-Ecotype) as farmers' preferred hybrids with high yield and good grain quality compared with local cultivars used as checks (Table 9).These new hybrids were added to in to the National Catalog in 2013.

Hybrid seed production technique:
To promote sorghum hybrids, 608 people, composed of seed farmers, private seed companies, farmers' associations and extension services, were trained in hybrid seed production techniques, thus increasing the availability of hybrid seed to the producers (Plate 1 and 2) (Table 10).Indeed, in 2011 and 2012, 19,658 tons of hybrid seeds was produced and traded in Mali by the various actors.Thus, it can be assumed that the introduction of sorghum hybrids in a farming environment has become a reality in Mali.

Conclusion
The performance of hybrids and their parents was different in two environments, without and with phosphorus fertilization, for yield, 50% heading and plant hight.Yield reduction, delay of heading, and reduction of the plant height was seen in trials without phosphorus supply (-P) compared to trials with phosphorus supply (+P).Five hybrids 150A/02-SB-F4DT-298 (Sewa), 150A/Grinkan (Grinkanyerewolo), 12A/Lata 3 Bala (Fadda), PR3009A/Lata 3 Bala (Omba), FambeA/Lata 3 Bala (Pablo) and three parents: Lata 3 Bala (male), 150A and FambeA (females), were suitable for farmers in   sorghum production in Mali.These new hybrids were added to the National Catalog.The farmers were well-trained in sorghum hybrid seed production techniques and acquired good skills in hybrid seed production techniques.
Thus, the introduction of sorghum hybrids in a farming environment has become a reality in Mali.
However, further studies should be carried out in order to consolidate the results obtained.
= Males and Females interaction Repetitions and Blocks M = Males F = Females E * M = Environment-Males interaction E * F = Interaction-Environment Females M * F = Male-Female Interaction E * M * F = Environment-Male-Female interaction • Estimation of GCA and SCA effects GCA was computed as:

Figure 1 :
Figure 1: Average yield of hybrids combinations low and high-P environments (2007 and 2008).

Table 2 :
List of sorghum hybrids and checks (controls) used in the study.

Table 3 :
List of trials, site and data sowing.

Table 4 :
Analyses of variance (ANOVA) of combining ability (GCA & SCA) of parents and their F 1 hybrids for grain yield in different environments.

Table 5 :
GCA effects of male parent for grain yield under low and high-soil phosphorus conditions in Samanko station(2007 and 2008).
** : Significant at the 0.01 probability level; * : Significant at the 0.05 probability level

Table 6 :
GCA effects of female parents for grain yield under low and high-soil phosphorus conditions in Samanko station(2007 and 2008).

Table 7 :
ANOVA of plant height and 50% heading across low and high-soil phosphorus conditions.

Table 10 :
List of locations, actors and number of persons trained in hybrid seed production techniques from 2010 to 2012.

Table 8 :
List of villages involved in Bougouni and Kati prefectures.