Causes of Low Seed Quality in Ilex paraguariensis A. St. Hil. Samples (aquifoliaceae)

This study sought to determine the possible causes of the low seed quality of I. paraguariensis A. St. Hil. Seeds from six samples collected at different sites were classified as empty, decayed, herbivorous and full. Viability was assessed by tetrazolium test in seeds filled with a visualized embryo. High amounts of empty and deteriorated seeds (54% to 93%) were verified in four of the evaluated samples, and insect attack was observed in two samples. The viability of visualized embryo-filled seeds was 83% to 100%. Low quality generally results from the presence of empty and deteriorated seeds, as well as dormancy.


INTRODUCTION AND OBJECTIVES
Yerba mate (Ilex paraguariensis A. St. Hil.) plays an important socioeconomic role in the Southern region of brazil, where it is produced. The leaves are dried to prepare a traditional tea that is consumed by millions of people in South America (Oliveira & Waquil, 2015).
Seedling production of the species is commonly performed by pyrenees (hereafter called "seeds"), which have around 5% germinability (Cuquel et al., 1994). This low germinability may be associated with physiological deformations such as absence of embryo and deterioration, as well as herbivory. Therefore, the presence of insects and empty and deteriorated seeds has been observed in Ilex species such as I. latifolia (Takagi & Togashi, 2013) and I. aquifolium (Arrieta & Suarez, 2004;Garcia et al., 2005). Thus, this study sought to determine the possible causes of the low quality of I. paraguariensis seeds.
Six seed samples were obtained from at least five matrices with a minimum distance of five meters and a maximum of five hundred meters in native areas of different locations and/or years: The fruits were ripe and with dark purple coloration -2.5/1 F: 5Y (Kollmorgen, 1975) (Figure 1). The seeds were extracted in a sieve and running water, dried on paper towels and used in the experiments.
Subsequently, 200 seeds, divided into four repetitions, were immersed in water for 24 h for each batch to facilitate longitudinal cutting. The seeds were evaluated visually with the aid of a Stemi-305® stereo microscope and classified into: Empty -without endosperm and embryo; Deterioratedwith necrotic endosperm and embryo or coalescent tissues (Catapan, 1998); Filled with unseen embryo; Filled with visualized embryo; and Herbivorous -with the presence of insect or signs of herbivory ( Figure 2).
For viability evaluation, seeds filled with a visualized embryo were immersed in 0.1% tetrazolium solution for 24 h at 35 °C (Catapan, 1998).
The experiments were installed in a completely randomized design. Data were tested for normality by the Shapiro-Wilk test and homogeneity by the Bartlett' s test, and analysis of variance was performed. Data with heterogeneous variance were transformed into arc sine √x/100. Means were compared by the Tukey test at 5% probability. The Pearson's correlation was used to correlate environmental conditions and seed sample characterization.
The characteristics of I. paraguariensis seeds differed according to the sample; however, higher percentages of decayed and empty seeds were generally observed ( Table 1).
The empty seeds (Figure 2a) were counted in greater number in the UR-16, UR-15, OC-16 and PU-16 samples with values between 27 and 44% (Table 1). These results may be related to pollination, distance between matrix trees, and the efficiency of natural pollinators (Ayub & Mariath, 1996apud Carvalho, 2003, since this is a dioecious species.  (Kollmorgen, 1975). Scale bar: 10 mm.
For the removal of empty seeds from a sample, Zanon (1988) suggests to immerse them in water and discard the supernatants. Sousa et al. (2003) observed that controlled pollination for I. paraguariensis is efficient and indicated for most of fruits and full seeds.
The PU-16 sample had the highest number of deteriorated seeds (Figure 2b and Table 1), differing from the others. Although the presence of pathogens was not evaluated, Oliveira (2013) reports that seed deterioration may be fungi-related.
Regarding herbivory, micro-hymenopterans belonging to the family Torymidae were identified (Figure 2c) by isolating the insect in OC-15 and UR-15 samples. These microwaps feed on the seed, and then exit through a hole that could be visually identified in the seeds (Figure 3).
In I. paraguariensis seeds from vintage 1982, provenances São Mateus do Sul-PR, Catanduvas-SC and Centenario-RS, Zanon (1988) verified the presence of unidentified micro hymenopteran, which caused approximately 50% of damaged seeds. The presence of full seeds with or without visualized embryos (Figures 2d and 2e) ranged from 6.11 to 92.03%, according to the sample (Table 1).
Correlation between climatic factors and seed quality was observed (Table 2). There was a strong linear correlation between empty seeds and altitude, showing that the higher the altitude, the greater the number of empty seeds. The relationship between temperature and rainfall was inverse, that is, the lower the temperature and rainfall, the higher the number of empty seeds. The correlation between temperature and empty seeds can be explained by pollination as it is entomophilous, and low temperatures may decrease insect visits. When evaluating pollination in I. paraguariensis, Pires et al. (2014) and Liebsch & Mikich (2009) observed there was greater visitation of flower pollinators in the periods with higher temperatures during the day. I. paraguariensis flowering occurs in the warmer months after winter, with average temperatures above 13 °C (Pires et al., 2014). Most of the seeds filled with visualized embryo were considered viable by the tetrazolium test (Table 1). Although viable, I. paraguariensis seeds are classified as dormant (Niklas, 1987;Heuser, 1990;Galíndez et al., 2018).
Based on the results, the low germination reported in I. paraguariensis seeds is not only related to dormancy, but also to factors that influence seed formation, considering the high number of empty/deteriorated seeds observed, in addition to the low temperature and higher altitude observed at the collection sites correlated with the high number of empty seeds.