Introduction

Brosimum alicastrum Swartz, a tree species with significant potential for both animal and human consumption, grows naturally in the southeast of Mexico (Santillán-Fernández et al., 2024). Several studies have contributed to the understanding of seed propagation, nursery production, mycorrhizal associations, establishment under natural conditions, and the species’ response to climate change, providing valuable insights for its silvicultural management. However, this body of knowledge remains limited, which has hindered the development of commercial plantations (Espinosa-Grande et al., 2023b). The successful establishment of such plantations requires trees with outstanding phenotypic traits; therefore, the propagation method used to obtain seedlings and clones is crucial for effective silvicultural management (Hernández-González et al., 2015).

The propagation of individuals with desirable traits for plantation establishment can be achieved either sexually, through seeds, or asexually, using techniques such as cuttings, grafting, and air layering. With advances in biotechnology, newer methods like in vitro propagation are becoming more widely used (Asadi & Shekafandeh, 2021). Regardless of the method chosen, effective nursery management is essential, as it directly influences the morphological quality of the seedlings and their capacity to adapt to field conditions (Rueda-Sánchez et al., 2012).

There is limited information available on propagation methods for B. alicastrum (Espinosa-Grande et al., 2023a). Among vegetative propagation techniques, air layering and grafting have shown the most promising results (Ahsan et al., 2019). These asexual reproduction methods enable individuals with outstanding phenotypic and genotypic traits, as they are designed to replicate a selected parent tree. Therefore, careful selection of donor trees is crucial and should be based on the species' characteristics and intended end use (Yang et al., 2019).

Air layering and grafting have played a key role in fruit, horticultural, and forestry production. These techniques allow for the mass propagation of individuals with desirable traits, management of soil and disease-related issues, extension of crop lifespan, propagation of species that are difficult to grow from seed, and the rapid acquisition of germplasm with valuable genetic information and economically important traits in shorter timeframes (Espinoza-Arellano et al., 2012). However, these techniques reduce the genetic diversity of the species by the uniformity of the offspring, so the selection of parent trees helps preserve this diversity (Xu et al., 2022).

Research on asexual propagation techniques for B. alicastrum is nearly nonexistent (Espinosa-Grande et al., 2023b). Therefore, research on effective mass propagation strategies for the species, while maintaining its genetic variation and productivity, is valuable (Hernández-González et al., 2015). This is particularly important given the species' role in the Yucatán Peninsula as an alternative plant resource for both animal and human nutrition in the context of food security and climate change (Ramírez-Sánchez et al., 2017). In this regard, the objective of the present study was to determine the asexual propagation method (grafting or air layering) that produces individuals with outstanding or suitable tree-measurement characteristics in the nursery, to evaluate their survival and growth in a plantation.

Materials and Methods

Study area and germplasm collection

The study was conducted at the experimental nursery located at the facilities of the Colegio de Postgraduados, campus Campeche (ColPos Campeche). The plant material of B. alicastrum was collected from locations near the research center (Figure 1). The region is characterized by evergreen tropical forests, clayey soils, and precipitation ranging from 800 to 1 200 mm, with dry periods lasting from three to seven months. The average annual temperature is 30 °C, with altitudes ranging from 30 to 80 m and a warm, humid climate (Comisión Nacional para el Conocimiento y Uso de la Biodiversidad [ CONABIO ], 2024).

Seed collection for rootstock production

In November 2021, 500 B. alicastrum fruits were collected from the evergreen tropical forest of Santo Domingo Kesté following the methodology described by Vallejos et al. (2010). Thirty trees with the best tree-measurement characteristics were selected, including greater height, larger diameter at breast height, and straight stem, with a minimum distance of 100 m between selected trees. The fruits were stored in sterilized plastic bags to be transported to the facilities of ColPos-Campeche in Champotón, Campeche, Mexico.

Damaged fruits, fruits with absent or damaged seeds, or fruits showing signs of pest and disease damage were discarded. The fruit was pulped to expose the seed, which was then washed with water to remove any remaining pericarp residues. On November 25, 2021, 400 seeds were placed to germinate in an area of 5 m x 5 m, which had been previously plowed and harrowed in the ColPos-Campeche forest area. The seeds were planted at a depth of 10 cm and spaced 25 cm apart, arranged in a rectangular grid pattern. Two weekly waterings of 1 liter per seedling were applied. The seedlings were maintained at the planting site for future use as rootstocks.

Asexual propagation with air layering

In late November 2022, an in vivo propagation experiment was carried out using air layering on young B. alicastrum trees (under 5 years old) from a plantation in Sihochac, Champotón, Campeche. During a field survey, healthy, vigorous, pest-free trees with straight stems and no branching were selected. A completely randomized experimental design with a factorial arrangement was used, where three substrate levels were evaluated (coconut fiber®, Peat moss®, and forest soil collected from areas where the species naturally grows) and two rooting hormone levels (Radix 10000® 5 g per experimental unit and no rooting hormone), resulting in a total of six treatments with 20 repetitions per treatment.

Air layering was carried out on branches located at the selected trees' mid-crowns. Vigorous branches with an average diameter of 1 cm were chosen. Air layers were established 80 cm from the tip of the branch to the tree stem. Using a knife, a ring was made by removing the bark over an area of 2 cm, and to prevent scarring, the cambium and phloem were scraped off. Transparent plastic bags containing the substrates from the six treatments, previously moistened, were placed around the branches. The ends were sealed with black bags to prevent light and moisture loss, and they were watered every 15 days with 15 mL of water using a syringe, following the methodology described by Durán-Casas et al. (2013).

The survival rate (%) for each treatment was evaluated after a period of 90 days from the start of the experiment. Through destructive sampling, six specimens were randomly selected per treatment, and both root and shoot variables were measured. Root measurements included the length of the primary root (cm), number of roots, and fresh root weight (g) after removing substrate residues. Shoot measurements included plant diameter (cm), plant height (cm), fresh weight (g) of the stem without leaves, and fresh weight of the leaves. Additionally, dry weight (g) of roots, stems, and leaves were recorded after placing the plant material in a forced-air oven at 70 °C for 24 h.

To assess the effects of the substrates, the rooting agent, and their interaction on the development of the air-layered plants, the data was analyzed using an ANOVA and mean separation with Tukey’s test (α = 0.05). Pearson correlation analysis (α = 0.05) was used to determine the relationships between shoot and root variables. Additionally, treatments with similar effects on plant development were grouped using principal component analysis (PCA, based on correlation methods). All statistical analyses were carried out using the open-access R statistical software, following verification of variable normality through the Shapiro-Wilk test (Venables & Smith, 2023).

Asexual propagation by grafting

In late November 2022, an in vivo grafting experiment was conducted at the experimental nursery of ColPos-Campeche. For the rootstock, B. alicastrum seedlings (one-year-old Ramón saplings) grown from seed were used. As scion material, vegetative cuttings (scion sticks) were collected from six mature Ramón trees-over five years old-located in the community of Hool. These trees were selected for their ability to produce healthy, vigorous, pest-free fruits and for having straight, unbranched stems. The identity of each source tree was maintained throughout the grafting process.

The selected scions were wrapped in a moist towel, transported from Hool to the campus Campeche, and placed vertically in containers with water to keep them hydrated until grafting, which took place the same day. Axillary buds of similar size and with green and erect meristems were extracted from some scions by means of a longitudinal cut.

The grafting techniques applied were crown grafting, bud grafting, side veneer grafting, and whip and tongue grafting. A completely randomized experimental design was implemented, with each grafting technique considered as a treatment and replicated 20 times. Graft success rate (%) was evaluated at 30, 60, and 90 days after grafting. At 90 days, additional measurements were taken, including increase in stem diameter (cm), height (cm), and the number of shoots produced by the grafted material. Data was analyzed using an ANOVA and Tukey’s test (α = 0.05) for mean separation, in the open-access R statistical software, after confirming the normality of the evaluated variables with the Shapiro-Wilk test (Venables & Smith, 2023).

Setting up a plantation air-Layered and grafted plants

Ninety days after the in vivo experiments were established, up to 10 surviving specimens per treatment, both air-layered and grafted, were randomly selected for establishment in a formal plantation 30 days later, at the facilities of the Colegio de Postgraduados, campus Campeche (Figure 2). Once separated from the mother plant, the air-layered specimens were transferred to 3 kg capacity bags containing their original substrates to allow continued root development

The selected grafted plants were removed from the area where their rootstocks had originally been planted. Additionally, in the plantation, B. alicastrum seedlings produced in the nursery from seed in November 2021, of the same age as the rootstocks, were included as controls. The plantation was established in late March 2023, following a completely randomized experimental design with a square planting arrangement, using a spacing of 3 meters between plants over a 1 000 m² area. Surviving individuals from the in vivo propagation phase were randomly distributed across the plot, including 30 grafted plants, 25 air-layered plants, and 10 control plants.

To prevent edge effects from influencing the experiment, B. alicastrum seedlings propagated from seed were planted around the perimeter of the plantation. The land was plowed and harrowed, and for the excavation of planting holes, dimensions of 30 cm in diameter by 40 cm in depth were used, with a 2-meter distance between experimental units. At the time of transplanting, worm compost was used to fill the planting holes of all experimental units in order to improve soil conditions. Given that it was a drought season for the region (April-July) (Santillán-Fernández et al., 2021a), a drip irrigation system was installed, and 10 liters of water were applied twice a week to each experimental unit.

The variables analyzed included survival rate (%), flowering (when at least one flower appeared, %), fruit production (when at least one fruit appeared, %), total height (cm), basal diameter (cm), and the number of leaves in each experimental unit for the treatments of air layering, grafting, and control. Measurements were taken at the end of each month from March to July 2023. The measurement from March was used as a baseline, and the monthly increments were analyzed from that point onward. Temporal representations of the increments for each treatment were created for the variables of total height, basal diameter, and number of leaves. Statistical differences were determined through the ANOVA and mean comparison using Tukey’s test (α = 0.05), with the aid of the open-access R statistical software, following verification of the normality of the evaluated variables using the Shapiro-Wilk test (Venables & Smith, 2023).

Results and Discussion

Sexual propagation

Out of the 400 seeds planted, 307 (76.75 %) germinated within 10 to 15 days. After 90 days, the survival rate was 64.25 % (257 plants), and after one year, 205 plants (51.25 %) survived, with an average height ranging from 80 to 100 cm and diameters between 12 and 15 cm. Santillán-Fernández et al. (2021b) recorded a germination rate of less than 60 % under similar conditions, a limitation explained by the recalcitrant nature of B. alicastrum seeds, which makes them more sensitive to dehydration and rapid loss of viability (Santillán-Fernández et al., 2023).

Asexual propagation by layering

For all treatments, survival at 90 days exceeded 80 %. The shoot variables were significantly influenced (Pr < 0.05) by the substrate factor, while the rooting hormone and the substrate × rooting hormone interaction did not statistically affect these variables (Pr > 0.05), except for leaf fresh and dry weight. In the case of root variables, the rooting hormone factor was also not statistically significant (Pr > 0.05), whereas the substrate factor and the substrate × rooting hormone interaction were statistically significant (Pr < 0.05) only for the number of roots and the length of the primary root (Table 1).

Table 1. Analysis of variance to determine the effect of substrate and rooting hormone on the viability of Brosimum alicastrum air layers

The influence of substrates on the development of B. alicastrum clones propagated via air layering has been previously reported by Santillán-Fernández et al. (2021b), who concluded that substrate type plays a more critical role in determining clone quality than fertilization or irrigation. Despite these findings, studies on this topic remain scarce, underscoring the need for further research into how different substrates affect the growth and development of B. alicastrum air layers (Espinosa-Grande et al., 2023a; Santillán-Fernández et al. 2021b).

Coconut fiber and peat moss substrates showed the highest mean values for both shoot and root system, compared to the forest soil substrate (Table 2). Santillán-Fernández et al. (2021b) also found that the best characteristics in B. alicastrum clone development were achieved using commercial inputs; however, this does not necessarily equate to higher clone quality, and they recommend further evaluation under formal plantation conditions

Table 2. Effect of the type of substrate on Brosimum alicastrum layering viability.

In plant physiology, roots and leaves are two essential components. Roots enable plants to adapt to the soil environment and absorb water and nutrients, while leaves carry out photosynthesis (Rodríguez-Pérez, 2006).

In the present study, the interaction between substrate and rooting hormone was statistically significant (Pr < 0.05) for certain root and leaf variables (Table 3). Peat moss with or without rooting hormone, resulted in the greatest primary root length; however, the highest number of roots was observed when only forest soil was used. According to Liao-Torres et al. (2012), in clones of tropical forest species, the development of a greater number of roots prior to separation from the mother plant depends on the substrate used. This is important, as it improves the new plant’s ability to adapt to firm soil.

Table 3. Effect of substrate and rooting hormone interaction on Brosimum alicastrum layering viability.

By correlating the shoot system variables of the air layers with the root system variables using Pearson’s correlation, a strong, directly proportional relationship (ρ > 0.95) was found between fresh and dry weights of root, stem, and leaf variables (Figure 3). However, plant diameter (DPlant) was inversely correlated with root length (LRoot, ρ = -0.82), root dry weight (DWRoot, ρ = -0.67), root fresh weight (FWRoot, ρ = -0.61), number of roots (NRoot, ρ = -0.41), and plant length (LPlant, ρ = -0.56). In contrast, plant length (LPlant) showed a positive correlation with root length (LRoot, ρ = 0.79), root fresh weight (FWRoot, ρ = 0.89), and root dry weight (DWRoot, ρ = 0.88). These results suggest that thinner branches showed better root development than thicker ones. Findings agree with those reported by Santillán-Fernández et al. (2021b) for the evaluation of B. alicastrum air layers in Campeche; however, there is limited scientific literature documenting the air layering propagation technique for this species (Espinosa-Grande et al., 2023a).

The principal component analysis (PCA) revealed that the first component (Prin1) grouped variables related to the root system and accounted for 51.75 % of the total variance. The second component (Prin2) included the shoot variables of the air layers, explaining 30.39 % of the variance. Together, these two components explained 82.14 % of the total variance. In Prin1, the most influential variables were root length (+0.44) and number of roots (-0.43), while in Prin2, the key variables were plant length (+0.51) and plant diameter (-0.50; Table 4). Negative values indicate an inversely proportional relationship, whereas positive values indicate a directly proportional one (Abdi & Williams, 2010).

Table 4. Eigenvectors from the principal component analysis grouping the main variables of shoot and root systems of Brosimum alicastrum air layers.

Group 1. Treatments with peat moss, characterized by air layers with the greatest shoot length and longest roots, but the smallest stem diameters and the fewest adventitious roots.

Group 2. Treatments using forest soil. These air layers were distinguished by having intermediate root length but the highest number of adventitious roots. Additionally, although shoot size was the smallest, plant stem diameters were the largest.

Group 3. Treatments with coconut fiber. These air layers developed shorter roots, an intermediate number of adventitious roots, and moderate values for shoot size and stem diameter.

The best shoot and root system characteristics were observed in air layers using peat moss, where the application of rooting hormone Radix 10000® had no significant effect. However, the air layers using forest soil without rooting hormone developed the highest number of adventitious roots, which may enhance the likelihood of successful adaptation to firm soil (Espinosa-Grande et al., 2023b). Alvarado-Aguayo and Munzón-Quintana (2019) reported that the effect of rooting hormones in tropical forest species is significant when combined with fertilization and irrigation, which in our case were limited.

Asexual propagation by grafting

The survival rate (%) of B. alicastrum experimental units propagated through grafting techniques under nursery conditions was evaluated at 30, 60, and 90 days (Table 5). At 30 days, all four grafting techniques showed some level of graft success. However, by the 90-day evaluation, the bud grafting technique resulted in total mortality of its experimental units. Santillán-Fernández et al. (2021b) reported that bud grafting is not a viable technique for this species, and our findings corroborate this conclusion.

The side veneer and whip and tongue grafting techniques had the highest mean graft success rates among the evaluated morphological parameters. These results are consistent with those reported by Santillán-Fernández et al. (2021b) in regions of Campeche, where the species grows naturally. The side veneer, whip and tongue grafting, and crown grafting are all scion grafting techniques, in which direct contact is established between corresponding tissues (epidermis, vascular tissue, and parenchyma) of both the rootstock and the scion. In contrast, bud grafting involves primarily bark tissue and a minimal portion of cambium in the bud, which is attached on a tangent of the cambium of the rootstock (Xu et al., 2022). This structural difference may explain why bud or patch grafts are not viable for B. alicastrum.

Table 5. Graft survival and its effect on the quality of Brosimum alicastrum clones.

Evaluation of a plantation using layering and grafting

At 120 days after the establishment of the in vivo experiments, grafting treatments showed survival rates exceeding 60 %, with crown grafting showing the highest survival (80 %, 16 experimental units), followed by whip and tongue grafting (70 %, 14 units) and side veneer (65 %, 13 units). Bud grafting, however, resulted in complete mortality, with no surviving experimental units. In the case of air layering treatments, the application of rooting hormone did not have a statistically significant effect. Furthermore, due to the destructive sampling process and mortality of some experimental units, treatments were grouped based on the type of substrate: forest soil (16 surviving units, combining those with and without rooting hormone), peat moss (14 units, combining those with and without rooting hormone), and coconut fiber (5 units).

In the initial plantation, five experimental units from the air layering treatment using coconut fiber as substrate were included; however, all five units failed to survive within the first 20 days. From April to July, the greatest increases in height and stem diameter were observed in individuals propagated by seed (control group) (Figure 5). According to Molina-Escalante et al. (2015), this can be attributed to the species’ physiology, as seedlings propagated sexually tend to show greater growth compared to those propagated asexually. Additionally, the control plants were propagated in November 2021, whereas the air layers and grafts were established in November 2022.

On the other hand, when analyzing the cumulative increases in tree-measurement variables from April to July (using the initial measurements taken at the time of planting in March as the baseline), it was found that the air layering_Forest soil group showed the highest growth increments, with the highest flowering (70 %) and fruit production (50 %) percentages (Table 6). In fact, the fruits from air-layered plants were visually larger compared to those from grafted plants (Figure 6). However, there is limited scientific literature reporting on propagation techniques using air layering and grafting for B. alicastrum (Espinosa-Grande et al., 2023a), which suggests that these findings may contribute to improving the propagation of a species with emerging economic potential.

Table 6. Increases in tree-measurement variables of Brosimum alicastrum individuals propagated by asexual techniques (air layering and grafting) and sexual techniques (seed) from April to July in Campeche, Mexico.

Conclusions

In air layering propagation, the type of substrate favored root development and the growth of Brosimum alicastrum clones, whereas the application of rooting hormones had no significant effect. Therefore, under conditions similar to those studied, the use of rooting hormones is not considered necessary. Bud grafting was also not a viable method. It is recommended to explore alternative grafting techniques, given the limited knowledge of their effects on this species, as well as to examine other management factors and the timing of propagation. It is important to note that while some air layering and grafting techniques showed positive results in nursery, the propagated individuals did not survive after transplantation. Clones propagated through air layering using forest soil as a substrate had the greatest increases in growth, flowering rates, and fruit production in plantation, likely due to enhanced root development.