The experiment was carried out at the National Forest Seeds Program laboratory and at the Mbao forest service site located in the Pikine districk in Dakar (Figure 1). Two experiments were carried out on the site: the first was carried out between May and June 2018 in the laboratory and the second between June and November 2018 in the nursery.

The vegetal material used was the seeds of M. charantia collected between November and January 2017 in the Niayes area (Dakar). Fruits were harvested at maturity to ensure good seed quality. After the fruit extraction phase, the seeds obtained were washed and dried to the open air in the sun (Figure 2). Then the dried seeds were sorted manually to eliminate the bad quality ones. Finally, the sorted healthy seeds were used for the germination test according to the pretreatments.

In the laboratory, four pretreatments were tested in germination boxes. Each pretreatment had two replicates containing 50 seeds. All the germination boxes were distributed according to a random arrangement (Figure 3(a) & Figure 3(b)). Four M. charantia seed lots were subjected to the pretreatments which are: (Lot1) control, that is to say untreated seeds; (Lot2) seeds soaked in tap water for 24 hours followed by rinsing the seeds three times in distilled water; (Lot3) seeds soaked in hot water at 90˚C for 24 hours followed by rinsing the seeds three times in tap water; (Lot4) soaking in boiling water at 100˚C for 24 hours followed by rinsing the seeds three times in distilled water (photo). Then the four lots of 100 seeds each are divided into lots of 50 seeds placed in different places. Of these 08 lots, four are exposed to light and the other four are exposed to darkness. For germination, sand sterilized in an oven at 150˚C for 24 hours to kill all microorganisms was used.

In this part, the substrate used was a mixture of erosive soil (sand) and manure. The mixture had been made up of the following different proportion/ratio: 2/3 sand + 1/3 manure. Black polyethylene bags of 21 cm × 7.5 cm were used for potting purposes. Before sowing in the bags, the seeds were subjected to pretreatments in the laboratory.

The sowing was done with two seeds per pot, to a depth of approximately 1.5 cm. The experiment was carried out in the PRONASEF (national forest seeds program) botanical garden at the Mbao water and forest site. All pots were placed in the nursery and watered regularly according to a completely random block arrangement with two treatments. The treatment (T1) is watered every day morning and afternoon, and the treatment (T2) watered every two days morning

and afternonn. Each experimental unit consists of four (4) rows of 100 pots for a total of 400 pots with each pot a seed. Watering was carried out according to the frequency of irrigation at the rate of half liter per pot. In the nursery, the emergence of seedlings was monitored every day from the third day of sowing between the beginning of July and the end of August 2018. The emergence was taken into account as soon as the seedling was visible. A systematic count of the seedlings that have emerged until the end of germination was carried out every day until the 30th day of germination. A random sample of four pots per repetition and per batch, is taken every week for two months to measure the height and the number of leaves. Several yield parameters such as the number of fruit per plant, the average number of seeds per fruit and the yield of seeds were evaluated.

To determine the germination rate, we considered the number of seeds emerged and the emergence of seedlings. Thus a seed was considered to have germinated when the cotyledons separate to allow the radicle to emerge. Therefore the total germination rate, the germination rates by pretreatment were calculated. Growth parameters such as height and number of leaves were evaluated. These parameters were determined at eight-day intervals from 10 to 30 days after sowing (DAS). After three (3) months, the experiment ended with the harvest of fruits in order to determine the number of fruit per plant, the number of seeds per fruit and the total yield per plot of 150 m².

The germination criteria and parameters are those defined by Soumana et al. [12] according to which the latency time (germination time) is the time elapsed between sowing and the first germination. The duration of germination or staggering is the time between the first and the last germination. A seed is considered germinated when the radicle pierces the seed coats. The counting of germinated seeds is carried out daily for 30 days. The parameters studied are: The germination rate (TG) which represents the number of seeds germinated compared to the number of total seeds germinated. This parameter is a better way to determine the germination capacity of seeds. It is expressed as a percentage and calculated from the type of pretreatment and the substrate according to the following formula [13]:

Germination rate ( % ) = Number of seeds germinated Total number of seeds sown × 100

The germination kinetics (the number of seeds germinated) was noted every 24 hours for 30 days. Average Germination Time (GMT) which was the number of seeds germinated relative to the number of days (time) is determined. In this study, data such as germination rate, height, number of leaves, number of fruit per plant, number of seed per fruit and total yield were collected and placed in tables of database using Excel 2013. ANOVAs were used to compare means of variables by Fisher’s test (multiple comparison tests with Tukey’s test) at the 5% level.

The change in the height overtime depended overtime on the pretreatments (Figure 10). The increasing trend of evolution of all the curves obtained was almost the same. But the height growth rate was faster in Lots 1 and 2 than in Lots 3 and 4.

The evolution of the number of leaves was higher in the plants from Lots 1 and 2 compared to the other two pretreatments (Figure 11). The slight drop observed at the level of the curves of the plants pretreated with hot water (Lot3) and those treated with boiling water (Lot4), would be due to an elongation of the main stem, which occurred between the 39th and 60th days after sowing (DAS).

The results showed that the frequency of watering had a significant influence on the growth parameters (Figure 12(a) & Figure 12(b)). In fact, watering every day promoted more growth in height (4.5 m) and number of leaves (98) than watering every two days (3.08 m and 86 leaves). The pretreatments did not significantly influence the growth parameters (Table 2). However, the average height of a M. charantia plant was 3.75 m with an average leaf count of 92.03.

Analysis of variance and Tukey’s test showed that the frequency of watering had a significant influence (p = 0.00) on fruit and seed yields (Figure 13(a) & Figure 13(b)). Watering every day produced more yield (30 fruits/plant; 15 seeds/Fruit; 2.4 kg of fruit and 700 g of seeds) than watering every Two days (18 fruits/plant; 9 seeds/fruit; 1.4 kg of fruit and 350 g of seeds). According to the results obtained, the pretreatments significantly influenced the yields parameters such the number of fruits per plant (p = 0.020), the weight of fruits (p = 0.003) and the weight of seeds (p = 0.019). On the other hand, the pretreatments had no significant effect on the number of seeds per fruit (p = 0.238). The higher yields in Lots 1 and 2 than Lots 3 and 4 were recorded (Table 3).

Germination was an important prerequisite for continued growth and other physiological processes, including productivity and fruit and seed yield. According to our results, the germination of the seeds was not influenced by the light factor (Figure 7) during the entire experiment. According to Johnson et al. [14] who worked in greenhouses, the presence of light could sometimes shorten germination time while reducing cumulative germination rates by 10% to 20%. What contradicts our results or the presence of light has no influence on the germination rate. In addition, the action of heat seems to be detrimental to the germination of the seeds, probably because of their high oil content [15]. However, the germination rate is strongly influenced by the different pretreatments. Indeed, the germination speed of seeds without pretreatment (Lot1) and those soaked in tap water (Lot2) is the fastest with a short lag time. Germination is started from the 4th and 5th day, it is rapid during the first days. On the other hand for the last two batches, we observe a longer latency time. So the pretreatments coated a slowdown in the speed of seed germination leading to a decrease in the germination rate. This result is in line with that obtained by Johnson et al. [14] who found that the improvement in germination was due to the weakening of the seed coats and thus making them permeable. Indeed, Benadjaoud and Aïd [16] reported that soaking in hot water at more than 70˚C did not allow any

NF/P: Number of fruits per plant; NS/F: Number of Seeds per fruit (kg); Y.F: Fruit yield; YS: Seed yield (g).

germination after 4 weeks of tests on argan seeds. According to the authors, it is likely that this type of treatment has a depressive, even lethal, effect on the seeds. Based on the results, immersing the seeds for 24 hours in tap water results in a high germination rate. This pretreatment allowed the softening of the seed coats by promoting germination. Indeed, the pre-soaking induces in the seed a drop in the water potential of the embryo which leads to the synthesis of new substances of low molecular weight and this leads to a production of the energy cited by [17]. Another hypothesis which could explain this difference in the germination rate would be an intense hormonal activity in the seed coats which would temporarily inhibit this germination. However, this inhibition would be lifted by soaking the seeds in lukewarm water for 24 hours. Our results are in agreement with those found by Pamo et al. [18] on seeds soaked in lukewarm water between 12 and 72 hours provide the best germination rate (90%). These germination rates are in phase with those obtained by Bembe et al. [4] on three cucurbits (Lagenariasiceraria,Citrullus lanatus and Cucurbitamoschata) with respectively 78.57%; 90.42%; 92.85%. Staying the seeds in tap water for 1 day did not have too much of an effect on the germination rate but reduced the germination time and delay by 3 days compared to the control. These results are in agreement with those reported by Ado et al. [5] on the seeds of DiospyrosMespiliformis. Furthermore, it should be noted that the duration of soaking the seeds in boiling water depends on the thickness and hardness of the seed coat [19]. The experiment carried out with the four (4) treatments showed that the seeds not pretreated and those treated with tap water gave a better percentage of germination than those treated with hot and boiling water. These results support the hypothesis of that each cultivar in cultivation requires a critical soaking time and it should be less than the safe limit [20]. According to these authors, having a thick seed coat does not allow good water absorption in the case of bitter gourd. This therefore results in late germination and also seed dormancy. Soaking seeds before sowing may be beneficial in this regard, but soaking time and heating are not very favorable for germination [20]. Soaking helps absorb water easily and also removes some pathogens from the surface of the seeds. Seed priming, the controlled hydration and dehydration of seeds, is widely used to increase the rate and uniformity of establishment of seedlings of commercial vegetable and flower seeds [21]. In addition, Kanwar et al. [22] reported that soaking may increase the activity of glyoxisome enzymes in bitter gourd seeds. In other words, germination and growth depend not only on the types of seeds, but also strongly influenced by many internal and external factors. For example, Harris et al. [20] reported that in winter germination of bitter gourd becomes late due to the soil temperature being lower than the required temperature. So at the moment the hot water treatment of the seeds allows better and earlier germination even in winter. A variety of cultivated species have shown faster germination, early emergence and vigorous sowing by soaking the seeds in hot water for the same time and then drying on the surface before sowing, which can result in a higher yield high cultures [20] [23]. The mean germination time is strongly influenced by the different pretreatments used. The best germination time (5 days) is obtained from seeds which have been soaked in tap water for 24 hours. Seeds treated with hot and boiling water also showed a short mean germination time (6 to 14 days). These results are consistent with those found by Achigan-Dako et al. [24] on three species of Egusi used in food in Benin. According to these authors, the germination time of these three species varies between 5 and 10 days. This same time interval was obtained by Zoro Bi et al. [2] on Phaseoluslunatus in Côte d’Ivoire.

The growth dynamics of seedlings in the nursery revealed important information. Like the germination percentage, the growth (height and number of leaves) of plants was influenced by the quality of the seeds and the frequency of watering. However, the pretreatment factor has no influence on plant growth. After 3 months of the test, the average height of the plants is between 2.4 m and 1.5 m respectively for watering every day and every two (2) days (Figure 10). There is therefore a clear correlation between growth in plant size and germination rates. So plants watered every day grow faster than those watered every other day. We can assume that growth hormones would be present in large quantities in these seedlings watered every day. Therefore, the growth rate of M. charantia is better for the plants grown from the seeds of Lot2 (tap water). In a study conducted by Braconnier et Bourou [17] in India on six varieties of millet (Eleusinecoracana L.) shows that plants grown from seeds pre-soaked with lukewarm water reduced the time to enter by 50% in bloom. Moreover, the study of juvenile growth of M. charantia has shown that this species grows rhythmically. It is a mode of growth whereby a stem or root meristem manifests cycles of morphogenetic activity passing through maximums and minimums or complete stops, at more or less equal time intervals [25]. This type of growth has also been demonstrated on plantlets of Terminaliaavicennioides and Piliostigmathonningii [26]. The rhythmic growth of the aerial stem of seedlings is a consequence of the rhythmic functioning of the apical meristem. The waves of growth are linked to the bud burst, which manifests itself in the opening of the leaves and the successive formation of internodes [26]. Momordicacharantia recorded the best height growth (4.5 m) in 60 days in the nursery. There is thus a variation in height growth and number of leaves with the type of watering. This variation could be explained not only by the physiological and genetic state of the seeds used, but also by the quality of the substrates used and the silvicultural care, in particular the watering given to the plants [27]. The leaves and the blade are important parameters in measuring growth and yield. This is because the seed when produced with healthier, well-watered seedlings, the plant will produce the best quality leaf, in length, width, color, size and shape. This may be due to earlier germination and less susceptible to pathogen attack as the treated seeds were washed in lukewarm water [28]. Moreover, the good aptitude for foliar production of Momordicacharantia from an early age would constitute a factor which determines its important medicinal use by the population [9]. In addition, watering stimulates and produces enzymes like amylase and lipase which activate storage materials in the seeds and mobilize anti-oxidant enzymes which subordinate to oxidation in the seeds which keep the seeds vigorous. Causing earlier emergence of limbo. Rehydration causes early emergence by influencing the prégerminatifs germination processes [29]. Ultimately, the watering frequency factor promotes crop growth, the number of branches/plant and plant yield [23]. According to the results obtained, the average time required for the appearance of the first flowers is 1 month 28 days. It emerges from this study that the speed of flowering of the feet is fast and variable among the plants. The flowering period for the 15 followed plants is the month of September and the average length of fruit maturity turns around 14 to 15 days. Furthermore, the results revealed a faster flowering speed for 8, 9, 10, 11 and 12 plants with a fixed maturity time of 15 days. These five (5) feet were under the shade of the trees. According to this experience, the ideal fruit harvest period would be October, three (3) and a half months after sowing. Experience has shown that it takes two months to produce flowering plants of M. charantia. However, variations in the number of days before the first flowering were found to be very important for different pretreatments. The days required for the first flowering were recorded with frequent observations. However, the study also revealed that such a height of the plants is acquired on condition of using bags (plastic bags) as container and a substrate of clay-silty structure. Indeed, the good growth of M. charantia seedlings was observed on a compost made up of two thirds (2/3) of clayey soil mixed with one third (1/3) of silty soil taken under cover of trees in the forest. . From the phenological data, it appears that the average time of appearance of flower buds of M. charantia is 38 days after sowing. These results are in line with those obtained by Bembe et al. [4] on Citrulluslanatus 31 days after sowing and 37 days after sowing in Lagenariasiceraria and Cucurbitamoschata. Thus, the duration of the production cycle of the species studied is approximately 120 days, which is identical to that observed by Zoro Bi et al. [2]. According to the author, like all other families of cultivated leafy and vegetable plants, Cucurbitaceae have more or less similar vegetative characteristics. It emerges from this study that the average maturity of the fruits is two weeks and the ideal harvest period is between October and September. These results are consistent with those obtained by Balde et al. [9] during the ethnobotanical investigation of M. charantia in Senegal. According to these authors, the maximum period of maturity of the fruit is between 15 and 20 days (85.00%) and the best time to harvest the fruit is the end of October and the beginning of November (91.50 In addition, Achigan-Dako et al. [24] observed in Cucumeropsisedulis a longer flowering time of 57 days. According to these authors, this species has the longest production cycle of all cucurbits ahead of the genus Momordica.

From the results obtained, the average yield of fruits and seeds is strongly influenced by the pretreatments and the frequency factor of watering. Thus, the best yield of fruit and seeds is obtained by pretreatment with tap water (lot2) and the control (lot1). It appears from this study that watering every day gives a better yield of fruits and seeds. Consequently, M. charantia produces more fruit per plant than that of Cucumeropsisedulis which produces an average of 5 fruits per plant [24]. Our results are superior to those obtained by Bembe et al. [4] on Lagenariasiceraria 332.80 g; Cucurbitamoschata 272.40 g and Citrulluslanatus 159.20 g. The relationship between seed yield and number of fruits per plant can be a linear regression. Analysis of the latter can be considered as the best estimator of seed production for this plant. However, compensation in total seed yield is possible when the number of fruits per plant increases. Under growing conditions in West Africa, the performance that interests both the producer and the end consumer is the seed yield for the cucurbits used as Egusi and the genus Momordica [2] [24]. Thus, in this region of the world, cultivation programs for M. charantia will have to focus instead on increasing the seed yield for its oil.