Selection of Habanero Pepper F1 Hybrids ( Capsicum chinense Jacq.) at the Yucatan Peninsula, Mexico with a High Potential for Di ﬀ erent Markets

: This study evaluated 29 F1 lines and the 11 genotypes of habanero peppers used in the crossbreeding program developed by the Scientiﬁc Research Center of Yucatan, M é xico. A randomized complete block design with four repetitions was used. Eight plants of each of the genotypes were studied per block. A total of 22 qualitative and 18 quantitative descriptors established in the manuals of the International Plant Genetic Resources Institute (IPGRI) and the National Service for Seed Inspection and Certiﬁcation (SNICS) was used. The multiple correspondence analysis of the qualitative traits explained 38.2% of the total variability. The trait that contributed the most to the qualitative variability identiﬁed was the presence of anthocyanins in the node. Principal component analysis showed that the ﬁrst two axes explained 85.1% of the total variability and that capsaicin content and fruit pericarp thickness were the major contributors to the variation recorded. Based on these results, four F1 hybrids of habanero pepper were selected because of their promising traits for the di ﬀ erent markets, i.e., high productive potential and / or high pungency. These traits are described in the section on Results. ,


Introduction
All peppers belong to the Capsicum genus, which includes around 42 species with broad diversity in the shape, color, and size of the fruit, and in the sensory attributes such as taste, aroma and hotness [1], which make them stand out from the rest of plants used as spices. Only five of these species have been domesticated and cultivated: C. annuum L., C. frutescens L., C. baccatum L., C. pubescens, and C. chinense Jacq. [2]. Besides being used fresh and in a broad variety of dishes in international cuisine, chili peppers are a raw material of many industries; among them the food industry (powder, pastas, sauces), the military (self-defense sprays, projectiles), pharmaceutical (creams, sprays, patches, ointments), and chemical (protective coatings for electrical wires and as an additive in ship paints) [3][4][5]. Habanero pepper is widely cultivated in the Yucatan peninsula, a region in Mexico acknowledged as the center of genetic diversity of the species (Capsicum chinense Jacq.). Its aroma, taste and hotness, set it is apart from the habanero peppers grown in any other part of the world. In 2010, these attributes earned it the denomination of origin "Habanero pepper of the Yucatan peninsula" [6]. Habanero pepper pungency is classified between 100,000 and 300,000 Scoville Heat Units (SHU) [7]. However, the native varieties of southeastern Mexico are considered among the hottest in the world, with pungency ranging between 145,950 and 892,719 SHU [8], probably due to the edaphoclimatic conditions of the region.
Recently, Muñoz-Ramírez et al. [9] evaluated the pungency of 'Bhut Jolokia', 'Trinidad Moruga Scorpion' and 'Carolina Reaper' peppers cultivated at Yucatan, Mexico, which have been published in the Guinness Book of World Records at different periods of time as the hottest peppers in the world [10,11], and the three varieties significantly surpassed the pungency for which they were originally acknowledged. The 'Carolina Reaper', grown in Yucatán and regarded as the hottest variety in the world with 2,200,000 SHU, reached 3,006,330 SHU. This allows the inference that in the region of the Yucatan peninsula some of the attributes of habanero pepper, particularly hotness, may be exacerbated. Despite the economic, social and cultural importance of the crop, no improved varieties of habanero peppers had been developed until recently to increase fruit yield and quality while preserving the traits that distinguish Yucatan's habanero pepper from those grown in other parts of the world.
Paradoxically, the yield of the native varieties of habanero pepper are low, fruits are small and heterogeneous, and fruits from the same plant may present different numbers of locules (2)(3)(4). For the last 15 years, special attention has been paid to the genetic improvement and technological development of the habanero pepper in southeastern México. Pungency is one of the most important traits considered in the improvement efforts that have been conducted. Canto-Flick et al. [8] worked with a collection of native habanero peppers and determined the capsaicinoids content of 18 accessions collected at the Yucatan peninsula. Whole fruit, placenta, and pericarp total capsaicinoids contents were determined by high-performance liquid chromatography (HPLC). The 18 accessions of habanero peppers assessed had a great variety of colors, shapes and sizes of the fruits. The results showed that 83.3% of the collection exceeded the pungency levels reported for habanero peppers [11] from other regions of the world. Interestingly, 33% of the accessions surpassed 500,000 SHU and 44.4% recorded above 600,000 SHU. A pungency level of 892,719 SHU was recorded in the whole fruit of accession NP1EG (yellow habanero), and the pericarp of accession NP3EC (orange habanero) reached 1,382,889 SHU. The pungency levels of both cultivars (NP1EG and NP3EC) were like those of 'Bhut Jolokia', but their shape and color were noticeably different, sharing the physical traits of the habanero peppers typical of the region.
In 2018, Santana-Buzzy et al. [12] presented 'Mayan Kisin', a high-yielding, red-fruited habanero hot pepper variety (C. chinense Jacq.). It is characterized by its high performance, and its bright red and very spicy fruits. 'Mayan Kisin' was the result of four selection cycles carried out on landraces collected in the region of Valladolid, in the state of Yucatan, Mexico. A study by Peña-Yam et al. [13] estimated the genetic parameters of seven agronomic characteristics for 11 habanero pepper genotypes. The aim of the study was to select potential progenitors to produce F1 hybrids. They found high values of the phenotypic coefficient of variation (PCV) of capsaicin content (CC). Heritability (h2) was high (0.98) for yield per plant (YP) and CC (0.93). The principal component analysis (PCA) revealed that the first three components explained 94.02% of the total variation; hence genotypes with high yield and high fruit weight (FW) were obtained (AKN-08, ASBC-09). The genotypes with a greater content of capsaicin were MBI-11and RES-05. The fruit of RNJ-04 had the greatest length. The greatest number of fruits per plant belonged to genotypes NBA-06, RKI-01, RHC-02, RHN-03, NKA-07, and MSB-12. Thus, the studied genotypes were found to be an excellent source of genetic material for habanero pepper improvement programs. Knowledge of the reproductive biology of a species is crucial to develop an efficient program of genetic improvement by hybridization. Peña-Yam et al. [14] conducted the first reported study on the floral biology of the habanero pepper. The aim of the study was to establish the best times to collect pollen, assess its viability and define the development state of the flower bud in which the anthers are closed and stigma is receptive. The goal was to use the acquired knowledge in the implementation of the crossbreeding program. The objective of the present study was to select, from among the diversity of habanero peppers in the region, F1 hybrids of high productivity, hotness, and potential for both the fresh and the industrial markets.

Materials and Methods
A total of 29 F1 lines obtained by a Top Cross design were evaluated; 11 improved varieties of habanero pepper that were used as progenitors in the breeding program were also assessed ( Figure 1). The improved varieties were obtained within the habanero pepper genetic improvement program developed in the Scientific Research Center of Yucatan (CICY) and are registered in the National Catalogue of Vegetables Varieties (CNVV).
Agriculture 2020, 10, x FOR PEER REVIEW 3 of 19 viability and define the development state of the flower bud in which the anthers are closed and stigma is receptive. The goal was to use the acquired knowledge in the implementation of the crossbreeding program. The objective of the present study was to select, from among the diversity of habanero peppers in the region, F1 hybrids of high productivity, hotness, and potential for both the fresh and the industrial markets.

Materials and Methods
A total of 29 F1 lines obtained by a Top Cross design were evaluated; 11 improved varieties of habanero pepper that were used as progenitors in the breeding program were also assessed ( Figure  1). The improved varieties were obtained within the habanero pepper genetic improvement program developed in the Scientific Research Center of Yucatan (CICY) and are registered in the National Catalogue of Vegetables Varieties (CNVV). A randomized complete blocks experimental design with four repetitions was used, assessing 10 fruits per replica. The plants grew under greenhouse conditions in the Scientific and Technological Park of Yucatan, located in Sierra Papacal, Merida, Yucatan at 21°07′20″ N 89°43′41″ O, at an altitude of 9 m above sea level, in the period between June 2018 and March 2019. Plants were transplanted into 1 m long bags (Pelemix, Guadalajara, México). The substrate was a thick and fine mix of coconut fiber in a 70:30 proportion; the distance between plants and rows was 20 cm and 160 cm, respectively. Total of eight plants of each genotype were randomly selected from each block and 22 qualitative and 18 quantitative traits were evaluated using the descriptors referenced in the IPGRI Manual [15] for Capsicum, and the SNICS Manual [16] for habanero pepper (C. chinense Jacq.) (Table 1). A randomized complete blocks experimental design with four repetitions was used, assessing 10 fruits per replica. The plants grew under greenhouse conditions in the Scientific and Technological Park of Yucatan, located in Sierra Papacal, Merida, Yucatan at 21 • 07 20" N 89 • 43 41" O, at an altitude of 9 m above sea level, in the period between June 2018 and March 2019. Plants were transplanted into 1 m long bags (Pelemix, Guadalajara, México). The substrate was a thick and fine mix of coconut fiber in a 70:30 proportion; the distance between plants and rows was 20 cm and 160 cm, respectively. Total of eight plants of each genotype were randomly selected from each block and 22 qualitative and 18 quantitative traits were evaluated using the descriptors referenced in the IPGRI Manual [15] for Capsicum, and the SNICS Manual [16] for habanero pepper (C. chinense Jacq.) ( Table 1).

Capsaicinoids Extraction and Quantification
The extraction and quantification of capsaicinoids followed the methodology reported by Collins et al. [17] with slight modifications. The separation and quantification of the capsaicinoids was conducted by HPLC (Agilent series 1200, Agilent Technologies, Santa Clara, CA, USA) equipped with an automatic injector and a fluorescence detector. The capsaicinoids were separated by a Zorbax Eclipse Plus C18 (Agilent Technologies, Santa Clara, CA, USA) column (4.6mmi.d. × 250 mm) at a temperature of 25 • C and an injection volume of 20 µL. The wavelengths used for the detection were 280 nm (excitation) and 338 nm (emission). The mobile phase was isocratic with 70% of solvent B (100% methanol) and 30% of solvent A (10% methanol solution v/v). The HPLC operating conditions to determine total capsaicinoids were flow rate of 1 mL/min and 15 min runtime. Standards of capsaicin and dihydrocapsaicin were used to develop a calibration curve based on the relationship of the maximum areas for the known concentrations of external standards. The stock solution was prepared in 100% (v/v) methanol and five different concentrations (20,60,100,200, and 300 ppm) were used to generate the calibration curve. The concentrations of the two major capsaicinoids (capsaicin and dihydrocapsaicin) were estimated using the calibration curve obtained. The concentration of capsaicinoids is reported in milligrams per gram of dry weight (DW).

Statistical Analyses
The data obtained from the qualitative traits were subjected to a multiple correspondence analysis (MCA), while data obtained from the quantitative traits were processed by variance analysis and the means were contrasted using the t Tests LSD (least significant difference) with p < 0.05 in order to determine the significance of the differences among variables. The associations between quantitative traits were determined using Pearson's correlation. The principal component analysis (PCA) used a traits correlation matrix. PCA results were plotted in a bidimensional plane. This was done with the IBM SPSS Statistics version 22 [18]. For the traits showing the greater contribution to variation in the PCA, a Euclidean distance matrix was calculated based on standardized data for cluster analysis using the unweighted pair group method with arithmetic mean (UPGMA) [19].

Variation of Qualitative Traits
The results of the analysis of the 22 qualitative traits assessed in 29 F1 Lines and 11 progenitors of habanero pepper (Table 2) showed that 82.5% of the genotypes presented three or more flowers per axil (NFA); the position of the flower (PF) was in the middle in most of the genotypes (55%). The anthocyanin color of the node (ACN) was absent in only 7.5% of assessed progenitors and F1 lines. The predominant shape of the stem (SS) was cylindrical (80%), and 80% of genotypes had an oval leaf shape (LS) while the shape of the rest was lanceolate. The predominant leaf position (LP) was horizontal (55%), and 95% had a ciliated leaf blade margin (LBM). Abbreviations are as in Table 1.
As to the color of the fruit, three different color shades were apparent in the immature fruit (IFC), with a predominance of green and dark green colors (45%). The genotypes of plants with ripe fruits of different colors (RFC)-yellow, orange, red and purple-were identified. Red was the most frequent color (82.5%) in the assessed genotypes. The shape of the fruit (FS) also exhibited great diversity. The most common was the triangular shape (45%), followed by the rectangular (27.5%); bell and square shapes were found with less frequency (10% and 17.5%, respectively). A total of 42.5% of the fruits had pointed apex (FSA). The smooth texture of the pericarp (FST) was most common (67.5%). Most of the fruits had three loculi (NL) with the density of the placenta (DPF) semi-distributed (80%). Data of the qualitative descriptors itemized for the F1 lines and the progenitors of habanero peppers are shown in Table S1.

Multiple Correspondence Factorial Analyses
The results of the multiple correspondence analyses (MCA) performed for the 22 qualitative traits evaluated in the 40 genotypes of habanero pepper identified two dimensions that explain 38.2% of the total variability (Table 3). The traits that contributed the most to axis 1 were anthocyanin coloration of nodes (ACN), the margin of the calyx of the fruit (MC), leaf position (LP), fruit shape (FS) and fruit undulation in cross-section (FUS). The traits contributing most to the second axis were leaf blade margin (LBM), the shape of the fruit apex (FSA) and fruit surface texture (FST). Figure 2 shows the distribution of the F1 lines and progenitors, as well as the modalities of the traits with a major contribution to axes 1 and 2. As can be appreciated, P33 and P34 move significantly away from the rest of the group. Both genotypes differ from the rest because their pericarps are markedly rugged (fruit surface texture), and their leaf blade margins undulated, in notable contrast with the rest of the genotypes. The underlined descriptors are those with the highest contribution in each dimension. Abbreviations are as in Table 1. Figure 2 shows the distribution of the F1 lines and progenitors, as well as the modalities of the traits with a major contribution to axes 1 and 2. As can be appreciated, P33 and P34 move significantly away from the rest of the group. Both genotypes differ from the rest because their pericarps are markedly rugged (fruit surface texture), and their leaf blade margins undulated, in notable contrast with the rest of the genotypes.

Variation of the Quantitative Traits
As can be seen in Table 4, most of the evaluated quantitative traits exhibit high variability, with the exception of the thickness of the pedicel of the fruit (FTP) that did not differ significantly (p < 0.05), so the trait was not considered in the principal component analysis (PCA). The results of the analysis of variance carried out for the evaluated traits in the plants and leaves are presented in Table 5. The germination values of F1 lines ranged between 60% and 100%, similar to those recorded for the progenitors (62-94%). Lines H8 and H22 showed 100% germination. The progenitor P34 had the greatest height (155.3 cm), while the lowest height (87.2 cm) belonged to Line H11. Variation was also found for the following traits: stem diameter (SD), with values ranging between 0.99 and 1.9 cm; length of blade (LLB) and width of blade (LWB) showed variations between 9.45 and 15 cm and between 5.2 and 8 cm, respectively; the length of the leaf petiole (LLP) variation values oscillated from 2.5 to 4.35 cm. Line H20 had the highest values for the traits of stem diameter (SD), length of blade (LLB) and width of blade (LWB) ( Table 5).  Germination percentage (G%), days to fruiting (DFR), average fruit length (FL), average fruit width (FWI), fruit weight (FW), number of seeds per fruit (NSF), fruit pericarp thickness (FWT), thickness the pedicel of the fruit (FTP) and the content of capsaicin (CC) are traits closely related with the yield and quality of the fruit [20]. The fruit traits variation for F1 lines and progenitors are shown in Table 6. According to the results, line H22 is characterized by greater fruit length (FL) (average value 5.52 cm), while P41 had the lowest length (3.41 cm). As for fruit width (FWI) and fruit weight (FW), the greater average values were recorded for P38 (4.32 cm and 16.82 g, respectively). P30 was the genotype with the greatest number of seeds per fruit (53.7). Genotype P41 stood out for having the thickest pedicel (FTP), while H7 had the thickest pericarp (FWT). Fruit yield per plant (FYP) showed a wide range of variation (846.78-99952.50 g·plant −1 ), and genotype H20 (F1 line) had the highest yield while progenitor P34 had the highest capsaicin content (CC) (147.11 mg·g −1 DW).

Correlation Analysis
The analysis of the quantitative traits' correlations ( Figure 3) indicated that the trait width of blade (LWB) presented a positive and significant correlation with the length of blade (LLB), while fruit width (FWI) correlated negatively with fruit length (FL). On the other hand, fruit width (FWI) had a positive correlation with fruit weight (FW) 0.642. Plant fruit yield (FYP) shows an association with stem diameter (SD), length of blade (LLB) and width of blade (LWB), which allows us to infer that there is a direct relation between the bearing of the plant and the productivity. Capsaicin content (CC) showed a positive and significant correlation (0.557) with plant height (PH), and a negative correlation with fruit pericarp thickness (FWT), with a value of −0.620, which indicates that fruits with a thin pericarp have a greater capsaicin concentration. The values of the correlations can be consulted in Table S2.

Correlation Analysis
The analysis of the quantitative traits' correlations ( Figure 3) indicated that the trait width of blade (LWB) presented a positive and significant correlation with the length of blade (LLB), while fruit width (FWI) correlated negatively with fruit length (FL). On the other hand, fruit width (FWI) had a positive correlation with fruit weight (FW) 0.642. Plant fruit yield (FYP) shows an association with stem diameter (SD), length of blade (LLB) and width of blade (LWB), which allows us to infer that there is a direct relation between the bearing of the plant and the productivity. Capsaicin content (CC) showed a positive and significant correlation (0.557) with plant height (PH), and a negative correlation with fruit pericarp thickness (FWT), with a value of −0.620, which indicates that fruits with a thin pericarp have a greater capsaicin concentration. The values of the correlations can be consulted in Table S2.

Principal Component Analysis
Results of the PCA (Table 7) showed that the first six axes represented 85.1% of the total variation, and that the first component explained most of the variation (%), ordering the genotypes by capsaicin contents (CC) and by fruit pericarp thickness (FWT). The second principal component was composed of the traits stem diameter (SD) and yield per plant (FYP), the third component was associated with fruit length (FL), the fourth with the number of seeds per fruits (NSF), the fifth component with the pedicel length (FPL), and the sixth component was related to stem length (SL).

Principal Component Analysis
Results of the PCA (Table 7) showed that the first six axes represented 85.1% of the total variation, and that the first component explained most of the variation (%), ordering the genotypes by capsaicin contents (CC) and by fruit pericarp thickness (FWT). The second principal component was composed of the traits stem diameter (SD) and yield per plant (FYP), the third component was associated with fruit length (FL), the fourth with the number of seeds per fruits (NSF), the fifth component with the pedicel length (FPL), and the sixth component was related to stem length (SL). Figure 4 shows the distribution of the evaluated traits for the first two components, according to their contribution to the explained total variation.  Figure 4 shows the distribution of the evaluated traits for the first two components, according to their contribution to the explained total variation.   Table 1.

Cluster Analysis
Five groups were identified in the dendrogram obtained from the cluster analysis ( Figure 5) based on the plant and fruit traits, which are described in Table 8. As can be seen, the F1 lines H20 and H26, which presented the highest values of stem diameter (SD) and fruit yield per plant (FYP), comprised group 1. Five Lines F1 (H9, H14, H15, H16, and H22) and two progenitors (P33 and P34) were located in group 2, characterized by presenting the highest mean values for stem length (SL), fruit pedicel length (FPL) and capsaicin content (CC). F1 lines (H7, H19 and H25) and two progenitors (P38 and P41) were allocated in group 3, with the highest mean values of fruit pericarp thickness (FWT) and of number of seeds per fruit (NSF). Four F1 lines (H10, H12, H18, and H28) and two progenitors (P37 and P40) with the lowest mean values for stem length (SL) comprised group 4. Group 5 was composed of most of the F1 lines (15) and five progenitors that exhibited the highest values of fruit length (FL).

Cluster Analysis
Five groups were identified in the dendrogram obtained from the cluster analysis ( Figure 5) based on the plant and fruit traits, which are described in Table 8. As can be seen, the F1 lines H20 and H26, which presented the highest values of stem diameter (SD) and fruit yield per plant (FYP), comprised group 1. Five Lines F1 (H9, H14, H15, H16, and H22) and two progenitors (P33 and P34) were located in group 2, characterized by presenting the highest mean values for stem length (SL), fruit pedicel length (FPL) and capsaicin content (CC). F1 lines (H7, H19 and H25) and two progenitors (P38 and P41) were allocated in group 3, with the highest mean values of fruit pericarp thickness (FWT) and of number of seeds per fruit (NSF). Four F1 lines (H10, H12, H18, and H28) and two progenitors (P37 and P40) with the lowest mean values for stem length (SL) comprised group 4. Group 5 was composed of most of the F1 lines (15) and five progenitors that exhibited the highest values of fruit length (FL).

Discussion
The variance analysis of traits with agricultural interest is critical for the development of programs aiming to obtain the varieties and hybrids with high productivity and/or that carry other important traits and attributes of agricultural interest [21]. The results of the analysis of the evaluated qualitative and quantitative traits of the 40 genotypes of habanero pepper revealed a broad genetic

Discussion
The variance analysis of traits with agricultural interest is critical for the development of programs aiming to obtain the varieties and hybrids with high productivity and/or that carry other important traits and attributes of agricultural interest [21]. The results of the analysis of the evaluated qualitative and quantitative traits of the 40 genotypes of habanero pepper revealed a broad genetic diversity, which notably fosters the work on the genetic improvement of the crop. As for the variation in the qualitative traits, the behavior of the ACN trait was interesting. Only 7.5% of the evaluated habanero pepper genotypes showed the absence of anthocyanins in the node of the plants, in contrast to what has been reported by Bozokalfa et al. [21] in another species of the genus (C. annuum L.). Their study found that node anthocyanins were absent in 47 of the 48 genotypes they assessed. This allows us to infer that, probably, this trait may be associated with the species.
Fonseca et al. [22] studied the genetic diversity of habanero peppers in the Amazon, and observed the predominance of some traits, such as the dark red color of the ripe fruit (42%), triangular-shaped fruits (42.1%) and rugged fruit surface (42.1%), in most of the evaluated genotypes. In our study, the MCA explained 38.2% of the total variability found, and identified some qualitative traits that contributed to clustering the genetic materials in a bidimensional plane. Castellón et al. [23], in a study on variations of native C. annuum L. accessions in Oaxaca, Mexico, found that MCA for the examined qualitative variables explained 88.5% of the variability, far higher than the findings in our study, which indicates greater variability in the habanero pepper. These same authors found statistically significant differences for all evaluated variables, except for the number of fruits per plant. In contrast, our study of 18 qualitative traits found that they all differed significantly (p < 0.05), except for the thickness of the pedicel of the fruit (FTP).
The width of the fruit (FWI) correlated negatively with fruit length (FL). On the other hand, the descriptor fruit width (FWI) had a positive correlation with fruit weight (FW), with a value of 0.642. Similar results were reported by Bharath et al. [24] and by Sharma et al. [25]. In our study, the trait pericarp thickness (FWT) had a positive association with fruit width (FWI) and fruit weight (FW). Moreira et al. [26] also worked with C. chinense and their results were similar to ours. Fruit yield per plant (FYP) shows an association with stem diameter (SD) and length of blade (LLB), indicating a direct influence between the bearing of the plant and the productivity. Likewise, capsaicin content (CC) had a significant positive correlation with plant height (PH), with a value of 0.557, and correlated negatively with fruit pericarp thickness (FWT), with a value of −0.620. Butcher et al. [27] identified an association between capsaicin and fruit wall thickness of −0.083, a lower value than the one we found in our study.
Both the PCA and the cluster analysis provide information showing the presence of an important genetic variability in the set of evaluated genotypes, which agrees with what has been reported by different authors [28,29]. Multivariate analyses have been used widely to evaluate the genetic variability of many crops [30,31]. Bozokalfa et al.'s results [21], including the PCA of the variation of 48 C. annuum accessions and lines, showed that the first six axes explained 54.29% of the variability, a value slightly lower than what we found in this study. Finally, the results of the cluster analysis obtained by Bianchi et al. [32] clustered Capsicum accessions in eight groups.
Our PCA showed that the traits capsaicin contents (CC), fruit pericarp thickness (FWT), stem diameter (SD) and fruit yield per plant (FYP) contributed the most to the accumulated variation. On the other hand, the cluster analysis generated five groups. Group 1 recorded the highest values for the following traits: Stem diameter (SD) and fruit yield per plant (FYP). Group 2 exhibited the highest values for stem length (SL), fruit pedicel length (FPL) and capsaicin content (CC), while the highest values for fruit pericarp thickness (FWT) and the number of seeds per fruit (NSF) were located in group 3. The lowest mean values for stem length (SL) were found in group 4 and group 5 had the highest value for fruit length (FL).

Conclusions
The Yucatan peninsula is a particularly privileged region of the world for habanero pepper genetic diversity. Based on the analysis of the data of the qualitative and quantitative traits, and taking into account the color, weight, shape and size of the fruit, as well as the yield and capsaicin content, we selected four F1 lines (H7, H14, H20, H27) ( Figure 6) because of their high yield potential and/or high pungency. The data presented in this study, associated with productivity and other important traits, as well as the broad morphologic description of the evaluated genotypes, allowed us to identify four F1 lines with high potential for different markets-on one hand, for the fresh fruit market, because of the high fruit quality, attractive colors, pretty shapes and high yields. On the other hand, the market is looking for a reliable and efficient source of high-quality capsaicin for different industries, particularly for the pharmaceutical industry. However, the most relevant result of our study is the fact of being able to obtain for the first time a habanero pepper hybrid (H14) with the greatest content of capsaicin to date (1,300,000-1,500,000 SHU). The hottest peppers in the world are not of the habanero type, and lack the attributes that give the habanero pepper from Southeast Mexico its high demand at international level. The germplasm of the habanero pepper that has supported this program of genetic improvement is stored in the Habanero Pepper Gene Bank of the Scientific Research Center of Yucatan (CICY), Yucatan, Mexico.
Agriculture 2020, 10, x FOR PEER REVIEW 16 of 19 market, because of the high fruit quality, attractive colors, pretty shapes and high yields. On the other hand, the market is looking for a reliable and efficient source of high-quality capsaicin for different industries, particularly for the pharmaceutical industry. However, the most relevant result of our study is the fact of being able to obtain for the first time a habanero pepper hybrid (H14) with the greatest content of capsaicin to date (1,300,000-1,500,000 SHU). The hottest peppers in the world are not of the habanero type, and lack the attributes that give the habanero pepper from Southeast Mexico its high demand at international level. The germplasm of the habanero pepper that has supported this program of genetic improvement is stored in the Habanero Pepper Gene Bank of the Scientific Research Center of Yucatan (CICY), Yucatan, Mexico.  Supplementary Materials: The following are available online at http://www.mdpi.com/2077-0472/10/10/478/s1, Table S1: Evaluated qualitative descriptors in F1 lines and progenitors of habanero pepper, Table S2 Funding: This work was supported by Consejo Nacional de Ciencia y Tecnología (CONACYT-México).

Conflicts of Interest:
The authors declare no conflict of interest.