J. Range Manage. 56: 466-473 September 2003 Observation: Leafy spurge control in western prairie fringed orchid habitat

DONALD R. KIRBY, RODNEY G. LYM, JOHN J. STERLING, AND CAROLYN HULL SIEG

Authors are Professor, Animal and Range Sciences Department, North Dnkota State Universiry. Fargo, N.D. 58105; Professor, Plrrnt Sciences Department, State Universi& Fargo, N.D. 58105; Research Assistant, Animal and Range Sciences Department, North Dhta State Universily. Fargo, N.D. 58105; and Research Ecologist, USDA Forest Service Rocky Mountain Research Station, 2500 S. Pine Knoll Drive, Flagst4 AriL 86001.

Abstract

The western prairie fringed orchid ( praeclara "Western prairie fringed orchid" ( Sheviak and Bowles) is a of the tallgrass Sheviak and Bowles) es una especie amenazada en los pastizales prairie. Invasion by Ieafy spurge (~uphorbiaesuka L.) is a s&ous altos. La invasi6n de "Leafy spurge" (Euphorbicr esuka L.) Es una threat to western vrairie frineed orchid habitat. The obiectives of seria amenaza para el hhbitat del :Western prairie fringed this study were td develop a rherbicide treatment to cox;trol leafy orchid". Los objetivos de este estudio fueron desarrollar un spurge while sustaining western prairie fringed orchid popula- tratamiento de herbicida para controlar el "Leafy spurge" mien- tions and to evaluate the soil seedbank composition of leafy tras se mantienen poblaciones de "Western prairie fringed spurge-infested sites to guide long-term management strategies. orchid" y evaluar la composici6n del banco de semillas del suelo Quinclorac (3,7-dichloro-8-quinolinecarboxylicacid), imazapic de sitios infestados con 'Zeafy spurge" para guiar estrategias de ((~)-2-[4,5-dihydro-4-methyl-4-(l-methylethyl)-5-oxo-1W-imida-manejo a largo plazo. En el otoiio de 2 aiios consecutivos se apli- zol-2=yl]-5-methyl-3-pyridinecarboxylicacid), and glyphosate caron los siguientes herbicidas: Quinclorac (hcido 3,7-dicloro-8- [N-(phosphonomethyl)glycine] plus 2,4-D (2,4-dichlorophenoxy quinolinecarboxilico ), Imazapic {(* )-2-[hcido 4,s-dihidro-4- acetic acid) were applied in the fall for 2 consecutive years, and metil-4-(I-rnetiletil)-5-oxo-IH-imidazol-2=yll-S-metil-3- changes in leafy spurge cover, density, yield, and herbaceous yield piridinecarboxilico) y Glifosato [N-(Fosfonometil)glicine] mas were assessed. In a separate study, quinclorac, imazapic, and 2,4-D (hcido 2,4- diclorofenoxiac6tico)y se evaluaron lacambios glyphosate plus 2,4-D were each fall-applied to 12 western prairie de cobertura, densidad, rendimiento de "Leafy spurge" y el fringed orchids and assessed for reoccurrence and density of rendimiento de herbaceas En un estudio separado el Quinclorac, orchids 1-year after treatment. Quinclorac and irnazapic, but not Imazapic, and Glifosato mas 2,4-D fueron aplicados cada otoiio a glyphosate plus 2,4-D, reduced leafy spurge cover, density, and 12 'Western prairie fringed orchids" y se evalu6 la reocurrencia yield without causing deleterious effects to associated native herba- y densidad de las orquideas 1 aiio despues de aplicar el ceous cover and yields. Western prairie fringed orchid reoccur- tratamiento. El Quinclorac e imazapic, pero no el glifosato mas rence and density were unaffected by any herbicide 1 year after 2,4-D, redujeron la cobertura, densidad y rendimiento del "Leafy treatment. Soil cores were removed in spring and fall following the spurge" sin causar efectos perjudiciales en la cobertura y first year herbicide treatment, washed and placed in trays. rendiiento de las herbhceas nativas asociadas. Despub de 1aiio Seedlings were allowed to germinate for 16 weeks in the green- de aplicados los tratamientos ninguno de ellos afect6 la reocur- house. Over 50 plant species were identified in the soil seedbank, rencia y densidad de "Western vrairie fringed orchid". Muestras of which approximately 60% were early sera1 species indicative of de suelb ftleron removidas en la-primavera-y otoiio siguientes del disturbance. Given the dominance of leafy spurge in the seed primer aiio de tratamiento de herbicida y fueron lavadas y colo- bank, a long-term management program to control this noxious cadas en charolas. Se permiti6 que las plhntulas germharan en el species is warranted. Although these results are promising, longer- invernadero durante un period0 de 16 semanas. Mas de 50 term studies need be conducted to ensure that repeated herbicide especies de plantas se identificaron en el banco de semilla del treatments do not harm the western prairie fringed orchid. suelo, de las cuales aproximadamente el 60% fueron especies serales indicativas de disturbio. Dada la dominancia de "Leafy spurge" en el banco de semilla, se garantiza un programs de Key Words: invasive species, herbicides, range improvement, soil manejo a largo plazo para controlar esta especie nociva. Aunque seedbank estos resultados son vromisorios, se necesita conducir estudios a largo plazo para asebrar que I& tratamientos repetidos de her- bicidas no daiian el 'Western prairie fringed orchid". Leafy spurge ( L.) invasion in southeastern North Dakota is a threat to the western prairie fringed orchid (Platanthera praeclara Sheviak and Bowles), a federally listed threatened plant species (U.S. Fish and Wildlife Service 1989, ious weed that infests approximately 0.5 million ha in North Wolken et al. 2001). Ltafy spurge is an introduced perennial nox- Dakota (Lym et al. 2000). It is a long-lived plant native to Europe and Asia that spreads prolifically by seed and adventitious buds on roots and crowns @unn 1979). Leafy spurge can tolerate a Manuscript accepted 4 Nov. 02. variety of environmental conditions and habitats from rocky

JOURNAL OF RANGE MANAGEMENT 56(5) September 2003 buttes to sandy shorelines and is difficult pyridinecarboxylic acid) and quinclorac is complex consisting to control. Piant communities where leafy (3,7-dichloro-8-quinolinecarboxylicacid), of at least 12 distinct plant communities spurge dominates often have reduced fre- which provide good to excellent leafy (Seiler and Barker 1985). The major plant quency and density of associated native spurge control with little or no injury to communities present on the study site herbaceous species (Selleck et al. 1962). other desirable broadleaf and grass species were sedge meadow, and Leafy spurge has been detected in habitats (Kuehl and Lym 1997, Markle and Lym mixed grass prairie. The sedge meadow supporting the western prairie fringed 2001). Also. glyphosate [N-(phospho- was dominated by various sedges (Carex orchid in North Dakota (Sieg and Bjugstad nomethy1)glycineJ plus 2,4-D can be'used spp.), bluegrasses (Poa spp.), reed grasses 1994) and (Winter 1994). and for leafy spurge control in areas with a (Calamogrostis spp.), switchgrass continued expansion of this invasive high water table such as the Sheyenne (Panicwn virgatum L.), and willows (Salix species could present a serious threat to National Grassland, and has no soil resid- spp.), which invaded the site. The tallgrass the western prairie fringed orchid (U.S. ual (Lym 1998). prairie was dominated by big bluestem Fish and Wildlife Service 1996). A herbicide demonstration trial on the (Andropogon gerardii Vitman), little The original range of the western prairie Sheyenne National Grassland in the early bluestem (Schizachyrium scoparium fringed orchid extended from Manitoba, 1990's indicated that herbicides could be Michx.). switchgrass, and blue- Canada through North Dakota and used to conh-01 leafy spurge and allow the grass (Poa pratensis L.); The mixed grass Minnesota to and western prairie fringed orchid to reestab- prairie was dominated by needlegrass (Bowles 1983). Western prairie fringed lish; however, after 2 years the hid had to (Stipa spp.), blue grama [~outeloua-pa- orchid populations, once widespread be discontinued because the threatened cilis (H.B.K.) Lag ex. Griffiths], Kentucky thmughout tallgrass prairies, are now lim- western prairie fringed orchid appeared in bluegrass, and the introduced leafy spurge. ited tg isolated prairie remnants due to some plots treated with fall-applied herbi- habitat loss. One of the largest populations cides. Therefore, the present study was Leafy Spurge Herbicide Trial remaining occurs on the Sheyenne initiated with the objectives of 1) evaluate Fall-applied herbicides that generally National Grassland, a 28,750 ha federally 3 fall-applied herbicides for leafy spurge control leafy spurge and can be used in the managed tallgrass prairie in southeastern control, 2) assess the impact of these her- mesic habitat that supports the western North Dakota. Leafy spurge infests bicides on reappearance and density of the prairie fringed orchid were evaluated. approximately 15 to 20% of the Sheyenne western prairie fringed orchid and produc- Treatments included glyphosate plus 2,4- National Grassland. If leafy spurge is tion of other species, and 3) quantify the D at 220 plus 330 and 440 plus 670 g ae allowed to continue to spread into western soil seedbank composition to determine ha-', imazapic at 70 and 140 g ae ha-' plus prairie fringed orchid habitat, it could out- what plant species are present to revege- methylated seed oil' at 2.3 liters ha" plus compete the western prairie fringed tate the site following control of leafy urea-ammonium nitrate fertilizer (28% N) orchid; yet, non-selective control methods spurge. at 2.3 liters ha". and quinclorac at 880 and also pose a potential threat to the western 1100 g at ha-' plus methylated seed oil2 at prairie fringed orchid (U.S. Fish and 2.3 liters ha-'. Herbicides were applied Wildlife Service 1996). Materials and Methods with a hand-held C02-pressurized sprayer Leafy spurge has been controlled with a with four. 8001 flat-fan .nozzles3 deliver- variety of herbicides (Lym 1998). Study Area ing 80 liters ha" at 280 kPa. The treat- However, herbicides cannot be used in The study was conducted on the ments were applied 2 consecutive years in areas where the western prairie fringed Sheyenne National Grassland (46' 28' N mid-September in 1997 and 1998, to orchid is located due to its status as a latitude. 97O 16' W longitude) approxi- mature leafy spurge with 3 to 10 cm of fall threatened species. The biological control mately 80 km southwest of Fargo, N.D. regrowth on stems and branches. The agent Aphthona spp. flea beetles have The Sheyenne National Grassland is part experimental design was a randomized become widely established and can greatly of a geologic unit known as the Sheyenne complete block, replicated 4 times, with 3- reduce leafy spurge infestations, but River Delta formed approximately 10,000 by 9.1-m plots. Aphthona spp. generally do not survive to 12.000 years ago. The Sheyenne Vegetation cover and density were esti- well in the sandy, mesic habitat of the National Grassland consists of stabilized mated using 6, randomly placed 0.1-m2 westem prairie fringed orchid (Lym 1998, sandy dune-like topography with well quadrats (Daubenmire 1959) in the sum- Mundal et al. 2002). drained hummocks (uplands) and subim- mer of 1998, 10 months after the first treat- Traditionally treatments of picloram (4- gated swales (lowlands). The soils of the ment, with the final evaluations made in amino-3,5.6-trichloro-2-pyridinecar- study were Serden (Mixed, frigid typic 1999, 22 months after the first treatment. boxylic acid) plus 2,4-D (2.4-dichlorophe- Udipsamments) and Maddock-Hecla Percent canopy cover was visually estimat- noxy acetic acid) either spring or fall (Sandy, mixed, frigid Entic Hapludolls- ed for litter, leafy spurge, forb, grass, applied have been used to control leafy Sandy, mixed, frigid Oxyaquic Hapludolls) shrub, and bare ground. Individual spurge in the region (Lym 1998). loamy fine sands, which are well drained were identified and counted by species to However, picloram csnnot be used in and susceptible to wind erosion. determine density in each treatment. areas with a high water table, which often Average annual precipitation for the Herbaceous yield was estimated in includes habitats that support the western area is 48 cm with approximately 70% August 1999 (23 months after the first prairie fringed orchid. Herbicides that received during the growing season have potential to control leafy spurge May-September. Precipitation received '~eth~latedseed oil was SUN-IT I1 by' AGSCO. without harming the western prairie during this study (1997 and 1998) was 60 1160 12'St. NE, Grand Forks. N.D. 58201. fringed orchid include imazapic {(+)-2- and 78 cm, respectively, which was above '~eth~latedseed oil was SCOIL by AGSCO, 1160 [4,5-dihydro-4-methyl-4-(I-methylethyl)-the long-term average. 12' St. NE, Grand Forks, N.D. 58201. 5-0x0- 1H-imidazol-2=yl]-5-methyl-3- '~ee~et,Spraying Systems Co.. P.O. Box 7900 The vegetation of the Sheyenne Wheaton Ill. 601 89.

JOURNAL OF RANGE MANAGEMENT 56(5) September 2003 treatment) by clipping five, 0.25-m2 Western prairie fringed orchid counts ica sand. Trays were watered daily, tem- quadrats and hand separating to leafy were conducted during peak flowering in perature was maintained between 20 and spurge, grass, grass-like, native forb, and July 1999. A 1-m2 quadrat was centered 28" C, and natural and supplemental light other categories. Quadrats were randomly on each permanently marked western at 450 :p Em-2 second.' resulted in 16 located on transects placed centrally prairie fringed orchid. Western prairie hours of available light. Following gem& through the length of each plot. Yield fringed orchids inside the quadrat were nation, seedlings were identified, counted, samples from each plot were placed in counted and recorded. Herbicide treatment and removed. Seedlings that could not be individual bags, dried at 45" C for 7 days, effects on western prairie fringed orchids identified were transplanted to larger pots and weighed. were tested using ANOVA for a complete- for continuation of growth until a positive Herbicide efficacy was tested using the ly random experimental design (SPSS identification could be made. Trays were general factorial analysis of variance 1994). Means were separated using evaluated until no seedlings emerged for (ANOVA) for a randomized complete Duncan's Multiple Range Test and consid- at least 1 week, resulting in an approxi- block design. Herbicide control of leafy ered significant at P < 0.05. mately 16-week germination period. spurge was evaluated along with 'injury to Above-average precipitation during the Scientific names of plants germinated other species (SPSS 1994). Means were study and plentiful rainfall throughout the from soil cores follow that of the Flora of separated using the Duncan's Multiple 1999 growing season resulted in a high the Great Plains (Great Plains Flora Range Test and considered significant at P water table. Study western prairie fringed Association 1986). Coefficients of conser- < 0.05. orchids were under water in September vatism were assigned to plant species 1999 at the time that second-year herbi- based on an assessment provided by the cide treatments were to be applied; conse- Northern Great Plains Floristic Quality Herbicide Effect on Western Prairie quently the study was terminated. Assessment Panel (2001). Coefficients of Fringed Orchid Reoccurrence Conservatism range from a value of 0 for Authorization to treat up to 96 western plant species that flourish in highly dis- prairie fringed orchids [permit PRT- Soil Seedbank Study Soil cores for seedbank analysis were turbed habitats td 10 for species indicative 704930 (subpermit sp 98-41.993 was of undisturbed, natural habitats. obtained from the U.S. Fish and Wildlife taken from the herbicide efficacy plots in May and October 1998 Each plot was Soil seedbank composition was analyzed Service. Herbicides were the same as using the ANOVA procedure for a ran- described in the first study, except only subdivided into 2, 1.5- by 9.1-m subsec- tions with 5 soil cores sampled centrally at domized complete block design replicated the highest use rate for each treatment was 4 times (SPSS 1994). Means were separat- included. Herbicides were applied as pre- 1.5-m intervals down the long axis of each subsection. Soil cores were taken to a ed using Duncan's Multiple Range Test viously described except a 2 noule hand- and considered significant at P < 0.05. held sprayer boom was used since the depth of 2.5 cm using a standard golf-cup treated area was a 1-m2 quadrat centered cutter (10 cm diameter). Soils were com- on each western prairie fringed orchid. posited by subsection, dried at 35" C for 7 Results and Discussion Each of the 4 treatments was randomly days and placed in a refrigerator at 3" C assigned to 12 flowering (replicates) west- prior to germination. ern prairie fnnged orchids for a total of 48 Greenhouse germination procedures fol- Leafy Spurge Herbicide Trial western prairie fringed orchids in the lowed those outlined by Ter Heerdt et al. Foliar canopy cover of leafy spurge 10 study. The herbicides were applied in rnid- (1996). The soil cors were washed and 22 months after first treatment was September 1998. This timing was optimal through coarse (4 mm) and fine (2 mm) reduced by quinclorac :ind imazapic fol- for leafy spurge control with these herbi- sieves to remove roots, pebbles, sand, lowing 2 annual fall-applied treatments cides and allowed the western prairie clay, etc. Each concentrated sample was regardless of application rate (Table 1). fringed orchid to complete its life cycle; then spread from 3 to 5 mm thick in 28- Glyphosate plus 2,4-D did not consistently most of the treated western prairie fringed by 56-cm trays on a mixture of stearn-ster- reduce leafy spurge canopy cover. Grass orchids had disseminated their seeds and ilized soil and commercial potting mix canopy cover 10 months after first treat- were senescent. topped by a 4-mm thick layer of sterile sil- ment increased where imazapic at 70 g ae

Table 1. Percent canopy cover following 2 annual fall-applied herbicide treatments for leafy spurge control on the Sheyeme National Grassland, North Dakota.

- plant class2 lo MAW 22 MA* Leafy Native Leafy Native ~reatment' Rate SPwEe Graminoids forbs spurge Graminoids forbs (g ae ha.') ------(%) ...... ------..------(8)-.--.-----.----..------Control 13 cd 31 ab 16 ab 18 b 33 b 4 a Glyphosate +2.4-D 220+ 330 8 bc 33 abc 8 a 7 a 50 c 4 a Glyphosate +2.4.-D 440+670 l8d 23 a 9 a 19b 18 a 6 a Imazapic + MSO~+ 28%~~ 70 4 ab 48 d 9 a 4 a 51 c 6 a Imazapic + MSO' + 28%~~ 140 2 a 39 bcd 13ab 2 a 48 c 6 a Quinclorac + MSO' 880 2 ab 45 cd 20 ab 3 a 60 d 5 a Quinclorac + MSO' 1100 1 a 31 ab 23 b 1 a 64 d 5 a plied in mid-Septemberin 1997 and 1998 to the same plots. '~eansin columns followed by a different letter differ (P c 0.05). '~bbreviations:MAFT. months ahfmt btatment; MSO, nrthylakd seed oil al2.3 liters ha.'; 28% N, urea-ammonium nitnte f&lizer at 2.3 liw ha.'.

468 JOURNAL OF RANGE MANAGEMENT 56(5) September 2003 ha-' and quinclorac at 880 g ae ha-' were Table 2. Leafy spwge stem density following 2 annual faLapplied herbicide treatments on the applied. Grass foliar cover 22 months after Sheyenne National GrPssland, North Dakota first treatment increased compared to the control for all treatments evaluated, except ~rcatment' Rate 10 MA&> 22 M&' glyphosate plus 2,4-D at 440 plus 670 g -- (g ae K') ------(no. rn-4 ---.------ha-'. The reduced grass cover from the Control 40c 79 b high rate glyphosate plus 2.4-D treatment Olyphosate + 2.4-D 2U)+ 330 35 c 75 b was due to herbicide injury, which has Glyphosate + 2.4-D 440 + 670 20b 74b been documented in prior studies in the Imazapic + MSd + 28%N3 70 2 a 19 a northern Great Plains (Lym and Kirby Imazapic + MSO) + 28%N3 140 0 a 6 a Quinclorac+MS03 880 3 a 16 a 1991, Lym et al. 1991). Native forb cover Quinclorac + MSO~ 1100 0 a 5 a 10 and 22 months after fist treatment was 'Applied in mid-scplemkr in 1997 md 1998 to tbe sam plou. not altered regardless of treatment. kcma in cdum~foUowed by a different kttcr differ 8 < 0.05). Leafy spurge stem density was reduced 'Abbrcvinticrm: W, months aft- first bwlmcnC MSO,methylaled seed oil at 2.3 lilus ha-'; 28% N, urea-ammoni- following single and repeated quinclorac urn nibate fertiliw at 2.3 liten ha-'. and imazapic treannents at both rates eval- uated (Table 2). Leafy spurge stem density the imazapic and glyphosate plus 2.4-D orchids had completed their life cycles and 10 months after fist treatment averaged 0 treatments, and 50% of the untreated plots. were senescent when the herbicides were to 3 stems m-' for quinclorac and imazapic These levels of reoccurrence of untreated applied, so little imazapic and quinchlorac treatments compared to 40 stems m-' in the wcstnn prairie fringed orchid are similar absorption and translocation into the west- untreated control. Leafy spurge stem den- to the previous report that 55% of perma- ern prairie fringed orchid roots was sity 22 months after first treatment nently marked flowering western prairie expected. Also some evidence suggests increased slightly to 5 and 19 stems mJ, fringed orchid in grazed and ungrazed pas- that quinclorac and imaiapic aremore but the stem count nearly doubled in the tures on the Sheyenne National Grassland likely to damage dicots such as leafy untreated control to 79 stems m-'. reappeared the year following marking spurge than they are to damage monocots Generally, glyphosate plus 2.4-D did not (Sieg and King 1995). Similarly, western such as the western prairie fringed orchid reduce leafy spurge stem density com- prairie fringed orchid densities 1 year after (Lym et al. 1991, Kuehl and Lym 1997, pared to the control. treatment were not reduced on treated Masters et al. 1998, Markle and Lym Leafy spurge yield 23 months after first plots relative to control plots (Fig. 2). 2001). Cover of grasses was reduced by treatment was reduced following 2 annual Western prairie fringed orchid densities most of the herbicide treatments relative to treatments of quinclorac and imazapic, averaged 1.9, 1.3, and 0.9 plants m" in the the control, but grass production 23 regardless of application rate (Table 3). In plots treated with quinclorac, glyphosate months after first treatment in the imam contrast, grass yield following 2 annual plus 2.4-D and imazapic, respectively, pic and quinclorac treatments was higher treatments of quinclorac and imazapic sig- compared to 1.I western prairie fringed compared to the untreated plots (Table 3). nificantly increased compared to the orchids m" i3 the untreated plots. In contrast, this study and others (Lym and untreated control but not with glyphosate We attribute these results mostly to the Kirby 1991, Lym et al. 1991) document plus 2.4-D. Yields of grass-likes, forbs and timing of herbicide treatments relative to that glyphosate treatments may reduce other plant species did not differ from the the phenology of the western prairie grass production. untreated control. fringed orchid. The western prairie fringed The response of leafy spurge and grass to herbicides was similar to previous reports. Kuehl and Lym (1997) and hlasters et al. (1998) reported good leafy spurge control with fall-applied quinclorac treatments without adverse affects on for- age yields. Masters et al. (1998) and Markle and Lym (2001) reported similar control of leafy spurge with imazapic to the present study with little to no loss in herbage production. However, imazapic has been injurious to grasses when a fall treatment was followed by a spring treat- ment compared to a single fall treatment (Masters et al. 1998). Control Glyphosate + Im8zapie Quinelorne Herbicide Effect on Western Prairie 2,QD Fringed Orchid Reoccurrence Herbicide treatments Three fall-applied herbicide treatments did not reduce reoccurrence of western Fig. 1. Efleet of fall-applied herbicides on reoc~urrenceof western prairie fringed orchids the prairie fringed orchid the first year follow- season following treatment on the Sheyenne National Grassland, NJ). Herbicide rates were ing treatment (Fig. 1). Western prairie glyphosete plus 2,4D at 440 plus 670 g ae hi', fmazapic at 150 g ae ha" plus methylated f~gedorchids reoccurred in 83% of the seed oil at 23 liters ha-', and qulnchlorac at 1100 g ae hi' plus methylated seed oil at 23 plots treated with quinclorac, 60% for both liters w'. Percent reoccurrence did not qer(P > 0.05) among treatments.

JOURNAL OF RANGE MANAGEMENT 56(5) September 2003 469 Table 3. Herbaceous yields in August 1999,23 months after the first treatment, following 2 annual fall-applied herbicide treatments for leafy spurge control on the Sheyenne National Ghd,North Dakota

Plant classZ Leafy Grass- Native ~reatment' Rate Spurge Grass li kc forb Other (g ae ha-') ------(g rn-2) ------Control 101 b 120a 9 ab 38 ab 636.4a Glyphosate + 2,4-D 220 + 330 75 b 86 a 3 a 47 b 34G.4 a Glyphosate + 2,4-D 440 + 670 70b 79 a 5 a 30 ab 656.9 a Imazapic + MS~+ 28%~' 70 9 a 172 c 9 ab 26 ab 566.0 a Imazapic + MS~+ 28%~' 140 2 a 224 d 1 a 34 ab 29i4.9 a Quinclorac + MSO' 880 13a 160 bc 25 b 10 a 3W.7 a Quinclorac + MSO' 1 100 4 a 200cd 15 ab 33 ab 29d.4 a 'Applied in mid-Septembu in 1997 and 1998 to the slmc plots. '~eansin columns followed by a different lettcr differ (P < 0.05). 'Abbreviations: MSO, methylated seed oil at 23liters ham';28% N. ma-ammonium nitrate fdlizer at 2.3 liters hs".

Finally, some evidence exists that herbi- ed big bluestem, sideoats gram (Bouteloua square meter of soil surface, approximate- cides may reduce interspecific competi- curtipendula (Michx) Tom.), prairie june- ly 1400 early seral and invasive species tion, and thus may encourage the western grass (Koeleria pyramidata (Lam.) seeds and propagules could potentially prairie fringed orchid to germinate and Beauv.), little bluestein, sand dropseed germinate and cover the site. Based on develop when adequate soil moisture is (Sporobolus cryptandrus (Tom.) A. Gray), these findings, a sustained weed manage- present. Several studies have documented needleandthread (Stipa comata Trin. & ment program would be needed to suc- episodic flowering and seed production in Rupr.) and green needlegrass (Stipa viridu- cessfully revegehte the site upon which the western prairie fringed orchid that is la Trin.). Carex. Juncus and Typh species this study was conducted following con- associated with adequate soil moisture were also identified in soil cores. trol of leafy spurge. (Sieg and King 1995), and others have There was an abundaxe of viable seed speculated on the importance of distur- in the soil seedbank of the study area bances such as fire or grazing for stimulat- (Table 5). The average seedling density to Conclusions ing westem prairie fringed orchid growth a 2.5 cm depth ranged from 1570 to 2090 (Bowles 1983). seedlings mJ. Of this, 38% to over 50% Quinclorac and imazapic applied in con- were leafy spurge seedlings. Perennial secutive falls decreased leafy spurge foliar Soil Seedbank Study grasses only comprised an average of 4% cover, density and yield. Herbaceous A total of 53 genera or species were of total seedlings in the samples. Total yields, other than leafy spurge, were not identified in spring and fall soil seedbank native perennial species, i.e., grisses, reduced and generally were stimulated fol- cores (Table 4). Of these, 60% were con- forbs and Carex spp., averaged 18%, lowing quinclorac and imazapic treatment. sidered early seral species indicative of while early seral and invasive species such Fall-applied glyphosate plus 2,4-D did not disturbed tallgrass prairie. Grasses that as leafy spurge, introduced grasses, Poa control leafy spurge or result in increased germinated, which are indicative of distur- spp., forbs and others totaled 82% of all herbaceous yield. bance were the annual species, common seedlings in the samples. Thus, for every witchgrass (Panicum capillare L.) and yellow foxtail (Setaria glauca (L.) Beauv.), and perennial species, redtop (Agrostis stolonifera L.) and Kentucky bluegrass (Poa pratensis L.). Thirty-six genera or species of forbs were identified. of which less than 50% were perennial and nearly 33% were introduced species. Introduced species included common lambsquarters (Chenopodium album L.), Canada thistle (Cirsium arvense (L.) scop.), leafy spurge, kochia (Kochia sco- paria (L.) Schrad.), prickly lettuce (Lactuca serriola L.), black medic (Medicago lupulina L.), sweetclover Control Glyphorrte + Inuupic Qulnclor~c (Melilotus spp.), field thistle (Sonchus % 4-D arvensis L.) and field pennycress (Thlapsi arverase L.). Fourteen grass genera or species were Fig. 2. Average density of western prairie fringed orchids reappearing in herbicide-treated present, of which approximately 80% were areas the season following treatment on the Sheyenne National Grassland, N.D. Herbicide perennial and native species of the tallgrass rates were glyphosate plus 2,4-D at 440 plus 670 g ae ha'', imampic at 140 g ae ha" plus prairie (Table 4). Native tallgrass prairie methylated seed oil at 2.3 liters ha", and quinchlorac at 1100 g ae ha-' plus methylated species germinating from soil cores inchd- seed oil at 23 liters hi'. Densities did not dirtrer (P > 0.05) among treatments.

JOURNAL OF RANGE MANAGEMENT 56(5) September 2003 Table 4. Scientinc name, common name, season encountered, Me form, origin, and cdfieient of conservatism of plant species identined on herbidde beatments for ledy spurge control in southeastern North Dakota.

Life Coefhient of Scientific name' Common name season2 fm3 origin4 conservatism5 Forb spcdes Amaranth rctroflurus L. Rough pigweed Both Ambrosia ancmi&foliaL. Common ragweed Fall Ambrosia psilostachya DC western ragweed Spring Artcmisia dmcunculus L. Green sagewort Both Artcmisiafrigida Willd. hinged sage Boa Alumisia 1udoviciaM Nun Gray sage. pasture sage Both Aster cricoides L. Whitc asta Spring Chenopodiwn album L. Commonlambsquartus Spring Cinium orvenrc (L.) Scop. Canada thistle Both Conyur c0Mdcmi.s (L.) Cronq. Horsewecd Both Epilobium cilianun Ref. Willow hab Spring Erigcron philaddphicus L Philadelphia fleabane Spring Erigcron strigosus Daisy fleabane spring Muhl. ex Willd. Erysimwn crspcrwn (Nua.) DC Western wallflowa Boa Euphorbia esvla L. Lcofy spurge Boa Euphorbia glyptospnna Englem. Ridge seodcd spurge Both Fragaria virginiana Duchn. Wild strawbmy Boa Geranium spp. Geranium spp. Both Kochia scoparia (L.) Schrad. Kochia Pall Loctuca scrriola L. Riddy latuce Fall Lcpidiwn dens~~onunSchrad. pewgrass Both Liatris spp. Liatris spp. Spring Lycopcrr nrpcr Grme Rough bugleweed Both Medicago lupdina L. Black medic Both Melilotus spp. Swatclova Boa Ocnothera biennis L. Common evening primrose Both Oxalis aricm L. Yellow wood soml Boa Physalis virginiana P. Mill. groundchary Fall Rvdbcckia himL. Black eyed susan Both SnLrola iberica Sam. & RU Russian thistle Spring Sohum ptycanthum Dun. ex CD Black nightshade Both Solidago gigantea Ait. Late goldenrod Fall Solidago mirzouricnsis Nun. Rairie goldenrod Spring Soncharumis L. Field thistle Both Thlarpi arvensc L. field pennycress Both Urtico dioica L. Stinging nettle Pall Grapsspedeu Agrostis stolonifera L. Redtop Both Andropogon gerardii Vim Big bluestcm Both Boutcloua cunipendula Sideoats gnma spring (Michx) Tom. Dichnnthcliwn wilcoxicurum Wicox dichanthelium Both (Vasey) Fmkmann Koclcria pyrwnidata Spring (Lam.) Bcauv. Muhlenbcrgia raccmosa Marsh rnuhly Spring (Michx.) B.S.P. Pm'cum capillarc L. Common witchgrass Both Pacomprcssa L. Canada bluegrass spring Papratcnris L. Kentucky bluegrass Fall Schiulrchyriwn scoparium Little bluestcm Both (Michx.) Nash Selaria glauca (L.) Beauv. Yellow foxtail spring Sporobolus crypdms Sand dropseed Spring (Tom.) A. Gray Stipa comata Trin. & Rupr. Needleandthread Spring Sripa viridula Trin. Green needlegrass Spring 0ther Cnrer spp. Sedge spp. Both Jrmcus SDD. Rush spp. Both .. -- Cattail spp.- - Both 'scientific nmm follow the nanurlsm oftbe Flara of the Gtut Flak (Gacat Plaim Plm Association 1986). 2~carcsif thc retdlinga wuc emin thc spring, fall, cr both sample pcriodr. bfe-fam - P, pucnnial; A, annunl; B, biennial. 40rigin - N, native: I. intmdd 'A cncriicicnt vdue of 0 is indicative of species htflourish in highly disturbed habitats. while higher values (to 10) an assigned to @a from undirturbeb natural arm (Tk N0nhr.m Great Plains Floristic QdtyAslcumcnt Panel ml).

JOURNAL OF RANGE MANAGEMENT 56(5)September 2003 Table 5. Combined spring and fail soil seedbank composition for 6 herbicide treatments for leafy spurge control in southeastern North Dakota.

Perennial native Total Seedin! Leafy native Poa Native Introduced perennial introduced Treatment' Rate density spurge grasses spp. fmbs forbs Other species4 specie$ (g ae hiL) (no. m-2) ...... ---(%) ---- Control 2090 37.6 a 4.7 ab 21.3 a 10.2 b 17.1 b 9.1 a 14.9 a 38.5 a Glyphosate + 220 + 330 1590 45.4ab 4.8 ab 28.0 a 7.2 ab 7.0 a 7.6 a 12.0 a 35.3 a 24-D Glyphosate + 440+ 670 1620 51.9 b 2.7 a 22.7 a 6.3 ab 7.5 a 8.9 a 9.1 a 30.3 a 2.4-D Imam ic + 70 1430 46.7ab 3.5 ab 23.0 a 6.4 ab 8.5a 11.9a 9.9 a 31.6a MSO ? + 28%~~ lmaza ic + 140 1770 45.7ab 3.5 ab 24.9 a 5.4 a 9.4a 11.1 a 8.9 a 34.5 a MSO ?+ 28%~~

Quinclorac + MSO~ 1100 1770 42.9ab 3.9 ab 21.7 a 8.7 ab 13.3 ab 9.5 a 12.6 a 35.2 a lied in mid-September in 1997 ad1998 to the same plots. 2~vwagcnumber of seedlings mQ in soil sampled to a 2.5 crn depth. '~eaosin a column followed bv a diffucnt letter differ (P c 0.05). %cludcr perenoial and native bsesmd forb. 'kafy spurge and Othw category arc not included in total pucentage of inaoduced species. 'MSO, mcthylated seed oil 81 23 liters.'; 28% N, urea-ammonium nitrate fcltilizCl at 2.3 liten ha".

Single season data on the impact of Literature Cited Markle, D.M. and R.G. Lym. 2001. Leafy these herbicides on the western ~rairie spurge (Euphorbia esula) control and herbage production with imazapic. Weed fringed orchid demonstrated no red;ctions Bowles, M.L. 1983. The tallgrass prairie in western prairie fringed orchid reoccur- Tecbnol. 15:474-480. orchids (Nutt.) Lindl. Masters, R.H., D.D. Beran, and F. Rivas- rence or density in treated plots. This and candidum Mulh. ex. result can most likely be attributed to the Pantoja. 1998. Leafy spurge (Euphorbia Willd.: some aspects of their status, biology esula) response to AC263,222. Weed timing of the herbicide application. In this and ecology and implications toward man- Technol. 12:602409. agement. Nat. Areas 1. 3:4-37. study, western prairie fringed orchids were Mundal, DA., D.L. Olson, and RJ. CarIson sprayed after they were senescent, thus Daubenmire, R.F. 1959. A canopy-cover method of vegetational analysis. Northwest 2002. The effect of Aphthona spp. flea beetle reducing potential translocation in western (Colcoptera:Chrysornelidae) larvae feeding prairie fringed orchid. Also following Sci. 33:43-66. Dunn, P.H. 1979. The distribution of leafy on leafy spurge. Euphorbia csula L., root senescence, reduction of herbicide absorp- spurge and other Euphorbia spp. in the systems and stem density in North Dakota: tion into the roots would be expected. . Weed Sci. 27:509-516. 19861996. Biol. Control (Submitted). Longer term data are needed to fully Great Plains Flora Association. 1986. Flora Northern Great Plains Floristic Quality assess the impact of herbicide treatments of the Great Plains. University Press of Assessment Panel, The. 2001. Coefficients on western prairie fringed orchid . Lawrence, Kans. p. 1392. of conservation for the vascular flora of the regrowth, flowering, and seed production. Kuehl, B.D. and R.G. Lym. 1997. Leafy Dakotas and adjacent grasslands. U.S. Geol. A total of 53 genera or species were spurge (Euphorbia esula) control with quin- Surv., Biolo. Res. Divi., Information and identified in the soil seedbank, of which clorac. Weed Technol. 11:265-269. Tech. Rep.USGS/BRDflTR-2001-0001.p. 60% were considered early seral species Lym, R.G. 1998. The biology and integrated 32. management of leafy spurge (Euphorbia Seiler, CJ. and W.T. Barker. 1985. Vascular indicative of disturbance. Approximately csula) on North Dakota rangeland. Weed flora of Ransom, Richland, and Sargent 1500 to over 2000 seedlings m-' of soil Technol. 12:367-373. counties, North Dakota. Prairie Nat. surface to a 2.5 cm depth. from seeds and Lym, R.G. 2000. Leafy spurge (Euphorbia 17: 193-240. propagules were present in soils from the esuln) control with glyphosate plus 2.4-D. J. Selleck, G.W., R.T. Coupland, and C. study site. Over 80% of the seedlings were Range Manage. 53:68-72. Frankton. 1962. Leafy spurge in early seral species and invasive species Lym, R.G. and D.R. Kirby. 1991. Effect of Saskatchewan. Ecol. Monogr. 32:l-29. including leafy spurge, which comprised glyphosate on introduced and native grasses. Sieg, C.H. and A.J. Bjugstad. 1994. Five over half of these seedlings. Thus, a long- Weed Technol. 5421-425. years of following the western prairie fringed term weed management program will be Lym, R.G., C.G. Messersmith, and R. orchid (Platanthera praeclara) on the necessary to successfully revegetate this Zollinger. 2000. Leafy spurge identification and control. North Dakota State Univ. Ext. Sheyenne National Grassland, North Dakota, site even after leafy spurge is controlled. Serv., Fargo, N.D. 58105. W-765 (R), p. 8. p. 141-146. In: R.G. Wickett, P.D. Lewis, A. Questions concerning the effect of repeat- Lym, RG., KG. Beck, P.K. Fay, M. Fed, Woodliffe, and P. Pratt (eds.) Proc. ed fall-applied herbicide treatments on the and M. Peterson. 1991. Leafy spurge con- Thirteenth North Amer. Prairie Conf., western prairie fringed orchid in long- a trol with glyphosate plus 2.4-D: a regional Windsor, , Canada. term weed management program remain research project. Roc. West Soc. Weed Sci. to be answered. 4433-35.

JOURNAL OF RANGE MANAGEMENT 56(5) September 2003 Skg, C.H. and RM. King. 1995. Influence of U.S. Fish and Wildlife Service. 1989. Wlnter, B. 1994. Managuneot of Plnrrmthcra environmental factors and preliminary demo- Endangered and threatened wildlife and praeclara on western Minnesota Nature graphic analysis of a threatened orchid, plants: determination of threatened status for Conservancy Preserves. Western prairie PlnrMthera praeclara. Amer. Midland Nat. Platanthera leucophaea (eastern prairie fringed mhid workshop. Crookston, Minn. 134:61-77. fringed orchid) and Platanthera praeclara Woiken, PM, C.H. Sieg, Pnd S.E. W- SPSS, Inc 1994. SPSS advanced statistics. 6.1. (western prairie fringed orchid). Federal 2001. Quantifying suitable habitat of the Chicago. Ill. Register 54:39857-39862. threatened westan prairie fringed mhid. 1. Ter Heerdt, G.N.J., G.L. VerweU, R.M. U.S. Fish and Wildlife Service. 1996. Range Manage. 54:611-616. Bekker, and J.P. Bakker. 1996. An Platanthera praeclara (western prairie improved method for seedbank analysis: fringed orchid) recovery pie. U.S. Fish and seedling emagence after removing the soil Wildlife Swvice. R Snclling, Minn. p. 101. by sieving. Punc. Ecol. 10:144-151.

JOURNAL OF RANGE MANAQEMENT 58(5) September '2003