Background and aims: Serotiny is common in the genus Banksia, so any seed collection is likely to be comprised of seeds that were produced in many different years. This study aimed to determine the impact of cone age and degree of serotiny on longevity in ex situ storage.
Methods: Cones of identifiable age classes were collected from three species of Banksia. Seeds were extracted from cones and the degree of serotiny calculated. An estimate of initial viability (K(i)), the time for viability to fall by one probit (σ) and the relative longevity of seeds (p(50)) for each species and cone age class was determined using a comparative longevity test (50 °C, 63 % relative humidity).
Key results: The degree of serotiny ranged from moderate (7·9) for Banksia attenuata to strong (40·4) for B. hookeriana. Survival curves for all seed age classes within each species could be described by regressions with a common slope (1/σ), but with different values for K(i). The time taken for viability to fall by one probit (σ) could be described by a common value (29·1 d) for all three species.
Conclusions: Differences in seed longevity between cone age classes and species was related to variation in initial viability (K(i)) rather than to differences in σ. While targeting the youngest mature seed cohort on a plant will maximize the viability of seeds collected, a wide range of age classes should be collected (but stored as separate cohorts if possible) for quality conservation/restoration seed collections where genetic diversity is important.