Abstract
Development projects that impact Arctostaphylos morroensis (Morro manzanita), a federally listed threatened species, typically require habitat restoration, mitigation or both. Replacement plantings for habitat restoration and mitigation can be obtained by establishing new plants from cuttings or seed propagation. Propagation of A. morroensis from seed has the advantage of increasing genetic diversity but typically has low germination rates. We investigated various propagation techniques with the goal of determining the most efficient and effective approach for generating A. morroensis from seeds in a greenhouse setting. We compared emergence rates for fruits from two sources, shrubs and leaf litter; examined germination in response to various scarification treatments including fire, heat, acidification, and no treatment; and compared germination in response to stratification, including cold and warm storage. We found that shrub-sourced fruits, and acidification and fire (gas torch) scarification yielded the best results. We also found that a warm stratification period worked best for shrub-sourced fruits and a cold stratification period worked best for litter-sourced fruits. Identifying the most efficient propagation methods can increase success and decrease costs associated with restoring and establishing healthy A. morroensis communities, and serve as a potential model for use in other manzanita species.
Restoration Recap
Arctostaphylos morroensis is a federally threatened coastal maritime chaparral species for which there is limited knowledge on propagation techniques for developing restoration plantings from seed.
Successful propagation from seed requires scarification by oven-heating, fire, or acidification to break seed coat dormancy.
Propagation of A. morroensis from shrub-sourced fruits is preferable; however, litter sourced fruits can also provide an adequate source for restoration projects.
From scarification to first emergence may take three to four months in A. morroensis and likely other manzanitas as well.
Habitat restoration and mitigation projects that include revegetation efforts generally rely on direct seeding, or planting container stock purchased from nurseries or generated in greenhouses, or both (Ceccon et al. 2016). Directly seeding sites is relatively inexpensive and quick (Young and Evans 2000, Palmerlee and Young 2010). However, installing container stock can greatly increase the likelihood of successfully establishing targeted species, especially those that have dormant seeds requiring specific germination cues (Baskin and Baskin 2020), and thereby shorten the time period to attain ecologically functioning vegetation communities (Young and Evans 2000, Bean et al. 2004). Thus, effective and efficient propagation methods are an important tool for achieving successful restoration outcomes. This is frequently the case for rare plant species, which may have more limited seed or stock due to constraints on seed collection (e.g., natural scarcity of reproductive adults, or conservation regulations) or low rates of success with germination (Carlson and Sharp 1975, Schemske et al. 1994, Menges et al. 2004, Baskin and Baskin 2004, Herrera and Takara 2006).
Arctostaphylos morroensis (Morro manzanita) is a rare, coastal maritime chaparral shrub species that is endemic to the vicinity of Los Osos in west San Luis Obispo County, California. It occurs in coastal dune scrub, maritime chaparral, and coast live oak woodland habitats, and most stands grow in Baywood fine sand (Soil Conservation Service 1984, USFWS 2022). Arctostaphylos morroensis is an obligate-seeder, meaning that following fire it does not resprout when the crown and stem are burned, and thus naturally occurring stands must re-establish from seeds in the soil seed bank. The species was listed as threatened by the U.S. Fish and Wildlife Service in 1994 due to threats from urban and residential development, habitat fragmentation, competition from non-native species, lack of protection on private lands and continued fire suppression (USFWS 1994). At that time, it occupied approximately 340 to 360 hectares, which is roughly one-third of its historic range (Mullany 1990). Estimates from USFWS 5-year reviews and independent efforts suggest the geographic range has remained stable since that time (USFWS 2022, Tyler and Kofron 2024).
As a federally listed threatened species, A. morroensis is subject to regulatory requirements under the federal Endangered Species Act (ESA). The ESA requires development project proponents to participate in conservation and to ensure that project activities will not jeopardize the continued existence of a listed species. Therefore, projects that involve unavoidable temporary and permanent impacts to A. morroensis may be required to restore temporarily impacted communities and mitigate for permanent impacts through the creation of new A. morroensis habitat. Nursery generation of manzanitas from seed typically has exceptionally low rates of success (Everett 1957, Berg 1974, Carlson and Sharp 1975, Herrera and Takara 2006). This is due to challenges in breaking seed dormancy (Keeley 1991, Keeley et al. 2005) as well as low seed viability (Tyler and Odion 2020). Establishing new plants from cuttings, while also challenging, can achieve higher rates of success, but eliminates the genetic diversity provided by propagation from seed. Determining the most efficient and effective seedling propagation techniques should improve the success of future habitat restoration and mitigation projects for A. morroensis, and may provide insight for propagation of other Arctostaphylos species.
One of the primary hurdles to producing seedlings in this and related obligate-seeding species is determining the most effective method for breaking seed dormancy. Seeds of obligate seeders are mainly or even completely refractory (Sweeney 1956, Keeley 1987, Keeley 1991); that is, germination is inhibited until primary dormancy is released by a specific mechanism. Two processes may contribute to overcoming delayed germination: scarification to soften or weaken an impermeable seed coat, and stratification, e.g., exposure to a period of cold temperature, to overcome internal dormancy (Berg 1974, Bonner et al. 1974). Previous studies with manzanita species have examined scarification treatments that attempt to mimic natural processes including fire (Emery 1988, J. Sayers personal communication), heat (Tyler and Odion 2020), smoke (Keeley et al. 2005, Jurado et al. 2011), acidification (Emery 1988), charred wood, use of wet or dry seeds, and other combinations of factors (Jurado et al. 2011, Tyler and Odion 2020). The importance of cold stratification in breaking dormancy has not been examined experimentally with A. morroensis, though prior studies on manzanita seed germination have often included a cold stratification period (Keeley 1987, Emery 1988, Tyler and Odion 2020).
Another consideration for successful propagation is seed source, which is consequential in terms of availability, genetics (e.g., local ecotypes), viability, and specific germination cues required. Prior germination and viability studies on A. morroensis have been conducted on seed from soil seedbanks (Odion and Tyler 2002, Tyler and Odion 2020). In those studies, the goal was to understand post-fire natural seedling recruitment, where any seed source that was not buried in the soil would be consumed in a burn. However, fresh and litter-stored seeds might be especially useful in restoration and thus, study of their germination requirements and success rates in producing seedlings is warranted.
Previous studies have focused on germination of individual seeds, cultured on a growth medium or other surface to allow detection of initial radical protrusion from the seed coat (Odion and Tyler 2002, Tyler and Odion 2020). This approach is essential for accurately determining germination rates. However, it is neither efficient nor cost-effective for the propagation of seedlings for restoration needs. A. morroensis fruits are drupes, covered by a thin exocarp, and containing a dry, mealy mesocarp surrounding multiple hard stones or “nutlets” that are unfused to highly adherent (Berg 1974, Parker et al. 2012). It is highly labor intensive to extract individual seeds from fresh fruits by separating them from the exo- and mesocarp, and then attempting to separate adhered nutlets. A more efficient approach employed by some practitioners (J. Sayers, personal communication) is to subject whole fruits to a scarification technique, manually crush the fruits to loosen the seeds and facilitate their separation, and then sow this material in shallow nursery flats. With this method, the outcome is measured as seedlings emerged rather than germination rate, as some seeds may germinate but fail to survive or produce a shoot. However, since the goal of propagation for restoration is to produce seedlings, emergence (of a shoot) is the most practical and relevant outcome. In addition, this approach makes comparisons using the number of fruits per treatment rather than an estimated number of seeds, which provides more useful guidelines to restoration practitioners for the targeted numbers of fruits recommended for field collection efforts.
We conducted a multi-factorial experiment focused on determining the most efficient method of propagating A. morroensis for the purpose of generating restoration plantings, while also hoping to advance the understanding of its ecological processes. Accordingly, we compared emergence rates using litter- and shrub-sourced fruits under warm and cold stratification that were exposed to scarification treatments of fire, heat, and acidification. We were uncertain whether shrub- or litter-sourced fruits would have better germination success, given that littersourced fruits may be dried out or damaged, but also may have benefited from natural scarification processes breaking their internal seed coat dormancy. We hypothesized that stratification, particularly cold stratification, would increase germination success, and that all scarification treatments would increase germination rates compared to controls.
Methods
Fruit Collection
Arctostaphylos morroensis fruit collection took place on July 28, 2023, and August 10, 2023. Fruits were hand-picked from shrubs and from the surface of the ground leaf litter, with the two sources (shrub, litter) stored separately. Fruits, from both shrub and litter, were collected from two different populations of A. morroensis located approximately 4.8 km apart (Figure 1). The first was a small, mature population estimated to be 60 years old, bordering South Bay Boulevard southeast of Turri Road, that developed after construction of the South Bay Boulevard bridge in 1966. Arctostaphylos morroensis in this location were uniform in size and intermixed with Ceanothus cuneatus (buckbrush), Adenostoma fasciculatum (chamise), and other species typical of maritime chaparral habitat, and with Quercus agrifolia (coast live oak) in oak woodland habitat. This population was thought to have germinated in response to the importation of Baywood fine sand as fill for the construction of South Bay Boulevard and the bridge spanning Los Osos Creek, a tributary to the Morro Bay estuary, in 1966. There were no records of fire impacting this location. At this site we collected fruit from at least 25 individuals.
Collection locations, Los Osos, San Luis Obispo County, California.
The second fruit collection site was a much larger, mature population on county and California State Parks land south of Los Osos (Hazard Canyon) (Figure 1). Arctostaphylos morroensis in this population were estimated to be at least 70 years old based on historical aerial photography dating back to 1949 and analysis of cross sections of C. cuneatus that occur within the stand (Tyler and Odion 1996). Collection from Hazard Canyon was outside the area to be impacted by the bridge project, so collections adhered to the USFWS guidance of collecting no more than 10% of available fruits from individual shrubs. Additionally, since fruit collections had occurred in the Hazard Canyon population in 2022, we adhered to the USFWS guidance of collecting from a single location within that region no more than once every two years. At this site we collected fruit from at least 50 individuals.
Collection efforts yielded over 4,000 shrub-sourced and over 1,320 litter-sourced fruits. The discrepancy between the number of shrub- and litter-collected fruits reflects the more limited availability of intact whole fruits in the litter. Shrub-collected fruits varied from green and firm to orange and slightly desiccated, whereas fruits from leaf litter were brown to black and desiccated. Fruits from shrub and leaf litter sources were kept separate for experimental trials. To estimate the number of seeds per fruit we performed seed counts by cutting open 25 randomly selected fruits from each of the shrub and litter collections.
Treatments
Arctostaphylos morroensis fruits from shrub and leaf litter sources were divided into eight treatments each, representing four different scarification treatments (oven heat, fire, acidification, and control) and two stratification treatments (warm and cold). Based on number of fruits collected, each shrub-sourced treatment had 500 fruits, and each litter-sourced treatment had 165 fruits. After dividing fruits among treatments, 385 remaining whole fruits from shrub and litter sources were sown directly in the greenhouse with no scarification or stratification treatment.
Scarification treatments were applied as follows. Firetreated intact fruits were evenly broadcasted atop wooden flats containing Baywood fine sand, covered with a 2-inch even mixture of shredded paper and A. morroensis leaf litter, and burned with a handheld, MAPP gas torch until all paper and leaf litter were charred. This methodology sought to mimic short duration but high intensity heat associated with chaparral fires (Hart 2005). Fruits were then sifted from the Baywood fine sand and charred material, and crushed with a rolling pin. To estimate surface temperatures experienced in this treatment, in three additional flats without fruits, we employed pyrometers that were copper tags with temperature-sensitive paint strips (300°F, 450°F, 650°F, 850°F, 1100°F, 1300°F), placing one in the center of each flat, on top of the soil and below the litter/paper fuel. We burned these flats using the same methods used in the actual treatments. All three tags indicated that the maximum temperature reached in our burning treatment was at least 300oF.
For the oven-heating treatment, intact fruits were placed in a commercial grade oven at 100°C for 5 minutes based on previous literature regarding A. morroensis germination (Tyler and Odion 2020). Fruits were then crushed with a rolling pin to separate the seeds.
Acid-treated fruits were soaked intact in a 93% sulfuric acid solution (H2SO4) for two hours. The soaking time was selected based on the scientific literature for other Arctostaphylos species (Berg 1974, Emery 1988, Hart 2005) adjusted for the relatively small fruits of A. morroensis compared to other manzanita species. After soaking in acid, fruits were rinsed with cold water, air-dried and crushed with a rolling pin.
Following the scarification treatments, crushed fruits were placed in cold or warm storage. Cold stratification entailed storing fruits in the dark at 5°C. Warm stratification entailed storing fruits at room temperature (~20°C) in the dark. Both stratification treatments lasted 2.5 months (November 1 to mid-January). All samples were stored in breathable paper bags to prevent molding.
After the stratification period, crushed fruits were sown 0.6 cm deep into 0.3 m × 0.5 m flats, which were filled with 50% Baywood fine sand and 50% standard nursery potting mix (one flat per treatment). These were maintained in a hoop house and monitored for seedling emergence. Hoop houses, commonly used in restoration, lack the climate controls found in a greenhouse, but do moderate temperature, protect from wind, protect from vertebrate predation, and deliver consistent irrigation. The flats were evenly misted daily for 11 minutes, completely saturating the soil.
All fruits within an individual treatment (e.g., 500 per treatment in the case of shrub-sourced) were placed in one flat rather than dividing them into replicates with smaller numbers of fruits per treatment for several reasons. First, it is known that germination rates of this species are very low (ranging from 1–4%), so if sub-divided, the replicates would likely have wide variation, with exceptionally low germination in most. Second, we did not expect that a fire simulation treatment that uses burning litter and other material would carry fire consistently across a smaller sample unit (e.g., within 10 cm × 10 cm trays). Finally, the aim of this study was to investigate and report on methods that would be useful to practicing restorationists, who regularly use large flats subjected to a recommended treatment to germinate seeds and establish seedlings. Given this design, our statistical analysis assessed differences between treatments without the assumptions of replicate samples.
The number of seedlings per treatment was recorded weekly from January 16 to May 8, 2024. The termination of the experiment was dictated by the threat of seedling stress due to mold and moss. All seedlings were transplanted into pots at the conclusion of the study (four months after completion of stratification) to be used for future plantings in restoration sites.
Statistical Analysis
For all statistical analyses we examined each seed-source separately—shrub- and litter-sourced—to assess differences among numbers of seedlings that emerged in various treatments. To compare the proportions of seedlings emerging (# seedlings per fruit) in different treatments, we used Z-tests, which are appropriate where sample sizes are large, but proportions being tested potentially small (Snedecor and Cochran 1989, Zar 1999). We calculated Z in Microsoft Excel. To examine the effects of stratification, we tested the hypothesis that within a scarification treatment (fire, oven-heated, or acid-treated) the number of seedlings differed between the two stratification treatments (cold vs warm storage). To examine the effects of scarification, we used Z-tests to test the hypothesis that within each stratification/storage treatment, the scarification treatments would result in significantly greater seedling emergence than the control; we also used Z-tests to assess for differences among the scarification treatments. We used two-sided tests for all comparisons.
Results
Although we used fruit counts for our data analysis, seed counts were used to estimate emergence rates for comparison to values reported in the literature. We found an average of 4.6 seeds per fruit (range of 2 to 9, estimated standard deviation 1.8) from shrub-sourced fruits and an average of 4.3 seeds per fruit (range of 2 to 9, estimated standard deviation 1.8) from litter-sourced fruits. We observed that many of the fruits (both shrub- and litter-sourced) contained seeds that appeared to be fused, so some segments counted as single seeds may have instead been multiples. Therefore, we consider our findings of 4.6 and 4.3 to be low estimates of number of seeds per fruit. We did not determine seed viability by examining each seed for an intact embryo, but observations of A. morroensis fruits collected from a nearby site found 60–80% viability for fresh seed (Tyler, unpublished data).
Shrub-sourced Seeds
First seedling emergence occurred February 20, 2024 (35 days after sowing, and 95 days after scarification treatments) from shrub-sourced fruits with acid treatment in cold and warm storage (9 and 5 seedlings, respectively) (Figure 2). Emergence occurred in all treatments through the end of the study (May 8, 2024), with the exception of controls (Figure 2). Counting the initial date of seedling observation (February 20) as day one, the rate of emergence was especially high in the acid treatment in warm storage (Figure 2) over the first 30 days, and for nearly all treatments, emergence plateaued by day 73. Overall, the final emergence rates varied substantially among the treatments (Figure 3), with the acid-soaked, warm-stratified fruits yielding the maximum of 90 seedlings from 500 fruits (proportion = 0.18).
Seedling emergence over time by treatment from shrubsourced fruits: scarification as Fire, Oven-heated, Acid, and stratification as Cold and Warm treatments; n=500; not shown are controls (cold and warm stratified), which had zero emergence.
We found significant differences between warm vs. cold stratification in two of the treatments. Warm stratification resulted in significantly greater numbers of seedlings emerged in two of the treatments: oven-heated and acidtreated (Z-tests p = 0.011, and p < 0.001, respectively) (Figure 3). Although the fire treatment produced more seedlings in warm-stratification, this difference was not statistically significant (p = 0.208).
Seedling emergence from shrub-sourced fruits. Shown is the proportion of fruits (out of 500) that yielded a seedling by the end of the study. Asterisk indicates a significant difference between cold and warm-stratification within a scarification treatment (Z-test, p<0.05).
All scarification treatments produced significantly more seedlings compared to the controls except for the ovenheated (cold stratified) treatment (p = 0.317) (Figure 3, Table 1). In addition, we found that all treatments within a particular stratification regime were significantly different from each other, with fire-treated seeds producing the greatest number of seedlings in the cold-stratification treatment, and acid-soaked seeds producing the highest seedling numbers in the warm-stored treatment (Figure 3, Table 1).
Effects of scarification on shrub-sourced seeds. Results of Z-tests to determine differences in proportions of seedlings (per fruit) between scarification treatments. Shown are p-values from paired comparisons within a storage treatment (e.g., first row provides result for cold stratified fire vs. cold stratified oven heated as p < 0.001) for shrub-sourced fruits. N = 500 fruits per treatment
Propagation rates based on seedlings emerged per fruit ranged from < 1% to 18% (Figure 3). Using our estimate of 4.6 seeds per fruit, the propagation rates for number of seedlings produced per seed was substantially lower, from < 1% to 4%.
Litter-sourced Seeds
Initial seedling emergence from litter-sourced seed occurred February 27, 2024, slightly later than for shrubsourced seed. However, similarly, earliest emergence occurred in the acid treatment in cold storage (Figure 4). With the exception of controls, which produced no seedlings, emergence occurred in all treatments through the end of the study (May 8, 2024) (Figure 4). As was observed for shrub-sourced seed, the rates of emergence were highest in the first 30–40 days. Overall, the final emergence rates varied among the treatments with the fire-treated, coldstratified fruits yielding the high of 16 seedlings from 165 fruits (proportion = 0.09), however the differences were not as striking as for shrub-sourced seeds (Figure 5).
Seedling emergence over time by treatment from litter sourced fruits: scarification as Fire, Oven-heated, Acid, and stratification as Cold and Warm stratification treatments; n=165; not shown are controls (cold and warm stratified), which had zero emergence.
Seedling emergence from litter-sourced fruits. Shown is the proportion of fruits (out of 165) that yielded a seedling by the end of the study. Asterisk indicates a significant difference between cold and warm-stratification within a scarification treatment (Z-test, p<0.05).
The only significant effect of stratification on littersourced seed was for those exposed to the fire scarification treatment (Z-test, p = 0.042) (Figure 5). However, unlike the shrub-sourced seed, the litter-sourced seed had higher emergence in the fire-scarification treatment when subjected to cold stratification/storage. For all other treatments there was no significant difference between warm vs cold stratification methods.
All scarification treatments produced significantly more seedlings compared to the controls (Figure 5, Table 2). The only significant difference between treatments within a particular stratification regime was for cold stratified seeds, with the fire treated seeds producing more seedlings than the oven-heated seeds (Figure 5, Table 2).
Effects of scarification on litter-sourced seeds. Results of Z-tests to determine differences in proportions of seedlings (per fruit) between scarification treatments. Shown are p-values from paired comparisons within a storage/stratification treatment (e.g., first row provides result for cold stratified fire vs. cold stratified oven heated as p = 0.042) for litter-sourced fruits. N = 165 fruits per treatment.
Not including the controls, propagation rates based on seedlings emerged per litter-sourced fruit ranged from 4% to 9% (Fig. 5). Using our estimate of 4.3 seeds per fruit, the propagation rates for # seedlings produced per seed was lower, from < 1% to 2%.
Additional Control Sample
The 385 fruits (combined remaining shrub- and littersourced) that were sown directly without treatment were monitored for six months (from November 1 to May 8). We detected no emergent seedlings in this sample throughout the entire study.
Discussion
In addition to providing new information about the basic ecology of this species, our propagation studies resulted in practical suggestions for restoration in terms of seed source, storage, germination cues, and methods.
Seed Source
Comparing different sources of seeds collected from the field, fresh, shrub-sourced fruits provided the best emergence results compared to litter-sourced fruits and are the preferred method to generate restoration plantings based on the results of this study. Close to 20% of shrub-sourced fruits, when subjected to the optimal treatment (scarification with acid, and warm stratified) produced a seedling. The next most successful treatments (scarified with fire in either warm or cold storage) also used shrub-sourced fruit, and yielded a seedling from 11–13% of the fruits. This is an encouraging finding, since shrub-sourced fruits can be collected in greater quantity and with less effort compared to litter sourced fruits. However, availability of shrubsourced fruits is dependent on the time of year. The fruits mature in spring-summer and drop when fully ripened. Tyler et al. (2023) found that the majority of fruits fell from the plants during June and early July one year, and August to early October in the following year. Thus, from roughly October through June there are no fresh fruits available for harvesting. In addition, the available quantity of fruit varies annually based on the individual plant sampled (Tyler et al. 2023) and some years fruit availability can be unpredictable.
As a practical alternative to fresh, shrub-sourced fruits, litter-sourced fruits proved to be a viable option. Though collection efforts are more labor intensive and fruit quality varies, litter-sourced fruits can be collected at any time of year. Nearly 10% of litter-sourced fruits yielded a seedling if receiving the optimal treatment (fire, cold-stratified). Emergence rates from litter-derived fruit with fire scarification in cold storage were close to the emergence rate for shrub-sourced fruits receiving this same treatment, suggesting that use of litter-sourced fruits can be worthwhile when shrub-sourced fruit is unavailable.
Stratification/Storage
With only one exception—litter-sourced, fire- and acidscarified treatments (Figure 5)—warm storage had higher rates of emergence than cold storage for all other treatments. This was an unexpected result based on the published literature. Some previous studies have found that other Arctostaphylos species benefitted significantly more from cold stratification than warm stratification (Carlson and Sharp 1975, Emery 1988, Jurado et al. 2011). In addition, other germination experiments conducted on manzanitas, including A. morroensis, have had positive emergence results for seeds exposed to cold stratification (Keeley 1987, Tyler and Odion 2020), though these studies did not include a comparison to seeds exposed to warm stratification. From a practical standpoint ours is a welcome result; warm storage is preferred because it is a simpler process. It can be accomplished without a refrigerator and it may be easier to prevent mold from developing during the storage period.
One possible explanation for the results observed in the present study for this species is that the mild maritime climate in which this plant grows and evolved may not have selected for such a cold dormancy period. Another explanation is that the fresh seed may not yet have developed secondary/internal dormancy that cold stratification helps to break. This second explanation is supported by our observation that cold stratification did benefit the litter-sourced seed in one scarification treatment (subjected to fire), which may be a function of seed age. The litter-sourced fruits are older—probably one to a few years old. The majority of dropped fruits of A. morroensis are removed quickly by predators, most likely small mammals and birds (Tyler et al. 2023). The fruits that do remain in the litter are eventually broken apart or buried by animals, so those that are intact and on the surface are likely to have been dropped within a few years from maturing. Compared to fresh, shrub-sourced fruits, the fruits on the top layers of litter, have hardened, dried exocarps on the surface of the fruits, and more importantly, hardened, stony endocarps (outer surface/seedcoat of the individual seeds). The additional stimulation of cold-stratification for breaking dormancy may be particularly useful to employ for this seed source.
Scarification
Fire-related cues such as heat and by-products of combustion have been identified as principal mechanisms that break seed dormancy of fire recruiters that rely on soilstored seed (Keeley 1991). In other manzanita species, germination rates are low, but enhanced with smoke, charred wood, heat shock, or some combination of these treatments (Odion 2000, Keeley et al. 2005, Jurado et al. 2012). In the present study, with only one exception (shrub-sourced, oven-heated, cold stratified), any scarification treatment yielded significantly better results than the controls. The most successful treatments were shrub-sourced fruits scarified with acid or with fire. For litter-sourced fruits, all scarification treatments were better than the controls, with the only difference between scarification treatments being fire vs. oven-heated in the cold stratified fruits. This suggests that for generating A. morroensis for restoration projects, acid scarification would be the preferred choice to maximize seedling emergence. However, fire scarification produced the second highest emergence rate and can be considered a viable alternative to using sulfuric acid.
Both treatments have safety considerations. Acidification requires use, storage, and disposal of sulfuric acid. Personal protective equipment is required during handling, and disposal may require neutralization, dilution, or treatment as a hazardous waste. However, acidification of fruits can be closely controlled through the sulfuric acid dilution and soak time for consistent treatment. Fire treatment involves the use of a gas torch which introduces a potential fire hazard. Additionally, test conditions are more difficult to control regarding consistency of heat applied to every fruit. From that perspective, acid treatment has an advantage over fire treatment. Using oven heating for scarification avoids these safety issues, and would be a good option for litter-sourced fruits. It was surprising that oven heating of fresh shrub-sourced fruits did not produce many seedlings in our study. Since the treatment was applied to whole fruits, it is possible that the mealy mesocarp served to insulate the nutlets/seeds, and prevented adequate heat to break the seed coat. It could be informative to investigate the impacts of oven heating on fresh fruit after first crushing them to expose the seeds.
From a practical perspective, it is useful to note that none of our control treatments (cold or warm stratified for either shrub- or litter-sourced fruit, or our untreated whole fruits) yielded even a single seedling. These findings are consistent with previous studies conducted with Arctostaphylos species using fresh fruit in which un-scarified controls did not germinate: A. pungens (Jurado et al. 2011), A. glandulosa (Keeley 1987), and A. canescens (Parker 1987). The fact that all controls in this study had zero emergence suggests that it would be advantageous to use scarification and stratification for restoration planting efforts.
Use of Whole Fruits
While previous studies of germination in Arctostaphylos have been conducted using extracted individual seeds as the unit of measurement, we recommend focusing on use of fruits, both crushed and treated, for propagation collection, processing, and sowing protocols, and for estimating seedling production. For example, in our study, we found that fresh, shrub-sourced fruits, optimally treated, yielded seedlings at a rate of 13–18%, meaning that if 100 seedlings are required for restoration, between 550–770 fruits are needed. Or if litter-sourced fruits are being used, to yield 100 seedlings, one would need 1000–2000 fruits (since the most successful treatments produced 5–9% seedlings per fruit.) This is a more practical and obtainable field objective, rather than directing a collector to gather 5 to 10 thousand seeds, for example. In addition, extracting individual seeds from each fruit to treat and sow is time-intensive and is unlikely to produce substantially superior results.
The only limitation from our approach is that comparison to prior studies on germination rates in other manzanita species relies on our rough estimates of total seed numbers. However, our results were consistent with germination rates reported in previous studies. Based on our findings of 4.6 seeds per fruit, our treatments resulted in germination-emergence rates of < 1–4% (percent of seed that germinated and produced a seedling) for fresh fruit, and < 1–2% for litter-sourced fruits. Tyler and Odion (2020) did not investigate germination in fresh fruit, but reported rates for litter-stored seed as 1.5–2.3%.
Acknowledgements
Thank you to all those who helped with fruit collection, study design, and in the greenhouse: Sal Zaragoza, Katie Drexhage, California Department of Parks and Recreation interns, John Sayers and Chris Kofron. We received helpful comments and suggestions from two anonymous reviewers and the associate editor, which improved the quality of this manuscript.
