Plant material

Fifteen eggplant accessions belonging to the genebank of Corporación Colombiana de Investigación Agropecuaria (Agrosavia) were evaluated as follows: C003, C006, C011, C014, C015, C025, C026, C027, C028, C032, C035, C036, C040, C042, and C049. The accessions were planted in an experimental plot of the Caribia Research Center (CI) of Agrosavia located in Zona Bananera [banana and plantain production zone] of Magdalena State, Colombia, from June to November 2017.

The CI Caribia is located according to the geographic coordinates at 10° 47' N latitude and 74° 10' W longitude at an elevation of 18 m. The region has an average annual temperature of 28 °C, with a relative humidity of 82% and an average annual rainfall of 1,280 mm, placing it according to Holdridge's climate classification (1967) in the Tropical Dry Forest (Bs-T) life zone.

The accessions were planted in the field under a completely randomized design with four repetitions. The experimental unit comprised 40 plants in a plot of 1×x 1 m. A drip irrigation system was used, using drip tapes with a discharge of 0.8 L h^-1 and irrigation until the soil reached field capacity. The CropWat 8.0 program was used to calculate water needs. For this, the amounts of water lost by the crop or crop evapotranspiration (ETc) were estimated. The ETc was estimated using Equation 1 .

Where Kc is the crop coefficient; for the current study, the Kc values used were the ones reported by ^{Allen et al. (2006)}. Reference evapotranspiration (ETo) was calculated using the FAO Penman-Monteith method (^{Allen et al., 2006}), with climatic data obtained from the Davis-Vantage2plus6162 automated climatological station of CI Caribia.

The calculation of degree days was estimated based on the standard equation or simple method (^{López et al., 2011}) according to Equation 2:

Where:

The Tb used was the one reported by ^{Maynard and Hochmuth (2007)}, i.e., 15.6 °C. Four phenological stages established for Solanaceae by ^{Meier (2001)} were used with adaptations (Table 1) based on the number of days in which each phenological stage was completed in 50% of the plants sown in each plot.

Table 1 List of stages established for 15 Solanum melongena L. accessions. 

The yield was expressed in tons per hectares (t ha^-1) during 14 harvests. Fruits that had not reached physiological maturity (unripe state) but that had the typical shape and size of their respective cultivar were considered fruits suitable for harvest. For data analysis, descriptive statistics and multivariate principal components analysis were used, using the Infostat program version 2017.

Climatic characterization of the experimental location

Because the municipality of Zona Bananera is in a low altitude region, the temperatures vary according to the altitude of the territory (temperature zones), and this municipality, having heights of less than 1,000 m.a.s.l., has a temperature behavior corresponding to a warm temperature zone, in which the average annual temperature is higher than 24 °C. According to IDEAM, the maximum average temperature of the municipality of Zona Bananera is 28 °C, remaining almost constant throughout the municipality, as well as a minimum of 24 °C (Municipio Zona Bananera, Plan básico de ordenamiento territorial, 2001).

As it can be seen in Figure 1, in the first vegetative growth stage of eggplant, which lasted approximately one month, temperatures were above 15 °C (the temperature that is limiting the crop); therefore, no event caused growth retardation. Likewise, the same occurred with the maximum temperatures, which did not exceed 35 °C, being optimal for the crop.

Figure 1 Characterization of temperature (°C) and relative humidity (%) during the phenological cycle of eggplant in Zona Bananera, the Caribbean region, Colombia. 

Similarly, this also occurred with the relative humidity showing values between 60 and 100%, typical of the humid Caribbean region.

Regarding the presence of diseases, even though the environment was suitable for the establishment of some limiting pathogens for eggplant such as Sclerotium rolfsii, its incidence was very low with 0.04% of infected plants (17 plants).

In Figure 2 and Figure 3, precipitation throughout crop phenology was higher than the reference evapotranspiration (Eto) and much higher than the Eto/2, which indicates that there were no critical drought scenarios.

Figure 2 Precipitation and reference evapotranspiration during the phenological cycle of eggplant in Zona Bananera, the Caribbean region, Colombia. 

Figure 3 Precipitation and crop evapotranspiration during the phenological cycle of eggplant in Zona Bananera, Caribbean region, Colombia. 

In the same way, during the four phenological crop stages, total precipitation was similar to crop evapotranspiration, which indicates that, in general terms, the water requirements were supplied through precipitation in the area and the shortage was obtained through irrigation. Moreover, the evapotranspirative demand of the crop during the four phases was 560 mm. However, the soil was monitored with tensiometers, one of 0-20 cm and another of 20-40 cm, to maintain soil humidity at field capacity.

Accumulation of degree days in 15 eggplant accessions

The accumulation of degree days for 15 eggplant accessions reached values from 892.14 to 1,077.71 °C (Table 2), with which we could select early or late accessions for these climatic conditions.

Table 2 Accumulation of degree days in 15 eggplant accessions for the phenological stages of leaf development, appearance of the floral organ, flowering, and fruit formation. 

On the other hand, accessions C040, C014, and C035 were those that produced fruits in less time with accumulated values in degree days of 892.14±76.21 and 910.4±49.05, respectively. This behavior is of great importance since these accessions were the ones that had the highest yields with values above 8,000 gram per plant, for a planting density of 10,000 plants per hectare, 80 t ha^-1 were obtained, i.e., a significantly higher value than the national yield, which is around 16 t ha^-1 (^{Araméndiz et al 2008b},^c) (Figure 4). The opposite occurred with genotypes C015, C011, C032, and C025; the last two are the latest to start producing fruits suitable for harvesting, as observed in the degree days accumulation for this phenological phase (Table 2). However, no significant differences (P>0.05) were found between accession for any of the phenological stages evaluated, which gives indications that those that begin to produce earlier fruits will be those that will have the highest final yields.

Figure 4 Yield (grams per plant) and degree days (°C) to fruits development suitable for harvesting in 15 eggplant accessions. 

Accessions C028, C032, and C025, were the latest ones with more than 1,000 °C accumulated to produce fruits suitable for first harvest (Table 2); moreover, these had a strong correlation with degree days to flowering, as observed in the multivariate biplot with the largest and most significant vectors for component 1 (Figure 5).

Figure 5 Double representation of 15 eggplant accessions with the variables evaluated. 

The accumulation of degrees days was similar for the 15 accessions in the first phenological stage (vegetative growth) with average values of 388.66 °C (Table 2); this effect was because this phenological stage was measured until the plants had produced nine leaves. Furthermore, it should be noted that the maximum vegetative development rate in the crop is reached when flowering begins.

An opposite behavior was reported by ^{Rouphael et al. (2010)} with a highly positive relationship between the number of leaves and the thermal time during the experiment in three aubergine cultivars. The response of the numbers of the leaves to the thermal time was curvilinear, with values 450 degree days. After 450 degree days, the increase in the number of leaves per plant was linearly related to the thermal time. However, ^{Maynard and Hochmuth (2007)} report for aubergine a base temperature of 15.6°C and a maximum temperature of 35 °C. These results could be useful to generate a model for the leaf area development and finally, a cultivation growth model for this crop.

This behavior is since crop development depends to a large extent on the temperature and the photoperiod; meanwhile, in the tropics, the temperature is the environmental variable with the greatest influence on crop development. The regulating role of temperature is through its action on the enzymatic reactions that directly or indirectly intervene in development processes (^{Baker and Reddy, 2001}).

This behaviour is explained by ^{Martín and Jerez (2017)}, who mentions that temperature is a basic factor that influences the development rate, particularly from emergence to flowering and maturity. Many species are adapted to particular temperature ranges; thus, as the environmental temperature or mean temperature (Tm) decreases, its development rates are reduced to such an extent, that if the temperature falls to the limit, the development stops. This temperature limit is called the base temperature (Tb). On the contrary, as the temperature increases, development rates increase to an optimum temperature (To), from which they decrease and eventually stop at their maximum development temperature (Tmax).

For eggplant cultivation limiting temperatures are those below 15.6 °C and above 32 °C, finding the optimum at 22 °C (^{Maynard and Hochmuth, 2007}; ^{Uzun, 2007}). Hence, the climatic offer of the region did not have stressful temperatures for the crop, and it had a rather additive effect, causing some accessions to begin their production in less thermal time or days.

In agriculture, the combination of time and temperature comprises the so-called thermal time (TT) or also known as the sum of heat, degree day, growing degree days, or physiological time. This approach has been used in numerous studies to describe the phenology of crops like corn (Zea mays L.), sugar cane (Saccharum spp.), melon (Cucumis melo L.), chickpea (Cicer arietinum L.) and cotton (Gossypium hirsutum L.) (^{Baker and Ready, 2001}).

Likewise, ^{Sadek et al. (2013)} and ^{Fealy and Fealy (2008)}, specify the importance of heat requirements on eggplant cultivation since these can be quantified and associated with the time when the fructification stage begins. This crop will not continue growing if the average temperature is below the base (12 °C), and the degrees of heat provides a reasonable estimate of the heat energy available for plant growth representing an important factor during the crop cycle.

On this aspect, ^{Uzun (2007)} points out a behavior similar to the one reported in this research, in which the heat requirements were different between varieties of eggplant. Meanwhile, these authors show a curvilinear effect of the variables associating yield with the temperature factors and light intensity, the optimum being 17-20 °C and 7-17 MJ m^-2 d^-1 for temperature and radiation, respectively; above these values number of fruits per plant as well as average fruit weight decreased, and therefore, yield.

Nevertheless, ^{Kürklü et al. (1995)} conducted a study to define the response of eggplant at different temperatures in controlled greenhouse environments. The temperatures established in the greenhouses were 14, 18, 22, 26, and 30 °C. Vegetative growth was much higher as the temperature increased to a point where the growth decreased, meanwhile fruit development (size, fruit weight, and yield) had an opposite behavior, increasing their values as the temperature increased to 30 °C. The total harvested yield of eggplant at different temperatures had the highest fruit yield at 19, and 16 °C (approximately 33, 32, and 28 kg, respectively), and plants that grew at 26 and 30 °C produced much less than plants grown at lower temperatures (18 and 15.5 kg, respectively). Therefore, the optimum temperature for fruit production in eggplant is approximately 22 °C.

Multivariate analysis by principal components

Authors, as ^{Pla (1986)} mentions that it is not valid to evaluate only one component to explain data variability. Because of, it was found that only 23.7% synthesize the variability in the first component; further, applying the criterion of selection of principal components of Kaiser (^{Pla, 1986}), the first three components were used obtaining a value of 54%.

In Figure 5 accessions C032, C025 and C028 have similar behavior and are associated with the variables degree days to leaf development, floral bud starting and flowering; meanwhile, the latter was more associated with the eighth harvest and degree days to leaf development. Similarly, a second group was formed with accessions C011, C042, and C015. In the opposite axes, three large groups are observed, accessions C003, and C036 associated mainly with harvests 6, 7, 10, 11 and 13; whereas, accessions C027, C049, C006, and C014, were associated with harvests 3, 4, 5, 9 and 14; and finally, accessions C040, C035 and C026 showed a strong association to the total variable yield.

^{García and López (2002)} explain that this behavior occurs when plants are exposed to thermal variations on the physical environment, and these have great influence on the different physiological, biochemical and metabolic processes leading to their growth and development; furthermore, these variations determine the leaf area and the accumulation of dry matter during the biological cycle of the plant. Additionally, growth and development are undoubtedly affected by factors other than temperatures, such as the flow and duration of photosynthetically active radiation, the availability of nutrients and water, and the loss of photosynthetic tissue.

In a production system, which is developed under protected environments and where there is permanent availability of nutrients and water, other bioclimatic factors (i.e., temperature, luminosity, and CO₂ concentration) can be exploited to increase yield or improve the quality of the final product. Optimizing the conditions that determine maximum crop yield with a minimum expenditure of energy in the systems cultivated in open fields, are fundamental to generate economic and environmentally sustainable technology.

In general terms, eggplant is a plant that requires an adequate temperature during its cultivation to reach yields above the national average (16 t ha^-1) (^{Araméndiz, 2008b}, ^c) and fruit quality. Eggplant has non-climacteric fruits (^{Arguedas-García, 2017}). Its optimum vegetative growth occurs with temperatures between 27 and 32 °C during the day, and between 21 and 27 °C during the night, although fruit growth is favored between 22 and 26 °C (^{Passam and Karapanos, 2008}).

Day temperature should oscillate between 25 and 30 °C together with good luminosity to satisfy the needs of the plant, and with it, greater efficiency in the photosynthesis. It should be carried out in a way that its products, when being sent to other parts of the plant, are not used entirety for respiration. It is achievable with better night temperatures (14-16 °C), which are favorable for plant and fruit growth.