Rice is the world’s largest cereal in terms of human food consumption [1] [2] . Once a luxury food and consumed during the festive season in the Democratic Republic of Congo (DRC), it is now the main staplefood in some provinces, but also in other least developed countries. Along with wheat, it is the most consumed cereal in the world [1] . However, rice cultivation in the DR-Congo is mainly rain-fed, although the country offers enormous potential for irrigated rice cultivation [2] . Despite the efforts made by the Government and partners to increase production through strategic development projects and improved varieties, domestic production of this commodity remains low [2] . Neverthless, rice is the second most consumed cereal by the population in the DR-Congo, after maize [3] . In tropical regions, the crop is subject to drastic biotic and abiotic constraints, leading to threats of genetic erosion [4] . In addition, it has been pointed out that agricultural production on land in most African countries is being treated by population pressure, which is growing faster than in other regions [5] . In addition, poor land management practices subsequently lead to soil nutrient depletion [6] . It is the levels and types of soil elements during the crop cycle that determine the quality of plant mineral nutrition and largely the quantitative yields of crops [7] . Considering current socio-economic and environmental challenges, it is imperative to look for other sources of nutrients that can allow sustainable agriculture at a time when the import of agricultural products is breaking records due to low crop yields.

To compensate for the decline in yields, several approaches can be considered, such as the use of organic fertilization and crop rotation [8] [9] . The use of mineral fertilization is cost prohibif due low rice farmer’s income in the DR-Congo. The implementation of crop rotation requires knowledge of specific techniques. Several studies have suggested that producers adopt organic fertilizers as a crop management alternative aimed at reducing or eliminating chemical fertilizers. The decomposition of organic residues significantly improves the level of nutrients and organic matter in the soil [10] . However, the use of organic amendments such as organic waste is poorly documented in the Mbujimayi region, even though they are an optimal source of nutrients and contain 50% to 90% organic matter. The mineral fertilization proposed as a solution remains fragile on soils such as oxisols with high mobility of aluminum and iron precipitating the other nutrients applied to the soil [8] [9] [11] . In addition, the low-income level of farmers in the region and the lack of training in the application of mineral fertilizers often constitute obstacles to the use of mineral fertilizers. In addition, the increase in population has led to an intensification of agricultural practices and an extension of cultivated areas, which has resulted in a reduction in the fallow time. This situation not only accentuates soil erosion but also leads to a rapid depletion of its nutrients, particularly nitrogen, phosphorus and potassium [12] .

The development of local resources that are not exploited, such as bat guano, Tithonia diversifolia, male inflorescences of oil palm, cow dung and chicken droppings are proposed [13] . These local resources are of interest to certain regions of the D. R. Congo in general, as they contain elevated amounts of plant nutrients [13] . T. diversifolia is rich in nitrogen, phosphorus and potassium and possesses other properties [14] - [17] . It decomposes rapidly after application, resulting in the improvement of soil physical, chemical and biological properties and in the increase of nutrient availability [15] . Hence, we hypothesize that the use of foliaceous biomass of T. diversifolia combined with NPK_17-17-17 + urea (46% N) would boost the rice yield compared to mineral fertilization with NPK_17-17-17 + urea (46% N) used without any combination with organic fertilizer in the edapho-climatic conditions of Mbujimayi. The main objective of this study was to evaluate the effect of T. diversifolia and the mineral combination (NPK_17-17-17 + Urea (45% N)) on morpho-agronomic characteristics of rice (Oryza sativa).

The results for morphological and agronomic characteristics are presented in Tables 2-3 . The effects of organic and inorganic fertilization were observed based the selected parameters studied. After statistical analysis of the data collected in the field, it is observed that the combination 1/2 (BFT – 4 kg + NP) generated similar diameter at the collar as the NP mineral fertilizer with 4.33 cm and 4.16 cm, respectively. Nevertheless, this diameter at the collar was significantly greater than those (between 3 cm and 3.33 cm) observed under control and other treatments tested inputs. The same trend was observed on almost other vegetative parameters such as plant height and the number of tillers per pocket ( Table 2 ). Indeed, plants treated with 1/2 combination (BFT – 4 kg + NP), BFT – 4 kg, and NP showed similar height (100.93 cm, 99.03 cm, and 98.63 cm, respectively) that were significantly higher than control and other treatments [1/2 (BFT – 2 kg + NP, BFT – 2 kg] ( Table 2 ). Higher number of tillers/pocket (6.33) was also observed in plants treated with (BFT – 4 kg + NP), BFT – 4 kg, and NP compared to the control and other organic and inorganic treatements.

In the columns, the means followed by the same letter are not significantly different at the 5% threshold following the LSD test.

In the columns, the averages followed by the same letter are not significantly different at the 5% threshold following the LSD test.

For agronomic charcateristsics, days to 50% flowering varied between 73.00 and 74 days with an average of 74 days. There were no clear significant differences among treatments for days to 50% flowering, days to 50% maturutity and the length of panicles ( Table 3 ). The length of the panicles varied between 24 and 29 cm with an average of 27.00 cm. The control and BFT – 4 kg showed significantly shorter panicles compared to other treatments. For yield components, 1/2 (BFT – 4 kg + NP) and the NP treatments generated a higher weight of 1000 grains. For yield per hectare, 1/2 (BFT – 4 kg + NP) induced significantly different levels of production than the control and other treatments incuding 1/2 (BFT – 2 kg + NP), BFT – 4 kg + NP, BFT – 2 kg, BFT – 4 kg. This is depicted in Figure 3. The rate of seed absorption was very low in all the treatments, an indication of the good growing conditions and genetic attribute of the Lioto variety used.

The correlation coefficients between agronomic traits revealed that with the exception of length of particle and the abortion rates, all the yield components (panicles per plant, seeds per panicle, weight of 1000 grains, and grail yield per plot) were strongly correlated with grain yield per hectare ( Table 4 ). In fact, the Pearson correlation values varied from 0.81 to 0.99 for these traits and the grain yild and only 0.41 for length of panicles and the abortion rate.

The results obtained in this study show that in relation to vegetative variables such as emergence rate, plant height in cm, and number of tills/pocket, the treatment based on the 1/2 combination (BFT – 4 kg + NP) was better, although not significantly different from rice under the addition of the chemical fertilizer NP. This can be justified by the fact that mineral fertiliser releases nutrients quickly and is used excessively by the plant, compared to organic fertiliser, which takes time to break down and maintain optimum nutrients that stimulate plant growth in the long term. De Ridder and Van Kaulem [19] and Muyayabantu et al. [20] reported that the use of inorganic and organic fertilizers often leads to synergy and improved efficiency in nutrient and water use. This could be the case for certain soil inputs such as the 1/2 combination (BFT – 4 kg + NP) which similarly increased the diameter at the collar as the NPK mineral input with 4.33 cm and 4.16 cm, respectively. Nevertheless, this diameter at the collar is significantly greater than those (between 3 cm and 3.33 cm) observed under control and other inputs to the ground, with no significant difference between them. The same trend was maintained on almost other vegetative parameters such as the height of the plants and the number of tillers per pocket. Indeed, rice under-treatment based on the 1/2 combination (BFT – 4 kg + NP) significantly shows better growth in height (100.93 cm) and a good number of tillers (6.33) compared to rice under other soil inputs. Rice under NP and BFT – 4 kg comes in second place, while the lowest height was found in rice that was not fertilized (control), although it generated a similar number of tillers to rice under other treatments. The variability in height observed during this study is likely due to soil variability.

The combination of organic (T. diversifolia) and inorganic (NP) nutrient [1/2 (BFT – 4 kg + NP] sources resulted in higher grain yields than all other treatments. This is likely due to increased nitrogen and phosphorus availability as well as improvements in other soil parameters. These results are consistent with the findings reported by Steiner et al. [21] . These studies concluded that replenishment of the nutrients available to the plants by the addition of mineral fertilizers alone is not satisfactory to maintain soil fertility in freely drained soils. In addition, as reported in many studies, the integration of organic and inorganic nutrient inputs increases the efficiency of the use of these fertilizers and provides a more balanced supply of nutrients to crops. Gao et al. [22] reported significant increases in maize yield following the application of green manures. Leaves incorporated into the soil (as green manure) at the beginning of the season decomposed and would have released nutrients, especially nitrogen, which improved crop performance. It should be noted that the amount of nutrients provided by organic matter is highly dependent on the amount of organic matter applied [21] [22] . In all the cases, the untreated plot always yielded lower grain production compared to organic and inorganic fertilisers.

As for the weight of 1000 grains, it appears at the same pace as for the number of grains per panicle. The soil contribution of the 1/2 combination (BFT – 4 kg + NP) further confirms its superiority with a weight of 1000 grains of rice (42.66), although statistically similar to the intake of NP (42.00), remains, however, significantly high to that of rice under other treatments. The lowest weight of 1000 rice grains was recorded under control, which remained similar to the intake of BFT − 2 kg and BFT − 1/2 (2 kg + NP). On the other hand, the greater the weight of 1000 grains, the more seed to be used per hectare, which could have a positive impact with the varieties that are evolving, given the low income of farmers in the region. However, rice under 1/2 combination (BFT – 4 kg + NP) gave a plot production (499.00 g) statistically (p = 0).05) compared to rice under other treatments followed by the addition of NP chemical fertilizers (431.0 g). Unfertilized rice gave a significantly lower return (348.67 g). As the yield is derived from the extrapolation of plot production, the same tenadance is maintained. Indeed, rice under 1/2 (BFT – 4 kg + NP) significantly retains its superiority over other treatments, with a yield of 2.21 t·ha⁻¹. The control rice was the one that gave the lowest possible yield, significantly lower (1.54 t·ha⁻¹) than the rice under other treatments, which remained intermediate. These results corroborate with those found at INERA Yangambi, in a controlled environment, the variety used in our study (Lioto) produced between 2000 and 3000 kg/ha; and in peasant areas, it produced between 1500 kg/ha and 2000 kg/ha. According to Nziguheba et al. [23] , Tithonia diversifolia is used alone as a fertilizer, but combined with phosphorus fertilizers, can double or even triple the harvests. In the same given environment, the interactions of plants are dependent on a complex whole, it can also be noted that the good yield in rice could not be due solely to the application of fertilizers. The absence of attacks, diseases, predators and good climatic conditions have also made it possible to have the best yields in this environment. The soil of the City of Mbujimayi is a sandy-clay soil made up of 15% clay. However, this rice would be best suited to a soil that can buffer climatic variations having a good water-holding capacity, contains a roughly equal proportion of clay, sand, silt and a pH varying between 4.5 and 8.7 [24] . However, the other constraint is the retention of soil moisture, which is important because upland rice depends above all on water can be major constraints for rice production.

Overall, the effects of T. diversifolia observed in this rice trial are consistent with data observed in other crops. Setyowati et al. 2018 demonstrated that Tithonia compost increase plant height, plant dry weight, curd diameter, as well as curd weight of cauliflower. Ewané et al. [25] showed that T. diversifolia leaves and stems treatment increases the number of shoots, the height and the diameter of shoots as well as the area of shoots leaves compared to the control in plantain crops. Dayo-Olagbende et al. [26] showed that the application of tithonia mulch improved growth, and yield indices of maize as well as soil physical and chemical properties. Likewise, Muyayabantu et al. [8] demonstrated that the combination of NP with T. diversifolia or E. abyssinica leaves resulted in the highest increase in maize grain yields in Gandajika. These authors laters showed that the use of T. Diversifolia resulted in the highest monetary advantage index (MAI) compared to the use of inorganic fertilizers and corp mixtures [9] .

Thanks to Dr. Kabwe Nkongolo, Laurentian University, Sudbury, Ontario, Canada, for reviewing the manuscript and “Université Officielle de Mbuji Mayi” for administrative support.