Seed Propagation of Carapa amorphocarpa W. Palacios Using Various Treatments of Substrates and Mechanical Seed Scarification in a Nursery in the Andean Area of Northwestern Ecuador

Abstract

Carapa amorphocarpa W. Palacios is a forest species known solely from Cerro Golondrinas in northwestern Ecuador. The species is threatened due to illegal exploitation of its wood and the encroachment of the agricultural frontier. Although information on its ecology and forestry was presented in 2012, there is still very little information available. This study investigated the effects of various pre-germinative treatments on the seed germination and early growth of Carapa amorphocarpa in a nursery. The treatments included combinations of substrates (forest soil alone and forest soil mixed with sand), along with pre-germinative mechanical seed scarification. Through these treatments, the seeds’ germination capacity, growth potential, and survival rates were evaluated in a nursery. Seed germination was found to be cryptocotylar hypogeal, commencing at 41 days and achieving an average of 83.13%. Survival of germinated seedlings was 99.23% at 60 days after germination. Furthermore, a high degree of correlation in growth was observed between basal diameter and plant height (R = 0.94). The mean values for the plants’ basal diameter and total height were 0.91 cm and 50.48 cm, respectively, at 60 days. Plant form was straight with excellent health. These insights offer the potential to enhance species populations, mitigate threat risks, and in the long term, foster sustainable use of Carapa amorphocarpa genetic resources. Moreover, this seed propagation information can be leveraged for other species within the genus exploited for oil extraction from their seeds, thereby broadening knowledge and facilitating sustainable forest resource management.

1. Introduction

Latin America is home to significant biodiversity, including species of high economic and ecological importance, but it faces significant challenges due to rampant deforestation [1]. This phenomenon, which is widespread in several countries in the region, not only endangers natural resources but also disrupts ecological balances and threatens the ability of local communities to sustain themselves [2,3]. Uncontrolled exploitation of natural resources, such as illegal logging, has resulted in the degradation of extensive forested areas crucial for ecosystem restoration [4]. In this context, forest restoration emerges as a critical strategy by which to mitigate the adverse effects of deforestation and ensure the provision of multiple ecosystem services [5,6].

Nurturing healthy and diverse forests for restoration and biodiversity conservation begins with understanding and effectively managing the propagation of forest species in nurseries. Insights from this process not only facilitate seed collection but also guide the production of successful seedlings for reforestation efforts [7]. Seed germination, influenced by various factors such as seed size, soil characteristics, and seed origin, plays a pivotal role in the success of seedling production [8]. In pursuit of understanding and enhancing plant species propagation processes, there is a need for studies focused on specific species and the regions where they are found. Habitat loss and the unsustainable exploitation of forest resources have led to a population decline of valuable species throughout Latin America. Addressing these challenges necessitates research and the implementation of appropriate propagation practices for the conservation and sustainable use of forest biodiversity [9,10].

Within this context, the focus of this study lies in the seed propagation of Carapa amorphocarpa W. Palacios, an endemic species found on the slopes of Cerro Golondrinas in northwestern Ecuador. Typically, it thrives at altitudes ranging between 2000 and 2300 m in highly humid forests characterized by epiphyte-covered trees and dense undergrowth [11]. This species shares threats due to illegal exploitation with other members of the genus, such as Carapa guianensis Aubl and Carapa procera DC, distributed across several Latin American countries [12]. The aim of this paper is to gain comprehensive insights into the germination processes and seedling establishment of Carapa amorphocarpa. This research endeavors to provide crucial data essential for management and conservation efforts, not only for this species but also for others facing similar challenges in the region [13,14].

On the Golondrinas hill in northwestern Ecuador, Carapa amorphocarpa faces illegal exploitation for timber purposes, with a lack of minimal management criteria. Furthermore, this species, confined to a diminutive range, grapples with a deficit of ecological and forestry data [11]. Therefore, the population dwindles exponentially, while its potential for natural regeneration remains enigmatic. The native habitat of the species has been heavily affected by deforestation, with the opening of the El Limonal-Chical road in 2006 alone representing a loss of 31% of the forest [15]. Ecuador harbors six recognized species of Carapa, each holding economic significance [16]. However, several species are on the brink of extinction, necessitating assisted human propagation [2,11,14]. The seed propagation of Carapa species underscores the importance of genetic and phytosanitary considerations, alongside environmental factors and seed predation dynamics [17]. This underscores the importance of integrating Carapa amorphocarpa into forest production schemes and pre-germination treatment evaluations.

Research on the germination and initial development of Carapa species has been limited, with most studies focusing on their ecosystem services, such as the oil extracted from seeds [18,19]. One such study, conducted by Camara et al. [20] in Basse Casamance, Senegal, aimed to determine the optimal conditions for germination and growth of Carapa procera seedlings in nurseries. It was observed that seeds from different locations had varying germination rates and morphological characteristics, and that the choice of specific substrates, such as pure soil, influenced the success of germination and plant growth. On the other hand, the study conducted in Brazil by Gonçalves et al. [21] also addressed the germination of Carapa guianensis seedlings, evaluating different substrates for their development. It was found that the Latosol Amarillo substrate favored greater growth of the seedlings, suggesting that substrate selection can have a significant impact on the success of germination and initial development of the species. Finally, the study carried out by Voui et al. [22] in the Ivory Coast focused on the vegetative propagation of Carapa procera, exploring various methods to reproduce the species without relying on seeds. It was determined that propagation by stem cuttings was the most effective method, with significant survival rates and potential for domestication of the species [22]. Together, these studies provide valuable information on cultural practices, environmental conditions, and propagation methods that can promote the establishment and conservation of Carapa procera in different regions of the world. Furthermore, the objective of propagation is crucial, as seed propagation generally captures a wider range of genetic diversity than clonal propagation [23]. This comprehensive understanding is crucial for promoting the sustainable use and conservation of this valuable species, which plays an important role in local biodiversity and economies.

The results of studies on the propagation of species within the Carapa genus have proven valuable in promoting their conservation. Furthermore, given the lack of research on silviculture of Carapa amorphocarpa, the significance of this work is underscored. Therefore, the novelty of this study lies in its contribution to the conservation and management of Carapa amorphocarpa by providing essential information on its propagation. Additionally, unique characteristics of the seed, such as its amorphous shape, were considered in identifying the appropriate position during planting. Also, the leathery sarcotesta that slows propagation prompted us to test methods of mechanical scarification [24]. The research results on seed structure, germination efficiency, plantation dynamics, seedling survival, and initial growth trajectories lay the groundwork for understanding the species’ silviculture. Additionally, they facilitate the exploration of differences among genotypes and various treatment approaches, including the use of diverse substrates and comprehensive analysis of additional variables.

While this study may have a regional focus, its implications extend to broader discussions within forestry and biodiversity conservation. The examination of the propagation methods for a threatened species offers valuable insights into silvicultural practices and ecosystem management strategies. Furthermore, the species under investigation, Carapa amorphocarpa, is within the genus Carapa, which faces significant exploitation for its seed oil [25]. Consequently, the findings from this study have wider applicability beyond the specific species studied, providing insights relevant to conservation efforts for other Carapa species with little scientific information on their seed propagation.

2. Materials and Methods

2.1. Experimental Location

This research was conducted in the southeastern extremity of the Cerro Golondrinas Protected Forest, located in the Carchi province of northwestern Ecuador (Figure 1). Nested within the biogeographical region of the Chocó, renowned as one of the 25 most biodiverse regions globally, this protected forest primarily comprises the Lower Montane Evergreen Forest. Additionally, a smaller portion transitions between Lower Montane Evergreen Forest and Montane Cloud Forest. Its flora is renowned for its richness, diversity, and endemism, making it a significant center of genetic diversity for the country. The elevation within the forest ranges from 1360 to 3060 m above sea level, with annual precipitation averaging between 2000 to 2500 mm, and a mean temperature of approximately 15 °C [15].

The experimental site was situated in the El Carmen sector of the Jijón y Caamaño parish, within the Mira canton, at an altitude of 2100 m above sea level. Initial preparations involved clearing the site of existing vegetation and marking boundaries using barbed wire fencing. Additionally, chain link fences were installed around seedbed areas to prevent seed predation.

The experimental area was divided into three distinct sections: a substrate mixing and storage area covering 8.75 m², a fruit maturation area covering 8.75 m², and a seedbed area spanning 87 m². Two seedbeds, each measuring 10 m in length and 2 m in width, were constructed for the experiment. Within these seedbeds, specific sections were designated for experimental treatments and germination trials. Furthermore, a separate area was dedicated to fruit ripening, as the majority of collected fruits were found to be immature.

2.2. Substrate Preparation

Substrate preparation activities involved relocating necessary components to the study area. Forest soil was directly extracted from the seed collection site without undergoing sieving or disinfection to preserve native vegetative matter. Also, the forest soil was mixed with sand to enhance drainage. The substrates subject to study were the following:

▪

Substrate 1 (S1): Forest soil mixed with sand in a 3:1 ratio.

▪

Substrate 2 (S2): Forest soil alone.

In total, approximately 2.50 m³ of substrate was utilized, which was filled into 7 × 9-inch polyethylene bags, leaving a 1 cm space from the edge to prevent air pockets. After filling the bags, they were arranged into four blocks and labeled according to the pre-germination treatment. A reserve of the substrate components was kept, as daily watering caused the substrate to overflow, leaving the seeds exposed in some cases.

2.3. Fruits and Seeds Collection

Carapa amorphocarpa is used for the high quality of its wood and seeds; and its endemic distribution contributes to the decrease in the population individuals. Field visits did not reveal abundant seed trees, so the minimum distance between selected trees was approximately 1 km, spatially distributed evenly in the study area to capture genetic variability. The selection of seed trees was carefully conducted considering factors such as the presence of natural regeneration.

To identify the Carapa amorphocarpa seedling trees, a meticulous count of mature fruits was conducted. On average, 10 fruits were observed on each identified tree, although some exhibited a maximum of 15 mature fruits. These fruits were characterized by a greenish-brown color and a cracked pericarp that detached from the tree individually over prolonged periods. Conversely, other individuals of the species showed minimal or no fruit production. Binoculars were used to facilitate precise fruit identification, aiding in discerning details even from a distance. Furthermore, collection activities were predominantly scheduled for the morning to capitalize on optimal visibility conditions and mitigate the effects of the cloudy conditions that often prevail in the afternoon. This rigorous and systematic approach ensured careful selection of the trees for seed Carapa amorphocarpa propagation trial.

The forest floor was covered with a substantial amount of leaf litter and fallen branches, making fruit collection challenging. Moreover, the terrain’s slope caused significant fruit displacement upon falling. Fifteen trees exhibiting optimal phenotypic characteristics were selected for fruit collection. These trees displayed good health, maturity, straight form, and at least one commercially viable sawn log. The study area exhibited a limited tree population, with evidence of natural regeneration, albeit with a scarcity of seed trees. Each selected tree was numbered with white paint, clearly visible at a height of 1.30 m above ground level.

Carapa amorphocarpa fruits were harvested directly from the trees due to the scarcity of fallen fruits on the ground. This suggests rapid consumption by animals like rodents, birds, and mammals. The harvest occurred in September during the rainy season. This was due to the observation that only pericarps were found on the ground, indicating rapid consumption by animals such as rodents, birds, and mammals [26,27,28,29]. The collected fruits displayed a brown pericarp, traditionally acknowledged as a sign of maturity, akin to the seeds, where the sarcotesta takes on a brown hue (Figure 2a) [14]. However, upon extraction, some seeds were found to be immature, identifiable by their white sarcotesta (Figure 2b). To assess maturity and classify the fruits, part of the pericarp was removed (2 cm approximately), and the seed color was observed, a key phenotypic indicator in fieldwork. The immature fruits underwent conventional anthropogenic ripening processes, involving two methods: one group of eight fruits was exposed to water, which was changed daily for two weeks, while another group of eight fruits was stored in forest soil. The storage of fruits and seeds in the nursery involved burying the fruits in total darkness at ambient temperature (approximately 15 °C) within raised beds containing forest soil rich in organic matter. While the soil was moist, it was not waterlogged.

It is noteworthy that mature seeds meeting specific phenotypic criteria, such as a light brown sarcotesta and a size exceeding 4 cm in length, were selected for the primary experiment. Preference was given to larger and heavier seeds, which often indicate greater health and vigor [30]. These seeds must also be in good phytosanitary condition, meaning they must be intact and free from visible damage such as cracks or spots. Additionally, they should not exhibit signs of pathogen infection, such as fungal or bacterial presence, indicating the absence of spots or rot [31]. Conversely, seeds obtained from the artificial ripening process were utilized for preliminary germination tests to determine optimal seed positioning and viability.

Table 1. Summary of the quantity of fruits and seeds utilized in the study obtained from 15 seed trees.

Fruits	Origin of the Fruits	Number of Fruits	Number of Seeds	Utilization
Mature	Harvested from the tree	23	324	Main test
Immature	Process of maturing in water	8	89	Preliminary
	Process of maturation in forest soil	8	94	Germination test
Total		39	507

shows the number of seeds and fruits collected.

Preliminary germination tests of Carapa amorphocarpa focused on the position of the seed hilum due to its amorphous shape, which makes identifying the sides of the seed challenging. However, it is possible to associate the convex side with the hilum. Thus, five seeds were sown with the convex side and the wider part of the hilum facing upward, and five seeds were planted upside down. Only forest soil substrate was used, with daily watering provided. Additionally, they were kept shaded under a shade cloth to replicate their natural environment conditions.

2.4. Treatments of the Carapa amorphocarpa Seeds

The treatments consisted of combinations of substrates S1 (forest soil mixed with sand in 3:1 ratio) and S2 (forest soil alone) with a pre-germinative mechanical scarification (MS) treatment. The first treatment used S1 with MS, labeled S1-MS. The second treatment involved S2 with MS, labeled S2-MS. The third treatment employed S2 without MS, labeled S2-NoMS. The fourth treatment used S1 without MS, labeled S1-NoMS. Seeds subjected to mechanical scarification were obtained from ripe fruits collected directly from the trees. The scarification process involved carefully removing the bark from the angular part of each seed (Figure 3). This method was chosen to mechanically break the sarcotesta, facilitating water and oxygen absorption [32]. By exposing the inner layers of the sarcotesta, the aim was to overcome dormancy barriers and improve seedling emergence [33], particularly in species with hard or impermeable sarcotesta like Carapa amorphocarpa. The angular part was specifically selected for its structural integrity and accessibility, ensuring a controlled and precise scarification process.

The seeds were sown in 17.5 cm × 22.5 cm polyethylene bags filled with the corresponding substrate for each treatment, initially filling the bags to 70% of their capacity. The seed was placed in the correct position identified in the germination tests, which involved placing the seeds with the convex side and the wider part of the hilum facing upward. Subsequently, the seeds were covered with 2 cm of soil, leaving 1 cm of the bag’s edge free to retain irrigation water in that space. Since Carapa amorphocarpa plants naturally grow in forested habitats or areas with tree canopy that provide partial shade, shading conditions or careful control of light intensity were implemented to mimic their natural environment. A shade cloth, palm and banana leaves were used to provide the necessary shade and prevent direct exposure to sunlight. Additionally, cultural practices such as weeding in the plant bags and nursery in general were carried out. Seedlings were watered daily in the afternoons to maintain moist substrates. Additionally, a mesh fence was installed to safeguard the seeds from rodents.

The germination capacity of each treatment was evaluated using Equation (1):

$$GP=\frac{NGS}{TNS}\times 100,$$

(1)

where GP is germination percentage, NGS is the number of Carapa amorphocarpa germinated seeds, and TNS is the total number of seeds sown. The germination time was in days.

For the germinated Carapa amorphocarpa plants, data regarding basal diameter, height, shape (straight: 1; crooked: 2; bifurcated: 3), and health were recorded following the criteria set by the Ecuadorian Center of Agricultural Services (CESA) [34], as shown in

Table 2. Plant health rating criteria for Carapa amorphocarpa seedlings.

Code	Category	Characteristics
0	Bad	Dead plants, dry
1	Regular	Plants with ≤50% green leaves
2	Good	Plants with ≥50% green leaves
3	Excellent	Plants with 100% green leaves

. Measurements of these variables were taken at two different time points: 30 days and 60 days after germination. These data were recorded on field sheets using a vernier caliper for diameter and a tape measure for height.

The survival percentage (SP) of the plants was also calculated, using Equation (2) [24]:

$$SP=\frac{NLP}{NTP}\times 100,$$

(2)

where NLP is the number of living plants and NTP is number of total plants originally planted. Data pertaining to diameter and height were analyzed 60 days post-germination. Root length was measured by extracting plants from their containers and gauging their primary roots with a flexometer.

2.5. Statistical Analysis

The unrestricted random design was applied with factorial arrangement A × B × C, where factor A: substrate (S1 and S2); factor B: scarification (MS and NoMS); factor C: time measurements (30 days and 60 days). Statistical tests were performed using RStudio version 4.1.1 [35]. Continuous quantitative variables were tested for normality using the Shapiro–Wilk test and the Levene test for homoscedasticity [36]. Once the assumptions were met (p-value > 0.05), an analysis of variance (ANOVA) was performed, followed by Tukey’s multiple comparison test with a significance level of 0.05 [37]. For ordinal qualitative variables and quantitative variables without a normal data distribution, the non-parametric Friedman test [38] was applied directly to related samples. This analysis was carried out at 30 days and 60 days after germination.

To determine the number of plants, Equation (3) for calculating the size of the finite sample was used [39].

$$n=\frac{N\times Z_{\mathsf{\alpha }/2}^2\times P\times (1-P)}{E^2\times \left(N-1\right)+Z_{\mathsf{\alpha }/2}^2\times P\times (1-P)},$$

(3)

where $n$ is the desired sample size, $N$ is the size of the population (15 seedling trees × 10 fruits × 12 seeds = 1800 seeds approximately), $Z$ is the statistical parameter dependent on the confidence level (CL = 88%), $E$ is the maximum acceptable estimation error (12%), $P$ is the probability of the studied event occurring (success = 0.5), and $Q=1-P$, which is the probability of the studied event not occurring. Once n is calculated, it is divided by the number of experimental units to obtain the number of plants for each repetition. The general characteristics of the trial for statistical analysis was 16 experimental units, 20 plants per experimental unit, 4 repetitions, 4 treatments, and 80 plants per treatment giving a total of 320 plants.

At the end of the evaluation, a correlation analysis was performed. Root length, basal diameter, and height variables were assessed using the Pearson correlation coefficients.

3. Results

3.1. General Characteristics of Fruits and Seeds

All the collected fruits had an amorphous shape (Figure 4a), with one or two rounded lobes, which confirms one of the distinctive characteristics of Carapa amorphocarpa. The pericarp was smooth and brown. The fruit size is 8 to 16.5 cm × 6 to 10 cm and contains between 9 and 15 seeds. The seeds are variable in shape and size (Figure 4b). One kilogram contains about 40 seeds and measures 1.5–5.5 cm long, 2–4 cm wide, and 2–5 cm deep. They have a thick, coriaceous, smooth, and brown sarcotesta that has four sides, i.e., three planes (due to compression between the seeds inside the fruit) and one convex side that follows the shape of the epicarp. At the front are the hilum (0.8–2.3 cm long) and the aliform micropyle. The embryo for the seeds of the Carapa species is axillary-foliate-lined, large, and occupies a quarter of the total lumen of the seed and is more central than peripheral.

It is important to mention that during the research process, evidence of larval infestation in Carapa amorphocarpa seeds was observed. Through photographic analysis and comprehensive literature review, it was determined that the larvae of Hypsipyla ferrealis were responsible for this infestation. These insects create galleries within the seeds, resulting in the complete destruction of the structures (Figure 5).

3.2. Preliminary Germination Tests

The direct collection of fruits from the tree resulted in confusion in identifying mature fruits. Some fruits with phenotypic characteristics similar to mature fruits turned out to be immature. After 10 days, the results for the eight fruits subjected to water ripening showed that the fruit pericarp decomposed and the seed sarcotesta acquired a red color, along with a foul odor, indicating fruit and seed putrescence. In contrast, fruits stored in forest soil matured properly, acquiring a brown pericarp with a soft texture. In these cases, the pericarp cracked, facilitating seed extraction, and the seeds had a light brown sarcotesta.

The seeds obtained from the maturation process, utilizing forest soil, were utilized for preliminary germination tests to assess viability and positioning. It was determined that it is important to focus on the position of the hilum seed since its amorphous shape makes it difficult to identify the sides of the seed. However, it is possible to relate the convex side with the hilum. In this way, five seeds were sown with the convex side and the wide part of the hilum facing upwards (Figure 6a), and five seeds were sown upside down. The best position was the first since the radicle will be directed downwards and the epicotyl upwards (Figure 6b). In the seeds with the second position, the radicle turned over to reach the ground, and the epicotyl made the same movement to emerge (Figure 6c). With this evidence, all the seeds of the trial were located in the first position. The results report 100% germination for seeds that “matured” in forest soil.

3.3. Germination Capacity in the Main Test

Germination began when the sarcotesta split and the radicle appeared between the hilum and the micropyle (Figure 7a–c); at the same time, but more slowly, the hypocotyl formed as a thickened and fleshy ring (Figure 7d), from whose center the epicotyl was formed (Figure 7e). The small hypocotyl (less than 0.5 cm for Carapa amorphocarpa) and vestigial and hypogeal germination (cryptocotylar seedling) are a characteristic of the genus Carapa [40,41].

The germination of Carapa amorphocarpa occurred gradually over a period of 52 days. The first seeds germinated at 41 days, and the final seeds sprouted at 93 days, illustrating a prolonged and progressive germination process. The analysis of variance (SC type III) for the germination time reported that Carapa amorphocarpa does not present statistical differences in any of the treatments, i.e., S1-MS (70 days), S2-MS (69 days), S1-NoMS (75 days,) and S2-NoMS (7.7 days). However, the analysis of the factors reported differences for factor A (p-value of 0.014), and according to the Tukey test (p-value < 0.05), it is evident that the application of mechanical scarification to the seed accelerated germination time (69.5 days). On the other hand, factor B did not report statistical differences (p-value of 0.8271), so the substrates did not influence the analyzed variable.

The germination rate achieved during the test was 83.13%. Significant statistical differences were observed for the germination variable in factor A (p-value of 0.0001) according to the results of the ANOVA. The Tukey test indicated that seeds without pre-germination treatment exhibited the highest average germination rate at 95%. The ANOVA analysis for factor B also showed significant differences (p-value of 0.0161), with the Tukey test indicating that the mixture of forest soil and sand resulted in the highest average germination rate (88.75%).

Multiple comparisons using the Tukey test at a 95% confidence level revealed significant differences for the following interactions: S2-MS and S1-NoMS (p-value 0.0002), S2-NoMS and S2-MS (p-value 0.002), and S1-NoMS and S1-MS (p-value 0.02). The treatments with the highest means were S1-NoMS (98.75%) and S2-NoMS (91.25%), which did not show significant differences between them. Therefore, these treatments were identified as the most effective for promoting germination (Figure 8). It is noteworthy that treatments without scarification yielded the best results. For a comprehensive overview of the statistical analyses, please refer to the Supplementary Material.

3.4. Growth in Height

According to the ANOVA for initial growth in height, Carapa amorphocarpa presented highly significant differences (p-value < 0.0001) for factor A at 30 and 60 days after germination. In this way, the multiple comparisons of the Tukey test at 95% showed that the seeds without scarification reported the highest mean, 49.76 cm for the first measurement and 59.11 cm for the second. Factor B did not present significant differences in the two measurements (p-value: 0.3828 at 30 days and 0.5472 at 60 days); therefore, it is shown that the substrates do not affect the variable growth in height.

Finally, the interactions between treatments showed significant differences for the two measurements (p-value: 0.0107 at 30 days and 0.0153 at 60 days). Thus, Tukey’s multiple comparisons identified S2-NoMS and S1-NoMS as the best treatments, since no statistical differences were reported between them. The averages of the best treatments identified for the first and second measurements are presented in the following Figure 9. The complete statistical analyses are available in the Supplementary Material.

ANOVA analysis revealed highly significant differences in seedling height for factor C (p-value < 0.001) and the interaction of factors A and B (p-value < 0.005). This indicates that both the scarification–substrate interaction and the time elapsed since germination significantly affect seedling height. However, the interaction among factors A, B, and C was not significant (p-value = 0.588). Upon examining Figure 10, the lines demonstrate a consistent increase in height as time progresses, regardless of the applied treatment. Although significant differences in height between treatments are identified, the interaction between treatments and days does not exhibit a notable influence on plant height. This suggests that while treatments individually affect plant height, there is no substantial interaction between treatments and time in terms of their effect on height.

In Figure 10, each treatment is represented by a different line color. These lines indicate the overall trend of increasing height growth values at 30 and 60 days after germination.

3.5. Growth in Diameter

The ANOVA of the basal diameter in the two measurements shows highly significant differences for factor A (p-value: <0.0001 at 30 days and 0.0003 at 60 days). Therefore, the seeds without scarification represented the highest mean: 0.90 cm in the first measurement and 0.99 in the second. In addition, factor B did not present statistical differences in the measurements (p-value: 0.1718 at 30 days and 0.4408 at 60 days), so substrates did not influence the initial growth in diameter.

The interactions between the treatments showed significant differences for the measurement at 30 days (p-value of 0.0258), determining S2-NoMS (0.91 cm) and S1-NoMS (0.89 cm) to be the best treatments (Figure 11). The multiple comparisons of factors using Tukey’s test at 95% reported significant differences for S2-MS and S1-NoMS (p-value of 0.008) and S2-NoMS and S2-MS (p-value of 0.0034). Additionally, S2-NoMS (0.98 cm) and S1-NoMS (1.0 cm) were identified as the best treatments.

ANOVA analysis for seedling diameter revealed highly significant differences between treatments (p-value < 0.001) and diameter growth over time (p-value = 0.00135). This indicates that both the applied treatments and the time period have a significant impact on seedling diameter. However, the interaction between treatments and diameter growth time was not significant (p = 0.93698), suggesting that the effect of treatments on diameter does not vary significantly over time. In Figure 12, the lines depict a general pattern of increasing plant diameter over time, regardless of the applied treatment. While significant differences in diameter between treatments are observed, the interaction between treatments and time period does not appear to be significant. The complete statistical analyses are available in the Supplementary Material.

Figure 12 illustrates the interaction between treatments and diameter growth over time, with each treatment represented by a different line color. These lines indicate the overall trend of increasing diameter growth values at 30 and 60 days after germination.

3.6. Phytosanitary Status and Shape of the Seedlings

At 30 and 60 days after germination, the Friedman tests for the seedling shape variable showed similar medians between treatments. Therefore, no p-value showed statistical differences. Finally, the representative median for this variable is a value of 3, which is equivalent to a straight-shaped plant.

Regarding the health variable, the Friedman tests for the data of the measurements at 30 and 60 days showed similar values in the medians of the treatments. Therefore, it was shown that there were no significant differences in any of the comparisons (p-value > 0.05). The median for all treatments is a value of 3, corresponding to excellent seeding health for Carapa amorphocarpa.

3.7. Survival

Survival at 30 days after germination is 100% since all the plants that germinated survived in this period. On the other hand, at 60 days, a mortality rate is reported; however, the data did not comply with the normality assumption (p-value < 0.05). For this reason, the non-parametric Friedman test was performed, and it was shown that there are no significant differences in any of the treatments (p-value of 0.6310). On average, 60 days after germination, the survival was 99.23% (Figure 13).

3.8. Correlations

The correlations between root length and both total height (r = 0.36, p-value = 0.16) and basal diameter (r = 0.29, p-value = 0.27) were not statistically significant (Figure 14b). Conversely, total height exhibited a highly significant positive correlation with basal diameter (r = 0.94, p-value = 0.00000011), indicating a strong relationship between these variables (Figure 14c).

3.9. Propagation Costs

In addition, we have included information regarding the production costs associated with Carapa amorphocarpa. A total of 360 quality seedlings were produced: 266 from the main test and an additional 94 from preliminary germination tests. The cost of the general test was USD 824. Labor was the highest value (USD 455), attributed, in large part, to harvesting fruits.

Hence, for the investigation, a cost of 2.28 USD per seedling was estimated. The price of the plants was high because the production was carried out in a temporary nursery that was established only for the experimental phase of the investigation. In this way, costs for setting up the trial and transportation of supplies are included. On the other hand, adequate tools were not available for fruit and seed harvesting activities, so labor costs increased. It should be noted that there is no exact information on the production.

4. Discussion

4.1. General Characteristics of Fruits and Seeds of Carapa amorphocarpa

This study accurately identified Carapa amorphocarpa, primarily through its characteristic amorphous fruits, which are a distinctive feature of this species [11]. These variations in fruit morphology are reflected in the quantity of seeds produced per fruit, which can significantly differ among species within the Carapa genus [29]. For example, while the study identified between 9 and 14 seeds per fruit for Carapa amorphocarpa, previous studies indicate that Carapa guianensis averages 6 to 7 seeds per fruit according to Mchargue and Hartshorn [42] and 7 to 8 seeds per fruit according to Smith [43] in Central American trees. However, population estimates in northern South America report considerably higher figures, with up to 16 seeds per fruit, while in Trinidad, an average of 12 seeds per fruit is reported [43]. This difference in seed quantity per fruit has significant implications for seed availability for natural regeneration of each species. Additionally, factors such as predation by rodents and by Hypsiphila ferrails can reduce the quantity of seeds available for natural regeneration of the species [44]. Although these factors may seem insignificant when analyzed individually, their cumulative effect, combined with human exploitation of the species, can result in a significant population decline. To carry out effective restoration, a comprehensive understanding of the species’ silviculture is crucial. This includes knowledge of its habitat requirements, its response to forestry practices, and optimizing propagation techniques and nursery establishment [45]. A comprehensive approach addressing ecological and silvicultural aspects could promote the long-term conservation of Carapa amorphocarpa and other valuable species within the Carapa genus [46].

One prominent characteristic of seeds from the Carapa genus is their recalcitrant nature, primarily evidenced by their relatively large size [47,48], sometimes exceeding four centimeters in length [14]. These recalcitrant seeds, due to their large size and thick seed coat, maintain high sensitivity to dehydration, allowing these characteristics to be used as parameters to define storage conditions [14]. Considering this, seeds were evaluated for storage in water, following Cordero’s suggestion [49], as well as in moist forest soil. It was observed that water was not useful and caused seed rot, while forest soil proved effective in seed conservation, even serving as a tool for artificial maturation for some fruits collected immature. This method allows the seeds to remain attached to the fruit, as extraction would result in viability loss if stored for more than two days due to dehydration [50].

Mature Carapa fruits typically fall on the ground upon reaching maturity, making collection from the ground easier [17,26,29]. However, in the conducted study, finding fruits on the ground proved difficult due to the sparse tree population and rodent predation. Consequently, fruits had to be collected directly from the tree, where distinguishing between mature and immature fruits was challenging. Identification of mature fruits relied on phenotypic characteristics such as the thick brown pericarp and large size of the fruits [29], yet some fruits with these traits turned out to be immature. Therefore, it is recommended to implement complementary fruit collection processes, such as placing blankets 2 m high around the trees, to catch the fruits before they reach the ground. This helps prevent fruits from rolling long distances or being preyed upon by rodents due to the terrain topography.

4.2. Germination Trials and Establishment of the Experiment

In the germination tests, several important aspects were identified. The presence of Hypsiphila ferrails larvae in some seeds was observed, which create galleries and damaged the interior of the seeds [44]. This infestation is common in seeds of species of the Carapa genus, so it is crucial to disinfect the seeds before storing them to avoid contaminating the entire seed bank [21]. Additionally, due to the possible collection of immature seeds by mistake in the identification of ripe and unripe fruits, artificial maturation tests were conducted. In these tests, the seeds were not extracted from the fruit, but the whole fruit was preserved and stored in water. The idea was for the seeds to continue acquiring nutrients and protection from the fruit’s pericarp to continue their maturation process. However, the fruit became saturated with water, which accelerated the rotting of the pericarp and the fibrous material (mesocarp) that serves as food for the seed, affecting both the seeds and the fruit. On the other hand, seeds stored in moist forest soil matured successfully, as the soil provided a suitable environment for their development and protection. Therefore, in this study, the only way to avoid discarding immature seeds was to promote their maturation in forest soil. Although the seeds were cut from the tree, they continued to acquire nutrients from the fibrous material of the fruit [41]. However, it is important to avoid water saturation, as this accelerates the decomposition of the fruit’s pericarp and affects the seeds by causing them to rot.

The challenge of properly placing amorphous seeds during sowing was also addressed. For this study, the convex side of the seed was associated with the position of the hilum, which generally indicates the appropriate position for sowing. In other Carapa species, this process may vary, underscoring the need for further research to better understand the differences between species, particularly because reported information on seed positioning in this genus is limited. While propagation studies are underway [21,40,46], the specific seed placement for planting has not been adequately documented.

4.3. Germination of Carapa amorphocarpa Seeds and Seedling Growth in Height and Diameter

The germination of Carapa amorphocarpa is gradual, occurring over a period of 52 days, with the first seeds germinating at 41 days and the last at 93 days. This behavior has also been reported for Carapa guianensis, a species of the same genus, with a germination time between 40 to 180 days [50], 28 to 36 days [51], and 30 to 90 days [52]. The statistical analysis did not show significant differences between treatments. However, when analyzing individual factors, scarification showed moderately significant differences, suggesting that it accelerated germination time. These findings align with Vianna’s statement [53], which suggests that variations in sarcotesta thickness depend on the regional characteristics where the species are found. Mechanical scarification, involving breaking the sarcotesta, influenced the germination time, consistent with the observations of Silva et al. [50], who noted a reduction in germination time due to this pre-germinative treatment.

The germination rate of Carapa amorphocarpa was high at 83.13%. Statistical results demonstrated that seeds without scarification treatment exhibited higher germination rates, and that a mixture of forest soil and sand resulted in the highest average germination rate. These findings are consistent with other studies, where untreated seeds of Carapa guianensis have shown germination rates ranging from 38% [54] to 92% [17]. In contrast, mechanical scarification of seeds has been reported to yield germination rates of 61% [51] and 70% [40], respectively. It is important to mention that Vozzo [41] and Scarano et al. [52] have reported that the embryo seed is large and occupies most of the interior of the seed. Therefore, it is possible that mechanical scarification was not well applied because during the cut, the embryo could have been affected (the sarcotesta was not completely removed). The authors reported that mechanical scarification can increase the percentage of germinated seeds; however, this treatment is recommended only for laboratory studies. The reason is that the protective structure of these seeds is removed, exposing them to the proliferation of fungi; this is why the substrates become an important factor for the germination of seeds with mechanical scarification. Furthermore, this procedure could disrupt the microflora present on the seed surface, including beneficial bacteria and fungi [55]. These microorganisms play a crucial role in promoting seedling growth and maintaining their health [56]. Therefore, it is essential to evaluate the impact of mechanical scarification and similar procedures on the seed-associated microbiota. The inadvertent removal of this microflora during scarification could lead to a decline in seedling development. One potential strategy by which to mitigate this negative effect could involve reintroducing beneficial microflora during the planting process [57].

The forest soil used in the study was characterized by high contents of organic matter, and according to Silva et al. [50], the seeds of the genus Carapa need substrates additional organic matter, even though they contain rich mineral nutrients. On the other hand, Barbaro et al. [58] stated that an ideal substrate must have porosity and great moisture retention capacity, fast drainage, and good aeration. In this way, the sand used in the substrate allowed the seed to germinate in favorable conditions; in addition, the sand has demonstrated several essential functions such as moisture retention and cationic exchange [59].

Carapa is one of the genera of the Meliaceae family that hosts Hypsipyla grandella Zeller and Hypsipyla ferrealis [44]. For example, the borer Hypsipyla grandella pierces the terminal shoot and kills it, causing a bifurcation and a twisted stem [60,61]. Additionally, Lyctus beetle and pin borer attacks have been reported on the bark and sapwood of Carapa guianensis [60]. However, these pest attacks occur on the ground, that is, outside of nursery care, where the seedlings are outdoors. Hence, Ferraz et al. [40] concluded that the excellent health and shape results obtained in the Carapa guianensis propagation study are due to the protection and proper management of the species in the nursery.

Based on some propagation studies of Carapa guianensis in the nursery, the initial growth averages in height vary between 30 and 40 cm in approximately 3 months [54,62]. Therefore, it is evident that Carapa amorphocarpa has reported better results in our study in less time (Figure 7). However, there is no comparative germination study between these species, so the variation of these results can be determined by other factors such as pregerminative treatments and climatic conditions of the experimental area. In the present study, better results of initial growth in diameter and height are evidenced for treatments without mechanical scarification. It should be noted that the sarcotesta acts as a protection of the seed to maintain its nutritional reserves during the first stages of growth [41,63]. In this way, it was shown that the cotyledons of the seeds with pregerminative treatment decomposed faster. Therefore, there is a possibility that seedlings that did not retain their cotyledons for a long time had limitations in the supply of nutrients that come from the seed.

It should be noted that the germination tests were carried out very close to the place where the species grows; that is, the seeds were subjected to conditions of humidity, temperature, and soil, similar to the conditions where the seeds germinate naturally. According to Ferraz et al. [40], the phytosanitary status of the plants—and, in turn, their survival—is directly related to the care given to seedlings in the nursery. The root developed poorly, although the plants reached significant heights, especially due to the length of the epicotyl. It can be assumed that large cotyledons contribute to the development of the seedling in the first months of growth since numerous cotyledosperm seeds with long, fleshy, reserve-rich cotyledons such as Carapa fulfill this function [41].

These findings suggest that while there is a significant relationship between root length and both total height and basal diameter, the correlation between root length and total height may not be as robust as initially expected. This highlights the complexity of the interactions between root development and above-ground growth in the species under study [64].

The study revealed high costs for the propagation of Carapa amorphocarpa in the study area. These costs are primarily due to the difficulty in obtaining genetic material (seeds). Another significant aspect is that it is uncommon to establish plantations of Carapa amorphocarpa in the northern zone of Ecuador, where there are extensive plantations of Eucalyptus spp. and Pinus spp. It would be important to analyze the establishment of Carapa amorphocarpa seedlings in the field to determine their performance across treatments using the plot volume index (PVI) [65]. This approach could help include this species in restoration programs or commercial plantations, leveraging the ecosystem services provided by the seeds, such as oil extraction. A further study would be necessary to explore the feasibility of extracting oil from the seeds and incorporating this practice into sustainable use programs [66].

5. Conclusions

A key finding of this study was the impact of mechanical scarification on Carapa amorphocarpa seed germination. While scarification led to faster germination times (around 70 days), it also resulted in lower germination rates compared to unscarified seeds (which achieved germination rates exceeding 98%). This suggests that scarification may damage the seed embryo and compromise germination success. Despite this, the overall germination rate across all treatments remained high (over 83%), and seedlings exhibited excellent health and a remarkable survival rate of over 99% at 60 days after germination. This indicates that scarification is not essential for Carapa amorphocarpa propagation, but prioritizing methods for collecting high-quality seeds might be more crucial. This study also investigated the influence of different substrates on germination and seedling development. While there were no significant differences between substrates, a mixture of forest soil and sand emerged as the most favorable for germination percentage. This can be attributed to the beneficial properties of sand, such as improved moisture retention, aeration, and drainage. Interestingly, seedlings grown from unscarified seeds displayed better initial growth in both height and diameter. This might be due to the protective seed coat providing essential nutrients during the early stages of development. One notable observation was the underdeveloped root system of the seedlings compared to their height. This could be explained by the reliance on nutrients from the large cotyledons in the early stages. The correlation between total height and basal diameter exhibited a very strong relationship, implying well-proportioned above-ground growth. However, the correlation between root length and height was considerably weaker, indicating a potential area for further investigation to enhance root development in Carapa amorphocarpa seedlings. Addressing this aspect will be vital for ensuring their long-term survival and successful establishment in the field.