Assessing Extreme Seed Longevity: The Value of Historic Botanical Collections to Modern Research.

Botanical, historical, and archaeological collections have been the source of extraordinarily long-lived seeds, which have been used to revive extinct genotypes or species. The longest-lived example of a viable seed of known age is the date palm, Phoenix dactylifera L., of which an estimated 2000-year-old seed was germinated in 2005. Seed longevity is important for agriculture and biodiversity conservation, and understanding the basis for the extraordinary longevity of seeds from botanical collections could help improve seed banking technology. In this work, we studied the viability and structural features of date palm seeds collected in Baghdad in 1873 and stored in the Economic Botany Collection (EBC) at the Royal Botanic Gardens, Kew, and seeds collected in 2004 and stored dry at -20°C in the Millennium Seed Bank (MSB). Viability was studied by attempted seed germination and in vitro culture of embryos, and structural features were studied by X-rays, transmission electron microscopy, and differential scanning calorimetry. We found that the seeds preserved in the MSB did not decrease in viability, with ultrastructural features similar to those in freshly harvested seeds. In contrast, the 144-year-old seeds were dead, and large ultrastructural changes were observed, particularly in the storage lipids (size, distribution, and melting properties) and other storage constituents. These results contrast with previous reports that date seeds could remain viable for ∼2000 years in uncontrolled storage environments. We did not find that the postharvest treatment of the EBC seeds in the 19th century, or their storage conditions at Kew, was more deleterious than that which was likely encountered by the ∼2000-year-old seeds. These results highlight the role of well-documented collections in establishing whether reports of extraordinary longevity are ordinarily repeatable.