EFFECT OF ROW SPACING, MINERAL AND ORGANIC FERTILIZERS ON YIELD AND GROWTH PARAMETERS OF PANICUM MAXIMUM ON FERRALITIC SOIL IN BENIN (WEST AFRICA).

This study proposes to evaluate the influence of the spacing row and NPK 15-15-15 fertilizer and goat manure on Panicum maximum seed production and fodder nutritive value harvested in south Benin on ferralitic soil from 2015 to 2017. The experiment was a factorial design comprising all combinations of 4 row spacings (20, 40, 60 and 80 cm) and 6 fertilizers: six levels of fertilizer: (F0) No goat manure (Control) (F1), 100 kg NPK 15-15-15, (F2) 20 t/ha goat manure, (F3) 15 t/ha goat manure, (F4) 10t/ha goat manure + 33kg NPK 15-15-15 and (F5) 5 tons + 66 Kg NPK/ha. Four replications were used in the experiment. The results showed that wider row spacing (60 cm) enhanced seed production in this experiment. Number of total tillers/tussock, fertile tillers/tussocks and spike length were significantly greater under the widest rows spacing (60 cm and 80 cm). The best seed yield got with row spacing 60 cm (392.1 kg.ha -1 ) and 45.5% of caryopses germinate after six months of storage against 14.5% for spikelet. Seed yield and its attributes increased significantly with NPK combined to goat manures (F4).

This study proposes to evaluate the influence of the spacing row and NPK 15-15-15 fertilizer and goat manure on Panicum maximum seed production and fodder nutritive value harvested in south Benin on ferralitic soil from 2015 to 2017. The experiment was a factorial design comprising all combinations of 4 row spacings (20, 40,  The results showed that wider row spacing (60 cm) enhanced seed production in this experiment. Number of total tillers/tussock, fertile tillers/tussocks and spike length were significantly greater under the widest rows spacing (60 cm and 80 cm). The best seed yield got with row spacing 60 cm (392.1 kg.ha -1 ) and 45.5% of caryopses germinate after six months of storage against 14.5% for spikelet. Seed yield and its attributes increased significantly with NPK combined to goat manures (F4).

…………………………………………………………………………………………………….... Introduction:-
Feed shortages and the poor quality of available feed are the major constraints to increased livestock productivity in sub-Saharan Africa (Hsu, 1999;Koura, 2015). In Benin, Natural pastures are the most important livestock feed resource (Lesse, 2016). They are low yielding and their production is insufficient and grazing periods are only favorable for four to five months per year. For several decades, grazing areas have been shrinking and likely to continue to do so because of rapid expansion of cultivated land for crop production to provide food for the everincreasing human population. As a result, there is always likely to be limited feed resources for the existing livestock population (Adjolohoun, 2008;Lesse, 2016;Mengistu et al., 2016). However, the extension and promotion of improved forage production packages is lagging behind and not progressing as expected. This is due to a number of limiting factors; among which scarcity of forage seed is the foremost ( The promotion of this crop in fodder production can reduce the need for livestock feed resources, increase the income of livestock breeders and reduce conflicts caused by transhumance between pastoralists and farmers among who have the land (Buldgen et al., 2001;Adjolohoun, 2008). According Gbenou et al. (2019), one of the most important factors that limit forage crop production in Benin is unavailability of seed. Seed production requires good soil fertility and mastery of production techniques. However, the soils in southern Benin are not very fertile. It is important to think of their fertilization for the production of fodder or forage seeds. Organic fertilizers including farmyard manure, poultry manure, cow manure and sheep manure may be used for the crop production as a substitute of the chemical fertilizers because the importance of the organic manures cannot be overlooked. Worldwide, there is growing interest in the use of organic manures due to depletion in the soil fertility (Gbenou et al., 2018a). Economic premiums for certified organic grains have been driving many transition decisions related to the organic farming (Delate and Camberdella, 2004). Continuous use of fertilizers creates potential polluting effect in the environment (Oad et al, 2004). Synthesis of chemical fertilizers consumes a large amount of energy and money. However, an organic farming with or without chemical fertilizers seems to be possible solution for these situations (Prabu et al., 2003) The aims of this study were to evaluate the influence of plant spacing and organic and inorganic fertilizers on the development, production and seed quality of Panicum maximum local in the southern Benin on a ferralitic soil.

Description of Study Area
The trial was conducted on an experimental field during rainy seasons of 2015 to 2017. It is located at Abomey-Calavi a Guinean Soudanian agroecological zone in southern Benin (West Africa) between Latitude 06°30'N and Longitude 2°40'E with an altitude of 50 m above sea level. The area is characterized by sub-equatorial climate with two rainy seasons (March to end of July and mid-September to November) and two dry seasons (August to mid-September and December to March). The area is dominated by ferralitic soil presenting the following characteristics: pH (2/5water) 6.2; N = 0.05 %; organic carbon = 0.4 %; Ca = 60 mg/100 g; Mg = 10 mg/100 g; P (extractable) = 0.2 mg/100 g and K(exchangeable) = 20 mg/100 g (Houndjo et al. 2018). The experimental site was previously cultivated without any organic fertilization, as reflected by the low organic carbon content of the soil. The analysis conducted on 0-10 cm soil-samples showed that sand, loam and clay represent approximately 84, 8 and 8 %, respectively.

Experimental Setup: Design and Treatments
The experiment consisted of factorial combinations of six levels of fertilizer: ( (24) treatments combinations were laid out in a split-plot design with plant row spacing assigned to the main-plot (block), while fertilizer (goat manure or NPK fertilizer) was assigned to the sub-plot (elementary plot).

18
The manure was well rooted goat manure that had stabilized for about 100 days. Elementary plot size was 7 m × 7 m (49 m²) with 3 m spacing between plots and 5 m between blocks. Each plot was replicated 4 times. Goat manure was applied 2 weeks before planting and the inorganic fertilizer NPK was applied 2 weeks after planting both for the single and combined applications. Goat manure was used in this study because of its availability in the study area. Manure level (20 t/ha goat manure) retained was based on assessment of the decomposition kinetic and the nutrient mineralization process of goat' dejection (Gbenou et al. 2018b). Also, the plants density and the inorganic fertilizer level (100 kg NPK) were chosen on the basis of the recommendations for density and fertilization of forage crops recommended by agricultural supervision services in the area and research results done on seed production of this specie (P. Maximum) and other grasses (Barbieri, 2000

Data Collection Physiologic data
Tiller density and time of floral initiation were recorded; inflorescence exsertion was deemed to occur when 50 per cent of the fertile tillers had exserted inflorescences. The plants were monitored on each plot to assess the frequency of emergence of the inflorescence. Likewise, 'days to seed maturity' was assessed as days taken from seedling emergence to first seed harvest. Seed maturity was recognized when the seed became hard, started changing color from green to brown, and when seed drop from the top racemes commenced.

Growth parameters, seed components and yield
Data were recorded throughout the experimental. Plant height and number of tillers per plant were assessed. Tillers were categorized as flowering or non-flowering and morphological characteristics of 10 flowering tillers selected randomly from each plot were measured for tiller height (base to the top of the inflorescence), tiller diameter (at the middle of the lowest internode) panicle length and panicle primary branch length. Number of panicle per tiller was recorded on 100 tillers selected randomly per plot. All seed within each plot were harvested each year over the period of seed maturity. Due to the varying row spacing, a net plot sampling area involved 10 and 5 rows of 2 m each for the 20 and 40 cm spacings respectively, then 4 rows of 1.8 m and 3 rows of 1.6 m for 60 and 80 cm spacings respectively. This corresponds to a sampling of a similar area of 4 m 2 for all treatments (row spacing).The Seed heads had been covered with a nylon net bag (Cover), with an outlet to collect the seed. Seed were collected every 3 days. Each year, after harvesting seed, between mi-July and late August, the plant tops were harvested to determine dry matter yield (Phaikaew et al., 2002;Kumar et al., 2005).

Seed processing and seed quality testing
Seeds of each experiments year were air-dried for 4 days after harvest in a seed shed before cleaning by using a jet of air. Seeds were afterwards stored in polyethylene bags at room temperature. Seeds were also collected from the rest of the field when shedding was on the way and stored in polyethylene bags at room temperature. Pure seed percentage was estimated using the international method using a 10 g sample for each plot, setting apart seeds and inert materials. 1,000 seed weight was estimated as the average of eight 100 pure seed replicates multiplied by 10 (ISTA, 2005). Germination tests were performed after both three and six months post-harvest storage at room temperature, because it have time period is reported as necessary for breaking dormancy and obtaining maximum germination (Matías et al., 1985). Four replicates of 100 seeds each were used in each treatment for obtaining the germination percentage.

Statistical analysis:-
Data were analyzed with the GLM (General Linear Model) procedure of SAS 8.02 software (SAS Inc., Cary, NC) using the following model : Yij= μ + Di + Fj +Yk + (D*F)ij + (D*Y)jk + (F*Y)jk +(D*F*Y)ijk + eijk. Where μ = mean, Di = year effect, row spacing effect, Fj = fertilizer effect, (D*F)ij, (D*Y)jk, (F*Y)jk, and (D*F*Y)ijk Yk their two or three ways interactions and eik the error term. Germination percentages were transformed to arcsine √%. When significant interaction was observed, data were re-analyzed separately by two-or one way analysis of variance. A significant was declared at P<0.05 (Phaikaew et al., 2002;Kumar et al., 2005).

Weather conditions
The average temperature and annual precipitation of the 2015 season were 26.5°C and 1048 mm. The 2016 season was slightly warm (27.0°C) and richer in precipitation (1054 mm). The average temperature and precipitation of the 2017 season were 26.8°C and 1011 respectively. The 2017 season was poorer in annual precipitation. Growth and development attributes and dry-matter production The effects of different treatments on various growth and development parameters are given in Table 3. Plant height increased significantly as row spacing increased with the main increase occurring between 60 and 80 cm (80cm˃ 60cm ˃40cm ˃20cm; p < 0.05). Highest dry matter yield (6.4 t/ha) was recorded at 60 cm row spacing in each year, being significantly greater than the 5.7 t/ha at 40 cm, 5.5t/ha at 80 cm and 4.4 t/ha at 20 cm row spacing (P < 0.01). The number of tillers/m 2 was highest at the narrow spacing (20 cm > 40 cm > 60 > 80 cm; P < 0.01) in the three experimental years but tiller diameter was also best at the widest spacing (P < 0.01). The data about these both growth parameters were similar between years (80cm˃ 60cm ˃40cm ˃20cm).
Application of fertilizer had a significant impact on growth parameters (P < 0.01). The greatest plant height and the best results of tillers number per m 2 , tiller diameter and dry matter were recorded at the F3 and F4 levels of fertilizer (15 t/ha goat manure and 10t/ha goat manure + 33kg NPK respectively).
On high plant density plots (20 * 20cm and 40 * 40 cm) flower emergence begins with a lateness of 6 to 10 days compared with low density of plants (60 cm * 60 cm and 80 cm * 80 cm) on which floral emerge started in 70 days following crop establishment (P <0,05). The gap is also very important for the time needed to obtain 50% of flowering emergence. Data on days to 50% flowering were significantly difference. Thus, the widely row spacing (80 cm) had an emergence frequency better than the others (80 cm < 60 cm< 40 cm < 20 cm). In general, flowering data showed a significant difference for start flowering time and days appropriate to obtain to 50% flowering between the fertilizer levels (F3< F4 < F1< F5 < F2 < F0, P < 0.05).

Seed yield and its attributes
Seed attributes were affected significantly by row spacing and fertilizer levels (p < 0.05) ( Table 4). The number of fertile tillers per plant, inflorescence number per tiller fertile, inflorescence length and spike length were significantly higher in the wide row spacing (60 * 60 cm and 80 cm * 80 cm) as compared with narrow row spacing treatments (20 * 20 cm and 40 * 40 cm). Number of fertile tillers per plant was significantly higher in low density crop than in high density crop 80> 60 > 40> 20 cm, P<0.05). The plant plots fertilized with 15 tons/ha goat manure and 10 tons goat manure + 33 kg NPK/ha produced more fertile tillers per plant, panicle/tiller fertile, panicle length and best spike length than others crops. But, fertilizer levels effects were inconsistent between treatments with on fertile tillers/tussock, Inflorescence length or spike length like growth and development attributes. However, the best results were obtained with 15 tons/ha and 10 tons + 33 kg NPK/ha fertilizers.
Seed yield and were affected significantly by both row spacing and fertilizer level (Table 5). Narrower spacing did not favor seed production of P. maximum in this experiment. In fact, the highest seed yield (392.1 kg/ha) was recorded from the 60 cm row spacing. This was significantly higher than yields for 80 cm (338 kg/ha), 40 cm (245.6 kg/ha) and 20cm (85.4 kg/ha) (P˂ 0.001). Overall, seed yield at the density 60 * 60 cm was 295% greater than that at the narrowest spacing seed at the widest spacing. Seed yield vary significantly with the different fertilizer (P˂ 0.001). The highest seed yield was recorded from the fertilizer F4 (10 tons + 33 kg NPK/ha) in second and third Years of the study.

Seed quality
Seed physical quality, in terms of 1,000 seed weight, showed significant differences (P<0.01) between treatments.
The greatest values 872.4 mg and 855.5 mg found at the 80 * 80 cm and 60 * 60 cm density respectively were not similar (P < 0.05) to those obtained in the 20 * 20 cm and 40 * 40 cm treatments (788.6 and 694 mg, respectively). Fertilizer using also significantly influenced the weight of 1,000 seeds (p <0.05). Plots receiving F3 and F4 fertilization levels (15 tons/ha and 10 tons + 33 kg NPK/ha) produced the best weights of 1,000 seed (860.3 and 836.7 mg respectively). Overall, the weight of caryopses accounted for 47.3% of the total weight of a seed (Table 6). In addition, plant spacing affected pure seed percentage (P<0.01) and the higher value (67 %) was found at the 80*80 cm distance. The second high value on seed purity level was got to the one found at the 60 * 60 cm (61.7 %). The 20 cm row spacing produced the low seed purity level (31.3%). No significant differences for levels fertilizer 20 influence (F1= F2 = F3 =F4= F5) on seed purity percentage between plots fertilized were found but the plots which didn't receive fertilizer (F0) produced the low seed purity level (42.7%) ( Table 7).
The table 8 shows the influence of row spacing and fertilizer input levels on seed germination. Row spacing and fertilizer using didn't significantly effect on P. maximum seed germination (P> 0.05). However, seeds harvested from wider spreading plots (60 cm and 80 cm) tend to germinate better than those from smaller spacing (20 cm and 40 cm). Caryopses germination percentages after 3 and 6 months were 7.5 (%) and 45.5% respectively, compared with germination percentages of the wrapped seeds (spikelet) which varied of 3.5 (%) to 14.5 (%) between 3 and 6 months (P < 0.01).

Discussion:-
Grass seed production is success favored in regions which are characterized by clear, sunny, warm days in combination with little or no rainfall during flowering. These climatic conditions promote good flowering of grass species and provide an environment conducive to pollinating activity of bees, two factors essential for seed production (Rincker et al. 1988). However, rain which fell readily during flowering during June would have had consequences on the efficiency of the tillers to give inflorescences ( Table 2). These rains would also have had negative effects on seed yield over the three years of the study. The choice of seeds harvest method by bags was to avoid massive losses. In contrast, the rain was less abundant during the harvest period (mid-July and late August).
Plants grown at low densities (60 cm * 60 cm and 80 cm * 80 cm) produce flowering stems faster than those grown in high density. On these plots where row spacing was widely, the frequency of emergence of inflorescences is high and 50% of inflorescences are produced in a short time. This might be due to the fact that wider inter row spacing had a better light interception as compared to the narrower row spacing. Further, more nutritional area available in wider row spacing might have caused the crop to flower earlier than the closer spacing (Farag et al.,1994). The synchronization of flowering stems emergence is useful to producer in defining the harvest period to avoid losses of seeds and working time. Knowledge of the processes related to plant development of P. maximum is extremely important for successful adoption of this species as forage crop. According to Hopkinson (1985), to understand the development of plants and their control which allows appropriate management for seed production, it is necessary to consider the morphology of the plant, the flowering physiology and the dynamics between the plant species and the environment at the time of flowering. The reproductive stage of most tropical and subtropical forage plant is influenced by photoperiod, temperature and humidity (Hopkinson 1985).
Data P. maximum species growth show that plant height is regulated by the genetic makeup of the plant and the environmental factors (Shahzad et al., 2007), and planting density determines the growing situation by affecting the competition for space and production resources. The current experiment showed significantly different plant height with varying plant density, the maximum plant height of 265 cm was obtained from 80 cm * 80 cm plant space followed by 60 cm * 60 cm plant space which was resulted with a plant height of 239.1 cm (Table 3). Minimum plant height was obtained with the lowest plant space (20 cm × 20 cm) which produced a plant height of 94 cm (Table 3) Panicles form the P. maximum inflorescence. It is the top part of the plant, carried on the last inter-node. Panicles are composed of primary ramifications (small branches) that carry secondary branches themselves carrying the pedicels which carry the spikelet. The length of panicle plays a vital role in Paniceae plants towards the spikelet per panicle and finally the seed yield (Shahzad et al., 2007). The analysis of variance for panicle and panicle primary branch lengths shows significant difference among the treatments. Thus, a relatively higher panicle length of 116.2 cm was 21 recorded from 80 cm * 80 cm plant space followed by 60 cm * 60 cm which produced 108.2 cm long spike. Further decrease in planting space beyond 60 cm * 60 cm also resulted in decline in the length of panicle (Table 3). Longer panicle at 60 cm * 60 cm plant space can be referred to the ideal plant population in this treatment, which resulted in optimum crop plant competition. These results were similar with those mentioned in literature that state that inflorescence size is correlated to plant density. This finding is in-line with the findings of Abbas et al. (2009) and Tewodros et al. (2017), who reported a decreasing trend in panicle length with increasing plant population. Similarly, Ozturk et al. (2006) reported more spike production per plant with increasing pattern of plant space in Oriza (Tritium aestivum). For example, the larger inflorescence was found in P. stratum ( Phaikaew et al., 2001b) at the 100*100 cm row and plant spacing, respectively, which was the largest spacing evaluated. Also, in C. ciliaris (Kumar et al.,2005) a larger panicle was observed when row width increased, and the largest at the 75 cm spacing. However, other authors (Matías et al., 1992;Febles et al., 1997;Sharma et al.,2002) report that plant spacing did not show significant effects on panicle length of tropical forage grasses. It is mentioned that inflorescence size is dependent on the initial size of the stem apex, on metabolite supply and on competition between the apex and other reproductive sites in plants (Humphreys et al., 1986).
The data showed statistically significant difference for seed yield. Plant space of 60 cm * 60 cm produced a higher seed yield of 392.1 kgha -1 followed by 80 cm * 80 cm plant space which produced a seed yield of 338 kg.ha -1 . The lowest grain yield was obtained from 40 cm * 40 cm plant space followed by 20 cm * 20 cm seed rate which produced 245.6 and 85.4 kgha -1 seed yield respectively (Table 4). This finding is in agreement with the finding of Baloch et al. (2010) and Tewodros et al. (2017) who reported a decline in yield of wheat with increasing plant density. This might be explained that dense plant population creates keen competitions between plants for production resources leading to a decreasing seed yield. Similar finding also reported by Hayatullah et al. (2000) who presented a decreasing pattern in grain yield with increasing plant density.
Seed yield varied significantly with fertilizer levels. Plots that received organic fertilizer alone or in combination with the mineral fertilizer gave the best yields. The highest seed yield (331.2 kg / ha) was recorded with the 10-ton fertilizer level + 33 kg NPK / ha. Good seed yield (305.4 kg / ha) was also obtained with the fertilization level of 15 tons / ha. The lowest yield was obtained with the fertilization level of 5 tons + 66 kg NPK / ha (282.0 kg / ha), an increase of 204% compared to control. The response to N and P fertilizer resulted mainly from a combination of increased number of fertile tillers, i.e. inflorescences, greater spike length and more spikelet/panicle. Loch et al. Combining goat manure with chemical fertilizer (NPK) treatments (F3 and F4) led to significantly higher yields and yield parameters than using chemical fertilizers alone. Therefore, it is clear that the combined application of organic manures and inorganic fertilizers is highly beneficial for sustainability in crop production (Khan et al., 2001). Based on the evaluation of soil quality indicators, Dutta et al. (2003) reported that the use of organic fertilizers together with chemical fertilizers, compared to the addition of organic fertilizers alone, had a higher positive effect on microbial biomass and hence soil health. Application of organic manure in combination with chemical fertilizer has been reported to increase absorption of N, P and K by the plant (Bokhtiar and Sakurai, 2005).
Seed physical quality indicated that plant distance influenced both 1,000 seed weight and pure seed percentage. The greater values for these parameters were obtained at the row spacing 80 * 80 cm followed by 60 * 60 cm, in coincidence with a larger pure seed yield. That shows that seed yield is a function of seeds containing caryopses (pure seed) and of seed weight. On this matter a positive effect of plant density on seed physical purity percentage is mentioned for other tropical forage grasses, P. Atratum (Phaikaew et al., 2001a) and B. decumbens cv. Basilisk (Matías et al., 1992) where plant density affected 1,000 seed weight. Regarding this, it is mentioned that the first 22 seeds to fall are those heavier and more mature (Humphreys et al., 1986), so pure seed yield drops. The interesting pure seed percentage and high 1,000 seed weight found in the 60 * 60 cm spacing are reflected in a larger seed yield. Although no significant effect of plant spacing and levels fertilizer on seed germination was observed, the germination rate of seeds varies according to their physical forms. Indeed 45.5% of caryopses germinate after six months of storage against 14.5% for spikelet. The germination rate varies considerably between the two test periods (3 and 6 months after storage). Other authors (Matías et al., 1992) report similar findings for seed germination in B. decumbens cv. Basilisk. However, some authors report that plant spacing increased seed germination rate in C. ciliaris (Kumar et al., 2005) and P. maximum cv. Común (Febles et al., 1997). This discrepancy could be due to several factors, as forage species, harvest date, storing period and conditions, germination rate methodology, postharvest management and weather at the production and storage periods.
Moreover, this difference can be explained by the dormancy state of the seeds. Dormancy is considered as a temporary suspension of growth of any plant structure containing a meristem (Rezvani et al., 2016). These phenomena can help plants to adapt to a variety of habitats and climates and survive in different ecosystems. Seed dormancy can serve to synchronize germination so that probability of seedling survival is optimized (Baskin and Baskin, 1998). Thus, our study shown that combinational dormancy occurs in P. maximum seeds that have both exogenous (physical) and endogenous (physiological) dormancy.

Conclusion:-
In conclusion, the results from the study indicated that row spacing and goat manure and inorganics (NKP) fertilizers had a significant influence on the phenology, growth, yield components and yield seed of Panicum maximum. Seed production potential of Panicum maximum can to be improved with both 10 tons (goat manure)+ 33 kg NPK/ha or 15 tons/ha (goat manure) fertilizers on 60 cm x 60cm plant density plots on ferralitic soils in Benin (West Africa). Increases in seed yield can be attributed to larger pure seed percentage, greater seed weight and high panicle length. Seed physiological quality, as defined by germination, did not improve significantly in any of the evaluated treatments. The soils fertility improvement and adequate plant density choice are required to the wider use of this species in livestock systems. However, as this is one location alone, the experiment has to be repeated over locations and seasons with inclusion of more ecotypes of P. maximum to reach at a more reliable conclusion. Application of fertilizer will increase seed yields but economic considerations would determine whether these increases are profitable. An added benefit of wider row spacings is that seed produced is bigger and results in more vigorous seedlings.   )  2015  2016  2017  2015  2016  2017  January  26  28  27  27  0  0  February  27  28  25  0  0  0  March  27  28  29  55  11  33  April  29  27  27  155  203  169  May  26  28  28  153  133  175  Jun  26  27  26  219  217  200  July  25  26  26  118  110  122  August  25  26  26  55  101  89  September  26  27  27  112  167  105  October  28  26  26  131  109  113  November  25  25  27  23  0  3  December  28  28  28  0  3  2  Total  --1048  1054  1011 23    For the same column and the same treatment, means followed by different lower case letters are significantly different at p ˂ 0.05;