Bioscience Journal | 2022 | vol. 38, e38004 | ISSN 1981-3163 1 Michelle Souza VILELA1 , José Ricardo PEIXOTO1 , Samara Dias Rocha RAMOS2 , Rosa Maria de Deus de SOUSA1 , Assussena Pereira de OLIVEIRA3 , Marcelo de Abreu Flores TOSCANO3 , Antônio Alves de OLIVEIRA JUNIOR3 1 Department of Agronomy and Veterinary Medicine, Federal University of Brasilia, Brasilia, Distrito Federal, Brazil. 2 Graduate Program in Agronomy, Federal University of Brasilia, Brasilia, Distrito Federal, Brazil. 3 Postgraduate Program in Agronomy, Federal University of Brasilia, Brasilia, Distrito Federal, Brazil. Corresponding author: Assussena Pereira de Oliveira Email: assussena.oliveira@hotmail.com How to cite: VILELA, M.S., et al. Agronomic assessment of 32 sour passionfruit genotypes in Federal District. Bioscience Journal. 2022, 38, e38004. https://doi.org/10.14393/BJ-v38n0a2022-54231 Abstract The production of passion fruit is important in Brazil. In order to contribute to the development of the most promising cultivars of passion fruit, this study aimed to evaluate the agronomic performance of 32 genotypes of passion fruit in Federal District of Brazil, and to estimate genetic parameters for use in breeding programs. Thirty-two genotypes were used in a randomized block design, with eight plants per plot and four replications. The experiment was conducted in field. Twenty-eight harvests were performed, and the variables analyzed were: productivity estimated, total number of fruits per hectare, average fruit weight and these characteristics following classification of fruits in five categories. The genotypes that presented the highest total yield estimated were MAR20 # 23, AR 01 and PLANTA 7. For industrial purposes, genotypes MAR 20 # 21 and BRS Gigante Amarelo were superior. For fresh consumption, the genotypes with the best performance were PLANT 7, AR 01 and MSC. Total productivity estimated and total number of fruits per hectare in the first-class classification showed high values of heritability and CVg/CVe ratio. These results indicate a favorable condition for selection. Keywords: Genetic parameters. Passiflora edulis Sims. Productivity. 1. Introduction In the national fruit culture, there are some fruits that launch Brazil to the position of great world producer. Passion fruit is one of the crops that contribute to this condition of Brazil as a world fruit producer, with an average productivity of 14t/ha in 2018 (IBGE 2018). The average productivity of passion fruit in recent years has varied from 12 to 15 tons per hectare, with potential for production of 30 to 35 tons per hectare (Silva et al. 2009). Elite genotypes, developed in research actions, can produce more than 50t/ha/year (Faleiro et al. 2011). Fruit production stands out in the Northeast, Southeast, and South regions of Brazil. Bahia is the main producer, with 160,902 tons in 15,660 hectares, followed by Ceará, with 147,458 tons produced in 6,862 hectares. In third place in national production is the state of Santa Catarina with 53,961 tons in 2,270 hectares (IBGE 2018). In the Southeast region, passion fruit is one of the eight most cultivated fruit species in the extensive system, being preceded only by the cultures of orange, banana, lemon, mango, mandarin orange, pineapple, and grape (Meletti 2011). AGRONOMIC ASSESSMENT OF 32 SOUR PASSIONFRUIT GENOTYPES IN FEDERAL DISTRICT https://orcid.org/0000-0002-0417-568X https://orcid.org/0000-0002-8885-2886 https://orcid.org/0000-0002-4250-7032 https://orcid.org/0000-0002-5983-4718 https://orcid.org/0000-0002-9279-1895 https://orcid.org/0000-0001-9094-6796 https://orcid.org/0000-0002-2605-3704 Bioscience Journal | 2022 | vol. 38, e38004 | https://doi.org/10.14393/BJ-v38n0a2022-54231 2 Agronomic assessment of 32 sour passionfruit genotypes in Federal District The productivity of the passion fruit culture is considered low. Many factors influence this characteristic, the cultivation of inappropriate varieties being one of them (Nogueira 2016). Others refer to the genetic characteristics of the plant, edaphic, environmental conditions, biotic agents and the action of man (Faleiro et al. 2011). According to Faleiro and Junqueira (2016), low productivity is one of the main problems of culture, emphasizing the need for research aimed at the development of improved varieties and the establishment of production technologies capable of providing increased productivity, possibility of increased survival of the culture and improvement of the quality of the fruits. It has been observed, in recent years, that there is a lack of genetic materials with high productivity, fruit quality and resistance to phytopathogens, mainly due to the lack of research work in the various areas of knowledge and especially with genetic improvement of passion fruit (Faleiro et al. 2018). According to Faleiro et al. (2011), the genetic improvement of passion fruit has several purposes depending on the product to be considered (fruits, leaves or seeds) and the region of cultivation. Increased productivity, fruit quality, resistance to diseases, nematodes and also an increase in the rate of fruit avenging are the main objectives of the improvement of the crop. In the open field, agronomic performance and resistance to phytopathogens require continuous work of genetic improvement, since there are few passion fruit cultivars available to Brazilian producers and their productivity is considered to be regular to low (Faleiro et al. 2011). Another problem faced by the crop is the small longevity of the crop. In several areas of sour passion fruit planted at the end of the last century, crops in full production with up to 7 to 8 years of age were observed. However, in these same areas, recently, crops have not exceeded two years of age, and in many cases, total death occurs at just one year of age. Productivity can be increased by reducing losses due to diseases, passion fruit culture has losses due to fungal, bacterial and virotic diseases. Phytosanitary control can be decisive in the production of passion fruit, but for the chemical control to be efficient, it is interesting that it is associated with a field with cultivars with a good genetic potential. Therefore, evaluating the performance of cultivars for the genetic improvement of passion fruit is important to ensure future production (Peruch et al. 2018). In Brazil, most breeding programs are related to fruit, both in terms of productivity and quality. In qualitative terms, it is considered that a fresh variety, developed for the market must have large and oval fruits, in order to achieve good commercial classification, besides having a good appearance, being resistant to transport and loss of quality during storage and commercialization (Faleiro et al. 2018). In this sense, the selection of passion fruit cultivars with good productivity and fruit quality is essential for the development of culture in Brazil. Thus, the work aimed to evaluate the agronomic performance of 32 genotypes of sour passion fruit in the Federal District, as well as to estimate genetic parameters to be used in genetic improvement programs of this culture. 2. Material and Methods The experiment was carried out at Fazenda Água Limpa, belonging to the University of Brasília (UnB), located in Vargem Bonita, 25 km south of the Federal District of Brazil, with latitude of 16 ° South, longitude of 48 ° West and 1100 m of altitude. The climate of the region is of the AW type, characterized by rains concentrated in the summer, from October to April, and dry winters from May to September (Cardoso et al. 2014). The experiment was installed in a Red-Yellow Latosol soil, clayey, deep, with good drainage. In the experimental area, liming and the incorporation of 1 kg of simple superphosphate per pit in pre-planting was carried out. The soil analysis showed the following results: Al (0.05 meq); Ca + Mg (1.9 meq); P (4.5 ppm); K (46 ppm); pH 5.4 and 4% Al saturation. The cover fertilizations were carried out in a circle, at a distance of 40 to 50 cm from the surface of the plant, while the superphosphate was incorporated into the soil. Thirty-two genotypes were used, in a randomized block design, with eight plants per plot and four replications. The genotypes used were: PLANTA 6, MAR 20 # 40, PLANTA 1, MAR 20 # 29, MAR 22 # 2005, ROXO AUSTRALIANO, MAR 20 # 15, MSC, RC3, RUBY GIGANTE, ARO1, ARO2, MAR 20 # 49, BRS SOL DO CERRADO, MAR 20 # 6, PLANTA 5, MAR 20 # 23, PLANTA 4, PLANTA 2, PLANTA 7, MAR 20 # 03, EC30, MAR 20 # 10, MAR 20 # 34, MAR 20 # 21, FB200, FP01, BRS GIGANTE AMARELO, EC-RAM, GA2, REDONDÃO and MAR 20 # 39. Bioscience Journal | 2022 | vol. 38, e38004 | https://doi.org/10.14393/BJ-v38n0a2022-54231 3 VILELA, M.S., et al. These genotypes were developed based on research carried out by the University of Brasília - UnB and Embrapa Cerrados. They originate from intraspecific and interspecific hybridizations and also from materials derived from mass selection made in productive orchards in southeastern Brazil. The seedlings were produced by sowing in polystyrene trays (120 mL per cell) containing artificial substrate based on vermiculite plus Pinus sp. Bark, later transplanted to plastic bags containing soil sterilized with methyl bromide, remaining for about 90 days in a greenhouse at the Experimental Biology Station of UnB. On November 19 and 20, 2008, the seedlings were transplanted to the field, following the spacing of 2.75 m between rows and 3 m between plants, making a total of 1,024 useful plants with external borders. The vertical spreader support system was used in the field, with posts five meters apart and two strands of straight wire nº 12, at 2.20 m above the ground (upper wire) and 1.60 m (lower wire), with formation pruning in the combed scheme. The irrigation system used was a daily drip (one day watering shift), applying around 5 mm per m2 (5 liters/m2). The drippers were 30 cm apart. The control of weed plants consisted of periodic mowing between lines and use of post-emergent herbicides in the lines - glyphosate, in the form of a directed jet. It was agreed not to do any chemical control of diseases during the whole work, until the end of the harvests. Manual pollination was not carried out. Agronomic performance evaluations were carried out after one year of planting, from November 2009 to June 2010, totaling 28 harvests. The harvests were carried out collecting fruits with full maturity point, that is, fruits that were on the floor of the experiment. Each plot of the experiment was collected separately in plastic boxes identified according to the sketch of the experimental area. The boxes were taken to a shed for post-harvest evaluation, for the weighing procedure, which was followed weekly throughout the analysis period. The variables analyzed were: estimated productivity (kg ha-1), considering 9,697 plants per hectare, number of fruits per hectare, average fruit mass (g), and these characteristics considering the classification of fruits in terms of equatorial diameter into five exemplified categories shown in Table 1. Table 1. Classification of fruits according to their equatorial diameter (mm), used in the evaluation of 32 genotypes grown at FAL - UnB, 2009 to 2010, according to a proposal by Rangel (2002). Classification Equatorial Diameter (mm) First class Diameter less than 55 1 B Fruit diameter greater than 55 and less than 65. 1 A Diameter greater than 65 and less than 75 2 A Diameter greater than 75 and less than 90 3 A Diameter greater than 90 The experimental data were transformed by x + 1 root, submitted to analysis of variance and compared by the Tukey average test at 5% probability. Estimates of the genotypic variances between accessions (Ô𝑔 2), phenotypic at the level of mean (Ô𝐹 2 ) and environmental mean (Ô𝑒 2), (heritability at the level of mean (h2), coefficients of environmental variation (CVe) and genetic (CVg) for total productivity characteristic were obtained, using the Genes program (Cruz 2013). Using the estimates of phenotypic, genotypic and environment variances and covariance, the CVg/CVe ratio and phenotypic correlations were determined with Genes estatistic software. Linear correlation analyzes were performed between all studied variables, based on the significance of their coefficients. In the classification of intensity of the correlation to 0.05 ≤ p ≥ 0.01, it was considered very strong (r ± 0,91 to ± 1,00), strong (r ± 0,71 to ± 0,90), average (r ± 0,51 to ± 0,70) and weak (r ± 0,31 to ± 0,50), according to Guerra and Livera (1999). 3. Results and Discussion From the analyzed data, it was possible to observe significant statistical differences, over the 28 harvests, in the estimated productivity, productivity of first Class, 1B, 1A, 2A and 3A fruits, number of fruits, number of fruits in each classification, with the exception of 1A class, average mass and average mass for classifications, except for the average masses of 1B, 1A and 2A classes (Table 2). Regarding the estimated productivity, based on the Tukey means comparison test, at 5% probability, five groups were distinguished. The genotype MAR20#23 had the highest productivity with 17,162 kg ha-1, Bioscience Journal | 2022 | vol. 38, e38004 | https://doi.org/10.14393/BJ-v38n0a2022-54231 4 Agronomic assessment of 32 sour passionfruit genotypes in Federal District followed by AR 01 (15,626 kg ha-1) and PLANTA 7 with 15,130 kg ha-1, differing statistically from the genotypes MAR20#29, PLANTA 5 and EC-3-0, which presented the lowest productivity, with 4,762, 4,625 and 4,097 kg ha-1, respectively (Table 2). Moreira et al. (2018), studying the same genotypes, obtained similar results in 20 harvests, with estimated productivities values for the genotypes MAR20#23 (15,474 kg ha-1), PLANTA 7 (14,663 kg ha-1) and AR01 (13,996 kg ha-1), differing statistically from genotypes MAR20#29 and EC-3-0, which had the lowest yields, with 4,219 kg ha-1 and 4,055 kg ha-1. It was possible to observe that the MAR20#29 inferior performance data obtained in this work, corresponded to data found by these authors. Cavalcante et al. (2016), evaluating agronomic characteristics of four commercial cultivars, obtained yields of 15,759 and 11,664 kg ha-1 for the BRS Sol do Cerrado and BRS Gigante Amarelo genotypes, respectively. In this study, the yield obtained for the BRS Sol do Cerrado were relatively lower (8,650 kg ha - 1), but the productivity of the BRS Gigante Amarelo were very similar (11,665 kg ha-1) to the one obtained by these authors. The differences between these results may have occurred because the genotypes used in this study were selected from allogamous plants, thus segregation may occur, which may result in yields differences. Dias et al. (2017), evaluating the effects of Nitrogen and Potassium fertilizations in different passion fruit genotypes on Minas Gerais state, obtained a higher yield for the BRS Gigante Amarelo and BRS Sol do Cerrado (44.89 and 45.96 t ha-1, respectively). This yields differences obtained by Dias et al. (2017) can be explained by the field management used by these authors, in that case, the higher fertilizer doses, which was, in some treatments, almost two times higher than the recommended dose. The results found in this work, show that the EC-3-0 genotype had the lowest productivity value among the studied genotypes (4,097 kg ha-1). According to IBGE (2018), the average national passion fruit production was 602,651 tons with an average yield of 14.10 t ha-1. It is interesting to note that there was no artificial pollination in the current study, a procedure that would probably increase the productivity of the experiment, as it increases the pollination success and, consequently, the quantity of fruits. This procedure is commonly carried out in most commercial orchards. Genetic improvement studies usually do not adopt artificial pollination, because this procedure can cause an interference in the results. Considering the fruit production per hectare, there was statistical differences between the evaluated genotypes. The genotypes that stood out with the greatest amount of fruit produced were PLANTA 1, FB 200 and MAR 20#23 with 156,026, 150,545, and 136,901 number of fruits per hectare, respectively. The PLANTA 5 genotype obtained the lowest amount of fruits per hectare (32,042 fruits) (Table 2), demonstrating the differences that can be observed between passion fruit genotypes. Table 2. Estimated productivity and number of fruits per hectare for 32 genotypes of passion fruit cultivated at Fazenda Água Limpa during 28 harvests. Brasília-DF, 2020. GENOTYPES Estimated productivity kg ha-1 Number of fruits per hectare PLANTA 6 12,770.00abc 95,482.00Ab MAR 20#40 10,001.00abc 89,402.00Ab PLANTA 1 9,802.00abc 156,026.00A MAR 20#29 4,762.00bc 41,210.00Ab MAR 22#2005 10,405.00abc 85,265.00Ab ROXO AUSTRALIANO 5,930.00abc 77,285.00Ab MAR 20#15 13,690.00abc 107,585.00Ab MSC 5,626.00abc 48,842.00Ab RC3 6,242.00abc 52,442.00Ab RUBI GIGANTE 11,026.00abc 94,865.00Ab AR 01 15,626.00ab 116,282.00Ab AR 02 10,001.00abc 82,945.00Ab MAR 20#40 8,837.00abc 78,401.00Ab BRS SOL DO CERRADO 8,650.00abc 71,825.00Ab MAR 20#06 13,457.00abc 111,557.00Ab PLANTA 5 4,625.00bc 32,042.00B MAR 20#23 17,162.00a 136,901.00Ab PLANTA 4 11,882.00abc 93,637.00Ab Bioscience Journal | 2022 | vol. 38, e38004 | https://doi.org/10.14393/BJ-v38n0a2022-54231 5 VILELA, M.S., et al. PLANTA 2 12,545.00abc 112,897.00Ab PLANTA 7 15,130.00abc 108,242.00Ab MAR 20#03 13,925.00abc 110,890.00Ab EC-3-0 4,097.00c 38,026.00Ab MAR 20#10 9,802.00abc 99,226.00Ab MAR 20#34 11,237.00abc 102,401.00Ab MAR 20#21 14,885.00abc 131,770.00Ab YELLOW MASTER FB200 10,817.00abc 150,545.00A FP 01 11,026.00abc 91,810.00Ab BRS GIGANTE AMARELO 11,665.00abc 114,922.00Ab EC-RAM 7,057.00abc 57,122.00Ab GA 2 11,665.00abc 101,762.00Ab RENDONDÃO 8,465.00abc 74,530.00Ab MAR 20#39 6,242.00abc 55,226.00Ab *Means followed by the same letters do not differ by the Tukey test at 5% probability. Regarding the estimated productivity and the number of fruits when related to the size classification, the 32 genotypes showed statistically significant differences, except for the 1A class, where the number of fruits did not show any statistical difference (Table 3). It is worth mentioning that First class and 1B fruits are considered ideal for the industry, as they are not accepted in markets due to their small size. The other classes (1A, 2A and 3A) are destined for the in natura commercial markets (Pires et al. 2011). Regarding the fruit average mass classification, the studied genotypes showed statistically significant differences, in the F test at 5% of significance, only in the first and 3A classifications. For the first-class fruits, the average mass ranged from 37g (genotypes Roxo Australiano, MAR 20 # 40 and EC-RAM) to 65g (PLANTA 7). In fruits classified as 3A, the highest average mass value was observed in the MSC genotype with 170g (Table 3). Moreira et al. (2018) observed similar results, with, first class average fruit mass ranging from 38 g in EC-RAM to 71 g in MAR20#15 and for the fruits classified as 3A, the MSC genotype also presented higher average mass, with 168g. Cavalcante et al. (2016) obtained average fruit mass values in the order of 256.91g for the BRS Gigante Amarelo genotype, the highest average fruit mass achieved in his research. However, the cultivar FB 200 had the lowest fruit mass with an average value of 182.7g. In a study with thirty-five hybrids evaluated by Zaccheo et al. (2012), the average fruit mass varied between 130 and 205g. Jesus et al. (2018) pointed out that the highest average fruit mass obtained for the cultivars BRS Gigante Amarelo and BRS Sol do Cerrado, were 215g and 210g, respectively. Again, the cultivar FB 200 presented the lowest average among the three with 128g. The genetic parameters for the estimated productivity, number of fruits and average mass variables are shown in Table 4. The observed heritability for estimated productivity was 62.6%. Heritability measures the degree of correspondence between phenotypic and genetic values, and high values of this parameter indicate that simple selection methods such as mass selection can lead to considerable gains, considering that the environment has little influence (Falconer 1987). The ratio between coefficient of genetic variation and environmental variation, CVg/CVe ratio, indicate the relation between the genetic and environmental variability, and a value higher than one unit indicate a favorable situation for the genetic selection. The ratio observed in this study was 0.64, which reflects an unfavorable condition for selection, since the genetic variance was lower than the environmental variance. According to Cruz et al. (2013), values of this magnitude indicate that the use of simple improvement methods (i.e., mass selection) will not provide significant gains during the selection process. The use of breeding methods based on the performance of families is more appropriate than those that use selection based on the performance of individual plants. For the number and average fruit mass, the heritability values and CVg/CVe ratio were 55% and 0.53; 22% and 0.26, respectively (Table 4). Among the classes, the number of fruits and estimated productivity showed the following heritability values (Table 5), in order: first class fruits (79% and 79%), 1B (54% and 62%), 1A (12% and 57%), 2A (44% and 61%) and 3A (62% and 61%). Bioscience Journal | 2022 | vol. 38, e38004 | https://doi.org/10.14393/BJ-v38n0a2022-54231 6 Agronomic assessment of 32 sour passionfruit genotypes in Federal District Similar results were found by Moreira et al. (2018), who observed a heritability of 65% and CVg/CVe value of 0.69 for total yield, with 32 genotypes and 20 harvest. Viana et al. (2017), working with 95 sour passion fruit progenies, obtained heritability values and CVg/CVe ratio for estimated productivity of 63.01%. Silva et al. (2012) found different results, in a work carried out with 140 genotypes of complete siblings, where the estimates of heritability coefficients ranged from 19.54 to 71.38%. Two importantevaluated characteristics, number of fruits (NF) and estimated production (PT), showed low heritability estimates, with 39.19 and 28.04% respectively. The CVg/CVe ratio was lower than one unit for most variables. However, for the estimated productivity and the number of first-class fruits classification (Table 5), the values of the CVg/CVe ratio were very close to 1 (0,98 and 0,96, respectively). These values indicate a favorable condition, since the genetic variance exceeds the environmental variance (Vencovsky 1987). The estimated heritability for these variables were also the highest found, contributing to the values of the CVg/CVe ratio. Correlation is an important parameter in breeding programs since it allows simultaneous or indirect selection, especially when the character of interest has problems with measurement and identification or low heritability (Cruz et al. 2012). Table 3. Number of fruits, Productivity (kg ha-1) and Average mass (g) by fruit classification regarding the equatorial diameter. 2020. PROGENIES PT 1ª NF 1ª MM 1ª PT 1B NF 1B MM1B PT 1A PLANTA 6 1,765.00abcde 28,901.00abcdef 50abc 7,570.00abc 50,177.00ab 145a 3,137.00abc MAR 20#40 2,602.00abc 40,805.00abcd 65abc 5,777.00abc 41,210.00ab 145a 1,370.00abc PLANTA 1 1,157.00abcde 21,610.00abcdef 50abc 5,930.00abc 48,401.00ab 122a 2,117.00abc MAR 20#29 677cde 12,770.00cdef 50abc 3,250.00abc 24,026.00ab 122a 785bc MAR 22#2005 1,850.00abcde 31,330.00abcde 65abc 6,085.00abc 42,437.00ab 145a 2,117.00abc ROXO AUSTRALIANO 901bcde 19,601.00bcdef 37c 3,845.00abc 42,850.00ab 101a 1,090.00abc MAR 20#15 2,117.00abcd 35,722.00abcde 65abc 8,465.00abc 57,601.00ab 145a 2,602.00abc MSC 290e 5,330.00f 50abc 2,602.00c 29,242.00ab 101a 1,850.00abc RC3 677cde 14,162.00cdef 50abc 3,722.00abc 29,242.00ab 122a 1,601.00abc RUBI GIGANTE 1,850.00abcde 33,857.00abcde 50abc 6,725.00abc 49,285.00ab 122a 2,210.00abc AR 01 1,850.00abcde 31,330.00abcde 50abc 8,837.00abc 63,505.00ab 122a 3,970.00a AR 02 901bcde 18,226.00bcdef 50abc 5,777.00abc 47,962.00ab 122a 2,810.00abc MAR 20#40 1,090.00abcde 24,650.00abcdef 37c 4,625.00abc 37,637.00ab 122a 2,501.00abc BRS SOL DO CERRADO 901bcde 16,385.00cdef 50abc 4,901.00abc 41,210.00ab 101a 2,402.00abc MAR 20#06 2,117.00abcd 35,345.00abcde 65abc 8,650.00abc 63.002.00ab 122a 2,602.00abc PLANTA 5 530de 8,650.00ef 65abc 2,602.00c 16,642.00b 145a 1,297.00abc MAR 20#23 3,026.00ab 45,797.00abc 65ab 10.001.00a 71,825.00a 145a 3,845.00ab PLANTA 4 962bcde 17,690.00cdef 50abc 7,057.00abc 55,226.00ab 122a 3,137.00abc PLANTA 2 2,602.00abc 42,026.00abcd 50abc 6,890.00abc 55,226.00ab 122a 2,501.00abc PLANTA 7 2,117.00abcd 30,977.00abcde 65a 8,101.00abc 54,757.00ab 145a 4,226.00a MAR 20#03 1,937.00abcd 34,970.00abcde 50abc 8,465.00abc 59,537.00ab 14a 3,026.00abc EC-3-0 677cde 11,665.00def 50abc 2,705.00bc 21,317.00ab 122a 626c MAR 20#10 2,305.00abcd 44,522.00abc 50abc 5,626.00abc 44,522.00ab 122a 1,765.00abc MAR 20#34 2,305.00abcd 41,210.00abcd 50abc 6,562.00abc 49,730.00ab 122a 2,026.00abc MAR 20#21 3,601.00a 55,226.00ab 50abc 9,802.00ab 68,645.00a 122a 1,522.00abc YELLOW MASTER FB200 2,026.00abcd 29,242.00abcdef 65ab 7,226.00abc 62,501.00ab 122a 1,445.00abc FP 01 1,522.00abcde 27,890.00abcdef 50abc 7,226.00abc 52,901.00ab 122a 2,117.00abc BRS GIGANTE AMARELO 3,482.00a 59,050.00a 50abc 6,242.00abc 46,657.00ab 122a 1,682.00abc EC-RAM 785bcde 17,425.00cdef 37bc 3,845.00abc 28,562.00ab 122a 1,937.00abc GA 2 1,522.00abcde 30,277.00abcdef 50abc 7,570.00abc 59,050.00ab 122a 2,210.00abc RENDONDÃO 1,445.00abcde 26,570.00abcdef 50abc 5,330.00abc 40.001.00ab 122a 1,601.00abc MAR 20#39 1,025.00bcde 19,045.00bcdef 50abc 3,722.00abc 28,225.00ab 122a 1,226.00abc NF1ª: Number of first class fruits/ha, PT1ª: Estimated productivity for first class fruits (kg ha-1), MM1ª: Average mass of first class fruits in g, NF1B: Number of 1B fruits/ha, PT1B: Estimated productivity for 1B fruits in kg ha-1, MM1B: Average mass of 1B fruits in Bioscience Journal | 2022 | vol. 38, e38004 | https://doi.org/10.14393/BJ-v38n0a2022-54231 7 VILELA, M.S., et al. g, NF1A: Number of 1A fruits/ha, PT1A: Estimated productivity for 1A fruits in kg ha-1, MM1A: Average mass of 1A fruits in g, NF2A: Number of 2A fruits/ha, PT2A: Estimated productivity for 2A fruits in kg ha-1, MM2A: Average mass of 2A fruits in g, NF3A: Number of 3A fruits/ha, PT3A: Estimated productivity for 3A fruits in kg ha-1, MM3A: Average mass of 3A fruits in g. Averages followed by the same letters do not differ by Tukey's test at the 5% level. Table 3 (continued). Number of fruits, Productivity (kg ha-1) and Average mass (g) by fruit classification regarding the equatorial diameter. 2020. PROGENIES NF 1A MM 1A PT 2A NF 2A MM 2A PT 3A NF 3A MM 3A PLANTA 6 13,925.00a 226a 290ab 1,297.00ab 226a 10b 37b 65a MAR 20#40 6,725.00a 197a 10b 37b 50a 0b 0b 0a PLANTA 1 52,901.00a 122a 485ab 7,570.00a 197a 5b 10b 26a MAR 20#29 3,845.00a 170a 5b 10b 26a 0b 0b 0a MAR 22#2005 9,802.00a 197a 226ab 901ab 226a 0b 0b 0a ROXO AUSTRALIANO 6,085.00a 170a 37ab 2,210.00ab 65a 10b 37b 17a MAR 20#15 11,882.00a 226a 145ab 577ab 145a 0b 0b 0a MSC 9,802.00a 197a 577ab 2,305.00ab 257a 122a 401a 170a RC3 8,101.00a 197a 145ab 730ab 197a 0b 0b 0a RUBI GIGANTE 10,405.00a 197a 170ab 677ab 145a 0b 0b 0a AR 01 18,770.00a 197a 626a 2,501.00ab 257a 17ab 37b 82a AR 02 14,401.00a 197a 290ab 1,090.00ab 257a 0b 0b 0a MAR 20#40 12,322.00a 197a 577ab 2,210.00ab 145a 0b 0b 0a BRS SOL DO CERRADO 11,882.00a 197a 197ab 842ab 226a 10b 26b 26a MAR 20#06 12,322.00a 197a 145ab 577ab 257a 0b 0b 0a PLANTA 5 5,626.00a 226a 145ab 577ab 122a 0b 0b 0a MAR 20#23 18,497.00a 197a 170ab 730ab 197a 0b 0b 0a PLANTA 4 16,901.00a 170a 485ab 2,210.00ab 226a 0b 0b 0a PLANTA 2 12,997.00a 17a 401ab 1,445.00ab 257a 17b 37b 26a PLANTA 7 19,601.00a 197a 485ab 2,026.00ab 145a 0b 0b 0a MAR 20#03 13,690.00a 226a 40ab 1,601.00ab 257a 17ab 50ab 82a EC-3-0 3,845.00a 170a 26ab 101b 65a 0b 0b 0a MAR 20#10 8,837.00a 197a 65ab 290b 101a 0b 0b 0a MAR 20#34 10.001.00a 197a 145ab 577ab 122a 0b 0b 0a MAR 20#21 7,397.00a 197a 50ab 226b 145a 0b 0b 0a YELLOW MASTER FB200 43,265.00a 122a 65ab 290b 122a 10b 26b 26a FP 01 10,202.00a 197a 65ab 226b 257a 0b 0b 0a BRS GIGANTE AMARELO 8,282.00a 197a 65ab 401ab 122a 0b 0b 0a EC-RAM 9,217.00a 197a 82ab 325b 122a 0b 0b 0a GA 2 11,237.00a 197a 122ab 626ab 197a 0b 0b 0a RENDONDÃO 7,745.00a 197a 50ab 226b 145a 0b 0b 0a MAR 20#39 6,725.00a 197a 82ab 362ab 122a 0b 0b 0a NF1ª: Number of first class fruits/ha, PT1ª: Estimated productivity for first class fruits (kg ha-1), MM1ª: Average mass of first class fruits in g, NF1B: Number of 1B fruits/ha, PT1B: Estimated productivity for 1B fruits in kg ha-1, MM1B: Average mass of 1B fruits in g, NF1A: Number of 1A fruits/ha, PT1A: Estimated productivity for 1A fruits in kg ha-1, MM1A: Average mass of 1A fruits in g, NF2A: Number of 2A fruits/ha, PT2A: Estimated productivity for 2A fruits in kg ha-1, MM2A: Average mass of 2A fruits in g, NF3A: Number of 3A fruits/ha, PT3A: Estimated productivity for 3A fruits in kg ha-1, MM3A: Average mass of 3A fruits in g. Averages followed by the same letters do not differ by Tukey's test at the 5% level. Bioscience Journal | 2022 | vol. 38, e38004 | https://doi.org/10.14393/BJ-v38n0a2022-54231 8 Agronomic assessment of 32 sour passionfruit genotypes in Federal District Table 4. Estimates of phenotypic (Vf), genotypic (Vg), environmental (Ve), wide sense heritability (ha2), genetic variation coefficient (CVg) and ratio between genetic and environmental variance coefficient (CVg/CVe), using 28 harvests data from 32 sour passion fruit genotypes in the Federal District, described for 3 response variables. Brasília, 2020. Genetic Parameters PT NF MMT Vf (average) 316.99 2,978.02 0.46 Ve(average) 118.36 1,382.71 0.36 Vg(average) 198.62 1,594.31 0.10 ha2(average family) 62.65 % 53.53 22.12 CVg 14.05% 13.40 3.00 CVg/CVe 0.64 0.53 0.26 PT: Estimated productivity in kg ha-1, NF: Number of fruits per ha, MMT: Average fruit mass (g). The phenotypic correlation values are shown in Table 6. From the evaluated data, it was possible to observe a strong positive phenotypic correlation between the estimated productivity and number of fruits (rf = 0.86). The estimated productivity and number of fruits for each classification also showed a positive and very strong phenotypic correlation (first class, rf = 0.98; 1B, rf = 0.96; 2A, rf = 0.82; 3A, rf = 1.00). Values of this magnitude indicate that the characters mentioned are directly related to the increase in the quantity of fruits, and estimated productivity observed in the experimental field. Strong positive correlation values were found between the estimated productivity and estimated production for First class class (rf = 0.80), as 1B (rf = 0.98) and as 1A (rf = 0.78). In addition, the estimated productivity was also positively and strongly correlated with the number of fruits of the first and 1B classes (rf = 0.78 and 0.94 respectively). For the 3A fruits, very strong positive correlations were observed for the estimated productivity of 3A fruits and the average mass of fruits 3A class (rf = 0.92), shown in Table 6. Similar results were found by Santos et al. (2017) and Moreira et al. (2018). Pimentel et al. (2008) affirm that these results favor the breeder's decision making, since when they select for plants with a higher number of fruits, it occurs indirectly selection for greater productivity. It was possible to observe that in all fruit classifications, the estimated productivity showed a higher phenotypic correlation with the number of fruits than with the mass of the fruits (Table 6). Similar data were found by Silva et al. (2015), in which the estimated productivity showed a greater correlation with the number of fruits (rf = 0.971) and with the fruit mass (rf = 0.196), indicating that high productivity necessarily involves the selection of plants with a larger number of fruits. Negative correlation values were found between the number and average fruit mass (rf = -0.25). Negative and significant phenotypic correlation values were observed between the number of 1A fruits and the average mass of 1A (rf = -0.60); average 1B fruit mass and estimated productivity, number of fruits and average fruit mass of 3A classification (rf = -0.5; rf = -0.53; and rf = -0.37, respectively). Similar results were observed in other works (Silva et al. 2015; Moreira et al. 2018). This result indicate that a greater quantity of fruits means a lower average fruit mass. From these results, it appears that with the increase in the number of fruits, there may be a progressive reduction in their size. The negative correlation between number of fruits and average weight of fruits is an indication that the excessive number of fruits can lead to the production of fruits with less mass, with less commercial value (Koetz et al. 2010). Negative correlations between these characters suggest that a breeding program need to increase the number of fruits to a level that does not cause excessive competition between fruits, causing a reduction in the average mass. Bioscience Journal | 2022 | vol. 38, e38004 | https://doi.org/10.14393/BJ-v38n0a2022-54231 9 VILELA, M.S., et al. Table 5. Estimates of phenotypic (Vf), genotypic (Vg), environmental (Ve), wide sense heritability (ha2), coefficient of genetic variation (CVg) and the genetic variation coefficient and the environmental variation coeficient ratio (CVg/CVe), using data from 28 harvests of 32 genotypes of sour passion fruit in the field in the Federal District. Brasília, 2020. Genetic Parameters PT 1ª NF 1ª MM 1ª PT 1B NF 1B MM 1B PT 1A Vf (average) 116.46 1620.93 0.22 198.19 1205.37 0.33 88.79 Ve(average) 23.88 34-.35 0.08 73.95 542.78 0.24 37.49 Vg(average) 92.54 1280.57 0.13 124.24 662.59 0.08 51.30 ha2(average family) 79.48 79.00 60.44 62.68 54.97 26.25 57.77 CVg 24.39 21.75 4.83 14.42 12.08 2.59 15.59 CVg/CVe 0.98 0.96 0.61 0.64 0.55 0.29 0.58 NF1ª: Number of first class fruits/ha, PT1ª: Estimated productivity for first class fruits (kg ha-1), MM1ª: Average mass of first class fruits in g, NF1B: Number of 1B fruits/ha, PT1B: Estimated productivity for 1B fruits in kg ha-1, MM1B: Average mass of 1B fruits in g, NF1A: Number of 1A fruits/ha, PT1A: Estimated productivity for 1A fruits in kg ha-1, NF2A: Number of 2A fruits/ha, PT2A: Estimated productivity for 2A fruits in kg ha-1, MM2A: Average mass of 2A fruits in g, NF3A: Number of 3A fruits/ha, PT3A: Estimated productivity for 3A fruits in kg ha-1, MM3A: Average mass of 3A fruits in g. Table 5 (continued). Estimates of phenotypic (Vf), genotypic (Vg), environmental (Ve), wide sense heritability (ha2), coefficient of genetic variation (CVg) and the genetic variation coefficient and the environmental variation coeficient ratio (CVg/CVe), using data from 28 harvests of 32 genotypes of sour passion fruit in the field in the Federal District. Brasília, 2020. Genetic Parameters NF 1A PT 2A NF 2A MM2A PT 3A NF 3A MM3A Vf (average) 1230.62 43.16 276.34 8.36 4.23 14.90 10.53 Ve(average) 1072.43 16.60 153.37 7.97 1.61 5.66 5.31 Vg(average) 158.19 26.55 122.97 0.38 2.62 9.24 5.21 ha2(average family) 12.85 61.52 44.49 4.64 61.91 62.01 49.54 CVg 11.51 38.13 37.58 4.84 82.57 108.39 82.59 CVg/CVe 0.19 0.63 0.44 0.11 0.63 0.63 0.49 NF1ª: Number of first class fruits/ha, PT1ª: Estimated productivity for first class fruits (kg ha-1), MM1ª: Average mass of first class fruits in g, NF1B: Number of 1B fruits/ha, PT1B: Estimated productivity for 1B fruits in kg ha-1, MM1B: Average mass of 1B fruits in g, NF1A: Number of 1A fruits/ha, PT1A: Estimated productivity for 1A fruits in kg ha-1, NF2A: Number of 2A fruits/ha, PT2A: Estimated productivity for 2A fruits in kg ha-1, MM2A: Average mass of 2A fruits in g, NF3A: Number of 3A fruits/ha, PT3A: Estimated productivity for 3A fruits in kg ha-1, MM3A: Average mass of 3A fruits in g. Table 6. Estimates of phenotypic correlation values between characters from 32 genotypes of sour passion fruit cultivated at Fazenda Água Limpa. Brasilia, 2020. PT NF MMT PT 1ª NF 1ª MM 1ª PT1B NF1B MM1B PT 1 0.86* 0.21 0.80* 0.78* 0.51* 0.98* 0.94* 0.48* NF - 1 -0.25 0.77* 0.77* 0.41* 0.87* 0.92* 0.18 MMT - - 1 -0.09 -0.16 0.33 0.15 -0.04 0.60* PT 1ª - - - 1 0.98* 0.54* 0.80* 0.74* 0.47* NF 1ª - - - - 1 0.40* 0.78* 0.73* 0.42* MM 1ª - - - - - 1 0.52* 0.40* 0.56* PT 1B - - - - - - 1 0.96* 0.49* NF 1B - - - - - - - 1 0.23 MM 1B - - - - - - - - 1 PT 1A - - - - - - - - - NF 1A - - - - - - - - - MM 1A - - - - - - - - - PT 2A - - - - - - - - - NF 2A - - - - - - - - - MM 2A - - - - - - - - - PT 3A - - - - - - - - - NF 3A - - - - - - - - - MM 3A - - - - - - - - - PT: Estimated productivity, NF: Number of fruits, MMT: Average fruit mass, NF1ª: Number of first class fruits/ha, PT1ª: Estimated productivity for first class fruits (kg ha-1), MM1ª: Average mass of first class fruits in g, NF1B: Number of 1B fruits/ha, P T1B: Estimated productivity for 1B fruits in kg ha-1, MM1B: Average mass of 1B fruits in g, NF1A: Number of 1A fruits/ha, PT1A: Bioscience Journal | 2022 | vol. 38, e38004 | https://doi.org/10.14393/BJ-v38n0a2022-54231 10 Agronomic assessment of 32 sour passionfruit genotypes in Federal District Estimated productivity for 1A fruits in kg ha-1, NF2A: Number of 2A fruits/ha, PT2A: Estimated productivity for 2A fruits in kg ha- 1, MM2A: Average mass of 2A fruits in g, NF3A: Number of 3A fruits/ha, PT3A: Estimated productivity for 3A fruits in kg ha-1, MM3A: Average mass of 3A fruits in g. * Significant at 5% probability. Table 6 (continued). Estimates of phenotypic correlation values between characters from 32 genotypes of sour passion fruit cultivated at Fazenda Água Limpa. Brasilia, 2020. PT1A NF1A MM1A PT2A NF2A MM2A PT3A NF3A MM3A PT 0.78* 0.45* 0.15 0.35* 0.19 0.47* -0.08 -0.10 0.08 NF 0.54* 0.70* -0.30 0.24 0.32 0.32 -0.03 -0.05 0.11 MMT 0.49* -0.17 0.65* 0.35* -0.10 0.35* 0.05 0.02 0.05 PT 1ª 0.33 0.17 0.08 -0.11 -0.19 0.05 -0.28 -0.29 -0.17 NF 1ª 0.32 0.14 0.10 -0.12 -0.16 0.05 -0.31 -0.32 -0.20 MM 1ª 0.28 0.29 -0.02 -0.02 -0.18 0.14 -0.12 -0.15 -0.09 PT 1B 0.69* 0.44* 0.09 0.22 0.10 0.42* -0.16 -0.18 0.01 NF 1B 0.66* 0.52* -0.06 0.26 0.21 0.44* -0.01 -0.02 0.12 MM 1B 0.34 -0.05 0.50* -0.04 -0.28 0.03 -0.51* -0.53* -0.37* PT 1A 1 0.47* 0.30 0.74* 0.49* 0.70* 0.13 0.11 0.26 NF 1A - 1 -0.60* 0.50* 0.62* 0.39* 0.18 0.16 0.30 MM 1A - - 1 0.07 -0.27 0.17 -0.09 -0.08 -0.07 PT 2A - - - 1 0.82* 0.69* 0.47* 0.45* 0.54* NF 2A - - - - 1 0.48* 0.42* 0.42* 0.50* MM 2A - - - - - 1 0.35* 0.33 0.41* PT 3A - - - - - - 1 1.00* 0.92* NF 3A - - - - - - - 1 0.92* MM 3A - - - - - - - 1 PT: Estimated productivity, NF: Number of fruits, MMT: Average fruit mass, NF1ª: Number of first class fruits/ha, PT1ª: Estimated productivity for first class fruits (kg ha-1), MM1ª: Average mass of first class fruits in g, NF1B: Number of 1B fruits/ha, PT1B: Estimated productivity for 1B fruits in kg ha-1, MM1B: Average mass of 1B fruits in g, NF1A: Number of 1A fruits/ha, PT1A: Estimated productivity for 1A fruits in kg ha-1, NF2A: Number of 2A fruits/ha, PT2A: Estimated productivity for 2A fruits in kg ha- 1, MM2A: Average mass of 2A fruits in g, NF3A: Number of 3A fruits/ha, PT3A: Estimated productivity for 3A fruits in kg ha-1, MM3A: Average mass of 3A fruits in g. * Significant at 5% probability. 4. Conclusions The genotypes that stood out with the highest total productivity estimated were MAR20 # 23, AR 01 and PLANTA 7. The genotype MAR 20 # 23 also showed one of the highest values in terms of total number of fruits produced per hectare. For industrial purposes, the highest productivity and the largest number of fruits per hectare were found in genotypes MAR 20 # 21 and BRS Gigante Amarelo. For fresh consumption, the genotypes with the best performance were, respectively, PLANT 7, AR 01 and MSC. Total productivity estimated and total number of fruits per hectare in the first-class classification showed high values of heritability and CVg/CVe ratio. These results indicate a favorable condition for selection. Authors' Contributions: VILELA, M.S.: conception and design, acquisition of data, analysis and interpretation of data, drafting the article; PEIXOTO, J.R.: acquisition of data, analysis and interpretation of data; RAMOS, S.D.R.: acquisition of data, drafting the article; SOUSA, R.M.D.: acquisition of data, analysis and interpretation of data, drafting the article; OLIVEIRA, A.P.: acquisition of data, analysis and interpretation of data, drafting the article; TOSCANO, M.A.F.: acquisition of data, analysis and interpretation of data, drafting the article; OLIVEIRA JUNIOR, A.A.: conception and design, acquisition of data, analysis and interpretation of data, drafting the article. All authors have read and approved the final version of the manuscript. Conflicts of Interest: The authors declare no conflicts of interest. Ethics Approval: Not applicable. Acknowledgments: The authors would like to thank the funding for the realization of this study provided by the Brazilian agencies CAPES (Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brasil), Finance Code 001, and CNPq (Conselho Nacional de Desenvolvimento Científico e Tecnológico - Brasil). The authors would also like to thank the University of Brasilia (UnB) for the support of infrastructure and technical. Bioscience Journal | 2022 | vol. 38, e38004 | https://doi.org/10.14393/BJ-v38n0a2022-54231 11 VILELA, M.S., et al. References CARDOSO, M.R.D., MARCUZZO, F.F.N. and BARROS, J.R. Classificação Climática de Köppen-Geiger para o Estado de Goiás e o Distrito Federal. Acta Geográfica. 2014, 8, 40-55. http://dx.doi.org/10.5654/acta.v8i16.1384 CAVALCANTE, N.R., et al. Productivity, fruit physicochemical quality and distinctiveness of passion fruit populations. Revista Brasileira de Fruticultura. 2016, 38(4), e-142. https://doi.org/10.1590/0100-29452016142 CRUZ, C.D., REGAZZI, A.J. and CARNEIRO, P.C. Métodos biométricos aplicados ao melhoramento genético. 4th ed. Viçosa, MG: Universidade Federal de Viçosa, Imprensa Universitária, 2012. CRUZ, C.D. GENES: a software package for analysis in experimental statistics and quantitative genetics. Acta Scientiarum. Agronomy. 2013, 35(3), 271-276. https://doi.org/10.4025/actasciagron.v35i3.21251 DIAS, D.G., et al. Production and postharvest quality of irrigated passion fruit after n-k fertilization. Revista Brasileira de Fruticultura. 2017, 39(3), e-553. https://doi.org/10.1590/0100-29452017553 FALCONER, D.S. Introduction to quantitative genetics. Viçosa, MG: Universidade Federal de Viçosa, Embrapa Arroz e Feijão, 1987. FALEIRO, F.G., et al. Germoplasma e melhoramento genético do maracujazeiro: histórico e perspectivas. Planaltina: Embrapa Cerrados, 2011. Available from Internet: https://ainfo.cnptia.embrapa.br/digital/bitstream/item/76032/1/doc-307.pdf FALEIRO, F.G., et al. 2018. Avanços e Perspectivas do Melhoramento Genético de Passifloras no Brasil In: MORERA, M.P., et al. Maracujá: dos recursos genéticos ao desenvolvimento tecnológico. Brasília, DF: Embrapa Cerrados, pp. 85-95. Available from Internet: https://www.embrapa.br/en/busca-de-publicacoes/-/publicacao/1101174/maracuja-dos-recursos-geneticos-ao-desenvolvimento-tecnologico FALEIRO, F.G. and JUNQUEIRA, N.T.V. Maracujá: 500 perguntas e 500 respostas. Brasília, DF: Embrapa Informação Tecnológica, 2016. Available from Internet: https://www.embrapa.br/en/busca-de-publicacoes/-/publicacao/1061917/maracuja-o-produtor-pergunta-a-embrapa- responde GUERRA, N.B. and LIVERA, A.V.S. Correlação entre o perfil sensorial e determinações físicas e químicas do abacaxi cv. Pérola. Revista Brasileira de Fruticultura. 1999, 21(1), 32-35. IBGE - Instituto Brasileiro de Geograria e Estatística. Produção Agrícola Municipal. 2018. Available from Internet: https://www.ibge.gov.br/estatisticas/economicas/agricultura-e-pecuaria/9117-producao-agricola-municipal-culturas-temporarias-e- permanentes.html?=&t=resultados JESUS, C.A.S. de, et al. Fruit quality and production of yellow and sweet Passion fruits in northern state of São Paulo. Revista Brasileira de Fruticultura. 2012, 40(2), e-968. https://doi.org/10.1590/0100-29452018968 KOETZ, M., et al. Qualidade de frutos do maracujazeiro-amarelo em ambiente protegido e natural produzidos sob diferentes regimes de irrigação. Revista Brasileira de Agricultura Irrigada. 2010, 4(2), 115-126. MELETTI, L.M.M. Avanços na cultura do maracujá no Brasil. Revista Brasileira de Fruticultura. 2011, 33, 83-91. https://doi.org/10.1590/S0100- 29452011000500012 MOREIRA, H.S.M., et al. Agronomic characterization and genetic parameter estimation in yellow passion fruit. Bioscience Journal. 2018, 34(6), 58-70. https://doi.org/10.14393/BJ-v34n6a2018-39919 NOGUEIRA, I. Caracterização agronômica e fisíco-química de progênies de maracujazeiro azedo (Passiflora edulis Sims) no Distrito Federal. Brasília: Faculdade de Agronomia e Medicina Veterinária, Universidade de Brasília, 2016. Masters Dissertation. PERUCH, L.A.M., COLARICCIO, A. and BATISTA, D.C. Controle de doenças do maracujazeiro: situação atual e perspectivas. Agropecuária Catarinense. 2018, 31(1), 37-40. https://doi.org/10.22491/RAC.2018.v31n1.2 PIMENTEL, L.D., et al. Seleção precoce de maracujazeiro pelo uso da correlação entre dados de produção mensal e anual. Pesquisa Agropecuária Brasileira. 2008, 43(10), 1303-1309. https://doi.org/10.1590/S0100-204X2008001000007 PIRES, M.M., JOSÉ, A.R.S. and CONCEIÇÃO, A.O. Maracujá: avanços tecnológicos e sustentabilidade. Ilhéus: Editus, 2011, pp. 237. RANGEL, L.E.P. Desempenho agronômico de nove genótipos de maracujazeiro-azedo cultivados sob três níveis de adubação potássica no Distrito Federal. Brasília. Universidade de Brasília, 2002. Masters Dissertation. SANTOS, C.L., et al. Relationship between yield and fruit quality of passion fruit c03 progenies under different nutritional levels. Revista Brasileira de Fruticultura. 2017, 39(2), e-691. https://doi.org/10.1590/0100-29452017691 SILVA, F.H.L., et al. Generating relevant information for breeding Passiflora edulis: genetic parameters and population structure. Euphytica. 2015, 208(3), 609–619. https://doi:10.1007/s10681-015-1616-8 SILVA, M.G.M., et al. Seleção recorrente intrapopulacional no maracujazeiro amarelo: Alternativa de capitalização de ganhos genéticos. Ciência e Agrotecnologia. 2009, 33(1), 170-176. https://doi.org/10.1590/S1413-70542009000100024 SILVA, M.G.M., et al. Biometria aplicada ao melhoramento intrapopulacional do maracujazeiro amarelo. Revista Ciência Agronômica. 2012, 43(3), 493-499. http://dx.doi.org/10.5654/acta.v8i16.1384 https://doi.org/10.1590/0100-29452016142 https://doi.org/10.4025/actasciagron.v35i3.21251 https://doi.org/10.1590/0100-29452017553 https://ainfo.cnptia.embrapa.br/digital/bitstream/item/76032/1/doc-307.pdf https://www.embrapa.br/en/busca-de-publicacoes/-/publicacao/1101174/maracuja-dos-recursos-geneticos-ao-desenvolvimento-tecnologico https://www.embrapa.br/en/busca-de-publicacoes/-/publicacao/1061917/maracuja-o-produtor-pergunta-a-embrapa-responde https://www.embrapa.br/en/busca-de-publicacoes/-/publicacao/1061917/maracuja-o-produtor-pergunta-a-embrapa-responde https://www.ibge.gov.br/estatisticas/economicas/agricultura-e-pecuaria/9117-producao-agricola-municipal-culturas-temporarias-e-permanentes.html?=&t=resultados https://www.ibge.gov.br/estatisticas/economicas/agricultura-e-pecuaria/9117-producao-agricola-municipal-culturas-temporarias-e-permanentes.html?=&t=resultados https://doi.org/10.1590/0100-29452018968 https://doi.org/10.1590/S0100-29452011000500012 https://doi.org/10.1590/S0100-29452011000500012 https://doi.org/10.14393/BJ-v34n6a2018-39919 https://doi.org/10.22491/RAC.2018.v31n1.2 https://doi.org/10.1590/S0100-204X2008001000007 https://doi.org/10.1590/0100-29452017691 https://doi:10.1007/s10681-015-1616-8 https://doi.org/10.1590/S1413-70542009000100024 Bioscience Journal | 2022 | vol. 38, e38004 | https://doi.org/10.14393/BJ-v38n0a2022-54231 12 Agronomic assessment of 32 sour passionfruit genotypes in Federal District VENCOVSKY, R. 1987. Herança quantitativa. In: PATERNIANI, E. and VIEGAS, G.P. (Eds.). Melhoramento e produção de milho no Brasil. Campinas: Fundação Cargil, pp. 137-214. VIANA, A.P., et al. Implementing genomic selection in sour passion fruit population. Euphytica. 2017, 212(10), Article number 228. https://doi.org/10.1007/s10681-017-2020-3 ZACCHEO, P.V.C., et al. Production and qualitative characteristics of the fruits of yellow passion fruit hybrids. Revista Brasileira de Fruticultura. 2012, 34(4), 1113-1120. https://doi.org/10.1590/S0100-29452012000400019 Received: 26 April 2020 | Accepted: 22 July 2020 | Published: 16 February 2022 This is an Open Access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. https://doi.org/10.1007/s10681-017-2020-3 https://doi.org/10.1590/S0100-29452012000400019