Mini-cutting rooting and plantlet growth in Erythrina crista-galli L.

Submitted on July 17 , 2020 and accepted on October 16 , 2020. This work is part of the second author’s final course assignment. 2Universidade Federal de Santa Maria, Departamento de Fitotecnia, Santa Maria, Rio Grande do Sul, Brazil. dilson.bisognin@ufsm.br 3Universidade Federal de Santa Maria, Santa Maria, Rio Grande do Sul. Brazil. lopes.gabriel376@gmail.com, angelcmalheiros@gmail.com 4Universidade Federal de Santa Maria, Colégio Politécnico, Santa Maria, Rio Grande do Sul, Brazil. renatrevisan@gmail.com 5Universidade Federal de Santa Catarina, Departamento de Agricultura, Biodiversidade e Florestas, Curitibanos, Santa Catarina, Brazil. kelen.lencina@ufsc.br *Corresponding author: dilson.bisognin@ufsm.br Mini-cutting rooting and plantlet growth in Erythrina crista-galli L.


INTRODUCTION
Erythrina crista-galli L., commonly referred to as swamp cork, is a deciduous tree occurring in swampy or marshy floodplains, from Maranhão to Rio Grande do Sul in Brazil. Under these conditions, swamp cork hosts epiphytic plants and attracts several birds, such as hummingbirds, and insects (Gratieri-Sossella et al., 2008), such as bees, which are its main pollinating agents (Galleto et al., 2000). The swamp cork has great ornamental potential, as well as utilization in unprotected and degraded areas. Plantations or natural regeneration are usually from seminal seedlings. Seeds produced from selffertilization, common in fragmentations of original forests, present lower fresh weight and germination than those from cross-fertilization (Galleto et al., 2000).
Vegetative propagation is an alternative for plantlet production that allows multiplication of selected individuals and establishment of uniform populations by cloning. Clonal plants can be produced from different vegetative propagation techniques. In the case of forest species, cutting or its variations (mini-cutting and microcutting), are the most used technique for mass production of plantlets. This is due to the fact that vegetative propagules are produced in clonal hedges established from seminal seedlings or rooted cuttings. Although these aspects are more notable for exotic species, scientific and technological development in vegetative propagation by mini-cutting of native species has been improved (Pimentel et al., 2016;Rodrigues et al., 2017;Burin et al., 2018).
The success of plantlet production by mini-cutting depends on adventitious rooting. Several factors are involved in this process, such as the species and the genotype to be propagated, the quality of the root system being formed, and the growth of the produced plantlet (Neves et al., 2006). In addition, moisture and temperature in the rooting environment, composition of the substrate, use and concentration of auxin, and type and size of the propagule (Pimentel et al., 2016;Rodrigues et al., 2017) affect the rooting percentage. Studies on native tree species have shown that the type of mini-cutting affects adventitious rooting and growth in Cedrela fissilis (Xavier et al., 2003), while the mini-cuttings origin, leaf area and size, and concentration of indolebutyric acid affect rooting competence in Handroanthus heptaphyllus (Rodrigues et al., 2017). Therefore, improvements in adventitious rooting competence at genotype level depends upon finetuning of some of these factors.
Considering that only herbaceous mini-cuttings of swamp cork plants up to one-year-old have high rooting potential (Gratieri-Sossella et al., 2008), and no reference was found discussing factors affecting mini-cutting rooting, this research was carried out to evaluate the application need of different concentrations of indolebutyric acid (IBA) for adventitious rooting, the different substrate composition for mini-cutting rooting and plantlet growth, and the rooting curve of two swamp cork clones.

MATERIAL AND METHODS
The experiments were conducted in an acclimatized greenhouse of the Center for Plant Breeding and Vegetative Propagation, Plant Science Department of the Federal University of Santa Maria, State of Rio Grande do Sul, Brazil. The mini-clonal garden was established in August 2016, with plantlets originated from rooted mini-cuttings of the clones 15SM05 and 15SM08. The mini-clonal garden was established and managed according to the methodology used for other native species, such as Cabralea canjerana (Burin et al., 2018) and H. heptaphyllus (Rodrigues et al., 2017).
After the establishment, the plantlets were coppiced and periodically pruned for mini-stump formation. The collected shoots were used for preliminary experiments and definition of treatments. For all experiments, approximately 20 cm long shoots of the 15SM05 and 15SM08 clones were collected in the mini-clonal garden and sectioned in mini-cuttings of approximately 3 cm in length with a single bud and two leaflets with 50% of the total area. For rooting, mini-cuttings were planted in 110 cm³ polyethylene tubes and placed in a humidity chamber, equipped with a misting system to maintain relative humidity above 80%. The humidity chamber is localized in an acclimatized greenhouse.
The experiments were carried out to study the need of indolebutyric acid (IBA) application (March 2017), substrate composition for mini-cutting rooting (April 2017) and plantlet growth (May 2017), and rooting curves of the clones 15SM05 and 15SM08 (August 2017). In one experiment, initiated in March 2017, mini-cuttings were immersed in a hydroalcoholic solution of 0, 1000, 2000, or 3000 mg L -1 of IBA for 10 s. The mini-cuttings were planted in polyethylene tubes containing a mixture of a commercial substrate with pine bark and vermiculite of medium granulometry in the proportion of 2:1 (v/v) and placed in the humidity chamber. The percentages of survival, rooting, and shooting, number and length of roots, and number and length of shoots were evaluated at 30 days. The experiment was conducted as a factorial (clones × IBA concentrations) in the completely randomized design, with six replications of five mini-cuttings.
In another experiment, initiated in April 2017, minicuttings were planted in tubes containing different substrates. For rooting, expanded vermiculite of medium granulometry, coarse sand, a commercial substrate with pine bark, and the combination of commercial substrate and vermiculite of medium granulometry (proportion 2:1v/v) were evaluated. At 30 days of cultivation in the humidity chamber, mini-cuttings were subjected to the same evaluations described in the previous experiment. The experiment was conducted as a factorial (clones × substrates) in the completely randomized design, with seven replications of five mini-cuttings.
The rooted mini-cuttings obtained in the previous experiment were planted in 110 cm³ tubes in May 2017 to evaluate the plantlet growth in different substrate compositions. The commercial substrate with pine bark and vermiculite of medium granulometry (proportion 2:1v/v), commercial substrate and soil of subsoil (proportion 1:1 -v/v), and commercial substrate and soil of subsoil (proportion 2:1 -v/v) were evaluated. In addition, 5 g L -1 of fertilizer Osmocote™ (N = 15, P 2 O 5 = 9 and K 2 O = 12) was added to all substrate compositions. The plantlets were kept in an acclimatized greenhouse for 120 days. Every 30 days, the plantlets were evaluated for survival percentage, number of leaves, shoot height, stem diameter, and shoot height:stem diameter ratio. The experiment was a factorial (clones × substrates) in the completely randomized design, with six replications of four plantlets.
In August 2017, mini-cuttings were also planted in polyethylene tubes containing expanded vermiculite of medium granulometry in order to study the rooting curve of each clone and determine the ideal time of mini-cuttings cultivation in the humidity chamber. The percentage of cuttings undergoing rooting and the number and length of formed roots were evaluated every seven days, until 42 days of cultivation in the humidity chamber. The accumulated rooting refers to the total number from the beginning until the evaluated day. Therefore, the total number was gotten at 42 days of cultivation in the humidity chamber for the percentage of rooting and the number and length roots. The daily current increment (DCI) and the daily mean increment (DMI) of rooting where calculated as DCI = X (i+1) -X (i) and DMI = X (i) / T (i) , respectively, where: i = time of evaluation; X (i+1) = percent of rooting, number, or length of roots at time (i+1); X (i) = percent of rooting, number, or length of roots at the evaluation time; and T (i) = number of days at evaluation (Ferreira et al., 2004), which intersection determines the ideal time of mini-cuttings cultivation in the humidity chamber. The experiment was conducted in the completely randomized design, with five replications of ten minicuttings.
The analysis of variance was used to test the effects of the treatments. The different levels of any factor that were found to lead to significant differences in the tested traits by the ANOVA F-test (p < 0.05) were further compared using Tukey's test, with a 5% type-I error probability (p < 0.05). To meet the assumptions of normality and homogeneity, the percentage data were transformed to arcsine of and counting data to . All analyses were done with the aid of the software Sisvar, version 5.6 and Excel 365.

RESULTS AND DISCUSSION
There was no interaction between clones and concentrations of IBA for all evaluated traits (Table 1). Mini-cuttings presented a high percentage mean of rooting (80.8%), with all coefficients of variation below 25%, except for the percentage of shooting. Clones did not differ for percentages of survival and rooting and number of roots, but the clone 15SM05 was superior to 15SM08 for the length of roots and percentage, number, and length of shoots. The IBA application only affected the percentage of rooting. Increasing the concentration of IBA negatively affected the percentage of rooting, while the application of 1000 mg L -1 did not differ from the control treatment. Thus, mini-cutting rooting of swamp cork did not depend upon the IBA application.
The clones of swamp cork presented high adventitious rooting competence, which was expected since these clones were previously selected for mini-cutting rooting. The clone 15SM05 showed a 7-fold higher percentage of shooting than the 15SM08, which is important in a new plant for reaching an equilibrium between root and shoot growth and development. These selected clones presented high rooting competence without the need of IBA application, which is very important for plantlet production by mini-cutting as IBA application is laborious and costly. Similar results were observed for other tree species, such as Vochysia bifalcata (Rickli et al., 2015), Eucalyptus grandis (Wendling & Xavier, 2005), Peltophorum dubium (Mantovani et al., 2017) and H. heptaphyllus (Rodrigues et al., 2017;Oliveira et al., 2015), for which the application of AIB did not improve the percentage of the mini-cutting rooting.
There was no interaction between clones and substrate compositions for all the evaluated traits ( Table 2), showing that the requirements for substrate composition were similar for both clones. High percentage mean of rooting (81%) and low coefficient of variation were also found for different substrate compositions. Clones of swamp cork did not differ for the percentages of rooting and survival, but the clone 15SM05 was superior than the 15SM08 one for the number and length of roots and the percentage, number, and length of shoots. Vermiculite and its combination with commercial substrate showed the best results for the percentage of rooting and the number and length of roots and shoots. Commercial substrate presented the highest percentage of shooting and, together with vermiculite, the highest percentage of rooting.
Although sand is a substrate that provides some aeration and porosity, and it is easy to acquire at low cost, it resulted in lower rooting percentages, probably due to   its high bulk density. The bulk density is directly affected by compaction, which results in mechanical resistance, reducing the porosity of the substrate and thus limiting the infiltration and redistribution of water, making it difficult to aerate (Mateus et al., 2017). The physical characteristics of the substrate, such as high density and small particle size, reduce aeration and affect respiration (Garcia et al., 2012), which is essential for rooting. In this experiment, vermiculite and its combinations with commercial substrate have lower bulk density than the combination of commercial substrate, vermiculite and coarse sand (Pimentel et al., 2016). As vermiculite has a very low bulk density (130 kg m -3 ), it contributes to lower substrate density and, therefore, increasing the percentage of rooting and the number and length of roots. The greatest rooting of H. heptaphyllus mini-cuttings was also obtained with the combination of commercial substrate and vermiculite, corresponding to a substrate with low bulk density and high aerial space, total porosity, and water retention capacity (Pimentel et al., 2016). Besides improving rooting percentage, vermiculite also favored the number and length of roots and shoots. Longer roots allow absorption of water and minerals at higher depths, increasing the adaptability to unfavorable soil and climatic conditions, such as greater resistance to water deficit periods. Rooted mini-cuttings were then planted in 110 cm³ tubes with different substrate compositions for growth evaluation in the greenhouse. Switching vermiculite to soil from subsoil increases the bulk density of the substrate composition from 297 kg m -3 to 590 kg m -3 and reduces the percentage of porosity from 78 to 73 (data not shown), without affecting any of the evaluated traits during 120 days of cultivation of swamp cork plantlets in the greenhouse (Figure 1). This should be associated with the adaptation of swamp cork plants to different soil types and conditions. Since vermiculite was important for minicutting rooting, the combinations of commercial substrate and vermiculite (CS+V 2:1 -v/v) should be the choice of substrate composition for plantlet production of swamp cork. In this experiment, the clone 15SM05 presented the highest number of leaves, shoot height, and shoot height by stem diameter ratio already at 30 days of cultivation in the greenhouse. After 120 days of cultivation, the clone 15SM05 was superior to 15SM08 for all evaluated traits, with 89% of plantlets alive. The relationship between shoot height and stem diameter of the plantlets at 120 days of cultivation was 6.8, 2.8, and 4.8 for the clones 15SM05, 15SM08, and average substrates, respectively. These values indicate a high morphological quality of the swamp cork plantlets, considering that the relationship should be less than 10 (Birchler et al., 1998), and presented a balanced growth between shoot height and stem diameter, which provides greater resistance and better fixation to the soil when planted in the field (Artur et al., 2007).
The cork swamp clones also differed for the current and mean daily increments, and the accumulated rooting percentage (Figure 2). After 21 days of cultivation in the humidity chamber, the clone 15SM05 already presented 87.5% of rooted mini-cuttings compared to 0% of the 15SM08 clone. The percentage of rooting at 28 days of the clone 15SM05 was 91.7%, whereas that of the clone 15SM08 was 39.6%, reaching 87.5% after 42 days. The intersection between the current and mean daily increments determines the ideal time for the removal of mini-cuttings from the humidity chamber, as indicated by Ferreira et al. (2004) and confirmed by Melo et al. (2011). The interception occurred at 28 days of cultivation for the clone 15SM05, seven days after the maximum daily current increment. This extra time of cultivation in the humidity chamber should improve mini-cutting rooting and plantlet survival during acclimatization (Melo et al., 2011).   Therefore, mini-cuttings of the clone 15SM05 should be maintained in the humidity chamber for 28 days for rooting. The current daily increment did not intercept the mean daily increment until the mini-cuttings of the clone 15SM08 were removed from the humidity chamber. Therefore, minicuttings of the clone 15SM08 were transferred for the acclimation in an earlier rooting stage compared to 15SM05. These probably explain the differences between clones for mini-cuttings rooting and shooting (Tables 1 and 2) and plantlet survival and growth (Figure 1). These results confirmed that mini-cuttings should be removed from the humidity chamber after the interception of current and mean daily increments (Ferreira et al., 2004;Melo et al., 2011), since early removal may affect plantlet survival and growth (Melo et al., 2011).
The number of roots followed the same pattern as the rooting percentage for both clones, with the maximum daily current increment at 21 days for 15SM05 and 28 days for 15SM08 (Figure 2). Regarding the length of the roots, the maximum daily current increment was at 35 days of cultivation for both clones. Based on the percentage of rooting and the number of roots, minicuttings of the clone 15SM05 should be cultivated in a humidity chamber for 28 days, while those of the clone 15SM08 should be cultivated for 42 days. Therefore, differences in rooting competence of the swamp cork clones (Tables 1 and 2 and Figure 1) were more associated with the day of evaluation, as they presented similar percentages of rooting after 42 days of cultivation in the humidity chamber. These differences in the adventitious rooting competence between swamp cork clones were also found in other forest species. Variations among clones and the position on the trunk where epicormic shoots were collected were found in Ilex paraguariensis . The number of rooted mini-cuttings per ministump allowed an early selection of clones for the propagation of C. canjerana by mini-cutting (Burin et al., 2018). Variations among clones were observed for IBA application and serial mini-cutting technique in E. grandis (Wendling & Xavier, 2005). These differences among clones corroborate the existence of genetic variability for adventitious rooting competence, enabling the identification of clones with higher production of rooted mini-cuttings in order to increase the multiplication rate (Burin et al., 2018).
The swamp cork clones also showed differences in the increment of rooted mini-cuttings with time, which is very important for defining the ideal time for the removal of the mini-cuttings from the humidity chamber. The clone 15SM08 needed 50% more time in a humidity chamber than 15SM05, increasing the possibility of disease incidence and the demand for this specialized environment for producing the same number of plantlets. In addition to the greater competence for adventitious rooting, the clone 15SM05 also showed a higher rate of induction and formation of roots than 15SM08. Similar results were found in the progenies of Anadenanthera macrocarpa halfsiblings, indicating genetic differences in the ideal time of mini-cuttings to complete the rhizogenic process in the humidity chamber (Dias et al., 2012). Furthermore, in hybrid clones of E. benthamii × E. dunnii, the ideal time for minicutting rooting in the humidity chamber varied between 35 and 42 days, depending on the evaluated clone (Brondani et al., 2012).
The results of these experiments show that minicuttings of swamp cork clones showed high rooting competence, with mean rooting percentage above 80% even without IBA application, and resulted in plantlets with high morphological quality. Vermiculite and its combination with commercial substrate resulted in the highest percentage of rooting, and number and length of roots and shoots. The clones showed differences regarding their adventitious rooting competence and the increment in rooted mini-cuttings with time in the humidity chamber. It probable happened because adventitious root formation is strongly controlled by a developmental program involving cell specification and reprogramming, initiation, and cell division leading to primordia formation, and emergence, and root outgrowth (Lakehal & Bellini, 2019). Therefore, for vegetative propagation by mini-cuttings, both percentage of rooting and spare time for rooting are very important traits, and strategies should be designed for the selection of clones with high rooting competence and rate of adventitious root formation. The swamp cork clone 15SM05 combines high rooting competence, rate of adventitious root formation, and vigorous growth of the plantlets, and is therefore most suitable for the mass production of plantlets.

CONCLUSIONS
Mini-cuttings of Erythrina crista-galli clones showed high rooting competence even without the application of IBA. Vermiculite and its combination with commercial substrate resulted in the highest percentage of rooting, number and length of roots and shoots. The clones showed differences regarding the adventitious rooting competence and the increase of rooted mini-cuttings over time. Clone 15SM05 combines high rooting competence, adventitious root formation rate and vigorous plantlets growth.

ACKNOWLEDGMENTS
To the Coordination for the Improvement of Higher Education Personnel (CAPES) and the National Council for Scientific and Technological Development (CNPq) for scholarship grants. The authors declare no conflict of interests.