Harvested chickpea seeds (cv. Nacional 29) were air-dried at room temperature to 6% moisture content (fresh mass basis) and then stored for 4 months at 4°C in the dark in hermetically sealed containers. Seeds with 6% moisture content were used in subsequent experiments, as recommended by ISTA (2005).
Three seed sizes of chickpea were compared: 55 ± 3 mg per seed (group 1); 72 ± 3 mg (group 2); 89 ± 3 mg (group 3). Physical characteristics relating to germination (radicle longer than 5 mm), emergence, plant height, number of leaves, total leaf weight; stem length, diameter and weight and main root length, diameter and weight were recorded for up to 21 d. In addition, levels of leaf chlorophyll a and b (Porra 2002), total protein (Bradford 1976), superoxide dismutase (SOD) (McCord and Fridovich 1969) and peroxidase (PER) (Pascual et al. 1983) activities were also recorded. Biochemical determinations were assayed using three independent samples with 100 mg of leaf material per sample.
Chlorophyll pigments were extracted with 750 µl methanol (100%). Samples were centrifuged (at 10 000 rpm and 4°C for 15 min), supernatants were collected and the absorbance was read at 652.4 nm (Porra 2002). Protein extraction was carried out with Tris-HCl 0.1 mol l− 1 buffer, pH = 8.5–8.8, 5 mmol l− 1 EDTA and 20 mmol l− 1 β-mercaptoethanol. The total protein content was determined according to Bradford (1976).
To determine superoxide dismutase and guaiacol peroxidase-specific activities, samples were finely ground in liquid nitrogen. Extraction was performed with Tris–HCl buffer (0.35 mol l− 1, pH 8.0), EDTA (20 mmol l− 1), cysteine (15 mmol l− 1), PVPP (50%) and PMSF (0.2 mmol l− 1). Samples were homogenized on ice with polytron apparatus (Ultra-turrax T25). The homogenate was filtered through a two-folded piece of gauze and centrifuged at 10,290 xg (Beckman J-21, Palo Alto, CA) for 30 min at 4°C. Superoxide dismutase activity was measured according to McCord and Fridovich (1969). The reaction mixtures included 80 µl of plant extract and 900 µl of potassium phosphate–KOH (50 mmol l− 1, pH 7.6), EDTA (0.1 mmol l− 1); cytochrome C (0.01 mmol l− 1), xanthine (0.05 mmol l− 1) and 20 µl xanthine oxidase (EC 1.2.3.22; 0.03 units). Absorbance (550 nm) was measured every 15 s for 3 min. Averages of linear section absorbance were used and cytochrome c extinction coefficient (21.1 mmol− 1 l cm− 1) was used. Superoxide dismutase activity was defined as U g− 1 plant fresh mass [U = enzyme quantity hydrolyzing 1 µmol of superoxide per hour (37°C)]. Specific activity was calculated as the rate of superoxide dismutase activity relative to protein content.
Guaiacol peroxidase activity was measured according to Pascual et al. (1983). The reaction mixtures included 100 µl of plant material extract, 1.0 ml Tris–HCl buffer (0.01 mol l− 1, pH 7.0), 150 µl guaiacol (100 mmol l− 1) and 20 µl hydrogen peroxide. Absorbance (470 nm) was measured every 15 s for 3 min. Averages of linear section absorbance were used and guaiacol extinction coefficient (5,570 x 10− 6 µmol− 1 l cm− 1) was used. Guaiacol peroxidase activity was defined as U g− 1 plant fresh mass [U = enzyme quantity hydrolyzing 1 µmol substrate per minute (37°C)]. Specific activity was calculated as the rate of guaiacol peroxidase activity relative to protein content.
The following agronomical traits were also measured during three months of growth in the field, as recommended by the International Board for Plant Genetic Resources (FAO/IPGRI 1994): plant height, fresh and dry weight; number of branches; time until anthesis in 50% of plants; total number of pods; number of filled pods; number of grains per pod, number and weight of grains per plant; weight of 100 grains; duration of plant cycle, and yield.
Germination was assessed by placing the three seed morphs on filter paper in Petri dishes (Ø: 10 cm moistened with 15 mL of distilled water and five replicates of 10 seeds per dish). To evaluate young plantlets up to 21 d of growth, seeds were planted into pots (500 mL volume, one seed per pot) containing Ferralytic-red soil and filter-cake-sugarcane ash (1:1, v:v). There were five replicates of 10 pots for each treatment (50 seeds per treatment). To study adult plants, 90 seeds of each seed morph were randomly selected and sown in a plant bed (with Ferralytic-red soil and filter-cake-sugarcane ashes) under field conditions. Seeds were planted 70 x 25 cm apart (3 replicates of 30 seeds each) in a randomized complete block design. The experiment was conducted at the Field Experimental Station of the Bioplant Centre, University of Ciego de Ávila, Cuba (21 52´48.6´´ N, 78 41´32.6´´ W; 53 meters above sea level; Sept 2020 – Jan 2021). Temperatures at 13:00 averaged 33oC during the experiment; relative humidity reached 80%; and the Photosynthetic Photon Flux was 1 140 µmol s− 1. Technical instructions as provided by the Cuban Ministry for Agriculture to cultivate chickpea were applied. Fertilizers were not supplied. Seeds were not inoculated with Rhizobium. Microjet irrigation was used to water the plants for 5 min every 8 h. Border plants, which had more space to grow, were not considered.
All statistical analyses (One-Way ANOVA and Tukey, p = 0.05) were carried out using SPSS (Version 8.0 for Windows, SPSS Inc., New York). The overall coefficients of variation (OCV) were calculated as follows: (standard deviation/average) * 100. In this formula, we considered the average values of the three treatments compared (seed sizes) to calculate the standard deviation and average. Therefore, the higher the difference between the three materials compared, the higher the OCV (Lorenzo et al. 2015). The OCVs were classified as Low from 1.73 to 24.42%, Medium from 24.42 to 47.11% and High from 47.11 to 69.79%. The OCVs were only calculated for those indicators with statistically significant differences according to ANOVA and Tukey tests.