Introduction

Seed dormancy is a key adaptive trait that controls the timing of germination and hence seedling establishment in variable environments (Baskin and Baskin, 2014). Among dormancy classes, physical dormancy (PY) caused by a water-impermeable seed coat is especially common in the Fabaceae and requires disruption of the seed coat (by abrasion, thermal shocks, or specific water-gap openings) before imbibition can occur (Baskin et al. 2000;Geneve et al. 2018). From an evolutionary perspective, the deployment of PY across lineages reflects adaptive responses to environmental cues and risk‐spreading strategies in seed plants (Willis et al. 2014). Moreover, recent research has emphasized the dynamic nature of PY: permeability may be seasonally altered by temperature fluctuations, moisture dynamics, or heat pulses, thereby permitting rapid water uptake and initiating germination (Lamont and Pausas, 2023;Morrison et al. 1998). Understanding these mechanisms is therefore important not only for seed ecology and the evolution of life‐history strategies but also for practical propagation and restoration of rare species.

Gueldenstaedtia verna (Georgi) Boriss., a member of the genus Gueldenstaedtia in the Fabaceae, is a rare perennial herb native to Korea and classified as Near Threatened (NT) under the Korean Red List. It is the only Korean representative of its genus, occurring in highly restricted populations in Gyeongsangbuk-do and Daegu Metropolitan City. This species, known for its small stature (5–10 cm) and violet papilionaceous flowers blooming in mid-summer, holds both conservation and ornamental value (KNA, 2025;Lee and Kim, 2017). However, germination-based propagation has been limited by low germination rates, likely due to intrinsic dormancy barriers associated with its hard, colliculate seed coat a characteristic typical of Fabaceae with physical dormancy (Geisler et al. 2016). Given the potential for both ex situ propagation and in situ restoration of remnant populations, clarifying whether PY alone or a combination of dormancy mechanisms operates in G. verna is a priority.

The present study therefore (1) assesses seed morphology and water‐imbibition dynamics to verify the presence of physical dormancy, (2) evaluates the effectiveness of different mechanical and chemical scarification treatments for dormancy release, and (3) investigates germination responses under various constant, alternating, and seasonally simulated temperature regimes. By linking mechanistic dormancy traits to practical dormancybreaking protocols, we aim to provide both ecological insight and actionable guidance for conservation, seed banking, and the sustainable utilization of this endemic legume species.

Materials and Methods

Seed Collection and Storage

Mature seeds of Gueldenstaedtia verna were collected on 23 May 2023 from Dodong, Dong-gu, Daegu Metropolitan City, Republic of Korea. After collection, the seeds were air-dried at room temperature (22–25°C) for two weeks, carefully selected, and placed in plastic containers with silica gel packets. The dried seeds were then stored in a cold chamber at 4°C until further experimental analyses.

Water Imbibition Test

To examine the impermeability of the seed coat and determine the presence of physical dormancy, water uptake tests were conducted at 20–25°C. Two sheets of filter paper (ADVANTEC No. 2, Ø 90 mm, Japan) were placed in 90 × 15 mm Petri dishes (Cat. No.: DH.W30015, DAIHAN Scientific, Wonju, Korea), and 20 seeds per replicate (three replicates) were positioned on the moistened paper. Distilled water was added sufficiently to ensure that the seeds were fully submerged, and additional water was supplied as needed to maintain adequate moisture. Seed mass was measured at 0, 3, 6, 9, 12, 24, 48, and 72 h after imbibition. The water uptake rate was calculated using the formula W_s (%) = [(W_h-W_i)/W_i] × 100, where W_s is the increased seed weight, W_h is the hydrated seed weight, and W_i is the initial seed weight.

Effect of scarification and temperature on germination

To determine the optimal germination temperature, both scarified and non-scarified seeds were incubated under different temperature regimes. Non-scarified seeds were tested at 4°C, 15/6°C, 20/10°C, and 25/15°C, whereas scarified seeds were incubated only under the two contrasting temperature regimes (4°C and 25/15°C). performed by needle prick scarification, making small punctures approximately 0.1 mm deep on the seed coat.

To simulate seasonal temperature fluctuations, two sequential temperature shift experiments (move-along tests) were performed: Move A & C (from winter to summer: 4°C (12 weeks) → 15/6°C (4 weeks) → 20/10°C (4 weeks) → 25/15°C (12 weeks)) and Move B & D (from summer to winter: 25/15°C (12 weeks) → 20/10°C (4 weeks) → 15/6°C (4 weeks) → 4°C (12 weeks)). In these experiments, Move A and Move B were conducted with non-scarified seeds, whereas Move C and Move D involved seeds subjected to needle prick scarification. Seeds were transferred stepwise according to the designated temperature schedule. All experiments were conducted in programmable incubators (JSMI-04CPL, JS Research Inc., Gongju-si, Korea) with a constant 4°C for low-temperature treatment, and alternating 15/6, 20/10, and 25/15°C for moderate- and high-temperature conditions. Chambers were maintained under alternating 12 h light / 12 h dark conditions, illuminated by cool white fluorescent lamps providing a photosynthetic photon flux density of 21.61 μmol·m^-2·s^-1.

To simulate seasonal temperature fluctuations, two sequential temperature shift experiments (move-along tests) were performed: Move A & C (from winter to summer: 4°C (12 weeks) → 15/6°C (4 weeks) → 20/10°C (4 weeks) → 25/15°C (12 weeks)) and Move B & D (from summer to winter: 25/15°C (12 weeks) → 20/10°C (4 weeks) → 15/6°C (4 weeks) → 4°C (12 weeks)). In these experiments, Move A and Move B were conducted with non-scarified seeds, whereas Move C and Move D involved seeds subjected to needle prick scarification. Seeds were transferred stepwise according to the designated temperature schedule. All experiments were conducted in programmable incubators (JSMI-04CPL, JS Research Inc., Gongju-si, Korea) with a constant 4°C for low-temperature treatment, and alternating 15/6, 20/10, and 25/15°C for moderate- and high-temperature conditions. Chambers were maintained under alternating 12 h light / 12 h dark conditions, illuminated by cool white fluorescent lamps providing a photosynthetic photon flux density of 21.61 μmol·m^-2·s^-1.Scarification Treatments

Scarification Treatments

To break the hard seed coat, three types of scarification treatments were applied: (1) mechanical scarification using 320-grit sandpaper to a depth of approximately 0.1 mm, (2) needle prick scarification by piercing the seed coat with a syringe (KOVAX 1 ml, KOVAX-SYRINGE, HANBAEK, Bucheon, Korea) needle to a depth of approximately 0.1 mm, and (3) chemical scarification by immersing seeds in 95% sulfuric acid (H₂SO₄, 95.0%, SAMCHUN Chemical Co., Ltd., Pyeongtaek, Korea) for 5 or 10 minutes. Before scarification, seeds were surface-sterilized, and after the scarification treatments, they were incubated at 25/15°C.

Seed Disinfection

Prior to germination test, seeds were thoroughly rinsed with distilled water at least three times and submerged in 50 mL of 2000 mg·L^-1 benomyl solution (Benomyl, FarmHannong, Seoul, Korea). The suspension was stirred continuously at 350 rpm for 24 hours using a magnetic stirrer (MSH20D, DAIHAN-Scientific). After soaking, the seeds were rinsed again with distilled water.

Germination Evaluation and Statistical Analysis

Germination was conducted in moistened Petri dishes sealed with Parafilm (Bemis Company Inc., Oshkosh, WI, USA) to prevent desiccation. Seeds were considered germinated when the radicle reached 1 mm in length. Each treatment included 20 seeds per replicate with three replicates. All statistical analyses were performed using SAS software version 9.4 (SAS Institute Inc., Cary, NC, USA). One-way ANOVA was applied, followed by Duncan’s multiple range test for mean separation at α = 0.05. Data visualization was carried out using Sigma Plot version 10.0 (Grafiti LLC, Palo Alto, CA, USA).

Results and Discussion

The water imbibition experiment of Gueldenstaedtia verna seeds for 72 h revealed a water permeability of 5.1 ± 2.6% relative to the initial seed weight (Fig. 1). This indicates the presence of a physical barrier in the seed coat that prevents water penetration (Fig. 5a,c), suggesting that the seeds possess physical dormancy (PY). Such a low water permeability is consistent with the threshold reported for seeds with physical dormancy (PY) in Fabaceae species, which generally absorb less than 10–20% of their initial weight under standard imbibition conditions (Baskin & Baskin, 2000, 2014). Similarly, in the Fabaceae species Astragalus koraiensis, the increase in water content relative to the initial seed weight was reported as 9.2 ± 4.1%, indicating restricted water uptake pathways (Jang et al. 2024). Some Fabaceae seeds exhibit limited water permeability due to the hardness of the seed coat, reinforced by lignin, cutin, wax, and densely packed cell layers (Baskin, 2003;Baskin and Baskin, 2000;Geisler et al. 2016). The anatomical characteristics of the seed coat and their role in dormancy are ecologically significant, as they determine the timing of germination and seedling establishment.

Temperature is an important environmental factor for relieving seed dormancy under natural conditions, promoting successful germination at appropriate times and enhancing seedling survival (Geneve, 2003). In this study, non-scarified seeds incubated at four different temperature regimes (4, 15/6, 20/10, 25/15℃) did not germinate (Fig. 2a), and the seasonal-mimicking temperature regime (move-along test) also failed to induce germination (Fig. 3). In contrast, mechanical scarification significantly increased germination. Seeds punctured with a syringe reached a final germination rate of 98.3% within 1 week under the Move A treatment (Fig. 3, Fig. 5d), and in Move C, radicle emergence was not observed during the 12 weeks at 4 °C but increased sharply once the temperature shifted to 15/6°C, with most seeds germinating at 20/10°C (Fig. 3). Likewise, when the same mechanical scarification was applied, seeds incubated at 25/15℃ reached 100% germination within 1 week (Fig. 2b, Fig. 5e), indicating that these relatively high temperatures promote germination, whereas lower temperatures (4℃) are not within the optimal temperature ·range for germination, even when physical dormancy is relieved.

To evaluate the effect of scarification type at 25/15℃, four treatments sandpaper, needle prick scarification, 95% sulfuric acid for 5 min, and 95% sulfuric acid for 10 min were applied, resulting in germination rates of 100%, 98.1%, 20.0%, and 58.3%, respectively (Fig. 4). Sandpaper and needle prick scarification achieved 100% final germination within 1 week, consistent with the temperature-dependent data (Fig. 2b and Fig. 4). In sulfuric acid treatments, 10 min exposure was more effective than 5 min (Fig. 5b). Similarly, Rhie et al. (2016) reported that in Lespedeza tomentosa, non-scarified seeds exhibited only 6% water uptake, and acid treatment (90– 180 min) was required to overcome dormancy.

Combinational dormancy, in which physical dormancy (PY) coexists with physiological dormancy (PD), requires a secondary process such as exposure to high or low temperatures to break physiological dormancy even after PY has been alleviated (Baskin and Baskin, 2014). Several Fabaceae species are known to exhibit combination dormancy (Soltani et al. 2020;Van Assche and Vandelook, 2010). In contrast, G. verna seeds appear to germinate immediately after PY is relieved, without any subsequent inhibition, suggesting that they lack physiological dormancy. The evolution of physical dormancy, depending on its extent, appears to protect seeds from predation and contribute to their long-term survival under field conditions (Daling et al. 2011;Paulsen et al. 2013;Souza and Marcos Filho, 2001). Physical and anatomical traits that can detect environmental cues may regulate the intensity of dormancy, which may require diverse strategies. Future studies should aim to characterize seed coat structure in detail, elucidate the mechanisms of water-gap formation and opening, and identify environmental triggers, such as temperature and moisture fluctuations, that regulate dormancy release in natural habitats. Investigating how seed morphology, dormancy intensity, and ecological conditions interact would further clarify the adaptive significance of physical dormancy and support optimized propagation protocols for both in situ and ex situ conservation of this valuable species.