For many species, there is no data on their seed storage behavior. In these cases, you could test the seeds yourself. The most complete protocol is the one published by Hong and Ellis (1996), “A protocol to determine seed storage behaviour.”
The full protocol requires testing the germination percentage at various drying stages, after measuring their moisture content through the oven method, and then after cooling and freezing for various lengths of time. Obviously, performing this test means using thousands of seeds and requires a lot of time. Other methods exist, including a simplified protocol proposed by Hong and Ellis (1996), where seed germination is tested on freshly collected seeds after drying at 10%–12% moisture content (MC), then at 5% MC, and then after freezing at −20°C for three months.
The Kew Millennium Seed Bank’s 100-seed test for desiccation tolerance uses only 100 seeds to determine if seeds tolerate drying to 5% MC, without testing their freezing response or partial desiccation tolerance. If the seeds are very small, you’ll need an accurate precision balance (at least four decimal places).
Intermediate seed storage behavior may be more prevalent among tropical and subtropical species. If seeds are freeze-sensitive, or do not tolerate freezing temperatures, this behavior may not be detected until two or more years of storage (due to very gradual compression of seeds’ cell contents over time). This is why we recommend using the protocol suggested by Chau et al. (2019) in their paper “Seed freeze sensitivity and ex situ longevity of 295 species in the native Hawaiian flora,” which includes testing both the drying and freezing responses up to two years after storage.The authors identified that many freeze-sensitive seeds can still be stored at refrigerated temperatures (approx 5°C) for several years.