Honeybush (Cyclopia spp.) pollen viability and surface morphology

Honeybush (Cyclopia spp.) pollen viability and surface morphology

South African Journal of Botany 128 (2020) 167 173 Contents lists available at ScienceDirect South African Journal of Botany journal homepage: www.e...

2MB Sizes 0 Downloads 27 Views

South African Journal of Botany 128 (2020) 167 173

Contents lists available at ScienceDirect

South African Journal of Botany journal homepage: www.elsevier.com/locate/sajb

Honeybush (Cyclopia spp.) pollen viability and surface morphology J. Koena,b,*, M.M. Slabbertb, M. Booysec, C. Bestera a b c

Agricultural Research Council Infruitec-Nietvoorbij, Private Bag X5026, Stellenbosch 7599, South Africa Department of Horticulture, Tshwane University of Technology, Private Bag X680, Pretoria 0001, South Africa Agricultural Research Council Biometric Services, Private Bag X5026, Stellenbosch 7599, South Africa

A R T I C L E

I N F O

Article History: Received 25 March 2019 Revised 28 October 2019 Accepted 3 November 2019 Available online xxx Edited by D Honys Keywords: honeybush pollen storage viability SEM

A B S T R A C T

Genotypes of Cyclopia, used to produce honeybush tea, are being improved through selective breeding for suitability in large-scale cultivation and consistent flavour/aroma qualities. Though high levels of natural fecundity have been reported in honeybush species, attempts at controlled crosses have hitherto resulted in virtually no seed set. While investigating this problem, it was found that many basic questions regarding the specific reproductive physiology of the genus have not been answered. Studies were undertaken to provide the necessary data about sexual reproduction in honeybush species, in support of existing breeding programmes. This paper presents data concerning pollen viability in relation to flower bud stage, storage and pollen surface morphology of C. longifolia, C. maculata and C. subternata. Pollen anthesis and viability (measured as in vitro pollen germination percentage - PGP1) were linked to visual stages of flower bud development. Pollen samples, dried and stored for 540 days at -18 °C, were tested for PGP at intervals (7, 30, 90, 180, 365, 540 days) during the storage period. While there was an overall decrease in PGP over storage time, four genotypes did not have any significant decrease in PGP over time. The effect of media sucrose content (MSC2) (0 25%) on in vitro PGP of honeybush pollen was tested and differences in response were observed between species. Pollen surface morphology was investigated using FE-SEM microscopy and no consistent differences in shape or surface ornamentation were observed between the species, though differences in grain size were observed. © 2019 SAAB. Published by Elsevier B.V. All rights reserved.

1. Introduction The genus Cyclopia, endemic to the fynbos region of South Africa, falls under the taxonomic family Fabaceae (legume), subfamily Papilionoideae (Faboideae) (Schutte, 1997). A number of honeybush (Cyclopia) species are harvested and processed to make a sweetly fragrant, herbal tisane. Because of the gaining popularity of this beverage, commercial production has become necessary to protect wild populations from over-exploitation (Bester, 2013). Breeding programmes have been established to produce improved genetic material for the honeybush industry, but not much is known about the specific physiology and propagation requirements of honeybush species. The present study investigated pollen viability in relation to flower bud stage and storage, as well as pollen surface morphology of three honeybush species. Floral studies form a branch of descriptive botanical science applicable to various scientific pursuits such as the delineation of taxonomy, the analysis of fecundity, the selection of breeding techniques,

* Corresponding author at: Private Bag X5026, Stellenbosch 7599, South Africa. E-mail address: [email protected] (J. Koen). 1 Pollen germination percentage 2 Media sucrose content https://doi.org/10.1016/j.sajb.2019.11.004 0254-6299/© 2019 SAAB. Published by Elsevier B.V. All rights reserved.

and the understanding of ecological interactions (Cordoba and Cocucci, 2011; Jia and Tan, 2012; Singh, 2004). There are many variations in flower form among legumes, but the flowers of the Papilionoideae subfamily are almost uniformly zygomorphic with papilionoid features (Schutte, 1997). The present study identifies the most appropriate flower bud stage(s) from which to collect pollen for use in controlled crosses. The ability to maintain viability of stored pollen is essential to breeding programmes when anthesis does not coincide between the desired parent plants (Moura et al., 2015; Ozcan et al., 2019). Because honeybush species flower only once per year, collected pollen may have to retain viability for 12 18 months in storage, until the next flowering season, in order for certain crosses to be made. Most studies that look at pollen storage viability involve a reduction of the pollen sample’s moisture content and storage at a low and constant temperature (Guerrant et al., 2013; Mesnoua et al., 2018). The in vitro germination of pollen grains is a technique used for estimating viability of a pollen sample. Optimal media formulation for the in vitro germination of pollen grains is known to differ between species. Agar, sucrose and boric acid form the basic formulation used in most studies, sometimes with additional components such as basal salts (Shivanna, 2019). Media sucrose content one of the first variables to be determined in creating a species-

168

J. Koen et al. / South African Journal of Botany 128 (2020) 167 173

specific media formulation (Lin et al., 2017; Sarkar et al., 2018; Shivanna, 2019). Pollen grains, produced in the anthers of the androecium, are mobile cells that contain the spermatic nuclei of angiosperms. The pollen grain surface is generally comprised of pollenkit (pollen coat), exine (outer layers) and intine (inner layers) (Punt et al., 2007). The exine of pollen grains, comprised mainly of sporopollenin, is frequently found to form species-specific patterns of ornamentation and aperture types. Palynological studies form part of the basic understanding of any species, but pollen morphologies of Cyclopia spp. have not been described in any publication to date. Acetolysis is a technique commonly used to remove pollenkitt lipids and/or other substances from the surface of the pollen coat in order to reveal distinguishing details of the exine (Hesse and Waha, 1989; Wang and Dobritsa, 2018). However, it is known that sensitive pollen and/or fragile pollen structures may also be damaged or destroyed when using this technique. Ethanol/methanol washes have been recommended as an alternative to acetolysis (Jones, 2014). In the present study, the storage potential of air dried (moisture content not determined) honeybush pollen, stored for 18 months at 18 °C, was evaluated. The effect of MSC (0 25%) on in vitro PGP of honeybush pollen was measured. The surface morphologies of honeybush pollen grains were also imaged using FE-SEM microscopy and described using standard palynological terminology (Punt et al., 2007). 2. Material and methods Plant materials (flower buds and pollen) were collected from open-pollinated, honeybush seed orchards at the ARC Nietvoorbij Research Farm (S 33° 540 23.41700 E 18° 520 14.19600 ), Stellenbosch. Collections took place between July and October of 2017 from three species: C. longifolia, C. maculata and C. subternata. 2.1. Identifying the appropriate flower bud stage for honeybush pollen collection Honeybush flower buds were collected and arranged according to their developmental sequence. Visually-identifiable developmental stages within the sequence were numbered and buds from each stage were dissected for observation and in vitro testing of pollen maturity. The in vitro pollen germination tests were conducted as described in 2.3. 2.2. Storage of honeybush pollen at

18 °C for 540 days

Pollen samples from three genotypes of each species were collected. The samples were dried for 12 h at room temperature in a desiccant chamber and sealed in 1.5 ml micro-centrifuge tubes. The tubes were then sealed in LDPE bags and stored at 18 °C for 540 days. Pollen samples were tested for germination on day 7, 30, 90, 180, 365 and 540. A new micro-centrifuge tube containing pollen was opened for each genotype on each observation day and discarded after use. The in vitro pollen germination tests and data analysis are described in 2.3. 2.3. Testing 0 25% sucrose content for in vitro germination of honeybush pollen Pollen samples from three genotypes of each honeybush species were collected. Solid in vitro pollen germination media, containing 0.4% bacteriological agar and 100 ppm boric acid, were freshly prepared with concentrations of 0, 5, 10, 15, 20 and 25% sucrose. The media were set in polystyrene Petri dishes, with four replicates per medium. Honeybush pollen samples were sown onto the media surface and allowed to germinate for 1 3 h at room temperature. Digital

photographs were then taken via dissecting microscopy and manual counts were performed. A pollen grain was considered to have germinated when the length of the pollen tube exceeded the diameter of the grain (Alburquerque et al., 2007; Craddock et al., 2000; De  jo Machado et al., 2014; Mondal and Ghanta, 2012; Sparks and Arau Yates, 2002; Van der Walt and Littlejohn, 1996). PGP was calculated with the formula: number of germinated pollen/number of pollen x 100 (Liu et al., 2013; Maryam et al., 2015; Moura et al., 2015). The experimental design was completely randomised design with four replications. Experiments were conducted over six days. The treatment design of an experiment was a split plot design with six media treatments as main plot and the subplot factor was three species with three genotypes within each species. The data was recorded as percentage germinated pollen, measured as the number of germinated pollen grains divided by the total number of pollen grains in a demarcated area (with a minimum of 50 grains in a demarcated area). The Levene test of homogeneity of experimental variances was not verified, therefore a combined analysis of variance over days was performed using a weighted least squares analysis of variance (John and Quenouille, 1977). The weight was the reciprocal of the mean square error of the experiment of each day. The data was subjected to the General Linear Models Procedure (PROC GLM) of SAS software (Version 9.4; SAS Institute Inc., Cary, USA). Shapiro-Wilk test was performed on the standardized residuals from the model to verify normality (Shapiro and Wilk, 1965). Fisher’s least significant difference was calculated at the 5% level to compare treatment means (Ott and Longnecker, 2001). A probability level of 5% was considered significant for all significance tests. 2.4. Imaging honeybush pollen surface morphology using SEM In this descriptive study, dry pollen samples were prepared for SEM by oven-drying at 50 °C for 6 h. Additional samples of honeybush pollen were either acetolysed (Jones, 2014) or rinsed in 95% EtOH to investigate these as potential techniques for removal of pollenkitt. Attempts were also made to image imbibed pollen grains by dehydration through EtOH and HDMS liquid phases (James Cook University, 2014). Critical-point drying was not attempted. Samples were gold sputter coated and images were taken with a Zeiss MerlinTM (FE-SEM) at Stellenbosch University’s Central Analytical Facilities. Pollen surface morphology was described using palynological terminology as defined in the glossary created by Punt et al. (2007). Measurements were taken using Digimizer (Digimizer, 2019). 3. Results and discussion 3.1. Identifying the appropriate flower bud stage for honeybush pollen collection Pollen taken from honeybush flower buds at stage 1 3 showed no sign of germination in vitro but pollen samples taken from bud stages 4 and upward were found to germinate in vitro (Fig. 1). Pollen dehiscence was found to occur from bud stage 5 to 6. Upon dehiscence, from stage 5 and upwards, pollen was found to form a soft, somewhat-cohesive pellet inside the tapered, distal end of the fused ventral petals (Fig. 2). In the investigated species, this pellet of pollen was often found to surround the stigma upon anthesis. Harvesting of dehisced pollen (bud stage 6 7) was found to be a simpler and quicker procedure than plucking individual, undehisced or partly-dehisced anthers from their filaments (bud stage 4 5). Half of the flower, comprised of one ventral petal, one lateral petal and half of the dorsal petal, can be peeled away and the pollen knocked or brushed out. An efficient way to accomplish this is by inserting one half of the tip of a narrow tweezer into the keel through the

J. Koen et al. / South African Journal of Botany 128 (2020) 167 173

169

Fig. 1. Stages of flower bud development in Cyclopia spp. 1: Calyx emerged from bract, no petals visible; 2 4: Petals emerging and lengthening; 5: Dorsal petal begins to unseal along its bottom margin, from the point closest to the calyx and outward, and calyx begins to indent at the base; 6 7: Lateral and ventral petals become visible, dorsal petal continues to unseal along its margin, calyx continues to indent; 8: Dorsal petal flexes backward and the flower is fully open. Scale bar = 10 mm.

Fig. 2. Papilionoid C. subternata flower. A = dorsal petal; B = lateral petal, C = ventral petal; D = sexual organs; Scale bar = 10 mm.

small, natural opening at it base, gripping the petals firmly and gently pulling away in a curled motion. Apart from flower bud stage, external factors may also affect pollen dehiscence and viability. For example, temperatures above or below the optimal range for a species have been found to inhibit pollen dehiscence and cause pollen sterility in some food legumes (Sita et al., 2017). Low temperatures were found to inhibit pollen dehiscence of pigeon pea, contributing to low seed-set (Choudhary et al., 2018). In honeybush, the genotype, age and condition of the donor plant may also play a role in determining fecundity (Motsa et al., 2017). It is not unusual for papilionaceous species to shed pollen directly onto their stigma (Chaturvedi et al., 2011). Cleistogamy and autogamy are common in legumes, but by no means universal. In a study of six papilionoid legume species in montane and subalpine regions of Canada (Kudo and Harder, 2005), all species showed strong self-incompatibility (obligate out-crossers). In honeybush, compatibility trials are currently underway.

3.2. Storage of honeybush pollen at

18 °C for 540 days

Significant differences in variation of response to length of cold storage between genotypes of the three species were observed (Table 1). Overall, PGP decreased between observation day 7 and observation day 540 by 9.9% for C. longifolia and 22% for C. subternata, but increased for C. maculata by 2.6% over the same period (LSDp=0.05=2.7) (Fig. 3). During the study, C. longifolia displayed a consistently higher PGP (mean 59.18 § 2.38 standard error) than C. maculata (42.19 § 2.34) or C. subternata (32.76 § 3.9) (Fig. 3). Cyclopia subternata had the lowest PGP overall, an unexpected result given than Motsa (2017) found C. subternata to have higher levels of natural fecundity when compared to C. genistoides. Factors such as plant donor age or condition may have affected the viability of the pollen samples. Dafni and Firmage (2000) discuss these and other factors (morphological, environmental, internal, etc.) that are known to affect pollen viability in some

170

J. Koen et al. / South African Journal of Botany 128 (2020) 167 173 Table 1 Pollen germination percentage for three species of Cyclopia by genotype (clone) and storage period at 18°C.

Table 2 Pollen germination percentage for three species of Cyclopia by clone and media sucrose content.

Storage period in days

Media sucrose concentration (%)

Species

Clones

7

30

90

180

365

540

Species

Clones

0

5

10

15

20

25

1

LHK35 LHK36 LHK51 MBV3 MBV4 MBV10 SKB4 SKB15 SKB18

51.8 63.7 54.9 43.7 27.3 44.4 41.1 50.9 33.4

54.8 68.7 63.7 52.9 34.6 52.9 48.8 40.0 30.7

52.8 63.9 56.7 45.8 29.4 50.9 57.0 33.4 21.3

53.5 72.8 64.3 49.2 35.8 49.3 55.0 12.1 28.3

55.6 77.5 69.9 30.8 29.8 59.5 42.1 0.0 36.2

33.3 53.5 53.8 43.0 30.7 49.5 31.9 0.0 27.4

1

LHK35 LHK36 LHK51 MBV3 MBV4 MBV3 SKB4 SKB15 SKB18

45.8 50.4 57.5 33.6 21.9 34.6 37.1 11.1 13.0

56.0 72.4 66.9 45.6 31.3 49.3 46.6 16.9 17.4

56.3 70.5 63.9 50.9 32.5 54.6 52.0 23.2 25.2

54.2 72.2 57.9 45.0 31.0 54.8 47.7 27.3 30.6

48.7 68.7 58.4 47.6 36.2 56.5 50.0 30.7 46.7

41.0 66.0 58.6 42.7 34.8 56.8 42.7 27.4 44.5

2

3

Species: 1 = C. longifolia (LSDp=0.05=5.8), 2 = C. maculata (LSDp=0.05=8.4) and 3 = C. subternata (LSDp=0.05=6.2)

cases. Large and significant differences in PGP observed between genotypes within a species have also been reported in studies on other  jo Machado et al., 2014; species (Dafni and Firmage, 2000; De Arau Moura et al., 2015). Over the storage period, fluctuations in PGP were observed for all species and clones within species. While PGP may be expected to decrease over time due to the entropic processes of biological aging (Popovic, 2018), unexpected increases in PGP between one observation day and the next were also recorded. Since in vitro pollen germination was conducted without temperature control in this study, it seems probable that honeybush pollen is sensitive to ambient temperature as a variable during in vitro germination. Higher or lower temperatures than optimal have been shown to affect pollen germination and pollen tube elongation in studies on other species (Sita et al., 2017; Sparks and Yates, 2002). However, Cheung (2011), commenting on several Arabidopsis ecotype studies, notes that in vitro PGP can vary significantly between one day and the next, even when test conditions seem to be identical. Only one of nine honeybush genotypes reached zero percent in vitro pollen germination during the storage period (SKB15). In contrast, four genotypes (LHK51, MBV3, MBV4, MBV10) did not decrease significantly in PGP after 540 days of storage at 18 °C. While a significant decrease in total PGP was observed (9.7%, species combined, LSDp=0.05=1.1), spikes in PGP recorded over the storage period and the longevity of 4/9 genotypes indicate the potential for dried honeybush pollen to be stored for 18+ months at 18 °C without significant loss of viability. Pollen collection and storage methods need to be further optimised in

Fig. 3. The effect of storage period at significantly different at p = 0.05.

2

3

Species 1 = C. longifolia (LSDp=0.05=4.0), 2 = C. maculata (LSDp=0.05=4.9) and 3 = C. subternata (LSDp=0.05=4.5)

order to obtain consistent results. It is also recommended that future studies determine what level of correlation exists between viability measurements obtained by in vitro germination of honeybush pollen and functional seed set following pollination. 3.3. Testing 0 25% media sucrose content for in vitro germination of honeybush pollen Significant differences in variation of the response to MSC between genotypes of the three species were observed (Table 2). Overall, highest PGP was obtained at 5 10% MSC for C. longifolia (max. 65.1%), at 10 25% MSC for C. maculata (max. 46.8%) and at 20% MSC for C. subternata (max. 42.5%) (LSD p = =0.05=2.67) (Fig. 4). For the honeybush species tested in this study, optimal MSC range varied significantly between species and genotypes within species. Differences in pollen viability and longevity have also been reported between genotypes of other species, for example Protea and Musa (Ikeh, 2014; Van der Walt and Littlejohn, 1996). It was not possible to measure pollen tube length in the current study, but longer pollen tube length seemed to be associated with higher PGPs. Future studies may confirm or refute this observation. Sucrose is thought to play a role in maintaining osmotic balance between pollen tissues and the germination medium, helping to prevent early pollen tube rupture (Dey et al., 2016; Liu et al., 2013; Moura et al., 2015). The addition of other media components such as smoke-water and/or salts (calcium, potassium, magnesium, etc.) have proven beneficial to the in vitro pollen germination of other

18°C on pollen germination percentage of three Cyclopia species, genotypes combined within species. Means with the same letter are not

J. Koen et al. / South African Journal of Botany 128 (2020) 167 173

171

Fig. 4. The effect of media sucrose content on pollen germination percentage of three Cyclopia spp., genotypes combined within species. Means with the same letter are not significantly different at p = 0.05.

species and may investigated for Cyclopia spp. in future (Abdelgadir et al., 2012; Maguire and Sedgley, 1997; Mondal and Ghanta, 2012; Papenfus et al., 2014). Media sucrose content was also tested over the storage period (see 3.2). It was found that optimal MSC range for each genotype did not change over time (data not shown). 3.4. Imaging honeybush pollen surface morphology using SEM The examined Cyclopia spp. pollen grains were found to be similar in shape and ornamentation to those of some species from other taxa, such as Salix spp. (Halbritter, 2016) and Sambucus spp. (Tweraser et al., 2016). The grains are dispersed as tricolpate monads. When dry, they are prolate and isopolar while the equatorial outline (polar view) is lobate. Dry grains have infoldings, which are three sunken, colpate apertures (Figs. 5-7). When imbibed, the grains are spheroidal in shape with a circular outline in the polar view and

variable, amorphous ornamentations visible on the aperture membranes (Fig. 8). For all three honeybush species, pollen surface morphology was similar. Ornamentation of the sexine was found to be reticulate (heterobrochate) with free-standing columellae present in the lumina (Figs. 5 7). The margo is differentiated from the remainder of the sexine in ornamentation and thickness, becoming smoother (fewer and smaller lumina) and thinner toward the aperture margin (Fig. 8). Differences in pollen grain morphology have been used as identifying characteristics in the taxanomic delineation of Cyperaceae species, cornelian cherry cultivars (Cornus mas L.) and Butia species (Mert, 2009; Mourelle et al., 2015; Van Wichelen et al., 1999). On the other hand, pollen morphology is reportedly very similar between cultivars of field pea (Pisum sativum L) and between species of Ingofera (Jiang et al., 2015; Zhao et al., 2016). Pollen grains may also be very similar in appearance between unrelated plant families.

Fig. 5. Untreated, oven-dried Cyclopia longifolia pollen grains. a = equatorial view; b = polar view, c = exine surface. Images taken with FE-SEM, digitally enhanced.

Fig. 6. Untreated, oven-dried Cyclopia maculata pollen grains. a = equatorial view; b = polar view, c = exine surface. Images taken with FE-SEM, digitally enhanced.

172

J. Koen et al. / South African Journal of Botany 128 (2020) 167 173

Fig. 7. Untreated, oven-dried Cyclopia subternata pollen grains: a = equatorial view; b = polar view, c = exine surface. Images taken with FE-SEM, digitally enhanced.

Fig. 8. Cyclopia spp. pollen: a = partially collapsed imbibed grain, acetolysed and dehydrated through ethanol and HMDS phases; b = exine surface following acetolysis and ethanol dehydration; c = exine surface following ethanol rinse without acetolysis. Images taken with FE-SEM, digitally enhanced.

Low et al. (1989) observed that pollen grain surface morphology of Salix spp. and Brassica spp. are so similar as to be indistinguishable from each other unless acetolysed. Taking into consideration slight variations which exist between grains from the same sample, no consistent differences in shape and ornamentation were observed between the three Cyclopia spp. investigated, even when acetolysed. The untreated honeybush grains were found to be lightly coated in pollenkitt, which partially fills the lumina, coating the surface of the foot layer and obscuring the infratectum (Figs. 5 8). No visible damage to the grains resulted from acetolysis, but simple ethanol washes seemed to produce similar results (Fig. 8). Critical drying may be preferable to dehydration via ethanol and HMDS phases, which resulted in the slight distortion of imbibed grains (Fig. 8). The HMDS dehydrating agent was also found to leave a slight but noticeable residue. Average pollen size for the honeybush species tested was found to be “medium”, similar to Brassica spp. (Diethart, 2016). Average dimensions measured for C. longifolia were 38 mm polar axis length by 20 mm equatorial diameter. Measurements for C. maculata were 40 mm x 20 mm and 42 mm x 22 mm for C. subternata. It is recommended that more genotypes from each species be analysed to confirm whether the observed differences in grain size are a persistent trait. In the present study, it was found that visual stages of flower bud development could be used as indicators of the level of pollen development within honeybush flowers. This knowledge will facilitate the harvest of viable pollen for breeding trials. In the pollen storage experiment, four honeybush genotypes did not have any significant decrease in PGP over time, suggesting that honeybush pollen may be successfully stored for periods longer than 18 months, given the right conditions. In the pollen germination experiment, differences between species were observed in response to media sucrose content. This finding will form part of the creation of in vitro viability testing protocols for honeybush pollen. In the pollen surface morphology study, no consistent differences in shape or surface ornamentation were observed between honeybush species, though differences in grain size were observed. Further palynological investigation is recommended in order to distinguish honeybush pollen from that of similar species.

Declaration of Competing Interest None. Acknowledgments We thank the Stellenbosch University Central Analytics Facility for their assistance in obtaining the FE-SEM images. This study was supported by the Department of Science and Technology (DST), Agricultural Research Council (ARC) and Tshwane University of Technology (TUT). Supplementary materials Supplementary material associated with this article can be found in the online version at doi:10.1016/j.sajb.2019.11.004. References Abdelgadir, H.A., Johnson, S.D., Van Staden, J., 2012. Pollen viability, pollen germination and pollen tube growth in the biofuel seed crop Jatropha curcas (Euphorbiaceae). South African Journal of Botany 79, 132–139. https://doi.org/10.1016/j. sajb.2011.10.005. Alburquerque, N., García Montiel, F., Burgos, L., 2007. Short communication. Influence of storage temperature on the viability of sweet cherry pollen. Spanish Journal of Agricultural Research 5, 86–90 Jan. Bester, C., 2013. A model for commercialisation of honeybush tea, an indigenous crop. Acta Horticulturae 1007, 889–894. Chaturvedi, S.K., Gupta, D.S., Jain, R., 2011. Biology of food legumes. In: Pratap, A., Kumar, J. (Eds.), Biology and Breeding of Food Legumes. Oxfordshire, CABI, pp. 35–37. Cheung, A., 2011. Pollen Methods Course. Brown University, Providence, RI. Choudhary, A.K., Sultana, R., Vales, M.I., Saxena, K.B., Kumar, R.R., Ratnakumar, P., 2018. Integrated physiological and molecular approaches to improvement of abiotic stress tolerance in two pulse crops of the semi-arid tropics. The Crop Journal 6, 99–114. https://doi.org/10.1016/j.cj.2017.11.002. Cordoba, S.A., Cocucci, A.A., 2011. Flower power: its association with bee power and floral functional morphology in papilionate legumes. Annals of Botany 108, 919– 931. https://doi.org/10.1093/aob/mcr196 Aug. Craddock, J.H., Reed, S.M., Schlarbaum, S.E., Sauve, R.J., 2000. Storage of flowering dogwood (Cornus florida L.) pollen. Hortscience 35, Feb:108 109. Dafni, A., Firmage, D., 2000. Pollen viability and longevity: practical, ecological and evolutionary implications. Plant Systematics and Evolution 222, 113–132.

J. Koen et al. / South African Journal of Botany 128 (2020) 167 173  jo Machado, C., Feitosa Moura, C.R., Pinto de Lemos, E.E., Ramalho Ramos, S.R., De Arau do, A., 2014. Pollen grain viability of coconut accessions at low Ribeiro, F.E., da Silva Le temperatures. Acta Scientific 36, Apr:227 232. 10.4025/actasciagron.v36i2.17346 Dey, K., Mondal, S., Mandal, S., 2016. Studies on in vitro pollen germination of Mitragyna parvifolia (Roxb.) Korth. International Journal of Current Microbiology and Applied Sciences 5, 768–777. Diethart, B., 2016. Brassica napus (WWW document). paldat - A Palynol. database. URL https://www.paldat.org/pub/Brassica_napus (Accessed 19 July 2018). Digimizer, 2019. Belgium: MedCalc Software. Guerrant, E.O., Havens-Young, K., Maunder, M., Raven, P.H., Conservation, C.P., 2013. Ex situ plant conservation: supporting species survival in the wild. The Science and Practice of Ecological Restoration Series. Island Press. Halbritter, H., 2016. Salix daphnoides (WWW document). paldat - A Palynol. database. URL https://www.paldat.org/pub/Salix_daphnoides (Accessed 19 July 2018). Hesse, M., Waha, M., 1989. A new look at the acetolysis method. Plant Systematics and Evolution 163, 147–152. Ikeh, E., 2014. Pollen diversity, viability and floral structure of some Musa genotypes. Nigerian Journal of Biotechnology 27, 21–27. James Cook University, 2014. Biological sample preparation for the SEM (WWW document). URL https://research.jcu.edu.au/archive/enabling/aac/cairns-aac/biologicalsample-preparation-for-the-sem (Accessed 28 October 2019) Jia, J., Tan, D.Y., 2012. Variation in style length and anther-stigma distance in Ixiolirion songaricum (Amaryllidaceae). South African Journal of Botany 81, 19–24. https:// doi.org/10.1016/j.sajb.2012.03.011. Jiang, Y., Lahlali, R., Karunakaran, C., Kumar, S., Davis, A.R., Bueckert, R.A., 2015. Seed set, pollen morphology and pollen surface composition response to heat stress in field pea. Plant, Cell & Environment 38, 2387–2397. https://doi.org/10.1111/ pce.12589. John, J.A., Quenouille, M.H., 1977. Experiments: Design and Analysis, 2nd ed. Charles Griffin & Company LTD, London. https://doi.org/10.1002/bimj.4710200313. Jones, G.D., 2014. Pollen analyses for pollination research, acetolysis. The Journal of Pollination Ecology 13, 203–217. Kudo, G., Harder, L.D., 2005. Floral and inflorescence effects on variation in pollen removal and seed production among six legume species. Functional Ecology 19, 245–254. https://doi.org/10.1111/j.1365-2435.2005.00961.x. Lin, Y., Wang, Y., Iqbal, A., Shi, P., Li, J., Yang, Y., Lei, X., 2017. Optimization of culture medium and temperature for the in vitro germination of oil palm pollen. Scientia Horticulturae (Amsterdam) 220, 134–138. https://doi.org/10.1016/J.SCIENTA.2017.03.040. Liu, L., Huang, L., Li, Y., 2013. Influence of boric acid and sucrose on the germination and growth of Areca pollen. American Journal of Plant Sciences 4, 1669–1674. https:// doi.org/10.4236/ajps.2013.48202. Low, N.H., Schweger, C., Sporns, P., 1989. Precautions in the use of melissopalynology. Journal of Apicultural Research 28, 50–54. https://doi.org/10.1080/ 00218839.1989.11100820. Maguire, T.L., Sedgley, M., 1997. Storage temperature affects viability of Banksia menziesii pollen. Hortscience 32, 916–917. Maryam, M., Fatima, B., Salman Haider, M., Abbas Naqvi, S., Nafees, M., Ahmad, R., Ahmad Khan, I., 2015. Evaluation of pollen viability in date palm cultivars under different storage temperatures. The Pakistan Journal of Botany47, Jan.:377 381. Mert, C., 2009. Pollen morphology and anatomy of cornelian cherry (Cornus mas l.) cultivars. Hortscience 44, 519–522. Mesnoua, M., Roumani, M., Salem, A., 2018. The effect of pollen storage temperatures on pollen viability, fruit set and fruit quality of six date palm cultivars. Scientia Horticulturae (Amsterdam). 236, 279–283. Mondal, S., Ghanta, R., 2012. Effect of sucrose and boric acid on in vitro pollen germination of Solanum macranthum Dunal. The Indian Journal of Fundamental and Applied Life Sciences 2, 202–206.

173

Motsa, M.M., Bester, C., Slabbert, M.M., Ngwenya, M.Z., Booyse, M., 2017. Natural fecundity and germination characteristics of selected Cyclopia (Honeybush) species: preliminary findings. Journal of Agricultural Science 9, 154‒167. https://doi.org/ 10.5539/jas.v9n6p154. do, A., da, S., 2015. In vitro germination and viabilMoura, C.R.F., Machado, C., de, A., Le ^ncia Agrono ^mica 46, 421– ity of pollen grain of coconut accessions. A Revista Cie 427. https://doi.org/10.5935/1806-6690.20150022. n, C., Gutie rrez, L., Mazzella, C., 2015. ComparMourelle, D., Gaiero, P., Speroni, G., Milla ative pollen morphology and viability among endangered species of Butia (Arecaceae) and its implications for species delimitation and conservation. Palynology 40, 160–171. https://doi.org/10.1080/01916122.2014.999955. Ott, R.L., Longnecker, M., 2001. An Introduction to Statistical Methods and Data Analysis, 5th ed Duxbury Press, Duxbury, MA. Ozcan, A., Sutyemez, M., Bukucu, S.B., Ergun, M., 2019. Pollen viability and germinability of walnut: a comparison between storage at cold and room temperatures. Fresenius Environmental Bulletin 28, 111. Papenfus, H.B., Kumari, A., Kulkarni, M.G., Finnie, J.F., Van Staden, J., 2014. Smoke-water enhances in vitro pollen germination and tube elongation of three species of amaryllidaceae. South African J. Bot. 90, 87–92. https://doi.org/10.1016/j. sajb.2013.10.007. Popovic, M., 2018. Thermodynamic mechanism of life and aging. arXiv:1801.08073. Punt, W., Hoen, P.P., Blackmore, S., Nilsson, S., Le Thomas, A., 2007. Glossary of pollen and spore terminology. Review of Palaeobotany and Palynology 143, 1–81. https:// doi.org/10.1016/j.revpalbo.2006.06.008. Sarkar, T., Sarkar, S.K., Vangaru, S., 2018. Effect of sucrose and boric acid on in-vitro pollen germination of guava (Psidium guajava) varieties. Journal of Advanced Research 15 (1), 1–9. https://doi.org/10.9734/AIR/2018/41145. Schutte, A.L., 1997. Systematics of the genus Cyclopia vent. (Fabaceae, Podalyrieae). Edinburgh Journal of Botany 54, 126–133. https://doi.org/10.1017/ S0960428600004005. Shapiro, S.S., Wilk, M.B., 1965. An analysis of variance test for normality (complete samples). Biometrika 52, 591–611. Shivanna, K.R., 2019. Pollen Biology and Biotechnology. CRC Press, Boca Raton, Florida. Singh, G., 2004. Plant systematics: an Integrated Approach, 2nd ed. Science Publishers, Enfield, NH. Sita, K., Sehgal, A., HanumanthaRao, B., Nair, R.M., Vara Prasad, P.V., Kumar, S., Gaur, P.M., Farooq, M., Siddique, K.H.M., Varshney, R.K., Nayyar, H., 2017. Food legumes and rising temperatures: effects, adaptive functional mechanisms specific to reproductive growth stage and strategies to improve heat tolerance. Frontiers in Plant Science 8, 1658. https://doi.org/10.3389/fpls.2017.01658. Sparks, D., Yates, I.E., 2002. Pecan pollen stored over a decade retains viability. Hortscience 37, 176–177. Tweraser, E., Halbritter, H., Schneider, H., 2016. Sambucus ebulus (WWW document). Paldat - A Palynol. database. URL https://www.paldat.org/pub/Sambucus_ebulus (Accessed 19 July 2018). Van der Walt, I.D., Littlejohn, G.M., 1996. Storage and viability testing of Protea pollen. The Journal of the American Society for Horticultural Science 121, 804–809. Van Wichelen, J., Camelbeke, K., Chaerle, P., Goetghebeur, P., Huysmans, S., 1999. Comparison of different treatments for LM and SEM studies and systematic value of pollen grains in Cyperaceae. Grana 38, 50–58. https://doi.org/10.1080/ 001731300750044708. Wang, R., Dobritsa, A.A., 2018. Exine and aperture patterns on the pollen surface: their formation and roles in plant reproduction. In: Roberts, J.-A. (Ed.), The Annual Review of Plant Biology. John Wiley Sons, pp. 1–40. Zhao, X.-.L., Gao, X.-.F., Xu, B., 2016. Pollen morphology of Indigofera (Fabaceae) in China and its taxonomic implications. Plant Systematics and Evolution 302, 469– 479. https://doi.org/10.1007/s00606-015-1275-1.