A Restoration Practitioner's Guide to the Restoration Gene Pool Concept

THEORY

A Restoration Practitioner’s Guide to the Restoration Gene Pool Concept

by Thomas A. Jones and Thomas A. Monaco

ABSTRACT Choosing plant materials for each desired species is often one of the most difficult steps in developing a restoration plan. The Restoration Gene Pool concept was developed to clarify the options available to the ecological restoration practitioner in terms of plant materials. We present a decision-making flowchart incorporating the issues delineated in the Restora - tion Gene Pool concept. We intend to provide practitioners with a framework to make objective and defensible plant materials choices in keeping with the objectives and philosophy of the restoration project. The flowchart consists of a series of boxes with single or multiple statements to be judged as true or false. The statements deal with issues such as genetic identity, plant metapopulation, functional guild, ecoregion, invasive weed presence, site soil and climate, endangered species presence, commercial seed availability, seed contracting potential, and availability of material bred for improved stress tolerance. Implementing the flowchart is a straightforward way to apply the Restoration Gene Pool concept to a particular project, but we encourage each practitioner to personalize the flowchart to make it as practical as possible for the situations that she or he most often encounters. Keywords: plant materials, genetic identity, metapopulation, ecoregion, Restoration Gene Pool concept, ecological restoration

ne of the first steps in form- ers 1997, Brown and Amacher 1999). Brown (2004) makes the case for Oulating an ecological restoration The Restoration Gene Pool (RGP) using the functional guild concept to plan is the development of a plant concept (Jones 2003) assigns plant condense the species list to a reasonable species list. While this is typically materials to one of four restoration number of species that 1) satisfy straightforward, the next step, making gene pools (primary to quaternary) in requirements for ecosystem structure choices among plant materials for a order of declining genetic correspon - and function and 2) deter establish - given species, may be more complex. dence to the target population for the ment of undesirable species. Func- In some cases the desired species may restoration project. In this approach tional guilds are used to condense not be commercially available, while to plant material selection, adapta - species lists into groups consisting of in other cases there may be multiple tion is decoupled from genetic iden - individual species that possess similar plant materials to choose from. In tity because they often do not ecological structure and function still other cases, the only commer - correspond, particularly when the (Walker 1992, Naeem and Wright cially available material of the desired ecosystem function of the disturbed 2003, Brown 2004)—an approach species may be deemed ecologically or site has been irreversibly or signifi - that may be more pragmatic than genetically inappropriate for meeting cantly altered. In such situations, inclusive of every species on the species project objectives. plant material with highest genetic list (Walker 1992). Inclusion of a func - Ecological restoration practitioners identity (the primary restoration gene tional guild may also reduce site vul - have long been faced with a choice pool) may not be successful, in which nerability to invasion by undesirable between native and introduced plant case material from a higher-order pool species of the same functional guild material—a choice confounded by a may be substituted. This decision can (Fargione and others 2003), while general lack of consensus concerning be made individually for each plant inclusion of multiple desirable species what constitutes “being native” (Mil - species in the restored plant commu - of a particular functional guild may lar and Libby 1989, Roundy and oth - nity within the scientific context that increase stability (Chapin and others ecosystem management demands. 1997). Others argue that this may be Ecological Restoration Vol. 25, No. 1, 2007 The RGP concept provides a place for an unrealistic expectation at large scales ISSN 1522-4740 E-ISSN 1543 -4079 cultivars of native species and nonin - (D’Antonio and Meyerson 2002). ©2007 by the Board of Regents of the vasive introduced plant material when Walker (1992) suggests that guilds University of Wisconsin System. use of native material is not feasible. should be subdivided according to

12 March 20 07 ECOLOGICAL RESTORATION 25:1 non-trivial functional attributes until indigenous material, but also because Restoration Gene Pools the remaining species are mostly of the potential for irreversible cont - Primary Restoration Gene Pool redundant. This may largely resolve amination of the local gene pool for the presence of residual functional cross-pollinated species in particular • 1° RGP material may be 1) collected variation among species within func - (Millar and Libby 1989, Knapp and from the restoration site or from a tional guilds as described by Brown Rice 1994). site or bulked across multiple sites (2004). For example, cool-season However, in situations where (metapopulations) that have been grasses could be subdivided according immediate repair of ecosystem func - determined either by presumption to growth habit, phenology, or life tion is deemed more critical than or, preferably, by genetic study to be history. While the functional guild possible long-term consequences of genetically connected to it by gene approach may appear to conflict with compromising species composition, flow (seed and/or pollen transfer) the view that each species is innately the quaternary RGP will be an (Option A); or 2) collected from a important to ecosystem function acceptable alternative (Ewel and site or bulked across multiple sites of (Ehrlich and Ehrlich 1981), Chapin Putz 2004). While acknowledging the same plant community type as and colleagues (1997) point out that the risks of non-native species, these the target site elsewhere in the same this conflict is resolved when each authors argue that greater risk-tak - Level III ecoregion (U.S. Environ - species is interpreted as a member of ing is warranted when environmen - mental Protection Agency 2006) a functional guild. In other words, tal conditions have been severely (Figure 1), but without genetic con - duplication within a functional guild modified by human activity. They nectivity (Option B). is less critical to ecosystem structure believe that introduced species that • Material is “natural” rather than and function than having a single provide essential ecological or socio- “genetically manipulated” (subjected representative of each functional economic services may be used to to selection and/or hybridization) guild present. good advantage. They acknowledge and may be certified as “source-iden - The RGP concept (Jones 2003) that, in some cases, these needs are tified” pre-variety germplasm (PVG) provides a framework for clarifying better met by introduced species (Figure 2; Jones and Young 2005). the issues involved in choosing plant than native species. materials for each species itemized in If an introduced species is to be While the practitioner’s typical pre - a condensed species list developed used, it should be well established ferred alternative is to use material using functional guilds. The purpose that it has no invasive properties. indigenous to the site (Option A), of the RGP descriptor (1°, 2°, 3°, 4°) Ewel and Putz (2004) note that some contract seed production is not always is to succinctly describe the rela- non-native species are relatively feasible because of expense or the pro - tionship between the target plant innocuous, whereas others have dra - ject’s scheduled time frame. Seed costs population and the restoration plant matic negative effects on ecosystem may be high because demand for such material. Each pool is briefly de- structure and function. They material may be limited beyond the scribed below, but we refer the reader addressed the risks of species intro - restoration project. Costs might be to an earlier publication (Jones 2003) duction by making several sugges - more reasonable if there were suffi - for a more thorough discussion. tions for assessing a species’ potential cient commercial demand for the We recognize that the four pools invasiveness. The subject of predict - product such that it could become are unlikely to have equal merit in ing potential for invasiveness has part of the business’ product line. the eyes of the ecological restoration been addressed by numerous invasive When Option A is not feasible, the practitioner. Nevertheless, we in- species biologists (Mack 1996, practitioner should investigate Option clude all of them here for the sake of Goodwin and others 1999, Kolar B, material of the same metapopula - completeness. Many practitioners and Lodge 2001, Hegen and Trepl tion as determined by a genetic study. will prefer strict use of the primary 2003, Sutherland 2004). Neverthe - Material from the same plant commu - RGP (indigenous material), while less, many restoration ecologists and nity type, for example, Shiflet’s (1994) others will tolerate the use of the practitioners will find the use of rangeland cover types, elsewhere in quaternary RGP (vicarious, non- introduced species under the moni- the Level III ecoregion (U.S. Environ - native species). Besides personal pref - ker of restoration inappropriate, pre - mental Protection Agency 2006) (Fig - erence, this is also a function of ferring to refer to projects that ure 1) may be used as a surrogate. project objectives. When genetic include them as rehabilitation or Levins (1966) originally described identity issues are paramount, pri - reclamation instead. metapopulations for animal popula - mary RGP material will be consid - tions as a group of subpopulations con - ered essential to project success. This nected by immigration or emigration. is the case not only because of the A plant metapopulation is analogous possibility of maladaptation of non - to an animal metapopulation, but the

March 20 07 ECOLOGICAL RESTORATION 25:1 13 gion be used as a surrogate for empir - ically determined metapopulation boundaries.

Secondary Restoration Gene Pool • The 2° RGP is considered when 1° RGP material is 1) unavailable or 2) insufficiently stress tolerant or com - petitive to succeed, as determined by the practitioner’s best judgment derived from past experience. • Material originates from a different metapopulation, a different plant community type within the same Level III ecoregion, or a different Level III ecoregion. Such material is presumed to be genetically discon - nected from the restoration site material because of the absence of gene flow between them, though not biologically isolated, meaning that no significant sterility is found in their hybrid. • Material may originate from a single source (natural) (Option A) or from hybridization or bulking of multiple sources from ecologically similar sites (natural) (Option B). Option A is preferred when genetic specificity is desired, and Option B is preferred when genetic diversity is considered more important. • Material may or may not have been Figure 1. Level III ecoregions of the coterminous United States (U.S. Environmental Protection Agency 2006). subjected to artificial selection using plant-breeding methodology, but if genetic manipulation occurs, this mechanisms of gene flow connecting statistically supported groups of blue - defaults the material to Option C. subpopulations are pollen movement bunch wheatgrass ( Pseudoroegneria • Option A could be certified as either and seed dispersal. Metapopulations are spicata ) populations. Many of these a cultivar or any class of PVG. important because they are much less 21 metapopulations had a geograph - Options B and C could be certified likely to go extinct than an individual ical correspondence to boundaries as a cultivar or PVG “selected” or subpopulation. Additionally, metapop - delineated on Level III (Figure 1) or “tested” classes, but not source-iden - ulations are the units upon which evo - Level IV ecoregion maps. The geo - tified class (Jones and Young 2005). lutionary forces operate (Ehrlich and graphical boundaries of the metapop - Raven 1969). ulation also serve as the boundaries of Single-source material (natural Genetic studies, particularly those the primary RGP. Thus, material track) with a specific local origin employing DNA markers, can be from a single site or multiple sites (Option A) may be preferred when used to define geographic boundaries within these boundaries may serve as genetic identity is paramount and of plant metapopulations (Larson primary RGP material for any target increased fitness is unnecessary. Oth - and others 2004, Jones 2005). For in- site within these boundaries. How - erwise, a multiple-origin and/or bred stance, by using amplified fragment ever, in the absence of prior genetic material (genetically manipulated length polymorphism (a selectively research, we suggest that the bound - track) may be preferred (Option B). neutral DNA marker system), Larson aries of the same plant community In cross-pollinated species in particu - and colleagues (2004) identified 21 type within the same Level III ecore - lar, hybridization is used to increase

14 March 20 07 ECOLOGICAL RESTORATION 25:1 When certifying seed, the Associa - tion of Official Seed Certifying Agencies (AOSCA) identifies partic - ular plant material as natural or “genetically manipulated” (Young and others 2005, Figure 2). When material has been increased from a single-site collection (2°A RGP) or bulked across ecologically similar sites (2°B RGP), it qualifies as “nat - ural” track (Figure 2, right-hand col - umn) (S.A. Young, pers. comm.). However, when either 1) hybridiza - tion or bulking of material from dis - parate sites or 2) artificial selection, as described above takes place, the resultant material (2°C RGP) is on the genetically manipulated track (Figure 2, left-hand column). The use of the two tracks to describe native plant materials is further dis - cussed in Jones and Young (2005).

Tertiary Restoration Gene Pool • The 3° RGP is considered when 2° RGP material is insufficiently stress tolerant or competitive to succeed. • Material is developed by hybridiza - tion of the desired taxon with a closely related taxon in order to introduce traits of interest into the plant material of choice. The closely related taxon is presumed to be par - tially biologically isolated from the desired taxon, that is, significant, but Figure 2. Pre-variety germplasm according to AOSCA (Young and others 2005). not complete, and sterility is found in their hybrid. For purposes of this genetic diversity of plant material. applied in the development of P-7 discussion, taxa could refer to differ - While there is potential for outbreed - bluebunch wheatgrass germplasm, a ent ploidy levels, genetically sup - ing depression (Hufford and Mazer multiple-origin polycross (Jones and ported subspecies, or closely related, 2003, Jones 2005), the introduction others 2002, Jones 2003). yet distinct, species. of “not-too-distant” genetic variation Restoration settings are often par - • Seed is always genetically manipu - should have a positive influence ticularly challenging because of biotic lated but could be certified as either (Templeton 1986). Advantages in- or abiotic modifications to ecosystem a cultivar or PVG-selected or tested clude increased genetic diversity upon processes or functions (Vitousek classes, but not source-identified which natural selection may operate 1990). Therefore, artificial selection (Jones and Young 2005). and enhanced fitness due to hybrid of plant materials (Fehr 1987) may be vigor, both of which are especially appropriate to enhance stress toler - The tertiary RGP is actually a spe - critical when populations are sub - ance or competitive ability (Option cial case. It may exist for the most jected to modified environments C). In addition, selection may be widely used restoration species, but (Moritz 1999). Knapp and Rice required to develop material that suc - because it requires a major research (1994) suggested a strategy of com - cessfully produces seed in cultivation. effort, it is typically not available for bining or inter-mating seed from sev - This may require attention to such species that are less commonly sold eral locales that are representative of traits as seed yield, seed retention, and in the seed trade. the target site. Such an approach was seed germinability.

March 20 07 ECOLOGICAL RESTORATION 25:1 15 Quaternary Restoration or introduced, to secure a site in the It should be noted that the flow - Gene Pool face of invasive weeds in order to facil - chart offers two options that have not • The 4° RGP is considered when the itate the establishment of desirable been previously discussed, but may be other RGPs are insufficiently stress natives. Using chemical and mechan - of great significance (see bold-out - tolerant or competitive to succeed or ical methods, Cox and Anderson lined text boxes, Figure 3). If primary when their seed is unavailable. (2004) successfully seeded native RGP material is not available and the • Material is of a species other than the species into sown crested wheatgrass indigenous seedbank can be expected desired taxon, although it has similar (Agropyron cristatum ) stands estab - to eventually recover on its own fol - ecological structure and function, that lished after wildfire to suppress cheat - lowing disturbance, use of fast-grow - is, it is a member of the same func - grass ( Bromus tectorum ). In contrast, ing, non-persistent plant material tional guild. It may be an unrelated little success was achieved by estab - may be used as a cover crop to stabi - native species (Option A) or a nonin - lishing directly into cheatgrass. lize the soil rather than introducing vasive, introduced species (Option B). non-indigenous plant material of the secondary RGP. This includes the • Seed for either option may be natural Decision-making or genetically manipulated and could common practice of seeding sterile Flowchart be certified as either a cultivar or any cereal hybrids to preclude soil erosion class of PVG (Jones and Young 2005). In Figure 3, we present a decision- by closing the gaps created by distur - making tool in the form of a flowchart bance, until the seedbank regenerates The quaternary RGP may be for use in applying the RGP concept. the indigenous flora (Robichaud and employed when the desired species is This tool presents pertinent issues to others 2000). not available or when it is no longer the practitioner in a stepwise fashion. Secondly, material not representa - able to function properly within a The flowchart delineates an objective tive of the functional guild and, there - damaged ecosystem (D’Antonio and approach to answer a rather subjective fore, not qualifying for any RGP, may Meyerson 2002). While implementa - question: Which plant materials be necessary in extreme circum - tion of the quaternary RGP may should be used in an ecological restora - stances. For example, reclamation- restore ecological processes, it obvi - tion project? Its purpose is to force the grade material may be used to ously does not place evolutionary practitioner to analyze options, given stabilize the site, protecting it from processes on a natural trajectory, as the the limitations and realities of a partic - recurring wildfire, invasive weed other RGPs do to varying degrees. For ular situation, rather than to simply species, or soil degradation. While this reason, many practitioners do not apply a rote technique. The flowchart this action could never be considered consider the quaternary RGP accept - is not intended to be immutable. an ecological restoration activity, it able for legitimate ecological restora - Rather, the practitioner should tailor it may be a necessary step to effect tion, preferring instead the use of the to fit his or her specific applications, restoration in the future through terms “rehabilitation” or “reclamation” and to reflect personal philosophies or assisted succession (Jones 1997). for such situations (Aronson and oth - institutional policies. We believe the ers 1993, Society for Ecological flowchart will clarify plant material Restoration International 2004). selection issues for the practitioner and Application of the However, in the RGP concept, if a permit better communication and jus - Flowchart species native to the continent, though tification of plant material decisions Box 1 (Figure 3) lists two statements not indigenous to the project site, is made for the project. (1 and 2) to judge whether the project preferred to an exotic from another In the flowchart, an all-statements- objectives are consistent with the goal continent, Option A may be imple - true/any-statement-false selection is of restoration with indigenous genetic mented instead of Option B. In prac - made at each decision point, ulti - material only and two statements (3 tice, the use of exotics is more mately leading to one of the four and 4) to verify that site condition is acceptable the further the current restoration gene pools or two special - good enough to accomplish this goal. biotic and abiotic conditions are from ized options in bold-outlined text If all four statements are true, and the the desired target (D’Antonio and boxes as explained below. Multiple site is expected to recover on its own Meyerson 2002, Ewel and Putz 2004). options are also given for primary without seeding (box 3), the recom - Considerable interest is present in (A/B), secondary (A/B/C), and qua - mendation is made to employ an using the quaternary RGP as a means ternary (A/B) RGPs. The natural and annual or short-lived perennial cover to an end by what has been termed manipulated tracks used by seed certi - crop to stabilize the site until the “assisted succession.” To the best of fication officials (Young and others ecosystem repairs itself. However, if our knowledge, this term was first 2005, Jones and Young 2005) are the site cannot be expected to recover used by Hironaka (1994) to describe indicated by dashed-line and solid- without seeding, then seed must be the use of a plant species, either native line text boxes, respectively. obtained for the project. If indigenous

16 March 20 07 ECOLOGICAL RESTORATION 25:1 Figure 3. Restoration Gene Pool decision-making flowchart.

(natural track) seed is available (box 4; downgrade to the secondary RGP or sidered next (box 9). Bred material 1°A RGP) or, secondarily, seed from greater (box 7). Note that when soils may be developed from germplasm of the same metapopulation or plant or precipitation are deemed severely the desired taxon alone (“true” for community type in the same Level III challenging to establishment success, box 9), in which case the flowchart ecoregion (box 5; 1°B RGP), such the false response to the second state - leads to the 2°C RGP option. Alter - natural-track seed may be applied to ment in box 7 steers the user toward natively, bred material may be devel - the site. Alternatively, either may be the tertiary RGP. If the response to oped through hybridization between produced under contract (box 6). box 7 is “true,” secondary RGP mate - the desired taxon and a closely related If, instead, any of the statements in rial is chosen (“true” for box 7), and taxon, typically another species, sub - box 1 are false, one may proceed one may have the option of choosing species, or ploidy level (box 10). By directly from box 1 to box 7, skipping between material of specific genetic definition, this is the tertiary RGP. the primary RGP. However, a provi - identity—single-source natural track If none of these options are avail - sion is made for situations when (2°A RGP)—and material of high able (“false” for box 10), then the threatened or endangered /special sta - genetic diversity—multiple-source quaternary RGP, those taxa within the tus species are present (box 2). In such natural track (2°B RGP) (box 8). If desired taxon’s functional guild, cases, special attention is given to the artificial selection is necessary for should be considered (box 11). Qua - primary RGP even if any box 1 state - restoration success, a third option is ternary RGP materials may be North ments are false. It may be desirable to provided—genetically manipulated American natives (4°A RGP) or, if maintain as genetically pristine a state track (2°C RGP). unavailable or ineffective (“false” for as possible on sites that must be man - From this point (box 7), if the pri - box 11), non-invasive introduced taxa aged with respect to legal limitations. mary RGP material is unavailable or (4°B RGP). In both cases, the mater - If the options in boxes 4-6 are soils or precipitation are severely chal - ial may be either natural or genetically unavailable, the alternative is to lenging, bred material should be con - manipulated. Finally, if these last

March 20 07 ECOLOGICAL RESTORATION 25:1 17 options have been exhausted (“false” Ehrlich, P.R. and P.H. Raven. 1969. Differ - Knapp, E.E. and K.J. Rice. 1994. Starting for box 12), a reclamation-grade plant entiation of populations. Science 165: from seed: Genetic issues in using material may be used regardless of 1228-1232. native grasses for restoration. Restora - tion & Management Notes 12:40-45. functional guild. Ewel, J.J. and F.E. Putz. 2004. A place for alien species in ecosystem restoration. Kolar, C.S. and D.M. Lodge. 2001. Frontiers in Ecology and the Environ - Progress in invasion biology: Predicting Acknowledgments ment 2:354-360. invaders. Trends in Ecology and Evolu - We extend our appreciation to Thomas A. Fargione, J., C.S. Brown and D. Tilman. tion 16:199-204. Jones’s fellow instructors at the Bureau of 2003. Community assembly and inva - Larson, S.R., T.A. Jones and K.B. Jensen. Land Management National Training Cen - sion: An experimental test of neutral 2004. Population structure in Pseudo- ter’s course 1730-12 “Selecting Native Plant versus niche processes. Proceedings of the roegneria spicata (Poaceae:Triticeae) Material to Meet Management Goals”— National Academy of Sciences USA modeled by Bayesian clustering of Carol Spurrier, Cini Brown, Rob Fiegener, 100:8916-8920. AFLP genotypes. American Journal of Keli Kuykendall, Scott Lambert, Roger Fehr, W.R. 1987. Principles of cultivar devel - Botany 91:1789-1801. Rosentreter, and Nancy Shaw, and course opment. Volume 1. Theory and technique. Levins, R. 1966. Strategy of model building facilitators, Ann Hutchinson and Mark New York: Collier Macmillan. in population biology. American Scien - Phillips—for brainstorming and trou - Goodwin, B.J., A.J. McCallister and L. tist 54:421-431. bleshooting the flowchart. Fahrig. 1999. Predicting invasiveness Mack, R.N. 1996. Predicting the identity of plant species based on biological and fate of plant invaders: Emergent References information. Conservation Biology and emerging approaches. Biological Aronson, J., C. Floret, E. Le Floc’h, C. 13:422-426. Conservation 78:107-121. Ovalle and R. Pontanier. 1993. Restora - Heger, T. and L. Trepl. 2003. Predicting Millar, C.I. and W.J. Libby. 1989. Disney - tion and rehabilitation of degraded biological invasions. Biological Invasions land or native ecosystem: Genetics and ecosystems in arid and semi-arid lands. 5:313-321. the restorationist. Restoration & Man - I. A view from the South. Restoration Hironaka, M. 1994. Medusahead: Natural agement Notes 7:18-24. Ecology 1:8-17. successor to the cheatgrass type in the Moritz, C. 1999. Conservation units and Brown, C.S. 2004. Are functional guilds northern Great Basin. Pages 89-91 in translocations: Strategies for conserving more realistic management units than S.B. Monsen and S.G. Kitchen (eds.), evolutionary processes. Hereditas 130: individual species for restoration? Weed Proceedings: Ecology and management 217-228. Technology 18:1556-1571. of annual rangelands. General Techni - Naeem, S. and J.P. Wright. 2003. Disentan - Brown, R.W. and M.C. Amacher. 1999. cal Report INT-GTR-313. Ogden, UT: gling biodiversity effects on ecosystem Selecting plant species for ecological Forest Service Intermountain Research functioning: Deriving solutions to a restoration: A perspective for land Station. seemingly insurmountable problem. managers. Pages 1-16 in L.K. Holz - Hufford, K.M. and S.J. Mazer. 2003. Plant Ecology Letters 6:567-579. worth and R.W. Brown (compilers), ecotypes: Genetic differentiation in the Robichaud, P.R., J.L. Beyers and D.G. Revegetation with native species. Pro - age of ecological restoration. Trends in Neary. 2000. Evaluating the effective - ceedings, 1997 Society for Ecological Ecology and Evolution 18:147-155. ness of postfire rehabilitation treat - Restoration Annual Meeting, Ft. Laud - Jones, T.A. 1997. Genetic considerations for ments. General Technical Report erdale, FL, 12-15 November 1997. native plant materials. Pages 22-25 in RMRS-GTR-63. Fort Collins, CO: Proceedings RMRS-P-8. Ogden, UT: N.L. Shaw and B.A. Roundy (eds.), U.S. Forest Service Rocky Mountain U.S. Forest Service, Rocky Mountain Proceedings of a Symposium: Using Research Station. Research Station. Seeds of Native Species on Rangelands. Roundy, B.A., N.L. Shaw and D.T. Booth. Chapin III, F.S., B.H. Walker, R.J. Hobbs, General Technical Report INT-GTR- 1997. Using native species on range - D.U. Hooper, J.H. Lawton, O.E. Sala 372. USDA Forest Service Integrated lands. Pages 1-8 in L.K. Holzworth and D. Tilman. 1997. Biotic control Research Station, Ogden, UT. and R.W. Brown (compilers), Revege - over the functioning of ecosystems. Sci - _____. 2003. The restoration gene pool tation with native species. Proceedings, ence 277:500-504. concept: Beyond the native versus non- 1997 Society for Ecological Restora - Cox, R.D. and V.J. Anderson. 2004. native debate. Restoration Ecology tion Annual Meeting, Ft. Lauderdale, Increasing native diversity of cheatgrass- 11:281-290. FL, 12-15 November 1997. Proceed - dominated rangeland through assisted _____. 2005. Genetic principles and the ings RMRS-P-8. Ogden, UT: U.S. succession. Journal of Range Manage - use of native seeds. Native Plants Jour - Forest Service, Rocky Mountain ment 57:203-210. nal 6:14-18, 20-24. Research Station. D’Antonio, C.M. and L.A. Meyerson. 2002. Jones, T.A., S.R. Larson, D.C. Nielson, S.A. Shiflet, T.N. 1994. Rangeland cover types of Exotic plant species as problems and Young, N.J. Chatterton and A.J. the United States. Denver, CO: Society solutions in ecological restoration: A syn - Palazzo. 2002. Registration of P-7 blue - for Range Management. thesis. Restoration Ecology 10:703-713. bunch wheatgrass germplasm. Crop Sci - Society for Ecological Restoration Interna - Ehrlich, P.R. and A.H. Ehrlich. 1981. ence 42:1754-1755. tional Science and Policy Working Extinction: The causes and consequences Jones, T.A. and S.A. Young. 2005. Native Group. 2004. SER International Primer of the disappearance of species. New York: seeds in commerce. Native Plants Jour - on Ecological Restoration. Tucson, AZ: Random House. nal 6: 286-293. SER International.

18 March 20 07 ECOLOGICAL RESTORATION 25:1 Sutherland, S. 2004. What makes a weed a Vitousek, P.M. 1990. Biological invasions weed: Life history traits of native and and ecosystem processes: Towards an exotic plants in the USA. Oecologia integration of population biology and Thomas A. Jones is a research geneticist for 141:24-39. ecosystem studies. Oikos 57:7-13. the USDA-Agricultural Research Service, Forage and Range Research Laboratory, Templeton, A.R. 1986. Coadaptation and Walker, B.H. 1992. Biodiversity and eco - outbreeding depression. Pages 105-116 logical redundancy. Conservation Biol - Utah State University, Logan, UT in M. E. Soulé (ed.), Conservation biol - ogy 6:18-23. 84322-6300; 435/797-3082, Fax: 435/797-3075, [email protected]. ogy: The science of scarcity and diver - Young, S.A., B. Schrumpf and E. Amberson. sity. Sunderland, MA: Sinauer. 2005. The AOSCA native plant con - U.S. Environmental Protection Agency. nection. URL: http://www.aosca.org/ Thomas A. Monaco is a research ecologist 2006. Level III ecoregions. URL: aoscanativeplantbrochure.pdf (accessed for the USDA-Agricultural Research http:// www. epa.gov/wed/pages/ecore - 19 April 2005). Service, Forage and Range Research gions/level_iii.htm (accessed 19 Sept. Laboratory, Utah State University, Logan, 2006). Corvallis, OR: Western Ecology UT 84322-6300; Fax: 435/797-7231, Division. 435/797-3075, [email protected].

March 20 07 ECOLOGICAL RESTORATION 25:1 19