Seed Source Reclassification

Adopting EPA Level III Ecoregions as a better descriptor of seed source and a more relevant arbiter for seed transfer

By Dylan Levy-Boyd and Richard Haard, July 2013

There seems to be general consensus within the restoration community that utilizing genetically fit, locally adapted native plant materials improves the likelihood of planting success (SER 2004). Beyond this agreement, there are wide ranging debates about how to best implement this strategy (Kaye 2002, reprinted in the 2007 Fourth Corner Nurseries catalog). Selecting the most appropriate high quality plant material for a project usually involves finding an acceptable compromise between the best available science and the cost and time required to implement that science. Much of the compromise and debate centers on the seed source(s)[1] of the plant material to be used. This article will touch on the issue of selecting seed sources and cover how Fourth Corner Nurseries is adapting to make selecting appropriate plant material more straightforward.

Many of the species propagated for restoration have vast ranges. For example, the distribution of redosier dogwood (Cornus sericea), one of the most commonly utilized restoration plants in the Pacific Northwest, extends over most of the continent (Figure 1). Differing environmental conditions across a species’ ranges exert selective pressures which in turn can lead to populations developing genetic differences. These genetic differences amount to local adaptations. Factors that contribute to local adaptation include soil type, elevation, climate, hydrology, solar aspect, and surrounding vegetation. An important corollary is that the possibility of local adaptation is unique to each species and environmental gradient. There are examples of local adaptation and, even sympatric speciation, occurring at spatial scales as small as a few hundred meters (Nevo 2012). At the other end of the spectrum, most of the invasive species that we work so hard to combat are examples of species thriving far from their origin. Species disposed to developing genetic differences across small environmental gradients are referred to as site specialists, and conversely species that only develop genetic differences across strong environmental gradients are called site generalists. So, when selecting plant material for a widely distributed species such as redosier dogwood, does the seed source matter?

Distribution Map

Figure 1 North American distribution of Cornus sericea (from USDA plants database)

In the mid 20th century the timber industry’s replanting experiences, notably the replantings that fared poorly, led to the recognition that locally sourced seed generally yielded better stands (Randall and Berrang 2002). When Fourth Corner Nurseries first began offering native plants (in the 1980s and ’90s) project specification writers were allowing ornamental strains of natives and native species with Holarctic distributions to be used in plantings. In the nursery we noticed many of these analog strains were either disease prone or had differing environmental responses and/or phenology compared to the wild collected seed that we were bringing in from the region. We started offering multiple seed sources for the commonly requested restoration species and were one of the first growers in this region who worked to educate our customers about seed sources.

Over the years we have struggled to find the correct terminology to describe our seed sources. Terms we’ve formerly used are ‘SW Washington’ and ‘Whatcom/Skagit’. These served as a general indicator your plants did not come from Montana or Yugoslavia, but using geopolitical descriptors for seed sources creates artificial categories that lack a true ecological linkage. Research exploring  the scales of local adaptations and the implications of seed transfer between sites show that local adaptation is correlated with genetic distance (e.g. the relatedness of populations) and environmental conditions, but not with spatial distance (aka site proximity) (Montalvo and Ellstrand 2001). It is somewhat counterintuitive to disassociate local adaption from the ‘closest source available’, when selecting a seed source for a project, but in the absence of genetic and environmental data there is no inherent benefit from doing so.

With hundreds of native species in cultivation and a lack of specific genetic data for each species, native plant nurseries such as ours are adopting EPA ecoregions as a better descriptor of seed source and a more relevant arbiter for seed transfer. These ecoregions, now established across all of North America, are based upon the work of US Geological Survey geographer James Omernik, who used a weight of evidence approach to biotic and abiotic factors as a means of classification (Omernik 1987, CEC 1997, 2006). The result is a classification of ecological regions reaching from the Arctic to Central America that are subdivided into four hierarchical levels, (I,II,III,IV). It should be noted that there are two alternate ecoregion numbering systems, one developed for the United States and the second for all of North America. We are using the (EPA) Level III classification system to catalog our plant offerings, with the exception that the CEC Level III ecoregions will be used for non-US sources.

Ecoregion Poster

Figure 2 Level III Ecoregions of the Pacific Northwest

Much like the Plant Hardiness Zones used by gardeners to determine whether a plant will survive their local winter, seed transfer zones (aka seed zones) inform restoration practitioners of  the geographic areas within which plant material can be moved with little risk of  being poorly adapted. A number of models have been proposed to serve as provisional seed zones for native plants until genetic data becomes available (e.g. Bower et al. 2010). Ecoregions fall toward the simpler end of the provisional seed zone spectrum, but a study with five herbaceous species in the Willamette Valley, Oregon, suggest that that Level III ecoregions perform reasonably well as seed transfer zones (Miller 2011). Still, it is also clear when comparing ecoregions to empirically derived seed zones, such as the new tree seed zones in Washington (Randall and Berrang 2002), that a one size fits all approach is a gross approximation of the natural patterns of local adaption for any given species.


Fourth Corner Nurseries is continuing to propagate multiple seed sources of the most common restoration species. Species are being organized according to the EPA Level III ecoregion from which the seed was sourced, two notable exceptions being: 1) conifer species will be organized by the industry’s tree seed zone maps; and 2) species for which the Level III ecoregion is unknown will be organized by state. In the absence of species specific seed zones for native plants, Level III ecoregions show promise as an approximate delimiter between populations under differing ecological conditions. If your project has objectives that encompass more than replanting success or you prefer to use alternate seed zones, we are happy to work with you  to determine if our seed sources conform to project guidelines. Also, Fourth Corner Nurseries continues to provide contract growing options which allow for propagation of particularly collected seed.

More ecoregion maps and GIS resources are available from the Environmental Protection Agency (, and Commission for Environmental Cooperation  ( The National Atlas provides a useful web browser supported ecoregion mapping tool (

[1] Most plant material is sourced from seed, but cuttings, rhizomes, and bulbs can be used as well.



Bower, A., St. Clair J.B., and V.J. Erickson. 2010. Provisional seed zones for native plants.

Commission for Environmental Cooperation. 1997. Ecological regions of North America: toward a common perspective. Commission for Environmental Cooperation, Montreal, Quebec, Canada. 71p. Map (scale 1:12,500,000). Revised 2006.

Kaye, T. N. 2002. Common ground and controversy in native plant restoration: the SOMS debate, source distance, plant selections, and a restoration oriented definition of native. Pages 5-12 in D. L. Haase and R. Rose, editors. Proceedings of the conference: Native plant propagation and restoration strategies. December 12-13, 2001. Nursery Technology Cooperative and Western Forestry and Conservation Association, Eugene, OR.

Miller, S.A., Bartow, A., Gisler, M., Ward, K., Young, A.S., and T.N. Kaye.  2011. Can an Ecoregion Serve as a Seed Transfer Zone? Evidence from a Common Garden Study with Five Native Species. Restoration Ecology 19(201):268-276.

Montalvo, A.M. and N.C. Ellstrand. 2001. Nonlocal transplantation and outbreeding depression in the subshrubLotus scoparius (Fabaceae).  Am J. Botany 88(2):258-269.

Nevo. E. 2012. “Evolution Canyon,” a potential microscale monitor of global warming across life. PNAS 109(8):2960-2965.

Omernik, J.M. 1987. Ecoregions of the conterminous United States. Map (scale 1:7,500,000). Annals of the Association of American Geographers 77(1):118-125.

Randall, W.K. and Berrang, P. 2002. Washington tree seed transfer zones. Olympia: Washington Department of Natural Resources. 84 p.

Society for Ecological Restoration International Science & Policy Working Group. 2004. The SER International Primer on Ecological Restoration. tucson: Society for Ecological Restoration International.