Research ArticleRefereed Research
Open Access

Optimizing regeneration protocols for native Seeds of Success–collected milkvetch (Astragalus spp.) genetic resources

Bailey Hallwachs, Elizabeth Martin, Barbara Hellier and Brian M Irish
Native Plants Journal, May 2025, 25 (3) 179-191; DOI: https://doi.org/10.3368/npj.25.3.179

Abstract

The USDA ARS National Plant Germplasm System conserves and promotes the use of important agricultural and ecological plant germplasm. In these collections, ~23,000 accessions representing 147 plant families have been acquired through the Bureau of Land Management Seeds of Success program. Unlike many of the cultivated crops in the collections, a significant proportion of these species lack basic biological information necessary for seed regeneration and long-term conservation. To fill knowledge gaps, we studied 5 native milkvetch (Astragalus L. [Fabaceae]) species to optimize germination protocols, characterize phenological and phenotypic traits, and test pollinator efficiency using various pollinator treatments. Germination trials tested 3 cold stratification (0, 2, or 6 wk in 4 °C [40 °F]) and 2 germination temperature (low: 15/25 °C [59/77 °F] and high: 20/30 °C [68/86 °F]) treatments in a full factorial design. Germination trial seedlings were outplanted at 2 sites (Prosser and Pullman, Washington) in fall 2020. We evaluated 3 pollination methods using honeybee (Apis mellifera Linnaeus [Apidae]), alfalfa leafcutting bee (Megachile rotunda Fabricius [Megachilidae]), and open pollination. Cold stratification increased germination in A. drummondii but had little effect on the other 4 species. Germination at the lower incubation temperatures was as effective, if not better, than germination at the higher temperatures. Only Canadian milkvetch (Astragalus canadensis L.) and two-grooved milkvetch (A. bisulcatus (Hook.) A. Gray) had sufficient survival in the Pullman location to move forward with pollinator treatments. Differences were not found in pollinator treatments for seed set or yield (P > 0.05). Astragalus canadensis and A. bisulcatus performed generally better than other species evaluated with the highest number of established and surviving plants, good seed set, and highest seed yields. This research provides critical information on germplasm management and suggests the need for further research on protocol development.

Key Words

The Seeds of Success (SOS) program, led by the Department of Interior, Bureau of Land Management (BLM), is a collaborative program whose mission is to collect wildland native seed for research, development, germplasm conservation, and ecosystem restoration (Haidet and Olwell 2015; Greene and others 2019; Volk and others 2023). SOS seed collections are managed by the United States Department of Agriculture (USDA), Agriculture Research Service (ARS), National Plant Germplasm System (NPGS) curatorial programs through seed recollection and regeneration as quantities (due to distributions) or quality (that is, viability) decline. In contrast to many agricultural crops, many species within these collections lack basic biological information regarding germination, growth, pollination, and reproductive biology, which is critical to long-term conservation efforts. With an increased rate of native habitat loss, and subsequent decline in biodiversity, practitioners have a pressing need to understand basic information for these plant genetic resources (PGR) in order to grow, regenerate, and deploy native forb and grass species in restoration and revegetation capacities (National Academies of Sciences 2023).

Milkvetch or locoweed (Astragalus L. [Fabaceae]) species are annual and perennial legumes, and they constitute one of the largest genera of plants in the world with approximately 3000 identified species (Hendron 2020). More than 350 species of Astragalus are native to the US (McCue and others 2001). The NPGS safeguards the genetic diversity of 2478 Astragalus accessions, representing 430 species from 65 countries. Of the species in the Astragalus collection, 46% are native to the US (USDA ARS GRIN-Global 2024). Although germination, growth, and pollination requirements have been documented for some species in the genus, these details are lacking for most. This information is especially important given the range of taxonomic diversity and geographic origins of the accessions in the NPGS collections in the genus (Seglias and others 2018).

A thorough knowledge of germination requirements and reproductive biology is critical for consistent and reliable seed production for genebank germplasm maintenance, for commercial seed production and, of most importance, for any large- scale restoration projects (Haidet and Olwell 2015; Greene and others 2019; National Academies of Sciences 2023; Volk and others 2023). Physical seed dormancy is well-documented and a common trait in Astragalus, indicating the need for scarification of seeds, but other germination requirements are less widespread (Soltani and others 2020 and references therein). Physiological or combinational (physical plus physiological) dormancy has been reported in select Astragalus species (Kaye 1997; Soltani and others 2020; Jones and others 2022). For these species, a period of cold stratification improved germination success, but depending on the study, the length of cold stratification varied from a few weeks to many months (Kaye 1997; Jones and others 2022). Similarly, the range of germination temperatures used in studies with Astragalus seeds varies. Propagation protocol database information from the online Native Plant Network reported germination temperatures for various Astragalus species range from 13/18 °C (55/64 °F) night/day to 12/30 °C (54/86 °F) night/ day (https://npn.rngr.net/propagation).

Regeneration of sufficient seed quantities with acceptable seed quality is dependent on breeding systems. Astragalus species are obligate or facultative outcrossers with the breeding biology of less than 1% of Astragalus species known, and manual (or insect) selfing (tripping) has been found to improve or is a requirement for fruit set (Platt and others 1974; Green and Bohart 1975; Richards and Myers 1997; Tanner and others 2013; Richards 2020; Soltani and others 2021; Fulkerson and Kinter 2023). For most species and populations, however, the optimal pollinator (or need for one at all) is unknown.

To address the gap in knowledge, we used 5 native milkvetch species acquired by the NPGS through the SOS program to optimize germination protocols, characterize phenological and phenotypic traits in the field, and test pollinator efficiency on seed set quantities and qualities. Specific study objectives were to 1) determine optimal cold stratification length and temperature to maximize germination success; 2) assess phenological and morphological traits of accessions when grown in an agricultural setting; and 3) assess the efficacy of commercially available managed insect pollinators compared to open-pollinated (naturally occurring) treatments on reproductive output during seed increases.

METHODS

Plant Species

Five species of Astragalus from the NPGS acquired though the SOS collection program were used in this study (Table 1). The 5 species were two-grooved milkvetch (A. bisulcatus (Hook.) A. Gray [Fabaceae]), Canadian milkvetch (A. canadensis L.), Drummond’s milkvetch (A. drummondii Douglas ex Hook), freckled milkvetch (A. lentiginosus Douglas ex Hook), and rushy milkvetch (A. lonchocarpus Torr.). Species were intentionally selected that had multiple accessions from at least 2 different Omernik Level III ecoregions and had been collected at approximately the same time (Omernik and Griffith 2014). Two accessions per species were included in the germination study, and one accession per species was subsequently used for field characterization and a pollinator study (Table 1). We used only one accession per species in the field to prevent cross-pollination in the open-pollinated plots attempting to keep original population genetic integrity for regenerated seed. Only original seed was used in this study, and it was wild-collected by BLM SOS teams between 2013 and 2016. Seed was held in cold (4 °C [40 °F]), dry (18–20% relative humidity) storage at the NPGS Plant Germplasm Introduction and Testing Research Unit (PGITRU) genebank in Pullman, Washington.

View this table:
Table 1

Geographical and other identifier information for the 5 milkvetch (Astragalus L. [Fabaceae]) species and accessions used in the study.

Germination

We tested 3 time periods of cold, moist stratification and 2 germination temperatures in a complete factorial design, for a total of 6 treatment combinations. Cold stratification was either 0, 2, or 6 wk in 4 °C (40 °F) cold storage. Germination temperature was either 15/25 °C (59/77 °F) night/day or 20/30 °C (68/86 °F) night/day. Germinators were set on a 16-h night/8-h day cycle. Seeds were wrapped in germination towels treated with Captan fungicide (50% WP) (Corteva Agriscience, Indianapolis, Indiana) for stratification and germination. Four replications of 25 seeds each were used for each treatment combination (stratification × temperature), for a total of 600 seeds per accession. Prior to use, seed was scarified by using a scalpel to individually nick each seed and was then soaked overnight in deionized/distilled water.

Germination, measured as radicle elongation of 1 mm, was recorded weekly, for a total of 2 wk, after seeds were transferred to the germinators. For each species, we analyzed the effect of stratification, germination temperature, and the interaction between stratification and temperature on the percent germinated seed with a generalized linear model (R Core Team 2023). Percent germination was arcsine-square root transformed to better meet model assumptions. Length of cold stratification and germination temperature were included as fixed effects in the model and accession was included as a random effect. Because little difference occurred in germination from week 1 to week 2, all results presented use percent germination at 1 week.

Field Characterization

In September 2020, greenhouse-grown seedlings for 1 accession of each Astragalus species were outplanted to 2 field sites in Washington State, one in Pullman (46.723790 N, 117.134972 W, Figure 1) and the other in Prosser (46.25235 N, 119.73799 W). These 2 Washington State University (WSU) research farm field sites differ in elevation and climate. The WSU PGITRU farm in Pullman, located at 716 m (2352 ft) elevation in eastern Washington, has 445.3 mm (17.53 in) average annual precipitation. The WSU Irrigated Agriculture Research and Extension Center (IAREC) Prosser field site, at 202 m (666 ft) of elevation in the Columbia River Basin of Washington, has 221.7 mm (8.73 in) average annual precipitation (NOAA 2023).

Each field site consisted of 45 plots, 2.7 × 3.4 m (9 × 11 ft) in size, which were used for both field characterization of plants and the pollinator study. The layout was a randomized complete block design (5 Astragalus species × 3 pollinator treatments × 3 replicates). Each plot was planted with 24 to 32 transplants of a single species, with the aim of having in the range of 20 surviving plants per plot in spring 2021. Plots were drip irrigated for initial establishment but not irrigated for the remainder of the study.

We measured the initial establishment of transplants by counting surviving individuals 1 mo after outplanting. Survival over the first winter was recorded in 2021, and survival over the duration of the study was recorded in 2022 for all plots. No plants survived the first winter at the Prosser site, limiting the remainder of the study to Pullman, Washington. Demographic, phenological, and morphological measurements were made during the 2021 and 2022 field seasons in Pullman. Data included percent reproductive plants/plot, date of first open flower, plant size/ground coverage mid-season (height and 2 perpendicular diameter measurements), and plant growth habit (measured as angle of main stem to the ground and classified from 1 to 10, erect to completely prostrate, respectively). These data were collected from the uncaged, open-pollinated plots for A. bisulcatus and A. canadensis and from all plants in all plots for A. drummondii, A. lentiginosus, and A. lonchocarpus because no cages were used to contain pollinator treatments for these species (see below).

Figure 1.
Figure 1.

Field site used in milkvetch (Astragalus L. [Fabaceae]) germplasm research in Pullman, Washington, over 2 growing seasons (2021 and 2022). Newly established plots: field site 1 mo after outplanting (A); established field site with experimental design and erected pollinator field cages (B).

Plant survival was modeled using a linear mixed effects model lmer function in R statistical software. Data were aggregated by year to determine plant survival as the response variable with species as the explanatory variable and plot included as a random effect. Plot-level aggregation was used to account for variability among plots, given limited data on the plant survival within this study. Phenotypic traits, including height, basal area, and growth habit, were analyzed with a linear mixed effects model as fixed effects and plot were included as a random effect. Basal area was calculated using the area of an ellipse (A= abп) and log-transformed to better fit model assumptions. We excluded A. lentiginosus and A. lonchocarpus from the statistical analysis because of inadequate sample sizes. Post-hoc pairwise comparisons were performed using Tukey’s method (α = 0.05) to isolate species differences in each analysis (R Core Team 2023).

Pollinator Study

Of the 5 species outplanted, only A. canadensis and A. bisulcatus had sufficient overwinter survival to merit moving forward with the pollination study. We tested 3 pollination treatments on these species: 1) honeybee (Apis mellifera Linnaeus [Apidae]); 2) alfalfa leaf cutter bee (Megachile rotunda Fabricius [Megachilidae]); and 3) open pollination. Insect-proof cages (PAKGlobal, Cornelia, Georgia) with a 40 mesh size were erected in the spring of 2021 prior to flowering for the honeybee and alfalfa leaf cutter bee plots. Rodent damage to one A. canadensis alfalfa leaf cutter plot reduced the replication in this treatment to 2 plots; all other treatments had a sample size of 3 plots.

Honeybee nucs (nucleus colony), with 4 frames per nuc, were added at the beginning of the season with a 1:1 sugar water mixture as supplemental food. Alfalfa leaf cutter bees were reared on-site, and new bees were added biweekly (~100 bees) to the corresponding pollinator plots. Plots were maintained with mechanical (hand) weeding and by using Beleaf (flonicamid) (FMC Corporation, Philadelphia, Pennsylvania), a pollinatorsafe insecticide for aphid population management.

At the end of the growing season, 15 inflorescences were collected from each plot for floral measurements. In the event there were less than 15 plants per plot remaining, we doublecollected some plants. To count the fully formed, aborted, and unfertilized embryos, we used 30 individual fruits, 2 adjacent fruits per raceme. Bulk seed was also harvested from each plot to measure yield. Seeds were stored following harvest in cold dry storage at 4 °C (40 °F) and 18–20% relative humidity. Seed quality was assessed by germination assays following previously described protocols. Statistical analyses were conducted using the lmer function in the R statistical software (R Core Team 2023), using a linear mixed effects model to test treatment and species as fixed effects on yield, seed set/quality, and germination, and plot was included as a random effect.

Results

Germination

For A. bisulcatus, A. lentiginosus, and A. lonchocarpus there was an interaction effect of germination temperature dependent on cold stratification (Table 2). Without cold stratification, more seeds germinated at the lower incubation temperature (15/25 °C [59/77 °F]); with cold stratification, seed germination was similar at both incubation temperatures (Table 2). Four species had comparable germination both with and without cold stratification. Only A. drummondii responded to stratification treatment with 2–14% germination without cold stratification compared to 74–94% germination with cold stratification (P < 0.001).

View this table:
Table 2

Mean percent germination over 4 replications (+/– 1 SD) of 5 different milkvetch (Astragalus L. [Fabaceae]) species with 3 cold stratification treatments (0, 2, or 6-wk cold stratification) and 2 germination temperatures (low: 15/25 °C [59/77 °F] and high: 20/30 °C [68/86 °F]).

For all species tested, germination was comparable between the 2- and 6-wk cold stratification treatments, indicating the length of cold stratification did not have an effect (Table 2). Of the 5 species tested, A. canadensis had the least variability among treatments. We measured an effect of temperature on germination with slightly higher germination rates at the lower incubation temperature (15/25 °C [59/77 °F]), but this difference was minimal, and germination was between 69 and 93% for all treatments (Table 2). Furthermore, all 5 species of Astragalus had 90% germination or higher in at least one tested treatment.

Field Characterization

Astragalus species played a significant role (P > 0.05) in survival rates based on temporal variation, year-to-year (Table 3). We observed a species-level effect with lower rates of survival (P < 0.001) for A. drummondii, A. lentiginosus, and A. lonchocarpus from the initial (2020) to the remaining (2022) populations when compared to A. bisulcatus and A. canadensis. Although not significant, A. bisulcatus survival decreased throughout the study from a mean of 19.4 plants (81% survival) in 2021 to 15.7 (65.2%) in 2022. In comparison, A. canadensis populations rebounded from a mean of 15.4 plants (75.5%) in 2021 to 17.6 plants (95.5%) in 2022 (Table 3).

Astragalus bisulcatus, A. canadensis, and A. lentiginosus, were reproductive in 2021. Flowering for these 3 species began between 11 May and 18 May 2021 and continued throughout the growing season. The 4 species that survived to 2022 (excluding A. lentiginosus) were reproductive. Flowering began between 23 May and 30 May 2022, which was approximately a week later than in 2021. The majority of A. canadensis plants were reproductive both years, with 89.3% and 100% of individuals flowering in 2021 and 2022, respectively. We recorded that A. bisulcatus had low levels of reproduction in 2021 (22% of plants), but all individuals were reproductive in 2022. The remaining 3 species all had low numbers of surviving plants (Table 3). Eight out of 10 (80%) A. lentiginosus plants flowered in 2021; 20 out of 32 (63%) A. drummondii plants flowered in 2022; and 2 out of 6 (33%) A. lonchocarpus plants flowered in 2022.

We documented phenotypic traits to determine the mean height, basal area, and plant growth form (habit) (Figure 2, Figure 3). There was an effect of year on many phenotypic characteristics measured, with plants exhibiting growth and development throughout the study period (P < 0.05) (Table 4). Astragalus accessions differed in size and growth habit when grown in the field setting. When comparing across years (2021– 2022), we observed differences for mean heights (P < 0.05). In both 2021 and 2022, A. bisulcatus plants were the tallest plants, followed by A. canadensis, and then A. drummondii (P < 0.05). Mean height for A. bisulcatus was 22.90 cm (9.02 in), compared to A. drummondii at 8.65 cm (3.41 in). In both 2021 and 2022, A. bisulcatus had larger basal areas (P < 0.05) with a range of 2634.1 to 11,551.0 cm2 (408.3 to 1790.4 in2). This range contrasted with A. canadensis that showed a 1156.0 to 9459.0 cm2 (179.2 to 1466.2 in2) range. We scored plant growth habit on a scale from 1 (erect) to 10 (completely prostrate); A. drummondii plants were more erect than A. bisulcatus and A. canadensis (P < 0.05) (Table 4). In 2022, only A. drummondii was statistically more erect than A. bisulcatus. Given an inadequate population size for A. lentiginous and A. lonchocarpus, only observational analyses were made and shown.

View this table:
Table 3

Establishment and overwinter survival in milkvetch (Astragalus L. [Fabaceae]) research plots in Pullman, Washington, over 2 growing seasons (2021–2022).

Figure 2.
Figure 2.

Two-grooved milkvetch (Astragalus bisulcatus (Hook.) A. Gray) germplasm used in research study in Pullman, Washington, in 2022. Raceme close-up (A), fruit/pod set close-up (B), fully uncovered plot view with abundant upright/erect vegetative growth (C).

Pollinator Study

At the end of the 2022 growing season, cleaned bulk seed yields from open-pollinated plots averaged 302.0 g/plot for A. bisulcatus (100 seed weight: 0.44 g, ~227 seeds/g) and 264.0 g/ plot for A. canadensis (100 seed weight: 0.13 g, ~731 seeds/g) (Figure 4). There were no significant differences in seed yield among the pollinator treatments for either of these 2 species (P > 0.05) (Figure 4). However, for A. bisulcatus, the alfalfa leaf cutter plots produced the lowest bulk seed yields per plot (192.0 g/plot) when we considered differential plant survival in plots. The bulk seed yield per plot for A. canadensis pollinated by honeybees was 192.7 g/plot, which was lower than the other two pollinator treatments: 275.2 g/plot for alfalfa leaf cutter and 264.0 g/ plot for open pollination (Figure 4). Using estimates based on the number of plants in each plot, A. bisulcatus plants yielded a mean seed mass of 19.6 g/plant with honeybees, 16.1 g/plant with open pollinated, and 11.8 g/plant with alfalfa leaf cutter bees. When using honeybees as pollinators, mean seed yield for A. canadensis was 13.8 g/plant, with a mean of 14 plants/plot. This was compared to the yield of 12.4 g seed/plant (21 plants/ plot average) using open pollinated and 12.0 g seed/plant (23 plants/plot average) for the alfalfa leaf cutter treatment (data not shown).

Figure 3.
Figure 3.

Canadian milkvetch (Astragalus canadensis L.) germplasm used in research in Pullman, Washington, over 2 growing seasons (2021 and 2022). Upright/erect vegetative growth (A), decumbent vegetative growth (B).

View this table:
Table 4

Phenotypic traits in milkvetch (Astragalus L. [Fabaceae]) open-pollinated research plots in Pullman, Washington, over 2 growing seasons (2021 and 2022).

Figure 4.
Figure 4.

Mean seed yield (g) per 7 x 3.4 m (9 x 11 ft) plot for 2 milkvetch (Astragalus L. [Fabaceae]) species (A. bisulcatus [ASBI] and A. canadensis [ASCA]) using 3 pollinator treatments (alfalfa leaf cutter bee [ALC], honeybee [HB], and open pollinated [OP]) evaluated in research in Pullman, Washington (2022). Pollinator treatments were not significantly different (P > 0.05).

We also tested the difference in pollinator efficiency by counting the percent of ovules per fruit that were successfully fertilized and matured into fully filled seed. There were no pollinator effects on the number of filled, aborted, or unfertilized ovules for pollinator treatments for either plant species (P > 0.05) (Figure 5). Astragalus bisulcatus fruits had an average of 11.2 ovules per fruit. Of these, 3.2 matured into filled seed, 5.0 were fertilized but later aborted, and 3.0 were unfertilized. We found A. canadensis fruits had an average of 19.6 ovules per fruit. Of these, 11.9 matured into filled seed, 3.5 were fertilized but later aborted, and 4.2 were unfertilized.

Differences in mean percent germination were observed when comparing across the 2 species (P < 0.05). However, pollinator treatment results were not different in seed quality assessed in germination assays for either Astragalus species. Although minor variations were noted for germinations between pollinator treatments, rates were consistently high for both A. bisulcatus (~80%) and A. canadensis (~90%) (Figure 6).

Figure 5.
Figure 5.

Mean seed set of fully developed embryos (count) in A. bisulcatus (ASBI) and A. canadensis (ASCA) using 3 pollinator treatments (alfalfa leaf cutter bee [ALC], honeybee [HB], and open pollinated [OP]) evaluated in research in Pullman, Washington. Pollinator treatments were not significantly different (P > 0.065).

Discussion

Germination

Physical seed dormancy is well-documented in the genus Astragalus, but physiological or combinational (physical plus physiological) dormancy has been reported only in select species (Soltani and others 2020). Of the species used in this study, only A. drummondii showed strong evidence of combinational dormancy improving from 2 to 14% germination to 68 to 94% with stratification of any length of time (see Table 2). For germination temperature, we found the lower 15/25 °C (59/77 °F) night/day germination temperature to be as effective, if not better, at breaking dormancy than the higher tested temperature of 20/30 °C (68/86 °F) for all species and accessions. Smreciu and others (1988) also found cold stratification had no effect on germination of A. bisulcatus seeds from Alberta, Canada, but unlike the present study, found cold stratification had no effect on A. drummondii. Because Smreciu and others (1988) assessed germination for 30 d, and we assessed germination for only 14 d, this could indicate A. drummondii has delayed germination without cold stratification. Alternatively, correlation of interpopulation variation in germination requirements with climate, elevation, and latitude is documented for many plant species (Cortes-Fernandez and others 2021; Brown and Allen 2023) and could explain germination differences between seed accessions from diverse geographic localities.

Field Characterization

Among the Astragalus species selected for this study, A. canadensis is the most geographically widespread species (USDA NRCS 2023), which could explain its adaptability, vigor, and higher rate of plant survival in Pullman, Washington. Astragalus bisulcatus is predominantly found in the central US and into Canada (MFG 2023), and in our study it had a similar survival rate and field establishment to A. canadensis. Following initial establishment, many plants were lost over the 2020–2021 winter, especially A. drummondii, A. lonchocarpus, and A. lentiginous. At the end of June 2021, the region experienced a “heat dome” in which extremely high temperatures were recorded for 5 d (Cotlier and Jimenez 2022; Emerton and others 2022; Thompson and others 2022; USDA Climate Hubs 2022). Excessive heat stress, even with enough soil moisture, can negatively impact plant physiology through reduced root development, inhibition of photosynthesis, and transpiration processes. These damaging impacts can lead to a reduction of plant performance in early vegetative stages, field establishment, and yield (Irmak 2016). The high temperatures could have contributed to the reduction in plant survival observed at the end of the study for some species and accessions. Surprisingly, we observed a rebound in the plant population for A. canadensis. This observation could potentially be explained by volunteer seedlings arising from shattering or possible regrowth from crowns from the previous growing year.

Figure 6.
Figure 6.

Mean percent seed germination for A. bisulcatus (ASBI) and A. canadensis (ASCA) using 3 different pollinator treatments (alfalfa leaf cutter bee [ALC], honeybee [HB], and open pollinated [OP]). Pollinator treatments were not significantly different (P > 0.05).

A combination of a historical climatic event coupled with species range, distribution, and environment where the original seed was collected, could be factors contributing to the overall survival and extensive plant mortality in the first year of this study. Further research examining these environmental variations among years and locations would benefit any speciesspecific adaptations that lead to positive outcomes for Astragalus and other PGR.

Our research indicates that when maximizing seed production for Astragalus species, several obstacles must be overcome. In addition to optimized germination and field establishment, challenges including prostrate growth habit, plant overcrowding, indeterminate flowering and (or) fruiting, pod dehiscence, seed shattering, and timing of harvest complicate the efficiency of reliable and consistent production. These PGR problems are manageable, but they are amplified in commercial seed production. By the second year of our study, A. canadensis and A. bisulcatus were crowded within the plot boundaries; therefore, plants may have benefited from more space to facilitate growth and development.

Pollination Study

The impact of the pollinator treatments tested by means of seed yield and quality via germination did not differ (P < 0.05) within species. Our study was possibly too small to account for differences between pollinator treatments in our Astragalus species. In our research, the managed insect pollinators were able to fertilize the 2 species evaluated, and seed yields were comparable to open pollination. More research is needed to determine the optimal number of plants combined with the reproductive biological requirements or species-specific functions of seed set outcomes for effective management strategies.

Multiple Astragalus spp. are obligate entomophiles, serving as an important forage flower to many native and non-native pollinators, primarily bee species (Platt and others 1974; Green and Bohart 1975; Richards 1987; Richards and Myers 1997; Tanner and others 2013; Richards 2020). Bumblebees (Bombus spp.) are the top contributor to successful pollination, higher seed set, and yield in Cicer milkvetch (Astragalus cicer L.) in agricultural settings (Stroh and others 1972; Johnston and others 1975; Richards 1987; Richards and Myers 1997). Pollinators have been used in the successful management of Cicer milkvetch, with commercially reared Bombus spp. (Richards and Myers 1997) and alfalfa leaf cutter bee (Richards 2020). According to Richards (2020), Cicer milkvetch sustains and returns an adequate commercial quantity of alfalfa leaf cutter bees for successive years of forage production. Bombus spp. also frequent and are characterized as important to A. canadensis in a pollination, predispersal seed predation, and plant density study of a tall-grass prairie in Iowa (Platt and others 1974). Clement and others (2006) found Bombus spp. to be one of the most prevalent visitors to Astragalus and Onobrychis plots at the Central Ferry, Washington, PGITRU research farm. In our study, naturally occurring Bombus spp. were noted foraging the open-pollinated treatments on many occasions, which warrants further exploration in their use for seed production in A. canadensis and A. bisulcatus (Figure 7). These, and more, are all avenues for future studies on maintenance strategies for ex situ, genetically diverse PGR populations and maximum production of sufficient seed quantities in a genebank and agricultural setting for long-term conservation and restoration outcomes.

Many inherent difficulties are associated with maintaining ex situ native PGR seed from diverse geographical locations.

Climate and soils will differ from the plants’ native range, and there will be variable survival of native plants in an agricultural setting regardless of source location. In our study, 3 of the accessions we studied had less than 25% survival over the first winter in Pullman (with no species survival in Prosser). The low number of surviving plants prevented us from moving forward with pollination treatments for these species. The low survival during fall outplanting for some of the species of Astragalus evaluated could indicate that spring outplanting or direct sowing might be favored.

Figure 7.
Figure 7.

Bombus species pollinating A. bisulcatus (Hook.) A. Gray in an open-pollinated plot.

Conclusions

Scarification alone is sufficient to break dormancy of 4 of the 5 Astragalus species studied, but A. drummondii may also require a short period of cold, moist stratification. Germination at

15/25 °C (59/77 °F) was as effective, if not better, than germination at 20/30 °C (68/86 °F) for all the species studied. Given the vegetative vigor, higher rates of survival, and increase in reproductive capacities in the second year of growth, A. canadensis and A. bisulcatus could be considered as suitable candidates for conservation and revegetation projects, as well as seed nursery production and other agricultural pursuits in Eastern Washington. Use of both honeybees and alfalfa leaf cutter bees for seed regeneration in A. canadensis and A. bisulcatus was as effective as natural pollination levels in our study.

Acknowledgments

This research was supported in part by the U.S. Department of Interior, Bureau of Land Management, Seeds of Success Program, and the U.S. Department of Agriculture, Agriculture Research Service, National Plant Germplasm System.

This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (https://creativecommons.org/licenses/by-nc-nd/4.0/) permitting copying and distributing the material in any medium or format in unadapted form only, for noncommercial purposes only, provided the original work is properly cited.

References

  1. ↵
    1. Brown A,
    2. Allen PS
    . 2023. Elevation impact on seed germination requirements for two Asteraceae species. Native Plants Journal 24(1):45–53.
    OpenUrlAbstract/FREE Full Text
    1. Clement SL,
    2. Griswold TL,
    3. Rust RW,
    4. Hellier BC,
    5. Stout DM
    . 2006. Bee associates of flowering Astragalus and Onobrychis genebank accessions at a Snake River site in eastern Washington. Journal of the Kansas Entomological Society 79:254–260.
    OpenUrl
  3. ↵
    1. Cortes-Fernandez I,
    2. Cerrato MD,
    3. Ribas-Serra A,
    4. Cardona C,
    5. Gonzalez C,
    6. Gil L
    . 2021. Evidence of interpopulation variation in the germination of Eryngium maritimum L. (Apiaceae). Plant Ecology 222:1101–1112.
    OpenUrl
  4. ↵
    1. Cotlier GI,
    2. Jimenez JC
    . 2022. The extreme heat wave over western North America in 2021: an assessment by means of land surface temperature. Remote Sensing 14(3):561.
    OpenUrl
  5. ↵
    1. Emerton R,
    2. Brimicombe C,
    3. Magnusson L,
    4. Robers C,
    5. Di Napoli C,
    6. Cloke HL,
    7. Pappenberger F
    . 2022. Predicting the unprecedented: forecasting the June 2021 Pacific Northwest heatwave. Weather 77(8):272–279.
    OpenUrl
  6. ↵
    1. [ESA] Ecological Society of America
    . 2023. Washington (DC): U.S. Department of Energy.
  7. ↵
    1. Fulkerson JR,
    2. Kinter CL
    . 2023. Reproductive ecology and bee associates of Packard's milk vetch (Astragalus packardiae), a rare plant endemic to southwestern Idaho. Western North American Naturalist 83(2):243–253.
    OpenUrl
  8. ↵
    1. Green TW,
    2. Bohart GE
    . 1975. The pollination ecology of Astragalus cibarius and Astragalus utahensis (Leguminosae). American Journal of Botany 62(4):379–386.
    OpenUrlCrossRef
  9. ↵
    1. Greene SL,
    2. Carver D,
    3. Khoury CK,
    4. Irish BM,
    5. Olwell P,
    6. Prescott L
    . 2019. Seeds of Success: collateral benefits to agricultural crop improvement, research, and education. Crop Science 59(6):2429–2442.
    OpenUrl
  10. ↵
    1. Haidet M,
    2. Olwell P
    . 2015. Seeds of Success: a national seed banking program working to achieve long-term conservation goals. Natural Areas Journal 35(1):165–173.
    OpenUrlCrossRef
  11. ↵
    1. Hendron J
    . 2020. Meet the marvelous milk-vetches. URL: https://www.fws.gov/story/2021-06/meet-marvelous-milk-vetches (accessed 17 Aug 2023). Carlsbad Fish and Wildlife Office: U.S. Fish & Wildlife Service.
  12. ↵
    1. Irmak S
    . 2016. Impacts of extreme heat stress and increased soil temperature on plant growth and development. CropWatch. URL: https://cropwatch.unl.edu/2016/impacts-extreme-heat-stress-and-increased-soil-temperature-plant-growth-and-development (accessed 5 Dec 2023). Institute of Agriculture and Natural Resources: University of Nebraska-Lincoln.
  13. ↵
    1. Johnston A,
    2. Smoliak S,
    3. Hanna MR,
    4. Hironaka R
    . 1975. Cicer milkvetch for western Canada. Ottawa (Canada): Canada Department of Agriculture. Technical Reference 1536. p 6.
  14. ↵
    1. Jones TA,
    2. Bushman BS,
    3. Crockett RT,
    4. Forsyth KC
    . 2022. Scarification and pre-chilling requirements for germination of the native forb Utah trefoil (Lotus utahensis Ottley). Native Plants Journal 23:148–155.
    OpenUrlAbstract/FREE Full Text
  15. ↵
    1. Kaye TN
    . 1997. Seed dormancy in high elevation plants: implications for ecology and restoration. In: Kaye TN, Liston A, Love RM, Luoma DL, Meinke RJ, Wilson MV, compilers. Proceedings: Conservation and management of native plants and fungi: 1995 Nov 15–17; Corvallis, OR. Proceedings NPSO. p 115–120.
  16. ↵
    1. McCue K,
    2. Belt E,
    3. Yurlina, M.
    2001. Propagation protocol for Astragalus bibullatus. Native Plants Journal 2(2):131–132.
    OpenUrlFREE Full Text
  17. ↵
    1. [MFG] Montana Field Guide
    . 2023. Two-grooved Milkvetch—Astragalus bisulcatus. URL: https://fieldguide.mt.gov/speciesDetail.aspx?elcode=PDFAB0F1B0 (accessed 5 Dec 2023). Montana Natural Heritage Program.
  18. ↵
    1. National Academies of Sciences, Engineering, and Medicine
    . 2023. An assessment of native seed needs and the capacity for their supply: final report. Washington (DC): The National Academies Press.
  19. ↵
    1. [NOAA] National Oceanic and Atmospheric Administration
    . 2023. URL: https://www.weather.gov/wrh/Climate?wfo=otx (accessed 4 Jan 2024). Washington (DC): NOAA's National Weather Service.
  20. ↵
    1. Omernik JM,
    2. Griffith GE
    . 2014. Ecoregions of the conterminous United States: evolution of hierarchical spatial framework. Environmental Management 54:1249–1266.
    OpenUrlCrossRefPubMed
  21. ↵
    1. Platt WJ,
    2. Hill GR,
    3. Clark S
    . 1974. Seed production in a prairie legume (Astragalus canadensis L.). Oecologia 17(1):55–63.
    OpenUrlCrossRefPubMed
  22. ↵
    1. R Core Team
    . 2023. R: a language and environment for statistical computing. URL: http://www.R-project.org/. Vienna, Austria: R Foundation for Statistical Computing.
  23. ↵
    1. Richards KW
    . 1987. Diversity, density, efficiency, and effectiveness of pollinators of cicer milkvetch, Astragalus cicer L. Canadian Journal of Zoology 65(9):2168–2176.
    OpenUrlCrossRef
  24. ↵
    1. Richards KW
    . 2020. Effectiveness of the alfalfa leafcutter bee Megachile rotundata Fab. to pollinate four perennial legumes. Journal of Apicultural Research 59(1):69–76.
    OpenUrl
  25. ↵
    1. Richards KW,
    2. Myers TW
    . 1997. Commercially managed colonies of bumblebees for pollination of cicer milkvetch. Acta Horticulture 437:293–298.
    OpenUrl
  26. ↵
    1. Seglias AE,
    2. Williams E,
    3. Bilge A,
    4. Kramer AT
    . 2018. Phylogeny and source climate impact seed dormancy and germination of restoration-relevant forb species. PLoS One 13(2):e0191931.
    OpenUrlPubMed
    1. Smreciu EA,
    2. Currah RS,
    3. Toop E
    . 1988. Viability and germination of herbaceous perennial species native to southern Alberta grasslands. Canadian Field-Naturalis 102:31–38.
    OpenUrl
  28. ↵
    1. Soltani E,
    2. Baskin JM,
    3. Baskin CC,
    4. Benakashani F
    . 2020. A meta-analysis of the effects of treatments used to break dormancy in seeds of the megagenus Astragalus (Fabaceae). Seed Science Research 30(3):224–233.
    OpenUrl
  29. ↵
    1. Soltani E,
    2. Benakashani F,
    3. Baskin JM,
    4. Baskin CC
    . 2021. Reproductive biology, ecological life history/demography and genetic diversity of the megagenus Astragalus (Fabaceae, Papilionoideae). Botanical Review 87(1):55–106.
    OpenUrl
  30. ↵
    1. Stroh JR,
    2. Carleton AE,
    3. Seamands WJ
    . 1972. Management of Lutana cicer milkvetch for hay, pasture, seed, and conservation uses. Wyoming Agricultural Experiment Station, University of Wyoming. Technical Reference 66. p 3.
  31. ↵
    1. Tanner DA,
    2. Clark C,
    3. Pitts JP
    . 2013. Pollination biology of Astragalus phoenix (Fabaceae) with notes on the natural history of its pollinator, Anthophora porterae (Hymenoptera: Apidae). Western North American Naturalist 73(3):373–381.
    OpenUrl
  32. ↵
    1. Thompson V,
    2. Kennedy-Asser AT,
    3. Vosper E,
    4. Lo YTE,
    5. Huntingford C,
    6. Andrews O,
    7. Collins M,
    8. Hegerl GC,
    9. Mitchell D
    . 2022. The 2021 western North America heat wave among the most extreme events ever recorded globally. Science Advances 8(18):eabm6860.
    OpenUrlPubMed
  33. ↵
    1. USDA ARS GRIN-Global
    . 2024. Germplasm Resources Information Network (GRIN Taxonomy) URL: https://npgsweb.ars-grin.gov/gringlobal/query/summary (accessed Feb 2024). Beltsville (MD): National Germplasm Resources Laboratory.
  34. ↵
    1. USDA Climate Hubs
    . 2022. Northwest Climate Hub (Extreme Weather). URL: https://www.climatehubs.usda.gov/hubs/northwest/topic/2021-northwest-heat-dome-causes-impacts-and-future-outlook. Olympia (WA): U.S. Department of Agriculture.
  35. ↵
    1. [USDA NRCS] USDA Natural Resources Conservation Service
    . 2023. The PLANTS database. URL: https://plants.usda.gov (accessed 5 Dec 2023). Greensboro (NC): National Plant Data Team.
  36. ↵
    1. Volk GM,
    2. Carver D,
    3. Irish BM,
    4. Marek L,
    5. Frances A,
    6. Greene S,
    7. Khoury CK,
    8. Bamberg J,
    9. del Rio A,
    10. Warburton M L,
    11. Bretting PK
    . 2023. Safeguarding plant genetic resources in the United States during global climate change. Crop Science 63(4):2274–2296.
    OpenUrl
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools
Optimizing regeneration protocols for native Seeds of Success–collected milkvetch (Astragalus spp.) genetic resources
Bailey Hallwachs, Elizabeth Martin, Barbara Hellier, Brian M Irish
Native Plants Journal May 2025, 25 (3) 179-191; DOI: 10.3368/npj.25.3.179
Twitter logo Facebook logo Mendeley logo
Bookmark this article

Jump to section

Keywords