Abstract
Seeds of some winged bean varieties have low germination due to the presence of water-impermeable hardseeds. Seeds of ‘Taitung No.1’ winged bean had only 31% germination because the remaining 69% of seeds had a water-impermeable seedcoat. Sandpaper abrasion and sulfuric acid immersion for 15 and 25 min effectively removed hardseededness of the seeds, resulting in more than 89% germination. As seed moisture content (MC) decreased from 14.8% to 7%, the percentage of seeds with a water-impermeable seedcoat increased. Seed lots with 14.8%, 11%, 9%, and 7% MC had 7%, 38%, 56%, and 78% of hardseeds, respectively, on day 7 in the germination test. It was found that the hilum was responsible for water loss from the whole seed, and the seedcoats began to become water-impermeable at 12% MC. The lens and micropylar regions were initial water entry sites in the Vaseline-blocking experiment.
Winged bean [Psophocarpus tetragonolobus (L.) DC.] is a legume vegetable crop commonly grown in the tropics primarily for its edible green pods (NAS, 1975). The green pods are used as a vegetable and are slowly becoming popular in Taiwan over the past few years. However, some winged bean seeds do not absorb water when it is readily available, such as during a germination test or in wet soil (these are known as “hardseeds”), and seed germination is delayed as a result. This undoubtedly poses a challenge in the cultivation of winged bean. Research on winged beans is scarce, and most of it was conducted during the 1970s and 1980s, with the primary focus on seed nutrition and crop cultivation (Cerny et al., 1971; Kadam et al., 1984; Nangju and Baudoin, 1979; Sathe et al., 1982). Although preliminary investigations into hardseededness in winged bean were carried out a few decades ago (Ellis et al., 1985; Rudrapal et al., 1992; Tham, 1980), there have been no recently published studies, and hardseededness remains a significant constraint on greater utilization of this crop.
Seed dormancy is a life history adaptation that spreads germination over time, and it helps time germination so that it occurs when environmental conditions are favorable for seedling establishment (Egley, 1989). Hardseeds are seeds that develop physical dormancy (PY), that is, a water-impermeable seed or fruit coat, during maturation development and are commonly found in species of Fabaceae (Baskin and Baskin, 2014). Seeds of food legumes such as soybean (Harris, 1987), lima beans (Pollock and Toole, 1966), and chickpeas (Frisbee et al., 1988), as well as seeds of forage legumes such as white clover (Martens et al., 1995) and subterranean clover (Nichols et al., 2013), are known to contain hardseeded lines. After dispersal, hardseeds do not germinate even if water is freely available, and they will persist in the soil until the dormancy is ended by suitable environmental signals such as seasonal temperature variation and high or low temperature (Gama-Arachchige et al., 2010; Van Assche et al., 2003).
In hardseeded legumes, the water impermeability of the seedcoat lies within the palisade layer, which is tightly packed with macrosclereids (Smýkal et al., 2014). Due to the architecture of the cell arrangement and heavily thickened cell walls of the macrosclereids, there is little intercellular space in the palisade layer, and this can impede water diffusion in the seedcoat (Ma et al., 2004). In addition, macrosclereids are often impregnated with water-repelling chemicals such as phenolics and tannins (Harris, 1987). One of the major phenolics isolated from the seedcoat of wild soybean (Glycine soja), epicatechin, exhibits a positive correlation with the hardseededness of the seeds (Zhou et al., 2010). Seeds with PY have a sensitive specialized microstructure, that is, a water gap that acts as a water entry site if the seed is exposed to a specific environmental signal such as high temperatures, alternating temperatures, light, and humidity (Geneve et al., 2018; Yang et al., 2019).
Seed MC is perhaps one of the most important factors during hardseed development. The water impermeability of the seedcoat in hardseeds is initiated only when the MC of the seeds is below a species-specific threshold level (Jaganathan, 2016). Seeds of Indian lotus (Nelumbo nucifera) with 10% MC had 77% hardseeds, compared with 10% hardseeds in 12% MC seeds (Jaganathan et al., 2017). Similarly, in seeds of adzuki bean (Vigna angularis), the seedcoat was permeable at 29.94% MC (dry basis) and became impermeable when MC was lowered to 14.96% (Miano and Augusto, 2015). As seedcoat becomes impermeable to water, the hilar fissure of the seed acts like a valve that allows water vapor to diffuse out of the seed under low surrounding relative humidity (RH), and it prevents water from entering the seed under high humidity (Hyde, 1954).
Seed coats of hardseeds can be made permeable by using scarification treatments such as acid immersion, heating, freezing, and mechanical abrasion (Kimura and Islam, 2012). Treatment with concentrated sulfuric acid is one of the most effective ways to remove hardseededness, but the immersion period needs to be carefully monitored because prolonged immersion will harm seeds (Hu et al., 2009; Wang et al., 2007). Seeds of cluster clover (Trifolium glomeratum) had germination of <5% due to hardseededness, but a 30-min treatment with concentrated sulfuric acid increased germination to >80% (Martín and Guerrero, 2014). However, longer immersion caused seed damage, whereas seeds remained hard when immersion time was insufficient. Sandpaper abrasion can be an effective mechanical treatment for hardseeds, although it is labor-intensive. Seeds of hardseeded cluster clover had close to 0% germination, whereas the seeds abraded by sandpaper had >95% germination (Mira et al., 2017). Nicking also removes hardseededness in seeds by directly damaging the seedcoat with a knife or nail clipper, which allows water entry. For example, nicked hardseeds of Leucaena leucocephala had 90% germination compared with 0% in nontreated seeds (Tadros et al., 2011).
In this study, we wished to investigate the following aspects of hardseeded winged bean seeds: 1) effectiveness of scarification treatments on seed germination, 2) effect of seed MC on development of hardseededness, and 3) structural difference in the seedcoat of hardseeds.
Materials and Methods
Materials.
Seed Lot 1: ‘Taitung No.1’ winged bean seeds were received from the Taitung District Agricultural Research and Extension Station (TTDARES). The MC of the seeds was 7.1%, and seeds had been in storage for <1 year. Seed Lot 2: ‘Taitung No.1’ mature winged bean pods were collected from plants at National Chung Hsing University (NCHU) between Feb. and Apr. 2021. Seed pods were harvested when they were fully black and were then left at room temperature for 10 d before shelling the seeds for later use. The MC of the shelled seeds was ≈14.8%. They were dried to 12%, 11%, 9%, and 7% MC by using drying beads and 3% RH in a humidity-controlled auto drying cabinet (SL-86CA; EDRY, Taiwan).
Scarification treatments on seed germination.
Seeds from TTDARES (Seed Lot 1) were used. Seeds were subjected to the following treatments. Sandpaper: seeds were scarified on the dorsal side with sandpaper until the seedcoat layer was noticeably disturbed and tiny white spots were visible. Sulfuric acid immersion: seeds were immersed in concentrated sulfuric acid (H2SO4, 95% concentration) for 5, 15, and 25 min, then washed under tap water for 1 min. Nontreated seeds were used as a control.
Effects of initial seed moisture content on hardseededness.
Moisture content of seeds with impermeable seedcoat.
Identification of initial water entry sites.
Seeds harvested at NCHU with 12% MC were used (Seed Lot 2). Seeds were covered with a layer of Vaseline at the micropyle, lens, or both sites (micropyle + lens) and germinated at 30 °C in a petri dish as described in the earlier section. Each replicate contained 15 seeds in a petri dish, and a total of three replicates were used. Hardseeds were counted at the end of day 5 and calculated as a percentage.
Seed coat structural observation by scanning electron microscope.
Permeable seed of ‘Taitung No.1’ winged bean with 13% MC and hardseed with 9% MC were examined under scanning electron microscope (SEM). In our preexperiment, we noticed that seeds with 13% MC had 0% hardseeds and imbibed normally. Only when seed MC dropped below 12% did hardseeds start to appear. Because all seeds at 13% MC were permeable, one was randomly selected from the batch. Hardseeds were selected by first placing seeds in a petri dish lined with wet filter paper for 14 d, then seeds that did not imbibe water (no size increase or wrinkled seedcoat) were used. The samples were coated with gold and examined with Hitachi S-3000N scanning electron microscope at 15 kV.
Statistical analysis.
All data are represented as mean values across replications. Analyses of variance were performed using the software SAS Enterprise Guide 7.1 and then mean values were compared using Fisher’s least significant difference test (P = 0.05). Germination, hardseeds, and nonimbibed seed percentages were arcsine transformed before analysis to improve normality.
Results
Effects of scarification treatments on seed germination.
Seeds of ‘Taitung No.1’ winged bean seeds without scarification treatments had 69% of hardseeds (Table 1). Sandpaper treatment completely removed hardseededness of the seeds and achieved 100% germination. In concentrated sulfuric acid (H2SO4) treatments, immersion for 5 min was not sufficient, and 16% of hardseeds remained; the seed took significantly longer to germinate compared with other treatments. Longer immersion for 15 and 25 min effectively reduced the hardseededness to <5% and resulted in germination >89%. However, seeds immersed in H2SO4 for 25 min had 9% abnormal seeds. Seeds subjected to sandpaper treatment and H2SO4 immersion for 15 and 25 min germinated significantly faster, with mean germination time (MGT) <4.9 d compared with 6.2 d in the control. Seeds in sandpaper treatments and H2SO4 immersion for 25 min also had the most uniform germination.
The effects of scarification treatments on the germination percentage, hardseeds percentage, abnormal seed percentage, mean germination time (MGT), and uniformity (T90–T10) of ‘Taitung No.1’ winged bean seeds incubated on moist filter paper in petri dish at 30 °C for 14 d.
Effects of initial seed moisture content on hardseededness.
Water uptake of the seed batches was positively correlated with their respective seed MC. All seed, regardless of MC, had very little water uptake on day 1 (Fig. 1). Seeds with 14.8% MC had a rapid increase in water uptake on day 2, and water uptake remained the highest throughout the course of imbibition. Seeds with 11% and 9% MC shared a rather similar water uptake curve, and rapid increase in water uptake occurred on days 3 and 5, respectively. Seeds with 7% MC had the lowest and slowest water uptake of all seeds. However, scarified seeds with 14.8% and 7% MC showed a similar water uptake curve, much faster than nonscarified seeds. For example, to reach water uptake of 0.3 g water/g seed DW, nonscarified seeds with 14.8%, 11%, 9%, and 7% MC required roughly 2.5, 4.5, 5.5, and 8.5 d, respectively, whereas scarified seeds with 14.8% and 7% required only 1.5 and 2 d (Fig. 1). Seeds with lower initial MC contained more hardseeds, as seen in the seeds with 7% MC, which had the most nonimbibed seeds during the 14-d experiment (Fig. 2). The percentage of nonimbibed seeds also decreased slowly in the batch that initially had 7% MC. For example, seeds initially with 7% MC had 78% of nonimbibed seeds on day 7, whereas seeds with 11% and 14.8% MC had 38% and 7% of nonimbibed seeds, respectively. The percentage of nonimbibed seeds decreased subsequently on day 14, whereas seeds initially with 14.8% MC had almost no hardseeds; seeds with initial 7% and 11% MC still had 36% and 16% hard nonimbibed seeds, respectively.
Water uptake by ‘Taitung No.1’ winged bean seeds with 14.8%, 11%, 9%, and 7% initial moisture content, and scarified seeds with 14.8% and 7% moisture content, incubated on moist filter paper in petri dishes at 30 °C. Error bars indicate the standard deviation of the means. DW = dry weight.
Citation: HortScience 57, 9; 10.21273/HORTSCI16683-22
Percentage of nonimbibed seeds in ‘Taitung No.1’ winged bean seeds with 14.8%, 11%, 9%, and 7% initial moisture content, incubated on moist filter paper in petri dishes at 30 °C. Error bars indicate the standard deviation of the means. Means with the same letter on days 7 and 14 are not significantly different by Fisher’s least significant difference test at 5% level.
Citation: HortScience 57, 9; 10.21273/HORTSCI16683-22
Moisture content in seeds with blocked hilum.
Seeds with a blocked hilum had a slow decrease in MC throughout the drying process (Fig. 3). For example, nonblocked seeds initially at 14.8% MC required only 2 d to decline to 12% MC compared with 20 d for seeds with a blocked hilums. At the end of 20 d, the MC of nonblocked seeds had dropped to 5.3%.
Moisture content of ‘Taitung No.1’ winged bean seeds, with and without hilum blocked, during drying at 3% relative humidity at 27 °C. Error bars indicate the standard deviation of the means.
Citation: HortScience 57, 9; 10.21273/HORTSCI16683-22
Identification of initial water entry sites.
In seeds with micropyle and lens blocked by Vaseline, 78% were hardseeded, which was the highest among the treatments (Fig. 4). Seeds with either site blocked had roughly 58% hardseeds, compared with 36% in nonblocked seeds.
The hardseed percentage of ‘Taitung No.1’ winged bean seeds (12% seed moisture content) with micropyle (Mic), lens, and micropyle and lens (Mic+Lens) blocked by Vaseline and nontreated seed (CK) on day 5. Error bars indicate the standard deviation of the means.
Citation: HortScience 57, 9; 10.21273/HORTSCI16683-22
Seed coat structure in SEM.
On the ventral side of ‘Taitung No.1’ seeds, there is clearly a visible hilum. The micropyle was visible on the permeable seed (Fig. 5A) but was rarely visible in the hardseed (Fig. 5B). Permeable seeds had an oval-shaped opened micropyle (Fig. 5E), whereas the micropyle stayed tightly shut in the hardseed, appearing to look like a slit (Fig. 5F). The micropyle of the seeds was much smaller than the hilum and was located very close to the hilum. The hilar fissure of permeable seed was only slightly more opened (Fig. 5C) than that of the hardseed (Fig. 5D). The lens was barely visible on permeable and hard seeds, and the two seeds had a rather smooth and similar surface (Fig. 5G and H). On closer inspection, at the lens region, small cracks appeared on permeable seeds (Fig. 6A), but there were no cracks on the hardseeds (Fig. 6B). On the surface of the same region, both permeable and hard seeds had a region with a dot-shaped pattern toward the hilum (left on the image), which was surrounded by a ring-shaped pattern. The surface became smoother father away from the hilum, near the lens (Fig. 6A and B). In the lens region, permeable seeds had a bumpy surface with a dot-like pattern (Fig. 6E). A similar but less bumpy pattern was visible on the surface of the hardseeds, which appeared to be covered by a thicker layer of deposit (Fig. 6F). On the seedcoat outside of the hilar region, permeable seed had a fabric-like pattern surface that was parallel in one direction, with a honeycomb-shaped pattern on top (Fig. 6G). On the other hand, the surface on the seedcoat of hardseeds appeared smoother and covered by a layer of deposit, but a similar honeycomb-shaped pattern could still be seen (Fig. 6H).
Scanning electron micrograph overviews of ‘Taitung No.1’ winged bean seed with (A, C, E, G) 13% moisture content and (B, D, F, H) hardseed with 9% moisture content, at the (A, B) hilar region, (C, D) hilum, (E, F) micropylar, and (G, H) lens. Hf = hilar fissure; Hi = hilum; Le = lens; Mi = micropyle.
Citation: HortScience 57, 9; 10.21273/HORTSCI16683-22
Scanning electron micrograph close-up views of ‘Taitung No.1’ winged bean seed with (A, C, E, G) 13% moisture content and (B, D, F, H) hardseed with 9% moisture content, at the (A, B) lens, surface of (C, D) micropylar region, (E, F) lens region, and (G, H) seedcoat outside the hilar region. *Dot-shaped pattern on surface near hilum; **ring-shaped pattern on surface near hilum; ***smooth surface near lens.
Citation: HortScience 57, 9; 10.21273/HORTSCI16683-22
Discussion
Scarification treatments on seed germination.
It is important to note that the seed scarification treatments were carried out on a seed lot that often contained both permeable and hardseeds (Davis et al., 1991). Because permeable seeds are more vulnerable to the damage caused by scarification, the severity of the treatment must be limited to prevent any substantial increase in abnormal seeds. The low germination of the ‘Taitung No.1’ seeds was due to a high percentage of hardseeds (Table 1). Similar observations were made by Ellis et al. (1985) and Tham (1980) on winged beans seeds. The impermeable seedcoat was responsible for the hardseededness, which is a common feature in seeds of Fabaceae species (Smýkal et al., 2014). In our study, sandpaper was the most effective among the treatments because it directly damaged the impermeable seedcoat of the seeds. Similar results on improving the germination of hardseeded winged bean seeds were obtained by Ellis et al. (1985). Winged bean seeds are large (0.86 cm × 0.75 cm × 0.87 cm) and thus can be easily scarified using sandpaper. Seeds of Trifolium subterraneum scarified with sandpaper are highly irregular due to their small size, which makes them vulnerable to damage during scarification (Martin and De la Cuadra, 2004). Immersion in concentrated sulfuric acid for 15 and 25 min was also effective in our study (Table 1). Immersion of hardseeds of Cicer canariense (a perennial wild relative of chickpea) in concentrated sulfuric acid (98%) for 120 min significantly improved the germination from 0% to 94% (Guma et al., 2010). During acid immersion, the seedcoat was digested, and more seeds become permeable as a result. Tiny holes were visible throughout the entire seedcoat of the acid treated seeds, and more and larger holes appeared as the immersion period increased. Similar observations were made by Hu et al. (2009). The quantity of seeds used per batch during the acid immersion treatment could influence effectiveness of the treatment. During our preexperiment, the same acid treatment was less effective and resulted in more seeds remaining hard when a larger batch of seeds was immersed in the acid all at once (data not shown). This was most likely due to seeds at the center of the batch having less contact with the acid than those on the outer edge, which reduced the level of corrosion of the seedcoats. We also found that hot water treatment at 70 to 90 °C for 1 to 15 min was not an effective scarification method because the number of abnormal seeds increased with longer treatment periods but the percentage of hardseeds did not decrease (data not shown). Generally, the remaining hardseeds were dead before the seedcoat became permeable. In addition, we tried ultrasonic treatment (40k Hz and 120 W), in which seeds were soaked in water and oscillated at 30 °C for 30 min, but to no avail; the water temperature would increase to 50 °C due to heat generated from the ultrasonic treatment, but the percentage of hardseeds did not decrease significantly (data not shown). In a previous study of winged bean seeds, Tham (1980) found that hot water treatment at 55 or 95 °C did not have any significant effect on reducing the hardseed percentage, whereas mechanical scarification (seedcoat pricking) and sulfuric acid treatment had great success in reducing the hardseeds from 45% to 1%. We propose using mechanical scarification and sulfuric acid immersion as the primary treatments for future investigations on hardseeded winged bean varieties. Large-scale mechanical scarification methods and adjustments to the concentration of sulfuric acid and immersion period can be further investigated.
Effects of initial seed moisture content on hardseededness.
The relationship between seed MC and hardseededness was first observed in the seeds collected from NCHU. We noticed that the collected seeds did not contain many hardseeds before drying. Seed MC could be categorized into three groups based on their water uptake (fastest to slowest): 14.8% initial MC, 11% to 9% MC, and 7% MC (Fig. 1). The difference in water uptake was due to the degree of hardseededness among seeds with different initial MC. Because seeds with lower MC had a higher percentage of hardseeds that also become permeable at a slower rate (Fig. 2), this limited the water uptake of the seeds at any given time. Most permeable seeds would finish imbibing within 4 d (data not shown), thus the continuous increase in water uptake by the seeds was the result of more seeds becoming permeable over time during the imbibition period. In other words, the water uptake rate was related to the breakdown rate of the hardseeds. With lower initial MC of seeds, the seedcoat was more water-impermeable and harder to make permeable. At an initial MC of ≈11% or less, the seedcoat was relatively impermeable because the water uptake by the seeds was shown to be significantly lower than in seeds with 14.8% initial MC (Fig. 1). By the end of day 14, a significant number of hardseeds in seed batches with ≤11% initial MC still remained. In pea (Pisum sativum), Ellis and Roberts (1982) showed that hardseededness began at or below 12% MC, and “absolute” hardseededness was detected at 9% MC. In our study, scarified winged bean seed with an initial 7% MC had much higher water uptake compared with nonscarified ones (Fig. 1). Clearly, the slow water uptake of the seeds was caused by the impermeability of the seedcoat; the barrier to permeability was removed when seeds were scarified by sandpaper. Although seeds with initial 7% MC had impermeable seedcoats, after being scarified, their water uptake on day 2 was tripled compared with that on day 1. Because the scarified spot on the seed was extremely small compared with the surface of the whole seed (<1% by observation), it is unlikely that the scarified spot was the sole entry site for the water. Rather, we speculate that as water entered through the scarified spot, the seedcoat became more permeable at some point during day 1, which then allowed a much higher water uptake rate on day 2 as the seedcoat had a much higher surface area. The exact mechanism is not known, but we hypothesize that as water was absorbed by the seed through the scarified spot, the expanding embryo was causing pressure on the seedcoat, and as a result, some of the water was absorbed by the inner side of the seedcoat during the hydration process and became more permeable. Future research is needed to determine the mechanism and dynamics of the transformation of an impermeable seedcoat into a permeable one.
Moisture content of seed with impermeable seedcoat.
The hilum of seed appeared to be responsible for the water loss of the whole seed: the MC of seeds with a blocked hilum decreased slowly from 14.8% to ≈12%, whereas MC in the control decreased steadily down to ≈5% over the same period (Fig. 3). Because the hilum is much smaller in area than the rest of the seedcoat, the seedcoat was likely remaining resistant to moisture loss as the MC only declined to 12%. This result is in agreement with Hyde (1954), who found that the rate of moisture loss of tree lupin (Lupinus arboreus L.) seeds with the hilum blocked suddenly declined when seeds were at 25% MC (dry basis). In our study, because seeds with 12% MC had close to 40% of hardseeds, it was evident that hardseededness had indeed begun to develop (Fig. 4). Thus, a seed MC of 12% could be regarded as a threshold for the seedcoats to transform from the permeable to impermeable state. In an investigation by Jaganathan et al. (2017), seeds of Indian lotus (Nelumbo nucifera) did not develop impermeable seedcoats when the MC was >10%, but seedcoat impermeability was induced when MC was <10%. In our study of winged bean seed, the drying curve for the control seed with ambient RH maintained at 3% could be used as a reference for seed drying of ‘Taitung No.1’ (Fig. 3). As moisture was absorbed from the seed, the drying capability of drying beads and silica gel decreased. This would result in different drying curves if used on seeds with a different initial MC. However, drying seed with controlled RH % technology would obtain the same drying curve regardless of initial seed MC, provided that the RH % was maintained at a constant level.
Identification of initial water entry sites.
Blocking the micropyle and lens of the seeds with Vaseline resulted in significantly more hardseeds than in the nonblocked control seeds (Fig. 4). The lens and micropyle of winged bean seed are two microstructures with much smaller area than the hilum, the area of which is also small compared with the total nonhilar region of the seedcoat (Fig. 5A, B, and E–H). Because blocking two very small parts of the seed surface had a significant effect on seed imbibition, we think it is reasonable to assume that the lens and micropyle are important sites for initial water entry. When hardseeds of Albizia julibrissim and black locust (Robinia pseudoacacia) were exposed to high temperatures, a water-gap formed in the lens region of the seed that allows seed to imbibe (Karaki et al., 2012; Yang et al., 2019). However, in our study, we did not find any structure on the whole seedcoat of winged bean seeds resembling the water-gap when exposed to dry and wet heat (data not shown). As for why some seeds with lens and micropyle blocked still imbibed normally, we offer the following speculations. First, it is likely that the region of the seedcoat in the near vicinity of the hilum, not just limited to the lens and micropyle, could be relatively permeable to water. Indeed, in a study of seeds of some Vigna species, Luan et al. (2017) observed that the area of the seedcoat around the hilum began to appear wrinkled when the seeds were soaked in water. Second, although the average MC of the seeds used in this study was 12%, there is likely to have been some unevenness in MC across the seed batch, resulting in variability in permeability from seed to seed. In the study by Jaganathan et al. (2017) of Indian lotus (Nelumbo nucifera), a seed batch with overall MC of 11% had 73% hardseeds, but further investigation revealed that the MC of the hardseeds was 8.5%, which implied that the permeable seeds had a much higher MC. In our study, it was not possible to determine the exact MC of the seeds without destroying them, thus high variation in the hardseededness of seeds could exist simply as a result of varying MC from seed to seed in a collected seed batch.
It is important to note that the use of only three replicates of 15 seeds each in our experiments was a limitation, especially if we consider the possible influence of growing environment and biological variation on seed dormancy, which is discussed shortly. Nevertheless, we do want to back up our claims with the fact that preexperiments were done numerous times and similar results were obtained before the experiments outlined in this study. It was unfortunate that we had only a limited quantity of seeds for use in the experiment. In Taiwan, growers are not able to produce plentiful supplies of winged bean seeds, as they can for more mainstream legume crops. Winged bean has a short-day requirement for flower and pod set, and thus there is only one limited season for seed harvesting. In addition, it is difficult to obtain a supply of seed that has been well stored; we often purchased seed lots that had already lost their viability.
An interesting observation in our study was that the winged bean seed harvested on different dates had differences in appearance and shape. A huge proportion of the seeds harvested in Feb. 2021 were uneven in shape, and some even had dimples on one or both cotyledons. The cotyledons in the seeds with dimples were sunken along with the seedcoat but remained intact and nondamaged, and the seeds germinated normally. However, almost no seeds harvested in March in the same year had dimples or uneven shape. The March seeds appeared more even in size and were rounder. The cause of this phenomenon is not known, but we speculate that it was related to the temperature during early stage of seed formation. Seeds harvested in February were pollinated during winter in Dec. 2020, whereas seeds harvested in March were pollinated in Jan. 2021, when the later temperature had risen considerably. Given that batches of seeds collected varied in their hardseededness, we consider it possible that distortion in shape of the seeds could play a part in their hardseededness.
Conclusion
Seeds of ‘Taitung No.1’ winged bean had low germination due to the presence of hardseeds. Hardseededness of the seeds could be effectively removed by either sandpaper or immersion in sulfuric acid immersion for 15 to 25 min, which ensured fast and high seed germination. Low seed MC was found to be crucial to hardseed development. When seed MC was reduced to ≤12%, the seedcoat began to become impermeable, and the amount of hardseeds increased as MC decreased. Lens and micropyle were speculated to be one of the water entry sites, whereas the hilum was responsible for moisture loss during drying.
Literature Cited
Baskin, C.C. & Baskin, J.M. 2014 Seeds: Ecology, biogeography, and evolution of dormancy and germination 2nd ed Elsevier San Diego, CA
Cerny, K., Kordylas, M., Pospisil, F., Svabensky, O. & Zajic, B. 1971 Nutritive value of the winged bean (Psophocarpus palustris Desv.) Br. J. Nutr. 26 2 293 299 https://doi.org/10.1079/bjn19710035
Davis, T.D., George, S.W., Upadhyaya, A. & Persons, J. 1991 Improvement of seedling emergence of Lupinus texensis Hook. following seed scarification treatments J. Environ. Hortic. 9 1 17 21 https://doi.org/10.24266/0738-2898-9.1.17
Egley, G.H 1989 Water-impermeable seed coverings as barriers to germination 207 223 Taylorson, R.B. Recent advances in development and germination of seeds. Springer Boston, MA https://doi.org/10.1007/978-1-4613-0617-7_16
Ellis, R.H. & Roberts, E.H. 1982 Desiccation, rehydration, germination, imbibition injury and longevity of pea seeds (Pisum sativum) Seed Sci. Technol. 10 501 508
Ellis, R.H., Hong, T.D. & Roberts, E.H. 1985 Preliminary seed germination and seed storage investigations with the winged bean [Psophocarpus tetragonolobus (L.) DC.] The Winged Bean Flyer 5 2 22 36
Frisbee, C.C., Smith, C.W., Wiesner, L.E. & Lockerman, R.H. 1988 Short term storage effects on dormancy and germination of chickpea (Cicer arietinum) J. Seed. Techno. 12 1 16 23 https://www.jstor.org/stable/23432692
Gama-Arachchige, N.S., Baskin, J.M., Geneve, R.L. & Baskin, C.C. 2010 Identification and characterization of the water gap in physically dormant seeds of Geraniaceae, with special reference to Geranium carolinianum Ann. Bot. 105 977 990 https://doi.org/10.1093/aob/mcq078
Geneve, R.L., Baskin, C.C., Baskin, J.M., Gehan Jayasuriya, K.M.G. & Gama-Arachchige, N.S. 2018 Functional morpho-anatomy of water-gap complexes in physically dormant seed Seed Sci. Res. 28 3 186 191 https://doi.org/10.1017/s0960258518000089
Guma, I.R., Padrón Mederos, M.A., Santos Guerra, A. & Reyes-Betancort, J.A. 2010 Evaluation of methods to remove hardseededness in Cicer canariense, a perennial wild relative of chickpea Seed Sci. Technol. 38 1 209 213 https://doi.org/10.15258/sst.2010.38.1.20
Harris, W.M 1987 Comparative ultrastructure of developing seed coats of ‘hard-seeded’ and ‘soft-seeded’ varieties of soybean, Glycine max (L.) Merr. Bot. Gaz. 148 324 331 https://doi.org/10.1086/337660
Hu, X.W., Wang, Y.R., Wu, Y.P. & Baskin, C.C. 2009 Role of the lens in controlling water uptake in seeds of two Fabaceae (Papilionoideae) species treated with sulphuric acid and hot water Seed Sci. Res. 19 73 80 https://doi.org/10.1017/S0960258509301099
Hyde, E.O.C 1954 The function of the hilum in some Papilionaceae in relation to the ripening of the seed and the permeability of the testa Ann. Bot. 18 241 256 https://doi.org/10.1093/oxfordjournals.aob.a083393
Jaganathan, G.K 2016 Influence of maternal environment in developing different levels of physical dormancy and its ecological significance Plant Ecol. 217 71 79 https://doi.org/10.1007/s11258-015-0560-y
Jaganathan, G.K., Song, D., Liu, W., Han, Y. & Liu, B. 2017 Relationship between seed moisture content and acquisition of impermeability in Nelumbo nucifera (Nelumbonaceae) Acta Bot. Bras. 31 4 639 644 https://doi.org/10.1590/0102-33062017abb0188
Kadam, S.S., Salunkhe, D.K. & Luh, B.S. 1984 Winged bean in human nutrition Crit. Rev. Food Sci. Nutr. 21 1 1 40 https://doi.org/10.1080/10408398409527395
Karaki, T., Watanabe, Y., Kondo, T. & Koike, T. 2012 Strophiole of seeds of the black locust acts as a water gap Plant Species Biol. 27 3 226 232
Kimura, E. & Islam, M.A. 2012 Seed scarification methods and their use in forage legumes Res. J. Seed Sci. 5 2 38 50 https://doi.org/10.3923/rjss.2012.38.50
Luan, Z.H., Zhao, J.M., Shao, D.K., Zhou, D.W., Zhang, L.H., Zheng, W. & Sun, Q. 2017 A comparison study of permeable and impermeable seed coats of legume seed crops reveals the permeability related structure difference Pak. J. Bot. 49 4 1435 1441 http://www.pakbs.org/pjbot/papers/1502352554.pdf
Ma, F., Cholewa, E., Mohamed, T., Peterson, C.A. & Gijzen, M. 2004 Cracks in the palisade cuticle of soybean seed coats correlate with their permeability to water Ann. Bot. 94 213 228 https://doi.org/10.1093/aob/mch133
Martens, H., Jakobsen, H.B. & Lyshede, O.B. 1995 Development of the strophiole in seeds of white clover (Trifolium repens L.) Seed Sci. Res. 5 3 171 176 https://doi.org/10.1017/s0960258500002798
Martin, I. & De la Cuadra, C. 2004 Evaluation of different scarification methods to remove hardseededness in Trifolium subterraneum and Medicago polymorpha accessions of the Spanish base genebank Seed Sci. Technol. 32 671 681 https://doi.org/10.15258/sst.2004.32.3.03
Martín, I. & Guerrero, M. 2014 Effect of sulphuric acid scarification on seed accessions of cluster clover (Trifolium glomeratum) stored in a genebank Seed Sci. Technol. 42 293 299 https://doi.org/10.15258/sst.2014.42.2.18
Miano, A.C. & Augusto, P.E.D. 2015 From the sigmoidal to the downward concave shape behavior during the hydration of grains: Effect of the initial moisture content on Adzuki beans (Vigna angularis) Food Bioprod. Process. 96 43 51 https://doi.org/10.1016/j.fbp.2015.06.007
Mira, S., Schnadelbach, A., Correa, E.C., Pérez-García, F. & González-Benito, M.E. 2017 Variability of physical dormancy in relation to seed mechanical properties of three legume species Seed Sci. Technol. 45 540 556 https://doi.org/10.15258/sst.2017.45.3.11
Nangju, D. & Baudoin, J.P. 1979 Performance of winged bean (Psophocarpus tetragonolobus (L.) DC.) Niger. J. Hortic. Sci. 54 2 129 136 https://doi.org/10.1080/00221589.1979.11514860
National Research Council 1975 The winged bean: A high-protein crop for the humid tropics National Academy of Sciences Washington, DC
Nichols, P.G.H., Foster, K.J., Piano, E., Pecetti, L., Kaur, P., Ghamkhar, K. & Collins, W.J. 2013 Genetic improvement of subterranean clover (Trifolium subterraneum L.). 1. Germplasm, traits and future prospects Crop Pasture Sci. 64 4 312 346 https://doi.org/10.1071/cp13118
Pollock, B.M. & Toole, V.K. 1966 Imbibition period as the critical temperature sensitive stage in germination of lima bean seeds Plant Physiol. 41 221 229 https://doi.org/10.1104/pp.41.2.221
Rudrapal, D., Okubo, H., Uemoto, S. & Fujieda, K. 1992 Comparison of the anatomy and physiology of seeds of two varieties of winged bean (Psophocarpus tetragonolobus) Scientia Hort. 51 13 24 https://doi.org/10.1016/0304-4238(92)90099-x
Sathe, S.K., Deshpande, S.S. & Salunkhe, D.K. 1982 Functional properties of winged bean [Psophocarpus tetragonolobus (L.) DC] proteins J. Food Sci. 47 2 503 509 https://doi.org/10.1111/j.1365-2621.1982.tb10112.x
Smýkal, P., Vernoud, V., Blair, M.W., Soukup, A. & Thompson, R.D. 2014 The role of the testa during development and in establishment of dormancy of the legume seed Front. Plant Sci. 5 351 https://doi.org/10.3389/fpls.2014.00351
Tadros, M.J., Samarah, N.H. & Alqudah, A.M. 2011 Effect of different pre-sowing seed treatments on the germination of Leucaena leucocephala (Lam.) and Acacia farnesiana (L.) New For. 42 3 397 407 https://doi.org/10.1007/s11056-011-9260-1
Tham, W.F 1980 Hardseededness in winged bean (Psophocarpus teragonolobus (L.) DC.) Proceedings of Legumes in the Tropics Universiti Pertanian Malaysia Selangor, Malysia 21 28
Van Assche, J.A., Debucquoy, K.L.A. & Rommens, W.A.F. 2003 Seasonal cycles in the germination capacity of buried seeds of some Leguminosae (Fabaceae) New Phytol. 158 2 315 323 https://doi.org/10.1046/j.1469-8137.2003.00744.x
Wang, Y.R., Hanson, J. & Mariam, Y.W. 2007 Effect of sulfuric acid pretreatment on breaking hard seed dormancy in diverse accessions of five wild Vigna species Seed Sci. Technol. 35 550 559 https://doi.org/10.15258/sst.2007.35.3.03
Yang, Y., Jaganathan, G.K., Biddick, M., Li, J. & Liu, B. 2019 Identification of the water-gap in Albizia julibrissim (Fabaceae: Mimosoideae) following wet and dry heat Plant Biosyst. 154 4 578 583 https://doi.org/10.1080/11263504.2019.1651781
Zhou, S., Sekizaki, H., Yang, Z., Sawa, S. & Pan, J. 2010 Phenolics in the seed coat of wild soybean (Glycine soja) and their significance for seed hardness and seed germination J. Agr. Food Chem. 58 10972 10978 https://doi.org/10.1021/jf102694k