Introduction

Tomatillo (Physalis ixocarpa Brot. ex Horm.) is produced in 30 states of the Mexican Republic. In 2020, it occupied a total planted area of 40,116.97 ha and had an average yield of 19.28 t·ha^-1 (^{Sistema de Información Agroalimentaria y Pesquera [SIAP], 2021}). Currently, tomatillo is the sixth most widely planted vegetable crop in the country. Due to its widespread cultivation, both for domestic consumption and export, its seed is in great demand. Despite the importance of this crop, little research has been carried out on seed physiology related to drying, whose purpose is to reduce costs, preserve physiological quality and make drying processes efficient. Some aspects to consider in this process are temperature, air speed, thickness of the seed layer, drying technique, and aeration (^{Coronado-González, Peña-Lomelí, Magaña-Lira, Sahagún-Castellanos, & Ybarra-Moncada, 2019}; ^{Martínez-Solís, Peña-Lomelí, & Montalvo-Hernández, 2004}).

Tomatillo, being a fleshy fruit, has seeds with a high moisture content. Therefore, it is recommended that they be subjected to the drying process immediately after being extracted from the fruit to avoid fermentation, which causes immediate damage. It has been determined that at 55 days after pollination, the tomatillo seed can be considered physiologically mature, that it rapidly loses viability with storage temperatures above 30 °C, and that drying increases its germination capacity, regardless of its stage of development (^{Pérez-Camacho, González-Hernández, Molina-Moreno, Ayala-Garay, & Peña-Lomelí, 2008}; ^{Pérez-Camacho et al., 2012}).

Drying is a fundamental process in the production of high quality seeds, since it reduces the amount of water to adequate levels and enables the preservation of quality during storage, in addition to avoiding physical and chemical alterations induced by excess moisture in the seed (^{Baudet, Villela, & Cavariani, 1999}). The drying process can be carried out in various ways; however, it requires special care in terms of the temperature used.

Depending on the species, exposure time and drying method, high temperatures can damage cell membranes, denature proteins and cause cracks. Such damage causes a reduction in the physical and physiological quality of seeds, either after drying (immediate effect) or during storage (latent effect) (^{Pérez-Camacho et al., 2008}, ²⁰¹²). One of the objectives of storage is to preserve the quality of the seed and reduce the deterioration process, which is influenced by the initial physiological conditions of the seeds, physical damage, storage conditions, type and incidence of pathogens, and the joint action of these factors, which can cause different behaviors among lots of stored seeds (^{Villela & Peske, 2003}).

At present, the tomatillo seed drying process is carried out under environmental conditions, either sun-dried or shade-dried (^{Martínez-Solís et al., 2004}), so it is necessary to generate information that helps overcome the problems inherent to this process, since there are few studies related to drying and storage, in the context of commercial seed production. Therefore, this research aimed to evaluate the effects of eight temperatures during the drying of seeds of four tomatillo varieties, stored for three periods, on their physical and physiological quality.

Materials and methods

The research was carried out from June to December 2014 in the Ecology and Seed Analysis laboratories of the Plant Science Department of the Universidad Autónoma Chapingo (UACh), Mexico. Tomatillo seeds produced in the spring-summer 2014 growing season were used in the study. The Diamante, Gema and Tecozautla 04 varieties were produced in isolated lots at the UACh Experimental Agricultural Field, and the Manzano Tepetlixpa variety at the Colegio de Postgraduados, Campus Montecillo. Transplanting to the open field was carried out in the first week of April, at a density of 30,000 plants·ha^-1 (^{Coronado-González et al., 2019}). The fruit harvest was carried out in the second week of June, 60 days after full flowering, and the seeds were extracted manually (^{Pérez-Camacho et al., 2008}).

The study factors were variety (Diamante, Gema, Manzano Tepetlixpa and Tecozautla 04; ^{Servicio Nacional de Inspección y Certificación de Semillas [SNICS], 2021}), drying temperature (ambient, 30, 35, 40, 45, 50, 55 and 60 °C) and storage time (zero, two and four months). The parameters of the drying process and physical quality of the seeds were studied under a full factorial treatment design, product of the combinations of the four varieties with the eight temperatures (32 treatments), in a completely randomized experimental design with four replications. The experimental unit consisted of 5 g of wet and drained seeds for the application of the different drying temperature levels. The control seeds were dried under laboratory conditions, without direct sunlight and with adequate ventilation. The temperature treatment ended once seed weight became constant. The environmental conditions during the drying period were recorded by means of a sensor (Hobo® HO8-004-02, Onset®, USA) and were as follows: temperature of 24.8 ± 2.71 °C, relative humidity of 23.76 ± 0.38 %, dew point temperature of 2.73 ± 2.5 °C and absolute humidity of 5.48 ± 0.98 g·m^-3.

The physical characteristics of the seeds were quantified before, during and at the end of the drying process. Initial (ISM) and final seed moisture (FSM) percentages were calculated using the formula of ^{Bewley and Black (1994)}. Seeds were dehydrated for 72 h at 72 °C in an oven (Stabletemp 52120-02, Cole-Parmer®, USA) and then the dry weight was determined. The drying time of each seed lot was recorded until they reached constant weight. Seeds were weighed every 2.5 h at room temperature levels up to 50 °C, and every 1.5 h for 55 and 60 °C. With the weight data of the samples at the beginning and end of the drying process, the percentage of weight removed from the seeds in the form of water by drying (PWRD) was calculated.

At the end of the temperature treatments, the seeds were sorted by means of a pneumatic separator (Seedburo®, USA), with an air speed of 9 m·s^-1. Seeds were stored in satin paper envelopes, under laboratory conditions and without direct sunlight. These seeds were used in subsequent vigor and germination tests. The environmental storage conditions of the seeds were recorded with the same type of sensor mentioned previously, and were as follows: temperature of 23.43 ± 0.64 °C, relative humidity of 23.55 ± 0.06 %, dew point temperature of 1.46 ± 0.56 °C and absolute humidity of 4.95 ± 0.18 g·m^-3.

To determine the physiological quality parameters of the seeds, a full factorial treatment design was established, product of the combinations of the four varieties with the eight temperatures and three storage periods. The experimental design was completely randomized with four replications. The physiological quality of the seeds of the 96 treatments was evaluated using standard germination and vigor tests (^{International Seed Testing Association [ISTA], 2004}).

In the germination tests, the experimental unit consisted of a Petri dish with a diameter of 95 mm and a depth of 10 mm, with 100 seeds evenly distributed on filter paper moistened with 5 mL of distilled water. In this test, the experimental units were placed in a germination chamber (Seedburo D-7140, Seedburo®, USA) at 30 ± 1 °C, 90 % relative humidity (^{Martínez-Solís, Mendoza, Rodríguez-Pérez, Peña-Lomelí, & Peña, 2006}) and darkness for 21 days according to ^{ISTA (2004)} standards. The necessary irrigations were carried out with distilled water. The variable evaluated during this test was the germination percentage, which was quantified based on the number of germinated seeds in relation to the total. Counts were made after the first (G1), second (G2) and third week (G3), as well as the total test (TG), which was the sum of the three weeks.

Vigor tests were performed according to ^{ISTA (2004)} standards. The experimental unit consisted of a Petri dish with a diameter of 135 mm and a depth of 22 mm, with 50 seeds evenly distributed on filter paper moistened with 10 mL of distilled water. The boxes were placed in a germination chamber (Precision 818, Thermo Scientific®, USA) at 30 ± 1 °C (^{Martínez-Solís et al., 2006}), under 24 h illumination during the 14 days of the test. Irrigations were applied daily with distilled water. Germinated seeds were counted every 48 h, considered as such when there was radicle emergence. The variables evaluated in the test were the germination speed index (GSI; calculated according to Maguire's proposal [1962] ), average seedling length (SL; at the end of the vigor test, a random sample of 10 seedlings per experimental unit was taken, their length was measured from the tip of the primary root to the tip of the cotyledonary leaves and the average in centimeters was obtained) and average seedling dry weight (SDW; seedlings evaluated in the previous parameter were dried to constant weight for 72 h at 72 °C in an oven [Wisconsin, Memmert, USA]; subsequently, they were weighed on an analytical balance [Oahus Pioneer, Oahus Corp, USA] and average seedling weight in milligrams was obtained).

Percentage data were transformed using the arcsine formula y/100 (^{Sokal & Rohlf, 1995}). Analysis of variance and Tukey's comparisons of means test (P ≤ 0.05) were performed for the factors studied and their interactions. In addition, Pearson's linear correlations between pairs of variables were obtained. Statistical analyses were performed with SAS version 9.0 software (^{SAS Institute Inc., 2004}).

Results and discussion

Analyses of variance indicate that the variety factor affected (P ≤ 0.01) ISM and PWRD, and had statistical significance in FSM (P ≤ 0.05). Temperature treatments affected (P ≤ 0.01) ISM, FSM and PWRD. The Variety x Temperature interaction was only significant in ISM and PWRD (P ≤ 0.05).

The results of the comparison of means among varieties (Table 1) showed that the four had different ISM (P ≤ 0.05), where the Manzano Tepetlixpa variety had the highest value (49.4 %) and Tecozautla 04 the lowest (43.1 %). These results can be attributed to the genetic condition of the varieties, since they come from different races, so they probably have different degrees of maturity and chemical composition (^{Peña-Lomelí, Ponce-Valerio, Sánchez-del Castillo, & Magaña-Lira, 2014}). The Manzano Tepetlixpa, Tecozautla 04 and Gema varieties had the highest FSM values, with 6.7, 6.6 and 6.4 %, respectively. In contrast, Diamante had the lowest FSM value (5.8 %), statistically less than Manzano Tepetlixpa (P ≤ 0.05).

PWRD had the same behavior as ISM; that is, the four varieties were different (P ≤ 0.05). Manzano Tepetlixpa seeds had the greatest weight loss (47.0 %), while Tecozautla 04 seeds had the least weight removed during the process (40.3 %). Diamante and Gema had intermediate values of 43.8 and 42.5 %, respectively. The amount of water removed from the seeds is related to temperature, relative humidity, air flow and drying method. However, these conditions were the same for all four varieties. What best explains the observed phenomenon among varieties is the characteristics of each one, such as the genotype, stage OF M, water content, membrane permeability, physical composition of the seed lot, and chemical composition (^{Camacho-García, Albuquerque-Barros, Teichert-Peske, & Lemos-de Menezes, 2004}). This has been demonstrated in Physalis by ^{Pérez-Camacho et al. (2008)}, ^{Pérez-Camacho et al. (2012)} and ^{Pichardo-González et al. (2010)}.

Comparison of means among drying temperatures (Table 2) showed that seeds dried at ambient temperature and at 30 °C had the highest moisture content at the end of the process (FSM), with 9.3 and 8.2 %, respectively (P ≤ 0.05). The driest seeds and those that lost the most weight at the end of the test were those exposed to 45 °C or higher (43.4 - 45.3 %), while seeds dried at 35 and 40 °C lost the least weight. Seeds dried at 60 °C reached constant weight at 255 min after the start of the test, 5.83 times faster than those dried at ambient temperature. It was observed that with increasing temperature, the drying time decreased, due to the higher rate of water removal. This coincides with what was reported by ^{Christ, Corrêa, and Mantovani-Alvarenga (1997)}, who state that the higher the temperature, the faster the drying speed, which leads to greater damage.

The PWRD results in the four varieties (Figure 1a) showed that Manzano Tepetlixpa was superior in terms of seed weight loss, while Tecozautla 04 was the one that lost the least, similar to what occurred in the comparison of means by variety. It was also observed that when drying the seeds at 35 °C the Diamante, Gema and Tecozautla 04 varieties lost about 40 % of their weight, while Manzano Tepetlixpa was superior to them, losing 4 % more. At 35 °C, the four cultivars lost less weight, and from this temperature onwards the water loss is greater, which corresponds to a shorter drying time, but also with lower germination (Tables 2 and 4). This indicates that a standard seed drying process of the four cultivars should be carried out at 35 °C, as drying temperatures ≥ 45 °C drastically reduced the physiological quality parameters of the seeds.

Figure 1 Behavior of the seeds of four tomatillo (Physalis ixocarpa Brot. ex Horm.) varieties during the drying process with eight temperatures. a) Percentage of weight removed in the form of water (PWRD), b) germination percentage after the first week (G1), c) total germination percentage (TG) and d) germination speed index (GSI).  

Regarding physiological quality, the results of the analysis of variance showed a significant effect (P ≤ 0.01) of the factors variety, drying temperature and storage period on six of the seven variables studied. The interaction of variety with temperatures was significant (P ≤ 0.05) in four of the seven variables, and with storage time only in seedling length.

The results of the comparison of means among varieties (Table 3) showed that Tecozautla 04 and Gema had the best G1 (78.9 and 76.4 %, respectively), while Diamante had the lowest value (51.4 %). In the second week, Diamante, Gema and Manzano Tepetlixpa had the highest G2 values (5.6, 5.0 and 4.5 %, respectively), and Tecozautla 04 the lowest (4.2 %), although it was only statistically lower than Diamante. For G3, there were no significant differences among varieties (data not shown). Gema and Tecozautla 04 had the highest TG (82.2 and 83.7 %, respectively), and these varieties, together with Manzano Tepetlixpa, also had the highest SDW (0.960, 0.945 and 0.942 mg, respectively). Tecozautla 04 also had the highest values in the variables GSI (15.7) and SL (5.2 cm), while the Diamante variety had the lowest performance (P ≤ 0.05) in TG (57.9 %), GSI (9.5) and SDW (0.8 mg), and numerically in SL (4.4 cm).

Since the seeds were produced under the same conditions, the difference in vigor behavior among the varieties can be attributed to, among other factors, their genetic condition, since the cultivars studied come from different races and, therefore, are genetically different. Manzano Tepetlixpa is from the Manzano race, Diamante is a product of the Puebla x Rendidora races, and Tecozautla 04 and Gema are from the Puebla race (^{Peña-Lomelí et al., 2014}). The origin of the varieties could influence germination due to the different degree of breeding, and even differences in earliness (^{Coronado-González et al., 2019} ), which generates seeds with different stages of maturity, which can be reflected in the germination values. In this regard, chemical composition can influence vigor, drying speed and growth potential (^{Camacho-García et al., 2004}). In this context, ^{Pivotto-Bortolotto, Lemos de Menezes, Camacho-Garcia, and Matheus-Mattioni (2008)} state that differences in vigor associated with seed characteristics are generally attributed to chemical composition, mainly in relation to the amount of reserves or metabolic efficiency.

Comparison of means for the effect of temperatures (Table 4) showed that seed drying at 30 °C favored germination (P ≤ 0.05) during the first week of the test, as values 7.09 and 9.94 % higher than those reached with seeds dried at ambient temperature and 35 °C, respectively, were obtained. Seed drying at 30 °C also favored TG and the GSI, this because they presented the highest values, which were above those obtained with ambient temperature by 5.69 % and 3.19, respectively (although statistically equal in TG). Seeds dried at ambient temperature and at 30 °C maintained a germination percentage higher than 85 %, which is the minimum established by SNICS for the certification of seeds for sowing (^{SNICS, 2014}).

The G1 mean was 69.39 %, and drying the seeds at 45 °C or higher resulted in lower values. The same phenomenon occurred with the variables TG, GSI and SDW. This indicates that the higher the drying temperature, the lower the seed quality, regardless of the cultivar. This coincides with what was reported by ^{Lemos-de Menezes, Lourenço-Pasqualli, Piccinin-Barbieri, Duarte-Vidal, and Massuquini-Conceição (2012)}, who evaluated drying in rice seeds. ^{Gowda, Talukdar, and Ramaiah (1992)} established that to obtain germination greater than 70 % in tomato (Solanum lycopersicum L.) seeds, the maximum permissible drying temperature is 45 °C. ^{Rosa, Pinho, Vieira, and Veiga (2000)}, when studying the damage caused by drying in maize seed, point out that at 45 °C, the highest drying temperature in their study, damage could occur in the cell membrane systems, which reduced their ability for restoration during imbibition and, consequently, affected the release of solutes into the medium.

Excessive heating of seeds during drying, for prolonged periods of time or at high temperatures, can cause damage, such as a reduction in the percentage and speed of germination, formation of abnormal seedlings, ruptures inside the seed and testa, and alteration of the coloration. The extent of damage depends on the interaction between temperature, exposure time and seed water content (^{Nellist & Hughes, 1973}).

The Variety x Temperature interaction had an effect on the variables G1, TG, GSI and SDW, and indicated that the behavior of the cultivars varied with the drying temperatures. Figure 1b shows that when seeds were dried at 30 °C, the Gema, Tecozautla 04 and Diamante varieties had the highest G1 (90 %), while the Manzano Tepetlixpa cultivar lagged behind with 78 %. Manzano Tepetlixpa reached its maximum germination (80 %) when dried at ambient temperature. On the other hand, when seeds were dried at 35 °C, G1 was similar in the four varieties (80 %).

Regarding TG (Figure 1c), it was observed that when the seeds were dried at 35 °C the germination of the four varieties was similar (from 82 to 86 %), as in the variable G1, but when they were dried at 30 °C, the Diamante, Gema and Tecozautla 04 varieties reached maximum germination (> 90 %), while the Manzano Tepetlixpa cultivar was below that level with 84 %. Likewise, it was observed that Diamante was the most sensitive to the high drying temperature, since above 40 °C its germination percentage decreased drastically, which coincided with the maximum drying temperature of 45 °C established by ^{Gowda et al. (1992)} for tomato (Solanum lycopersicum L.) seeds.

Figure 1d shows that drying at 30 °C favored the GSI, as the Tecozautla 04, Gema and Diamante cultivars reached the highest values (22, 20 and 19, respectively), while Manzano Tepetlixpa recorded 15.5. However, Manzano Tepetlixpa reached a GSI of 17.5 at 35 °C, while the others decreased considerably. As the drying temperature increased, the GSI of the four cultivars decreased. These results coincide with those reported by ^{Höfs, Braga-Schuch, Teichert-Peske, and Albuquerque-Barros (2004)}, who state that high temperatures negatively influence the expression of seed vigor, which affects the initial growth of seedlings and their ability to accumulate biomass. In the GSI, the behavior of Diamante was similar to its performance in TG, since after 40 °C its values decreased until they reached 1 (when dried at 60 °C). This coincides with what was observed by ^{Lemos-de Menezes et al. (2012)}, who point out that the higher the drying temperature, the greater the decrease in the physiological quality of the seeds.

The comparison of means of the storage period (Table 5) shows that storing the seeds for two months favored G1, with values 5 and 8 % higher (P ≤ 0.05) than those obtained with seeds kept at zero and four months, respectively; GSI and TG also improved significantly (P ≤ 0.05), although in this latter variable the value was statistically equal to that of the zero month of storage. Considering the second month of storage, at the fourth month there was a drop in the values of the variables G1, TG and GSI, and an increase in G2, which represents a decrease in physiological quality. The above behavior indicates the presence of a possible dormancy period in the seeds, which in this research disappeared after two months (60 days) of storage. This behavior is also reported by ^{Marín-Sánchez, Mejía-Contreras, Hernández-Livera, Peña-Lomelí, and Carballo-Carballo (2007)} in tomatillo, and ^{Randle and Homna (1981)} in pepper (Capsicum annum L.).

Storing tomatillo seeds for two months also improved their SDW, compared to unstored seeds. ^{Martínez-Solís et al. (2006)} explain this result by relating it to a higher germination percentage, which in turn allows a higher germination speed. That is, the seedlings were more quickly exposed to light and, consequently, had more time to photosynthesize, which was reflected in the increase in SDW. For the SL trait, the highest values were obtained without storage, which is explained by the greater amount of reserves at zero months of storage (^{Lemos-de Menezes et al., 2012}).

G1 had a high correlation (P ≤ 0.05) with the variables TG, GSI, SDW and SL. This indicates that seeds with the best G1 achieved a higher germination speed and, therefore, had more time to carry out photosynthesis, which is reflected in the increase in SDW and SL (^{Martínez-Solís et al., 2006}). This also explains the correlations of TG and SDW with GSI, and of SL with SDW.

Conclusions

Seed drying at 30 °C favored the expression of the physiological quality characteristics of the Gema, Tecozautla 04 and Diamante varieties, while Manzano Tepetlixpa performed better at 35 °C. Drying temperatures equal to or above 45 °C drastically reduced the physiological quality parameters of tomatillo seeds. A standard process for drying the seeds of the four varieties tested should be done at 30 °C. After two months of storage, the seeds expressed the maximum values in terms of physiological quality characteristics.