Ecological environment effects on germination and seedling morphology in Parkia biglobosa in nursery (Côte d’Ivoire) and greenhouse (France)

— Néré (Parkia biglobosa) is a wild species preferred and overexploited for its multiple uses by rural populations in Sub-Saharan Africa. The study of its germination and seedlings could constitute a prerequisite for its domestication, necessary for its conservation. This study aimed to assess the germination and morphology of seedlings taking into account distinct habitats from its natural environment.A total of 2160 seeds from different mother plants and 540 seedlings from germination were selected and evaluated. The trials were conducted on three sites (two nurseries in Côte d'Ivoire vs one greenhouse in France) with different microclimates. The results showed that the larger the mother trees are, the larger the seeds they produce, which in turn generate more vigorous seedlings. This study showed that the species grows better in a milder environment that is different from its region of origin (fertile soil with a stable or humid tropical climate: Montpellier greenhouse and Daloa nursery). Overall, parent trees did not statistically influence each germination and seedling development parameter for the three sites combined (P > 0.05). However, analysis of variance showed that germination and seedling development parameters differed between experimental sites (P < 0.05). These results are useful and could be used as decision support tools to guide conservation (domestication) and agroforestry programmes based on Parkia biglobosa. This study could be extended to other endangered species in order to preserve biodiversity.


INTRODUCTION
The existence of many species is threatened by poverty combined with a galloping demography in most tropical countries (Chupezi et al. 2009;Houndonougbo et al. 2020). Indeed, phytogenetic resources, in particular native forest, fruit and agroforestry species constitute an important source of subsistence and income for rural populations. In Africa, studies have shown that around 55% of plant species are endangered in forests, of which 10% are already extinct in the wild (Aké 1999; Maréchal et al. 2014). As a scientific challenge, it seems urgent to develop conservation strategies to compensate for the ever-increasing pressures on certain species that could lead to their extinctions, while taking human needs into account.In African savannah zone, trees and shrubs play a key role in the ecological balance in the face of the advance of the desert (Maazou et al. 2017). These trees are also the main source of goods and services essential to populations (Avana-Tientcheu et al. 2019). However, these trees are overexploited for their multiple roles (food, medicine, timber and firewood, cultural, rituals etc.) by the populations, but also threatened mainly by the strong demography, climatic variations and systems of land use  Maazou et al. 2017). The species is distributed from Senegal to Uganda and is highly valued for its seeds fermented in "Soumbara" or African mustard; a very nutritious and highly prized product to enhance the taste of various dishes in many Sahelian countries (Azokpota et al. 2006). Likewise, the pulp of néré seeds is eaten directly or combined with wheat flour, corn or millet to make donuts Despite its important socio-economic role and the threat to the genetic diversity of its stands, Parkia biglobosa is so far in the wild and few studies are available on its regeneration, conservation and sustainable management. In our opinion, taking human needs (food security) into account in developing a strategy for the sustainable management of the species should also consider its domestication (artificial regeneration). This domestication could open up long prospects for in-depth research, in particular on the efficient use of its agroforestry potential and its conservation. In this current context of climate change, the study of the germination and development of seedlings of this species with respect to a changing environment could appear as a good start to its domestication in quantity and quality.Several research questions were developed to specify the objectives of our study. These are: (i) Could the ever-changing environment have an effect on seed germination and seedling growth? (ii) are the dendrometric characteristics of the mother plant (ideal choice of seed trees) necessary to obtain a good germination rate and vigorous seedlings? (iii) is the choice of vigorous seedlings resistant to climatic stress necessary? (iv) do seedlings of this species adapt to different types of environment? (v) could the germination or morphology parameters of the seedlings guide the choice of an environment conducive to the implementation of an agroforestry program based on Parkia biglobosa? etc. This study was carried out in an attempt to answer all of these questions. The objective of this study is to contribute to the conservation of Parkia biglobosaby its domestication on a large scale in Côte d'Ivoire, taking into account environments foreign to its natural environment. Specifically, it involves (1) testing its adaptive power to different new climates, (2) evaluating the effect of three distinct environments on the germination of its seeds and the development of its seedlings and (3) of evaluate the effect of characteristics of mother plants on the germination of its seeds and the development of its seedlings.

Plant material
The plant material is composed firstly of seeds obtained after dehulling of ripe fruits (figure 1a), from six distinct mother trees spaced about 400 m apart in the same stand of Parkia biglobasa and secondly of three-month-old seedlings (figure 1b), resulting from the germination of seeds harvested under the six mother trees. All seeds were collected at the same time in late April 2019 from trees in good physiological condition at the experimental station (DeFo) of the CNRA (Centre National de Recherche Agronomique) in the Korhogo department of Côte d'Ivoire. The dendrometric characteristics of the mother trees and seeds are listed in Table 1. The plant material used is the property of the CNRA of Côte d'Ivoire and the authorisation to use this plant material was given to us within the framework of this study thanks to a partnership agreement signed and available on request between the said structure and our study project (EFISA).

Study sites
The trials were implemented from May to September 2019 in three sites, two in Côte d'Ivoire and one in France with different microclimates. Of the two sites in Côte d'Ivoire, one was at the CNRA forest experimental station (hereafter DeFo: Développement des forêts) in Korhogo, and the other at theUJLoG (UniversitéJean Lorougnon Guédé) University in Daloa. The trial in France took place in a controlled environment, in Greenhouse 8 at CIRAD (The French agricultural research and international cooperation organization working for the sustainable development of tropical and Mediterranean regions) in Montpellier. The characteristics of the study sites are listed in Table 2. Mature fruits were harvested in April and May 2019 on the mother trees using long wooden sticks forks attached or by knocking the top of the tree with stones. The mature fruit collected under each mother tree was husked by hand to remove the thorny shells from the seeds. The seeds were then divided into three batches. Each batch contained seeds from all six mother trees, i.e. 120 healthy seeds were selected per mother tree and per study site after sorting all the seeds collected (120 seeds x 6 mother trees x 3 test sites giving a total of 2,160 seeds of Parkia biglobosa).

Environment 1 and 2: Korhogo and Daloa Nurseries
Polyethylene black bags with drainage holes measuring 20 x 10 cm were filled with local soil and arranged in one block comprising six sub-blocks. Each sub-block was labelled with the mother trees serial number and geographic coordinates and contained seeds harvested on and under one mother tree. Each sub-block contained 60 bags of soil prepared to receive two seeds each. The seeds from each mother tree were soaked in water for 24 hours to break seed dormancy and then sowed directly at a depth of approximately 2 cm in the bags at a rate of two seeds per bag. Before planting, the seeds were treated with granulated FURADAN to control rodents and after seedling emergence, the pre-leaves were treated with DECIS to limit insects' attacks. Nursery maintenance consisted of daily watering and manual weeding.

Environment 3: CIRAD greenhouse in Montpellier
Polyester black pots with drainage holes measuring 30 x 15 cm (figure 1b) were filled with a mixture of potting compost as specified above ( Table 2). The pots were arranged in labelled blocks and sub-blocks in metal bins arranged in the same way as in Korhogo and Daloa. The seeds were sown in the same way as those from the other two sites in Côte d'Ivoire. Biological protection consisted of treatment with BIOBEST against greenhouse whiteflies. The pots were watered daily (10 cm3 per week). all the pots occupied an area of about 12 m2.

Seeds germination and seedling development parameters
A total of 2160 seeds were evaluated for the entire study. 120 seeds per mother tree (constituting a sub-block) were evaluated per study site (120 x 6 = 720 seeds in total on each site). However, a total of 540 seedlings were sorted and then evaluated for this entire study. The morphology of 30 vigorous and three-month-old seedlings resulting from the germination of the seeds of each mother tree (in each sub-block) was evaluated on each study site (30 x 6 = 180 seedlings in total for each site of study). All morphological parameters were measured using a ruler graduated in centimetres and an electronic caliper in millimeter. Five germination parameters and nine development parameters were evaluated, there are recorded in Table 3.

Beda Innocent Adji et al.
International Journal of Horticulture, Agriculture and Food Science (IJHAF)  The length connecting two nodes or the length of two points of successive insertions of organs or leaf scars, from the base to the apex of the seedling

Data analysis
The statistical analyses were first performed using one-dimensional descriptive statistics, link analysis (linear regression, correlation, and covariance) and multidimensional analysis (principal component analysis PCA) with XLSTAT 2020 version 7.5. The difference between the germination and morphology parameters was assessed using a two-factor multivariate analysis (MANOVA) with SAS software version 9.4. The Student-Newman-Keuls test at the 5% threshold was used for posthoc comparisons.

Description of the germination process for Parkia biglobosa seeds
The germination of Parkia biglobosa is hypogeal generally with a thick epicotyl at the base reaching an average of 2.07 mm in diameter at the collar and an average of 8.3 cm in length about 22 days after sowing. The pre-leaves are composed and bipinnate with generally 4 primary leaflets who composed an average of 11 pairs of secondary leaflets. The leaves following the pre-leaves are generally composed of two primary leaflets. Subsequently, the seedling produces bipinnate compound leaves with more and more primary leaflets (up to 10 primary leaflets with an average of 36 pairs of secondary leaflets at three months of age). Phyllotaxis is alternating spiro-disc at this stage and each leaf carries two stipules, one of which has the leaf sheath and another at the end of the leaf (at the point of insertion of the last two primary leaflets, at the end of the rachis).

Environment 1: Korhogo nursery
At the Korhogo nursery, the first germinations started on the 7th day and the 9th day with seeds from mother trees number 4 and 1 respectively. the shortest time to germination were obtained from seeds of mother tree number 1, with a variation of 9 to 24 days around an average of 18.04 ± 4.18 days. As for the best germination speed, it was 18 days and was obtained from mother tree number 1 as well. Short germination spread times were observed in mother trees number 1, 3 and 4 in 15, 18 and 18 days respectively. Finally, the good germination rates were 90% and 93.33% respectively in the seeds of mother trees 1 and 3.

Environment 2: Daloa nursery
In Daloa nursery, the smallest germination latency was 4 days and was observed in seeds of mother tree number 4. The smallest germination delays were observed in seeds of mother trees number 1, 3 and 4 with respective variations from 10 to 21 days around an average of 16.13 ± 2.78 days, from 13 to 30 days with an average of 17.98 ± 4.01 days and from 4 to 27 days around an average of 16.64 ± 5.68 days. The fastest germination speeds were 15, 16 and 18 days, respectively observed in seeds of mother trees number 1, 4 and 3. The smallest germination duration was 11 days and observed in seeds of the mother tree number 1. As for the best germination rates, they were observed in the seeds of mother trees number 1, 2, 3, 4 and 6 with respectively rates of 100%, 95.50%, 100%, 92, 50% and 100%.

Environnent 3: Montpellier greenhouse
In Montpellier greenhouse, the smallest germination latency times were 9 and 14 days and were observed in seeds of mother trees number 6 and 3 respectively. the shortest time to germination was observed in seeds of mother tree number 5 with a variation of 21 to 46 days around an average of 29.43 ± 11.81 days. the seeds of mother tree number 6 obtained a very good germination speed (17 days) unlike the others. The seeds of mother tree number 5 had short germination times (25 days) compared to the others. As for the germination rates, they were higher than in the seeds of mother trees number 1, 5 and 6 (respectively 80%, 70% and 70%).

Environment 1: Korhogo nursery
At the Korhogo nursery, the greatest sowing heights were obtained from seedlings resulting from the germination of seeds of mother tree number 3 (20.4 ± 2.61 cm). Large mean diameters were observed in seedlings grown from seed germination of mother tree 3 (2.61 ± 0.12 mm) and 4 (2.51 ± 0.43 mm). The number of leaves was higher in seedlings resulting from seed germination of mother tree number 6 (4.02 ± 0.42). As for the length of the leaves, it was greater in seedlings from mother tree 1 (11.11 ± 0.59 cm) and 6 (10.11 ± 1.34 cm). Leaf width was greatest in seedlings originating from seed germination of mother tree number 1 (10.2 ± 0.23 cm). The number of primary leaflets was higher in seedlings from mother trees 5 (5.63 ± 0.34) 3 (5.21 ± 0.26). Primary leaflet length was greater in seedlings originating from seed germination of mother trees 3 (7.51 ± 0.14 cm) and 6 (7.46 ± 0.21 cm). The largest primary leaflet widths were observed in seedlings from mother trees number 4 (4.21 ± 0.34 cm) and 3 (3.81 ± 0.21cm). finally, the longest internodes were observed in seedlings from the seeds of mother tree 1 (3.57 ± 0.2 cm).

Influence of environment and mother trees on seeds germination
The comparison of seeds germination compared to the experimental sites, revealed a significant variability between the three study sites for all the germination parameters evaluated (P <0.05) ( Table 4). The variance analysis of germination parameters (Table 4) shows, however, that all the germination variables observed overall are statistically identical from one mother tree to another (p> 0.05) on the three experimental sites.

Development parameters
Influence of environment and mother trees on seedlings development The comparison of experimental sites compared to the seedling morphology (Table 5) revealed that there is a significant difference between the study sites for the variable's height, diameter, number of leaves, length of leaves, length of primary leaflets and internode length (P <0.05). Unlike leaf width, number of primary leaflets and width of primary leaflets which were statistically identical regardless of the study site (p> 0.05). The results of variance analysis (Table 5) of seedlings morphology in relation to their origin (mother tree) showed that the set of development parameters evaluated were identical in the whole of a mother tree to another (p> 0.05).  Figure 3 shows the projection of study site and morphological parameters of all the seedlings on the PCA 1-2 (biplot). The analysis of the matrix of factor weights allowed the extraction of two components that explain 100% of the variability and therefore the total variation between the morphological characteristics of the seedlings and the environments. Plan 1-2 is characterized by eigenvalues of 97.57% for axis F1 and 2.43% for axis F2. The different descriptors contributing to the formation of the first (F1) and second component (F2) form a single group consisting of the Daloa and Montpellier (greenhouse) sites characterized by all morphological parameters higher than those of the seedlings in Korhogo site. Seeds and seedlings from seed germination of mother trees number 1, 3 and 6 performed best regardless of the study site; probably because of the size of the seeds collected on and under these mother trees. These three mother trees have the greatest dendrometrics characteristics among all the mother trees selected for this study. These mother trees could be identified as very good parents for the production of vigorous, resistant genotypes and elite trees for the establishment of permanent plots within the framework of the execution of a reforestation or agroforestry program. This study shows that large trees with large diameters produce large seeds which in turn generate vigorous seedlings. This observation could be studied on a large number of mother trees in order to consolidate the results and popularize them in a paying Overall, the germination parameters were as good as our expectation and imagination. Despite the nontreatment of the seeds with sulfuric acid (H2SO4), nor the scarification of the seeds or hot water for the lifting of dormancy, we obtained rates of up to 100% in natural areas as foreign to this species. In Niger, Douma et al. (2019) treated Parkia biglobosa seeds with sulfuric acid (H2SO4), which resulted in a two-day lag time, a four-day germination rate, and a germination rate of 80-92%. With the scarified seeds of Parkia biglobosa, the same authors obtained a latency time of three days, a germination rate of four days and a germination of 100%. When they heated the seeds, the germination rate was only 16-24%. Yet without nursery treatment, Amonum et al. (2016) obtained a latency time of six days, a germination rate of 33 days, the total germination time was 31 days, and the seed germination rate was 58.86%. Similarly, in a study on the viability of Parkia biglobosa seeds in Burkina Faso, Millogo (2014) obtained with seed banks a low germination rate (0.83-14.67%). According to him, this was due to a remarkably high loss of genetic diversity. The results of this study on seedling morphology are comparable to those obtained in Niger by Douma et al. 3 cm, an average diameter of 6.6 mm with an average of 8.5 leaves. We had thought that Parkia biglobosa seeds would not grow in an environment other than its natural range. However, the seeds germinated there (Daloa nursery and Montpellier greenhouse) more efficiently with more vigorous seedlings than those in its area of origin (Korhogo). This is undoubtedly the cause of more lenient environments (fertile soil, higher air humidity, favorable temperature, etc.) than in its natural zone (low humidity, high temperature, poor soil). This species shows its ability to adapt to different foreign environments through this study. For guaranteed conservation (domestication), this species could be introduced into permanent agroforestry systems throughout the territory of Côte d'Ivoire. In addition, it is a species that restores soil fertility (leguminous plant). It could therefore be a replacement solution for the chemical fertilizers used permanently in the fallow areas of the savannah (natural environment for the species) and forest areas (environment foreign to the species) of the country.
The analysis of variance showed that the mother trees do not significantly influence each parameter of germination and seedling morphology for all three sites (P> 0.05). Indeed, this is normal since it is the seeds and seedlings of the same mother trees that were evaluated on all three sites. But, taken in isolation site by site, the parameters of germination and seedling morphology differ from one mother tree to another for each site. We believe that the dendrometric characteristics of mother trees should be considered in seed collections for studies of germination and production of seedlings for species conservation (domestication). However, the comparison of the germination and development parameters of the seedlings according to each of the three experimental sites indicated that the three sites differ significantly for all the parameters evaluated (P <0.05). In fact, each site is made up of its own distinct microclimatic characteristics (type of soil, availability of water in the soil and in the ambient air, temperatures, etc.). Korhogo is located in a savannah zone (poor, arid soil with high temperatures, dry tropical climate), Daloa is located in a forest zone (rich and humid soil, mild temperature, humid tropical climate) and the Montpellier greenhouse is stable with a potting soil is made up of a mixture of substrates rich in mineral elements. Several studies have shown the effect of climate or climatic zone, soil, mother trees and seed provenance on plant growth and development (Giordano 1972

VI. CONCLUSION
This study showed that the dendrometric characteristics of mother trees could be a factor in the selection of vigorous seed and seedlings in the context of conservation (domestication) of Parkia biglobosa. The larger the mother trees are, the larger the seeds they produce, which in turn generate more vigorous seedlings. This study also showed that this species is adaptable to foreign environments. It grows better in a milder environment that is different from its native area. Overall, the mother trees did not statistically influence each germination and development parameter for all three sites combined. However, analysis of variance showed that germination and seedling development parameters differed from one experimental site to another. A rich soil and a stable or humid tropical climate (Montpellier greenhouse and Daloa site) guarantee good germination and seedling development in Parkia biglobosa. These results are useful and could be used as decision support tools to guide conservation (domestication) and agroforestry programmes based on Parkia biglobosa. This study could be extended to other endangered species in order to safeguard humanity.

VII.
AUTHOR'S CONTRIBUTION Beda Innocent Adji: Conception, methodology, supervision, software, formal analysis, writing the paper, resources, data acquisition and analysis. Sélastique Doffou Akaffou: Project administration, methodology, resources, data acquisition, supervision, writing the paperoriginal project, research and acquisition of funding. Sylvie Sabatier: Project administration, methodology, resources, data acquisition, supervision, writing the paperoriginal project, research and acquisition of funding.

ACKNOWLEDGEMENTS
This study was financed by the Ministry of Higher Education and Scientific Research of Côte d'Ivoire, the French Development Agency and IRD (Institut de Recherche pour le Developpement) within the framework of PRESeD-CI 2 (Renewed Partnership for Research for Development in Côte d'Ivoire) and C2D (Debt Reduction Contract) of the AMRUGE-CI project (Support for the Modernization and Reform of Universities and Grandes Ecoles of Côte d'Ivoire). We would like to thank the CNRA (Centre National de Recherche Agronomique) of Côte d'Ivoire for allowing access to its scientific research site and for supplying the plant material (Pterocapus erinaceusPoir., 1804) used for this study. The authors are grateful to the Centre de Coopération International de Recherche Agronomique pour le Développement(CIRAD) for providing a greenhouse (controlled environment) and the technical equipment necessary to conduct the study.

DATA AVAILABILITY
The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.