Implications for restoration
Restoration practitioners frequently use D. californica to reseed oak savanna and grassland ecosystems along the West Coast of the United States, although chasmogamous seeds have until now been used almost exclusively for this purpose (Maslovat 2002, Hayes & Holl, 2011, Lindh 2018). Our results from the HD common garden suggest that cleistogamous seeds are more vigorous than chasmogamous seeds, and that northerly-sourced seeds could germinate at a higher rate than both local and southerly-sourced seeds. We cannot, however, claim that higher germination probability will necessarily translate to greater fitness advantages over the plants’ life history, or that this pattern is likely to be replicated across the Willamette Valley. Despite finding a significant seed type effect at both planting locations, more common garden site replication would be needed to suggest a regional phenomenon (but see Bischoff et al. 2006, Miller et al. 2011, Gallagher & Wagenius 2016). Future research at multiple planting sites and involving multiple life stages across several years is required to better address questions regarding seed translocation for restoration planting. Such a study should include multiple species of restoration importance in the Willamette Valley to investigate whether trends are consistent across species. This information would support ongoing efforts to create seed transfer zones of species used in the restoration of Willamette Valley ecosystems (Miller et al. 2011, Ramalho et al. 2017).
On a practical note, the methods we used to facilitate cleistogamous seed preparation substantially reduce the processing time for outplanting cleistogamous seeds. Our sheathed and soaked manipulations did not affect germination success for either cleistogamous or chasmogamous seeds, demonstrating that these methods can be used to make cleistogamous seed planting an accessible complement to chasmogamous planting. Although post-soaked individual cleistogamous seed extraction is a laborious process when done by hand, planting groups of cleistogamous seeds within their intact stems is much less labor-intensive. Our sheathed treatment demonstrates that planting cleistogamous seeds that are still enclosed within their stem can be an easy method to successfully germinate cleistogamous D. californica seeds at a scale necessary for ecosystem restoration. Restoration practitioners may benefit from incorporating cleistogamous seed planting as an insurance policy in the event of reduced chasmogamous germination in much the same way that the plants themselves do (Zeide 1978, Schoen & Loyd). This practice may be especially beneficial when local seed sourcing ability is limited.
Although the detailed pathogen census performed by Mackin et al. (2021) demonstrates the differences in cleistogamous and chasmogamous pathogen communities, we were unable to assess the impact pathogens had on in-situ germination and thus cannot be certain that pathogen escape explains the intermediate distance advantage in cleistogamous seeds at the HD common garden. Still, because pathogen pressure is likely a factor influencing in situ D. californica germination, and because pathogen communities vary between nearby sites, plug planting of larger individuals could also present an effective method of restoring D. californica populations, provided sterilized soil is used. Although this method is more time consuming and expensive than direct re-seeding (Gallagher & Wagenius 2016), accurately predicting the efficacy of re-seeding approaches may require an intimate understanding of epiphytic, endophytic and soil pathogen communities on both seed types, which was beyond the scope of this study. When seeding D. californica for restoration, practitioners should consider the pathogen communities of both source and planting sites, use cleistogamous seeds in addition to chasmogamous seeds, and consider sourcing cleistogamous seeds from more distant northerly populations than their chasmogamous counterparts.