When Should You Haul Spaded Sweet Clover and Wild Parsnip?

By: Clara Barton

Key Takeaways:

  • Plants that were still flowering showed no viable seed production in our tests.
  • Plants with only green seeds produced very little germination, and the few seedlings that emerged appeared weak.
  • For both wild parsnip and sweet clover, it appears safe to leave spaded plants on site while they are still flowering or when seeds remain entirely green. Once seeds begin browning, hauling plants to a burn pile is likely worth the extra effort.
wild parsnip (Pastinaca sativa)
white sweet clover (Melilotus alba)
yellow sweet clover (Melilotus officinalis)

Anyone who has spent time controlling wild parsnip or sweet clover knows that hauling plants out of a site can be very time-consuming and laborious. In remote areas, many stewards simply sever the root and leave the plant where it falls, assuming it will die before producing viable seed. But is that assumption always correct?

To help answer that question, we tested seed viability from wild parsnip (Pastinaca sativa), yellow sweet clover (Melilotus officinalis), and white sweet clover (Melilotus alba) collected at different stages of development. Our goal was to determine when it becomes necessary to remove spaded plants from a site to prevent future seed production.

In early July, spaded plants were collected from seven treatment groups:

  • Yellow sweet clover with mostly green seeds
  • Yellow sweet clover with browning seeds
  • Yellow sweet clover still in flower
  • White sweet clover still in flower
  • White sweet clover with mostly green seeds
  • Wild parsnip still in flower
  • Wild parsnip with green seeds

The plants were dried in paper bags until late September simulating the conditions that would occur if cut plants were left on the landscape.

Green Parsnip Seeds
Hand milling seeds set up

To determine whether seeds remained capable of producing new plants, two different tests were used.

Float Test

A subset of seed groups was first tested using a float test.

Seeds were placed in 300 mL of water and observed for 15 minutes. In some species, viable seeds are more likely to sink, while immature or empty seeds tend to float.

The float test was conducted on:

  • Wild parsnip with green seeds
  • White sweet clover with green seeds
  • Yellow sweet clover with green seeds
  • Yellow sweet clover with browning seeds
Float test set up before photos

After 15 minutes, the only treatment that showed any seeds sinking was the green-seeded wild parsnip treatment.

This suggested that at least some parsnip seeds may have developed sufficiently to remain viable.

None of the sweet clover seeds sank. Follow-up research suggested the float test may not be reliable for sweet clover because the seeds possess an exceptionally hard, water-resistant seed coat. As a result, a lack of sinking should not be interpreted as evidence that sweet clover seed is nonviable.

Float test after 15 minutes

To test the repeatability of the observation, a second float test was performed on the wild parsnip seeds and produced similar results. Once again, only a small portion of seeds sank while most remained floating.

Parsnip test two next to parsnip test one both showing some sinking seeds

Germination Test

The most direct measure of viability was a germination test.

On September 30, seeds from all seven treatment groups were placed on damp paper towels, sealed in labeled plastic bags, and monitored for germination over a 15-day period.

Germination Results

TreatmentGermination Observed?Relative Amount
Yellow sweet clover – still floweringNoNone
Yellow sweet clover – green seedsYesLow
Yellow sweet clover – browning seedsYesHigh
White sweet clover – still floweringNoNone
White sweet clover – green seedsYesLow
Wild parsnip – still floweringNoNone
Wild parsnip – green seedsYesLow

The treatments classified as “low” germination produced very few seedlings relative to the number of seeds tested. Each treatment contained more than 30 seeds.

Approximate germination counts were:

  • Wild parsnip green seeds: 2-3 seedlings
  • White sweet clover green seeds: 4-6 seedlings
  • Yellow sweet clover green seeds: approximately 10 seedlings

Even among seeds that germinated, growth was limited. Seedlings remained small throughout the 15-day trial and produced neither substantial root systems nor true leaves.

By contrast, the yellow sweet clover treatment with browning seeds produced considerably higher germination rates, indicating that the seed had reached a much more viable stage of development.

Conclusions

These results suggest that timing matters when deciding whether spaded plants must be removed from a site. Plants that were still flowering produced no germinating seed in our study. Plants with entirely green seed produced only low levels of germination, and the resulting seedlings appeared weak. Once seeds began turning brown viability increased substantially.

Based on these observations, stewardship crews can likely leave spaded sweet clover and wild parsnip on site when plants are still flowering or when seeds remain completely green. However, when seed heads begin browning, the additional effort required to haul plants to a burn pile or disposal area appears justified.

Wild Parsnip with green seed forming in the central umbel. Photo taken by Dee Hudson

Related Posts

Does Yellow Sweet Clover Resprout After Mowing?

Does yellow sweet clover resprout after mowing? The short answer is yes, but the story doesn’t end there. By tracking mowed plants at Nachusa Grasslands throughout the summer, this project found that early regrowth did not lead to long-term survival, offering practical insights on the effectiveness of mowing.

Clara Barton's avatar by Clara Barton

Wild parsnip (Pastinaca sativa) control

By Don Osmond, volunteer steward Wild parsnip is a monocarpic perennial, usually taking 2-3 years to flower.  I chose a 4 acre area in southwestern Wisconsin to evaluate my control efforts over time.  Habitats ranged from dry degraded prairie to mesic lowland planted with forage.  Spring burns occurred in 2021 & 2023.  All the work…

Grassland Restoration Network blog's avatar by Grassland Restoration Network blog

Something went wrong. Please refresh the page and/or try again.

Little Bluestem Seed Collection Success Using Pull-Behind Seed Stripper

By: Clara Barton

For an upcoming winter planting project, the stewards managing the area envisioned a prairie dominated by shorter grasses rather than the expansive stands of big bluestem and feather grass often found in our restorations. To help achieve that goal, we set out to collect a large quantity of little bluestem seed while minimizing the amount of taller grass seed that ended up in the harvest.

Before collecting seed, the crew headed into the prairie and gave the area a “haircut.” Seed heads from taller grasses were clipped and removed. This made an area where little bluestem seed could be harvested using a seed stripper without worry of the taller grasses getting in the mix.

Pull-Behind Seed Stripper

Once the area was prepared, the crew drove a pull-behind seed stripper back and forth across the site.

The rotating brush head on the seed stripper efficiently removed ripe seed from the stems while leaving much of the standing vegetation intact.

Before
After

This machine was very effective in harvesting the seed!

We were extremely pleased with the results of this collection method. By removing the seed heads of taller grasses first, we were able to target little bluestem more effectively and collect seed tailored to the goals of the upcoming planting project.

This approach won’t be applicable for every restoration effort, but it proved to be an effective way to shift seed collection toward a desired species when conditions allow.

We look forward to seeing how this seed performs when it is planted this winter.

Happy seed collecting!

Establishment mowing effects on community establishment during tallgrass prairie reconstruction

Written and led by: Andrew Kaul – Restoration Scientist at the Missouri Botanical Garden Team Members: Matthew Albrecht, Cal Maginel, Brad Delfeld, Noah Dell

Key Takeaways:

  • Establishment Mowing during the first 1-2 years of development after seeding a prairie is commonly practiced to suppress weeds and improve recruitment of target species.
  • Past research on this practice has yielded mixed results and has not considered how mowing efficacy may depend on the number and timing of mowing events.
  • We initiated an experiment in 2025 at Shaw Nature Reserve in Missouri testing how establishment mowing effects prairie establishment. We mowed plots 0, 1, 2, or 4 times in the first year. We also mowed one half of all plots in the second year to test whether two years of mowing are necessary.
  • Preliminary results indicate that in year 2, differences between treatments were small, but plots mowed multiple times in year 1 had taller vegetation, and slightly lowered richness of target species. We will continue to sample this experiment to see how mowing effects change over the next three years.

Mowing can play a strategic role in managing grasslands during ecosystem restoration. Most research investigating mowing effects on grassland communities has focused on treatments applied to mature vegetation, often designed to simulate natural effects of fire or grazing, but the focus of this study is on the practice of “establishment mowing” (EM), where mowing is conducted in the first 1-2 years of vegetation development following a prairie planting into bare ground. EM is an adaptive approach used to suppress the growth of non-target “weed” species (native or non-native), which establish and grow faster than the seeded native species. Prairie species primarily grow below ground in their first year and they do not produce substantial aboveground biomass until the second or third year. In contrast, annuals and ruderal perennials grow rapidly in the first year and may suppress establishment of seeded prairie species.

The exact method of EM implementation may vary depending on region, seasonal weather, and weed composition, because these factors influence the distribution of plant growth throughout the year. The key features of EM include the number of mowing events, mowing height, and timing of mowing, especially in the first year. This practice dates back at least to the 1970s, but was popularized by Carl Kurtz’s 2001 book A Practical Guide to Prairie Reconstruction. Kurtz recommended mowing as low as 3″ to suppress weeds early in the first growing season, and then moving the mower deck up for subsequent mowing events roughly every three weeks from June through September. Some practitioners also recommend mow treatments in the second year. For example, Kurtz suggests mowing at a height of 6-8” in the second year as often as is needed. EM is widely recommended in nearly every practitioner guideline for prairie restoration and has become common in practice.

Despite these recommendations, there is limited research indicating this practice works as intended. Some managers do not use EM when starting new reconstructions, potentially due to costs and lack of evidence for return on investment. Mowing can be expensive, especially considering the time required to cover large management units multiple times a year. Additionally, EM may be counterproductive if trampling by tractor tires causes mortality of sown species as seedlings.

Past research on the value of EM to restoration outcomes has yielded mixed results. I first became interested in this topic while working on a synthesis of prairie restoration methods and outcomes in graduate school at Iowa State University. We sampled 93 planted prairies across Iowa and tested for the importance of several aspects of the restoration process on predicting diversity and composition of the resulting plant community (Kaul & Wilsey 2021). We found that mowing practice was one of the significant predictors of species richness, with sites established with EM having 4 more species on average compared to those with no mowing. However, that study was observational, and the sites had many other differences. Stronger evidence for the utility of EM would require a controlled experiment to isolate only the effects of mowing.

Several excellent studies on EM have been published in recent years. Experimental manipulations comparing EM to no mowing have found benefits in the first 1-2 years of restoration (Meissen et al. 2020, Glidden et al. 2023, Soley and Wilsey 2025), but not after longer periods (Meissen et al. 2020, Deever et al. 2024). A weakness of these studies is that they did not isolate the effects of EM based on timing or number of mowing events during the first year. This inspired us to develop an experiment that included several types of EM applications.

Our project addresses the following questions: does mowing reduce weeds and promote seeded species, and do mowing effects depend on the number of mowing events, the timing of mowing events, or the addition of mowing in the second year? Previous work has focused on the first quest, without considering variation in the method of EM implementation. In addition to control plots, our four mowing treatments include the following: a single early mow (May), a single late mow (July), two mows – early and late (May and July), or four mows – the “recommended” method (May, June, July, August). These treatments were implemented in a new prairie planting in 2025 at the Missouri Botanical Garden’s Shaw Nature Reserve. In 2026, we sampled the plant communities and implemented the 2nd-year treatment, mowing half of all the plots in late May (a split-plot design). All mowing treatments were conducted at approximately 6” using a rotary mower deck.

Pictures showing mowing treatment effects on vegetation structure in August of 2025. The t-post colored white shows a “control” plot that was not mowed; the red plot was mowed three months prior in May, the yellow was mowed one month prior in July, and the green was mowed only days prior.

At the GRN conference hosted at Shaw Nature Reserve this past May 2026, my anecdotal impression was that after touring this experiment, attendees were largely split on whether EM is important for prairie establishment. Some attendees swore by it while others insisted it was unnecessary. I am hoping this experiment will put that debate to rest one way or the other! We plan to sample the plant community annually for four years. Below are preliminary results after sampling treatments in 2025 and 2026.

As one might expect, mowing decreased the height of the plant canopy and slightly increased the amount of light reaching the ground, but only in plots mowed in July or later. PAR stands for Photosynthetically Active Radiation.

Interestingly, Year 1 mowing treatments also affected the structure of the plant community the next growing season. All treatments that included a July or later mow in year 1 had slightly taller canopies on average in June of year 2 (Red boxplots). The year 2 mow treatment (applied May 21st) resulted in shorter vegetation when measured on June 3 (Blue boxplots).

Based on the plant community in year 2, mowing treatments did not appear to affect total species richness. However, there was a slight trend with treatments mowed in July or later having fewer seeded species detected.

Following a similar trend, treatments mowed in July or later tended to have somewhat lower proportional abundance (cover) from seeded species and all native species (seeded and volunteers).

Prairie composition has high turnover in the first 3-4 years of assembly, so these early indications only provide part of the story. We will continue to track these plots, and next year we will be able to test how adding the 2nd-year mow affected vegetation as well.

References

Deever DT, Soley NM, Fullin K, Wilsey BJ (2024) Recruitment limitation of early- and late-flowering grassland forbs can be overcome with transplanting in prairie restorations. Recruitment limitation of early- and late-flowering grassland forbs can be overcome with transplanting in prairie restorations. Journal of Applied Ecology 61:223–235

Glidden AJ, Sherrard ME, Meissen JC, Myers MC, Elgersma KJ, Jackson LL (2023) Planting time, first-year mowing, and seed mix design influence ecological outcomes in agroecosystem revegetation projects. Planting time, first-year mowing, and seed mix design influence ecological outcomes in agroecosystem revegetation projects. Restoration Ecology 31:e13818

Kaul AD, Wilsey BJ (2021) Exotic species drive patterns of plant species diversity in 93 restored tallgrass prairies. Exotic species drive patterns of plant species diversity in 93 restored tallgrass prairies. Ecological Applications 31:e2252

Kurtz C (2001) A Practial Guide to Prairie Reconstruction. 1st ed. University of Iowa Press

Meissen JC, Glidden AJ, Sherrard ME, Elgersma KJ, Jackson LL (2020) Seed mix design and first year management influence multifunctionality and cost-effectiveness in prairie reconstruction. Seed mix design and first year management influence multifunctionality and cost-effectiveness in prairie reconstruction. Restoration Ecology 28:807–816

Soley N, Wilsey B (2025) Factors that extend flowering phenology for pollinators in prairie restorations. Factors that extend flowering phenology for pollinators in prairie restorations. Perspectives in Plant Ecology, Evolution and Systematics 66:125848

Testing Herbicide Treatments for Butter and Eggs

By: Clara Barton

Key Takeaways

  • Milestone + Garlon 3A + MSO provided the good suppression but caused extensive non-target damage.
  • Crossbow caused rapid injury but failed to provide lasting control.
  • Telar XP provided the best balance between suppression and selectivity.
  • Adding Talent surfactant appeared to increase the speed of injury.
  • Physical removal (spading) alone was ineffective because of extensive underground rhizomes.
  • Milestone alone appeared to have little impact which was the catalyst for this study.
  • Telar XP paired with a non-ionic surfactant will likely be our approach for future management.

Butter and Eggs (Linaria vulgaris)

Background

Butter and Eggs (Linaria vulgaris) is a rhizomatous perennial invasive plant capable of spreading through extensive underground root systems. Established populations can be difficult to control because new shoots may emerge from surviving rhizomes long after aboveground stems have been injured or removed.

At Nachusa Grasslands we have noticed Milestone treatments seem to have no impact on this plant. Mike Carr, long-time volunteer steward, started looking into other solutions. His research led him to Telar XP, which produced promising results in several infestations. During visits to these treatment areas from last year, we observed substantial differences between sites treated only with Milestone and sites that later received Telar XP. One area receiving the combined treatment remained largely free of Butter and Eggs a year after treatment, while nearby infestations remained visible. These observations became the foundation for the treatment comparisons included in this project.

Mike’s observations also raised several management questions. Telar appeared promising, but it was unclear how it compared to other treatment options. I was also interested in whether adding a surfactant would improve Telar performance, how herbicide treatments compared to physical removal, and whether the apparent benefit of applying Telar after Milestone could be reproduced in a separate infestation.

This summer, I evaluated seven treatment approaches. Two of these were meant  to replicate Mike Carr’s experiment from last year and the other 5 were meant to test other solutions. This weed is not a major problem at Nachusa Grasslands however there are some spots it is persisting in. Testing these treatment options will help us decide how to tackle this plant going forward.

Experimental Design

To compare treatment effectiveness, two large patches of Butter and Eggs infestations were divided up allowing for easy comparison.  Site 1 was dedicated to trying new treatments while site 2 was mainly used to recreate Mike Carr’s experiment.

Site 1

PlotTreatment
Q10.5% v/v Milestone + 1% v/v Garlon 3A + MSO
Q22% v/v Crossbow + MSO
Q32 g/gal Telar XP
Q42 g/gal Telar XP + Talent surfactant
Q5Spading

Site 2

PlotTreatment
AMilestone + MSO
BMilestone + MSO followed by 2 g/gal Telar XP + Talent surfactant after 2 weeks

Results

Milestone + Garlon 3A + MSO

This treatment produced the greatest level of suppression observed within the study. By mid-August, Butter and Eggs was largely absent from the treatment area and no evidence of regrowth remained. Surrounding vegetation was also heavily impacted, with much of the treated area appearing dead or severely injured

Area treated with Milestone and Garlon 3A. Note damage to off target surrounding plants.

Crossbow

Crossbow produced substantial early injury. Plants yellowed rapidly and appeared unlikely to recover during early monitoring visits. By August, however, Butter and Eggs stems were again abundant throughout much of the treatment area. Many plants appeared healthy and vigorous despite the earlier symptoms, and the infestation remained clearly visible within the plot.

Yellowed crossbow treated plants early  vs resprouts that came back later

Telar XP Treatments

Both Telar XP treatments provided substantial suppression throughout the growing season. By August, most treated stems were blackened and dead while surrounding grasses and forbs remained largely unaffected. Plants treated with Telar XP and Talent surfactant appeared to develop symptoms sooner than those treated with Telar XP alone, though by August both treatments showed similar levels of suppression.

Butter and Eggs plants completely blackened following treatment while surrounding vegetation remained largely unaffected.

Spading

Physical removal produced the weakest results of any treatment included at Site 1. Numerous shoots emerged throughout the treatment area following spading, and the infestation appeared to recover rapidly. Complete removal of belowground structures proved difficult, and surviving roots produced new growth throughout the season.

Butter and Eggs root system showing rhizomes branching at sharp angles. These underground connections make complete removal difficult and allow populations to recover quickly following spading. Repeated visits would likely be necessary to substantially reduce an infestation.

Milestone Treatments

The Big Jump infestation provided an opportunity to compare Milestone alone to Milestone followed by Telar XP and Talent surfactant. Plants treated only with Milestone remained abundant and continued flowering throughout the season. In contrast, flowering largely ceased in the area receiving the follow-up Telar XP treatment, and much of the infestation became difficult to locate by mid-August. The Milestone followed by Telar XP treatment also appeared to produce symptoms more rapidly than the Telar XP treatments at Site 1.

Area boom sprayed with Milestone, small yellow flowers seen are the butter and eggs

Discussion and Conclusions

Interpreting the control of Butter and Eggs is complicated by the plant’s extensive rhizome system. Many treatments produced obvious aboveground injury during the growing season, but visible damage does not necessarily indicate long-term control. Because this study was monitored for only a single season, the results should be viewed primarily as measures of aboveground suppression rather than definitive evidence of rhizome mortality.

The presence or absence of in-season resprouts may provide some indication of how thoroughly a treatment affected the underground root system. Treatments that produced little or no regrowth likely impacted rhizomes more effectively than treatments where plants recovered quickly. However, because Butter and Eggs can store substantial resources belowground, even heavily damaged populations may persist for some time before fully recovering or dying. As a result, caution should be exercised when interpreting a single season of observations.

The Milestone + Garlon 3A + MSO treatment produced the greatest level of suppression observed during the study, with no observable resprouts appearing within the treatment area. However, it also caused extensive injury to surrounding vegetation. Because Butter and Eggs often grows intermixed with desirable prairie species, broad use of this treatment would likely create substantial areas of bare ground and could increase opportunities for reinvasion by other undesirable species.

Crossbow produced a very different result. Plants initially appeared unlikely to recover, but by August the infestation remained readily visible throughout much of the treatment area. These observations suggest that the treatment caused substantial injury to aboveground growth while having a more limited impact on the underground rhizome network.

The Telar XP treatments provided the best balance between suppression and selectivity. Both treatments substantially reduced the infestation while leaving most surrounding vegetation unaffected. Plants receiving Telar XP and Talent surfactant appeared to develop symptoms sooner than those receiving Telar XP alone, although differences between the treatments became less apparent later in the season. A small number of green stems remained within both plots. Because the original infestation was dense, it is difficult to determine whether these stems represented true resprouts or simply plants that received incomplete herbicide coverage during application.

Observations from Big Jump further supported patterns previously reported by Mike Carr. In both cases, Butter and Eggs treated with Milestone followed by Telar XP showed substantially greater suppression than populations treated with Milestone alone. Similar results were observed in treatment areas established by Mike in previous years and in the Big Jump comparison monitored during this project. Together, these observations suggest that the treatment sequence consistently outperforms Milestone alone, which showed little evidence of suppression.

One question raised by these observations is why Butter and Eggs appeared to respond more strongly when Telar XP followed a Milestone treatment than when Telar XP was applied alone. The Milestone-only treatment showed no evidence of suppression, yet plants treated with Telar XP after Milestone appeared to decline more rapidly than those treated with Telar XP alone. This suggests that Milestone may be affecting the plant in ways that are not immediately visible aboveground or may be interacting with the follow-up Telar treatment in a way that enhances control. The mechanism responsible for this response remains unclear and warrants additional study. From a management perspective, however, the approach presents an important tradeoff. While the results were promising, applying two separate herbicide treatments requires additional field visits and may be less practical than a single well-timed application of Telar XP.

The patterns observed throughout this study also highlight the importance of herbicide residual activity. Crossbow produced rapid injury shortly after application, but much of the infestation recovered by August. This response is consistent with a treatment that effectively injures existing shoots but provides less prolonged suppression of emerging growth. In contrast, both Telar XP treatments continued suppressing Butter and Eggs throughout the growing season despite initially appearing less dramatic. Telar XP (chlorsulfuron) is known for providing both foliar and residual control, which may help explain the continued suppression observed in these plots. The Milestone + Garlon 3A + MSO treatment produced no observable resprouts, potentially due in part to the residual activity of aminopyralid, the active ingredient in Milestone. For a rhizomatous species such as Butter and Eggs, herbicides that remain active long enough to affect emerging shoots and underground buds may ultimately provide greater control than treatments that primarily produce rapid top-kill.

Based on this year’s observations, Telar XP combined with a non-ionic surfactant provided the best balance between suppression and selectivity. The treatment substantially reduced Butter and Eggs while causing minimal injury to surrounding vegetation. At the same time, the strongest suppression observed during this project occurred where Telar XP followed a Milestone treatment, lending further support to observations previously made by Mike Carr. Although the mechanism behind this response remains uncertain, the consistency of the pattern across multiple infestations suggests that the treatment sequence deserves further investigation.

This project began with a relatively simple question: which herbicide works best on Butter and Eggs? The results suggest that the answer is more complicated than expected. Some treatments produced rapid visible injury but failed to provide lasting suppression, while others continued affecting the infestation long after application. While the crew’s mixture did work, I think the use of a more selective herbicide like Telar XP should be our strategy for this weed in the future. Special thanks to Mike Carr for starting this experiment in 2025 and supplying me with his insights and observations to continue it this year.

Photo taken by Dee Hudson

Testing Herbicide Options for Dahurian Lespedeza Control

By: Clara Barton

Key Takeaways:

  • All three herbicide treatments tested resulted in complete mortality of treated Dahurian lespedeza plants within one month.
  • Garlon 3A (triclopyr) produced the fastest visible response, with plants appearing dead within one week.
  • Journey (imazapic) also acted quickly and was highly effective with long lasting residual properties, but impacted non-target plants.
  • Milestone (aminopyralid) controlled Dahurian lespedeza successfully but acted more slowly than the other treatments.

Background

Dahurian lespedeza (Lespedeza durica) is a perennial legume native to eastern Asia that was accidentally introduced to Nachusa Grasslands through a purchased seed mix. More on that here: Beware what seed you buy | grassland restoration network

Map from Kew Gardens mapping the distribution of Lespedeza durica areas it is native to are in green and areas its been introduced are in purple.

Dahurian lespedeza is currently present in a handful of locations in the United States. While its distribution remains limited today, many invasive plants begin as isolated populations before expanding into new areas. Because of this, there is an opportunity to act early and prevent Dahurian lespedeza from becoming a larger management challenge. Understanding how to effectively control the species now may help prevent future spread both within Nachusa Grasslands and beyond.

Observations of Dahurian lespedeza on iNaturalist

For several years, the restoration crew has controlled Dahurian lespedeza using a broad-spectrum herbicide mixture containing, Milestone (aminopyralid) and Garlon 3A (triclopyr), that targets multiple invasive species simultaneously. While the mixture has consistently provided effective control, it was unclear which herbicide was primarily responsible for killing Dahurian lespedeza. To answer that question, we separated two key components of the mix, Milestone and Garlon 3A, and tested them individually. If one herbicide proved responsible for most of the control, future treatments could potentially be simplified without sacrificing effectiveness.

We also included Journey (imazapic)  in the trial. Nachusa Project Manager Bill Kleiman had previously used Journey to treat Dahurian lespedeza in the field due to him having nothing else on hand when finding these plants. He observed strong control using Journy in the field, but its performance had never been directly compared to the individual components of the crew’s mixture.

By comparing Milestone, Garlon 3A, and Journey, we aimed to identify which herbicides provide reliable control of Dahurian lespedeza and inform future efforts to prevent its spread.

Experimental Design

In mid-July, we located and flagged Dahurian lespedeza plants within the study area. 31 plants were assigned to one of three herbicide treatments. Plants were vegetative at the time of treatment. More plants were found after our initial treatment and were treated but not included in this study. This plant can be hard to find and ID here is a blog post that might help: Lespedeza: friend and foe | grassland restoration network

GroupTreatment
1Garlon 3A 1% v/v + 0.5% MSO
2Milestone 0.5% v/v + 0.5% MSO
3Journey 1.5% v/v + 0.5% MSO

Results

The week following treatment was characterized by extremely hot weather, which may have accelerated herbicide activity and plant stress. The treatment groups exhibited noticeably different symptoms. Plants treated with Garlon 3A showed the fastest response. Every plant appeared dark brown and dead within one week of treatment. Milestone-treated plants displayed yellowing and wilting. While some green tissue remained, plants were beginning to brown and showed clear signs of decline. Journey produced severe distortion symptoms. Plants were browned, curled, and exhibited rapid injury. Large plants appeared especially susceptible, showing extensive damage within the first week.

Garlon 3A
Milestone
Journey
Plants after one week

After a month, all marked plants across all three treatment groups were completely brown and dead. While target plant results were identical among treatments, notable differences remained in the amount of collateral vegetation damage observed around treated plants. The Journey treatment group produced the greatest amount of off-target injury, this is not surprising given Journey is a broad-spectrum herbicide. Areas surrounding Journey treated plants showed visibly greater impacts to neighboring vegetation compared to the Milestone and Garlon 3A treatments. The Milestone group eventually reached complete mortality but required more time to do so than either Journey or Garlon 3A. The Garlon 3A plants showed little change between monitoring visits because they had already died rapidly following treatment.

Discussion and Conclusions

The findings provide useful insight into the performance of the mixture used by the crew. Since both Garlon 3A and Milestone successfully controlled Dahurian lespedeza when applied independently, either herbicide could be contributing substantially to the effectiveness of the mixture. However, the rapid mortality observed in the Garlon 3A treatment suggests that triclopyr may play an especially important role.

Journey also appears to be a highly effective Dahurian lespedeza treatment. However, Journey’s effectiveness came with increased off-target damage. The visible injury to surrounding vegetation suggests caution may be warranted when using Journey in high-quality restoration areas where protecting non-target species is a priority. In the study area we observed bare ground patches from where Journey was sprayed in previous years, indicating longer-lasting effects on germinating or emerging vegetation. Journey isn’t the only herbicide in this experiment to have longer residual activity. Both Milestone and Journey can provide continued suppression of susceptible seedlings after application unlike Garlon 3A, which primarily targets existing plants with relatively limited soil persistence. For Dahurian lespedeza, this residual activity could be beneficial by preventing new plants from establishing from the seed bank. However, the same properties that improve long-term control may also increase the risk of impacts to desirable broadleaf species.

From a management perspective, all three herbicides appear capable of controlling Dahurian lespedeza. Which one is used should be determined based on factors such as surrounding vegetation, estimations of the seedbank density, and the quality of the area being sprayed. While additional testing across multiple sites would be valuable, the results clearly demonstrate that all three herbicides can provide effective control of Dahurian lespedeza.

spot the hidden Dahurian lespedeza!
Sometimes plants are easy to spot, sometimes they are hidden in taller vegetation so be on the lookout for this plant!