Plant Growth Regulators on ‘Letícia’ Plum Fruit Set, Yield Performance and Fruit Quality Parameters in Southern Brazil

Abstract

Plant growth regulators (PGRs) such as aminoethoxyvinylglycine (AVG), 1-methylcyclopropene (1-MCP), and thidiazuron (TDZ) are widely used to improve fruit set and quality in stone fruits. This study evaluated the effects of these PGRs on fruit set, yield performance, and fruit quality parameters of the Japanese plum cultivar ‘Leticia’ under the edaphoclimatic conditions of the highland region of southern Brazil during the 2021/22 and 2022/23 growing seasons. The treatments (AVG, MCP, and TDZ) were applied in full bloom in a randomized complete block design with four replications, and the data from both seasons were analyzed by principal component analysis (PCA). All PGRs significantly affected fruit set, yield performance, and fruit quality parameters. The strongest associations were found with 182 mg L⁻¹ TDZ for fruit set, and with 62.5 mg L⁻¹ and 125 mg L⁻¹ AVG, and 21.43 mg L⁻¹ 1-MCP for yield performance-related trails. Applications of 125 mg L⁻¹ AVG, 21.43 mg L⁻¹ 1-MCP, and 182 mg L⁻¹ TDZ produced fruits with larger diameters and higher fresh weights. The PCA results indicated that TDZ at 182 mg L⁻¹ was closely associated with fruit set and yield performance, suggesting a strong multivariate relationship among these parameters and demonstrating its potential to enhance the productivity of ‘Leticia’ plum under the edaphoclimatic conditions of southern Brazil during the 2021/2022 and 2022/2023 growing seasons.

1. Introduction

The plum (Prunus domestica L.) is the second most economically important stone fruit worldwide, with approximately 2.5 million hectares cultivated and an annual production of 12.3 million tons [1]. In Brazil, commercial production is dominated by P. salicina Lindl., with cultivars such as ‘Letícia’ and ‘Fortune’ being the most widely grown. The total cultivated area is around 3500 hectares, with production exceeding 50,000 tons in the 2021/2022 season, mostly concentrated in the southern states of Santa Catarina and Rio Grande do Sul [2]. Despite growing interest, domestic production still fails to meet market demand, and imports remain essential [3]. This situation is largely due to a shortage in the number of available cultivars adapted to local conditions and to gametophytic self-incompatibility, which prevents self-pollination and reduces fruit set. Effective fruit set depends on several factors, including cross-pollination, the presence of compatible pollinizers, active pollinating agents, and synchronized blooming between cultivars [4]. In Southern Brazil, adverse weather conditions such as excessive rainfall, low temperatures, or cloudy periods during bloom can negatively impact pollination efficiency, pollen viability, and fruit set, ultimately reducing yield and fruit quality [5,6].

In addition to cross-pollination requirements and floral synchronization, effective fruit set depends on a balanced hormonal regulation, particularly between abscission-promoting hormones such as ethylene and abscisic acid, and those that inhibit abscission, including auxins, cytokinins, gibberellins, and polyamines [7]. Modulation of this hormonal balance through the application of plant growth regulators (PGRs) has proven to be a promising strategy for improving fruit retention, increasing yield, and potentially enhancing fruit quality [8,9,10].

Among the PGRs with commercial potential are aminoethoxyvinylglycine (AVG), 1-methylcyclopropene (1-MCP), and thidiazuron (TDZ). AVG inhibits ethylene biosynthesis by blocking 1-aminocyclopropane-1-carboxylic acid (ACC) synthase, thereby reducing flower and fruit abscission and senescence [5]. 1-MCP acts by binding to ethylene receptors and inhibiting its perception, thus delaying fruit drop and senescence [8]. TDZ, a synthetic cytokinin, has shown positive effects on fruit set and yield in mango [10], apple [11], and pear [12,13].

Despite the individual use of AVG, 1-MCP, and TDZ being previously reported in other temperate fruit crops such as apple, pear, and sweet cherry, their combined comparative evaluation in Japanese plum has not been conducted under subtropical Brazilian conditions. Previous studies have mainly focused on single PGR applications or on postharvest fruit quality effects rather than preharvest fruit set regulation [6,10,12]. The current research provides a novel and integrative approach by simultaneously evaluating these three PGRs, applied under field conditions in southern Brazil, where fluctuating winter chill and erratic bloom weather represent key challenges for fruit set and productivity. Therefore, this study aimed to evaluate the effects of AVG, 1-MCP, and TDZ on fruit set, yield performance, and fruit quality parameters in the Japanese plum cultivar ‘Letícia’ under the edaphoclimatic conditions of southern Brazil during the 2021/2022 and 2022/2023 growing seasons.

2. Materials and Methods

2.1. Experimental Area

The experiments were conducted in a commercial orchard of ‘Letícia’ plum (Prunus salicina Lindl.) with the pollinizer cultivar ‘SM6’, established in 2005 in the municipality of Lages, Santa Catarina, Brazil (27°46′2.85″ S, 50°10′55.13″ W), at an altitude of 880 m above sea level, during the 2021/2022 and 2022/2023 growing seasons. The climate of the region is classified as humid mesothermal (Cfb) according to the Köppen classification [14]. Daily data on maximum and minimum temperatures, rainfall, and solar radiation were recorded by the A880 automatic weather station of the National Institute of Meteorology [15]. The soil is classified as a typical dystrophic Bruno Oxisol [16], with a high clay content (430 g kg⁻¹) and organic matter (95 g kg⁻¹). The orchard was trained in a “Y” system with a planting density of 2000 trees per hectare, spaced 1.0 m × 5.0 m between trees and rows, respectively.

2.2. Experimental Protocol

The treatments consisted of the application of three plant growth regulators in full bloom (BBCH 65), using the commercial formulation concentrations: aminoethoxyvinylglycine (AVG, 150 g kg⁻¹), thidiazuron + diuron (TDZ, 120 g L⁻¹ of thidiazuron and 60 g L⁻¹ of diuron), and 1-methylcyclopropene (1-MCP, 17.15 g L⁻¹). The choice of PGR products and concentrations was based on previous research in temperate fruit species, preliminary observations in local orchards, and commercially recommended doses for field application in southern Brazil. The tested doses, based on active ingredient concentration, were as follows: control (no application); AVG at 31.3, 62.5, 93.6, and 125.0 mg L⁻¹; 1-MCP at 21.4, 42.9, and 64.3 mg L⁻¹; and TDZ at 182.0 mg L⁻¹. All applications were performed using an electric backpack sprayer with constant pressure of 20 bar, delivering a spray volume of 1000 L ha⁻¹.

2.3. Fruit Set

Fruit set was assessed by counting the number of flowers in full bloom and the number of fruits 30 days after treatments application, on four branches per plant. At the pre-harvest stage (physiological maturity), the following yield performance and fruit quality parameters were assessed using a sample of 20 fruits from each experimental plot.

2.4. Yield Performance

(a)

Yield and productivity were determined by counting the total number of fruits per plant and calculating the weight in kilograms (kg) and in tons per hectare (t ha⁻¹), respectively, by multiplying the yield per plant by the number of plants per hectare;

(b)

Fresh mass (FM), expressed in grams (g) per fruit, was determined by weighing a sample of 80 fruits from each treatment;

(c)

Equatorial diameter and height (in centimeters, cm) were measured using a graduated ruler, by aligning the fruits side by side.

2.5. Fruit Quality Parameters

(a)

Pulp firmness (PF) was measured in Newtons (N) using a texture analyzer equipped with a 9 mm diameter probe. A portion of the fruit epidermis was removed from two opposite sides of the equatorial region using a ‘peeler’ to allow proper assessment of PF.

(b)

Soluble solids content (SSC), expressed in °Brix, was determined using a digital refractometer (model ITREFD-45). The readings were taken from juice extracted from median slices of fruits collected from each experimental plot.

(c)

Fruit color was measured using a Minolta colorimeter (model CR 400), and expressed in terms of lightness (L), chroma (C), and hue angle (Hue).

2.6. Experimental Design and Data Analysis

The experiments followed a randomized complete block design with four replicates, each plot consisting of seven ‘Letícia’ plum trees of uniform vigor and canopy structure, totaling twenty-eight trees per treatment. Statistical analysis was based on the mean values obtained across the two growing seasons. All plots were located within the same orchard block, which presented homogeneous soil characteristics, topography, and management history. Standard orchard practices, including fertilization, pruning, irrigation, and pest and disease control, were uniformly applied across all treatments. The growing season was treated as a fixed effect to describe seasonal influence on treatment performance. The normality of the variables was assessed through ‘kurtosis’ and ‘skewness’. The optimal number of clusters was determined using the ‘silhouette’ coefficient, and cluster formation was carried out using the non-hierarchical k-means algorithm. Subsequently, a multivariate approach was applied using principal component analysis (PCA). PCA was performed to reduce the dimensionality of the measured variables and to identify the principal components that accounted for the greatest proportion of variance among treatments, particularly for fruit set, yield performance, and fruit quality attributes. K-means clustering was applied to group treatments exhibiting similar multivariate response patterns based on the PCA scores. The optimal number of clusters was determined using the silhouette coefficient, ensuring that treatments within each cluster shared similar response characteristics. These multivariate approaches enabled the identification of associations between plant growth regulators and physiological or yield performance that might not be evident from univariate analysis. Statistical analyses were conducted in R software, version 4.3.1 (R Core Team, R Foundation for Statistical Computing, Vienna, Austria), using the statistical environment at RStudio version 4.3.1 (Posit Software, PBC; Boston, MA, USA) [17].

3. Results

According to meteorological data collected in the municipality of Lages, SC, from January 2021 to September 2023 (Figure 1), notable climatic differences were observed between the 2021–2022 and 2022–2023 growing seasons, particularly during the critical stages of “Leticia” plum development. During the winter dormancy period (April to August), the 2022 growing season showed lower average temperatures and a greater number of days with minimum temperatures below 7.2 °C compared to 2021 growing season. As illustrated in Figure 2, both chilling hours and chill units were higher in 2022 growing season, indicating a colder and more prolonged chilling period.

The beginning of sprouting, observed in late August and early September, occurred under different thermal conditions. In 2021 growing season, sprouting started under slightly warmer temperatures, with daily means around 16 °C. In 2022 growing season, cooler temperatures persisted during this phase, with daily means below 14 °C, which may have delayed or influenced the uniformity of bud break.

The flowering phase, which occurred predominantly in early to mid-September, followed these initial sprouting conditions. In 2022, flowering took place under more stable and moderate temperatures (12 to 18 °C), while in 2021, temperatures during this stage were often higher, exceeding 20 °C during the day.

During fruit set and initial development (October to November), the 2021–2022 growing season experienced irregular rainfall, including peaks in October. In contrast, the 2022–2023 growing season had more balanced precipitation and moderate temperatures, contributing to more stable conditions during early fruit development.

The fruit growth and maturation period (December to February) was hotter in 2021–2022 growing season, with maximum temperatures often above 30 °C, especially in December and January. In 2022–2023 growing season, temperatures during this phase were milder, with fewer peaks of high temperature and more even rainfall distribution. By the time of physiological and commercial maturity, reached in February in both growing seasons, average temperatures were slightly lower in 2022–2023 growing season than in the previous season.

The Principal Component Analysis (PCA) biplots (Figure 3 and Figure 4) illustrate the distribution of treatments along the first two principal components (PC1 and PC2), which together explained 68.4% of the total variance for yield performance and 21.6% for fruit quality traits. Treatments that are positioned close to each other in the plot exhibited similar multivariate behavior, whereas those separated along the principal axes displayed distinct physiological or productive responses.

The PCA of Figure 3 and Figure 4 performed using fruit set, yield performance and fruit quality parameters data for the 2021/22 and 2022/23 growing seasons showed clear separation among the treatments with different plant growth regulators (PGRs).

Table 1. Yield performance from principal component analysis of ‘Letícia’ plum cultivar under application of plant growth regulators, during the 2021/22 and 2022/23 growing seasons.

Treatments	Fruit Set 30 DAA *	Fruits per Plant	Production	Yield
	(%)	n plant−1	kg plant−1	t ha−1
Control	3.82 ± 0.31	291 ± 23	21.37 ± 1.71	42.73 ± 3
31.3 mg L−1 (AVG)	3.81 ± 0.30	277 ± 22	16.96 ± 1.36	33.92 ± 2
62.5 mg L−1 (AVG)	3.61 ± 0.29	255 ± 20	18.79 ± 1.50	37.57 ± 3
93.6 mg L−1 (AVG)	3.19 ± 0.26	203 ± 16	15.29 ± 1.22	30.59 ± 2
125.0 mg L−1 (AVG)	4.07 ± 0.33	270 ± 22	20.41 ± 1.63	40.83 ± 3
64.3 mg L−1 (1-MCP)	3.42 ± 0.27	256 ± 20	19.46 ± 1.56	38.91 ± 3
42.9 mg L−1 (1-MCP)	4.76 ± 0.38	235 ± 19	17.34 ± 1.39	34.68 ± 2
21.4 mg L−1 (1-MCP)	3.48 ± 0.28	214 ± 17	17.11 ± 1.37	34.21 ± 2
182.0 mg L−1 (TDZ)	4.98 ± 0.40	251 ± 20	20.50 ± 1.65	41.01 ± 3

* Days After Application; AVG: aminoethoxyvinylglycine; TDZ: thidiazuron; 1-MCP: 1-methylcyclopropene; Values represent mean ± standard deviation (n = 4).

,

Table 2. Fruit parameters from principal component analysis of ‘Letícia’ plum cultivar under application of plant growth regulators, during the 2021/22 and 2022/23 growing seasons.

Treatments	Fresh Fruit Mass	Fruit Height	Fruit Diameter
	g	cm	cm
Control	73.22 ± 5.85	4.72 ± 0.09	4.82 ± 0.10
31.3 mg L−1 (AVG)	75.74 ± 6.06	4.69 ± 0.09	4.79 ± 0.10
62.5 mg L−1 (AVG)	73.87 ± 5.91	4.68 ± 0.09	4.73 ± 0.09
93.6 mg L−1 (AVG)	76.00 ± 6.08	4.76 ± 0.09	4.31 ± 0.08
125.0 mg L−1 (AVG)	77.35 ± 6.19	4.76 ± 0.09	4.87 ± 0.10
64.3 mg L−1 (1-MCP)	76.64 ± 6.13	4.69 ± 0.09	4.75 ± 0.08
42.9 mg L−1 (1-MCP)	75.09 ± 6.01	4.77 ± 0.10	4.81 ± 0.10
21.4 mg L−1 (1-MCP)	81.66 ± 6.53	4.91 ± 0.10	4.96 ± 0.10
182.0 mg L−1 (TDZ)	80.89 ± 6.47	4.85 ± 0.10	4.91 ± 0.10

Values represent mean ± standard deviation (n = 4).

and

Table 3. Fruit quality parameters from principal component analysis of ‘Letícia’ plum fruits under application of plant growth regulators during the 2021/22 and 2022/23 growing seasons.

Treatments	Colorimetry			Flesh Firmness	Soluble Solids
	EL	EC	EH	N	°Brix
Control	43.92 ± 3.07	29.05 ± 2.09	36.28 ± 2.90	11.66 ± 0.93	10.94 ± 0.44
31.3 mg L−1 (AVG)	44.95 ± 3.16	28.75 ± 2.06	35.23 ± 2.82	16.11 ± 1.29	10.85 ± 0.43
62.5 mg L−1 (AVG)	45.05 ± 3.17	29.46 ± 2.12	36.04 ± 2.88	21.98 ± 1.76	10.35 ± 0.41
93.6 mg L−1 (AVG)	46.84 ± 3.35	28.85 ± 2.07	46.20 ± 3.70	23.78 ± 1.90	10.90 ± 0.44
125.0 mg L−1 (AVG)	42.82 ± 3.07	28.60 ± 2.05	36.02 ± 2.88	16.16 ± 1.29	10.35 ± 0.41
64.3 mg L−1 (1-MCP)	44.47 ± 3.20	29.86 ± 2.16	35.75 ± 2.86	14.59 ± 1.17	10.75 ± 0.43
42.9 mg L−1 (1-MCP)	44.71 ± 3.22	28.36 ± 2.03	35.05 ± 2.80	21.74 ± 1.74	10.60 ± 0.42
21.4 mg L−1 (1-MCP)	43.17 ± 3.09	29.43 ± 2.12	32.73 ± 2.62	16.07 ± 1.28	10.99 ± 0.44
182.0 mg L−1 (TDZ)	44.81 ± 3.22	28.91 ± 2.07	37.67 ± 3.01	19.95 ± 1.60	10.46 ± 0.42

Values represent mean ± standard deviation (n = 4); Epidermis lightness (EL); Epidermis Chroma (EC); Epidermis hue Angle (EH).

present numerical values derived from PCA analyses. These tables are complementary to the PCA figures and are provided to facilitate a clearer understanding of the data structure and variable contributions underlying the principal components.

Treatments with higher yield, including AVG at 62.5 mg L⁻¹, AVG at 125.0 mg L⁻¹, 1-MCP at 64.31 mg L⁻¹, and the untreated control, clustered together, reflecting similar levels of productivity regardless of fruit set percentage. Treatments such as AVG at 31.25 mg L⁻¹, 1-MCP at 21.43 mg L⁻¹, and AVG at 93.60 mg L⁻¹ showed intermediate behavior for both variables (Figure 3 and Table 1).

In contrast, TDZ at 182 mg L⁻¹ showed the highest fruit set percentage (4.98%), which was different from most treatments, while also maintaining a high yield (41.01 t ha⁻¹), comparable to the top-yielding treatments. 1-MCP at 42.90 mg L⁻¹ presented similar fruit set values to TDZ, though its productivity was lower. Meanwhile, treatments such as AVG 31.25 mg L⁻¹, 1-MCP 21.43 mg L⁻¹, and AVG 93.60 mg L⁻¹ showed intermediate values for both fruit set and productivity (Figure 3 and Table 1).

As shown in Figure 4 and detailed in Table 2 and Table 3, fruit quality parameters varied among treatments. TDZ at 182 mg L⁻¹ resulted in significantly larger fruits, with higher average fruit mass and greater height and diameter, confirming its effectiveness in maintaining fruit size even under high fruit set conditions. Among the treatments with similar productivity to TDZ, only 1-MCP at 21.43 mg L⁻¹ produced fruits of comparable size and weight. The remaining high-yielding treatments, including AVG at 62.5 mg L⁻¹, AVG at 125.0 mg L⁻¹, and the control, resulted in fruits with smaller dimensions and lower individual weight (Table 2).

Regarding fruit color attributes, most treatments showed similar values for epidermal lightness (EL), chroma (EC), and hue angle (EH), indicating no major differences in visual appearance. An exception was AVG at 93.60 mg L⁻¹, which produced fruits with slightly lower red intensity, as reflected by a higher hue angle. No significant differences were observed among treatments in terms of pulp firmness (PF) or soluble solids (SS; °Brix), with values remaining within commercially acceptable ranges for both growing seasons (Table 3).

4. Discussion

Environmental factors, particularly temperature, precipitation, and chilling accumulation, played a key role in the variation in fruit set and yield performance between the two growing seasons. The highland region of southern Brazil is characterized by a humid subtropical mesothermal climate (Cfb), with annual rainfall of around 1400 mm and an average temperature of approximately 15.6 °C; however, interannual variability is frequent and may significantly impact floral development and fertilization [3].

The highland region of southern Brazil experienced different chilling accumulation patterns over the two growing seasons. In 2021, chilling accumulation was higher, contributing to more uniform budbreak and flowering [5], whereas in 2022, warmer winter temperatures and less consistent chilling likely caused irregular sprouting and reduced flowering uniformity [11,18]. These climatic differences likely impacted the hormonal responsiveness and fruit set outcomes across treatments.

Unfavorable weather conditions during bloom, such as high rainfall or low solar radiation, can interfere with pollinator activity and pollen tube growth, thereby limiting fertilization success [4]. This is especially relevant in self-incompatible cultivars like ‘Letícia’, which rely on cross-pollination. Reductions in effective pollination period or pollen viability have been linked to lower fruit set in previous studies [6].

These climatic and pollination constraints highlight the importance of hormonal balance and PGR-mediated regulation as complementary tools to improve fruit set under adverse environmental conditions.

The multivariate analyses using Principal Component Analysis (PCA) and k-means clustering provided an integrated understanding of the combined effects of AVG, 1-MCP, and TDZ on fruit set, yield performance, and fruit quality parameters. PCA was applied to reduce the dimensionality of the dataset and to identify the principal components that explained most of the variance among treatments, while k-means clustering grouped treatments with similar multivariate response patterns. The first two principal components (PC1 and PC2) explained 68.2% of the total variance, showing that TDZ was more associated with yield performance traits, whereas AVG and 1-MCP were linked to higher fruit retention and improved quality parameters. Treatments that clustered closely together exhibited similar physiological behavior, confirming the complementary effects of these PGRs under the edaphoclimatic conditions of southern Brazil.

The application of PGRs significantly affected fruit set, yield performance, and fruit quality, and their effects were strongly influenced by concentration and seasonal climatic conditions. TDZ at 182 mg L⁻¹ notably increased fruit set percentage compared to other treatments and seasons. This improvement is likely related to the cytokinin-like activity of TDZ, which stimulates floral bud retention and parthenocarpic development [10,19], while inhibiting cytokinin degradation and reducing flower abscission [8]. Similar results were observed in apples and plums [5,20]. Moreover, this response can be physiologically explained by the interaction between cytokinins, auxins, and ethylene, which collectively regulate flower abscission and fruit retention. Elevated cytokinin activity favors cell division and delays senescence, maintaining auxin transport from developing fruits to the pedicel and thereby preventing activation of the abscission zone [21,22]. Conversely, ethylene promotes cell wall degradation and organ separation; thus, treatments such as TDZ, AVG, and 1-MCP act synergistically to suppress ethylene biosynthesis or perception, strengthening fruit retention under environmental stress [23,24,25].

1-MCP and AVG also contributed to fruit set and yield performance, acting mainly through ethylene suppression. AVG inhibits the activity of ACC synthase, a key enzyme in ethylene biosynthesis [21], delaying senescence and reducing early fruit drop. In the present study, AVG at 62.5 and 125.0 mg L⁻¹ resulted in lower fruit set percentage than TDZ, but produced the highest yields among treatments. This effect is likely related to improved fruit retention and more efficient resource partitioning resulting from ethylene inhibition, since AVG blocks ACC synthase, a key enzyme in ethylene biosynthesis, thereby delaying senescence and reducing early fruit drop [21]. Similar findings were reported in ‘Rocha’ pears treated with AVG [12]. TDZ not only enhanced fruit set percentage but also maintained fruit size, as evidenced by PCA results, which showed a strong association of this treatment with productive traits such as fruit set and fruit weight (Figure 3 and Figure 4). This multivariate relationship indicates an effective balance between reproductive load and assimilate supply [6]. This contrasts with the typical trade-off reported for other fruit crops, in which higher fruit set commonly leads to smaller fruit size due to competition for assimilates [22,23].

Although not tested here, our findings align with previous reports showing that other growth regulators such as 1-naphthaleneacetic acid (NAA), gibberellic acid (GA₃), and 6-benzylaminopurine (6-BA) reduce flower and fruit abscission and improve quality traits like pulp firmness and diameter [24,25]. These observations support the conclusion that hormonal regulation can directly influence both physiological and commercial traits.

Regarding fruit quality, our findings indicated that soluble solids content and pulp firmness were not significantly affected by PGR treatments, consistent with previous reports suggesting stronger environmental and genetic influences [23,26]. The treatment AVG 93.6 mg L⁻¹ also resulted in brighter skin coloration, possibly due to reduced anthocyanin accumulation from ethylene inhibition.

Overall, our results demonstrate that the interaction between PGR type, dose, and seasonal climatic conditions was critical for treatment success. While TDZ and AVG proved consistent across both growing seasons, 1-MCP performance was more variable, highlighting the importance of proper timing and environmental compatibility. These findings reinforce the need for adaptive management of PGRs, considering both cultivar and climate, particularly in regions with high interannual variability. The combined use of TDZ and AVG with proper timing may help stabilize fruit set and yield in seasons with unfavorable weather conditions, providing practical guidance for growers in PGR management. By improving fruit retention and maintaining fruit size, these treatments can contribute to higher marketable yields and better economic returns, offering a valuable tool for optimizing plum production under the edaphoclimatic conditions of the highland regions of southern Brazil.

5. Conclusions

The use of plant growth regulators positively influences effective fruit set and yield performance in ‘Letícia’ plum trees under specific regional conditions of this study. The application of TDZ, especially at 182 mg L⁻¹, promotes greater effective fruit set and increases fruit fresh weight, while TDZ and AVG also contribute to an increase in fruit transverse diameter, showing potential as complementary tool to improve fruit production. No significant differences were observed among treatments for pulp firmness and soluble solids content under edaphoclimatic conditions of southern Brazil.