Abstract
Hispaniolan pine (Pinus occidentalis Swartz [Pinaceae]) is an endangered tree species endemic to Haiti and the Dominican Republic. Species recovery is hindered by lack of reputable propagation knowledge, challenges associated with seed procurement, and limited resources available locally for propagation. A suite of germination and sanitation studies were conducted to examine the effects of seed moisture content, cold stratification length, and chemical treatment on germination. No evidence supports that the odds of a seed germinating are statistically different among seed moisture content levels, cold stratification lengths, or sanitation treatments. Mean germination time (MGT) was unaffected by seed moisture content but was greater when seeds were subjected to ≥ 14-d cold stratification compared to unstratified seeds. MGT was also greater when seeds treated with a 10-min soak in 2:3, 5.25% sodium hypochlorite:water solution followed by a 30-min running water rinse were compared to untreated seeds. Treatment with a 1-h 3% hydrogen peroxide soak followed by a 1-h running water rinse significantly reduced seedborne fungi in comparison to untreated seed. Results suggest that P. occidentalis has non-dormant seed and does not benefit from the current practice of a pre-germination soak or from cold stratification, but a 1-h 3% hydrogen peroxide soak followed by a 1-h running water rinse can control seedborne fungi without compromising germination.
Hispaniolan pine (Pinus occidentalis Swartz [Pinaceae]) seedlings, 2 wk old.
Hispaniolan pine (Pinus occidentalis Swartz [Pinaceae]) is a critical species for restoration and afforestation in Haiti and the Dominican Republic, but the isolated location of remaining tree stands makes seed procurement difficult (Posner and others 2010; Hubbel and others 2018). Once seeds are obtained, nurseries must use the most effective methods for propagation to grow high-quality seedlings. Protocols must be science-based and use materials that are available and affordable to local nurseries (Haase and Davis 2017). To our knowledge, no study has been conducted to determine how treating seeds prior to sowing influences P. occidentalis germination, specifically with consideration of pre-germination treatments such as reducing seedborne fungi and evaluating seed dormancy.
Pines (Pinus spp.) have a diverse array of dormancy types, but most temperate and tropical American pines are either non-dormant or physiologically dormant (Baskin and Baskin 2014). Many conifer seeds are considered “shallowly dormant,” requiring stratification under cold, moist conditions to germinate (Jones and Gosling 1994). The 2 most common seed pretreatments used to break dormancy and allow germination are soaking seeds in water to reach a specific moisture content and placing seeds in cold stratification (Kolotelo and others 2001; Krugman and Jenkinson 2008).
Water performs critical roles within seeds: It is required for proper functioning of enzymes, it acts as a solvent within which chemical reactions can occur and molecules can be transported, and it is a reactant in the breakdown of storage compounds (Woodstock 1988). Evidence also supports that some pine species contain germination inhibitors in the seedcoat that are removed during soaking (Martinez-Honduvilla and Santos-Ruiz 1978). A seed’s ability to break dormancy and its response to changes in temperature are dependent on moisture level (Vertucci and Leopold 1986). Imbibition rate varies across conifer species (Dumroese 2000; Kolotelo and others 2001), and seeds that are sown dry can have delayed germination because of the time it takes to absorb enough water from the soil or potting media to initiate germination (Kolotelo and others 2001).
Germinating Pinus occidentalis seeds.
Many conifer seeds require stratification at 1 to 5 °C (33.8–41.0 °F) (Gosling 1998), but requirements vary among species and even among seedlots of the same species (Carpita and others 1983; Landis and others 1998). Cold stratification increases the growth potential of the embryo and allows it to overcome the mechanical restraint of the seedcoat (Carpita and others 1983). In loblolly pine (P. taeda L.), the rate of radicle expansion was greater for seeds that had undergone cold stratification (Carpita and others 1983). Conversely, stratification was detrimental to the germination of longleaf pine (P. palustris Mill.), Caribbean pine (P. caribaea), and Mexican yellow pine (P. oocarpa Schiede ex Schltd.) (Moreno 1985; Barnett and Jones 1993).
Optimal seed moisture levels for effective cold stratification vary by species (Kolotelo and others 2001). For example, in the Pinaceae family, Sitka spruce (Picea sitchensis (Bong.) Carriére), western hemlock (Tsuga heterophylla (Raf.) Sarg.), and ponderosa pine (P. ponderosa Kawsib & C. Lawson) seeds require low moisture contents (<30%) during stratification. Lodgepole pine (P. contorta Douglas ex Loudon) seeds require a moisture content between 30 and 32% for cold stratification while Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco), western white pine (P. monticola Douglas ex D. Don), and grand fir (Abies grandis (Douglas ex D. Don) Lindl.) seeds require moisture contents >32% (Kolotelo and others 2001). Seeds of some tropical pines, such as P. maximinoi H.E. Moore, P. montezumae Lamb., and P. leiophylla Schiede & Deppe, have increased germination rates if soaked before sowing, whereas germination rates for seeds of other tropical pine species, such P. caribaea Morelet and P. patula Schiede ex Schltd. & Cham., are unaffected by soaking (Ghosh and others 1974; Donald 1981; Vozzo 2002).
Seeds carry a microbiome that contains both beneficial and harmful microorganisms (Agarwal and Sinclair 1997). Seedborne pathogens are defined as organisms carried on or within seeds that have the potential to cause disease in the seed or emerging seedling (Maude 1996; Cram and Fraedrich 2009). Some seedborne pathogens will infect seeds during storage, killing damaged and low-vigor seeds and reducing overall germination (Cram and Fraedrich 2009). Pathogens can kill seedlings before the seedling emerges aboveground (pre-emergence damping-off) or after it emerges (post-emergence damping-off, shoot blight, or root rot) (Cram and Fraedrich 2009; Cram 2017). Other pathogens will cause only temporary growth reduction and shoot deformation, but healthy seedlings are able to outgrow the pathogens (Rees and Phillips 1994). Some common conifer seedborne fungal pathogens that cause seed or seedling death are in the genera Fusarium, Diplodia, Alternaria, and Rhizoctonia (Rees and Philips 1994; Lilja and others 1995; Talgo and others 2010).
Preventing seeds from becoming contaminated and controlling pathogen levels on seeds once they become infected require an integrated management approach using a combination of cultural sanitation practices and chemical treatments during seed collection, storage, and sowing. Seed handling during cleaning, storage, and seed pretreatments is often when most pathogens increase on seeds and spread among seedlots (Littke 1996). One of the most efficient and effective ways to control seedborne pathogen levels is by disinfecting seeds (Kolotelo and others 2001); although such treatments are effective at removing contaminating pathogens on the outside of seedcoats, they have no effect on those that have penetrated the internal tissues of the seed (Wenny and Dumroese 1987).
Two chemical treatments commonly used to disinfect conifer seeds are soaking in hydrogen peroxide or in sodium hypochlorite (bleach). Hydrogen peroxide is less phytotoxic (James and Genz 1981) and is the most common chemical used to disinfect conifer seeds in North America (Kolotelo and others 2001). It also stimulates germination in some conifer species, such as P. ponderosa and Pseudotsuga menziesii (James and Genz 1981; Dumroese and others 1988). For thin-coated or porous seeds, performing a hydrogen peroxide soak after imbibition and stratification, when seeds are full of water, prevents hydrogen peroxide from entering the seeds and damaging internal tissues (Dumroese and others 1988). Depending on the length of the soak, bleach can cause seed etching and can decrease the viability of thin-coated conifer seeds such as true firs, larches, and spruce (Wenny and Dumroese 1987). Unlike hydrogen peroxide, bleach does not penetrate undamaged seedcoats, and the treatment can be applied either before or after imbibition and stratification (Dumroese and others 1988). Recommendations for the composition of bleach solutions range from 2 parts bleach to 3 parts water to 1 part bleach to 5 parts water, depending on the species (Dumroese and others 1988; Kolotelo and others 2001).
The easiest seed sanitation treatment is a running-water rinse, which can be used while seeds are being imbibed before being placed in stratification. A running-water rinse is more effective than a still water soak at removing spores while allowing seeds to fully imbibe (James and Genz 1981), but it is not effective at removing all spores and can still lead to disease in treated seeds (James 1987) and can even spread spores if uninfected seeds are soaked with infected seeds (Kolotelo and others 2001). A running-water rinse does not decrease seed viability and is effective for species that have thin and permeable seedcoats that are sensitive to chemical treatments (Dumroese and others 1988).
Objectives of this research were to test the effects of seed moisture content and cold stratification length on germination of P. occidentalis seeds and to investigate whether different sanitation treatments affect germination rate and the prevalence of seedborne fungi.
MATERIALS AND METHODS
Pinus occidentalis seeds, obtained from the Boca de Nigua National Seed Bank, Dominican Republic, were sourced from a single population near Jarabacoa in La Vega Province, Dominican Republic, in September 2013. The seedlot was stored at 5 °C (41 °F) with desiccant until use. We randomly selected seeds from the seedlot using industry protocol for hand sampling (AASCO 2006). In July 2017, a tetrazolium test performed at the Oregon State University Seed Laboratory (Corvallis, Oregon) found the seedlot to have 84% viability.
Seed Imbibition
We estimated the rate of water uptake for P. occidentalis seeds to determine soaking lengths needed to attain the desired seed moisture contents. Water content for seeds in dry storage was determined by drying 3 samples of 25 seeds each in an oven at 103 °C (217.4 °F) for 26 h (Elias and others 2012). To create the water uptake curve, we randomly selected 7 samples of 25 seeds each and recorded the starting weight of each sample. Each sample was placed in a mesh bag and submerged in aerated distilled water. Each sample was blotted dry and weighed every 3 h for 12 h, then every 12 h until hour 84, then again at hours 108 and 130.
Seed Moisture Content and Cold Stratification
Seed moisture contents and lengths of cold stratification tested were selected based on previous research that determined the optimum moisture content for the germination of temperate conifer seeds and the optimum length of cold stratification of some tropical pines (Kolotelo and others 2001; Vozzo 2002). We tested 5 levels of seed moisture: 5% (dry seed), 27%, 31%, 35% (control), and 39%. We tested 5 levels of cold stratification length: 0 d (control), 7 d, 14 d, 21 d, and 28 d. The 35% moisture content, 0-d cold stratification is the current seed pretreatment protocol used in Dominican Republic nurseries, and we considered this to be the control treatment (personal communication, Wilman Placido-Made). We compared all combinations of 5 seed moisture levels and 5 lengths of cold stratification, resulting in a total of 25 treatment combinations.
We randomly selected seeds and divided them into 75 replicates containing 25 seeds each. Every replicate was randomly assigned to a seed moisture level and cold stratification length combination, with 3 replicates assigned to each treatment combination. Each replicate was considered an experimental unit and placed into a mesh bag. Seeds were cleaned by soaking in a solution of 3 parts water to 1 part 3% hydrogen peroxide for 5 min then rinsed in running water for 10 min. Bags were placed in aerated distilled water for differing amounts of time to achieve the desired moisture content. We determined soaking times using the water uptake curve best fitted line equation as described above. Once removed from soaking, each bag was blotted dry and underwent cold stratification using the standard naked stratification protocol for temperate conifers, in which bags were hung in a single refrigerator (average temperature of 6 °C [42.8 °F]) with a dish of water below the seeds (Kolotelo and others 2001).
Once seeds had completed their assigned stratification period, we removed seeds from a single bag and placed them into a 9-cm (3.5 in) plastic Petri dish on top of 2 layers of Whatman No. 1 filter paper moistened with 2.5 ml of distilled water. We repeated this process for each bag of seeds. Dishes were placed into a single germination chamber (Hoffman Manufacturing, Corvallis, Oregon) set at 25 °C/15 °C (77 °F/59 °F) at 12-h intervals with 12 h of light. Germination and dish moisture levels were monitored daily for 30 d. To maintain uniform moisture status in every dish, distilled water was added to a Petri dish if that dish fell below 75% moisture content, determined by weight. We considered seeds to have germinated when the radicle exceeded 1 mm in length and pointed downward. Seeds were removed from the dish once germinated, and the number of germinated seeds was recorded daily. Seeds with fungi were wiped clean using a paper towel moistened with distilled water and placed back into the dish. Location of the Petri dishes within the germination chamber was randomized daily.
Seed Sanitation
Seed sanitation treatments were based on a literature search of the most effective methods to control pathogens on temperate conifer seeds while taking into consideration materials that could be obtained on Hispaniola (Table 1). Treatments used varying dilutions of bleach (5.25% sodium hypochlorite) and 3% hydrogen peroxide and varying rinse times in running water, as well as a running-water-only rinse and a control (no sanitation treatment). We randomly selected seeds and divided them into 44 replicates of 25 seeds each. Each replicate was considered an experimental unit and each sanitation treatment contained 4 replicates. Once replicates of seeds had undergone their assigned sanitation treatment, the same germination protocols described above were used. In addition to recording germination daily, we recorded the number of germinated seeds with evident fungal growth. After the end of the germination experiment, all ungerminated seeds were sent to the Oregon State University Plant Clinic (Corvallis, Oregon) where all fungi with reproductive structures were identified to genus.
Pinus occidentalis seeds on moist blotter paper in the germination chamber.
Statistical Analysis
All analyses were done with R (Version 3.4.3, The R Foundation for Statistical Computing, 2017). All dishes were assumed to be independent, and all effects of interest were estimated from the corresponding models. Differences among treatments for each parameter were analyzed using either one-way or two-way ANOVAs, and comparisons were made between the control and treatments using Dunnett’s test (α = 0.05).
The moisture levels needed for the moisture content and cold stratification length study occurred between 3 and 24 h of imbibition, so we used only data between those times to obtain the best fitted line for the rate of water uptake. A time series linear mixed model was used to determine the seed moisture levels during imbibition, and a best fitted line equation was created using this model.
Sanitation treatments used in this study consisted of various combinations of soaks in diluted 5.25% sodium hypochlorite or 3% hydrogen peroxide and running-water rinses.
A generalized linear model using a quasi-binomial distribution with the logit link was used to test the effect of seed moisture content and cold stratification length on the odds of germination. The mean germination time (MGT) was determined for each moisture level and stratification length combination using the formula from Bewley and Black (1985):
MGT = Σ(T*N)/Σ(N)
T = Time (d)
N = Number of seeds germinating on d T
A linear model was used to test for differences in the mean MGT between moisture content and stratification length treatments.
A generalized linear model using a binomial distribution with the logit link was used to test the effect of the seed sanitation treatment on the odds of germination. A generalized linear model using a quasi-binomial distribution with the logit link was used to test the effect of the seed sanitation treatment on the odds of a germinated seed having fungi. A linear model was used to test for differences in MGT among sanitation treatments and the control.
RESULTS
Seed Imbibition
Water uptake took 5 h and 49 min to achieve 27% seed moisture content, 8 h and 34 min to achieve 31% seed moisture content, 12 h and 2 min to achieve 35% seed moisture content, and 17 h and 25 min to achieve 39% seed moisture content (Figure 1). The equation to determine seed moisture content at a given time between 3 h and 24 h is:
Moisture content (% fresh weight) = 16.302436 + 2.113130*time - 0.046526*time2
Water uptake curve (fresh weight basis) for Pinus occidentalis. Each point represents the mean (± s.e.) of 7 replicates.
Seed Moisture Content and Cold Stratification
No evidence supports that the odds of a seed germinating are statistically different among seed moisture content levels (P = 0.9596) or cold stratification lengths (P = 0.2683), nor was there evidence of an interaction between moisture content and cold stratification length (P = 0.9417) (Table 2). The ratios of estimated odds of germination, as well as 95% confidence intervals, for specific seed moisture contents (Figure 2A) and lengths of cold stratification (Figure 2B) both showed little treatment effect. MGT did not differ among seed moisture content levels (P = 0.2319) but did differ statistically among stratification lengths (P < 0.0001) (Table 2). No interaction occurred between seed moisture content and cold stratification length (P = 0.6155). MGT for seeds subjected to 14, 21, and 28 d of cold stratification was significantly longer than for unstratified (0 d) seed (P = 0.039, P < 0.0001, and P = 0.047, respectively).
Seed Sanitation
No evidence supports a statistically significant difference in the odds of a seed germinating among any sanitation treatment compared to the control (Table 3; Figure 3A). MGT for seeds treated with a 10-min soak in a solution of 2 parts bleach to 3 parts water followed by a 30-min running-water rinse (Treatment 5) was significantly greater than that for seeds in the control treatment (P = 0.0013) (Table 3). There was also evidence of a difference in the odds of a germinated seed having fungi when subjected to different sanitation treatments (Figure 3B). Treatment 7 (1-h hydrogen peroxide soak followed by a 1-h running-water rinse) resulted in significantly fewer seeds with fungi compared to the control treatment (P = 0.044) (Table 3). Among seeds that did not germinate, 15 genera of fungi were identified (Table 4).
Germination odds (with 95% confidence intervals) and mean germination time (MGT) (with 95% confidence intervals) for each level of seed moisture content and length of cold stratification.
Ratios of estimated odds of germination for each seed moisture content when compared to the control (35%), with 95% confidence intervals. The dashed line at 1.0 on the y axis indicates no statistical difference, and the gray background fill represents the area of no biological significance between 0.53 and 2.58.
Ratios of estimated odds of germination for each cold stratification length when compared to the control (0 d), with 95% confidence intervals. The dashed line at 1.0 on the y axis indicates no statistical difference, and the gray fill represents the area of no biological significance between 0.53 and 2.58.
DISCUSSION
Germination tests suggest that P. occidentalis seeds are non-dormant. Imbibing seeds prior to sowing (as is currently practiced) did not affect germination, with rates consistent with the 84% viability determined just prior to the study. Similarly, different moisture content levels in sown P. occidentalis seeds did not lead to a statistically significant difference in MGT. In addition to testing for a statistically significant difference, this study sought to determine if a biologically significant difference in germination rate or MGT could affect nursery production. For this study, a biologically significant response would be a change in mean germination rate of 14 percentage points (ISTA 2018), indicated by an odds ratio below 0.53 or above 2.58. The confidence intervals for all odds ratios comparing each moisture content level to the control indicate that no biologically significant difference occurs in germination rate among seed moisture levels.
Germination odds (with 95% confidence intervals), mean germination time (MGT) (with 95% confidence intervals), and percentage of germinated seeds with mold (with 95% confidence intervals) for each sanitation treatment and the control.
Ratios of estimated odds of germination for each sanitation treatment when compared to the control, with 95% confidence intervals. The dashed line at 1 on the y axis indicates no statistical difference, and the gray fill represents the area of no biological significance between 0.55 and 2.19. Treatments 2, 3, 5, 9, and 10 have the potential of decreasing the odds of a seed germinating when compared to the control, and treatments 1, 4, 6, 7, and 8 have increased odds of a seed germinating when compared to the control.
Ratio of the estimated odds of a germinated seed having mold when compared to the control, with 95% confidence intervals. The dashed line at 1 on the y axis indicates no statistical difference. An asterisk (*) represents a significant difference from the control.
Genera of fungi identified on ungerminated seeds from each sanitation treatment and the control.
Germinating Pinus occidentalis seeds 15 d after a 1-h hydrogen peroxide soak followed by a 1-h running-water rinse. This treatment was the only one to statistically reduce fungi on the germinating seeds.
Unlike this study, most research on conifer seed treatments have found an increase in germination rate when seeds are imbibed prior to sowing (Barnett 1976; Kolotelo and others 2001; Himanen and others 2013), though P. palustris showed decreased germination rates when soaked for progressively longer periods of time (Barnett and Jones 1993). Sowing dry seeds could have a positive production benefit for nurseries located on Hispaniola, many of which do not have reliable access to clean running water or a dedicated workforce with advanced training in seed handling (Haase and Davis 2017). Eliminating the time seeds spend soaking in contaminated still water could reduce the level of pathogens found on the seeds and reduce resource use.
Although we measured no statistical difference in germination rates among cold stratification lengths tested, the 7-d and 14-d cold stratification lengths have the potential to decrease the odds of a seed germinating in P. occidentalis. The lower confidence intervals for the comparison between 7-d and no stratification and between 14-d and no stratification is just below what would be considered a biologically significant result (that is, a change in the mean germination rate of 14 percentage points). Based on the confidence intervals, a change in germination rate of more than 14 percentage points is possible. Further, the biologically significant response in MGT would be a change of 5 d from the control (Mexal and Fisher 1987). Thus, the increase in MGT among seeds subjected to 14-d, 21-d, and 28-d stratification compared to unstratified seed, although statistically significant, may not be biologically meaningful.
Common nursery protocol is to perform the longest stratification period that does not decrease germination rate, since most species are able to germinate more uniformly and over a larger range of temperatures after stratification (Moreno 1985; Downie and others 1998; Barnett 2008). Although this may be practical for nurseries with access to reliable equipment and electricity, this would be difficult to practice at some nurseries on Hispaniola, where electricity may be intermittent or unavailable and refrigeration equipment prohibitively expensive (Perry 2020). In addition, the benefits of being able to germinate over a wide range of temperatures may not be relevant on Hispaniola, where seasonal temperatures do not fluctuate as widely as in temperate North America. Without a clear benefit to germination rate or MGT, the process of stratifying seeds may be impractical for P. occidentalis.
Among sanitation treatments, although we measured no statistical difference in germination rates, there was a potential biological difference. We did not record a pattern to which treatment could potentially increase or decrease germination, and the potential biological difference could be attributable to large germination rate confidence intervals because of the small sample sizes used (4 replicates per treatment). MGT increased only for treatment 5 (a 10-min soak in a solution of 2 parts bleach to 3 parts water followed by a 30-min running-water rinse). The confidence interval for MGT for this treatment contained values >5 d when compared to the control, which is the biologically significant value that could lead to seedlings being smaller than seedlings whose seeds had received other treatments (Mexal and Fisher 1987).
The only treatment to have a statistically significant effect on the number of germinated seeds with evident fungal growth was treatment 7 (a 1-h 3% hydrogen peroxide soak followed by a 1-h running-water rinse). Past studies have found bleach and hydrogen peroxide effective at decreasing the level of fungi on seeds (James and Genz 1981; Wenny and Dumroese 1987; Dumroese and others 1988), but only 1 of the 9 treatments in this study found those chemicals to have a significant effect. Notably, although we saw no difference in the number of germinated seeds with evident fungal growth, based on the fungi identified on ungerminated seeds, all bleach and hydrogen peroxide solutions eliminated genera known to cause seed or seedling death in otherwise healthy seedlings (for example, Alternaria, Fusarium, and Rhizoctonia) (Mittal and Wang 1993; Lilja and others 1995; Talgo and others 2010). Fungal genera that are weakly pathogenic and can cause damping-off in seeds or seedlings (Botryosphaeria, Chaetomium, Epicoccum, and Ulocladium) were identified in most treatments, regardless of the active chemical or dilution factor (Rees and Phillips 1994; Lilja and others 1995; Talgo and others 2010). These genera can cause disease when seeds are damaged during cleaning or storage, or may reduce seedling vigor during early growth, but they do not affect healthy seeds or seedlings (Kolotelo and others 2001; Cram and Fraedrich 2009). Species of Aspergillus, Cladosporium, and Penicillium are considered storage fungi. They are saprophytic, feeding on decaying or dead seed tissue, and are associated with poor quality seeds or seeds that have been stored incorrectly (Rees 1983; Agarwal and Sinclair 1997). It is not surprising that this seedlot would contain storage fungi as it had been transferred between locations multiple times and experienced changes in storage temperature and humidity. Other genera detected on ungerminated seeds are not known causal agents of pine diseases.
Germinating Pinus occidentalis seeds infected with fungi 15 d after being placed in a germination chamber with no sanitation treatment.
Although most sanitation treatments did not reduce the number of germinated seeds with evident fungal growth, note that the mere presence of fungi on a seed is not necessarily indicative of whether the seed will germinate and produce a healthy seeding (Mittal and Wang 1993; Cram and Fraedrich 2009; Himanen and others 2013). For example, seedlings that germinate from high-vigor seeds can outgrow some fungal contaminants (Kolotelo and others 2001). Further, without the use of more robust diagnostic tools, we cannot know whether fungal taxa identified on ungerminated seeds were indeed pathogenic and whether they were representative of fungi detected on germinated seeds. Future studies should employ more sensitive and specific diagnostic tools for pathogen detection (for example, molecular tools as described in Mancini and others 2016) and should grow seedlings in conventional potting media to determine which sanitation treatments are most effective at reducing seedborne pathogens that cause seedling mortality.
CONCLUSIONS
Propagation protocols are lacking for many native species, preventing production for use in restoration and afforestation projects. Providing nurseries with science-based protocols that use materials easily procured on-site and that do not require specialized training is critical for success, particularly in developing nations where restoration is needed (Haase and Davis 2017). Based on the results of this study, P. occidentalis does not have dormant seed and does not require a dormancy-breaking pretreatment. To reduce fungal contamination, P. occidentalis seeds should be soaked in a 3% hydrogen peroxide solution for 1 h followed by a 1-h running-water rinse. These propagation protocols should reduce seed and seedling loss to fungi, supporting the affordable production of high-quality seedlings in nurseries in Haiti and the Dominican Republic.
ACKNOWLEDGMENTS
Financial support for this research was provided by Oregon State University (Corvallis, Oregon) and Oxbow Farm and Conservation Center (Carnation, Washington). Oxbow and its staff also provided much appreciated space and assistance throughout the project. We are grateful to Wilman Placido-Made for providing seed for the experiment; John Mikkelson and Jim Kiser for assistance and use of research equipment; Ariel Muldoon and Lisa Ganio for suggestions with the statistical analysis; and Amy Ross-Davis for providing editorial comments on an earlier version of this manuscript.
Footnotes
NOMENCLATURE
Photos by Christina St John
This open access article is distributed under the terms of the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0) and is freely available online at: http://npj.uwpress.org
Jump to section
Related Articles
Cited By...
- No citing articles found.