Seed germination in natural populations of Hancornia speciosa Germinação de sementes em populações naturais de Hancornia speciosa

The objective of the current study was to assay the germination behavior of seeds from fruits collected from three natural populations of H. speciosa, located in remnant areas of Cerrado in the municipality of Porto Nacional, Tocantins state, Brazil. The studied populations were: Canaã, São Judas Tadeu and Providência. Seven genotypes were selected from each population and twenty fruits of each genotype were collected. After ripening, the fruits were manually pulped and seeds of the same genotype pooled. After cleaning, removing adhered pulp and exterior sterilization, seeds were sown in a nursery. The experimental used a Completely Randomized Design (CRD) with three replications. Four seeds of each genotype were sown per polyethylene bag and the germination monitored until the 45th day after sowing. Variables used to assay germination behavior were: Germination (G,%); Mean Time of Germination (MTG, days); Average Speed of Germination (ASG, d-1), Germination Time Variation Coefficient (VCt) and Germination Synchrony Index (Z). For germination data, nonparametric Kruskal-Wallis analysis of variance was performed, followed by Dunn's test at 5% probability. There was a difference between populations for all germination variables, with the Providência population performing the best. The variables MTG, ASG and Z were significantly higher for the Providencia population. In absolute terms, the Providencia population had the highest percentage of germination (G), but did not differ significantly from the São Judas Tadeu population. The results generated are important for understanding the germination of seeds of different populations of H. speciosa. This information is relevant for the ex situ spread of the species.


INTRODUCTION
Hancornia speciosa Gomes (mangabeira) is a native fruit tree of the Brazilian Cerrado, where it typically occurs on sandy and acidic soils with low fertility (COSTA et al., 2015). This species occurs in the midwest, north, northeast and southeastern regions of Brazil. Its wide distribution, across a wide variety of habitats, indicates a natural ability to adapt to different environments (RODRIGUES et al., 2017). In the Cerrado region the species commonly flowers between August and November, peaking in October, and fruit production from September to November (VIEIRA et al., 2017).
Interest in this species has been growing in recent years, with the fruit (mangaba) as the main commercial focus (ALMEIDA et al., 2018). Mangaba has good digestibility and high nutritional value, with higher protein content than most commercialized fruits (NASCIMENTO et al., 2014). It is rich in ascorbic acid, potassium, iron and zinc (LIMA et al., 2015), and can be consumed fresh or processed as juices, cookies, liqueurs, jams, candies, desserts, cakes, ice cream (OLIVEIRA et al., 2017).
Despite the economic potential, there are few commercial plantations of H. speciosa in Brazil and in most states, the fruits that reach the market and industries come from extractivism (ARRUDA et al., 2016). As well as other fruits of the Cerrado, H. speciosa presents commercial demand far above the supply capacity by means of extractive and this has stimulated a more technified production, encouraging research that seeks new knowledge about this species which is not yet fully domesticated (OLIVEIRA et al., 2018).
Deforestation, real estate speculation, plantations and monocultures have greatly reduced the original area of the ecosystems in which H. speciosa occurs, so that the species is now one of the most endangered native fruits in the Cerrado domain (DA SILVA et al., 2011). It is among the ten species considered CNPq to be of highest

Página | 378
information can be used to develop mechanisms for the management and conservation of genetic resources and to allow inferences concerning the behavior of the species in environments currently undergoing transformation, such as the Cerrado (PINHEIRO et al., 2018).
Accordingly, the current study aimed to record germination behavior of H. speciosa seeds from fruits collected in three natural populations of located in remnant areas of the Cerrado in Porto Nacional, Tocantins State, Brazil.

Study area
The work was carried out in three natural populations of H. speciosa var. speciosa occurring in areas of typical cerrado, on private properties located close to the city of Porto Nacional, Tocantins State, Brazil (Fig. 1).
The São Judas Tadeu population is located 24 km from Porto Nacional along the TO-070 highway, connecting Porto Nacional to Brejinho de Nazaré. The Canaan population is located approximately 10 km from the urban area of Porto Nacional towards Monte do Carmo on the TO 230 highway, and Providência population is located 18 km from Porto Nacional on the TO 050 highway connecting Porto Nacional to Palmas.
The municipality of Porto Nacional lie at 212 meters altitude, and the climate of the region has two seasons, a dry season (from May to September) and a rainy season (from October to April). The climate is Aw according to the Köppen classification.

Fruit collection
In November 2017, twenty physiologically mature fruits were collected from seven genotypes of each population (total of 140 fruits per population), georeferenced with the aid of a Global Positioning System (GPS) and numbered with aluminum tags, in order to facilitate locating and identifying them in the field. Determination of the genotypes for sampling was based on earlier studies (FREITAS et al., 2012), which had identified which genotypes were best representated by the phenotypic variability present in each population.
After complete maturation, the fruits were manually pulped and seeds from the same genotype then grouped. Seeds were cleaned in order to remove as much of the adhering pulp as possible without damaging them and subsequently disinfected with sequential immersions, first in 70% alcohol for one minute, then in 2.0% sodium hypochlorite (NaClO) for ten minutes and, finally, in distilled and then autoclaved water Nursery shading rate of 50% was used, with seeds sown at a depth of one centimeter (OLIVEIRA et al., 2018) and germination monitered until the forty-fifth day after sowing. Irrigation was performed periodically to maintain ambiente soil mostiure.
Daily average temperature and daily precipitation data were obtained from the Teaching and Research Meteorological Database (BDMEP) of the INMET (National Institute of Meteorology) station, located in Porto Nacional, Tocantins.

Soil analises
Three seperate soil samples were collected and mixed to obtain a composite sample for later physicochemical analysis. Results of its physicochemical analysis of the soil used as a substrate appear in Tab. 1. In general, it is a nutrient-poor soil with high acidity, physically classified as possessing a medium texture. Such profiles are typical of Cerrado soils.

Data analises
The variables used to evaluate germination were -Germination The criterion for ''germination'' was the emergence of any part of the seedling.
The number of seeds meeting such a criterion was counted daily. For germination data, nonparametric Kruskal-Wallis analysis of variance was performed, followed by Dunn's test at 5% probability.

RESULTS AND DISCUSSION
In the current study, we used emergence of any part of the seedling to define seed germination. Seedling emergence was followed for 45 days after sowing ( The overall experimental average for the germination variable (G) was 63.99%.
This average is not considered low, since the seed germination percentage of the species is often reduced due to recalcitrance, plus the inhibitory action of the pulp conferring (LORENZI, 2009;OLIVEIRA et al., 2014). In the current study, methods were adopted so that these factors did not influence the obtained results. According to Soares et al. (2007) H. speciosa seeds should be sown immediately or within 48 hours after removal from the fruit, as from the fourth day germination power drops rapidly.
A possible explanation for the germination rates on the current substrate is that H. speciosa grows mostly in soils with low organic matter content, high acidity and low nutrient availability and exchangeable bases (FERREIRA; MARINHO, 2007  The difference in germination velocity between seeds from different populations of the same species may depend, among other factors, on seed size (NOGUEIRA et al., 2003).
Several authors recommend that the largest and heaviest seeds be used to standardize seedling emergence and to obtain seedlings of similar size or greatest vigor (DA SILVA  Carvalho and Nakagawa (2012), the positive effect of large seed size is related to the fact that they are likely to have well-formed embryos and larger amounts of reserves, potentially making for more vigorou seedlings.
Preliminary analysis of fruit and seed morphology from the populations studied showed that the seeds of the Canaan population were smaller in size and mass than Factors that influence seedling germination and emergence include temperature, humidity, oxygen availability, soil structure and seed depth (COSTA; DANTAS, 2009).
The degree with which these external factors influence germination varies between Temperature interferes with germination rates and percentage success, notably by changing the rate of water absorption and altering the speed of chemical reactions that will mobilize or degrade stored reserves and the rate of synthesis of seedling growth substances (BEWLEY; BLACK, 1994). There is no one uniform effect for all species, as values are species-specific and dependent on the climatic conditions of their regions of origin at the time of natural germination (ALVES et al., 2015). Each species has an optimal germination temperature at which maximum germination occurs in the shortest time, and maximum and minimum temperature limits beyond which germination does not occur (MELO et al., 2008). Most tropical species, for example, germinate at temperatures between 20 and 30 °C (ARAUJO et al., 2016). According to Brancalion et al. (2010) the 25 °C median is considered the most suitable temperature for most species in the Cerrado biome. Graphical analysis of temperature data (Fig. 2) shows that the mean temperature during the experiment was 28 ºC, ranging across 25 ºC to 32 ºC, and so within the optimal temperature range for the species. Consequently, temperature was an unlikley interference factor for seed germination in the current study.  studying the germination of mangabeira seeds as a function of substrate and temperature, and recommended that laboratory tests of germination and GMV be performed using blotting paper as the substrate and temperatures between 25 ºC and 30 ºC, and paper roll substrate used at a temperature of 30 °C.
The humidity of the substrate on which seeds are sown is one of the essential factors for triggering the germination process (CARVALHO; NAKAGAWA, 2012).
During this process, the water absorption is key to promoting coat seed softening and embryo and reserve tissues enlargement, as well as favoring rupture of the testa, gas diffusion and the emergence of the primary root. Water is also important for the protoplasm dilution, allowing the diffusion of hormones and consequently enzymatic system activation. As a result, digestion, translocation and assimilation of reserves develops, resulting in embryo growth (RAMOS et al., 2006). Water deficit reduces cell expansion and may affect the process of cell division and this interferes with plant development. In addition, there is often a decrease in leaf water potential, which induces stomata closure, resulting in reduced gas exchange and consequently biomass production. On the other hand, flooded or excessively moist soils, limitations on oxygen diffusion may also slow down germination rates, largly due to the absence or scarcity of oxygen, so promoting with-in cell production of ethanol which is toxic to normal metabolism, so causing tissue acidification and death (GORDIN et al., 2016). During the experiment, water was constantly supplied throughout the germination period, keeping the field capacity of the substrate irrigated in a manner that avoided soil saturation. However, as shown in the rainfall graph (Fig. 2), extensive rainfall naturally occurs during the experimental period, with precipitation averaging around 6.9 mm/day (range 0 to 44.5 mm), which would favor moisture retention for an extended period. In addition, germination peaks occurred during periods with lower precipitation and milder temperature (around 26 ºC) ( Fig. 2 and 3).
The large volume of rainfall during the experiment, generated briefly (few days) an excessive water accumulation in the substrate, which may have favored the multiplication of rotting microorganisms and decreased survival of H. speciosa seeds.
According to Lewis and Clements (1999), large amounts of such water can provide suitable habitats for the development of microorganisms that negatively affect seed germination.

Página | 387
The Providência population had a more homogeneous germination process than the other two populations (Tab. 2). Why this was so may be related to the place of location of the population, as it is lies closest to the Luiz Eduardo Magalhães reservoir and so occurs in a region where the water table lies closer to the surface. Since the experimental period was accompanied by a large amount of rain (Fig. 2), the accompanying flooding may have reproduced the environment to which plants of this population are more adapted. Another factor that has a major influence on germination is light, with perception and translation of the light stimulus occuring in the embryo is responsible (DOUSSEAU et al., 2008). Seeds have a different degree of sensitivity to light, with species having seeds that are positively or negatively affected (LOPES et al., 2005). In  (1994) found that a full sun environment with 100% luminosity gave the highest and fastest germination rates. The current experiment was carried out in a nursery covered by a black screen providing 50% shading. In a study on protected environments and substrates for H. speciosa seedling production, Arrua et al. (2016) found both the 50% shading black screen and the 50% aluminized heat-reflective screen to be suitable for H. speciosa seedling production. Consequently, it is unlikely that luminosity was a factor negatively interfering with seed germination in the current experiment.
Oxygen is necessary for the promotion of a variety of key metabolic reactions in the seed, especially respiration. Although respiration in the early stages of germination is generally anaerobic, it soon becomes oxygen dependent (BORGES; RENA, 1993).
According to Melo et al. (2008) the need for oxygen is affected by such factors as humidity and high temperatures. In saturated or excessively moist soils, limitations on oxygen diffusion may also halt germination, probably because oxygen absence or scarcity favors ethanol production in cells, and this is toxic to normal metabolism causing tissue acidification and seed death (GORDIN et al., 2016). As seen in Tab penetration, also exerts physical impediments to seedling growth until they reach the soil surface and no longer depend on cotyledon reserves (TOLEDO et al., 1993). Depth of seed deposition can affect germination, conditioned by temperature, water content, seed peculiarities, soil physical and chemical properties, and climate, among other factors (BOTTEGA et al., 2014). According to Oliveira et al. (2018), the ideal sowing depth guarantees the most homogeneous germination profile, with shorter emergence times and more vigorous seedlings. As pointed out by Koakoski et al. (2007), the greater the depth of planting, the greater the volume of energy consumed while emerging, as well as the greater the damage caused by low temperatures and oxygen levels. On the other hand, at shallower depths the seed has an enhanced possibility of hydric stress.
According to Oliveira et al. (2018), H. speciosa seeds are influenced by sowing depth and substrate type and should be planted 1 cm deep in cerrado soils that contain up to 15% of washed sand. In the present study, sowing depth was 1 cm, and so probably did not negatively influence seed germination.

CONCLUSION
There was a difference between populations for all germination variables, with the Providência population performing the best. Factors such as seed size and environmental conditions may have favored the Providência population.
Average germination time, average germination speed and synchronization index were highest for the Providência population.
In absolute terms the Providência population had the highest germination percentage, but did not differ significantly from the São Judas Tadeu population. and its responses to temperature. Scientia Horticulturae, v. 197, p. 399-403. 2015