1. Introduction
A main tool to improve yield stability, crop resilience, food and nutritional security with changing climatic conditions is crop diversification with traditional, indigenous, or underutilized crops [
1,
2]. To achieve the Sustainable Development Goals (SDGs), which aim to balance the demands on land and food production (SDG 2) as well as biodiversity (SDG 15), maximizing the potential of underutilized crops is crucial, because these crops can be productive under marginal farming conditions with greater returns on ecological services [
3]. Opportunities for strengthening the management of underutilized crops and their seed quality are paramount, considering the current climatic variability that frequently has a deleterious effect on seed establishment and performance.
A main underutilized crop is okra (
Abelmoschus esculentus L.). It is a multi-use crop due to its edible pods, fresh leaves, stems and seeds. It is a valuable source of micronutrients, such as vitamins, calcium, iron, magnesium, proteins, antioxidants and is a rich source of calories [
4]. Okra requires little agricultural input, is resilient to climatic change, and thrives well in degraded soil or marginal lands [
5,
6]. It is a tropical and subtropical crop and prefers temperatures of 20–30 °C for growth, flowering and fruiting [
7]. Okra flowers can be self-pollinated, but cross-pollination by insects can improve effective fruiting, yields and the seed quality of okra [
8].
Okra’s floral physiology (bilaterally symmetrical whitish-yellow flowers with sticky pollen) encourages insect pollination rather than wind pollen dispersion [
8]. In the absence of insect pollination, the crop still produces seeds through self-pollination. Perera and Karunaratne [
9] reported that the deprivation of okra from pollination reduced seed production and may affect seed set and seed quality leading to lower seed viability.
Currently, okra cultivation and production are low in South Africa due to its semi-arid growing environment [
10,
11]. Van der Walt and Fitchett [
12] reported a pronounced seasonal difference in the soil temperature affecting cropping practices in South Africa. A shortage of improved okra seeds, poor establishment, poor germination, and unfavourable soil surface temperature are among the major factors limiting okra productivity in South Africa [
13]. A poor and non-synchronous germination is caused by delayed permeability through seed-coat-imposed dormancy, causing imperviousness of seed coats to water and gases, which complicates the management of okra production.
Several studies have aimed to overcome seed dormancy hurdles in okra through seed stratification, or chemical and mechanical scarification [
14]. Ebert et al. [
15] reported that the percentage of hard seeds in okra increased significantly in all cultivars with an increase in seed maturity. There is limited knowledge about the effects of insect pollination and seed coating with organic growth stimulation substances, such as biochar, on improving seed germinability and vigour at varying temperatures. Seed coating can stimulate germination and the early growth of seedlings under adverse weather conditions, and can reduce the impact of environmental stresses by improving water availability [
16].
Biochar is a carbon-based product obtained by slow pyrolysis of organic materials at high temperatures (300–450 °C) under low or no oxygen conditions [
17]. Biochar seed coating is a cheap technique that makes use of local materials, while mineral or organic fertilizer could be optionally added to enhance nutrient supply to the seedling [
18]. Biochar contains smoke derivatives and growth stimulators compounds called Karrikin, which provide strong chemical cues to stimulate seed germination and seedling growth [
19] through improved soil physical properties and soil water-holding capacity [
20]. Smallholder farmers save seed from the previous season and apply seed coating with biochar [
21]. Seed coating offers non-germinated seeds protection against adverse environmental conditions, such as variation in soil temperature, and it provides nutrients to seedlings [
22], stimulates root development, and helps to prevent seed predation by insects and birds. The beneficial role of ecological service providers, such as pollinators, on the seed quality of underutilized crops, such as okra, are rarely reported in the literature [
1].
Therefore, this study focuses on ways to improve seed germination and seedling vigour in okra, through the indigenous practice of seed coating. Moreover, we aim to assess the benefits of insect pollination on seed quality. We model the variation in soil temperature experienced during the okra seed establishment stage by using varying controlled temperature conditions. The objective of this study is to examine the beneficial impacts of insect pollination and biochar coating on okra seed germination and vigour performance under varying temperature conditions.
2. Materials and Methods
This study presents the results from two experiments: a seed germination and an emergence experiment.
2.1. Seed Germination Experiment
The germination experiment was conducted in a growth chamber of the Department of Soil Crop and Climate Sciences, University of the Free State, South Africa. For the germination experiment, a 2 × 3 × 3 full factorial experimental design with four replications was used [
23] to identify the impact of seed coating, pollination and temperature on germination variables.
2.1.1. Seed Source
The seeds were obtained from a field trial with okra at Kenilworth Farm, 15 km northwest of Bloemfontein, the Free State, South Africa (−29°06′ S; 26°15′ E, 1350 m asl). Part of the original okra seed for planting was saved. To obtain non-pollinated seed, a pollinator exclusion approach was used, which entailed the complete exclusion of the flower from insects. Exclusion was ensured using a tulle cloth bag with a 1mm mesh size covering the pod nodes before flowering, while other plant parts, such as leaves, were excluded. As the pods expanded during flowering, the bags were periodically adjusted to avoid contact with the florets. The flowers were labelled at the time of anthesis and four samples of 10 pods each were selected from the same plant (node) position, and manually harvested at 65 days after anthesis (equivalent to 17 WAP). Pollinated seeds were obtained by allowing insects to freely visit the flowers. The pods from both treatments were spread on glasshouse benches and air-dried. The seeds were manually threshed and air-dried to <12% moisture content. The mean hundred seed weight was 5.98 g for the pollinated seeds and 4.64 g for the non-pollinated seeds.
2.1.2. Preparation of Biochar
Pine biochar was obtained from C Fert™ Johannesburg, South Africa, which has a similar carbon content as other wood-derived biochar [
24]. The raw material was pinewood (15–25% moisture content; pyrolyzed at 400 to 450 °C). The biochar pellets were finely ground using a coffee blender (Safeway model SBCS167, South Africa). One gram of dried powdered biochar was dissolved in 10 mL distilled water, which was equivalent to 10% (
v/
v) concentration. The mixture was shaken for two hours at 120 rev. min
−1 in the dark, then filtered and the filtrate was used as aqueous biochar for watering the germination paper and seedlings during seed germination and emergence [
20].
2.1.3. Chemical Analysis
Three random samples from the powdered biochar were taken for chemical analyses according to the methods of the Soil Fertility Analytical Services Section, Department of Agriculture and Environmental Affairs, KwaZulu-Natal, South Africa. The pH and electrolytic conductivity (EC) were determined with Combo pH and EC waterproof meters HI 98,130 (Hanna Scientific, Woonsocket, RI, USA) respectively. The mineral nutrients, exchangeable K
+, Ca
2+,, Mg
2+, Fe
3+, Cu
2+, and Zn
2+ were analysed using an atomic absorption spectrophotometer (AA-7000 Shimadzu, South Africa) with the specific wavelength set for each mineral element [
25,
26]. The available phosphorus content was determined via the Lancaster method using a spectrophotometer (UV-1800; Shimadzu, Kyoto, Japan). The automated Dumas dry combustion method was used to determine total C and N, and organic carbon was determined using the Walkley–Black method [
25].
2.1.4. Seed Germination Assay
Pollinated seeds, non-pollinated seeds, and the original planting seed from the field trial (saved seed) were used. Saved seed was included to model the smallholder farmers’ practices where saved seeds are used to propagate next season’s cultivation. Seed coating treatments consisted of biochar-coated seed and uncoated seeds. Two hundred seeds from each seed source were primed with 50 mL distilled water (non-coated) or 10% biochar solution (coated seed) (see
Section 2.1.2). The seeds were primed for 6 h. Subsequently, the seed coating was strengthened with biochar in the ratio 1:5 (seed weight: biochar). The seed coating was air-dried for 12 hours to reduce the moisture content before use.
Germination assays were carried out according to International Seed Testing Association protocols [
27] (ISTA 2020). This was conducted in a growth chamber of the Department of Soil, Crop and Climate Sciences, University of the Free State, South Africa. One hundred seeds from each seed lot were taken, and 25 seeds were arranged in a paper towel and rolled, giving four replications. The rolled paper towels were placed in sealed plastic bags to avoid moisture loss and incubated in Labcon growth chambers (Labcon laboratory Equipment Germany L.T.I.E.) at a temperature regime of 25/16 °C (low temperature), 30/25 °C (optimal temperature), and 35/30 °C (extreme temperature) for 12 light/12 dark hour cycles for 7 days.
A daily count of germinated seeds was conducted with germination defined as radicle protrusion of at least 2 mm. The final germination percentage represented the percentage of normally germinated seeds on the 7th day. In addition, seeding morphological indices, such as shoot length (SL) and root length (RL), were manually obtained with a ruler on the 7th day. Root and shoot dry weights were determined after drying the samples at 80 °C for 24 h.
Mean time to germination (MGT) was calculated according to Bewley and Black (1994) as follows:
where: MGT = mean germination time, F = the number of seeds completing germination on day X, X = number of days counted from the beginning of the germination test, I = day one to day j and j = final day of germination.
Germination velocity index (GVI) which is a measure of the speed of seed germination was calculated according to Maguire (1962) as follows:
where: GVI = germination velocity index, G1, G2…Gn = number of germinated seeds in first, second… last count, and T1, T2…Tn = number of sowing days at the first, second… last count.
Seedling vigour: At the final count (14 days), all seedlings that had complete morphological parts without lesions or defects were selected as vigorous seedlings, and the average seedling length and weight of seedlings were measured for calculating the seedling vigour index as [
28]: VI = germination (%) × average seedling length (mm).
2.2. Seedling Emergence Experiment
The emergence experiment was conducted in a climate-controlled growth cabinet of the Department of Soil Crop and Climate Sciences, University of the Free State, South Africa. The experimental design for the emergence assay was a 2 × 3 full factorial experimental design [
23] with two independent variables: seed coating (two levels) and seed source (three levels). Two seeds per pot were sown in air-dried sieved soil pots (10 × 10 × 12 cm). Each treatment consisted of eight pots and was replicated four times. Seed coating (biochar-coated seed and uncoated seed) and seed source (pollinated seed, non- pollinated seed and saved seed) were the factors. The pots were placed in a climate-controlled growth cabinet (Conviron E15; Controlled Environments) with a long light period of 16 h, a photosynthetic photon flux density of 350 µmol·m
−2 s
−1, a day/night temperature of 30/25 °C, and relative humidity of 60%. Temperature and relative humidity in the chamber were continuously monitored by Conviron series controllers (CMP3243 Controlled Environments Ltd., Winnipeg, MB, Canada). The pots were rotated three times a week to ensure uniform growing conditions in the growth chamber. Before the experiment, the field capacity of the soil (FC) of each pot was evaluated using the formulae below:
Throughout the experiment, the water application rate was determined by measuring the individual pot weight.
A daily count of emerged seeds was conducted. In addition, 21 days after the start of the experiment, seedling morphological indexes, such as shoot length (SL) and root length (RL), were measured. Root and shoot dry weights were determined. Mean time to emergence (MET) was calculated according to [
29] as follows:
where: MET = mean emergence time, F = the number of seeds completing emergence on day X, X = number of days counted from the beginning of the emergence test, I = day one to day j and j = final day of emergence.
Emergence velocity index (EVI) which is a measure of the speed of seed emergence was calculated according to [
30] as follows:
where: EVI = emergence velocity index, E1, E2…En = number of emerged seeds in first, second… last count, and T1, T2…Tn = number of sowing days at the first, second… last count.
Physiological parameters of the plants were measured at 14 and 21 DAP: leaf chlorophyll content index (CCI), plant height (PHT), leaf number (LN). The CCI was measured using a portable SPAD meter (SPAD-502-PLUS chlorophyll meter, Konica Minolta, Ramsey, NJ, USA) on the adaxial leaf surface.
2.3. Statistical Analysis
Seed germination characteristics were analysed by a three-way analysis of variance, with temperature regime, seed coating and pollination status as factors, using the IBM SPSS Statistics package 25.0 (IBM Corp., Armonk, NY, USA). First, the data were tested for homogeneity of variance by Levene tests. Similarly, seedling emergence variables were analysed by a two-way analysis of variance, with seed coating and seed source as factors. The interaction effects of seed coating and seed source on leaf chlorophyll content index and plant growth were determined through a two-factor analysis of variance. The differences between means were assessed with Tukey’s test using a significance level of 0.05. The P-values highlighted in bold denote significance levels at p < 0.05. Linear regression was performed on XLSTAT. Principal component analysis (PCA) was constructed through SPSS by using a correlation matrix.