Plants are going extinct tens to hundreds of times faster than the background rate (Humphreys et al., Reference Humphreys, Govaerts, Ficinski, Nic Lughadha and Vorontsova2019), and Hawai‘i has the largest number of recorded plant extinctions (Humphreys et al., Reference Humphreys, Govaerts, Ficinski, Nic Lughadha and Vorontsova2019; Wood et al., Reference Wood, Oppenheimer and Keir2019). This is a result of climate change, invasive species and other anthropogenic effects (Sakai et al., Reference Sakai, Wagner and Mehrhoff2002). The need for effective plant conservation is more urgent than ever, and seed banking has been shown to be the most cost-effective and efficient means of ex situ conservation of plant germplasm (Guerrant et al., Reference Guerrant, Havens and Maunder2004; Cochrane et al., Reference Cochrane, Crawford, Monks, Cochrane, Crawford and Monks2007). Under conventional seed banking standards, seeds are desiccated to 10–25% relative humidity (RH), hermetically sealed, and then stored at −18 ± 3 °C (FAO, 2014). Under these conditions, seed lifespan is predicted to be extended decades to centuries (Ellis & Roberts, Reference Ellis and Roberts1980; Walters et al., Reference Walters, Wheeler and Grotenhuis2005).
Ambient herbarium conditions are inappropriate for long-term seed storage and retention of high germinability (e.g. Ellis & Roberts, Reference Ellis and Roberts1980). Nonetheless, herbaria have been suggested as overlooked repositories of viable seeds (Godefroid et al., Reference Godefroid, Van de Vyver, Stoffelen, Robbrecht and Vanderborght2011), and a global priority list of de-extinction candidate species using herbarium specimens was recently created (Albani Rocchetti et al., Reference Albani Rocchetti, Carta, Mondoni, Godefroid, Davis and Caneva2022a). Germination trials using seeds stored at ambient herbarium conditions have been undertaken with varying success. Schiedea kauaiensis (Caryophyllaceae) germinated after 3 years at ambient herbarium conditions (Wolkis et al., Reference Wolkis, Jones, Flynn, DeMotta and Rønsted2022). Common and rare species of Fabaceae germinated after 13 (five species of Acacia; Doran et al., Reference Doran, Turnbull, Boland and Gunn1983), 50 (Prosopis juliflora; Ffolliot & Thames, Reference Ffolliot and Thames1983) and 97 (Astragalus neglectus; Bowles et al., Reference Bowles, Betz and DeMauro1993) years, and Bupleurum tenuissimum (Apiaceae) germinated after 144 years in ambient herbarium conditions (Godefroid et al., Reference Godefroid, Van de Vyver, Stoffelen, Robbrecht and Vanderborght2011). However, no seeds germinated in a study of Metrosideros polymorpha (Myrtaceae) harvested from herbarium specimens 4–38 years old (Wolkis & Deans, Reference Wolkis and Deans2019).
Sesbania tomentosa Hook. & Arn. (Fabaceae), or ‘ōhai in ʻlelo Hawaiʻi (the Hawaiian language), is federally listed as endangered in the USA. This shrub, endemic to the Hawaiian Islands, is primarily adapted to coastal and dry lowland areas (Wagner et al., Reference Wagner, Herbst and Lorence2005; Cole & Morden, Reference Cole and Morden2021; USFWS, 2023). It has been proposed for categorization as Vulnerable on the IUCN Red List as there are < 1,000 individuals (D1) and there is an estimated ongoing decline of 10% in the next 10 years (C1) (Rønsted, Reference Rønstedproposed). Sesbania tomentosa is listed as orthodox in the seed information database (SER et al., 2023), and likely orthodox in the Hawaiʻi Seed Bank User's Guide (Hawaiʻi Seed Bank Partnership, 2023) based on an ex situ longevity assessment of 295 species in the Hawaiian flora (Chau et al., Reference Chau, Chambers, Weisenberger, Keir, Kroessig and Wolkis2019). In 2021, a sealed 4-l glass jar containing silica gel crystals (a commonly used desiccant) indicating ambient RH (c. 55%) and 12 lots of S. tomentosa seeds collected during 1990–1994, with passport data, was discovered in herbarium PTBG (the Herbarium of the National Tropical Botanical Garden; NTBG). We assessed the germinability of each seed lot to investigate longevity and to determine whether accessioning each lot into the NTBG Seed Bank and Laboratory was warranted.
As the seeds were stored in a jar and not physically attached to specimens, we did not consider the decision framework for harvesting seeds from herbarium specimens (Albani Rocchetti et al., Reference Albani Rocchetti, Davis, Caneva, Bacchetta, Fabrini and Fenu2022b). For each accession, one replicate of 15 randomly selected seeds was sown. Inferred from a popular gardening handbook (Lilleeng-Rosenberger, Reference Lilleeng-Rosenberger2005), Baskin & Baskin (Reference Baskin and Baskin2014) concluded that S. tomentosa exhibits physical dormancy. In such seeds, embryos are fully developed and the seed (or fruit) coat is impermeable to water (Baskin et al., Reference Baskin, Baskin and Li2000). Seeds were therefore scarified with a single-edged razor blade opposite the hilum to allow the seeds to imbibe. They were then placed in 60 mm Petri dishes on seed-germination paper (Anchor Paper Company, Saint Paul, USA) and placed in a sealed chamber above a saturated potassium sulphate solution (achieving an RH of c. 98% at 25 °C) overnight to avoid imbibition damage. The seeds were removed from the 98% RH sealed chamber and moved to a germination chamber between 11 and 15 November 2022. The seeds and seed-germination paper were moistened with a solution of 0.1% plant preservative mixture (Plant Cell Technology, Washington, DC, USA) in distilled water, to inhibit fungal growth without affecting germination (A. Guri, pers. comm., 2017). Dishes were then sealed with plastic paraffin film to reduce water loss. Seeds were exposed to daily alternating regimes of 12 h light (c. 41 mmol/m2/s cool white (4100 K) fluorescent light) and 12 h dark at corresponding temperatures of 25/15 °C in a germination chamber (GR36L, Percival Scientific, Perry, USA). Seeds were monitored every 7–14 days until every seed either germinated or was parasitized by fungus, or until 6 months had elapsed since sowing. Germination was defined as radicle emergence, and all resulting seedlings as well as non-germinated seeds were transferred to the NTBG Conservation Nursery for further propagation. Per cent germination was calculated by dividing the number of seeds germinated by the number of seeds sown and multiplying by 100.
Germination was observed after 7 days and had ended by day 34. Mean final germination was 88.9 ± SD 0.1 % (range 73–100%) (Table 1). The seedlings that were transferred to the NTBG Conservation Nursery appeared healthy and vigorous, and as of 6 September 2023, at least three plants per accession had survived, most of which were already flowering (R. Campbell, NTBG Nursery Manager, pers. comm., 2023).
The high germination rates indicate the seeds are of high conservation value, and therefore each seed lot was accessioned into the NTBG Seed Bank and Laboratory for conservation and continued longevity monitoring. This resulted in 100–3,101 seeds banked per accession, totalling 11,639 seeds (Table 1). An assessment of longevity in the Hawaiian flora that included data on two accessions of S. tomentosa stored for up to 15.5 years at a target equilibrium relative humidity (eRH) of c. 20% and −18 °C found that germinability declined to ≥ 30% of maximum (the re-collection threshold suggested by the authors of that study) in 5–10 years (Chau et al., Reference Chau, Chambers, Weisenberger, Keir, Kroessig and Wolkis2019). This is in contrast with our findings of up to 100% germinability in seeds stored > 32 years, which could at least triple the re-collection threshold for the species. In other accessions of S. tomentosa seeds stored at −18 °C in the NTBG Seed Bank, 49 of 50 seeds germinated after 25 years (accession no. 960201), and 15 of 15 seeds germinated after 20 years (accession no. 010019) (D. Wolkis, unpubl. data). However, neither Chau et al. (Reference Chau, Chambers, Weisenberger, Keir, Kroessig and Wolkis2019) nor ourselves analysed survival curves (e.g. Ellis & Roberts, Reference Ellis and Roberts1980). To understand the longevity trait fully, which is important for management of the species, seed survival curves and p 50, the time during storage when germinability has declined to 50%, should be investigated (Hay et al., Reference Hay, Davies, Dickie, Merritt and Wolkis2022).
It has been suggested that for some species exhibiting physical dormancy characterized by a water-impermeable seed coat (hard seededness), drying induces the impermeable trait (Jaganathan et al., Reference Jaganathan, Li, Biddick, Han, Song and Yang2019, and references therein). The unique anatomy of a water-impermeable seed coat allows the release of water from the seed (desorption), but water cannot be absorbed until the seed coat is damaged (i.e. a water gap is opened; Hay et al., Reference Hay, Rezaei, Wolkis and Mcgill2023). Although the desiccant inside the jar with the seeds indicated ambient RH (c. 55%), it was probably much lower when the jar was sealed. As S. tomentosa seeds have water-impermeable seed coats (Baskin & Baskin, Reference Baskin and Baskin2014), it is likely that the eRH of the seeds remained at this lower level, which could explain the high germinability observed after 30 years.
Many studies of seeds stored at herbarium conditions report poor germination. For example, only 1 in > 2,000 seeds harvested from herbarium specimens of rare Hawaiian species germinated (Wolkis et al., Reference Wolkis, Jones, Flynn, DeMotta and Rønsted2022); no seeds of M. polymorpha harvested from herbarium specimens 4–38 years old germinated (Wolkis & Deans, Reference Wolkis and Deans2019); and only eight seeds germinated (all from a single species) in > 2,600 sown (Godefroid et al., Reference Godefroid, Van de Vyver, Stoffelen, Robbrecht and Vanderborght2011). Our finding of high germinability of S. tomentosa seeds suggests nevertheless that some species can be resurrected from seeds stored in overlooked repositories. This should, however, be carried out as soon as possible, before viability declines further, and if harvesting from herbarium specimens directly, the prudent use of seeds should be considered (e.g. Albani Rocchetti, Reference Albani Rocchetti, Davis, Caneva, Bacchetta, Fabrini and Fenu2022b). Although we found high germinability at ambient conditions after 30 years, we would expect greater longevity (i.e. p 50 sensu Hay et al., Reference Hay, Davies, Dickie, Merritt and Wolkis2022) for orthodox seeds that were dried, hermetically sealed, and stored at low temperatures (i.e. conventional storage conditions; Ellis & Roberts, Reference Ellis and Roberts1980, FAO, 2014). Therefore, we recommend depositing seeds into seed banks where facilities and expertise can optimize longevity.
Author contributions
Conceptualization: ES, DW; germination studies: ES; statistical analysis: DW; writing, revision: all authors.
Acknowledgements
We thank the original collectors of the seeds Heidi Bornhorst, Steve Perlman, C. Rowland, Kerin E. Lilleeng-Rosenberger and those not listed; whoever had the foresight to conserve the seeds 30 years ago; PTBG Herbarium Curator Tim Flynn for finding the seeds; NTBG volunteers Jeffrey Frelinger and Leslie Ridpath for assistance in monitoring germination; NTBG Nursery personnel Rhian Campbell and Hayley Walcher for growing seedlings to mature plants; and Kupu/AmeriCorps for the opportunity for ES to study within the Conservation Leadership Development Program. NTBG provided institutional support. This research received no specific grant from any funding agency, or commercial or not-for-profit sectors.
Conflicts of interest
None.
Ethical standards
The research was carried out under Hawaiʻi State Department of Land and Natural Resources for Natural Area Reserve, Rare Plant, and Native Invertebrate Research Permit Numbers I2986 and I5371, and otherwise abided by the Oryx guidelines on ethical standards.
Data availability
The data supporting the findings of this study are available within the article.