December 2012 Number 1 - Utah Native Plant Society

Calochortiana December 2012 Number 1 

December 2012 

Number 1 

CONTENTS 

Proceedings of the Fifth Southwestern 

Rare and Endangered 

Plant Conference 

Calochortus nuttallii (Sego lily), 

state flower of Utah. By Kaye 

Thorne. 

Calochortiana, a new publication of 

the Utah Native Plant Society . . . . . 3 

The Fifth Southwestern Rare and Endangered 

Plant Conference, Salt Lake 

City, Utah, March 2009 . . . . . . . . . . 3 

Abstracts of presentations and posters 

not submitted for the proceedings . . . 4 

Southwestern cienegas: Rare habitats 

for endangered wetland plants. 

Robert Sivinski . . . . . . . . . . . . . . . . . 17 

A new look at ranking plant rarity for 

conservation purposes, with an emphasis 

on the flora of the American 

Southwest. John R. Spence . . . . . . . 25 

The contribution of Cedar Breaks National 

Monument to the conservation 

of vascular plant diversity in Utah. 

Walter Fertig and Douglas N. Reynolds 

. . . . . . . . . . . . . . . . . . . . . . . . . 35 

Studying the seed bank dynamics of 

rare plants. Susan Meyer . . . . . . . . . 46 

East meets west: Rare desert Alliums 

in Arizona. John L. Anderson . . . . . . 56 

Spatial patterns of endemic plant species 

of the Colorado Plateau. Crystal 

Krause . . . . . . . . . . . . . . . . . . . . . . . . 63 

Continued on page 2 

Copyright 2012 Utah Native Plant Society. All Rights Reserved.

Utah Native Plant Society 

Utah Native Plant Society, PO Box 520041, Salt Lake 

City, Utah, 84152-0041. www.unps.org 

Editor: Walter Fertig (walt@kanab.net), 

Editorial Committee: Walter Fertig, Mindy Wheeler, 

Leila Shultz, and Susan Meyer 

Copyright 2012 Utah Native Plant Society. All Rights 

Reserved. Calochortiana is a publication of the Utah 

Native Plant Society, a 501(c)(3) not-for-profit organization 

dedicated to conserving and promoting stewardship 

of our native plants. 

CONTENTS, continued 

Biogeography of rare plants of the Ash Meadows National Wildlife Refuge, Nevada. Leanna Ballard . . . . . . . . . 77 

Assessing vulnerability to climate change among the rarest plants of Nevada’s Great Basin. Steve Caicco . . . . . 91 

Sentry milkvetch (Astragalus cremnophylax var. cremnophylax) update. Janice Busco . . . . . . . . . . . . . . . . . . . . 106 

A tale of two single mountain alpine endemics: Packera franciscana and Erigeron mancus. James F. Fowler, 

Carolyn Hull Sieg, Brian M. Cassavant, and Addie E. Hite . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110 

Long-term population demographics and plant community interactions of Penstemon harringtonii, an endemic 

species of Colorado’s western slope. Thomas A. Grant III, Michele E. DePrenger-Levin, and Carol Dawson . . 115 

Conservation and restoration research at The Arboretum at Flagstaff. Kristin E. Haskins and Sheila Murray . . . . 120 

The digital Atlas of Utah Plants: Determining patterns of biodiversity and rarity. Leila M. Shultz, R. Douglas 

Ramsey, Wanda Lindquist, and C. Garrard. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122 

Molecular genetic diversity and differentiation in Clay phacelia (Phacelia argillacea Atwood: Hydrophyllaceae). 

Steven Harrison, Susan E. Meyer, and Mikel Stevens . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127 

A taxonomic revision of Astragalus lentiginosus var. maricopae and Astragalus lentiginosus var. ursinus two 

taxa endemic to the southwestern United States. Jason Alexander . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134 

Ecology of Rusby’s milkvetch (Astragalus rusbyi), a rare endemic of northern Arizona ponderosa pine forests. 

Judith D. Springer, Michael T. Stoddard, Daniel C. Laughlin, Debra L. Crisp, and Barbara G. Phillips . . . . . . 157 

Long-term responses of Penstemon clutei (Sunset Crater beardtongue) to root trenching and prescribed fire: 

Clues for population persistence. Judith D. Springer, Peter Z. Fulé, and David W. Huffman . . . . . . . . . . . . . . . . 164 

¡Viva thamnophila! Ecology of Zapata bladderpod (Physaria thamnophila), an Endangered plant of the Texas- 

Mexico borderlands. Dana M. Price, Christopher F. Best, Norma L. Fowler, and Alice L. Hempel . . . . . . . . . 172 

Intraspecific cytotype variation and conservation: An example from Phlox (Polemoniaceae). Shannon D. 

Fehlberg and Carolyn J. Ferguson . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189 

Prioritizing plant species for conservation in Utah: Developing the UNPS rare plant list. Walter Fertig . . . . . . . . 196 

2


Calochortiana, a New Publication of the Utah Native Plant Society 

Hundreds of scientific journals already exist for the dissemination of research on botany and ecology (including 

several fine publications based in Utah and the west). Nonetheless, space and financial constraints prevent many useful 

papers from being published in first and second-tier journals, relegating such work to the gray literature. In June 

2012, the board of the Utah Native Plant Society (UNPS) recognized the need for a peer-reviewed, electronic journal 

for unpublished gray-literature reports that pertain to Utah botany and vegetation. The board voted to establish an 

annual, technical journal that would complement its bimonthly member’s magazine, the Sego Lily. The objective of 

the new publication, named Calochortiana (“of or relating to Calochortus or Sego Lily”, the state floral emblem of 

Utah), is to provide a forum for professional and amateur scientists to share their findings on Utah botany and ecology 

with their colleagues. Calochortiana will focus primarily on monitoring or status surveys of rare species, seed 

propagation protocols, floristic checklists, genetic studies, vegetation mapping, natural history research, or other topics 

that might not otherwise be accepted in existing journals. All submissions will be peer-reviewed and the journal 

made available for free on the UNPS website. The journal is put together by an all-volunteer editorial board, though 

supported by UNPS. Readers, of course, are encouraged to show their appreciation by becoming members of UNPS! 

This first issue of Calochortiana contains papers presented at the 5th Southwestern Rare and Endangered Plant 

Conference, hosted by UNPS in March 2009. These papers were originally intended for publication by the US Forest 

Service as part of a proceedings volume. Unfortunately, staff changes, budget shortfalls, and new policy review requirements 

greatly delayed publication of the proceedings by the Forest Service. In October 2012, UNPS assumed 

responsibility for disseminating the conference papers to help launch its new journal. The second issue of Calochortiana 

will be published on the UNPS website (www.unps.org) in the fall of 2013. Submissions for that issue will be 

accepted through 30 April 2013. For more information, please contact me (walt@kanab.net). - Walter Fertig 

The Fifth Southwestern Rare and Endangered Plant Conference 

Salt Lake City, Utah, March 2009 

In late 2007, botanists in the southwestern United States began discussions about holding a region-wide rare plant 

conference modeled after the Fourth Southwestern Rare and Endangered Plants meeting held in Las Cruces, New 

Mexico in 2004. It was widely acknowledged through the botanical grapevine that it ought to be Utah’s turn to host 

the event. Mindy Wheeler, who was chair of the Utah Native Plant Society (UNPS) at the time, proposed that the 

Society take the lead in organizing a conference, slated for early spring 2009. UNPS already had experience with cohosting 

the annual state rare plant meeting with Red Butte Garden, so how hard could a regional conference be? 

Without going into the gory details, the months of developing an agenda, finding a venue, creating a website, signing 

up sponsors, sending out invitations to speakers and attendees, organizing field trips, hiring caterers, and completing 

hundreds of other tasks all just seemed to whisk by. On the evening of March 16, 2009, UNPS was proud to host 

the first event of the Fifth Southwestern Rare and Endangered Plant Conference - an informal mixer at historic Fort 

Douglas on the campus of the University of Utah in Salt Lake City. Fortified by good food, fine spirits, and excellent 

company, the organizers and participants of the conference were off to a good start. 

The conference officially began the following morning. Noel Holmgren, curator emeritus of the New York Botanical 

Garden, gave the keynote address in which he briefly outlined the history of the Garden’s Intermountain Flora 

project and described patterns of species richness and endemism in the Great Basin, Colorado Plateau, and the rest of 

the Southwest. UNPS presented Noel and Pat Holmgren with hand-crafted lanyards (for their hand lenses) in appreciation 

of their decades of work on the Intermountain Flora. 

Over the next three days, 36 additional speakers gave presentations or workshops and an additional 20 posters 

were displayed at an evening reception. Presentations covered a variety of topics, ranging from seedling ecology and 

rare plant biology to distributional modeling, impacts of climate change, plant biogeography, and fire ecology. 

The conference concluded with a Friday field trip to Stansbury Island along the south side of the Great Salt Lake. 

Despite the unusually warm temperatures of mid-March, relatively few plants were flowering, though attendees were 

treated to a display of violet buttercup (Ranunculus andersonii var. andersonii) in bloom. 

3


All told, over 150 botanists attended the week-long conference. Much of the success of the conference could be 

attributed to the hard work of the planning and program committees, both chaired by Mindy Wheeler with the able 

assistance of Bill Gray, Ann Kelsey, Bill King, Therese and Larry Meyer, Robert and Susan Fitts, Loreen Allphin, 

Rita (Dodge) Reisor, and Leila Shultz. A number of volunteers from UNPS and Red Butte Garden helped with registration, 

food, and behind the scenes work, including Elise Erler, Tony Frates, Celeste Kennard, Kipp Lee, Bill Nelsen, 

Kody Wallace, Sue Budden, Pamela and Robert Hilbert, Allene Keller, Jena Lewinsohn, Marilyn Mead, and Bev 

Sudbury. Artist Laura Call Gastinger provided a beautiful painting of Dwarf bearclaw poppy (Arctomecon humilis) 

for the conference program and souvenir mug. The following corporate and institutional sponsors assisted financially 

or by other means: The Nature Conservancy of Utah, US Forest Service Rocky Mountain Research Station, University 

of Utah Department of Biology, the Flora of North America project, Providia, Utah Natural History Museum, 

Utah Botanical Center, Red Butte Garden and Arboretum, the state of Utah Department of Natural Resources, and 

Bio-West, Inc. 

Twenty presenters at the conference kindly prepared manuscripts for this inaugural issue of Calochortiana, which 

will serve as the official proceedings document for the conference. Thanks to all the contributors for their willingness 

to share, and for their patience. - Walter Fertig 

Abstracts of Presentations and Posters 

not Submitted for the Proceedings 

Biogeography of the Intermountain Region and 

Connections to the Southwestern USA 

Noel H. Holmgren, Curator Emeritus, New York Botanical 

Garden. 

Abstract: The Intermountain Region lies between the 

east base of the Sierra-Cascade mountain chain and the 

west side of the Rocky Mountains. Its southern boundary 

overlooks the warm deserts of the Southwest and the 

northern boundary lies along the base of the Oregon and 

Idaho batholiths. The Great Basin occupies nearly twothirds 

of the Region, and the Wasatch and Uinta Mountains 

and the Utah segment of the Colorado Plateau occupy 

the eastern third. In combination, the plant associations, 

vegetation zones, and plant species distinguish the 

region as a reasonably natural floristic unit, but there are 

many geologic-historical relationships with the Southwest. 

With the changing climate, even greater similarities 

may be anticipated in the future. The basin and 

range topography of the Great Basin offers a perfect 

place to monitor possible species migration from south 

to north and from valley to mountain. 

Flora of the Arizona Strip 

Duane Atwood, Brigham Young University, retired 

(1870) C.C. Parry (1874-1875); and later by A.L. Siler 

and Marcus E. Jones. Generally speaking, most Arizona 

botanists have given little attention to this area. The first 

concentrated effort of collecting on the Strip was by 

Ralph K. Gierisch, a retired Forest Service employee, 

who worked primarily as a volunteer for the BLM Arizona 

Strip District located in St. George, Utah. Ralph 

made hundreds of collections, which are deposited at 

that office with many duplicates at BYU and NAU. My 

interest and first collection from the Strip was made 27 

May 1968 1 mile north of Fredonia to secure the type 

for Phacelia constancei, while working on a revision of 

the crenulatae group of Phacelia (Hydrophyllaceae). 

Then later in 1970 while living in Fredonia and working 

on the Kaiparowits Environmental Impact Studies with 

BYU, thru 1975; and as the first botanist for BLM and 

the second one nationally for the Cedar City BLM District 

(1975-1977). Collection trips to this unique area 

continued through to the present, often with Larry C. 

Higgins. An annotated list of vascular plants has been 

generated for the entire Strip and the National Parks and 

Monuments within its borders. Six new endemic taxa 

have been described from the area: Phacelia higginsii, 

P. furnisii, P. hughesii, Camissonia dominguezescalantorum, 

Physaria arizonica var. andrusensis and 

Tetradymia canescens var. thorneae. 

Abstract: The "Arizona Strip" is a unique botanical 

area isolated from the rest of Arizona by the Colorado 

River. Our knowledge of its flora has been slow and 

incremental with a few collections from the early botanists 

who visited southern Utah such as Edward Palmer 

4


Biogeography and the Evolution of Rare, Endemic 

Species: Insights from two Mustard Genera in the 

Southwest (Draba and Boechera, Brassicaceae). 

Loreen Allphin, Department of Plant and Wildlife Sciences, 

Brigham Young University, Provo, UT and 

Michael D. Windham, Duke University, Durham, NC. 

Abstract: With a growing number of plant species in 

danger of extinction due to human induced threats, 

species-by-species approaches to management are becoming 

unrealistic. Conservation of rare plants would 

be improved by a clearer understanding of the evolutionary 

forces that give rise to rare, endemic species. 

These data might facilitate the development of management 

strategies applicable to a wide range of rare species. 

For this study we conducted a detailed survey of 

species within the genera Draba and Boechera from the 

Southwest United States (genera and a region with the 

high concentrations of endemic species). We collected 

data on geographic distribution, degree of endemism, 

chromosome number, ploidy level, breeding system, 

reproductive fitness and presumed mode of speciation. 

The study revealed some interesting evolutionary and 

biogeographic patterns. Some rare endemic species in 

these genera were primarily diploid, outcrossing, paleoendemic 

species with relatively low fecundity. Conversely, 

other endemic species in these groups were primarily 

polyploid, autogamous or apomictic, neoendemics 

with relatively high fecundity. These patterns appear 

to reflect both the type of speciation that occurred and 

the geologic/biogeographic history of the region. The 

geography of rarity and endemism in these genera appears 

to be an expression of primary divergence, reticulate 

evolution, and evolutionary time. 

Physical and Chemical Characteristics of Xeric Soils 

in Eastern Great Basin Determines the Natural Plant 

Associations, but Recently Ruderal Species have Become 

an Important Factor. 

Rodd Hardy, Bureau of Land Management, Salt Lake 

City, UT 

Abstract: What are major physical and chemical properties 

of the soil profiles that are key factors for different 

plant associations? What are the major natural plant 

associations based on these soil properties? What particular 

ruderal species, to what degree, and when did 

these invaders become a major role within plant associations 

today? What recommendations are needed to mitigate 

the impacts of ruderal species to natural plant communities? 

This paper will note the ecotone sharpness in 

which plant communities in dry climates change from 

one type to another for major plant species and the 

chemical properties of the soil which determine specific 

plant communities. Winterfat and gray molly sites have 

particularly been vulnerable to annual grass and goosefoot 

forbs, but invasive species effects upon endemic 

species such as Pohl’s milkvetch and Small spring parsley 

has also been notable. 

Predictive Habitat Models for Arctomecon californica 

Torrey & Frémont and Eriogonum corymbosum 

Bentham var. nilesii Reveal for the Upper Las Vegas 

Wash Conservation Transfer Area, Nevada. 

Amy A. Croft, Thomas C. Edwards, Jr., Janis L. Boettinger, 

Glen Busch, James A. MacMahon, US Geological 

Survey and the Ecology Center, Utah State University 

Abstract: The Upper Las Vegas Wash Conservation 

Transfer Area (ULVWCTA), situated northwest of 

Las Vegas, Nevada provides habitat for two of the 

state’s special status species, Arctomecon californica 

Torrey and Frémont and Eriogonum corymbosum Bentham 

var. nilesii Reveal. In an effort to aid the Bureau of 

Land Management Las Vegas Field Office in conservation 

based decision making, we built a family of statistical 

models capable of predicting likely locations of each 

species in the ULVWCTA. To predict locations of the 

plant species, emphasis was placed on sensitivity, the 

ability of the models to predict where the species were 

located. A. californica sites were characterized by soils 

with low shear and compressive strength values, a low 

percentage of rock, and a physical soil crust. The most 

common soil type and vegetation association occupied 

by A. californica was the Las Vegas type (spring deposits) 

and the Ambosia dumosa-Atriplex confertifolia 

vegetation association. Models for A. californica had 

moderate to excellent predictive capabilities, with accuracies 

as reflected by sensitivity ranging from 75% to 

95%. Small sample sizes precluded construction of any 

models for E. corymbosum var. nilesii. Instead, we were 

able to successfully predict likely locations of E. corymbosum 

var. nilesii with the A. californica models. Overall, 

the models had predictive capabilities of sufficient 

accuracy to be used in conservation decisions for the 

ULVWCTA. 

Comprehensive Interactive Plant Keys for the Southwest 

Bruce S. Barnes, Flora ID Northwest, Pendleton, OR 

Abstract: Plant conservation and management for any 

given locality is a complex process which depends on 

reliable and continually updated information regarding 

what species are found and where. These critical data 

5


often require time-consuming initial and ongoing plant 

surveys. Computerized interactive keys produced over 

the past 14 years by the author greatly facilitate plant 

surveys by reducing the time to key unknown species by 

90% or more. This presentation will demonstrate the use 

of the plant identification software, to provide the audience 

with an understanding of the potential applications 

of this resource. The keys include all known vascular 

plants, both native and introduced, which grow outside 

of cultivation in 16 states and 4 Canadian provinces, 

including California, Nevada and Utah, with Arizona 

and New Mexico to be added in 2010. Plant characteristics 

may be selected in any order, with no forced 

choices. Terms are defined and illustrated, extensive 

references are included, and color photos are provided 

for over 99% of species. Synonyms and menus of genera 

and families are provided to reduce problems of 

changing nomenclature. The software is continuously 

updated with name changes, new plant finds, and new 

photos, with free annual updates available for purchasers. 

The keys are available by state or larger region. In 

most cases descriptive information is provided for separating 

subtaxa when present. These keys are a powerful, 

innovative tool to assist in providing timely plant survey 

data for plant conservation and management. 

The Climate Puzzle of Global Warming: It is not just 

about Chilies! 

Robert R. Gillies, Director/State Climatologist, Utah 

Climate Center at Utah State University 

Abstract: In arid and semi-arid Western North America, 

observations of climate change point to an increase 

in average temperature that is greater than the rest of the 

world’s average. In line with such a warming trend in 

climate, several studies of the precipitation regime for 

the region have documented less snowfall as evidenced 

by decreases in snowpack as well as earlier snow melt, 

increased winter rain events and reduced summer flows. 

An ensemble of global climate model (GCM) projections 

for Western North America reflect just such conditions 

in that they suggest intensifying drying conditions 

to be the norm for the Southwest region due primarily to 

Hadley Cell intensification. Regions that lie to the 

Northwest, the GCMs have as benefiting from increased 

precipitation but in transitional zones, i.e., between the 

wetter and drier zones, any gains in projected precipitation 

are offset by the likelihood of an increased frequency 

of above normal temperatures during the summer 

months; such results suggest that an overall deficit 

in water resources is on the cards for much of the Intermountain 

West. 

Long-Term Perspectives on Vegetation: Paleoecology 

as a Tool for Conservation and Ecosystem Management 

Mitchell J. Power, Utah Museum of Natural History, 

Department of Geography, University of Utah 

Abstract: Long-term studies on vegetation history have 

demonstrated the role of climate in controlling the 

composition and distribution of species through time. 

Paleoecological studies that use fossil plant and pollen 

offer many lessons from the past, including: 1) plant 

species respond individualistically to climate change, 2) 

vegetation composition during the last Ice Age, 21,000 

years ago, was very different than today, and 3) plants 

have “migrated” across hundreds to thousands of kilometers 

since the last Ice Age in response to changing 

climate and disturbance regimes. These lessons from 

paleoecology can be used to inform conservation efforts 

towards the protection of plant species that face unprecedented 

climate change. Traditionally, most “longterm” 

conservation studies span less than 50 years and 

therefore characterize historical variations in plant communities 

within a limited temporal domain. Conservation 

efforts aim to restore natural habitats and protect 

landscapes, but the question remains; what to restore 

things to? Through merging paleoecological knowledge 

with conservation objectives, land managers and conservationists 

are better positioned to make informed decisions 

to protect plant species as we experience the rapidly 

changing climate of the 21st century. 

The Southwest Region ‘GAP’ Program for Mapping 

Vegetation and Species 

Doug Ramsey, Department of Geography and Earth Resources, 

Utah State University, Logan, UT 

Abstract: The Southwest Regional Gap Analysis Project 

(SWReGAP) is an update of the Gap Analysis 

Program’s mapping and assessment of biodiversity for 

the five-state region encompassing Arizona, Colorado, 

Nevada, New Mexico, and Utah. It is a multi-institutional 

cooperative effort coordinated by the U.S. Geological 

Survey Gap Analysis Program. The primary objective 

of the update is to use a coordinated mapping 

approach to create detailed, seamless GIS maps of land 

cover, all native terrestrial vertebrate species, land stewardship, 

and management status, and to analyze this information 

to identify those biotic elements that are underrepresented 

on lands managed for their long term 

conservation or are ‘gaps.’ 

6


Responses of Colorado Plateau Drylands to Climate 

Change: Variability due to Land Use and Soil- 

Geomorphic Heterogeneity 

Mark E. Miller and Jayne Belnap, U.S. Geological Survey, 

Southwest Biological Science Center, Moab, UT 

Abstract: Dryland ecosystems comprise well over 50 

percent of the Colorado Plateau province and are subjected 

to land uses such as livestock grazing, recreation, 

and energy development. Low and variable amounts of 

precipitation constrain dryland resilience to land-use 

activities, making drylands particularly susceptible to 

persistent changes in structure, function, and capacity 

for providing key ecosystem services such as soil stabilization. 

Through multiple effects on soil and vegetation 

attributes, land use also mediates ecosystem responses 

to climate. Ecosystem responses to interactive effects of 

land use and climate vary spatially in relation to soil 

geomorphic properties such as texture, depth, horizonation, 

and topographic setting due to effects of these 

properties on water and nutrient availability, soil erodibility, 

and site susceptibility to hydrologic alteration by 

soil-surface disturbances. We use existing data from 

Colorado Plateau drylands to illustrate these concepts 

and to develop a set of testable hypotheses about climate-land-use 

interactions (i.e., how climate and land use 

each affect ecosystem resilience to the other) in relation 

to soil-geomorphic properties. For example, we predict 

that climate-land-use interactions in Colorado Plateau 

drylands will be greater on deep soils than on shallow, 

rocky soils because the former support grasslands and 

shrub steppe ecosystems that have been most extensively 

used and modified by livestock grazing. We also 

predict that climate-land-use interactions will be greater 

on relatively fine-textured soils than on coarse-textured 

soils because the former tend to be more susceptible to 

exotic plant invasions and hydrologic alteration following 

disturbance, and because they exhibit greater fluctuations 

in resource availability in response to precipitation 

variability. Variable ecosystem responses to climate 

due to land use and soil have implications for scientists’ 

efforts to predict ecological consequences of climate 

change with sufficient detail to inform management decisions, 

and for decision makers’ efforts to prioritize and 

evaluate risks of different management strategies. 

Colorado Rare Plant Conservation Initiative, Saving 

Colorado’s Wildflowers 

tage by improving the stewardship of Colorado’s most 

imperiled plants. One hundred thirteen native plant species 

in Colorado are considered imperiled or critically 

imperiled by the Colorado Natural Heritage Program, 

meaning they are at significant risk of extinction. Of 

these species, 63 are endemic, growing only in Colorado 

and no place else in the world. Nearly 50% of our 

state’s imperiled native plants are considered poorly or 

weakly conserved. Unlike animals, Colorado has no 

state-level recognition or protection for plants. Impacts 

to Colorado’s rare plants are at an all-time high due to 

our rapidly expanding human population. Primary 

threats include habitat loss and fragmentation associated 

with resource extraction, motorized recreation, housing 

and urban development, and roads. Many rare plants are 

also at risk due to a simple lack of awareness regarding 

their precarious status. Despite the size and scale of 

these threats, we still have a chance to make a difference 

through strategic conservation actions, since healthy 

populations of many imperiled plants still exist. The 

goal of the Rare Plant Conservation Initiative is to conserve 

Colorado’s most imperiled native plants and their 

habitats through collaborative partnerships for the preservation 

of our natural heritage and the benefit of future 

generations. 

Rare Plant Management and BLM Policy 

Carol Spurrier, Bureau of Land Management, Washington, 

DC. 

Abstract: Rare plant conservation continues to be part 

of the multiple use mission of the Bureau of Land Management 

(BLM) in the United States. With continuing 

increases in the demand for all types of energy and other 

goods provided by the public lands, as well as landscape 

scale changes in natural vegetation due to increased 

wildfire and climate change, we wondered if the public 

lands that have been designated as part of the Natural 

Landscape Conservation System (NLCS) hold significance 

for protection of rare plant resources in BLM. We 

examined 2006 occurrence data on BLM lands from 

NatureServe within NLCS unit boundaries to determine 

rare plant species occurring in each unit. In this paper 

we discuss our findings for the different types of 

designations (wilderness, wilderness study areas, National 

Monuments and National Conservation Areas) 

within the System. 

Brian Kurzel, Colorado Natural Areas Program 

Abstract: The Colorado Rare Plant Conservation Initiative 

is a diverse partnership of public and private organizations 

dedicated to conserving our state’s natural heri- 

7


Thinking and Acting Together to Preserve Uinta 

Basin Rare Plants 

Joan Degiorgio, Northern Mountains Regional Director, 

The Nature Conservancy, Salt Lake City, UT 

Abstract: Utah’s Uinta Basin is home to dozens of endemic 

plant species. Nine of those species are at risk 

due to increased levels of energy development. Representatives 

from state, federal and local agencies, the Ute 

Tribe, private consultants, conservation groups, researchers, 

industry and others have come together as the 

Uinta Basin Rare Plant Forum to collectively address 

threats to these plants. Through a transparent, open 

planning process key ecological attributes of each plant 

were identified and rated for viability, stresses identified 

and specific strategies developed. This planning effort 

has been an excellent tool to “capture” the collective 

wisdom of local experts; and, with this knowledge, develop 

comprehensive strategies for the nine species simultaneously. 

With an agreed upon strategy, the Forum 

will work together implementing the highest leverage 

strategies and engaging more partners and dollars because 

of the solid foundation built through this planning 

process. 

Interagency Management Agreement for TES 

Species in Central Utah 

David Tait, botanist, Fishlake National Forest, Richfield, 

UT. 

Abstract: The Bureau of Land Management Richfield 

Field Office, Fishlake National Forest, and Capitol Reef 

National Park share management responsibilities for 

many of the same Threatened, Endangered & Sensitive 

plant species (TES). To enable each of these agencies to 

better manage their shared TES species, they developed 

an interagency agreement in 1999 that enables them to 

pool their funding. This funding, which has been minimal 

at times, has been used to employ an interagency 

botanist and hire seasonals to survey and monitor these 

TES species throughout their ranges, regardless of 

agency boundaries. The BLM Price Field Office and the 

US Fish and Wildlife Service were added in 2007. This 

project has allowed us to: (1) conduct intensive surveys 

for target species on potential habitat within the project 

area, (2) determine potential for impact by visitor, recreational 

or livestock use on long-term viability of these 

rare plants, and (3) conduct long-term monitoring on 

several of the rare species. Between 1999 and 2008 approximately 

100,580 acres were surveyed for some 30 

TES plants species by the IA team. approximately 

27,500 acres on BLM, 37,920 acres on Capitol Reef, 

and 35,160 acres on lands administered by the Fishlake. 

Conservation Success for a Rare Idaho Endemic: 

Conservation Agreement and Botanical Special Interest 

Area for Christ’s Paintbrush 

Kim Pierson-Motychak, Sawtooth National Forest, 

Twin Falls, ID and Jeffrey E. Motychak, Motychak Environmental 

Consulting, Twin Falls, ID. 

Abstract: Christ’s Indian Paintbrush (Castilleja christii) 

is a rare species known from only one population in 

Southwestern Idaho, Cassia County. Due to its restricted 

distribution and vulnerability to threats, C. christii is 

designated as a Candidate species for Federal listing 

under the Endangered Species Act (ESA). Conservation 

Strategies were completed in 1995 and 2002 for the establishment 

of long-term monitoring protocols. These 

have been implemented from 1995 to present. In 2003, 

the portion of the population not included in the Mount 

Harrison Research Natural Area (RNA) was designated 

as the Mount Harrison Botanical Special Interest Area 

(BSIA). In 2005, a ten year Candidate Conservation 

Agreement (CCA) was signed between the US Fish and 

Wildlife Service and the USDA Forest Service. This 

CCA identified the key threats to the population which 

included: 1) non-native plant introduction and establishment, 

2) recreational impacts, 3) hybridization, 4) unauthorized 

livestock impacts, 5) road construction, maintenance, 

and facilities, and 6) natural threats. A total of 42 

conservation action items were committed to in the 

CCA. Results from the implementation of these conservation 

action items include aggressive non-native plant 

treatment, increased interpretive education, agency and 

public interaction, long-term demographic and reproductive 

monitoring, host-specificity determination, and 

preliminary pollination ecology. Population trends indicate 

that while plant densities within the communities 

have declined over the 13-year period, individual reproductive 

output (flowering stems/plant) has increased. 

Modeling Distributions of Rare Plants in the Southern 

Great Basin of Utah 

Marti Aitken, Utah State University and US Forest Service 

Pacific Northwest Research Station, Portland, OR; 

Leila M. Shultz, College of Natural Resources, Utah 

State University, Logan, UT; and David W. Roberts, 

Department of Ecology, Montana State University, 

Bozeman, MT. 

Abstract: Field-validated landscape level predictive 

models identify potential plant habitat for rare plants in 

the Great Basin of western North America. Four rare 

species (Jamesia tetrapetala, Penstemon nanus, Primula 

domensis, and Sphaeralcea caespitosa) endemic to the 

southern portion of the eastern Great Basin (SW Utah) 

8


were chosen to include a range of environmental variability, 

growth form, and plant communities. Herbarium 

records of known occurrences were used to identify initial 

sample sites. We established multiple field sites to 

determine the geographic coordinates, environmental 

attributes (slope, aspect, soils, parent material) and 

vegetation data (associated species) in order to develop 

two predictive models for each species: a field key and a 

probability-of-occurrence or predictor map. The field 

key was developed from environmental attributes and 

associated species data collected at the sites and used 

only field data. Predictive maps were developed with a 

geographic information system (GIS) containing slope, 

elevation, aspect, soils, and geologic data – then randomly 

tested. Classification-tree (CT) software was 

used to generate dichotomous field keys and the maps of 

occurrence probabilities. Predictions from both models 

were randomly field-validated during the second phase 

of the study, and final models were developed through 

an iterative process. Data collected during the field validation 

were then incorporated into subsequent predictive 

models. The models identified potential habitat by 

combining elevation, slope, aspect, rock type, and geologic 

process into habitat models for each species. The 

cross-validated models were >96% accurate and generally 

predicted presence with accuracy >60%. 

Ute Ladies’-Tresses in the Diamond Fork Watershed: 

An Update 

Bridget M. Atkin and Steve R. Ripple, BIO-WEST, 

Logan, Utah 

Abstract: Ute ladies’-tresses (Spiranthes diluvialis) 

(ULT) was listed as threatened in 1992. The largest 

known population is in the watershed of Diamond Fork 

Creek and its tributary, Sixth Water Creek. Between 

1916 and 2004, these streams were used as canals, and 

they conveyed irrigation water diverted from Strawberry 

Reservoir to the Wasatch Front. Increased peak flows 

altered the stream channel and aquatic ecosystem, creating 

unique conditions that allowed the rare orchid to 

thrive. In 2004 a system of pipes was installed to divert 

water directly into the Spanish Fork River, thereby reducing 

the flows in Diamond Fork and Sixth Water 

Creeks. Studies of ULT populations have been conducted 

since 1992 under the direction of Utah Reclamation 

Mitigation and Conservation Commission. Results 

show that ULT colonies are still maintaining large 

numbers. However, monitoring of ULT has been difficult. 

The unique life-cycle characteristics of ULT, along 

with its dynamic habitat, create many challenges. 

Highly variable yearly ULT counts are very difficult to 

interpret or correlate with environmental parameters. In 

2005 other studies were initiated and more associated 

plant species data are now being systematically collected 

to track changes that may indicate whether the 

decreased flows are impacting ULT habitat. During 

2007 ULT numbers showed at least two flushes, in early 

August with tiny individual plants. Conversely, in 2008 

ULT numbers were highest in mid-September and 

plants were large. These observations have wide reaching 

implications pertinent to many species, indicating 

that unless a population is observed carefully, data could 

easily be misinterpreted. 

Arizona Cliffrose (Purshia subintegra), An Arizona 

Endemic 

Debra Crisp, Coconino National Forest, Flagstaff, AZ, 

and Barbara G. Phillips, Zone Botanist, Coconino, Kaibab 

and Prescott National Forests 

Abstract: The Arizona cliffrose is a long-lived shrub, 

endemic to white Tertiary (Miocene and Pliocene) 

limestone lakebed deposits that are high in lithium, nitrates, 

and magnesium and is an Endangered species. It 

occurs in four disjunct populations spread across an area 

of approximately 200 miles in central Arizona. Threats 

to Arizona cliffrose include livestock grazing, mineral 

exploration, road and utility corridor development, offhighway 

vehicle use, urban development and drought. 

In this poster we summarize the results of some longterm 

monitoring transects initiated in 1987. These transects 

are in the Cottonwood population, which is 

thought to be the healthiest and contains the most diverse 

age structure of the four known populations. Data 

on these transects were collected three times, in 1987, 

1996 and in 2008. 

Demography and Pollination Biology of Graham's 

Penstemon (Penstemon grahamii), a Uinta Basin Endemic; 

5-year results. 

Rita [Dodge] Reisor and Wendy Yates, Red Butte Garden 

and Arboretum, University of Utah, Salt Lake City, 

Utah 

Abstract: Penstemon grahamii is a Uinta Basin endemic 

which grows on oil-shale outcrops of the Green 

River Formation. Long-term monitoring plots were established 

for P. grahamii to collect basic life history 

data, study pollination biology, and survey critical habitat. 

Research was conducted over 5 years (2004 to 2008) 

during May – June, at the Blue Knoll/Seep Ridge and 

Buck Canyon population sites located on BLM land. 

Data gathered includes rosette diameter, number of inflorescences, 

inflorescence height, flowers per inflorescence, 

number of fruiting individuals, and herbivory. 

The breeding systems study used the following treat- 

9


ments: autogamy, geitonogamy, xenogamy, and vector 

pollination as a control group. Surveys were based on 

historic Element Occurrence (EO) reports and surrounding 

habitat. Demographic data suggest that P. grahamii 

population size has remained fairly stable over the study 

period. Annual survivorship rates range from 47 – 82%, 

and mortality ranging from 6 – 36%. Flowering events 

are highly variable annually, ranging from zero flowering 

plants in 2006, up to 44% in 2004. As expected for 

the breeding systems results, the vector (control) produced 

the most fruits, then xenogamy, geitonogamy, 

and autogamy with the least fruits produced. Survey 

results found existing populations at each historic EO 

visited, and expanded the range and size for some occurrences. 

Current threats include high rates of herbivory, 

habitat loss, and fragmentation due to oil and gas 

development. It is unknown how the reproductive success 

of P. grahamii may be influenced by other development 

related impacts such as dust production and pollinator 

disturbance. 

Micropropagation Studies in Astragalus holmgreniorum 

Aaron R. Fry, Brett A. McGowan, and Julianne 

Babaoka, Ally Bench, Renée Van Buren and Olga R. 

Kopp, Department of Biology, Utah Valley University, 

Orem, Utah, 84058 

Abstract: Astragalus holmgreniorum, a species endemic 

to the northern areas of the Mojave Desert is 

listed as a federally Endangered species. Threats to the 

species stem from habitat destruction arising primarily 

from commercial and residential development, overgrazing 

by livestock, recreational vehicles, and mining 

operations. In an attempt to develop a micropropagation 

technique aimed at aiding in recovery efforts for the 

species, we report successful induction of shoots from 

callus tissue. Explants were taken from leaves (abaxial 

and adaxial surfaces) and from petioles. These were incubated 

in MS medium amended with 2,-4 D and BA to 

induce callus formation. Murashige and Skoog medium 

amended with 7 mg/L of 2,-4 D and 2 mg/L of BA induced 

the formation of embryos and plantlets. Current 

work focuses on the effects of varying concentrations of 

NAA, IBA, and IAA on root formation. Following root 

induction, we plan to acclimatize plantlets by incubating 

them in potting soil. Ultimately, we hope that this research 

may be used to aid in recovery efforts of this species. 

Geochemical Analysis of Tuffaceous Outcrops Associated 

with the Narrow Endemic, Penstemon idahoensis 

Welsh & Atwood 

Paul R. Grossl, William A. Varga, and Richard M. 

Anderson, Utah State University, Plant, Soils, and Climate 

Department and Utah Botanical Center, Utah State 

University. 

Abstract: Idaho penstemon (Penstemon idahoensis 

Welsh & Atwood) is a Sawtooth National Forest Sensitive 

plant species narrowly endemic to the Goose Creek 

drainage in northern Box Elder County, Utah, and adjacent 

southern Cassia County, Idaho. Idaho penstemon is 

a short, glandular, perennial forb comprised of several 

stems which emerge from a semi-woody caudex and 

topped with showy, blue flowers. Its distribution is 

restricted to dry, light-colored, sparsely vegetated, tuffaceous 

outcrops of Tertiary Salt Lake Formation 

sediments. Botanists have long recognized the association 

of endemic, often rare, plant species with unusual 

soils. Edaphic endemism is prominent among those 

plant associations which include ultramafic bedrock, 

such as serpentine, or calcareous bedrock such as limestone, 

chalk, dolomite, or gypsum. Edaphic endemic 

species frequently arouse conservation concern due to 

their restricted distributions or small plant population 

sizes. The focus of this investigation considered the 

question whether Penstemon idahoensis utilizes unusual 

soil conditions that exclude other taxa and thus provide 

low competition environments. To answer this question 

we attempted to ascertain via geochemical analysis any 

selective or restrictive constituents or composition including 

the presence or absence of gypsum, unusual soil 

pH levels, soil texture, salinity (EC), organic matter, or 

atypical element distributions associated with common 

soil components. 

What's Happened to Siler Pincushion Cactus? 

Lee E Hughes, Ecologist, Arizona Strip Field Office, St. 

George, UT, retired 

Abstract: The Siler Pincushion Cactus has, like all 

vegetation in the southwest, been under the influence of 

a ten year drought. The effects of this drought are evident 

in the data gathered on the cactus. The poster will 

show the data from the six demographic plots on the 

cactus. The data starts in 1986 and goes through to 

2008. It summarizes the mortality data. The size structure 

for each plot is shown graphically to demonstrate 

the affect of the drought on the size composition of the 

cactus. In summary the drought has reduced the small 

cactus significantly. Also shown, is the effect (or no 

effect) from livestock and ATVs being present in the 

10


plots. The plot data shows a cactus with drought problems, 

but in some areas is doing well also. 

Clark County (Nevada) Rare Plant Modeling and 

Inventory 

Sonja R. Kokos, Clark County Desert Conservation Program, 

Las Vegas, NV; David W. Brickey and Larry R. 

Tinney, TerraSpectra Geomatics, Las Vegas, NV; and 

Analie R. Barnett and Robert D. Sutter, The Nature 

Conservancy, Southeastern Region, Durham, NC 

Abstract: To understand the distribution of rare plants 

covered under the Clark County Multiple Species Habitat 

Conservation Plan, Clark County and Terra-Spectra 

Geomatics developed two predictive GIS models. The 

models used ASTER Imagery and Landsat ETM+ Imagery, 

soils data (NRCS SSURGO), geologic data, and 

presence/absence data for eight rare and endemic plant 

species. The first model was used to predict the distribution 

of three gypsum loving species, the Las Vegas 

bearpoppy (Arctomecon californica), Sticky ringstem 

(Anulocaulis leiosolenus var. leiosolenus), and Las 

Vegas buckwheat (Eriogonum corymbosum var. nilesii). 

The second model was used to predict the distribution of 

five sand or potentially sand loving species, the Threecorner 

milkvetch (Astragalus geyeri var. triquetrus), 

Pahrump Valley buckwheat (Eriogonom bifurcatum), 

Sticky buckwheat (Eriogonum viscidulum), Beaver Dam 

breadroot (Pediomelum castoreum), and Whitemargined 

beardtongue (Penstemon albomarginatus). 

Using these models, Clark County can now describe the 

occurrence of all eight species in terms of high, medium 

and low probabilities. During the 2009 and 2010 field 

seasons, the county will test both models using a sampling 

protocol developed jointly by Clark County and 

The Nature Conservancy. Intuitively, we expect these 

models to predict the distribution of some species better 

than others, and further model refinement will be 

needed. However, the results to date have produced 

some interesting hypotheses regarding the life history 

and biology of these species. We expect the results will 

be valuable to Clark County and the federal land management 

agencies charged with managing these species. 

Post-Fire Monitoring of Erosion Resistance and Dust 

Emission on the Milford Flat Fire, West-Central 


Mark E. Miller, National Park Service, Moab, UT 

(formerly U.S. Geological Survey, Southwest Biological 

Science Center, Kanab, UT) 

Abstract: Soil stabilization is a major objective of postfire 

emergency stabilization and rehabilitation (ES&R) 

projects, yet monitoring data are rarely sufficient to determine 

whether treatments effectively achieve this objective. 

To address this information need, the U.S. Geological 

Survey and Bureau of Land Management are 

collaboratively monitoring effects of ES&R treatments 

on soil-surface stability and rates of dust emission in 

low-elevation portions of the 147,000-ha Milford Flat 

Fire that occurred in west-central Utah in July 2007. In 

August 2008, monitoring plots were established to 

evaluate the effectiveness of three types of ES&R treatments 

(aerial seeding and chaining, seeding with a 

rangeland drill, and seeding with a rangeland drill after 

herbicide application) in areas where field observations 

and satellite imagery indicated high rates of dust emission 

during spring 2008. Monitoring attributes include 

indicators of erosion resistance (soil stability, ground 

cover, and sizes of gaps between plant canopies) in addition 

to measures of plant cover and community composition. 

Seasonal rates of dust emission are currently 

monitored with BSNE dust samplers. Sampling in August 

2008 indicated that average soil-surface stability 

was highest in unburned control plots and in burned 

plots that were not treated. Average soil stability was 

lowest in burned plots that were seeded with a rangeland 

drill following herbicide application. During the August-October 

2008 period, rates of wind-driven soil 

movement varied over three orders of magnitude and 

were greatest in plots that received ESR treatments, 

were in exposed landscape settings, and had soils that 

were most susceptible to wind erosion. 

Using GIS and Remote Sensing to Predict Dominant 

Plant Species Distributions in Rich County, Utah 

Kate Peterson, Doug Ramsey, Leila Shultz, John Lowry, 

Alexander Hernandez, and Lisa Langs-Stoner, Remote 

Sensing/GIS Laboratory and Floristics Lab, Dept. of 

Wildland Resources, Utah State University, Logan, UT 

Abstract: This research shows models of the potential 

spatial distribution of key upland plant species in Rich 

County, Utah. We used geospatial data layers of abiotic 

factors and remotely sensed (RS) imagery in conjunction 

with field-collected vegetation data. Plant species 

distribution maps are used to objectively and costeffectively 

correlate soil maps units with GIS data in the 

production of Ecological Site Descriptions (ESD’s). 

These were produced for Rich County in accordance 

with NRCS (Natural Resources Conservation Service) 

standards. Inasmuch as abiotic factors and vegetation 

associations can be used to predict the potential distribution 

of rare plants, we believe these analyses can be 

used to guide field searches for populations of endemic 

species. 

11


A Newly Discovered Gutierrezia on the Colorado 

Plateau 

Al Schneider, www.swcoloradowildflowers.com and 

Peggy Lyon, Colorado Natural Heritage Program 

Abstract: The authors will present information on the 

newly discovered species, Gutierrezia elegans or 

Lone Mesa Snakeweed. They discovered G. elegans 

August 4, 2008 while doing a plant survey in the new 

Lone Mesa State Park, 30 miles north of Dolores, Colorado. 

See http://www.swcoloradowildflowers.com/ 

Yellow%20Enlarged%20Photo%20Pages/gutierrezia% 

20elegans.htm for details, including the full description 

published in the December, 2008 issue of the Journal of 

the Botanical Research Institute of Texas. 

Incorporting Demography, Genetics, and Cytology 

into Long-Term Management Plans for a Rare, 

Endemic Alpine Species: Draba asterophora 

Emily Smith and Loreen Allphin, Department of Plant 

and Wildlife Sciences, Brigham Young University, 

Provo, UT. 

Abstract: Draba asterophora (Brassicaceae), a rare and 

endemic mustard, is known from three population clusters 

(North, Southeast, and south) occupying a narrow 

range of alpine habitats surrounding Lake Tahoe. The 

southern population cluster has been segregated as variety 

macrocarpa, whereas the other two clusters have 

been assigned to variety asterophora. Because this 

small, matted, perennial occurs at alpine sites, the species 

faces impending threats to its habitat through ski 

run expansion and development as well as from global 

climate change. With funding from the USDA Forest 

Service and local ski resorts, we are conducting morphological, 

ecological, chromosomal, and genetic studies 

of both varieties of D. asterophora to provide a 

framework upon which future management plans and 

mitigation can be developed. Preliminary results suggest 

that there are significant differences between the three 

population clusters. These include differences in soil 

composition, soil chemistry, plant density, demographics 

reproductive success, and genetics. Chromosome 

counts from the northern populations (Mt. Rose, Nevada) 

are tetraploid (n=20). Allozyme banding patterns 

support the hypothesis that these have arisen through 

autopolyploidy. The southeastern population has shown 

both diploid and triploid counts. Because the species 

includes more than one ploidy level, it should not be 

treated as a single panmictic taxon for purposes of conservation. 

Endangered Milkvetches of Washington County, 


Wendy Yates, Red Butte Garden and Arboretum, University 

of Utah, Salt Lake City, UT; and Ally Bench 

Searle and Renee Van Buren, Utah Valley State University, 

Orem, UT. 

Abstract: Astragalus ampullarioides (Welsh) Welsh 

and A. holmgreniorum Barneby are two federally listed 

Endangered species endemic to Washington County, 

Utah. A. ampullarioides is known from only four known 

populations, and A. holmgreniorum from only three 

populations. Prior research conducted on these species 

by Van Buren and Harper (2003) focused on vegetative 

and demographic characteristics. There have been no 

prior studies on the soil seed bank or seed viability. This 

study focused on determining the density of the soil 

seed bank and the percent seed viability for both species. 

For Astragalus ampullarioides soil was removed 

from two of the four known populations. For Astragalus 

holmgreniorum soil was removed from ten densely 

populated areas. Seeds were sifted from the soil and collected. 

The seeds extracted were then tested for viability 

by allowing them to germinate. Those seeds that did not 

germinate were further tested using the Tetrazolium test. 

This study found Astragalus ampullarioides had a soil 

seed bank density of 50 seeds/m 2 soil and viability was 

68.2%. Astragalus holmgreniorum had a soil seed bank 

density of 1.8 seeds/m 2 soil and viability was 87.7%. 

Collaborative Conservation for Washington County, 

Utah’s Federally-Listed Plants 

Elaine York and Gen Green, The Nature Conservancy, 

Salt Lake City, UT. 

Abstract: Washington County, Utah is home to fourfederally 

listed plants: the Dwarf bear poppy (Arctomecon 

humilis), Siler pincushion cactus (Pediocactus 

sileri), Holmgren milkvetch (Astragalus holmgreniorum) 

and Shivwits milkvetch (Astragalus ampullarioides). 

Each faces pressing threats, especially habitat 

loss and degradation from urban development, invasive 

plants, and off-road vehicle use. Through U.S. Fish 

and Wildlife (USFWS) coordination and the efforts of 

partners, many conservation actions have been completed 

including land acquisition, habitat fencing, habitat 

restoration, the establishment of Areas of Critical 

Environmental Concern, seed germination and pollinator 

research, community education efforts, a habitat 

management endowment and more. Conservation actions 

have been implemented by USFWS, Bureau of 

Land Management, Dr. Renée Van Buren, Dr. Kimball 

Harper, Dr. Susan Meyer, U.S. Geological Survey, 

12


Washington County, Red Cliffs Desert Reserve, the 

Shivwits Band of the Paiute Tribe, Zion National Park, 

Utah Native Plant Society, Utah Natural Heritage Program, 

School and Institutional Trust Lands Administration, 

The Nature Conservancy, and more. 

Spatial Landscape Modeling: The Land Manager’s 

Tool Box 

Elaine York and Louis Provencher, The Nature Conservancy, 

Salt Lake City, UT. 

Abstract: Facilitated by The Nature Conservancy, the 

Spatial Landscape Modeling Project quantitatively modeled 

the reference and current conditions for seventeen 

major vegetation types in the Grouse Creek Mountains 

and Raft River Mountains, a 1.1 million acre landscape 

in northwest Utah. Partners – including Utah Partners 

for Conservation and Development, Bureau of Land 

Management, Sawtooth National Forest, Utah Division 

of Wildlife Resources, National Resources Conservation 

Service, Quality Resource Management, and private 

landowners – shared management data to explore effectiveness 

of current management and developed computer-generated 

alternative management scenarios to 

consider options for optimal land health. Cutting-edge 

technology from remote sensing, GIS analysis and partner-informed 

computer models produced a number of 

tools to assist land managers in their understanding of 

large-scale vegetation dynamics, long-term management 

options and the importance of management cooperation 

across land-ownership borders. 

An Update on Ecological Investigations of the 

Shivwits Milk-Vetch (Astragalus ampullarioides), 

Washington County, Utah 

Mark E. Miller and Rebecca K. Mann, formerly US 

Geological Survey, Southwest Biological Science Center, 

Kanab, UT; Rebecca Lieberg, Cheryl Decker and, 

Kathy Davidson, Zion National Park, and Harland Goldstein 

and James D. Yount, U.S. Geological Survey, 

Earth Surface Processes Team, Denver, CO 

Abstract: The Shivwits milk-vetch (Astragalus ampullarioides) 

is one of four federally protected plant species 

restricted to particular geologic substrates at the edge of 

the Colorado Plateau and Mojave Desert in Washington 

County, Utah. Since 2006, the U.S. Geological Survey 

and National Park Service (Zion National Park, ZNP) 

have been studying this species in relation to geology 

and soils, herbivory, exotic plants, and mycorrhizal 

fungi. Habitat studies in 2006 documented the species 

on a new geologic substrate and across a broad range of 

soils, expanding the concept of potential habitat. Consumption 

of inflorescences by native herbivores reduced 

reproductive output in a ZNP subpopulation by 90% in 

2006 (low production year) and 75% in 2008 (high production 

year). Preliminary analyses indicate no significant 

effects of exotic red brome (Bromus rubens) biomass 

on growth or reproductive output of established 

milk-vetch plants in the same subpopulation during 

spring 2008. Effects of brome biomass on seedling recruitment 

remain unclear because low precipitation in 

2006 and 2007 prevented seed collection required for 

experimental studies. In 2007, median soil seed bank 

density in plots at ZNP was 45.7 seeds m 2 , with an extremely 

high density (2741 seeds m 2 ) in the plot with the 

sandiest soil. Coarse textured soils in this plot may reduce 

germination frequency, thereby resulting in longterm 

seed accumulation. Overall, results to date indicate 

that caging to exclude native herbivores may be the least 

expensive way to improve the viability of extant populations 

by enhancing reproductive output. 

Population Genetic Structure of an Endangered 

Utah Endemic Astragalus ampullarioides (Welsh) 

Welsh (Fabaceae) 

Jesse W. Breinholt, Utah Valley University, Orem, UT 

and Brigham Young University, Provo, UT; and Renee 

Van Buren, Olga R. Kopp, and Catherine L. Stephen, 

Utah Valley University, Orem, UT 

Abstract: The Shivwits milkvetch, Astragalus ampullarioides 

(Welsh) Welsh, is a perennial herbaceous plant 

in the family Fabaceae. This Utah edaphic endemic was 

federally listed as Endangered in 2001 because of high 

habitat specificity and low numbers of individuals and 

populations. All known occupied habitat for A. ampullarioides 

was designated as critical habitat by the US 

Fish and Wildlife Service in 2006. We used AFLP 

markers to assess genetic differentiation among the 

seven extant populations and quantify genetic diversity 

in each. Six different AFLP markers resulted in 217 unambiguous 

polymorphic loci. We used multiple methods 

to examine how population genetic structure in this species 

has changed over time. The genetic data indicate 

that, relatively recently, A. ampullarioides consisted of a 

single large contiguous genetic unit that fragmented 

over time into 3 genetic regions. These regions further 

fragmented and extant populations have differentiated 

through genetic drift. Populations exhibit low levels of 

gene flow, even between geographically close populations. 

We suggest plans for population establishment or 

augmentation carefully consider the genetic makeup of 

each of the extant populations. 

13


Summary of Sclerocactus Monitoring in the Uinta 

Basin 

Maria Ulloa, formerly Bureau of Land Management, 

Richfield, UT 

Abstract: Sclerocactus brevispinus is a small barrel 

cactus endemic to the Pariette Draw and S. wetlandicus 

is a larger barrel cactus endemic to the Green River 

benches. Both geographic locations are in the Uinta Basin. 

These species were listed as Threatened with Sclerocactus 

glaucus. Currently the species are under review 

by USDI-FWS; the agency is working on its final 

ruling to make the taxonomic changes to separate these 

3 species. In May of 1997, 37 monitoring plots of S. 

brevispinus and S. wetlandicus were established in the 

Uinta Basin, including 9 plots with transplanted individuals. 

Ecosphere Environmental Service (Ecosphere) 

entered a Cooperative Agreement with the Vernal Field 

Office of the Bureau of Land Management (BLM) to 

study the genus Sclerocactus in the Pariette Drainage. 

All cacti within a 15m-radius of the plot center point 

were mapped and tagged. In 1998, the plots were read 

by Ecosphere. Funding to continue the monitoring was 

not allocated. During the winter of 2004, BLM decided 

to relocate the plots for the 2005 field season to see if 

the tagged cacti could be found. The BLM was successful 

at relocating the plots and the tagged cacti and decided 

to continue the monitoring for 4 years. In addition 

to finding the fate of the tagged cacti, all new individuals 

have been mapped and tagged. Other information 

collected has been number of flowers, number of capsules, 

and a small sampling of how many seeds per capsule. 

During the field season of 2008, a random sampling 

of the distance of cacti from the center of ant’s 

nests was measured to determine if ants influenced distribution 

and dispersal of seeds. 

Demographics of Sclerocactus Species in the Uintah 

Basin 

Lynda Sperry, SWCA Inc., Salt Lake City, UT 

Abstract: Botanical surveys associated with oil and gas 

development in the Uintah Basin provide large databases 

for listed plant species in compliance with the Endangered 

Species Act. We have extensively surveyed 

two threatened cacti species over the past three years. A 

total of 8,793 individuals were identified in 2008, 4,780 

were Sclerocactus wetlandicus, 3,663 were S. brevispinus, 

and 350 were identified as possible hybrids. We 

found the greatest concentration of S. brevispinus on 

north facing slopes (20%), followed by flat surfaces 

(15%), and the least amount on east facing slopes (6%). 

The aspect was not as significant for S. wetlandicus with 

18%, 17%, and 16% on north, south, and flat surfaces 

respectively. The aspect with the least number of S. wetlandicus 

was northwest-facing with 6% of the individuals 

surveyed. Both species were found more often in 

communities dominated by Atriplex, including A. confertifolia, 

A. canescens, or A. corrugata. Using the 

SSURGO data layer, S. wetlandicus was found more 

frequently on Motto-Rock outcrop complex, whereas S. 

brevispinus was more frequently found on Badland- 

Rock outcrop complex. Both soil types are dominated 

by clay and have high salinity ratings. Utilizing survey 

data collected as part of the permitting process for oil 

and gas development provides a unique opportunity to 

gain basic ecological and demographic information for 

federally listed species. 

Breeding System Characterization of a Threatened, 

Cliff Dwelling, Narrow Endemic Primula 

Jacob B. Davidson and Paul G. Wolf, Biology Department, 

Utah State University, Logan, UT. 

Abstract: The maguire primrose (Primula cusickiana 

var. maguirei or Primula maguirei) is a Threatened cliff 

dwelling endemic plant found only in northern Utah’s 

Logan Canyon. Although a small number of populations 

are close to one another, these populations have highly 

differentiated genetic structure. Maguire primrose, like 

most species in the genus Primula, exhibits reciprocal 

herkogamy in its morphology. Pin morphs have stigmas 

that extend to the opening of the corolla tube and anthers 

found near the bottom of the tube. Thrum morphs 

have stigmas found near the bottom of the corolla tube 

and anthers that are near the mouth of the corolla tube. 

Despite the spatial separation of anthers and stigmas, 

self fertilization has been observed for a number of Primula 

species. In the spring of 2008, I made hand pollinations 

using intramorph illegitimate outcrossings, legitimate 

intermorph outcrossings, and various autogamy 

and geitonogamy tests, while excluding pollinators. Resulting 

seed set was examined from each treatment. Several 

temperature and relative humidity monitors were 

placed near the plants, to see if environmental conditions 

affected hand pollination success. Temperature 

fluctuations at each study site ranged widely between 

freezing temperatures and 20ºC. We share our preliminary 

results from this initial field season here. 

14


The Taxonomic Distinctness of Eriogonum corymbosum 

var. nilesii 

Mark Ellis, Biology Department, Utah State University, 

Logan, UT. 

Abstract: We examined populations of perennial, 

shrubby buckwheats in the Eriogonum corymbosum 

complex and related Eriogonum species in the subgenus 

Eucycla, to assess genetic affiliations of the recently 

named variety E. corymbosum var. nilesii. We compared 

AFLP profiles and chloroplast DNA sequences of 

plants sampled from Colorado, Utah, Nevada, northern 

Arizona, and northern New Mexico. We found evidence 

of genetic cohesion among Nevada's Clark County 

populations as well as their genetic divergence from 

populations of other E. corymbosum varieties and 

Eriogonum species. The genetic component uncovered 

in this study supports the morphological findings upon 

which the nomenclatural change was based, attesting to 

the taxonomic distinctness of this biological entity. 

Drymocallis and Other Generic Segregates from 

Potentilla (Rosaceae) 

Barbara Ertter, UC Berkeley, Curator of Western North 

American Flora 

Abstract: Generic delimitation in tribe Potentilleae 

(Rosaceae) has historically vacillated between a 

broadly circumscribed Potentilla and recognition of 

various segregate genera. Recent convergence of morphological 

and molecular studies has shown that several 

segregates are in fact more closely related to Fragaria 

than to core Potentilla. These are accordingly treated as 

Comarum palustre, Dasiphora fruticosa, Sibbaldia procumbens, 

Sibbaldiopsis tridentata, and multiple species 

of Drymocallis in a pending volume of Flora of North 

America (FNA). The last genus includes Potentilla arguta, 

P. fissa, and P. glandulosa in North America, as 

well as 10-20 Eurasian species (e.g., D. rupestris). However, 

rather than simply transferring the existing subspecies 

or varieties of P. glandulosa into Drymocallis, a 

provisional revision was undertaken to more closely 

approximate the natural variation that occurs in western 

North America. As a result, 15 species of Drymocallis 

are recognized in FNA, some with additional varieties: 

D. arguta, D. arizonica, D. ashlandica, D. campanulata, 

D. convallaria, D. cuneifolia, D. deserertica, D. fissa, 

D. glabrata, D. glandulosa, D. hansenii, D. lactea, D. 

micropetala, D. pseudorupestris, and D. rhomboidea. 

Some species and varieties are newly described, and 

additional variation was noted as potentially deserving 

taxonomic recognition or conservation attention. This 

revision of Drymocallis acknowledges the existence of 

wide zones of intergradation and ambiguous populations, 

countered by the philosophy that conservation and 

other needs are poorly served by too broad taxonomic 

circumscriptions that gloss over valid components of 

biodiversity in an ecogeographic setting. 

Doing Adaptive Management: Improving the Application 

of Science to the Restoration of a Rare 

Tahoe Plant 

Bruce Pavlik and Alison Stanton, BMP Ecosciences, 

South Lake Tahoe, CA 

Abstract: Tahoe yellow cress (Rorippa subumbellata), 

a plant endemic to the shores of Lake Tahoe, has been a 

candidate for protection under the Endangered Species 

Act since 1999. In 2002, a conservation strategy that 

described an adaptive management process for directing 

research, management, and restoration of the species 

was adopted by 13 signatory stakeholders. Although the 

implementation phase is at least four years from completion, 

we believe it provides an operative example of 

science-driven decision making. Specifically, we have 

found that implementation of adaptive management can 

be successful if: 1) the conceptual model of the adaptive 

management process is modified to include benefits to 

biological resources in situ, 2) all stakeholders are included 

upfront in the adaptive management working 

group to participate in the strategy and design of the 

whole program, 3) key management questions are used 

to focus data collection and identify essential management 

actions, and 4) information flow and the sequence 

of project stages (actions) are designed to facilitate 

stakeholder responses. In addition, the chance of success 

is greatly increased when agencies carefully choose target 

resources that meet several corollary requirements. 

A program of experimental reintroductions of Tahoe 

yellow cress from 2003 to 2006 not only produced a 

wealth of knowledge useful to managers, it also released 

1.5 million new seeds and 10,000 new plantlets into appropriate 

habitats around Lake Tahoe. Such tangible 

benefit to the species prompted the U.S. Fish and Wildlife 

Service to downgrade the priority status of the species 

under ESA. 

15


An Annotated List of the Endemic Species of 

Arizona 

Andrew Salywon, Wendy Hodgson, Desert Botanical 

Garden, Phoenix, AZ; Todd Ontl, Desert Botanical Garden 

and Department of Ecology and Evolutionary Biology, 

Iowa State University, Ames, IA; and Martin 

Wojciechowski, School of Life Sciences, Arizona State 

University, Tempe, AZ 

Abstract: In the Flora of Arizona, Kearney and Peebles 

(1964) estimated that roughly 5% (ca.164 spp.) of the 

flora is endemic to the state, and identified southern Arizona 

as harboring nearly double the number of endemic 

species compared to other parts of the state. However, 

no list of endemic taxa was provided. Therefore, in order 

to make meaningful comparisons of the endemic 

diversity with other states and to identify “hotspots” of 

endemicity within Arizona, we are compiling and annotating 

a list of the endemic plant taxa in Arizona. The 

annotations include taxonomic synonomy, publication 

and typification information in addition to distributional, 

ecological and evolutionary relationship data. Our working 

list is composed of ca. 250 taxa from 43 families 

and identifies the northern portion of the state (namely 

the Arizona Strip and the Grand Canyon) as harboring 

the highest percentage of endemics, in contrast to Kearney 

and Peebles analyses. It is hoped that insights into 

the relationships between geographical patterns and biological 

processes that can be gained from the list, including 

comparisons of the timing and mode of evolution 

of different groups. For example, Astragalus, 

Perityle, Agave, Eriogonum and Penstemon have been 

identified as the genera with the most endemic species. 

Not surprisingly these genera are composed of mostly 

annuals to short-lived perennials, with the exception of 

Agave, and are in groups that have undergone rapid and 

recent diversification in the Quaternary. In contrast, the 

woody endemics Berberis harrisoniana, Rhus kearneyi 

ssp. kearneyi, Sophora arizonica (= Calia) & Purshia 

subintegra are most likely of Tertiary origin and relictual. 

16


Southwestern Ciénegas: Rare Habitats 

for Endangered Wetland Plants 

Robert Sivinski, 

New Mexico Forestry Division, Santa Fe, NM, retired 

Abstract. Ciénega refugia for rare plants are medium to low elevation wet meadows characterized by stable springs 

and seeps in arid regions. Ciénega soils are usually alkaline and highly organic. Several southwestern plant species 

are confined to these habitats, making ciénega biotic communities distinct from other riverine or lentic wetlands in 

the region. A comprehensive inventory of southwestern ciénegas has not been completed; however, these habitats are 

clearly rare and diminishing in extent. Groundwater depletion, erosion, conversion to agriculture or aquaculture, abusive 

grazing, and exotic weeds threaten most remaining ciénega habitats and the rare species confined to them. Government 

and non-profit conservation agencies are attempting to restore and preserve a few remnant pieces of previously 

large ciénegas in Arizona, New Mexico, and Texas. Healthy ciénega habitats require active and continuous 

management at great effort and expense, especially for weed control. 

DEFINITION AND DESCRIPTION 

‘Ciénega’ is Spanish for a swamp, bog, or marsh. It 

is also spelled ‘ciénaga’ throughout much of the Spanish-speaking 

world – especially South America and the 

Caribbean. The ‘ciénega’ spelling is prevalent in the 

American southwest and often used in northern México. 

The origin of the word ‘ciénega’ is thought to be a contraction 

of the Spanish words ‘cien aguas’ meaning ‘a 

hundred waters or fountains’ (Crosswhite 1985). This is 

an allusion to springs, seeps and wet ground over a large 

area instead of a single pool, slough, or stream channel. 

Ciénegas gained acceptance as distinct climax communities 

of ecological significance when Hendrickson 

and Minckley (1985) made a thorough assessment of the 

ciénegas of southeastern Arizona. They defined the ciénega 

climax community as mid-elevation (1,000-2,000 

m) freshwater wetlands with permanently saturated, 

highly organic, reducing soils occupied by a lowgrowing 

herbaceous cover of mostly sedges and rushes. 

Few woody plants occur in the ciénega flora and often 

only as riparian tree species around the drier margins. 

Ciénegas occur in arid landscapes with high rates of 

evaporation, so the soils at the drying wetland margins 

usually have surface crusts of alkali or salts that are the 

deposited dissolved solids of evaporated or transpired 

soil solutions. 

Ciénega biotic communities of the southwestern 

United States and northern México are almost always 

features of springs and spring seeps (Brown 1982, Hendrickson 

and Minckley 1985, Dinerstein et al. 2000). 

Not all springs support ciénegas, but almost all ciénegas 

are supported by springs. These arid-land springs arise 

where stable aquifers intercept the ground surface in 

artesian basins or along geologic faults and fractures. 

They are generally not associated with fluctuating alluvial 

aquifers in channels that are flood-scoured, so are 

more likely to be found in the upper reaches of small 

drainages near geologic faults and igneous extrusions, in 

karst topography, and on gentle slopes where waterbearing 

strata have been exposed by river erosion or 

scarps. Size of individual ciénegas varies greatly from 

less than one acre to several hundred acres and is an expression 

of spring flow and topography. 

Ciénega vegetation is usually highly productive and 

dense. A list of plant species for southeastern Arizona 

ciénegas was assembled by Hendrickson and Minckley 

(1985). Milford and others (2001) and Sivinski and 

Bleakly (2004) produced lists of ciénega plants for the 

Rio Pecos Basin of eastern New Mexico. Most individual 

ciénegas have relatively low plant species diversity, 

but contribute a productive and rare subset of wetland 

species and habitats to an otherwise arid landscape. The 

most common ciénega plants of the southwestern region 

are the open water (when present) emergents of bulrush 

(Schoenoplectus spp.) and cattail (Typha spp.); sedges 

and rushes of water-saturated soils (Eleocharis spp., 

Carex spp., Bolboschoenus maritimus (L.) Palla, Fimbristylis 

spp.); salt- and alkali-tolerant inland saltgrass 

(Distichlis spicata (L.) Greene), scratchgrass 

(Muhlenbergia aperifolia (Nees & Meyer) Parodi), and 

Mexican or Baltic rush (Juncus arcticus Willdenow 

vars. mexicanus (Willdenow) Balslev or balticus 

(Willdenow) Trautvetter) on seasonally saturated and 

sub-irrigated soils; and alkali sacaton (Sporobolus airoides 

(Torrey) Torrey) on the drier ciénega margins. 

Woody plants are usually not a significant part of ciénega 

vegetation cover (Figure 1), but patches of shrubby 

willows (Salix spp.) or willow-baccharis (Baccharis 

salicina Torrey & Gray) may occur and the drier ciénega 

margins will often have riparian trees such as cottonwood 

(Populus spp.), Arizona ash (Fraxinus velutina 

Torrey), and tree willows. 

17


Figure 1. Blue Hole Ciénega in Guadalupe County, New Mexico (September 2008) with Wright’s marsh thistle 

(right-center) and Pecos sunflower (distant yellow background). 

As climax communities associated with aridland 

springs of the southwest, ciénega-types of vegetation 

associations should be expanded to include more floristic 

regions and physical conditions than just the southeastern 

Arizona ciénegas described by Hendrickson and 

Minckley (1985). Ciénega synonyms in the Great Basin 

deserts, Sonoran and Chihuahuan deserts can sometimes 

include ‘vega’, ‘wet meadow’, ‘saltgrass meadow’, 

‘alkali meadow’ and even ‘oasis’. They are not necessarily 

confined to medium elevations and can also occur 

around desert springs at low elevations. Water coming 

from ciénega springs may be fresh or somewhat salty. In 

short, most ciénega-type habitats are the wet meadows 

that form around aridland springs and seeps. 

It is the relative permanence of the spring features 

that make many ciénega habitats biologically distinct 

from other types of wetland communities. Ciénegas are 

typically positioned in the upper reaches of small drainages 

or above river channels where they are protected 

from the scouring floods that frequently modify river 

marshes and floodplains. Ciénega spring flows may 

vary, but are less susceptible to the flooding and drying 

than playa basin wetlands during moist and arid cycles 

of the climate. Sediment cores from San Bernardino 

Ciénega in southeastern Arizona show wetland conditions 

for most of the last 7,000 years (Minckley and 

Brunelle 2007) and at Cuatro Ciénegas in Coahuila the 

fossil pollen in sediments indicate nearly identical ecological 

conditions for more than 30,000 years (Meyer 

1973). Such springs are refugia for species that may 

have been more widespread and common during wetter 

periods of the Quaternary. Several fish and invertebrate 

species are now confined to only one or a few aridland 

springs and the ciénegas they support are small remnants 

of stable habitat for some rare and endangered 

plants. 

LOSING (WET) GROUND 

The interaction of humans with aridland springs and 

ciénegas is a prehistoric tale of early and prolonged dependence 

(Hanes 2008; Rhea 2008) with a more recent 

history of almost universal destruction or diminution 

during the last two centuries (Unmack and Minckley 

2008). Hendrickson and Minckley (1985) documented 

the history and demise of many southeastern Arizona 

ciénegas – mostly by arroyo cutting that dropped spring 

aquifers below the ground surface. The tragic loss of 

most large springs and ciénegas by water withdrawals 

and aquifer depletion in the Chihuahuan desert of Trans- 

Pecos Texas is also well documented by Brune (1981), 

El-Hage and Moulton (1998), and Poole and Diamond 

(1993). Of all the southwestern states, the ciénegas of 

18


New Mexico are the least studied and documented in 

published literature. The following are a few examples 

of some historic and extant New Mexico ciénegas that I 

have personally studied. 

San Simon Ciénega was one of the wetland jewels in 

the crown of the southwest. It was a wet valley bottom 

about five miles long and a half mile broad that straddled 

the New Mexico/Arizona border in the upper-most 

reach of the San Simon Valley of Hidalgo County. The 

perennial spring-run creek had emergent marsh vegetation 

surrounded by wet meadow bordered with riparian 

woodland in sharp contrast to the adjacent Chihuahuan 

Desert scrub. When I first visited in 1994, San Simon 

Ciénega was dead and being covered with mesquite 

(Prosopis glandulosa Torrey). Many decadent cottonwood 

and willows trees still survived as reminders of 

the former wetland. The situation is the same today, but 

a little grimmer as these senescent tree remnants continue 

to decline. 

The lower part of San Simon Ciénega in Arizona was 

destroyed just before the turn of the twentieth century 

by regional overgrazing of cattle and an arroyo erosion 

cut that lowered the water table (Hendrickson and 

Minckley 1985). The arroyo headcut was arrested by a 

dam at the New Mexico border and seemed to spare the 

New Mexican part of the ciénega. Then irrigated cotton 

farming moved into the Arizona side of the valley and 

intercepted the spring aquifer emanating from the Chiricahua 

Mountains. An anonymous and undated report at 

the New Mexico Energy, Minerals and Natural Res- 

ources Department, Forestry Division documents most 

of the following events. The spring-run creek stopped 

flowing in 1952, shortly after irrigated agriculture 

started. When the Mexican duck was listed as an Endangered 

species, the New Mexico Game and Fish Department 

and federal Bureau of Land Management attempted 

to create some open-water nesting habitat in the 

still wet valley bottom by detonating powerful explosives. 

The resulting crater pools were not suitable nesting 

habitat, created habitat for weedy plants, and the 

valley bottom continued to dry from irrigated farming in 

the adjacent uplands. A nesting pond was excavated and 

frequently pumped full of water, but was eventually 

abandoned after the Mexican duck was removed from 

the list of Endangered species for taxonomic reasons. 

All permanent wetlands quickly disappeared from the 

valley. 

Another loss of desert springs and ciénegas occurred 

at a cluster of large springs near the dry mouth of the 

Rio Mimbres in Grant County. The fates of Apache 

Tejo Spring, Cold Spring, Kennecott Warm Spring, and 

Kennecott Cold Spring were to be completely captured 

by wells to supply water to the copper mill at Hurley in 

the early twentieth century. Walking from creosote desert 

into these former wetlands to find dusty gray organic 

soil supporting only clumpy alkali sacaton and 

surrounded by the decades-old carcasses of big cottonwood 

trees can only be described as shocking and depressing 

(Figure 2). Only two springs in the area remain 

wet to this day – Faywood Hotspring on its travertine 

Figure 2. Dead riparian woodland surrounding dry ciénega at former Cold Spring in Grant County, New Mexico 

(April 1993). 

19


hill with much reduced flow and no ciénega, and nearby 

Faywood Ciénega, which is supplemented and maintained 

by piped-in water from a distant upland spring. 

And what might we have lost from these wetlands? The 

type collection of Cleome sonorae A. Gray (= Cleome 

multicaulis A.P. de Candolle) was made “near the Mimbres” 

in 1851 and not seen again in the state. Perhaps 

one of these dead spring ciénegas was the habitat for 

this rare wetland species that is now likely extirpated 

from New Mexico. 

Most New Mexicans think of Lake Valley as an 

abandoned mining town in Sierra County, but the original 

Lake Valley is three miles north of the town at Berrenda 

Creek. An igneous extrusion across the broad 

Lake Valley segment backed-up a series of seasonal 

marshes and a permanent spring and seeps with ciénega 

vegetation. Berrenda Creek has only intermittent flow 

during wet seasons and storm events, but the runoff captured 

in Lake Valley sediments slowly discharged into a 

perennial spring run at the base of the valley. A diversion 

levee around Lake Valley was constructed about a 

century ago that dried the wetlands, which were converted 

to irrigated agriculture. The diversion outlet 

caused the lower spring run to erode into a deeply incised 

channel that still supports riparian woodland, but 

has lost its ciénega habitats. 

Some of the best remaining examples in New Mexico 

of ciénega habitats around aridland springs occur in 

the Rio Pecos valley on karst topographies near Santa 

Rosa (Guadalupe County) and Roswell (Chaves 

County). Even these are degraded and shrinking in size. 

Hundreds of acres of seeping ground with spring runs 

and sinkhole lakes have become surrounded by the City 

of Santa Rosa. About half of the original ciénega habitats 

have been damaged or destroyed by excavations for 

fish hatchery ponds or public fishing ponds; filling for 

sport fields, buildings, roads and parking lots; and dense 

infestation by Russian olive (Elaeagnus angustifolia L.) 

and salt cedar (Tamarix chinensis Loureiro) forests. 

A 25-mile stretch of Rio Pecos valley from Roswell 

south to Dexter has sinkhole lakes, resurgent creeks, 

spring runs, and seeps – sometimes with extensive ciénegas. 

Some of these have also been damaged by fish 

hatchery operations (Dexter National Fish Hatchery) 

and recreational development (Bottomless Lakes State 

Park and Roswell Country Club). Aquifer depletion below 

a municipal well completely dried a mile-long ciénega 

north of Dexter and other springs and seeps are 

likely impaired by the irrigated agriculture that dominates 

the landscape on the west side of the river. Fortunately, 

large seeps and spring runs from the highly alkaline 

Salt Creek aquifer support some ciénega habitats on 

Bitter Lake National Wildlife Refuge near Roswell. Unfortunately, 

the wildlife focus on this refuge has caused 

extensive damage to ciénega habitats with the numerous 

dikes, diversions and drains constructed to enhance fish 

and waterfowl habitats – although this is recently changing 

with more attention being paid to rare wetland 

plants. 

RARE PLANTS OF SOUTHWESTERN 

CIÉNEGAS 

While several animal species are endemic to particular 

aridland springs or areas of spring features, very few 

ciénega plants are so narrowly endemic. Some notable 

exceptions include Amargosa niterwort (Nitrophila mohavensis 

Munz & Roos), Ash Meadows mousetails 

(Ivesia kingii S. Watson var. eremica (Coville) Ertter), 

Ash Meadows gumplant (Grindelia fraxinopratensis 

Reveal & Beatley), and spring-loving centaury 

(Centaurium namophilum Reveal, Broome & Beatley), 

which are endemic to Mohave Desert springs around the 

Ash Meadows region of southwestern Nevada and adjacent 

California. Another very rare ciénega plant with a 

very small geographic distribution is reclusive lady's 

tresses orchid (Spiranthes delitescens Sheviak). This 

Endangered orchid is presently known from only four 

ciénegas in close proximity near the international border 

of southern Arizona (Coleman 2000). 

Most rare ciénega plants have very broad distributions 

of several hundred, or sometimes more than a 

thousand, miles in length. They are rare species because 

their ciénega habitats are very rare. Widespread wetland 

species usually do not get much attention from rare 

plant botanists because of multiple-state (or country) 

distributions and the difficulties of accessing a class of 

habitat that is predominantly on private property. A few 

widespread ciénega species, however, are starting to get 

some much needed scrutiny by botanists and land managers 

in southwestern states. 

Only three extant populations of Parish’s alkali grass 

(Puccinellia parishii A.S. Hitchcock) were known at the 

time this species was proposed for inclusion on the Endangered 

Species list in 1994. This annual grass occupies 

the highly alkaline soils of aridland springs and 

ciénegas. The proposal to list gave southwestern field 

botanists the incentive (funding) to search for new 

populations, which located or confirmed a total of 30 

sites at seeps and ciénegas – 17 in New Mexico, 11 in 

Arizona, 1 in eastern California, and 1 in southwestern 

Colorado (FR Vol. 63, No. 186, 51329-51332, 9/25/98). 

USDI-Fish and Wildlife Service withdrew the proposal 

to list in 1998, but New Mexico still lists this plant as 

state endangered because it is a wetland species with 

less than 100 acres of total known occupied habitat 

(New Mexico Rare Plant Technical Council 1999). 

The Pecos sunflower (Helianthus paradoxus Heiser) 

is another good example of a widespread ciénega plant 

that is a threatened species. It occupies only alkaline 

spring ciénegas from western Texas to west-central New 

20


Mexico. The dramatic and almost complete demise of 

aridland ciénegas from aquifer depletion in the Chihuahuan 

Desert of Texas left only two populations of Pecos 

sunflower in a region that probably contained several. 

Some of the New Mexico populations are also damaged 

or threatened by aquifer depletion and nearly all are degraded 

by exotic tree infestations (USDI-Fish & Wildlife 

2008). Pecos sunflower was listed as a federally 

threatened species in 1999 and its ciénega habitats are 

finally receiving some management attention specific to 

the needs of this plant. 

Wright’s marsh thistle (Cirsium wrightii A. Gray) 

sometimes occurs in the same New Mexican ciénegas 

occupied by Pecos sunflower, but there appear to be 

fewer thistle populations in the United States. It ranges 

from southeastern New Mexico to southeastern Arizona 

and northern Chihuahua and Sonora. The type locality 

and single Arizona location at San Bernardino Ciénega 

has not been seen again since that ciénega was dried by 

down-cutting of the adjacent Black Draw. Some New 

Mexico populations at Lake Valley, Sacramento Mountain 

springs, and the City of Roswell (Country Club) 

have also been extirpated (New Mexico Rare Plant 

Technical Council 1999, Sivinski 1995, 2005). This is 

clearly a threatened ciénega species in the United States; 

however, the status of this plant in México is unknown. 

A dismal trend of aridland spring loss in México (Contreras 

and Lozano 2002, Unmack and Minckley 2008) 

offers little hope that this species is faring better south 

of the border. Cirsium mohavense (Greene) Petrak 

(synonym = Cirsium virginense Welsh) is a related wetland 

thistle that may be occupying the same sinking ship 

in the Mojave Desert except that the Mojave thistle is 

not exclusively a ciénega plant and also occurs in some 

hanging garden and riparian habitats (Utah Native Plant 

Society 2008). 

Leoncita false foxglove (Agalinus calycina Pennell) 

also co-occurs with Pecos sunflower and Wright’s 

marsh thistle in a ciénega at Bitter Lake National Wildlife 

Refuge in southeastern New Mexico. It is otherwise 

only known from an extant population at the Diamond 

Y Spring ciénega in western Texas, another historic and 

ambiguous collection in western Texas, and an historic 

collection in Coahuila (Poole et al. 2007). This is another 

species with almost no data available on its status 

in México. It seems to be exceedingly rare, but much 

additional research must be accomplished to support the 

initial appearance of rarity. 

Additional botanical surveys of all ciénegas in the 

southwestern United States and northern México will be 

needed to fully understand ciénega plant distributions 

and the threats to their habitats. Botanists should consult 

southwestern ichthyologists, herpetologists and aquatic 

invertebrate biologists who have been much more ag- 

gressive in locating and gaining access to aridland 

springs. They can help determine which springs support 

ciénega habitats and may already know many of the 

landowners. 

MANAGEMENT CHALLENGES 

Some remnant southwestern ciénegas have been acquired 

by federal and state governments and The Nature 

Conservancy as natural preserves or wildlife refuges. 

These have usually been protected because of the rare or 

endangered animals inhabiting the actual spring features, 

but the rare ciénega plants also need to be considered 

in preserve management. Ciénegas are productive 

and dynamic biotic communities that have attracted use 

by large herbivores for millions of years. A protective 

fence and hands-off approach for preserve management 

may only yield a ciénega that is overgrown, thatchy, 

drying, and pest-ridden (Kodric-Brown and Brown 

2007, Unmack and Minckley 2008). Needs for grazing 

or fire prescriptions, aquifer protection or restoration, 

and weed control calls for active management. 

Restoration and management of ciénegas affected by 

arroyo cuts that have lowered the potentiometric surface 

of adjacent springs and seeps will require the very difficult 

task of aggrading incised channels (Minckley and 

Brunelle 2007, Turner and Fonseca 2008). The ground 

water of a dead or damaged ciénega may still be close to 

the surface, but requires significant sedimentation and 

restoration of sheet flow to bring the potentiometric surface 

back to ground level and re-establish a “living” ciénega. 

On the other hand, former ciénegas supported by 

spring aquifers that have been depleted by groundwater 

pumping are unlikely to resume surface flow and become 

“living” again for the foreseeable future. 

Blue Hole Ciénega in Santa Rosa, New Mexico was 

purchased by the State Forestry Division’s Endangered 

Plant Program in 2005 to preserve critical habitat for the 

endangered Pecos sunflower and Wright’s marsh thistle. 

This 116-acre ciénega was about one-third infested with 

Russian olive trees (plus salt cedar to a lesser extent), 

suddenly ungrazed by livestock, and illustrative of some 

vegetation management challenges in a ciénega preserve 

(Figures 3 and 4). 

Weed tree control was an immediate concern because 

the entire ciénega was rapidly trending towards Russian 

olive woodland. Inmate work crews with chainsaws and 

backpack herbicide sprayers spent a total of 3,600 manhours 

cutting trees, spraying stumps, and broadcasting 

slash during the late summer and autumn months when 

the soil surface was dry over much of the ciénega. Winter 

to summer was an unsuitable period for weed control 

because effective herbicides could not be used while the 

soil surface was constantly wet or pooling water. The 

initial percent kill for tree stumps was about 80%. 

21

Figure 3. Work crew felling Russian olive trees at Blue 

Hole Ciénega (September 2007). 


wide by October when a work crew spent another 1,000 

man-hours treating the resprouts with herbicide. 

After almost three years of weed tree control, controlled 

fire, and expenditure of $80,000, Blue Hole Ciénega 

might be restored to a point where a small crew 

can annually destroy new weed tree seedlings within a 

few days. The Albuquerque Chapter of the Native Plant 

Society of New Mexico has volunteered to annually inspect 

the ciénega for other weed species that could arrive 

and become established, if not quickly detected. 

Accumulations of dead, thatchy, native vegetation will 

still have to be managed. Since high densities of Pecos 

sunflower are desirable at this preserve, intensive livestock 

grazing during the growing season is not an option. 

Therefore, controlled fires at suitable frequency 

(yet to be determined) will be needed for the foreseeable 

future. 

Most southwestern ciénegas are in private ownership 

because the spring features associated with them are 

valuable assets in an arid region. Restoration, protection 

and management of these rare and unique habitats are 

costly and require perpetual effort. Government programs 

that acquire ciénegas or assist landowners with 

their management are greatly needed in the southwestern 

states. 

When the slash had dried for three to five months, 

the entire ciénega was burned in December. Winter is 

the only feasible burn period because Pecos sunflowers 

set seed and die in October and the seeds begin to germinate 

in late February. The burn and mop-up took two 

days and was conducted by three State Forestry wildland 

fire crews, one USDA-Forest Service wildland fire 

crew and a pumper crew from the Santa Rosa Municipal 

Fire Department. The fire carried very well through the 

thick fine fuel of the grass and rush cover, but was generally 

not hot enough to consume tree slash more than 

one-inch in diameter. 

The bare, open ground of the burned ciénega created 

optimum habitat for germination and growth of annual 

ciénega plants such as Pecos sunflower and seepweed 

(Suaeda calceoliformis (Hooker) Moquin-Tandon) 

(Figures 5 and 6). Perennial ciénega plants quickly developed 

a lush growth that covered the charred one- to 

twelve-inch diameter tree slash lying on the ground. 

Most of the Wright’s marsh thistle rosettes that had been 

burned made new rosettes and many bolted flower 

stalks the following autumn. Unfortunately, about 20% 

of the weed tree stumps resprouted, and the still living 

roots around each dead stump sprouted one or more new 

weed tree saplings. These were three to four feet tall and 

Figure 4. Controlled burn of Blue Hole Ciénega 

(December 2007). 

22


Figure 5. Photo reference point at Blue Hole Ciénega before treatment (August 2006). 

Figure 6. Photo reference point at Blue Hole Ciénega after treatment (September 2008). 

23


LITERATURE CITED 

Brown, D.E. 1982. Biotic communities of the American 

Southwest – United States and México. Desert 

Plants 4(1-4): 1-341. 

Brune, G. 1981. Springs of Texas. Vol. 1. Branch- 

Smith, Inc., Fort Worth, TX. 566 p. 

Coleman, R.A. 2000. Orchids at a range limit in Arizona 

and New Mexico. North American Native Orchid 

Journal 6(3): 193-200. 

Contreras-Balderas, S. and M. de L. Lozano-Vilano. 

1994. Water, endangered fishes, and development perspectives 

in arid lands of México. Conservation Biology 

8: 379-387. 

Crosswhite, F.S. 1985. Editorial introduction to Desert 

Plants 6(3). 

Dinerstein, E., D. Olson, J. Atchley, C. Loucks, S. 

Contreras-Balderas, R. Abell, E. Iñigo, E. Enkerlin, C. 

Williams, and G. Castilleja. 2000. Ecoregion-based conservation 

in the Chihuahuan Desert: A biological assessment. 

World Wildlife Fund. Available: www.worldwild 

life.org/what/wherewework/chihuahuandesert/WWF 

Binaryitem2757.pdf [27 February 2009]. 

El-Hage, A. and D.W. Moulton. 1998. Evaluation of 

selected natural resources in parts of Loving, Pecos, 

Reeves, Ward, and Winkler counties, Texas. Texas 

Parks & Wildlife, Resource Protection Division: Water 

Resources Team, Austin. 40 p. 

Hanes, C.V. 2008. Quaternary cauldron springs as 

paleoecological archives. In: L.E. Stevens and V.J. Meretsky 

(eds.) Aridland Springs in North America: Ecology 

and Conservation. The University of Arizona Press 

and The Arizona-Sonora Desert Museum, Tucson. pp. 

76-97. 

Hendrickson, D.A. and W.L. Minckley. 1985. Ciénegas 

– Vanishing climax communities of the American 

Southwest. Desert Plants 6(3): 131-175. 

Kodric-Brown, A. and J.H. Brown. 2007. Native 

fishes, exotic mammals, and the conservation of desert 

springs. Frontiers in Ecology and the Environment 5 

(10): 549-553. Available: www.frontiersinecology.org 

[27 February 2009]. 

Meyer, E.R. 1973. Late-Quaternary paleoecology of 

the Cuatro Ciénegas Basin, Coahuila, México. Ecology 

54(5): 982-995. 

Milford, E., E. Muldavin, Y. Chavin and M. Freehling. 

2001. Spring vegetation and aquatic invertebrate 

survey 2000. Final Report to Bureau of Land Management, 

Roswell Field Office. 91 p. Available: http:// 

nhnm.unm.edu/ [27 February 2009]. 

Minckley, T.A. and A. Brunelle. 2007. Paleohydrology 

and growth of a desert ciénega. Journal of Arid Environments 

69: 420-431. 

New Mexico Rare Plant Technical Council. 1999. 

New Mexico Rare Plants. Albuquerque, NM: New Mexico 

Rare Plants Home Page. http://nmrareplants.unm. 

edu (Latest update: 22 January 2009) [27 February 

2009]. 

Poole, J.M. and D.D. Diamond. 1993. Habitat characterization 

and subsequent searches for Helianthus paradoxus 

(puzzle sunflower). In: R. Sivinski and K. Lightfoot 

(tech. eds.) Southwestern rare and endangered plants. Proceedings 

of the Southwestern Rare and Endangered 

Plant Conference; March 30‐April 2, 1993; Santa Fe, 

NM. NM Forestry Division, Miscellaneous Publication 

No. 2, Santa Fe. p. 53-66. 

Poole, J.M., W.R. Carr, D.M. Price and J.R. Singhurst. 

2007. Rare Plants of Texas. Texas A&M University Press, 

College Station. 640 p. 

Rhea, A.M. 2008. Historic and prehistoric ethnobiology 

of desert springs. In: L.E. Stevens and V.J. Meretsky 

(eds.) Aridland Springs in North America: Ecology and 

Conservation. The University of Arizona Press and The 

Arizona-Sonora Desert Museum, Tucson. p. 268-278. 

Sivinski, R.C. 1995. Wright’s marsh thistle (Cirsium 

wrightii). 1995 Progress Report (Section 6, Segment 10) 

to USFWS Region 2 Office, Albuquerque, NM. 8 p. 

Sivinski, R.C. 2005. Intermittent monitoring of Parish’s 

alkali grass (Puccinellia parishii), Wright’s marsh 

thistle (Cirsium wrightii), and Mescalero milkwort 

(Polygala rimulicola var. mescalerorum). Section 6, 

Segment 19 report to USFWS Region 2 Office, Albuquerque, 

NM. 6 p. 

Sivinski, R.C. and D. Bleakly. 2004. Vascular plants of 

some Santa Rosa wetlands, east-central New Mexico. The 

New Mexico Botanist 29: 1-5. Available: http://aces. 

nmsu.edu/academics/rangescienceherbarium/ [27 February 

2009]. 

Turner, D. and J. Fonseca. 2008. What is a ciénega? 

(and why do we care?). Restoring Connections, Newsletter 

of the Sky Island Alliance 11(2): 1-4. 

Unmack, P.J. and W.L. Minckley. 2008. The demise 

of desert springs. In: L.E. Stevens and V.J. Meretsky 

(eds.) Aridland Springs in North America: Ecology and 

Conservation. The University of Arizona Press and The 

Arizona-Sonora Desert Museum, Tucson. p. 11-34. 

Utah Native Plant Society. 2008. Utah Rare Plant 

Guide. Salt Lake City, UT: Utah Rare Plant Guide 

Home Page. http://www.utahrareplants.org [27 February 

2009]. 

USDI-Fish and Wildlife Service. 2008. Environmental 

assessment for designation of critical habitat for 

Pecos sunflower. Region 2, Albuquerque, NM. 64 p. 

24


A New Look at Ranking Plant Rarity for Conservation Purposes, 

With an Emphasis on the Flora of the American Southwest 

John R. Spence, 

Glen Canyon National Recreation Area, Page, AZ 

Abstract. A new rarity ranking system for prioritizing vascular plants for conservation and research is developed. 

This new system, termed the “At-Risk System” (ARS), ranks species using six variables, with each variable scored 

from 0-3; rarity type, biology (life-from, breeding system, pollination ecology, and dispersal ecology), population 

trend, anthropogenic threats, climate change vulnerability, and number of populations. Scores can range from 0 to 

18, with higher numbers indicating greater potential at-risk status. Selected species from the American Southwest are 

scored using the new system. Scores range from 2 for Pinus ponderosa to 18 for the critically endangered Arctomecon 

humilis. We know little about the biology and status of many rare species in the Southwest. This lack of 

knowledge was incorporated by using an automatic 3 score for the variables for which data are not available, which 

highlights uncertainty. For many species, the ARS score is not always strongly correlated with other ranking systems 

such as its ESA status or NatureServe G rank. 

The American Southwest supports one of the richest 

floras in North America, with perhaps as many as 6,000 

indigenous species distributed among the deserts and 

mountains of the region. The area includes six major 

arid and semi-arid biomes: the Chihuahuan, Colorado 

Plateau, Great Basin, Mohave, and Sonoran Deserts, and 

the Madrean region that extends from Mexico into 

southern New Mexico and Arizona. A recent compilation 

of rare species (Spence unpublished) has put the 

number of G1 and G2-ranked species at ca. 700. This 

preliminary list does not include the numerous local varieties 

of more common and widespread species, nor 

those species restricted to the Mexican portions of the 

biomes. With scarce resources, relatively few field botanists, 

and impending major climate change there is an 

urgent need to prioritize these species for conservation 

purposes. Unfortunately, very little is known about the 

basic ecological and biological characteristics of many 

of them. Thus, although there are ca. 260 G1 species in 

the American Southwest, we know relatively little about 

how to prioritize this list based on the likelihood of 

near-term extinction. Any attempt to rank large numbers 

of species for conservation funding must thus use a 

"triage" approach, based on general biological characteristics 

when more detailed quantitative data are not 

available. Rarity is also an elusive characteristic that 

may not be well reflected in its G rank, as it can change 

through time and space, and is not easily defined or 

quantified. Thus what defines rarity can vary across 

different scales (Harper 1981). 

There are three principal ranking systems that exist 

for rare plants, two of which have been applied locally. 

The California Native Plant Society ranking system has 

been developed for the flora of California (California 

Native Plant Society 2001). Their system includes four 

lists indicating general status inside and outside of California, 

combined with a subjective 3-number code indicating 

local distribution, rarity and risks. These values 

are based on fairly detailed information, which is often 

not available for rare plants elsewhere in the Southwest. 

The California 1B list of rare endemic species includes 

over 1,000 taxa. The Nature Conservancy developed the 

G ranking system, which was then applied to state-level 

species lists throughout the West. However, the G ranking 

is a rather coarse-scale tool given the large number 

of species with either a G1 or G2 rank regionally. The 

third system, the IUCN threat categorization (Mace 

1994; Mace and Lande 1991), has not been applied to 

southwestern rare plants. Other more detailed systems 

have been developed (e.g., Bond 1994; Kwak and Bekker 

2006) but often require detailed information on genetics 

and estimates of pollen flow and fecundity. There 

remains a need among field biologists, land managers 

and conservation planners for a general system that can 

quickly rank and prioritize species within the G1-G2-G3 

levels. 

A significant amount of research has been done in 

attempting to correlate species extinction and rarity with 

biological traits (see reviews in Brigham and Schwartz 

2003; Krupnick and Kress 2005; Kunin and Gaston 

1997). In a seminal paper, Rabinowitz (1981) developed 

a rarity matrix that was based on geography, habitat specialization, 

and local population abundance. This was 

further discussed and applied by Kruckeberg and Rabinowitz 

(1985). However, this matrix approach to rarity 

has been little utilized in rare plant conservation planning. 

Kruckeberg and Rabinowitz suggested several 

avenues of research to pursue that may provide insight 

into rarity, including molecular/genetic, habitat, demographic 

and breeding system characteristics. More re- 

25


cently, Bevill and Louda (1999) examined the literature 

and documented 71 plant variables that had been used in 

the study of rarity and extinction. Many of these require 

detailed demographic or lab-based genetics research, 

and are thus impractical at a scale of the American 

Southwest in the near-term. Also, many reviews and 

studies have failed to find a strong link between specific 

variables and rarity (e.g., Aizen et al. 2002; Lavergne et 

al. 2004), although other studies have found such links 

(e.g., Sjöström and Gross. 2006). The general consensus 

is that demographic and breeding system data and 

molecular investigations overlaid with climate envelope 

analyses provide the best research approaches to understanding 

rarity (Brigham and Schwartz 2003; Gitzendanner 

and Soltis1999; Kunin and Schmida 1997). Beyond 

these more detailed techniques, some regional surveys 

have found traits that are correlated with rarity, 

including life-form and longevity (Harper 1979; 

Sjöström and Gross 2006;), breeding system (Sakai et 

al. 2002; Sjöström and Gross 2006; Sodhi et al. 2008;), 

pollination ecology (Robertson et al. 1999; Sakai et al. 

2002; Sodhi et al. 2008) and dispersal ecology (Bond 

1994). 

Any ranking system should be general enough to be 

widely applicable, but also provide the ability to rank 

species from potentially low to potentially high "at-risk" 

status. It should also be based on what is known about 

the biology of rarity. I have developed a preliminary 

ranking system, with an emphasis on the arid and semiarid 

flora of the American Southwest, based on six elements. 

This system should be viewed as a draft attempt 

to provide an early warning ranking for those species 

that potentially may be most at-risk in the Southwest. 

The time-frame is the next 50 years, when anthropogenic 

threats and climate change are likely to push many 

rare plants towards extinction. The elements are rarity 

type, biology, population trends, number of populations 

(TNC element occurrences), direct anthropogenic 

threats, and climate change vulnerability. Each of these 

is discussed below, with examples and ranking criteria 

in tabular form. 

ELEMENTS OF THE RANKING SYSTEM 

Each element is scored from 0 (low risk) to 3 (high 

risk). The six elements are rarity, biology, population 

trends, anthropogenic threats, climate change vulnerability, 

and G rank (here called the N rank, see below in 

Table 3). The final score is called the at-risk score 

(ARS) for each species. Each of the elements is described 

below. 

1. Rarity Type 

Table 1 lists the seven forms of rarity as defined by 

Rabinowitz (1981), scored either 0 or 1 for each level of 

the three categories. Each of the eight types is also 

coded from RI-RVIII. Since the three variables used in 

the system are subjective, they need to be defined in the 

local context of the American Southwest. Widespread 

species are defined as those that are typically found 

across one or more provinces (more than one subprovince) 

as defined by McLaughlin (2007). Thus a 

widespread species could occur in the Southwestern 

Region, Sonoran Province, and both the Sonoran and 

Mohavian subprovinces. Geographically local species 

are defined as those endemic to a single subprovince, 

such as the Colorado Plateau or Great Basin. Habitat 

specialists are defined as species that are generally restricted 

to one or a very few similar soils or substrates, 

while habitat generalists occur across a variety of substrates 

and soil types. Wetland species are also typically 

characterized as habitat specialists. Abundance at local 

sites (population or element occurrences) is more difficult 

to define, and remains subjective in my system. 

Many endemic species can be quite abundant locally (cf. 

Lesica et al. 2006; Spence unpublished data), while 

other species always seem to be uncommon or sparse 

where they are found. A species can have a final score 

of 0 for widespread common habitat generalists, to 3 for 

local sparse habitat specialists. 

2. Biology 

Detailed studies have shown that traits most closely 

related to rarity include low population size, demo- 

Table 1. The Rabinowitz rarity matrix (Rabinowitz 1981) scored from 0-3 for the ranking system. 

Abundance at 

Sites 

Geographic Range Widespread (0) Local (1) 

Habitat Specificity Generalist (0) Specialist (1) Generalist (0) Specialist (1) 

Common (0) RI 0 RII 1 RIII 1 RIV 2 

Sparse (1) RV 1 RVI 2 RVII 2 RVIII 3 

26


graphic factors, competitive abilities, inbreeding depression, 

low pollen/ovule ratios and pollen limitation, or 

highly specialized breeding, pollination and dispersal 

systems (Brigham and Schwartz 2003; Byers and 

Meagher 1997; Cole 2003; Gitzendanner and Soltis 

2000; Purdy et al. 1994; Schemske et al. 1994; Walck et 

al. 1999). In some cases rarity may also occur due to 

extreme and rapid habitat loss in formerly common species 

(eg., Ge et al. 1999). In the absence of data on these 

variables, proxies are needed that can be linked to the 

biological characteristics of rare species. Although the 

literature is somewhat ambivalent (e.g., Aizen et al. 

2002; Bevill and Louda 1999; Sakai et al 2002), some 

studies have suggested that a variety of basic plant characteristics 

may be used to provide preliminary indications 

of possible vulnerability and general extinction 

probabilities. 

Thus self-incompatible breeders, species with highly 

specialized and rare pollinators, and species that depend 

on highly specialized dispersers are often considered to 

be more at risk than species with more generalized biology 

(Aizen et al. 2002; Bond 1994; Buchmann and Nabhan 

1996; Kwak and Bekker 2006). I have developed a 

set of four proxy variables based on the literature, with 

the caveat that these are only general indicators of potential 

at-risk status. Detailed studies are clearly needed 

for most species in the Southwest in order to determine 

the exact causes of rarity, which are likely to be taxon 

(genus, species) specific (Bevill and Louda 1999). The 

four proxy variables (traits) are life-form, breeding system, 

pollination ecology, and dispersal ecology. Each of 

these is discussed with examples in Table 2. This element 

is scored for a species by selecting the single highest 

score among the four biological traits, rather than 

averaging them. 

Table 2. The biology element scored using four principal biological traits to characterize the vulnerability 

of a species to extinction. 

Biological Traits Ranked Explanation Score Examples 

Long-lived woody species 

Short-lived woody species or 

long-lived herbaceous species 

Short-lived perennial 

herbaceous species 

Annual or biennial 

Life-form and longevity 

Long life buffers against environmental 

change (>100 yrs) 

Some vulnerability, but generation times 

tend to buffer against short-term 

changes (>25 yrs) 

Vulnerable, short generation time may 

not be able to track environmental 

changes (3-25 yrs) 

Extremely vulnerable, with seed bank 

longevity a critical factor in persistence 

of populations 

Breeding System 

0 Conifers, Coleogyne 

1 Atriplex, Ericameria, 

Pediocactus 

2 Astragalus, Eriogonum 

Penstemon 

3 Cryptantha, Ipomopsis, 

Phacelia 

Autogamous 

Mixed mating 

Facultative outcrossing; 

some autogamy 

Obligate outcrossing; xenogamy, 

dioecy, loss of sexual reproduction, 

or breeding system 

unknown 

Can set seed despite small numbers of 

individuals through selfing 

Flexible system that allows for reproduction 

with or without pollinators, selfing 

occurs 

Generally requires pollen transfer between 

individuals, with low selfing rates 

often associated with reduced fitness 

Requires more than one individual, specialized 

pollen transfer 

27 

0 Many annuals 

1 Many generalized insect 

pollinated taxa 

2 Many insect pollinated taxa 

3 Orchidaceae, dioecious species


Table 2. continued 

Biological Traits Ranked Explanation Score Examples 

Wind, water, self-pollination 

Generalized insects and vertebrates 

Generalized vertebrates 

Specialized animal pollination, 

or pollination system unknown 

Wind, water 

Generalized animal dispersal 

No structural or specialized features 

Specialized animal dispersal 

Pollination Ecology 

No requirements for animal vectors, can 

pollinate and set seed regularly 

Generally can pollinate and set seed 

through visitation of many different animal 

groups 

Generally can pollinate and set seed as 

long as hummingbirds, or other bird or 

bat species are present 

Requires a specialized, and often rare, 

insect or vertebrate species for pollination 

and seed set 

Dispersal Ecology 

Can be transported easily by wind or 

water 

Many bird and mammal species can 

disperse fruits 

No requirements for animal-mediated 

dispersal 

Requires a particular specialized animal 

vector to disperse seeds, such as Clark's 

nutcracker, Red crossbill, some antdispersed 

species 

0 Betulaceae, Conifers, Populus, 

aquatic species, many annuals 

1 Asteraceae, other open "dinnerplate" 

flowers 

2 Bat and hummingbirdpollinated 

species 

3 Agavaceae, Asclepias, 

Orchidaceae 

0 Spores, dust seeds, wind dispersed 

species 

1 Berry-producing species, sticktights, 

etc. 

2 Smooth seeds, short-distance 

dispersal only 

3 Fruits/seeds designed for specialized 

dispersal; Pinus albicaulis, 

Viola sp. 

3. Population Trends 

Mace (1994) suggested that population declines or 

loss of populations could be used to rank rare species. 

When data are available, this element can provide an 

accurate and straight-forward determination of the atrisk 

status of a species, but detailed information on 

population size and demographic data are difficult to 

collect and such data sets remain rare. General anecdotal 

information can be of some use in scoring a species 

in this element, however, such as loss of populations or 

general observational data showing declining numbers 

in populations. It is recommended that if time series, 

demographic, or abundance data are not available, that a 

species should be scored conservatively (higher), including 

a 3 when relevant as "unknown trends" (Table 

3). 

4. Direct Anthropogenic Threats 

Anthropogenic threats can include direct and indirect 

threats through climate change. I have separated out 

climate change from other more direct anthropogenic 

threats as the specific characteristics used to score them 

are different. Direct threats include water diversion or 

ground water pumping, recreational activity (e.g., OHV 

use), domestic livestock grazing, mining, agricultural 

development, introduction of invasive exotics, or urbanization. 

Threat levels can vary from minimal such 

as is found in many protected areas, to severe on lands 

around rapidly growing urban centers. This element is 

scored based on approximate values for impacts to habitat, 

ranging from generally minimal to >90% of a species 

habitat impacted. GIS analysis with ground surveys 

are often necessary to quantify the extent of impacts, but 

generally a rough first approximation can be made 

based on the knowledge of researchers familiar with the 

species and its habitat and associated threats. Thus the 

species scores remain somewhat subjective, based primarily 

on levels of degradation or impacts to the species 

overall habitat. Often this is not well known or is difficult 

to categorize. Especially difficult to quantify is the 

magnitude and imminence of these threats, and more 

work is needed to develop a more objective set of criteria. 

For now, the element is scored based on the presence 

of impacts regardless of intensity and timing. Gen- 

28


erally, when threats are poorly known and hard to quantify 

given available information, it is recommended that 

a conservative (higher) score be used. This element is 

scored range wide in this preliminary analysis (Table 3). 

5. Climate Change Vulnerability 

The climate envelope of a species can be modeled to 

predict distributional changes with global warming (e.g., 

Miles et al. 2004). Although we lack data for many species, 

a proxy variable that can provide a first approximation 

to the vulnerability of a species to climate change is 

its elevation range. If a species has a relatively broad 

elevation range, populations along the gradient will be 

adapted to local climates, and the species genetic ability 

to adapt will likely be greater than in a species with a 

very small elevation range exposed to a much smaller 

range of climate variability. Elevation ranges from 

1000 meters are used to score each species 

for climate change vulnerability (Table 3). 

6. Number of Known Populations 

This element in the ranking system is based on the 

general concept of element occurrences developed by 

the Nature Conservancy and further refined by Nature- 

Serve (Faber-Langendoen et al. 2009). The five categories 

are the same as the G and S ranks on the various 

state and state natural heritage databases, although I 

have provided more explicit criteria for the G4 and G5 

levels. Species with widespread continuous populations, 

such as many conifers, are ranked G4 if found within a 

single subprovince, and G5 if they are in more subprovinces. 

Thus a widespread species such as Saguaro 

(Carnegiea gigantea) is ranked as a G4, despite its large 

population size, because of its restriction to the Sonoran 

Desert. Since there are slight differences between this 

ranking element and the G scale, I have given this element 

the letter N for number of known populations. The 

scores are inverted from the N rank, thus a N1 is scored 

a three, an N2 a two, a N3 a one, and N4N5 a zero. 

Table 3. Four additional elements of the scoring system for ranking at-risk species, with scores from 0-3, 

and explanations for how each is scored. 

Ranking Elements 3-6 Score Explanation 

3. Population Trend Historical trends (if known) since approximately 1900, or when the species was 

first discovered. 

Stable or increasing 0 Populations are secure for the foreseeable future, with trends exhibiting natural 

short and long-term variability 

Minor declines 1 Data are available to indicate that some declines have occurred, (10% decline in total numbers 

per year 

4. Direct Anthropogenic 

Threats 

Scores are typically applied range-wide, although they could be used in a more 

narrowly defined region if needed. 

No direct threats 0 Few if any impacts from recreation or domestic livestock grazing, no invasive exotics 

present, no mining activity, recreational impacts largely absent, etc.;


Table 3. continued. 

Ranking Elements 3-6 Score Explanation 

5. Climate Change 

Vulnerability 

6. Number of known 

populations 

Estimate the elevational range based on population locality information. 

Low 0 Climate envelope and/or elevation range wide (>1000 m), apparently secure from 

climate change for the next 50 yrs 

Moderate 1 Climate envelope and/or elevation range moderate (>500 m), some vulnerability 

to climate change with moderate declines expected in next 50 yrs, secure for next 

10 yrs 

High 2 Climate envelope and/or elevation range small (100-500 m), vulnerable to climate 

change in the next 10 yrs, long-term (>50 yrs) probability of survival moderately 

low, with major population declines 

Severe 3 Climate envelope/elevation range very small (50 yrs) very low to none, declines 

or local extirpation expected within 10 yrs; or elevation range unknown; or 

an obligate wetland species in American Southwest 

N0 -- Extinct, at least in wild 

N1 3 1-5 known populations, endangered (=G1) 

N2 2 6-20 known populations, threatened (=G2) 

N3 1 21-100 known populations, vulnerable (=G3) 

N4 0 101-300 known populations, or relatively widespread and in more or less continuous 

stands, apparently secure (=G4) 

N5 0 >300 known populations or widespread and continuous, secure (=G5) 

For many species the definition of what constitutes a 

discrete population is difficult. In the case of widespread 

common species that occur in large continuous 

stands, the need to define discrete populations is of less 

importance than for rare species with localized scattered 

distributions In most cases, G1, G2, and G3 species 

tend to occur in scattered populations with gaps, where 

they are absent even in suitable habitat. For many of 

these species, gene flow and dispersal are likely to be 

relatively low, thus isolated populations, even if only a 

few hundreds of meters apart, can be effectively considered 

discrete and non-interacting populations. In Table 

3, those species with larger more continuous species 

(G4 or G5) can be scored without reference to the number 

of discrete populations. 

SCORING SPECIES USING THE SYSTEM 

For each species, the at-risk score is the sum of the 

scores for each of the six elements: rarity, biology, declines, 

anthropogenic threats, climate change vulnerabil- 

ity and N populations (see Tables 1-3), thus: 

At-risk Score (ARS) = rarity type + biology+ declines + 

threats + climate change + N score 

A local sparse habitat specialist with only 1-5 known 

occurrences would be scored very high. The larger the 

score, the more potentially at-risk the species should be 

considered as a first approximation. The overall conservation 

score can vary from 0 to 18. Examples for selected 

species are given in Tables 4 and 5 and are discussed 

below. 

General information on the status of a species should 

always be included for conservation planning (see Table 

5). Species should be characterized as part of the ranking 

system by at least some basic characteristics such as 

life-form, nativity, geographic distribution and protection 

status. Nativity would include three categories: indigenous 

(native, widespread), endemic (native, restricted 

to a subprovince), and paleoendemic (local, may 

30


be in one or more subprovinces, phylogenetically and 

geographically isolated; sensu Stebbins and Major 

1965). Geographic distribution is best characterized by 

the classification of McLaughlin (2007), who has developed 

the most current and detailed analysis of species 

distributions and floristic regionalization for the American 

Southwest. Protection status (land management 

categories) range from highly protected NGO preserves, 

many national parks, etc., through general public lands 

including wilderness, to state lands, and finally private 

lands. The categorization of Scott et al. (1993) provides 

a useful approach to categorize species. The percentage 

of the total number of populations in each of the land 

management categories should be determined, as it will 

provide useful additional information to make informed 

decisions about which species to prioritize for conservation 

funding. 

RESULTS 

If the system is to be useful, it must accurately reflect 

the status of known at-risk species. Results of scoring 

and ranking for 20 selected species using the system can 

be found in Tables 4 and 5. These scores are based primarily 

on species I have some familiarity with, and in 

some cases the ranking should be considered tentative. 

A variety of species were used as examples, ranging 

from critically endangered endemics such as Artomecon 

humilis and Ranunculus aestivalis, sparse widespread 

specialists such as Epipactis gigantea, and common endemics 

such as Chrysothamnus stylosa, to widespread 

Table 4. Selected species of the American Southwest tentatively scored with the ranking system. 

The final at-risk score (ARS) is listed. Scores in parentheses reflect uncertainty or lack of data about some aspect of 

the scoring. The ARS can vary from 0 (low at-risk) to 18 (critically endangered). 

Species Rarity Biology Trend Threats 

31 

Climate 

Change N ARS 

Arctomecon humilis 3 3 3 3 3 3 18 

Ranunculus aestivalis 3 2 3 3 3 3 17 

Ipomopsis sancti-spiritus 3 2 (3) 2 3 3 (16) 

Astragalus ampullarioides 3 3 2 2 3 3 16 

Pediocactus bradyi 3 3 2 2 3 2 15 

Puccinellia parishii 2 (3) 2 3 3 2 (15) 

Actaea arizonica 3 3 2 2 3 2 15 

Penstemon albomarginatus 3 2 2 2 2 2 13 

Camissonia atwoodii 3 3 1 1 3 1 12 

Ostrya knowltonii 2 (3) (3) 1 2 1 (12) 

Spiranthes diluvialis 2 2 2 1 3 1 11 

Cycladenia humilis 2 2 1 1 3 1 10 

Epipactus gigantea 2 3 1 2 2 0 10 

Carnegiea gigantea 1 3 1 2 2 0 9 

Cirsium rydbergii 2 2 1 1 2 1 9 

Salix gooddingii 1 2 1 2 3 0 9 

Erigeron maguirei 3 2 1 0 1 1 8 

Chrysothamnus stylosa 1 1 1 2 1 0 6 

Quercus gambelii 0 3 0 1 1 0 5 

Pinus ponderosa 0 2 0 0 0 0 2


common species like Pinus ponderosa. The conservation 

scores range from 18 for A. humilis, 17 for R. aestivalis, 

and 9 for E. gigantea, 6 for C. stylosa, and 2 for 

P. ponderosa. If the score is in parentheses, this means 

that some aspect of its threat status, biology, distribution, 

etc., is unknown. Uncertainty resulting from a lack 

of data or understanding of the species status and biology 

should be included in the ranking system (W. Fertig, 

pers. comm. 2009). When an element cannot be 

scored due to a lack of data, that element is automatically 

scored a 3, thus tending to elevate the species in 

the ranking list. 

DISCUSSION AND FUTURE WORK 

The first aspect of this approach to ranking that 

needs to be understood better is how the at-risk score 

(ARS) is distributed across a large number of rare taxa. 

Are there distinct breaks in the score that reflect underlying 

causes of rarity, or are the scores more likely to 

exhibit a continuous distribution? If breaks occur, can 

they be related to other ranking systems, such as the 

IUCN categories of critically endangered, endangered, 

and threatened? The significance of a high score is 

probably clear in this system as in others, as those species 

with high scores are likely to be already endangered. 

But we still need to know what kind of vulnerability 

to human-caused threats and climate change intermediate 

scores might reflect. Ultimately, the system 

may need to be refined and altered as ranking of species 

is completed and patterns begin to emerge. 

Since the ARS reflects a preliminary triage approach 

to “potential” vulnerability to near and mid-term threats 

(the next 10-50 years), a first use of the system would 

be to seek funding for research focused on medium to 

high-ranked species to see if in fact they are endangered 

or are likely to become endangered. Some previous 

work shows that species that initially were considered 

rare and in some cases listed under the Endangered 

Table 5. General ranking, rarity type, nativity and geographic range for selected species of the American 

Southwest. The geographic ranges are derived from McLaughlin (2007; Figure 3). 

Species Rarity ARS Score Nativity Geographic Range 

Arctomecon humilis RVIII 18 Endemic Mohave Desert 

Ranunculus aestivalis RVIII 17 Endemic Colorado Plateau 

Ipomopsis sancti-spiritus RVIII (16) Endemic S. Rocky Mountains 

Astragalus ampullarioides RVIII 16 Endemic Mohave Desert 

Pediocactus bradyi RVIII 15 Endemic Colorado Plateau 

Puccinellia parishii RVI (15) Indigenous Southwestern Region 

Actaea arizonica RVIII 15 Paleoendemic Colorado Plateau + Madrean 

Penstemon albomarginatus RVIII 13 Endemic Mohave Desert 

Camissonia atwoodii RVIII 12 Endemic Colorado Plateau 

Ostrya knowltonii RVIII (12) Paleoendemic Colorado Plateau + s NM 

Spiranthes diluvialis RVI 11 Indigenous Colorado Plateau + Great Plains 

Cycladenia humilis RIV 10 Paleoendemic Northern and Southwestern Regions 

Epipactis gigantea RVI 10 Indigenous Widespread North American 

Carnegiea gigantea RIII 9 Endemic Sonoran Desert 

Cirsium rydbergii RIV 9 Endemic Colorado Plateau 

Salix gooddingii RII 9 Indigenous Eastern and Southwestern Regions 

Erigeron maguirei RVIII 8 Endemic Colorado Plateau 

Chrysothamnus stylosa RIII 6 Endemic Colorado Plateau 

Quercus gambelii R1 5 Indigenous Southwestern Region 

Pinus ponderosa RI 2 Indigenous Northern and Southwestern Regions 

32


Species Act have after further surveys been found to be 

relatively common. A good example of this is the 

Maguire Daisy, Erigeron maguirei, which was federally 

listed as threatened prior to comprehensive surveys of 

potential habitat. It has since been found to be fairly 

common and is recommended to be upgraded to G3 

status (Clark and Tait 2007). Its ARS score (8) is lower 

some other widespread species such as Epipactus gigantea 

and Salix goodingii. 

There are several future directions that can be taken 

with this plant rarity ranking system. First, a workshop 

needs to be held where botanists with knowledge of 

various rare species get together and test the ranking 

system with a species list. Such a workshop is tentatively 

planned for 2011 to examine and finalize the 

G1G2 list of 700 species in the American Southwest. A 

second direction is for individual researchers to use the 

ranking system in their own area of expertise and determine 

how well it can be applied, how relevant it may be, 

and what refinements may be needed. It is hoped that 

readers of this paper will be interested in doing this. I 

encourage those of you who attempt to use the system to 

contact me with comments, criticisms, and suggestions. 

Finally, if in the future this ranking system, or some version 

of it, is deemed to be useful for conservation planning, 

it then needs to be incorporated into future action 

plans, regional planning efforts, conservation documents, 

and heritage databases. 

ACKNOWLEDGEMENTS 

I would like to thank Walter Fertig and Steve Caicco 

for stimulating discussions on how to rank plant rarity, 

and to Walt for reviewing an early draft of the manuscript. 

Emily Palmquist, Kyle Christy, and Meredith 

Jabis provided help in literature searches and data compilation. 


Aizen, M.A., L. Ashworth and G. Leonardo. 2002. 

Reproductive success in fragmented habitats: do compatibility 

systems and pollination specialization matter? 

Journal of Vegetation Science 13: 885-892. 

Bevill, R.I. and S.M. Louda. 1999. Comparisons of 

related rare and common species in the study of plant 

rarity. Conservation Biology 13: 493-498. 

Bond, W.J. 1994. Do mutualisms matter? Assessing 

the impact of pollinator and disperser disruption on 

plant extinction. Philosophical Transactions of the 

Royal Society of London, B. 344: 83-90. 

Brigham, C.A. and M.W. Schwartz (eds.). 2003. 

Population viability in plants. Conservation, management, 

and modeling of rare plants. Springer, NewYork. 

Buchmann, S.L. and G.P. Nabhan. 1996. The forgotten 

pollinators. Island Press, Covello, CA. 

Byers, D.L. and T.R. Meagher. 1997. A comparison 

of demographic characteristics in a rare and a common 

species of Eupatorium. Ecological Applications 7: 519- 

530. 

California Native Plant Society. 2001. Inventory of 

rare and endangered plants of California, 6 th Ed. California 

Native Plant Society Special Publication No. 1. 

Clark, D.J. and D.A. Tait. 2007. Interagency rare 

plant team inventory results – 1998 through 2003. Pp. 

32-38 In Barlow-Irick, P., J. Anderson and C. McDonald 

(eds.). Southwestern rare and endangered plants. 

Proceedings of the 4 th Conference. U.S. Forest Service 

Proceedings RMRS-P-48CD. Rocky Mountain research 

Station. Fort Collins, CO. 

Cole, C.T. 2003. Genetic variation in rare and common 

plants. Annual Review of Ecology and Systematics 

34: 213-237. 

Faber-Langendoen, D., L. Master, J. Nichols, K. 

Snow, A. Tomaino, R. Bittman, G. Hammerson, B. Heidel, 

L. Ramsay, and B. Young. 2009. NatureServe conservation 

status assessments: Methodology for assigning 

ranks. NatureServe, Arlington, VA. 42 pp. 

Ge, S., K.-Q. Wang, D.-Y. Hong, W.-H. Zhang and 

Y.-G. Zu. 1999. Comparisons of genetic diversity in the 

endangered Adenophora lobophylla and its widespread 

congener, A. potaninii. Conservation Biology 13: 509- 

513. 

Gitzendanner, M.A. and P.S. Soltis. 2000. Patterns of 

genetic variation in rare and widespread plant congeners. 

American Journal of Botany 87: 783-792. 

Harper, J.L. 1981. The meanings of rarity. Pp. 189- 

203 In Synge, H. (ed.). The biological aspects of rare 

plant conservation. J. Wiley & Sons, New York. 

Harper, K.L. 1979. Some reproductive and life history 

characteristics of rare plants and implications of 

management. Great Basin Naturalist Memoirs 3: 129- 

138. 

Krupnick, G.A. and W.J. Kress (eds.). 2005. Plant 

conservation. A natural history approach. University of 

Chicago Press, Chicago. 

Kunin, W.E. and K.J. Gaston (eds.). 1997. The biology 

of rarity: causes and consequences of rare-common 

differences. Chapman and Hall, New York. 

Kunin, W.E. and A. Schmida. 1997. Plant reproductive 

traits as a function of local, regional and global 

abundance. Conservation Biology 11: 183-192. 

Kruckeberg, A.R. and D. Rabinowitz. 1985. Biological 

aspects of endemism in higher plants. Annual Review 

of Ecology and Systematics 16: 447-479. 

Kwak, M.M. and R.M. Bekker. 2006. Ecology of 

plant reproduction: extinction risks and restoration perspectives 

of rare plant species. Pp. 362-386 In Waser, 

N.M. and J. Ollerton (eds.). Plant-pollinator interactions: 

from specialization to generalization. University 

of Chicago Press, Chicago. 

33


Lavergne, S., J.D. Thompson, E. Garnier and M. Debussche. 

2004. The biology and ecology of narrow endemic 

and widespread plants: a comparative study of 

trait variation in 20 congeneric pairs. Oikos 107: 505- 

518. 

Lesica, P. R. Yurkewycz and E.E. Crone. 2006. Rare 

plants are common where you find them. American 

Journal of Botany 93: 454-459. 

Mace, G.M. 1994. Classifying threatened species: 

means and ends. Philosophical Transactions of the 

Royal Society of London, B. 344: 91-97. 

Mace, G.M. and R. Lande. 1991. Assessing extinction 

threats: toward a reevaluation of IUCN threatened 

species categories. Conservation Biology 5: 148-157. 

McLaughlin, S.P. 2007. Tundra to tropics: The floristic 

plant geography of North America. Sida Botanical 

Miscellany No. 30. 

Miles, L., A. Grainger and O. Phillips. 2004. The 

impact of global climate change on tropical forest biodiversity 

in Amazonia. Global Ecology and Biogeography 

13: 553-565. 

Purdy, B.G., R.J. Bayer and S.E. Macdonald. 1994. 

Genetic variation, breeding system evolution, and conservation 

of the narrow sand dune endemic Stellaria 

arenicola and the widespread S. longipes (Caryophyllaceae). 

American Journal of Botany 81: 904-911. et al. 

Rabinowitz, D. 1981. Seven forms of rarity. Pp. 205- 

217 In Synge, H. (ed.). The biological aspects of rare 

plant conservation. J. Wiley & Sons, N.Y. 

Robertson, A.W., D. Kelley, J.J. Ladley, and A.D. 

Sparrow. 1999. Effects of pollinator loss on endemic 

New Zealand mistletoes (Loranthaceae). Conservation 

Biology 13: 499-508. 

Sakai, A.K., W.L. Wagner and L.A. Mehrhoff. 2002. 

Patterns of endangerment in the Hawaiian flora. Systematic 

Biology 51: 276-302. 

Schemske, D.W., B.C. Husband, M.H. Ruckelhaus, 

C. Goodwillie, I.M. Parker and J.G. Bishop. 1994. 

Evaluating approaches to the conservation of rare and 

endangered plants. Ecology 75: 584-606. 

Scott, J.M., F. Davis, B. Csuti, R. Noss, B. 

Butterfield, C. Groves, H. Anderson, S. Caicco, F. 

D’Erchia, T.C. Edwards, Jr., J. Ulliman and R.G. 

Wright. 1993. Gap analysis: a geographic approach to 

protection of biological diversity. Wildlife Monographs 

No. 123: 3-41. 

Sjöström, A. and C.L. Gross. 2006. Life-history 

characters and phylogeny are correlated with extinction 

risk in the Australian angiosperms. Journal of Biogeography 

33: 271-290. 

Sodhi, N.S., L.P. Koh, K.S.-H. Peh, H.T.W. Tan, 

R.L. Chazdon, R.T. Corlett, T.M. Lee, R.K. Colwell, 

B.W. Brook, C.H. Sekercioglu and C.J.A. Bradshaw. 

2008. Correlates of extinction proneness in tropical angiosperms. 

Diversity and Distributions 14: 1-10. 

Stebbins, G.L. and J. Major. 1965. Endemism and 

speciation in the California flora. Ecological Monographs 

35: 1-34. 

Walck, J.L., J.M. Baskin and C.C. Baskin. 1999. 

Relative competitive abilities and growth characteristics 

of a narrowly endemic and a geographically widespread 

Solidago species (Asteraceae). American Journal of 

Botany 86: 820-828. 

34


The Contribution of Cedar Breaks National Monument 

to the Conservation of Vascular Plant Diversity in Utah 

Walter Fertig and Douglas N. Reynolds, 

Moenave Botanical Consulting, Kanab, UT 

Abstract. Like most national parks in Utah, Cedar Breaks National Monument was initially established to protect its 

spectacular scenery rather than to preserve biological diversity. At less than 2500 ha, the monument is one of the 

smallest in the region and has a relatively small vascular flora of 354 documented species. Based on conventional 

measures of species richness (alpha diversity), Cedar Breaks might not seem like an important component of the protected 

area network of Utah. However, nearly 10% of the flora of Cedar Breaks is comprised of local or regional endemics 

that are mostly restricted to the Claron Formation or volcanic substrates. Many of these are rare species of 

high management interest. Our surveys in 2007-2008 documented nearly 1200 point locations for 17 of the monument’s 

rarest plants, including first records for Aster welshii and Jamesia americana var. rosea. The monument is 

especially significant in terms of beta diversity or complementarity, as it protects 63 plant species that are not otherwise 

found in NPS units in the state. As measured by an averaged Jaccard’s Coefficient of Similarity, Cedar Breaks 

National Monument has the second most unique flora among the parks in Utah. 

The Cedar Breaks Amphitheater is a large bowlshaped 

valley carved from orange and white limey sandstone 

layers of the Eocene Claron Formation on the west 

face of Cedar Mountain, about 18 miles east of Cedar 

City in southwestern Utah (Figure 1). Local Indian 

tribes called the area “the circle of painted cliffs” or the 

“place where the rocks are sliding down all the time”. 

Early Mormon settlers named it Cedar Breaks for the 

abundance of juniper (known locally as ‘cedar’) and the 

precipitous badland cliffs or breaks. President Franklin 

Roosevelt acknowledged the area’s “spectacular cliffs, 

canyons, and features of scenic, scientific, and educational 

interest” in designating Cedar Breaks as a national 

monument under the Antiquities Act in 1933 

(Evenden et al. 2002, Fertig 2009b). Administration of 

the monument was transferred from Dixie National Forest 

to the National Park Service (NPS), to be managed 

to “conserve unimpaired” the area’s natural and cultural 

resources and values “for the enjoyment of this and future 

generations” (NPS 2000). 

Protection of native biological diversity was not one 

of the rationales for creating Cedar Breaks National 

Monument, though this would eventually become an 

important part of NPS’s mandate to conserve natural 

resources. The botanical significance of the Cedar 

Breaks area was just beginning to be discovered in the 

early 1930s. Botanists George Goodman and C. Leo 

Hitchcock collected the holotypes of Breaks draba 

(Draba subalpina) and Cedar Breaks wild buckwheat 

(Eriogonum panguicense var. alpestre) from the rim of 

the Cedar Breaks Amphitheater in 1930 and Bassett 

Maguire added the holotype of Cedar Breaks daisy 

(Erigeron proselyticus, or Erigeron sionis var. trilob- 

atus) in 1934 (Fertig 2009b). These species are among a 

suite of nearly two dozen Claron formation endemics 

restricted to the Cedar Breaks area and the vicinity of 

Bryce Canyon in south-central Utah (Madsen 2001). 

Today Cedar Breaks National Monument is part of a 

network of highly protected areas that conserve biological 

diversity. This network includes other NPS units 

(national parks, monuments, recreation areas, and historic 

sites), designated wilderness areas, research natural 

areas, BLM-managed national monuments, and private 

nature preserves such as those managed by The Nature 

Conservancy. In Utah, these lands cover nearly 14% of 

the state (Prior-Magee et al. 2007). Though extensive, 

the Utah network does not yet capture a representative 

sample of the full array of the state’s biological diversity. 

Protection remains biased towards common and 

widespread species and vegetation types of low economic 

use (Fertig 2010a, Prior-Magee et al. 2007). 

The purpose of this paper is to examine the contribution 

of Cedar Breaks National Monument to the state’s 

preserve network by comparing the monument’s floristic 

composition, species richness (alpha diversity), degree 

of endemism, and number of rare species with that 

of other parklands. We hope to demonstrate that despite 

the monument’s small size, low alpha diversity, and 

relatively homogeneous vegetation, Cedar Breaks is 

significant because of its large number of plant species 

that are not protected elsewhere (i.e., the monument has 

high complementarity or beta diversity). We also hope 

to show how comparing annotated species checklists 

can be useful for identifying and prioritizing specific 

taxa that are missing or under-represented in the preserve 

network. 

35


METHODS 

From 2005-2008 we developed a revised checklist of 

the vascular plant flora of Cedar Breaks National Monument 

(Fertig 2009b, Fertig et al. 2009c). This entailed 

re-examination of over 700 specimens from the Cedar 

Breaks herbarium, as well as relevant collections housed 

at Brigham Young University, the University of Wyoming, 

and the New York Botanical Garden’s Virtual 

Herbarium. Additional species records were obtained 

through a review of previous park checklists, weed surveys, 

and vegetation studies (Buchanan 1992, Dewey 

and Andersen 2005, Jean and Palmer 1987, Roberts and 

Jean 1989, Springer et al. 2006). Field surveys were also 

conducted to corroborate unvouchered reports and in 

conjunction with a systematic rare plant inventory 

(Fertig and Reynolds 2009). The final checklist was annotated 

with information on synonyms, taxonomic problems, 

population size, growth form, global distribution, 

nativity, flowering period, and habitat. 

Eighteen rare local or regionally endemic plant species 

from Cedar Breaks National Monument and adjacent 

portions of the Ashdown Gorge Wilderness Area of 

Dixie National Forest were surveyed in 2007-2008 

(Fertig and Reynolds 2009). Populations of target species 

were mapped using a Global Positioning System 

(GPS) device and data were collected on population 

size, associated species, habitat conditions, and potential 

threats. Most species were mapped as individual points 

representing the centrum of relevé-like plots of approximately 

25-30 square meters. Populations of Arizona 

willow (Salix arizonica) occurred in sufficiently dense 

patches to be mapped as polygons. 

We assembled additional species checklists for other 

NPS units and the BLM’s Grand Staircase-Escalante 

National Monument (Figure 2, see Table 3 for citations). 

These checklists and our data for Cedar Breaks 

National Monument were compiled into a master statewide 

checklist in Microsoft excel format. The checklist 

followed the taxonomy and nomenclature of Welsh and 

others (2008). Only Utah-specific data were entered for 

those parks that crossed state lines (Dinosaur and 

Hovenweep National Monuments and Glen Canyon National 

Recreation Area). Unfortunately, complete species 

lists are not available for most wilderness areas, 

research natural areas, TNC preserves, and other highly 

protected areas, so these were excluded from the analysis. 

Simple queries were run in excel to compare overall 

species richness between parks. Jaccard’s index of similarity 

1 was calculated across pairs of parks to quantify 

beta diversity. 

1 Jaccard’s Coefficient of Similarity 

is calculated by the formula 

C/(N 1 + N 2 – C), where C = the 

number of taxa shared between two 

samples, N 1 = the number of taxa in 

sample one, and N 2 = the number of 

taxa in sample two. 

Figure 1. Cedar Breaks National 

Monument and Ashdown Gorge 

Wilderness Area, Iron County, 

Utah. Map courtesy of Zion National 

Park Resource Management 

& Research GIS. 

36


RESULTS 

Species Richness 

Based on our re-examination of herbarium specimens, 

review of literature, and new field work, 354 species 

and varieties of vascular plants (Table 1) are currently 

known from Cedar Breaks National Monument 

(Fertig 2009b, Fertig et al. 2009c). At least 74 of these 

taxa have been discovered since 2005. Overall, the 

monument’s flora has increased by 21% since Roberts 

and Jean (1989) reported 277 species for the area. The 

Figure 2. National Park Service units and BLMmanaged 

national monuments in Utah. Park acronyms: 

ARCH (Arches National Park), BRCA (Bryce Canyon 

National Park), CANY (Canyonlands National Park), 

CARE (Capitol Reef National Park), CEBR (Cedar 

Breaks National Monument), DINO (Dinosaur National 

Monument), GCNRA (Glen Canyon National Recreation 

Area), GOSP (Golden Spike National Historic 

Site), GSENM (Grand Staircase-Escalante National 

Monument), HOVE (Hovenweep National Monument), 

NABR (Natural Bridges National Monument), RABR 

(Rainbow Bridge National Monument), TICA 

(Timpanogos Cave National Monument), ZION (Zion 

National Park). Map courtesy of Zion National Park 

Resource Management & Research GIS. 

flora of Cedar Breaks contains 9.7% of the 3659 native 

and naturalized plant species documented in Utah by 

Welsh and others (2008) and 36% of the 156 reported 

families. 

Native species account for 94.9% of the flora of Cedar 

Breaks National Monument (336 species). Nearly 

82% of the species range widely across western North 

America and are common in Utah (Table 1). Eighteen 

taxa are categorized as local endemics that occupy a 

total area of less than 16,500 square kilometers and are 

restricted to the immediate vicinity of Cedar Breaks or 

adjacent high plateaus of south-central Utah (mostly the 

Tushar Range and Paunsaugunt Plateau) (Fertig 2009b). 

An additional 20 taxa are found only in the Colorado 

Plateau area of southern Utah, northeastern Arizona, 

northwestern New Mexico, and southwestern Colorado. 

Together these 38 local and regional endemics account 

for 10.7% of the monument’s flora. Just over 2% of the 

flora consists of species that occur sporadically across 

Utah (sparse taxa) or have populations in Cedar Breaks 

that are widely isolated from their main, contiguous 

range (disjunct) (Table 1). 

Only 18 introduced species (those not historically 

native to Utah or North America) have become established 

in the monument, representing 5.1% of the total 

flora (Table 1). The percentage of introduced species at 

Cedar Breaks is less than half that reported for the entire 

flora of Utah (13.5%) (Fertig 2007, Welsh et al. 

2008). None of the introduced plant taxa in the monument 

are listed by the state of Utah as official noxious 

weeds. 

Rare Species 

The Conservation Data Center (CDC) of the Utah 

Division of Wildlife Resources (1998) recognizes 22 

species from Cedar Breaks National Monument as species 

of concern (Table 2). Most of these are local or 

regional endemics restricted to the Claron Formation or 

species that are widespread outside of Utah but have 10 

or fewer extant populations in the state. We consider 

two additional species from the monument to be deserving 

of recognition by the CDC. Madsen’s daisy 

(Erigeron vagus var. madsenii) is a southern Utah endemic 

that was only described as a new taxon in 2008 

(Welsh et al. 2008). Rosy cliff jamesia (Jamesia americana 

var. rosea) was not recognized as occurring in the 

state of Utah until we verified populations in Cedar 

Breaks and the Ashdown Gorge Wilderness Area in 

2008. Previously, populations of this taxon were 

thought to represent var. zionis, a local endemic of 

southern Utah listed by the CDC as a species of concern 

and by the US Forest Service and BLM as sensitive. 

Var. rosea was formerly known only from California 

and Nevada (Fertig and Reynolds 2009, Holmgren and 

Holmgren 1989). In all, seven species from Cedar 

37


Table 1. Statistical Summary of the Flora 

of Cedar Breaks National Monument* 

Category 

Taxonomic Diversity 

Species & 

Varieties 

Full Species 

Only 

No. of Taxa 

Confirmed 

Present 

No. of Taxa 

Reported 

(not confirmed) 

Total 

347 7 354 

335 5 340 

Families 56 0 56 

Biogeographic Diversity 

Introduced 17 1 18 

Native 330 6 336 

Locally 

Endemic 

Regionally 

Endemic 

18 0 18 

20 0 20 

Disjunct 2 0 2 

Peripheral 0 0 0 

Sparse 6 0 6 

Widespread 284 6 290 

The number of taxa and plant families is based on Welsh and 

others (2008). Biogeographic diversity categories refer to the 

distribution of a species within Utah and the state’s contribution 

to its overall global range (Fertig 2009b). Introduced taxa 

are not native to Utah or North America but have become 

naturalized (breeding on their own without human assistance). 

Local Endemics have their entire global range restricted 

to an area of less than 16,500 square km (ca 6370 sq 

miles, or 1 degree of latitude x 2 degrees of longitude). Regional 

Endemics have global ranges of 16,500-250,000 

square km (an area about the size of Wyoming). Disjuncts are 

isolated from the contiguous portion of their range by a gap 

of more than 800 km (ca 500 miles). Peripherals are widespread 

globally but occur at the margin of their contiguous 

range in Utah and occupy less than 5% of the state’s area 

(usually only within a few miles of the state border). Sparse 

taxa occur widely across Utah or North America but their 

range within Utah is small and patchy, with populations restricted 

to specialized or uncommon habitats. Widespread 

taxa have global ranges exceeding 250,000 square km and 

occur over at least 10% of the state. 

* See Addendum for additional species documented since 

2009. 

Breaks National Monument are presently listed as sensitive 

by the US Forest Service or BLM and 14 were once 

candidates for listing as Threatened or Endangered under 

the US Endangered Species Act (Table 2). 

During 2007-2008 we targeted 16 of Cedar Breaks’ 

rarest local endemics and CDC species of concern 

(Table 2) for survey. The number of target species increased 

to 18 with the discovery of extant populations of 

Madsen’s daisy and the first records of Welsh’s aster 

(Aster welshii) for the monument. We recorded at least 

one population of 16 of the target species at 546 different 

sampling points within Cedar Breaks National 

Monument or the Ashdown Gorge Wilderness Area. 

Since more than one target species was often present at 

each location, we actually documented 1181 different 

sample points for these species. For the clonal species 

Salix arizonica we delineated 16 discrete polygons in 

three main population clusters that cover a total area of 

just over one hectare (Fertig and Reynolds 2009). 

Ten of our target species occurred in over 10% of our 

samples. These species were found mostly on the red or 

white limey-sandstone layers of the Claron Formation 

along the rim and slopes of the Cedar Breaks Amphitheater. 

Cedar Breaks wild buckwheat (Eriogonum panguicense 

var. alpestre) was the most widespread and 

abundant of the rare species, being found in 36% of all 

samples and having a population estimated at 35,200- 

100,000 individuals (Fertig and Reynolds 2009). This 

plant also has the smallest geographic range of any 

taxon in our study, being known only from the Cedar 

Breaks area within the monument and the adjacent 

Dixie National Forest and Ashdown Gorge Wilderness. 

Only three other species were estimated to have populations 

of over 10,000 plants: Least lomatium (Lomatium 

minimum), Markagunt aster (Aster wasatchensis var. 

wasatchensis), and Least spring-parsley (Cymopterus 

minimus). The least abundant and most restricted species 

in the study area were Rosy cliff jamesia (known 

from only about 100 plants in two main areas; Madsen’s 

daisy (approximately 400 plants in three main areas), 

Reveal’s paintbrush (Castilleja parvula var. revealii 

with 500 plants in two main populations), Podunk 

groundsel (Senecio malmstenii with about 1500 individuals 

in five sites), and Welsh’s aster (with an estimated 

1700 plants scattered along Ashdown and Rattle 

creeks in the bottom of the amphitheater). 

We were unsuccessful in relocating just one of the 18 

target species, the Zion draba (Draba asprella var. zionensis). 

This species is known from a single herbarium 

specimen (Dickman s.n. CEBR) collected from “Cedar 

Breaks National Monument” in 1977. Unfortunately, 

nothing more precise is known about the original collection 

site. Zion draba occurs commonly in Zion National 

Park on Navajo Sandstone cliffs and canyons. Comparable 

Navajo Sandstone outcrops are not exposed at Cedar 

38


Table 2. Vascular Plant Species of Concern of Cedar Breaks National Monument 

Family 

Apiaceae 

(Umbelliferae) 

Apiaceae 


Asteraceae 

(Compositae) 

Asteraceae 

(Compositae) 

Asteraceae 

(Compositae) 

Asteraceae 

(Compositae) 

Asteraceae 

(Compositae) 

Asteraceae 

(Compositae) 

Asteraceae 

(Compositae) 

Asteraceae 

(Compositae) 

Asteraceae 

(Compositae) 

Asteraceae 

(Compositae) 

Brassicaceae 

(Cruciferae) 

Species/ 

Common Name 

Cymopterus minimus 

Least spring-parsley 

Lomatium minimum 

Least lomatium 

Agoseris glauca var. agrestis* 

Field agoseris 

Antennaria pulcherrima* 

Showy pussytoes 

Aster wasatchensis var. 

wasatchensis 

Markagunt aster 

Aster welshii 

Welsh’s aster 

Erigeron sionis var. trilobatus 

Cedar Breaks daisy 

Erigeron vagus var. madsenii 

Madsen’s daisy 

Haplopappus zionis 

Cedar Breaks goldenbush 

Machaeranthera commixta* 

Bigelow’s aster 

Senecio malmstenii 

Podunk groundsel 

Townsendia montana var. 

minima 

Bryce Canyon townsendia 

Draba asprella var. zionensis 

Zion draba 

TNC Rank Legal & UTCDC 

Status 

G1G2Q/ 

S1S2 

G3/S3 

USFS: Sensitive 

USFWS: former C2 

UTCDC: Rare 

USFWS: former 3C 

UTCDC: Watch 

G5T5/S1S2 UTCDC: Taxonomic 

Problems 

G5?/S1 

Abundance in CEBR 

Ca 12,500 plants, in 

35% of sample plots 

Ca 33,600 plants, in 

13.2% of sample plots 

Not known 

UTCDC: Peripheral Not known 

G2/S2 UTCDC: Watch Ca 15,100 plants in 


G2/S2 UTCDC: Watch Ca 1700 plants in 4.6% 

of sample plots 

G2/S2 


UTCDC: Watch 

Ca 4200 plants in 


G4T?/SNR Ca 4000 plants in 2% 


G2/S2 

G4G5T3?/ 

S3? 

G1/S1? 

G4T3/S3 

UT BLM: Sensitive 


UTCDC: Watch 

UTCDC: Watch 



UTCDC: Rare 


UTCDC: Watch 

G4T3?/S3? USFWS: former 3C 

UTCDC: Watch 



Not known 

Ca 1500 plants in 4% 




Not known, may be 

falsely reported 

Brassicaceae 

(Cruciferae) 

Draba subalpina 

Breaks draba 

G3/S3 


UTCDC: Watch 



Brassicaceae 

(Cruciferae) 

Caryophyllaceae 

Equisetaceae 

Physaria rubicundula var. 

rubicundula* 

Breaks bladderpod 

Silene petersonii 

Peterson’s campion 

Equisetum variegatum* 

Northern scouring-rush 

G3//S3 


UTCDC: Watch 

G2G3/S2S3 USFS: Sensitive; 


UTCDC: Watch 

G5/S1 

Not known 


12.1% of plots 

UTCDC: Peripheral Not known 

39



Family 

Species/ 

Common Name 

TNC Rank Legal & UTCDC 

Status 

Abundance in CEBR 

Fabaceae 

(Leguminosae) 

Hydrangeaceae 

(Saxifragaceae) 

Polemoniaceae 

Polygonaceae 

Pyrolaceae 

(Ericaceae) 

Salicaceae 

Scrophulariaceae 

(Orobanchaceae) 

Astragalus limnocharis var. 

limnocharis 

Navajo Lake milkvetch 

Jamesia americana var. rosea 

Rosy jamesia 

Ipomopsis tridactyla 

Tushar gilia 

Eriogonum panguicense var. 

alpestre 

Cedar Breaks wild buckwheat 

Pyrola picta* 

White-veined wintergreen 

Salix arizonica 

Arizona willow 

Castilleja parvula var. revealii 

Reveal’s paintbrush 

G2T1/S1 





G5T3/SNR Ca 100 plants in 1.1% 


G5T2/S2 UTCDC: Rare Ca 2800 plants in 9.3% 


G3T2T3Q/ 

SSYN 

G4G5/S1 

G2G3/S2 

G2/S2 


UTCDC: Taxonomic 

Problems 

35,200-100,000 plants 

estimated, in 36% of 

sample plots 

UTCDC: Infrequent Not known 


USFWS: former 

Candidate 

UTCDC: Rare 



16 small to medium 

sized clones in 3 main 

population clusters 

covering 1.06 ha 

Ca 500 plants, in 4.8% 


*Species not surveyed in 2007-2008. 

Derived from Utah Department of Wildlife Resources (1998), Fertig (2009b), Fertig and Reynolds (2009), and Fertig and others 

(2009c). Codes: TNC rank assesses abundance and conservation priority on a scale of 1-5 (1 being extremely vulnerable and 5 

being secure) for full species (G) and varieties or subspecies (T) across their entire range and within each state (S). A “?” indicates 

uncertainty in the rank, Q = taxonomic questions, NR = not ranked, and SYN = species is considered a synonym and not 

ranked under the given name. Under legal status, USFWS = US Fish and Wildlife Service. C2 = category 2 candidate (a former 

category used for taxa that might warrant being proposed for Threatened or Endangered status following additional research). 3C 

= category 3 candidates (species dropped from consideration for listing). USFS = US Forest Service. BLM = Bureau of Land 

Management. UTCDC status includes conservation categories adopted by the state natural heritage program to prioritize endemic 

and rare plant taxa (Utah Division of Wildlife Resources 1998). These categories include: Rare (plants with rangewide viability 

concerns), Watch (regional endemics without rangewide viability concerns), Peripheral (rare or uncommon in Utah, but more 

common rangewide), Infrequent (plants occur infrequently over western US), and Taxonomic Problems (validity of species, subspecies, 

or variety has been questioned). 

Breaks National Monument, though similar sandstone 

cliffs of the Straight Cliffs or Wahweap formations are 

present in the bottom of Ashdown Canyon. These areas 

were searched in 2008, but no populations were found. 

It is possible that the label of the specimen is erroneous 

and the collection was actually made in Zion National 

Park (Fertig and Reynolds 2009). 

Similarity to Other Park Floras 

Cedar Breaks National Monument ranks tenth out of the 

14 NPS and BLM managed parklands considered in this 

study in both area and species richness (Table 3). As of 

2008, Grand Staircase-Escalante National Monument is 

the largest protected area at over 761,000 ha and has the 

highest number of vascular plant species with 999 2 . In 

general, vascular plant species richness is positively correlated 

with total area, with the main exception being 

Zion National Park with the second highest number of 

species (991) but ranking sixth in total area. When park 

2 More than 10 new species were documented in Zion National 

Park in 2009, allowing it to pass Grand Staircase-Escalante 

National Monument as the Utah parkland with the highest 

vascular plant species richness. See Fertig and others (2012) 

for updates on the flora of each NPS unit assessed in this 

study. 

40


Table 3. Number of Vascular Plant Taxa in National Park Service and Bureau of Land 

Management Parks, Monuments, Historic Sites, and Recreation Areas in Utah 

Management Area Size (ha) Total # 

of Taxa 

# of Rare 

Taxa a 

# of 

Endemic 

Taxa 

# of 

Unique 

Taxa b 

# of Taxa 

per log 

(area) 

Source 

Arches NP 30,966 524 32 82 12 50.7 Fertig et al. 2009a, 

2009c 

Bryce Canyon NP 14,502 587 51 71 28 61.3 Fertig and Topp 

2009 

Canyonlands NP 136,610 600 55 107 16 50.8 Fertig et al. 2009b, 

2009c 

Capitol Reef NP 97,895 888 62 142 90 77.3 Fertig 2009a 

Cedar Breaks NM 2,491 354 24 38 63 45.3 Fertig 2009b, Fertig 

et al. 2009c 

Dinosaur NM 85,096 757 (485 

in UT) 

Glen Canyon NRA 505,868 889 (863 

in UT) 

80 76 69 (UT) 66.7 Fertig 2009c, Fertig 

et al. 2009c 

60 176 73 (UT) 67.7 Hill 2005, Spence 

2005 

Golden Spike NHS 1,107 149 6 6 20 21.3 Fertig 2009d, Fertig 

et al. 2009c 

Grand Staircase- 

Escalante NM 

761,070 999 142 193 68 73.8 Fertig 2005 & unpublished 

data 

Hovenweep NM 318 340 (240 

in UT) c 17 31 6 59.0 Fertig 2009e 

Natural Bridges NM 3,009 428 21 72 6 53.4 Fertig 2009f 

Rainbow Bridge NM 65 224 12 33 4 53.7 Fertig 2010b 

Timpanogos Cave NM 101 235 8 15 39 50.9 Fertig and Atwood 

2009 

Zion NM 59,900 991 189 133 199 90.1 Fertig and Alexander 

2009 

a Derived from Utah Division of Wildlife Resources (1998) 

B 

Defined as species found in only one of the 14 parklands considered here 

c Erroneously reported as 214 taxa in Utah in Fertig (2009e) 

41


size is normalized by taking the natural log of area, 

Zion National Park emerges as the largest flora in the 

study area at 90.1 species/ln(area), followed by Capitol 

Reef National Park and Grand Staircase-Escalante 

National Monument (Table 3). Cedar Breaks National 

Monument drops to thirteenth in alpha diversity at 

45.3 species/ln(area) when area is normalized, exceeding 

only Golden Spike National Historic Site. 

In our sample, Cedar Breaks National Monument 

shares the most vascular plant taxa in common with 

Bryce Canyon National Park (227 species), Zion National 

Park (202 species), and Grand Staircase- 

Escalante National Monument (172 species) (Table 4). 

Not surprisingly, these are the three parklands closest 

in proximity to Cedar Breaks (Figure 2). The monument 

shares the fewest species in common with 

Golden Spike National Historic Site (29 species), 

Rainbow Bridge National Monument (32 species), and 

the Utah sections of Hovenweep National Monument 

(37 species). Factoring in discrepancies in the relative 

sizes of different park floras using Jaccard’s Coefficient 

of Similarity (JCS), Cedar Breaks National 

Monument is still most similar to Bryce Canyon National 

Park (JCS = 0.318), but is significantly less 

similar to Zion National Park and Grand Staircase- 

Escalante National Monument (JCS = 0.177 and 

0.146, respectively). Timpanogos Cave National 

Monument, which shared only 75 species in common 

with Cedar Breaks, has a Jaccard’s coefficient equal to 

that of Grand Staircase, despite being much farther 

distant. Based on the Jaccard’s Coefficient, Cedar 

Breaks National Monument is least similar to Rainbow 

Bridge National Monument and Golden Spike 

National Historic Site (Table 4). Cedar Break’s average 

coefficient of similarity among the other thirteen 

parklands in the study is 0.128. This is the second 

lowest value among all the parks analyzed, and is 

bested only by Golden Spike National Historic Site 

(average JCS = 0.121). By comparison, Canyonlands 

National Park (average JCS = 0.326) and Grand Staircase-Escalante 

National Monument (average JCS = 

0.312) share the most species on average with other 

Utah parklands. 

The low Jaccard’s Coefficient value for Cedar 

Breaks is a consequence of the area’s low overall species 

richness and the relatively high number of plant 

taxa that are unique to the monument. Of the 354 taxa 

documented for Cedar Breaks National Monument, 63 

species (17.8% of the local flora) are not found in any 

of the other protected areas in this study (Table 3). 

Only Zion National Park has a higher percentage 

(19.9%) of its flora that is protected nowhere else in 

the state. Ten of the 24 rare species of Cedar Breaks 

are protected only in the monument and another ten 

occur just in Cedar Breaks and Bryce Canyon National 

Park. 

DISCUSSION 

Creating a protected area network that contains full 

representation of the entire array of native biological 

diversity is one of the core principles of contemporary 

conservation biology (Groves et al. 2002, Margules and 

Pressey 2000, Margules and Sarkar 2007, Scott et al. 

1993, Stein et al. 2000). The foundation of such a network 

already exists in Utah, consisting of national 

parks, monuments, recreation areas, historic sites, designated 

wilderness, research natural areas, national 

wildlife refuges, and Nature Conservancy preserves. 

Unfortunately, many of the best protected sites in the 

state were originally created for their scenic and historic 

values or recreation potential rather than the preservation 

of biological diversity. Significant gaps remain in 

the state’s protected area network, especially for many 

rare and endemic species and plants from lowland habitats 

that have been largely converted to agriculture or 

residential use (Fertig 2010a, Prior-Magee et al. 2007). 

Building a modern preserve network is expensive 

and difficult, requiring costly land acquisition, changes 

in economic or regulatory policy, and expenditure of 

political capital. Because of these costs, it is critical that 

conservation strategies be both effective and efficient 

(Margules and Sarkar 2007, Stein et al. 2000). To maximize 

efficiency, conservation biologists often focus 

their efforts on preserving hotspots of unusually high 

species richness, representative examples of major 

vegetation types (that serve as surrogates for all biodiversity), 

and areas with high complementarity. Hotspots 

and vegetation types are especially useful for capturing 

broad swaths of diversity when building networks from 

scratch, but can become less efficient as preserve systems 

grow and gaps become more obvious (Williams et 

al. 1996). With its emphasis on species that tend to be 

rare and localized, complementarity can be an effective 

alternative to hotspots and vegetation types when filling 

specific holes in the network (Margules and Sarkar 

2007, Williams et a. 1996). 

Cedar Breaks National Monument has been part of 

the Utah protected area network for over 75 years. But 

if Cedar Breaks were not already protected, would it be 

a worthy addition to the network? In terms of overall 

species richness, the answer at first glance appears to be 

no. The monument contains only 354 vascular plant 

species, compared to the average of 558 taxa for the 

other parklands considered in this study. The area’s low 

alpha diversity can be attributed to the monument’s 

small size and relatively homogeneous vegetation 

(especially compared to larger parks in the Colorado 

Plateau portion of the state). Relative to other parks, 

42


Cedar Breaks also has a fairly low number of endemic 

and rare species. 

It is in terms of complementarity that Cedar Breaks 

makes its primary contribution to the state’s protected 

area network. Jaccard’s Coefficient of Similarity is a 

useful formula for measuring complementarity as it 

weights similarities in floristic composition between 

areas by discrepancies in total species richness. With an 

average JCS of 0.128, Cedar Breaks National Monument 

has the most dissimilar flora of any parkland considered 

in this study other than Golden Spike National 

Historic Site. The monument’s low JCS score is a consequence 

of its relatively high number of unique plant 

species (63 taxa or nearly 18% of the local flora) that 

are not protected elsewhere in the state. While many of 

these unique species are Claron endemics, the majority 

are taxa restricted to elevations above 3000 meters. The 

average elevation of Cedar Breaks is 2800 meters, a figure 

that exceeds the highest elevation of all other parks 

in the Utah reserve network. The 14 parklands in the 

preserve network analyzed here protect 2007 of Utah’s 

3659 native and naturalized vascular plant species 

(54.8% of the total state flora). Cedar Breaks’ unique 

contributions account for 3.1% of the total. Ten of the 

24 rare plant species recognized for Cedar Breaks and 9 

of the 38 local or regional endemics are among the species 

protected nowhere else. 

The flora of Cedar Breaks is most similar to that of 

Bryce Canyon National Park, with the two parks sharing 

227 plant taxa. This degree of similarity is not surprising 

given the proximity of the areas and their similar 

elevation, vegetation, and geology (both parks have extensive 

outcrops of reddish-orange Claron Formation 

badlands). Although there is redundancy in the protection 

of some species in both parks, this is not necessarily 

disadvantageous, as having multiple populations in the 

protected area network can reduce the risk of catastrophic 

loss and enhance the preservation of multiple genotypes 

(Noss and Cooperrider 1994, Margules and Sarkar 

2007). Similar redundancy occurs among common montane 

forest and meadow species found in both Cedar 

Breaks and Timpanogos Cave national monuments, 

which despite their great distance from each other share 

similar vegetation types and a relatively high Jaccard’s 

Coefficient of Similarity. 

Based on our analysis of vascular plant checklists 

from selected parklands, nearly 45% of the native and 

naturalized flora of Utah is not represented in the state’s 

existing preserve network. Some of the omissions may 

be an artifact of incomplete sampling or data synthesis, 

especially of wilderness areas, research natural areas, 

national wildlife refuges, private nature preserves, and 

other protected areas that could not be included in this 

analysis for lack of data. Other holes will only be filled 

by targeting specific plant taxa identified by analyzing 

system-wide complementarity. Fortunately, many of the 

missing species occur in specific geographic areas or 

habitat types which are, themselves, poorly represented 

in the current network. Fertig (2010a) identified just 12 

geographic areas that, if protected, would capture 70% 

of the missing plant species in the preserve network on 

Utah’s Colorado Plateau. These are mostly “hotspots” 

of unprotected endemism and include the La Sal, Abajo, 

Henry, Tushar, Boulder, and Pine Valley mountains, 

Book Cliffs, Tavaputs, and Fish Lake plateaus, Uinta 

Basin, Sevier Valley, and San Rafael Swell. Statewide, 

important gaps also exist in the Great Basin, lowland 

riparian areas, and the foothills and montane zones of 

northern mountains. Future additions to the network 

may well be sites like Cedar Breaks that are of relatively 

small size or modest species richness but which have 

significant beta diversity. 

For any reserve network to function, it will be increasingly 

important to keep score of what species are 

represented, how many populations are captured, and 

whether these populations are of sufficient size or quality 

to persist. Annotated species lists and databases of 

distributions are critical tools for identifying gaps in the 

reserve network. Efficient planning and implementation 

of a complete reserve system requires that intelligent 

choices be made in selecting geographic areas and species 

to target for inclusion. Measuring complementarity, 

as we have done for Cedar Breaks and other parklands, 

is a key step in the prioritization process. 


We would like to thank Paul Roelandt, Superintendent 

of Cedar Breaks National Monument, for encouragement 

and support; Cynthia Wanschura and Matt Betenson, 

formerly of the Zion National Park GIS lab, for 

help with maps; and Amy Tendick and Sarah Topp of 

the NPS Northern Colorado Plateau Network for sharing 

data on new species occurrences. Thanks also to our 2 

anonymous reviewers for helpful suggestions. 

REFERENCES 

Buchanan, H. 1992. Wildflowers of southwestern 

Utah. Bryce Canyon Natural History Association and 

Falcon Press, Helena, MT. 119 pp. 

Dewey, S. and K. Andersen. 2005. An inventory of 

invasive non-native plants in Cedar Breaks National 

Monument (2004): Final report. Utah State University. 

Weed Science Research Project #SD0515A. 

Evenden, A., M. Miller, M. Beer, E. Nance, S. Daw, 

A. Wight, M. Estenson, and L. Cudlip. 2002. Northern 

Colorado Plateau Vital Signs Network and Prototype 

Cluster, Plan for Natural Resources Monitoring: Phase 1 

Report, October 1, 2002 [two volumes]. National Park 

Service, Northern Colorado Plateau Network, Moab, 

UT. 138 pp. + app. 

43


Fertig, W. 2005. Annotated checklist of the flora of 

Grand Staircase-Escalante National Monument. Moenave 

Botanical Consulting, Kanab, UT. 54 pp. 

Fertig, W. 2007. Introduced and naturalized plants of 

Utah. Sego Lily 30(5):7-11. 

Fertig, W. 2009a. Annotated checklist of vascular 

flora: Capitol Reef National Park. Natural Resource 

Technical Report NPS/NCPN/NRTR-2009/154. 172 pp. 

Fertig, W. 2009b. Annotated checklist of vascular 

flora: Cedar Breaks National Monument. Natural Resource 

Technical Report NPS/NCPN/NRTR-2009/173. 

92 pp. 

Fertig, W. 2009c. Annotated checklist of vascular 

flora: Dinosaur National Monument. Natural Resource 

Technical Report NPS/NCPN/NRTR-2009/225. 182 pp. 

Fertig, W. 2009d. Annotated checklist of vascular 

flora: Golden Spike National Historic Site. Natural Resource 


50 pp. 

Fertig, W. 2009e. Annotated checklist of vascular 

flora: Hovenweep National Monument. Natural Resource 


112 pp. 

Fertig, W. 2009f. Annotated checklist of vascular 

flora: Natural Bridges National Monument. Natural Resource 


96 pp. 

Fertig, W. 2010a. Finding gaps in the protected area 

network in the Colorado Plateau: A case study using 

vascular plant taxa in Utah. In: Van Riper III, C., B.F. 

Wakeling, and T.D. Sisk, eds. The Colorado Plateau IV: 

Shaping Conservation Through Science and Management. 

University of Arizona Press, Tucson, AZ. Pp 

109-119. 

Fertig, W. 2010b. The flora of Rainbow Bridge National 

Monument. Sego Lily 33(6):4-14. 

Fertig, W. and J. Alexander. 2009. Annotated checklist 

of vascular flora: Zion National Park. Natural Resource 


196 pp. 

Fertig, W. and N.D. Atwood. 2009. Annotated 

checklist of vascular flora: Timpanogos Cave National 

Monument. Natural Resource Technical Report NPS/ 

NCPN/NRTR-2009/167. 76 pp. 

Fertig, W. and D.N. Reynolds. 2009. Survey of rare 

plants of Cedar Breaks National Monument: Final report. 

CPCESU Cooperative Agreement #H1200-004- 

0002. Colorado Plateau Cooperative Ecosystem Studies 

Unit and Southern Utah University. 93 pp. 

Fertig, W. and S. Topp. 2009. Annotated checklist of 

vascular flora: Bryce Canyon National Park. Natural 

Resource Technical Report NPS/NCPN/NRTR- 

2009/153. 130 pp. 

Fertig, W., S. Topp, and M. Moran. 2009a. Annotated 

checklist of vascular flora: Arches National Park. 

Natural Resource Technical Report NPS/NCPN/NRTR- 

2009/220. 117 pp. 

Fertig, W., S. Topp, and M. Moran. 2009b. Annotated 

checklist of vascular flora: Canyonlands National 

Park. Natural Resource Technical Report NPS/NCPN/ 

NRTR-2009/221. 130 pp. 

Fertig, W., S. Topp, M. Moran, and R. Weissinger. 

2009c. New vascular plant species discoveries in the 

Northern Colorado Plateau Network: 2008 update. Moenave 

Botanical Consulting, Kanab, UT. 24 pp. 

Fertig, W., S. Topp, M. Moran, T. Hildebrand, J. Ott, 

and D. Zobell. 2012. Vascular plant species discoveries 

in the Northern Colorado Plateau Network: Update 

for 2008-2011. Natural Resources Technical Report 

NPS/NCPN/NRTR –2012/582. 

Groves, C.R., D.B. Jensen, L.L. Valutis, K.H. Redford, 

M.L. Shaffer, J.M. Scott, J.V. Baumgartner, J.V. 

Higgins, M.W. Beck, and M.G. Anderson. 2002. Planning 

for biodiversity conservation: Putting conservation 

science into practice. Bioscience 52(6):499-512. 

Hill, M. 2005. Vascular flora of Glen Canyon National 

Recreation Area, Utah and Arizona. Master’s 

thesis, Northern Arizona University, Flagstaff, AZ. 

Holmgren, N.H. and P.K. Holmgren. 1989. A taxonomic 

study of Jamesia (Hydrangeaceae). Brittonia 41 

(4):335-350. 

Jean, C. and B. Palmer. 1987. Vascular plant species 

list, Cedar Breaks National Monument. 8 pp. 

Madsen, M. 2001. A floristic study of the Tertiary 

Claron Formation in the Paunsaugunt region of southern 

Utah. Master’s thesis, Department of Botany and Range 

Science, Brigham Young University, Provo, UT. 

Margules, C.R. and R.L. Pressey. 2000. Systematic 

conservation planning. Nature 405:243-253. 

Margules, C.R. and S. Sarkar. 2007. Systematic Conservation 

Planning. Cambridge University Press, Cambridge, 

UK. 270 pp. 

National Park Service (NPS). 2000. Management 

policies. US Department of Interior, National Park Service, 

Washington, DC. Publication # NPS DI416. 137 

pp. 

Noss, R.F. and A.Y. Cooperrider. 1994. Saving Nature’s 

Legacy: Protecting and Restoring Biodiversity. 

Island Press, Washington, DC. 416 pp. 

Prior-Magee, J.S., K.G. Boykin, D.F. Bradford, W.G. 

Kepner. J.H. Lowry, D.L. Schrupp, K.A. Thomas, and 

B.C. Thompson, editors. 2007. Southwest Regional Gap 

Analysis Project Final Report. US Geological Survey, 

Gap Analysis Program, Moscow, ID. 

Roberts, D.W. and C. Jean. 1989. Plant community 

and rare and exotic species distribution and dynamics in 

Cedar Breaks National Monument. Department of Forest 

Resources and Ecology Center, Utah State University, 

Logan, UT. 144 pp. 

44


Scott, J.M., F. Davis, B. Csuti, R. Noss, B. 

Butterfield, C. Groves, H. Anderson, S. Caicco, F. 

D’Erchia, T.C. Edwards, J. Ulliman, and G. Wright. 

1993. Gap analysis: a geographic approach to protection 

of biological diversity. Wildlife Monographs 123. 41 

pp. 

Spence, J.R. 2005. Notes on significant collections 

and additions to the flora of Glen Canyon National Recreation 

Area, Utah and Arizona, between 1992 and 

2004. Western North American Naturalist 65(1):103- 

111. 

Springer, A.E., L.E. Stevens, and R. Harms. 2006. 

Inventory and classification of selected National Park 

Service springs on the Colorado Plateau. Northern Arizona 

University, Flagstaff, AZ. NPS Cooperative 

agreement # CA 1200-99-009; Task # NAU-118. 

Stein, B.A., L.S. Kutner, and J.S. Adams. 2000. Precious 

Heritage, the Status of Biodiversity in the United 

States. The Nature Conservancy and Association for 

Biodiversity Information, Oxford University Press, New 

York. 399 pp. 

Utah Division of Wildlife Resources. 1998. Inventory 

of sensitive species and ecosystems in Utah. Endemic 

and rare plants of Utah: An overview of their distribution 

and status. Report prepared for the Utah Reclamation 

Mitigation and Conservation Commission and 

US Department of the Interior. 566 pp. + app. 

Welsh, S.L., N.D. Atwood, S. Goodrich, and L.C. 

Higgins. 2008. A Utah Flora, 2004-2008 summary 

monograph, fourth edition, revised. Brigham Young 

University Print Services, Provo, UT. 1019 pp. 

Williams, P., D. Gibbons, C. Margules, A. Rebelo, 

C. Humphries, and R. Pressey. 1996. A comparison of 

richness hotspots, rarity hotspots and complementarity 

areas for conserving diversity using British birds. Conservation 

Biology 10:155-174. 

Addendum 

Since this paper was written in the summer of 2009, 31 new vascular plant taxa have been collected or reported for 

Cedar Breaks National Monument, increasing the flora to 385 species and varieties (Fertig et al. 2012). Among the 

more uncommon or noteworthy additions to the monument's flora are Botrychium lunaria (sparse in Utah and the first 

species of Ophioglossaceae for Cedar Breaks), Cirsium clavatum var. clavatum (a regional endemic, first collected in 

1982 and located in a search of specimens at the Brigham Young University herbarium), and Penstemon caespitosus 

var. suffruticosus (a local endemic of southern Utah). 

In recent years, local political leaders in Iron County, Utah, have proposed changing Cedar Breaks from a national 

monument to a national park and having the park annex the Ashdown Gorge Wilderness on its western boundary. I 

conducted a floristic survey of the Ashdown Gorge Wilderness in the summer of 2009 (Fertig 2009g) and documented 

308 vascular plant taxa. Of these species 247 were previously known from Cedar Breaks National Monument, 

while 61 were new species for the local area. If the Ashdown Gorge Wilderness Area were added to Cedar 

Breaks National Monument the total flora would increase to 426 taxa (Fertig 2009g). 

Fertig, W. 2009g. Vascular plant flora of the Ashdown Gorge Wilderness Area and additions to the flora of Cedar 

Breaks National Monument. Moenave Botanical Consulting, Kanab, UT. 45 pp. 

45


Studying the Seed Bank Dynamics of Rare Plants 

Susan E. Meyer, 

USDA Forest Service Shrub Lab, Provo, UT 

Abstract. Seed bank studies are important for understanding the population biology of rare plants, especially shortlived 

species from unpredictable environments. Seed retrieval studies are used to determine how long seeds persist in 

the soil and how seed dormancy changes through time. Even short-term studies (three to five years) can help determine 

whether the seed bank is transient or persistent and characterize the shape of the seed depletion trajectory. Laboratory 

germination studies can provide clues about seed bank persistence, and in situ seed bank studies can quantify 

seed bank size, but only retrieval studies provide the information needed to inform population viability analysis. 

In order to assess the status of a rare plant population, 

botanists need some way to measure and quantify 

demographic (life history) variables. This can be as 

straightforward as quantifying the number of plants present 

in an area using a simple monitoring protocol, or it 

can involve classifying the individuals present by age, 

size or stage (for example, seedling, juvenile, vegetative 

adult, reproductive adult). It can also include an evaluation 

of the potential contribution of an individual to the 

next generation, that is, its reproductive output in terms 

of numbers of seeds or vegetative ramets produced and 

the quality of those seeds or ramets. If the plants are 

permanently marked and their attributes quantified on 

multiple occasions, it becomes possible to perform 

quantitative life history analysis (Brigham and Schwartz 

2003). One of the goals of life history analysis for plants 

of conservation concern is the formal inclusion of life 

history information into mathematical models that predict 

the probability of population persistence under different 

scenarios. The term used for the creation, execution, 

and interpretation of such models is population 

viability analysis (PVA; Beissinger and McCullough 

2002). 

Population viability analysis for plants is a relatively 

new area of research (Doak et al. 2002). One feature of 

plant life history that makes PVA difficult is the presence 

of a largely invisible life cycle stage, namely seeds 

in the soil seed bank. This paper deals both with assessing 

the need for including soil seed bank data in life history 

analysis for a particular plant species and tackling 

the problem of obtaining seed bank data when necessary. 

WHEN ARE SOIL SEED BANKS IMPORTANT? 

Doak and others (2002) provide an excellent discussion 

of problems associated with assumptions about 

seed banks in the context of PVA for plants. They first 

present the two extremes for plant life history strategy. 

At one extreme is a life history much like that of the 

vertebrates for which PVA was first developed. In these 

organisms, for example, giant sequoias, newborns are 

subject to high and variable mortality, the juvenile phase 

lasts a long time and is characterized by slow growth 

and increasing annual survival, and adults are longlived. 

Plants with this life history usually produce seeds 

that have a very short tenure in the seed bank, less than 

a year. At the other extreme are plants that have short 

and risky juvenile stages and short life spans as adults, 

but whose seeds have high dormancy and high survivorship 

in the soil, resulting in a large, persistent seed bank. 

For plants in the former category, the seed bank can 

safely be ignored in PVA, because the transition from 

seed production to seedling takes place within a single 

year. But for plants that produce long-lived seeds, the 

dynamics of the seed bank can play a very important 

role in population biology, and false assumptions based 

on inadequate data can lead to major problems with the 

resulting PVA. 

Doak and others (2002) identify two plant life history 

attributes most critical for deciding how important accurate 

measurement of seed bank dynamics is likely to be. 

The first, as implied above, is plant longevity, and the 

second is the effect of a variable environment on adult 

survival and reproduction. A long and stable mature 

stage with multiple opportunities for reproduction is 

likely to minimize the importance of a persistent seed 

bank. On the other hand, short-lived plants which are at 

variable risk of mortality and/or reproductive failure as 

adults are likely to be associated with long-lived seeds 

and persistent seed banks. Different combinations of 

positions along these two gradients (life span and environmental 

risk to adults) result in different assessments 

of the importance of seed banks. Even relatively longlived 

shrubs may have persistent seed banks if there is a 

risk of catastrophic mortality to adult plants, for example 

through fire or epidemic disease. And even annuals 

may have relatively short-lived seeds if the chances of 

adult survival to seed production are high. 

Another source of clues about the importance of the 

persistent seed bank is in the germination behavior of 

46


the seeds in a laboratory setting. If the seeds are nondormant 

at dispersal and cannot readily be induced into 

dormancy, it is unlikely that they will be able to form a 

persistent seed bank under field conditions. Similarly, if 

the seeds are dormant at dispersal but lose dormancy in 

response to an environmental cue likely to be encountered 

before the optimum germination time within the 

year, they are also unlikely to form a persistent seed 

bank. This type of dormancy is called cue-responsive or 

predictive dormancy. It functions to time germination 

optimally within the year following production by allowing 

the seeds to sense their environment and respond 

appropriately. Many spring-germinating species in the 

temperate zone have this type of seed dormancy, with 

moist-chilling or cold stratification that simulates winter 

conditions as the cue. This prevents precocious germination 

the autumn following production but allows complete 

germination the following spring. 

Not all cue-responsive dormancy is associated with 

short-lived seed banks, however. Sometimes the cue is 

associated with an episodic event such as fire (Keeley 

1987), or tillage that exposes weed seeds to light 

(Baskin and Baskin 1985). Such seeds may persist in the 

soil for many years, but will germinate synchronously 

when the cue is received. These seeds germinate readily 

in response to a laboratory-administered cue. The trick 

is in recognizing that such a cue is unlikely to be encountered 

under field conditions within any particular 

year, so that seeds with this kind of cue-responsive dormancy 

form a persistent seed bank. 

The best laboratory clue that seeds of a species are 

likely to form a persistent seed bank under field conditions 

is the presence of cue-non-responsive dormancy. 

These seeds will germinate to only small percentages no 

matter what dormancy-breaking treatment is applied. 

Sometimes it is possible to determine what pretreatment 

is needed to make the seeds become responsive to a particular 

cue, but more often even this is very difficult. 

The individual seeds are programmed to come out of 

dormancy at different times over a protracted period, 

and there seems to be no way to speed or circumvent 

this process. Often the only way to break cue-nonresponsive 

dormancy is to resort to unnatural treatments 

like injuring the seed, and sometimes even these draconian 

measures fail to trigger germination. 

METHODOLOGIES FOR STUDYING SEED 

BANKS 

For many people, the most obvious way to begin a 

study of seed bank dynamics is to attempt to quantify 

the in situ seed bank. This involves taking seed bank 

samples from the field and somehow measuring the 

number of seeds these samples contain. A common 

measuring method involves spreading the soil samples 

out in shallow pans, applying water, and counting and 

removing germinants as they emerge. Usually the soil 

sample is turned multiple times to encourage subsequent 

flushes of germination and emergence, and the seed 

bank is considered depleted when no further emergence 

occurs. Obviously, this methodology involves numerous 

assumptions about the dormancy status of the seeds, 

because seeds that do not germinate and emerge as seedlings 

are not included in the quantification. Sometimes 

the seedling emergence methodology includes multiple 

cycles of application of dormancy-breaking cues, for 

example, moist-chilling, which increases the chances of 

complete germination. But for truly cue-non-responsive 

species, these methodologies are clearly inadequate. 

Another commonly applied method for quantifying 

the in situ seed bank involves flotation, usually using a 

chemical such as potassium carbonate at high molarity. 

This method has been shown to yield more seeds on 

average than the emergence (germination) methodology 

(Ishikawa-Goto and Tsuyuzaki 2004), but the extracted 

seeds are no longer viable. This inability to distinguish 

between viable and nonviable seeds and to evaluate seed 

dormancy status represents a major disadvantage to this 

method. 

A third method for quantifying in situ seed banks is 

rather labor-intensive, but avoids some of the problems 

associated with the previous two methods (Meyer et al. 

2007). The samples are dry-screened (or wet-screened, 

depending on the soil type) using sieve sizes that eliminate 

fine soil and large particles such as gravel and root 

chunks. The remaining fraction, which contains the 

seeds of interest, is hand-processed to remove the seeds, 

which can then be subjected to germination testing and/ 

or viability evaluation. This method works best for medium 

to large seeds. Its accuracy is increased by inclusion 

of numerous small samples rather than fewer large 

samples, given an equal volume of sampled material. 

Sampling regime is a critical aspect of in situ seed 

bank evaluation, because the lateral distribution of seeds 

in soil is usually extremely heterogeneous. This means 

that stratified sampling regimes and often very large 

sample sizes are needed to get replicable data. It is 

highly advisable before launching into such a study to 

make sure that it is well-designed and will yield the desired 

information. In order to quantify the persistent 

seed bank, it is important to sample after germination is 

complete for the year but before any input of seed rain 

from current-year production, so that only seeds at least 

a year old will be sampled. 

Sampling the in situ seed bank cannot provide any 

information about seed bank persistence beyond a single 

year, because there is no way to know the age of seeds 

removed from the samples. Seeds from the previous 

production year could be the only ones present prior to 

dispersal of current-year seeds, or there could be an accumulation 

of seeds from an unknown number of prior 

47


production years. In situ sampling provides a quantification 

of the seed bank at a given point in time, but does 

not directly address seed bank dynamics. 

The method of choice for determining how long 

seeds can persist in the seed bank is the seed retrieval 

(seed burial or artificial seed bank) experiment. In this 

method, seeds of known age are introduced into the seed 

bank in a form that makes their subsequent retrieval and 

evaluation possible. This usually involves placing them 

in some sort of mesh bag that allows free passage of 

water and air, but not seeds. We usually use nylon mesh 

bags made from mosquito netting, which has a fine 

enough mesh size to contain even very small seeds. For 

larger seeds, fiberglass window screening works well. 

The nylon lasts for many years as long as it is covered 

with soil so that it cannot photo-degrade, and fiberglass 

screening can be placed directly on the surface. 

Some have criticized seed retrieval experiments because 

they place the seeds in an environment somewhat 

different from that experienced by seeds in the natural 

seed bank. An alternative method, or one that can be 

used in conjunction with a retrieval experiment, is an 

emergence experiment, where seeds of known age are 

planted in precise locations using a template or field 

marking system and monitored for emergence. For both 

retrieval and emergence experiments, it is important to 

include adequate replication and to use block designs to 

avoid localized effects that generate large experimental 

error. 

The retrieval method involves destructive sampling, 

so that it is necessary to include a different set of retrieval 

bags for each evaluation date. For rare plants, 

this can involve a prohibitively large number of seeds. 

But even yearly retrievals for three years, using a few 

hundred seeds, can provide valuable information that is 

difficult to obtain in any other way. 

Processing retrieval bags can be a zen experience, 

not suitable for those inclined to attention deficit. Particularly 

when the samples are wet after a germination 

event, it can take some time to untangle and quantify the 

germinants. Remaining ungerminated seeds are then 

incubated under controlled conditions and classified as 

germinable, dormant, or nonviable. With frequent sampling 

and adequate replication, it is possible to get a 

very good picture of the phenology of dormancy loss 

and germination, and also secondary dormancy induction, 

if it occurs. 

It is important to distinguish if possible between 

seeds that are lost from the seed bank through germination 

and those that are lost through pre-germination 

mortality. These two processes have potentially very 

different demographic consequences. Getting accurate 

information on field-germinated seeds requires retrieval 

soon after the germination event, and it can be difficult 

to anticipate the correct retrieval timing. Laboratory data 

can be helpful here. For example, if you know the seeds 

can germinate at near-freezing temperatures, it is reasonable 

to expect them to germinate during winter in 

places where the snow cover is deep enough to insulate 

the seed bed from freezing. This is a common strategy 

in environments where the soil dries quickly in the 

spring. 

PATTERNS OF SEED BANK DEPLETION 

Seed retrieval data can be examined graphically by 

plotting the percentage of initially viable seeds still present 

as viable seeds in the seed bank (actually, in the 

artificial seed bank) as a function of time in the field. 

For any given seed population, this will yield a characteristic 

seed depletion trajectory through time (fig. 1). 

For species that do not form persistent seed banks, this 

trajectory will essentially be a line that goes to a value 

of zero within a year of the initiation of the retrieval experiment 

(Figure 1a). Species with this type of seed depletion 

trajectory are said to have transient seed banks. 

In situ seed bank sampling after germination is complete 

but before current-year seed rain for species with transient 

seed banks is expected to yield no viable seeds. 

Basin wildrye (Leymus cinereus) is an example of a 

species with nondormant seeds and a transient seed 

bank (Meyer et al. 1995; Figure 2). Seed collections 

from three populations were placed in retrieval experiments 

in early fall at salt desert shrubland, sagebrush 

steppe, and mountain meadow sites. Germination did 

not take place in autumn, even though seeds were nondormant, 

because of their relatively long germination 

time. All the seeds germinated by the end of the following 

spring, regardless of population of origin or habitat 

at the seed retrieval site. 

Blackbrush (Coleogyne ramosissima) is another example 

of a species with a transient seed bank (Meyer 

and Pendleton 2004; Figure 3). Its seeds are dormant at 

dispersal in mid-summer but have cue-responsive dormancy, 

losing dormancy both in the dry state at high 

summer temperatures and during moist chilling. As a 

consequence, if winter rainfall is adequate, seeds germinate 

to high percentages during early spring. A few 

seeds may carry for a year over if the winter is unusually 

dry, but this is the exception rather than the rule. 

Seeds with cue-responsive dormancy that responds to 

a disturbance-associated cue show a similar pattern to 

those that form transient seed banks, except that there is 

an extended time delay prior to the seed bank-depleting 

germination event (Figure 1 b). These species are difficult 

to study in retrieval experiments, because the cue, 

such as fire or soil disturbance, is not only unpredictable 

but also would tend to destroy a retrieval experiment. 

Information on this type of seed bank depletion trajectory 

generally comes either from emergence experiments 

or from in situ seed bank studies that show abrupt 

48


100 

80 

A 

100 

80 

B 

60 

60 

Viable Seed Percentage 

40 

20 

0 

0 1 2 3 4 5 6 7 8 9 10 

100 

80 

60 

40 

20 

C 

40 

20 

0 

0 1 2 3 4 5 6 7 8 9 10 

100 

80 

60 

40 

20 

D 

0 

0 

0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 7 8 9 10 

Retrieval Year 

Figure 1. Schematic seed bank depletion trajectories for: (A) a species with a transient seed bank and a depletion trajectory 

that reaches zero within one year, (B) a species with cue-responsive dormancy and a seed bank that persists 

until a specific dormancy-breaking cue is received, (C) a species with a short-persistent seed bank, a constant loss 

rate, and a negatively exponential depletion trajectory, and (D) a species with cue-non-responsive seeds and a linear 

seed bank depletion trajectory. 

decreases in seed density following receipt of the germination 

cue. 

In many herbaceous perennial species, most of the 

seeds are programmed to germinate during the first year 

following production, but some possess a mechanism 

permitting carryover for at least a year, even when conditions 

for dormancy release and germination the first 

year are optimal. These species are said to have shortpersistent 

seed banks. This germination response pattern 

results in a seed depletion trajectory that is essentially 

negatively exponential (Figure 1c). If rate of loss of a 

cohort of seeds in the seed bank is constant across years, 

this is the type of seed depletion trajectory that will be 

generated. For example, if 80% of the seeds are lost the 

first year, 80% of the remaining 20% are lost the second 

year, and 80% of the remaining 4% are lost the third 

year, this would generate a negatively exponential loss 

trajectory. 

Lewis flax (Linum lewisii) is an example of a species 

that may exhibit a negatively exponential loss trajectory 

(Meyer and Kitchen 1994; Figure 4). Seed dormancy 

loss and germination phenology in this species and its 

close relative L. perenne (Euopean blue flax) also vary 

as a function of both population of origin and habitat at 

the seed retrieval site. In a two-year retrieval experiment 

at three sites, the “Appar” release of European blue flax 

had nondormant seeds that formed only a transient seed 

bank, regardless of retrieval site habitat. In contrast, the 

montane Strawberry seed collection of Lewis flax was 

largely dormant at the initiation of the retrieval experiment 

and required chilling to become nondormant 

(Figure 4). It lost dormancy and germinated completely 

by the end of the first spring at its site of origin in the 

mountains, exhibiting the transient seed bank pattern. It 

carried over a substantial fraction through the end of the 

second year at the foothill and especially the salt desert 

site. Seeds placed outside of their environmental context 

can show very different germination patterns than those 

placed in the habitat of origin. These seeds did not receive 

sufficient chilling to break dormancy at the drier 

sites and tended to form a persistent seed bank under 

those conditions. 

The foothill Provo Overlook seed collection of Lewis 

flax was nondormant at the initiation of the retrieval 

experiment, but contained a fraction that could be induced 

into secondary dormancy early during chilling the 

first year (Figure 4). This resulted in a divergence of 

seed sub-populations, so that a sizeable fraction germi- 

49


VIABLE SEED PERCENTAGE 

100 

80 

60 

40 

20 

0 

100 

80 

60 

40 

20 

0 

100 

80 

60 

40 

20 

0 

PINTO COLLECTION 

9/20 10/2011/2012/20 1/20 2/20 3/20 4/20 5/20 6/20 

INDIAN CANYON COLLECTION 

9/20 10/2011/2012/20 1/20 2/20 3/20 4/20 5/20 6/20 

STRAWBERRY COLLECTION 

9/20 10/2011/2012/20 1/20 2/20 3/20 4/20 5/20 6/20 

DATE 

WHITEROCKS DESERT SITE 

POINT OF THE MOUNTANI FOOTHILL SITE 

STRAWBERRY MONTANE SITE 

Figure 2. Seed bank depletion trajectories for three collections 

of basin wildrye (Leymus cinereus) placed in 

retrieval experiments at montane, foothill, and salt desert 

sites. All seeds germinated within a year of initiation 

of retrieval experiments. Adapted from Meyer and 

others (1995). 

nated the first year in all three habitats, but 20-30% remained 

in dormancy through the spring and carried over 

to the second year, generating the characteristic negatively 

exponential pattern. Three additional years of retrieval 

data for the Provo Overlook collection at these 

sites showed continued gradual decline in viable seed 

numbers each year (Meyer unpublished data). 

Seed bank depletion for shadscale (Atriplex confertifolia) 

exhibited a pattern somewhat similar to that of 

Lewis flax, but in this case the pattern resulted in a more 

persistent seed bank (Figure 5). There was no germination 

at all during the first year in the field, but a sizeable 

germination pulse occurred the second spring after retrieval 

initiation. During eight subsequent years in the 

field, the size of the remaining viable seed fraction 

slowly diminished until it was at or near zero for most 

collections (Figure 5; Meyer unpublished data). Laboratory 

germination experiments help to explain this pattern. 

Shadscale seeds are dormant at dispersal and tend 

to be nonresponsive to the chilling cue that triggers 

them to emerge at the correct time for establishment in 

early spring (Meyer et al. 1998). They must after-ripen 

in the dry state to become responsive to the chilling cue, 

and the rate of increase in the chilling-responsive fraction 

is an exponential function of temperature during 

dry storage (Garvin and Meyer 2003). This delays germination 

for at least a year, because the high summer 

temperatures needed for after-ripening are first experienced 

only after the first winter in this autumn-ripening 

species. Each summer another fraction becomes chilling-responsive, 

and this fraction is able to germinate the 

following spring if its chilling requirement is met. The 

resulting depletion trajectory is like a slow-motion version 

of the negative exponential pattern for species with 

short-lived seed banks. Once the first pulse of germination 

is complete, the trajectory becomes essentially linear 

through time. 

In contrast to the slow-exponential depletion trajectory 

seen for shadscale, species with truly cue nonresponsive 

dormancy tend to exhibit a linear depletion 

trajectory, with no large germination pulse in any one 

year and certainly not in the early years (Figure 1d). 

Hard-seeded legumes like Utah ladyfinger milkvetch 

(Astragalus utahensis) are typical of this group. Hardseededness 

refers to the inability of a seed to imbibe 

water because of physical barriers imposed by the seed 

coat or endocarp. It is often possible to trigger synchronous 

germination in hard-seeded species by injuring the 

seed coat, and it has been thought that such scarification 

must be a part of the natural dormancy-breaking regimen. 

Under field conditions, however, hard-seededness 

seems to be very gradually lost through time without 

any specific scarification mechanism. Because the seeds 

lose hard-seededness at different rates, this functions to 

spread germination over many years. 

50


PERCENTAGE 

OF INITIALLY VIABLE SEEDS 

100 

75 

50 

25 

0 

7-91 10-91 1-92 4-92 7-92 

RETRIEVAL DATE 

% DORMANT SEEDS 

% VIABLE UNGERMINATED SEEDS 

% VIABLE + GERMINATED SEEDS 

Figure 3. Change through time in the dormant seed fraction, the viable seed fraction, and the viable plus germinated 

seed fraction for blackbrush (Coleogyne ramosissima) seeds placed in a retrieval experiment in the habitat of origin at 

Arches National Park. All seeds germinated within a year of initiation of the experiment soon after dispersal. 

Adapted from Meyer and Pendleton (2004). 

Figure 4. Change through time during a two-year period for dormant seed percentage, viable ungerminated seed percentage, 

and viable plus germinated seed percentage for three collections of Linum placed in retrieval experiments at 

montane (Strawberry), foothill (Hobble Creek), and salt desert (Rush Valley) study sites. Adapted from Meyer and 

Kitchen (1994). 

51


PERCENTAGE (BASED ON INITIALLY VIABLE SEEDS) 

100 

80 

60 

40 

20 

0 

100 

80 

60 

40 

20 

0 

100 

80 

60 

40 

20 

0 

A 

8-COLLECTION MEAN 

11/91 3/92 7/92 11/92 3/93 7/93 11/93 3/94 7/94 11/94 3/95 7/95 11/95 3/96 7/96 

B 

PIPE 

SPRINGS 

11/91 3/92 7/92 11/92 3/93 7/93 11/93 3/94 7/94 11/94 3/95 7/95 11/95 3/96 7/96 

C 

NORTH OF MILFORD 

11/91 3/92 7/92 11/92 3/93 7/93 11/93 3/94 7/94 11/94 3/95 7/95 11/95 3/96 7/96 

RETRIEVAL DATE 

DORMANT NOT CHILL-RESPONSIVE 

DORMANT 

DORMANT+GERMINABLE 

DORMANT+GERMINABLE+FIELD-GERMINATED 

Figure 5. Patterns of change in the precentage of non-chill-responsive dormant seeds, chill-responsive dormant 

seeds, dormant plus germinable seeds and viable plus germinated seeds over a five-year period for: (A) The mean of 

eight seed collections, (B) the Pipe Springs seed collection, and (C) the north of Milford seed collection. Adapted 

from Meyer and others (1998). 

52


Two or three years of retrieval data are usually sufficient 

to determine whether a species will show a negatively 

exponential or a linear depletion trajectory. When 

Utah ladyfinger milkvetch seeds were placed into retrieval 

experiments at three contrasting sites, depletion 

trajectories were clearly linear at all three sites (Figure 

6). The slope of the depletion trajectory was quite similar 

across habitats and showed no clear pattern as a 

function of habitat, indicating that rate of loss of hardseededness 

was not tightly tied to environmental conditions. 

For A. utahensis, regression equations based on the 

first two years of retrieval data at each site were able to 

predict the approximate size of the remaining fraction in 

the subsequent four years. These equations were also 

used to estimate maximum longevity of this seed population 

in the seed bank, which was about 14 years at the 

montane site, 10 years at the foothill site, and 12 years 

at the salt desert site. Including the later retrievals in 

these regressions did not change them significantly, 

even though the fit of the lines for these later dates was 

not as good. This retrieval experiment had only two replications 

per retrieval date, resulting in considerable error 

in the estimate of hard-seededness, especially in later 

years, when values dropped far below 100%. The regression 

equation may be the best indicator of actual 

rate of seed bank depletion under this scenario. 

100 

Rush Valley - Desert Study Site 

80 

60 

40 

Percentage of Initially Viable Seeds 

20 

0 

100 

80 

60 

40 

20 

0 

100 

80 

Hard Seeds 

Hard + Germinable Seeds 

9-90 3-91 9-91 3-92 9-92 3-93 9-93 3-94 9-94 3-95 9-95 3-96 9-96 

Hobble Creek - Foothill Study Site 

9-90 3-91 9-91 3-92 9-92 3-93 9-93 3-94 9-94 3-95 9-95 3-96 9-96 

Strawberry - Montane Study Site 

60 

40 

20 

0 

9-90 3-91 9-91 3-92 9-92 3-93 9-93 3-94 9-94 3-95 9-95 3-96 9-96 

Date 

Figure 6. Patterns of change over a six-year period in the percentage of hard seeds and hard plus germinable seeds 

(total viable seeds) for a collection of Utah ladyfinger milkvetch (Astragalus utahensis) placed in seed retrieval experiments 

at three sites. Regression lines are fit to total viable seed percentage values at each site based on the first 

two years of retrieval (Rush Valley site: Percentage of viable seeds = -0.022 (days)+100.4, d.f.=13, R 2 = 0.794, 

P


All of the species discussed so far are common plants 

that were included in seed bank studies in order to understand 

their establishment ecology, rather than to provide 

seed bank data to inform PVA. There are very few 

long-term seed retrieval studies for rare plants. The 

Snake River Plains endemic Lepidium papilliferum is 

one of the few rare species whose seed bank dynamics 

have been included in PVA (Meyer et al. 2005, 2006). 

This spring ephemeral species has cue non-responsive 

seeds that show the characteristic linear decrease as a 

function of time in the proportion of initially viable 

seeds remaining in the seed bank (Figure 7; Meyer et al. 

2005). There was little or no germination during the first 

two years of this eleven-year retrieval study, so that at 

least three years of retrieval data would have been 

needed to estimate the slope of seed bank depletion. The 

maximum longevity in soil for two seed collections in- 

cluded in the study was estimated to be 12 years. Seeds 

of this species can be induced to germinate by piercing 

imbibed seeds and subjecting them to 2-4 weeks of 

moist chilling. Development of this technique has facilitated 

greenhouse production of seeds for reintroduction 

experiments, but seems to shed little light on how the 

seeds gradually become nondormant in the field. 

Rare plants can be expected to have seeds that run 

the gamut of germination response patterns and associated 

seed bank depletion trajectories. As discussed 

above, it is frequently important to investigate this aspect 

of rare plant population biology, especially when 

the goal is to understand the likely future status for a 

population or species. The information obtained from 

seed bank studies can also be critical for planning management 

and mitigation activities. 

% OF INITIALLY VIABLE SEEDS 

100 

80 

60 

40 

20 

0 

100 

80 

60 

40 

20 

0 

1991 SEEDS 

8/92 8/93 8/94 8/95 8/96 8/97 8/98 8/99 8/2000 8/2001 8/2002 8/2003 

1992 SEEDS 

DORMANT 

DORMANT PLUS GERMINABLE 

DORMANT PLUS GERMINABLE PLUS GERMINATED 

TOTAL I NITIALLY VIABLE 

8/92 8/93 8/94 8/95 8/96 8/97 8/98 8/99 8/2000 8/2001 8/2002 8/2003 

DATE 

Figure 7. Patterns of change through time in dormant seed percentage, dormant plus germinable seed percentage, and 

viable plus germinated seed percentage over an eleven-year period in a seed retrieval experiment with two collections 

of Lepidium papilliferum. The difference between viable plus germinated seed percentage and initially viable seed 

percentage represents seeds that lost viability prior to germinating. Adapted from Meyer and others (2005). 

54



Baskin, J.M., and C.C. Baskin. 1985. The annual 

cycle in buried weed seeds: a continuum. BioScience 

35:492-998. 

Beissinger, S.R and D.R. McCullough (eds). 2002. 

Population Viability Analysis. University of Chicago 

Press, Chicago, Illinois. 577 p. 

Brigham, C.A, and M.W. Schwartz, eds. 2003. Population 

viability in plants. Conservation, management 

and modeling of rare plants. Ecological Studies, 

Vol.165. Springer, Berlin. 362 p. 

Doak, D.F., D. Thompson, and E.S. Jules. 2002. 

Population viability analysis for plants: understanding 

the demographic consequences of seed banks for population 

health. p.312-337. In: Beissinger, S. R and D. R. 

McCullough (eds). Population Viability Analysis. University 

of Chicago Press, Chicago, Illinois. 

Garvin, S.C., and S.E. Meyer. 2003. Multiple mechanisms 

of seed dormancy regulation in shadscale 

(Atriplex confertifolia: Chenopodiaceae). Canadian 

Journal of Botany 81:301-610. 

Ishikawa-Goto, M., and S. Tsuyuzaki. 2004. Methods 

of estimating seed banks with reference to longterm 

seed burial. Journal of Plant Research 117: 245- 

248. 

Keeley, J.E. 1987. Role of fire in seed germination of 

woody taxa in California chaparral. Ecology 68: 434- 

443. 

Meyer, S.E., J. Beckstead, P.S. Allen, and H. Pullman. 

1995. Germination ecophysiology of Leymus 

cinereus (Poaceae). International Journal of Plant Sciences 

156: 206-215. 

Meyer, S.E., S.L. Carlson, and S.C. Garvin. 1998. 

Seed germination regulation and field seed bank carryover 

in shadscale (Atriplex confertifolia: Chenopodiaceae). 

Journal of Arid Environments 38: 255-267. 

Meyer, S.E. and S.G. Kitchen. 1994. Life history 

variation in blue flax (Linum perenne: Linaceae): Seed 

germination phenology. American Journal of Botany 81: 

528-535. 

Meyer, S.E. and B.K. Pendleton. 2005. Factors affecting 

seed germination and seedling establishment of a 

long-lived desert shrub (Coleogyne ramosissima: 

Rosaceae). Plant Ecology 178:171–187. 

Meyer, S.E., D. Quinney, D.L. Nelson, and J. 

Weaver. 2007. Impact of the pathogen Pyrenophora 

semeniperda on Bromus tectorum seedbank dynamics in 

North American cold deserts. Weed Research 47: 54– 

62. 

Meyer , S.E, D. Quinney, and J. Weaver. 2005. A life 

history study of the Snake River Plains endemic 

Lepidium papilliferum (Brassicaceae). Western North 

American Naturalist 65:11-23. 

Meyer, S.E., D. Quinney, and J. Weaver. 2006. A 

stochastic population model for Lepidium papilliferum 

(Brassicaceae), a rare desert ephemeral with a persistent 

seed bank. American Journal of Botany. 93: 891–902. 

55


East Meets West: Rare Desert Alliums in Arizona 

John L. Anderson, 

US Bureau of Land Management, Phoenix, AZ, retired 

Abstract. Two previously poorly known desert species of Allium in Arizona, A. bigelovii and A. parishii, were investigated 

to determine their actual rarity and conservation status. Each approached Arizona from different deserts 

and opposite directions: Allium bigelovii from the Chihuahuan Desert in New Mexico and Allium parishii from the 

Mohave Desert in California. The status of Allium bigelovii was made difficult to understand by the many misidentifications 

of herbarium specimens. A review of herbarium specimens and field visits determined that Allium bigelovii is 

a Chihuahuan Desert species that enters southeastern Arizona and is disjunct farther west into the Sonoran Desert on 

unusual lacustrine soils. Allium parishii was only known from historic collections in two mountain ranges in western 

Arizona. These historic locations were relocated. Although these two Allium species normally occur nearly 500 km 

apart and in different deserts, they approach within less than 100 km in the Sonoran Desert of western Arizona. 

There are 96 species of Allium L. in North America 

north of Mexico with thirteen species in Arizona 

(McNeal and Jacobsen 2002). The taxonomy of Allium 

in Arizona is largely unchanged since the monograph of 

Ownbey (1947). Three of the species recognized there 

are now treated as varieties: Allium nevadense S. Wats. 

var. cristatum (S. Wats.) Ownbey as A. atrorubens S. 

Wats. var. cristatum (S. Wats.) D. McNeal; A. palmeri 

Wats. as A. bisceptrum S. Wats. var. palmeri (Wats.) 

Cronquist; and A. rubrum Osterhout as A. geyeri S. 

Wats. var. tenerum Jones. The thirteen Arizona species 

are rather equally divided through the diverse Arizona 

habitats of deserts and mountains, aridlands and wetlands, 

low and high elevations, and northern and southern 

floristic affinities. Two of the desert species, Allium 

bigelovii S. Wats. (Figure 1) and A. parishii S. Wats. 

(Figure 2), have been poorly known in Arizona, but for 

different reasons. The taxonomic identity of the former, 

and consequently its real range and habitat in Arizona, 

has been confused by the many misidentifications of 

herbarium specimens; and, the geographic status of the 

latter in Arizona was made unclear by its few historic 

records in the state. 

The range of Allium bigelovii was described by Ownbey 

(1947) as “…southwestern New Mexico, northwestward 

across central Arizona to Mohave County.” In Arizona 

he cited five historic collections: (isotype, Palmer 

532 NY 1876 (Figure 3); Rusby 839 NY 1883; Crooks 

et al ARIZ 1939; Crooks & Darrow ARIZ 1938; and 

Benson & Darrow POM 1941), all from central Arizona 

(Figure 4). The type collection of A. bigelovii is from 

Cook’s Springs (Bigelow s.n.) in southwestern New 

Mexico (Watson 1871). Several collections at RSA 

(Eastwood 8276, Greene s.n.., Jones s.n.) and at NY 

(Greene s.n., Rusby s.n.., and Holmgren 6891) document 

its historic occurrence in southwestern New Mexico. 

Sivinski (2003) described the habitat and range of 

Figure 1. Allium bigelovii. 

Figure 2. Allium parishii. 

56


Figure 4. Map of the Arizona distribution of Allium 

bigelovii based on Ownbey (1947). 

Figure 3. First collection of Allium bigelovii in Arizona, 

Palmer 532 Walnut Grove, Yavapai County. 

Allium bigelovii in New Mexico as “…a desert species 

of southwestern New Mexico”, therefore, a Chihuahuan 

Desert species. At the start of this study twenty six collections 

at Arizona herbaria (ARIZ, ASU, ATC, Grand 

Canyon National Park (GC), Museum of Northern Arizona 

(MNA), and BLM Safford [BLMS]), had been 

identified or annotated as A. bigelovii (Table 1). Based 

upon these annotations, A. bigelovii was widely variable 

in habitat from desert to chaparral, grassland, oak woodland, 

pinyon-juniper woodland, and ponderosa pine and 

wide ranging geographically from southeastern Arizona 

westward beyond Wickenburg into the Sonoran Desert 

and northwestward to the Hopi lands and the Grand 

Canyon on the Colorado Plateau (Figure 5). This situation 

made its identity as a species and its natural habitat 

in Arizona puzzling. 

By examining these collections, I determined that 

over half had been misidentified and only twelve were 

actually Allium bigelovii (Table 1). The misidentified 

collections were annotated by me to several other species, 

the majority (8) as A. bisceptrum var. palmeri, 

three as A. macropetalum, two as A. acuminatum, and 

one as A. atrorubens var. cristatum (Table 1). The large 

number of misidentifications of Allium bigelovii as A. 

bisceptrum var. palmeri was due to their pairing in the 

key in Kearney and Peebles (1960) and the vague species 

differentiation based on qualitative characters there. 

To make species determinations I used the two species’ 

descriptions in Ownbey (1947) that included quantitative 

morphological characters as well as information on 

herbarium labels of habitat and geographic data. The 

basis for species definition was thus an evolutionary 

combination of morphology and ecology, a species’ 

physical characters, and the niche a taxonomic entity 

occupies in nature. The refinement of the identity of A. 

bigelovii in Arizona combined with a knowledge of its 

habitat and range in New Mexico demonstrated that A. 

bigelovii is a Chihuahuan Desert species (Figure 6) that 

extends from southwestern New Mexico into southeastern 

Arizona (Figure 7) (Gunder AZ930-8 ASU; Lunt 6 

BLMS). 

The localities of the remaining Arizona collections of 

Allium bigelovii followed an interesting disjunct pattern 

of distribution that extended the range of A. bigelovii 

discontinuously on Mid-Late Tertiary lacustrine deposits 

across central Arizona. This relictual “steppingstone” 

pattern had been documented by me (Anderson 

1996) for many species of various floristic affinities including 

other Chihuahuan Desert species: Anulocaulis 

leisolenus (Torrey) Standl., Polygala scoparioides Chodat., 

and Thamnosma texana (A. Gray) Torrey. The 

lacustrine deposits (Nations et al. 1982) containing A. 

57


Table 1. Author annotations of specimens at Arizona herbaria: ARIZ, AU, ATC, BLM Safford (BLMS), 

Grand Canyon National Park (GC), and Museum of Northern Arizona (MNA), previously identified as Allium 

bigelovii . 

Allium bigelovii Allium acuminatum Allium atrorubens 

var cristatum 

ARIZ: Crooks s.n. 

1938; Crooks s.n. 

1939; Darrow 10906. 

ASU: Butterwick 

4504, 6165; Gunder 

s.n. ATC: Kierstad 80 

-1; Morefield 1324; 

Llambreschte 30. 

BLMS: Lunt s.n. 

MNA: Haskell & 

Deaver 2449; Wetherill 

s.n. 

ARIZ: Reichenbacher 

1391. ASU: Lehto 

21378. 

Allium bisceptrum 

var. palmeri 

GC: Stochert s.n. ARIZ: Fishbein 304; 

Warren 248. ATC: 

Kasch s.n. 1979, s.n. 

1980; Llamphear s.n.; 

Reese s.n. BLMS: 

Bingham 3225. MNA: 

Kewanwytewa s.n. 

Allium macropetalum 

ARIZ: Wright s.n. 

ASU: Duran s.n.; 

Toleman 6-N. 

Figure 5. Map of the Arizona distribution of Allium 

bigelovii based on twenty six collections identified as 

Allium bigelovii at Arizona herbaria. 

Figure 6. Revised map of Arizona distribution of Allium 

bigelovii based on author’s annotations of collections at 

Arizona herbaria. 

58


Figure 7. Chihuahuan Desert habitat of Allium bigelovii 

in Greenlee County, Arizona. 

bigelovii include, from southeast to northwest, Tonto 

Basin (Crooks s.n. 1939 ARIZ); Verde Formation - 

Verde Valley (Morefield 1324 ATC; Lambreschte 30 

ATC; Haskell & Deaver 2449 MNA; Wetherill s.n. 

MNA); Rock Springs beds - Table Mountain (Kierstead 

80-1 ATC); Milk Creek beds - Walnut Grove (Palmer 

532 NY); Burro Creek (Crooks s.n. 1938 ARIZ; Darrow 

10906 ARIZ; Butterwick 4504 ASU; Anderson 2008-07 

ASU; and Chapin Wash Formation - Anderson Mine 

(Otton 1981) (Butterwick 6165 ASU; Anderson 2008-03 

ASU [Appendix 1]). These disjunct localities brought 

the Chihuahuan Desert species, Allium bigelovii, into 

the Sonoran Desert of west central Arizona as far west 

as Burro Creek, Mohave County (Figures 8, 9). Allium 

bigelovii is rarer in Arizona than previously thought. It 

is now known from approximately eight to ten localities 

(some collections from the Verde Valley have vague 

locality data on the herbarium labels). Because it occurs 

throughout southwestern New Mexico and is “…occasionally 

abundant…” there (Sivinski 2003), A. bigelovii 

is not a rare species overall. 

Allium parishii is a rare Mohave Desert species from 

California with peripheral localities in western Arizona 

at the eastern edge of its range (Figure 10). The type 

collection of A. parishii is from Cushenbury Springs, 

San Bernardino County, CA (S. B. Parish 1344 NY) 

(Watson 1882). A recent review of Allium parishii by 

White (2005) documented its current range and status. 

In California it primarily occurs in the San Bernardino 

Mountains (San Bernardino County), Little San Bernardino 

Mountains (Riverside County) and eastward into 

Joshua Tree National Park (JTNP). White (2005) recommends 

CNPS List 1B status. Its range in Joshua Tree 

National Park has recently been expanded by T. La 

Doux, botanist at JTNP (pers. comm. 2008). 

In Arizona, Allium parishii was once known only 

from an historic collection by Marcus Jones in 1903 

Figure 8. Sonoran Desert habitat of Allium bigelovii on 

lacustrine habitat at Burro Creek, Mohave County, Arizona, 

(westernmost occurrence). 

Figure 9. Close up of Allium bigelovii at Burro Creek. 

(Jones s.n., POM) from the Chemehuevi Mountains 

(Figure 11) which are now identified as the Mohave 

Mountains just east of Lake Havasu City, Mohave 

County. In 2005, students from Northern Arizona University 

made Allium collections in the Mohave Mountains. 

These collections were originally labeled as Allium 

atrorubens (Aamodt 9 ATC) and A. nevadense 

(Dow 13 ATC), but I subsequently identified them as A. 

parishii. These collections of Allium parishii were thus 

from the same mountain range as the historic Jones collection, 

but they included specific GPS locality data. In 

2008 I relocated A. parishii in this area (Figure 12) and 

recorded habitat data, associated species, and GPS locations 

(Anderson 2008-05 ASU [Appendix 1]). Interestingly, 

the two sites I recorded were only a mile apart but 

the plants grew on soils from different geological substrates: 

granite at Scott’s Well (Figure 13) and metamorphic 

gneiss near Arrastra Well (Figure 14). Also, the 

former site contained a diverse Sonoran/Mohave desert 

59


Figure 12. Allium parishii in the Mohave Mountains. 

Figure 10. Map of Allium parishii distribution in Arizona. 

Figure 13. Mohave Mountains habitat, Mohave County, 

Arizona, of Allium parishii with diverse Sonoran/ 

Mohave Desert shrubs on granitic soils. 

Figure 11. Jones s.n. 1903 collection of Allium parishii 

from the Chemehuevi Mountains. 

Figure 14. Mohave Mountains habitat of Allium parishii 

in low diversity burned habitat on metamorphic gneiss 

soils. 

60


shrub community whereas the latter site had much less 

shrub diversity and a dense growth of Bromus rubens L. 

(red brome) (Figure 15) due to a wildfire as evidenced 

by old burn scars on Yucca brevifolia stumps. 

More recent collections have been made from the 

Kofa Mounains near Quartzite, in La Paz/Yuma Counties, 

50 miles south of the Mohave Mountains, beginning 

in 1937 (Nichols s.n. ARIZ) and continuing in 

1960 (Monson s.n. ARIZ) and 1976 (Irwin 33 ARIZ). 

Also in 2005, Allium parishii was relocated in the Kofa 

Mountains by Karen Reichhardt from the BLM Yuma 

Field Office (Reichhardt 2005-100 ARIZ). The Kofa 

Mountains are a well known locality for many relict 

species including Washingtonia filifera Wendl. and Berberis 

harrisoniana Kearney and Peebles. I visited this 

site and recorded the habitat, associated species, and 

GPS location (Anderson 2008-04 ASU [Appendix 1]). 

In the Kofa Mountains A. parishii occurs on yet another 

edaphic habitat: volcanic soils derived from andesite, in 

a diverse Sonoran Desert shrub community (Figure 16). 

The Kofa Mountains site showed no evidence of wildfire 

and contained little Bromus rubens (Figure 17). A. 

parishii seems to be a resilient species due to the ecological 

variability in its geological and edaphic habitats 

and its ability to survive the increasing wildfire frequency 

scourging the Sonoran and Mohave Deserts due 

to the growing presence of Bromus rubens. 

The biogeographical effect of past climatic and geological 

history on plant species migrations can result in 

unusual patterns of distribution. The research presented 

here has demonstrated the surprising geographic proximity 

of these two desert species of Allium previously 

known primarily from opposite sides of Arizona (Figure 

18). The accurate delineation of Allium bigelovii as a 

Chihuahuan Desert species with a disjunct population as 

far west as Burro Creek and the rediscovery of A. 

Figure 16. Kofa Mountains habitat, La Paz County, Arizona, 

of Allium parishii on volcanic andesitic soil. 

parishii, a California Mohave Desert species, at the 

eastern edge of its range in the Mohave Mountains, 

brings these two species, usually 500 km apart, to within 

less than 100 km of each other in the Sonoran Desert of 

Mohave County, Arizona. 

ACKNOWLEDGMENTS 

I want to thank Edward Gilbert and Elizabeth Makings 

from Arizona State University for help in making 

maps and J. Mark Porter from Claremont College, Rancho 

Santa Ana Botanic Garden, for help in making 

Figure 15. Mohave Mountains burned habitat of Allium 

parishii with thick cover of Bromus rubens. 

Figure 17. Kofa Mountains habitat of Allium parishii 

showing lack of Bromus rubens. 

61


Sivinski, R.C. 2003. Annotated checklist of the genus 

Allium (Liliaceae) in New Mexico. The New Mexico 

Botanist 27: 1-6. 

Watson, S. 1871, U. S. Geol, Expl. 40 th Par. 487, pl 

38, fig 8,9. 

Watson, S. 1882. Proc. Amer. Acad. Arts 17: 380. 

White, S.D. 2005. Status review: Allium parishii 

Watson. Crossosoma 30: 5-16. 

APPENDIX 1 

All voucher specimens are deposited at Arizona State 

University (ASU). 

Figure 18. Map of Allium bigelovii and A. parishii 

showing distributional proximity in Mohave County, 

Arizona. 

maps and figures. Christine Bates, Karen Reichhardt, 

and Dick Zeller provided field assistance. Dr. D. Mc- 

Neal, Dr. D. Pinkava, and Mr. R. Sivinski reviewed the 

manuscript. 


Anderson, J.L. 1996. Floristic patterns on late Tertiary 

lacustrine deposits in the Arizona Sonoran Desert. 

Madrono 43: 255-272. 

Kearney, T.H. and R.H. Peebles. 1960. Arizona 

Flora, second edition, with supplement, by J.T. Howell 

and E. McClintock. Univ. of Calif. Press, Berkeley, CA. 

McNeal, D.W. and T.D. Jacobsen. 2002. Allium. In: 

Flora of North America Editorial Committee, eds. 

1993+. Flora of North America North of Mexico. 12+ 

vols. New York and Oxford. Vol. 26: 224-261. 

Nations, J.D., R.H. Hevley, D.W. Blinn, and J.J. 

Landye. 1982. Location and chronology of Tertiary 

sedimentary deposits in Arizona: a review. Pp. 107-122 

in RV. Ingersoll and M. O. Woodburne (eds.), Cenozoic 

nonmarine deposits of California and Arizona. Pacific 

Section, Society of Economic Paleontologists and 

Mineralogists. 

Otton, J.K. 1981. Geology and enesis of the Anderson 

mine, a carbonaceous lacustrine uranium deposit, 

western Arizona: a summary report. U. S. Geological 

Survey Open- File Report 81-780. 

Ownbey, M. 1947. The genus Allium in Arizona. 

Research Studies of the State College of Washington 

15:211-232. 

Allium bigelovii S. Wats. 

Arizona: Yavapai Co.: Anderson Mine site, late Tertiary 

lacustrine outcrop; with Canotia holocantha, Ambrosia 

dumosus, Fouquieria splendens, Nolina bigelovii, Yucca 

brevifolia, Pleuraphis rigida, Calochortus flexuosus; 

Tiquilia canescens; Locally common; 12S 0290870 

3798207 1945 ft. John L. Anderson 2008-03, Apr 7, 

2008. 

Arizona: Mohave Co.: Burro Creek Cliffrose site, ca 2 

miles above Six Mile Crossing of Burro Creek; late Tertiary 

lacustrine outcrop; with Canotia holocantha, Polygala 

acanthoclada, Chrysothamnus nauseosus var. 

junceus, Ziziphus obtusifolia, Calochortus flexuosus, 

Pleuraphis rigida, Aristida purpurea, Dichelostemma 

pulchra; Locally common; 12S 0283300 3828471 

2436 ft. John L. Anderson 2008-07, Apr 24, 2008. 

Allium parishii S. Wats. 

Arizona: La Paz Co.: Kofa Mts., High Tank Seven side 

canyon off of Burro Canyon; north-facing andesite hillside; 

with Simmondsia chinensis, Bernardia incana, 

Eriogonum fasciculatum, Fouquieria splendens, 

Ephedra aspera, Canotia holocantha, Viguieria deltoides, 

Acacia greggii, Krameria grayi, Gallium stellatum, 

Xylorhiza tortifolia, Pleuraphis rigida, Stipa speciosa, 

Calochortus kennedyi, Dichelostemma pulchra, 

Opuntia chlorotica, Agave desertii; uncommon (ca 50 

plants on one acre surveyed); 11S 0777891 3698772 

2788 ft. John L. Anderson 2008-04, Apr 17, 2008. 

Arizona: Mohave Co.: Mohave Mts., side canyon with 

Scotts Well, ca ½ miles NE of Scotts Well; north-facing 

granite hillside; with Canotia holocantha, Nolina bigelovii, 

Eriogonum fasciculatum, Encelia virginensis, Galium 

stellatum, Acacia greggii, Viguieria deltoides, Lotus 

rigida, Ephedra aspera, Opuntia acanthacarpa, Thamnosma 

montana, Janusia gracilis, Xylorhiza tortifolia, 

Pleuraphis rigida, Stipa speciosa; uncommon (ca 100 

plants on ten acres surveyed); 11S 758164 3829072 

3220 ft; T14N R 18W S7 NENW; John L. Anderson 

2008-05, Apr 23, 2008. 

62


Spatial Patterns of Endemic Plant Species of the Colorado Plateau 

Crystal M. Krause, 

Northern Arizona University, Biological Sciences, Flagstaff, AZ 

Abstract. The Colorado Plateau region supports one of the highest levels of endemism in the United States. Of the 

6,800 vascular plants of the region more than 300 are endemic. Endemic species may have a higher risk of extinction 

due to their restricted geographic range. This risk may be increased with climate change. To better understand the risk 

to endemics, ecological niche modeling can provide a better understanding of the dynamics of environmental factors 

on a species range. For the endemics of the Colorado Plateau, a changing climate may modify species range. But underlying 

factors such as substrate and specialized habitat will also play a role in how a species range may change. The 

focus of this study is to understand spatial patterns and factors that predict endemism and then model species potential 

distribution. 

The Colorado Plateau ecoregion supports one of the 

highest levels of endemism in North America, ranking 

in the top three ecoregions on the continent for the total 

number of endemics in all taxonomic groups (Ricketts 

et al 1999). The Colorado Plateau also has the highest 

rate of endemism in terms of actual numbers of species 

(Kartesz and Farsted 1999). The harsh, dry environment 

of the Colorado Plateau has historically placed intense 

environmental stress on the flora. Factors such as soil, 

climate, and water scarcity among others seem to limit 

the geographic range of many species. More than 300 

species of vascular plants on the Colorado Plateau are 

endemic. Many of the endemic plant species are edaphic 

endemics restricted to one soil type, but larger geographic 

patterns of plant endemism can also be seen in 

the Colorado Plateau’s “sky island” habitats. Other important 

areas are those below 2000m; these lower elevations 

have a greater number of endemics than higher 

elevations (Welsh 1978). 

The Colorado Plateau contains 122,805,655 acres of 

land, of these 3,622,942 acres (3 percent) are protected 

lands in National Parks and Monuments, another 

64,748,735 acres (52 percent) are federally owned 

(Figure 1). Land ownership is also unique for the Plateau 

with the third most federally controlled land per 

area of all other ecoregions. Protected areas of the Colorado 

Plateau have a pivotal role to play in enabling species 

and ecosystems to persist. Protected areas can remove 

or control many of the threatening processes such 

as habitat loss and fragmentation. 

The distribution of endemic plants in protected areas 

is not fully known and very little work has been completed 

in modeling distribution shifts in response to climate 

change. To better understand the complexity and 

variability climate change may have on the distribution 

of plants and animals, ecologists have recently developed 

the concept of computationally based Ecological 

Niche Models (ENMs) (Peterson, Soberon and Samcjez- 

Cordero 1999; Peterson and Vieglais 2001; Stockwell 

and Peters 1999). ENMs integrate a wide range of environmental 

data (including point location data) to define 

potential species habitats. The output of ENMs is a set 

of grids of potential habitat based on the co-occurrence 

of known species locations and various environmental 

conditions. Each grid is assessed for accuracy by comparing 

a set of reserved species locations to the predicted 

habitat distributions. 

The predicted ENM habitats can be projected onto 

past, current, and modeled future landscapes thus providing 

testable habitat conditions that can be compared 

to known conditions to assess model accuracy and improve 

environmental predictions (Peterson et al. 2002). 

Projection onto the current landscape indicates the present 

day geographic distribution of suitable conditions 

for these plant species - the species potential habitat. 

Comparing these projections to historical and current 

location data provides information on the changes that 

plants have made in their dispersal in response to disturbances 

and environmental changes in the recent past. 

Projecting the model onto future landscapes provides 

information of how climate change may affect the species 

distribution and dispersal ability. 

METHODS 

MaxEnt 

Data collection is the first step in modeling a species 

distribution; species location point records and environmental 

data are needed for the model. Location point 

records used in this study are from herbaria and Natural 

Heritage Programs from the Four Corners region, National 

Park Service and the Bureau of Land Management. 

Species with occurrence points from less than 10 

populations were excluded from modeling because prior 

studies have demonstrated that fewer than 10 populations 

is not meaningful without extensive habitat re- 

63


Figure 1: Colorado Plateau Study Area Map 

quirement data and climate envelope constraints that 

may not be available (Stockwell and Peterson 2002). 

Environmental data listed in Table 1 are from www. 

worldclim.org and the USGS Hydro1K digital elevation 

model data. These variables include precipitation, temperature, 

aspect and slope. Geologic layers are from the 

USGS 1995 soils data. The geologic layers were converted 

from polygon to raster and downscaled to 1 km 

to match the climate and elevation layers. 

The modeling technique used in this study was Maximum 

Entropy or MaxEnt. MaxEnt is a general-purpose 

machine learning method of ecological niche modeling 

(Phillips, Anderson and Schapire 2006). MaxEnt estimates 

a species probability distribution by finding the 

probability distribution of maximum entropy, subject to 

a set of constraints that represent the information about 

the species distribution (Phillips, Anderson and Schapire 

2006). 

An important factor for choosing MaxEnt was that it 

allows for the use of presence-only data and categorical 

variables. In addition, MaxEnt has been shown to perform 

better than other algorithms for modeling distributions 

with limited data points (Elith et al 2006 and Pear- 

son et al 2007). MaxEnt provides mechanisms to assess 

the relative importance of each independent variable, 

which provides a better understanding of range shifts 

due to climate change. 

To calibrate the models all location points were randomly 

divided into training (70 percent) and testing (30 

percent) datasets. To evaluate the accuracy of the model 

and each variable’s predictive power, the Receiver Operating 

Characteristic (ROC) curve (Hanley and McNeil 

1982) was used for both training and test data. The ROC 

curve represents the relationship between the percentage 

of presences correctly predicted (sensitivity) and one 

minus the percentage of the absences correctly predicted 

(specificity). The Area Under the Curve (AUC) measures 

the ability of the model to classify correctly a species 

as present or absent. AUC values can be interpreted 

as the probability that when a site with the species present 

and a site with the species absent are drawn at random, 

the former will have a higher predicted value than 

the latter. Following Araujo and Guisan (2006), a rough 

guide for classifying the model accuracy is: 0.5- 

0.6=insufficient, 0.6-0.7=poor, 0.7-0.8=average, 0.8- 

0.9=good and 0.9-1=excellent. 

64


Table 1. Environmental Variables 

BIO 1-Annual Mean Temperature 

BIO 2-Mean Diurnal Range 

BIO 3-Isothermality 

BIO 4-Temperature Seasonality 

BIO 5-Maximum Temperature of Warmest Month 

BIO 6-Minimum Temperature of Coldest Month 

BIO 7-Temperature Annual Range 

BIO 8-Mean Temperature of Wettest Quarter 

BIO 9-Mean Temperature of Driest Quarter 

BIO 10-Mean Temperature of Warmest Quarter 

BIO 11-Mean Temperature of Coldest Quarter 

BIO 12-Annual Precipitation 

BIO 13-Precipitation of Wettest Month 

BIO 14-Precipitation of Driest Month 

BIO 15-Precipitation Seasonality 

BIO 16-Precipitation of Wettest Quarter 

BIO 17-Precipitation of Driest Quarter 

BIO 18-Precipitation of Warmest Quarter 

BIO 19-Precipitation of Coldest Quarter 

DEM-Digital Elevation Model 1km 

Slope Angle-Derived from DEM 

Slope Aspect-Derived from DEM 

Geology-Landform Description 

Alternative Suitability Models 

Three final models were built for each species: 1) 

full model, 2) pruned model, and 3) topo model. The 

full model contained all variables. The pruned model 

was based on findings from a jackknife analysis, used to 

evaluate individual variable importance in model development. 

The jackknife method evaluates variable predictive 

strength by excluding each variable and creating 

a tentative model with the remaining variables (Phillips, 

Anderson and Schapire 2006). Then tentative models 

are created using each variable in isolation (Phillips, 

Anderson and Schapire 2006). Tentative models are 

then compared to the full model. The pruned model is 

then produced with only important predictive variables 

found during the jackknife analysis. The goal of the 

pruned model was to remove redundant variables and 

provide a better fit to the most important environmental 

predictors, when compared with the full model. The 

topo model used only elevation, slope angle and slope 

aspect as variables. There is a known correlation between 

elevation and climate (precipitation and temperature). 

The topo model was used to see how predictions 

of current species distributions based on topography 

alone (elevation, slope and aspect) compared to models 

with climate data (temperature, precipitation, slope and 

aspect). All three models are evaluated for accuracy 

with an AUC score. 

Spatial Comparison of Model Output 

The habitat suitability maps produced by the three 

models were then compared spatially to identify places 

of predictive agreement between models (consistently 

predicted present or absent), and places where predicted 

area of suitable habitat were in disagreement. A spatially-explicit 

(by pixel) comparison was performed following 

the methods proposed by Parolo and others 

(2008), which produced two output maps. The first map 

identifies areas of maximum agreement between the 

models and the second map identifies areas of minimum 

agreement between models. 

RESULTS 

Two hundred and eleven endemic plants were identified 

to have 10 or more representative location points 

for accurate modeling. The least number of points used 

for training data were seven with three test points, and 

the largest was 168 training points with 71 testing 

points. Model accuracy varied across species with AUC 

values ranging from 0.6423 to 1. The jackknife analysis 

demonstrated that slope aspect was the least predictive 

variable for the most species and Precipitation Seasonality 

(BIO15) was the highest predictive variable for the 

most species (Table 2). 

Example of ENM for Sclerocactus mesae-verdae 

One example of a rare endemic plant modeled is the 

Mesa Verde cactus (Sclerocactus mesae-verdae). Mesa 

Verde cactus is listed as Threatened by the US Fish and 

Wildlife Service and a recovery plan was written in 

1984 (Heil 1984). The recovery plan suggested that 

populations should be monitored to determine their stability 

(Ladyman 2004). This cactus is restricted to populations 

in the Four Corners region of New Mexico and 

Colorado. Mesa Verde cactus occurs in salt-desert scrub 

communities, typically in the Fruitland and Mancos 

shale formations, but has also been found to grow in the 

Menefee Formation overlaying Mancos shale (Roth 

2001). It is most frequently found on the tops of hills or 

benches and along slopes and at elevation ranging from 

4900 to 5500 ft (Roth 2001). Annual precipitation varies 

from approximately 8 to 20 cm. Average temperatures 

65


Table 2. Total number of training and test points used in the model and the overall AUC value for the 

model. Predictor variable with the highest AUC value when used in isolation and the variable with the least 

predictive power for each species. 

Species 

# Training 

Samples 

# Test 

Samples 

Test 

AUC 

66 

Least Predictive 

Variable 

AUC 

Highest Predictive 

Variable 

Abronia argillosa 28 12 0.9759 Slope Aspect 0.5681 BIO 9 0.9039 

Agave utahensis ssp. 

kaibabensis 

AUC 

12 5 0.9933 BIO 19 0.372 BIO 15 0.9724 

Aliciella haydenii 14 5 0.9903 BIO 5 0.4052 BIO 11 0.8754 

Amsonia jonesii 47 20 0.9789 Slope Angle 0.6246 BIO 18 0.8829 

Amsonia peeblesii 31 12 0.9979 Slope Aspect 0.7798 BIO 14 0.9554 

Aquilegia grahamii 8 3 0.9955 BIO 17 0.3579 BIO 15 0.9371 

Aquilegia loriae 10 4 0.9994 BIO 10 0.4245 BIO 11 0.9758 

Argemone arizonica 8 3 0.9817 BIO 19 0.3105 Slope Angle 0.8387 

Argemone corymbosa ssp. 

arenicola 

7 2 0.9998 BIO 2 0.3846 BIO 16 0.9845 

Asclepias cutleri 20 8 0.8857 Geology 0.5897 BIO 14 0.9104 

Asclepias welshii 15 6 0.9802 BIO 2 0.649 BIO 11 0.9563 

Astragalus ampullarius 26 10 0.9957 Slope Angle 0.5299 BIO 15 0.9392 

Astragalus beathii 21 9 0.9994 Slope Angle 0.5269 Geology 0.9521 

Astragalus consobrinus 11 4 0.997 Slope Aspect 0.5 BIO 11 0.9555 

Astragalus cronquistii 48 20 0.9989 Slope Angle 0.6139 BIO 15 0.9437 

Astragalus debequaeus 63 26 0.89 Slope Aspect 0.6442 BIO 9 0.975 

Astragalus desperatus var. 

conspectus 

Astragalus desperatus var. 

petrophilus 

17 6 0.9989 Slope Aspect 0.4179 BIO 11 0.9945 

21 8 0.9991 Slope Angle 0.3093 BIO 12 0.9643 

Astragalus deterior 83 35 0.9997 Slope Aspect 0.6219 BIO 8 0.9842 

Astragalus detritalis 24 10 0.9075 Slope Aspect 0.4536 BIO 11 0.9654 

Astragalus duchesnensis 56 23 0.9981 Slope Angle 0.7779 BIO 7 0.9799 

Astragalus eastwoodiae 14 6 0.9851 BIO 10 0.4144 BIO 11 0.906 

Astragalus episcopus var. 

lancearius 

8 3 0.9595 BIO 8 0.4212 BIO 16 0.836 

Astragalus hamiltonii 11 4 0.759 Slope Angle 0.4996 BIO 16 0.9884 

Astragalus henrimontanensis 

10 4 0.9996 Slope Aspect 0.4454 Geology 0.9316 

Astragalus humillimus 19 7 0.9988 Slope Aspect 0.6221 BIO 4 0.9775 

Astragalus iodopetalus 33 14 0.9985 Slope Aspect 0.7224 BIO 7 0.9746 

Astragalus iselyi 16 6 0.999 Slope Aspect 0.7312 BIO 8 0.9718

Species 

# Training 

Samples 


# Test 

Samples 

Table 2. Continued 

Test 

AUC 


Variable 

AUC 


Variable 

Astragalus linifolius 14 6 0.9721 BIO 2 0.5 BIO 8 0.9492 

Astragalus malacoides 16 6 0.9887 Slope Aspect 0.4433 BIO 16 0.9338 

Astragalus micromerius 10 3 0.9897 BIO 5 0.5042 BIO 11 0.941 

Astragalus moencoppensis 52 22 0.9854 Slope Aspect 0.6095 BIO 16 0.9076 

Astragalus musiniensis 63 26 0.9868 Slope Aspect 0.4348 BIO 16 0.9137 

Astragalus naturitensis 62 26 0.9953 Slope Aspect 0.5935 BIO 8 0.9406 

Astragalus nutriosensis 21 8 0.9982 Slope Aspect 0.4817 Geology 0.9908 

Astragalus oocalycis 8 3 0.9958 BIO 19 0.4252 BIO 16 0.9727 

Astragalus perianus 20 8 0.9749 BIO 2 0.6125 Geology 0.9398 

Astragalus piscator 16 6 0.9999 Slope Angle 0.6152 BIO 9 0.9976 

Astragalus proximus 12 5 0.9772 BIO 12 0.4203 BIO 16 0.9341 

Astragalus rafaelensis 83 35 0.9999 Slope Angle 0.707 BIO 7 0.9709 

Astragalus rusbyi 21 9 0.9982 Slope Angle 0.7211 BIO 3 0.9328 

Astragalus schmolliae 18 7 0.9996 Slope Angle 0.7367 BIO 11 0.9848 

Astragalus serpens 19 7 0.9976 Slope Aspect 0.4439 BIO 8 0.9671 

Astragalus sesquiflorus 24 10 0.9983 Slope Aspect 0.491 BIO 6 0.9368 

Astragalus sophoroides 28 11 0.9994 BIO 2 0.8478 BIO 12 0.9723 

Astragalus striatiflorus 24 10 0.9959 Slope Aspect 0.5023 BIO 15 0.952 

Astragalus tortipes 8 3 0.988 Slope Angle 0.4142 Geology 0.9211 

Astragalus troglodytus 52 22 0.9989 Slope Angle 0.6131 BIO 3 0.9536 

Astragalus welshii 16 6 0.7897 Slope Aspect 0.4063 BIO 15 0.8887 

Astragalus wetherillii 12 5 0.9845 BIO 19 0.4486 BIO 15 0.9848 

Astragalus xiphoides 45 18 0.9912 Slope Angle 0.5483 BIO 1 0.9824 

Camissonia atwoodii 50 21 1 Slope Aspect 0.6725 BIO 12 0.9868 

Camissonia eastwoodiae 19 7 0.9848 BIO 17 0.5237 BIO 13 0.8891 

Camissonia exilis 38 16 0.9739 Slope Aspect 0.6027 BIO 15 0.948 

Carex curatorum 19 7 0.9907 Slope Aspect 0.2923 BIO 15 0.8895 

Carex specuicola 77 32 0.9969 Slope Aspect 0.4858 BIO 11 0.9395 

Castilleja aquariensis 168 71 0.9999 Slope Aspect 0.6332 Geology 0.9901 

Castilleja kaibabensis 24 9 0.9999 Slope Aspect 0.6471 BIO 15 0.9874 

Castilleja revealii 14 6 0.9782 Slope Aspect 0.5 Geology 0.9151 

Chrysothamnus molestus 35 15 0.9993 Slope Aspect 0.6306 Geology 0.9756 

AUC 

67

Species 

Chrysothamnus viscidiflorus 

ssp. planifolius 

# Training 

Samples 


# Test 

Samples 


Test 

AUC 


Variable 

AUC 


Variable 

AUC 

8 3 0.9916 BIO 2 0.4925 Slope Angle 0.7825 

Cirsium chellyense 11 4 0.9864 BIO 10 0.4346 BIO 11 0.9344 

Cirsium murdockii 11 4 0.7394 BIO 4 0.2853 BIO 15 0.9444 

Cirsium perplexans 50 21 0.9924 Slope Angle 0.6892 Geology 0.9253 

Cirsium rydbergii 26 11 0.985 Slope Aspect 0.459 Geology 0.9099 

Clematis hirsutissima 19 7 0.9999 Slope Aspect 0.4112 BIO 12 0.9771 

Cleomella palmeriana 17 6 0.9784 Slope Angle 0.4802 BIO 13 0.889 

Crataegus saligna 19 8 0.9261 Slope Aspect 0.4154 BIO 6 0.9537 

Cryptantha atwoodii 22 9 0.9994 Slope Aspect 0.5488 BIO 16 0.9697 

Cryptantha capitata 33 14 0.9858 Slope Aspect 0.5139 BIO 15 0.9494 

Cryptantha cinerea var. 

arenicola 


Cryptantha creutzfeldtii 33 14 0.9997 Slope Aspect 0.6437 BIO 6 0.9685 

Cryptantha elata 10 4 0.9997 BIO 1 0.402 Slope Angle 0.9762 

Cryptantha johnstonii 12 4 0.9987 Slope Aspect 0.2564 BIO 6 0.9714 

Cryptantha jonesiana 20 8 0.6423 Slope Aspect 0.2748 BIO 9 0.9782 

Cryptantha longiflora 17 7 0.8946 BIO 5 0.4916 BIO 15 0.8834 

Cryptantha mensana 9 3 0.8785 Slope Angle 0.217 BIO 15 0.9045 

Cryptantha osterhoutii 37 15 0.9886 Slope Aspect 0.5552 BIO 13 0.914 

Cryptantha paradoxa 11 4 0.9657 BIO 10 0.4083 BIO 7 0.8628 

Cryptantha semiglabra 21 9 1 Slope Angle 0.7076 BIO 7 0.9914 

Cycladenia humilis var. 

jonesii 


Cymopterus duchesnensis 39 16 0.9988 Slope Aspect 0.7142 BIO 11 0.9706 

Cymopterus megacephalus 24 9 0.9492 Slope Aspect 0.5658 BIO 15 0.9087 

Cymopterus minimus 34 14 0.9916 Slope Aspect 0.4755 BIO 15 0.933 

Dalea flavescens 33 13 0.9489 Slope Aspect 0.5197 BIO 13 0.8887 

Draba graminea 10 4 0.9983 Slope Aspect 0.4768 BIO 10 0.9948 

Eremocrinum albomarginatum 

10 4 0.9826 BIO 9 0.2196 BIO 7 0.9466 

Ericameria zionis 8 3 0.7485 BIO 19 0.3877 Slope Aspect 0.839 

Erigeron kachinensis 60 25 0.9972 Slope Aspect 0.7543 BIO 15 0.9619 

Erigeron maguirei 21 8 0.9926 Slope Aspect 0.397 BIO 9 0.9714 

68

Species 

# Training 

Samples 


# Test 

Samples 


Test 

AUC 


Variable 

AUC 


Variable 

Erigeron mancus 17 6 0.9998 Slope Aspect 0.6755 BIO 15 0.9914 

Erigeron proselyticus 10 4 0.9798 BIO 12 0.3529 BIO 10 0.918 

Erigeron religiosus 13 5 0.9252 BIO 5 0.3833 BIO 15 0.9528 

Erigeron rhizomatus 23 9 0.9999 Slope Aspect 0.6843 BIO 9 0.9796 

Erigeron sionis 7 3 0.7639 BIO 19 0.445 BIO 15 0.8189 

Erigeron sivinskii 14 6 0.9732 Slope Angle 0.4754 BIO 8 0.9373 

Eriogonum aretioides 14 5 0.9821 Slope Angle 0.5789 BIO 15 0.9153 

Eriogonum bicolor 19 8 0.995 BIO 2 0.5918 BIO 4 0.9464 

Eriogonum clavellatum 45 19 0.9977 Slope Angle 0.6982 BIO 7 0.9668 

Eriogonum contortum 14 5 0.9996 BIO 2 0.6408 BIO 13 0.9855 

Eriogonum jonesii 26 11 0.9904 Slope Angle 0.5547 Geology 0.9317 

Eriogonum leptocladon 

var. ramosissimum 

AUC 


Eriogonum pelinophilum 111 47 0.9999 Slope Angle 0.716 BIO 7 0.9776 

Eriogonum ripleyi 31 13 0.9998 Slope Aspect 0.5704 BIO 15 0.9741 

Eriogonum scabrellum 12 5 0.9896 BIO 9 0.4757 Geology 0.9042 

Eriogonum subreniforme 21 9 0.8827 Slope Aspect 0.4721 BIO 12 0.9149 

Eriogonum tumulosum 14 5 0.9914 BIO 5 0.4753 BIO 7 0.9641 

Errazurizia rotundata 19 7 0.9991 BIO 3 0.7049 BIO 7 0.9724 

Euphorbia aaron-rossii 39 16 0.9991 Slope Aspect 0.5234 BIO 12 0.9537 

Euphorbia nephradenia 10 3 0.9928 Slope Aspect 0.4562 BIO 16 0.9519 

Gaillardia flava 14 6 0.9975 Slope Angle 0.6717 BIO 6 0.9812 

Gilia caespitosa 24 9 1 Slope Aspect 0.817 BIO 18 0.9578 

Gilia stenothyrsa 13 5 0.9989 BIO 10 0.496 BIO 9 0.9748 

Gilia tenuis 8 3 0.9998 Slope Aspect 0.1882 BIO 12 0.9445 

Glaucocarpum suffrutescens 

46 19 1 Slope Aspect 0.7096 BIO 9 0.9904 

Grindelia fastigiata 10 4 0.9902 Slope Angle 0.2632 BIO 9 0.9077 

Grindelia laciniata 12 5 0.9722 BIO 3 0.3623 BIO 17 0.8674 

Hackelia gracilenta 21 9 0.9998 Slope Aspect 0.6853 BIO 15 0.9844 

Hedeoma diffusa 53 22 0.9998 Slope Aspect 0.6175 BIO 15 0.9649 

Hedysarum occidentale 

var. canone 


69

Species 

# Training 

Samples 


# Test 

Samples 


Test 

AUC 


Variable 

AUC 


Variable 

Hesperodoria salicina 28 11 0.9129 Slope Aspect 0.5241 BIO 15 0.9367 

Heterotheca jonesii 12 5 0.9941 BIO 2 0.4925 Geology 0.8956 

Hymenoxys jamesii 25 10 0.9953 Slope Aspect 0.5684 BIO 11 0.9456 

Ipomopsis polyantha 17 6 0.9462 BIO 1 0.4734 BIO 7 0.8384 

Lepidium huberi 10 3 0.9728 BIO 17 0.498 BIO 9 0.9058 

Lepidium montanum var. 

neeseae 

AUC 


Lesquerella congesta 23 9 1 Slope Aspect 0.6799 BIO 6 0.9971 

Lesquerella kaibabensis 10 3 0.8568 Slope Aspect 0.5 Geology 0.9772 

Lesquerella navajoensis 12 5 0.9999 Slope Angle 0.5 Geology 0.989 

Lesquerella parviflora 86 36 0.9981 Slope Aspect 0.6236 BIO 15 0.9626 

Lesquerella pruinosa 13 5 1 Slope Aspect 0.58 BIO 16 0.9931 

Lesquerella vicina 24 10 0.9987 Slope Angle 0.6923 BIO 9 0.9722 

Lomatium concinnum 129 54 0.9998 Slope Aspect 0.6419 BIO 4 0.9586 

Lomatium latilobum 9 3 0.9962 BIO 17 0.4158 BIO 4 0.9019 

Lupinus crassus 38 16 0.9863 Slope Aspect 0.5558 BIO 8 0.9477 

Lygodesmia doloresensis 32 13 0.9999 Slope Aspect 0.7192 BIO 8 0.9717 

Mentzelia marginata 14 5 0.942 Slope Angle 0.3475 BIO 15 0.8468 

Mentzelia rhizomata 54 22 0.9999 Slope Aspect 0.716 BIO 15 0.9939 

Mentzelia shultziorum 8 3 0.6819 BIO 17 0.4103 BIO 4 0.9093 

Mimulus eastwoodiae 41 17 0.979 BIO 5 0.5615 Geology 0.8184 

Myosurus nitidus 12 4 0.79 BIO 5 0.435 Geology 0.9316 

Nama retrorsum 39 16 0.9892 Slope Aspect 0.5523 BIO 11 0.9098 

Oenothera acutissima 57 24 0.9982 Slope Aspect 0.611 BIO 13 0.9686 

Opuntia aurea 35 15 0.9957 Slope Aspect 0.5969 BIO 1 0.9615 

Oreoxis trotteri 9 3 0.997 BIO 17 0.4995 Slope Angle 0.9762 

Packera franciscana 10 4 0.9977 Slope Aspect 0.5 BIO 15 0.9707 

Parthenium ligulatum 13 5 0.9637 BIO 8 0.43 BIO 15 0.8352 

Pediocactus bradyi 29 12 0.9992 Slope Angle 0.7323 BIO 16 0.9645 

Pediocactus despainii 17 7 0.9966 Slope Aspect 0.4568 BIO 8 0.9807 

Pediocactus paradinei 28 12 0.9966 Slope Angle 0.7985 BIO 15 0.9701 

Pediocactus peeblesianus 

var. fickeiseniae 

40 17 0.9937 BIO 2 0.6728 BIO 11 0.9523 

70

Species 

# Training 

Samples 


# Test 

Samples 


Test 

AUC 


Variable 

AUC 


Variable 

Pediocactus winkleri 10 4 0.9997 Slope Aspect 0.3578 BIO 9 0.9799 

Pediomelum pariense 21 8 0.9839 Slope Aspect 0.4554 BIO 15 0.9679 

Penstemon ammophilus 19 7 0.9999 Slope Aspect 0.6623 BIO 15 0.9808 

Penstemon atwoodii 15 6 0.9978 Slope Aspect 0.4472 BIO 15 0.9662 

Penstemon bracteatus 18 7 0.9893 Slope Aspect 0.7072 BIO 9 0.9267 

Penstemon breviculus 9 3 0.6775 Slope Angle 0.2816 BIO 15 0.854 

Penstemon clutei 56 24 0.9973 Slope Aspect 0.5949 BIO 15 0.9707 

Penstemon distans 16 6 1 BIO 5 0.6628 BIO 1 0.9889 

Penstemon flowersii 19 7 1 Slope Aspect 0.7017 BIO 7 0.9976 

Penstemon gibbensii 8 3 1 Slope Angle 0.3782 Geology 0.9964 

Penstemon goodrichii 17 7 0.9999 Slope Angle 0.762 BIO 16 0.9904 

Penstemon grahamii 90 38 0.9996 Slope Aspect 0.69 BIO 9 0.9865 

Penstemon lentus var. 

albiflorus 

AUC 

16 6 0.9986 BIO 5 0.625 BIO 15 0.9771 

Penstemon marcusii 12 5 0.9991 Slope Angle 0.4632 BIO 18 0.9852 

Penstemon nudiflorus 79 33 0.9971 Slope Aspect 0.6484 BIO 15 0.9493 

Penstemon pseudoputus 39 16 0.9896 Slope Aspect 0.5151 BIO 15 0.9375 

Penstemon strictiformis 9 3 0.7698 Slope Aspect 0.5716 Slope Angle 0.6493 

Penstemon uintahensis 13 5 0.808 Slope Aspect 0.4688 BIO 15 0.9571 

Perityle specuicola 12 5 0.9926 Slope Angle 0.4812 BIO 8 0.9507 

Phacelia cephalotes 31 13 0.9851 Slope Angle 0.6395 BIO 4 0.9068 

Phacelia constancei 35 15 0.991 Slope Aspect 0.6325 Geology 0.952 

Phacelia crenulata var. 

angustifolia 

26 10 0.9787 BIO 19 0.5829 BIO 15 0.871 

Phacelia glechomifolia 46 19 0.9974 Slope Aspect 0.5353 BIO 15 0.9332 

Phacelia rafaelensis 17 7 0.9748 Slope Angle 0.5166 BIO 15 0.9301 

Phacelia splendens 17 6 0.9513 BIO 3 0.5761 BIO 8 0.9122 

Phacelia welshii 18 7 0.9987 Slope Angle 0.724 Geology 0.9767 

Phlox caryophylla 11 4 1 Slope Angle 0.4067 BIO 16 0.9953 

Phlox cluteana 23 9 0.9987 Slope Aspect 0.4581 Geology 0.9465 

Physaria obcordata 28 12 0.9895 Slope Aspect 0.5847 BIO 15 0.9721 

Physaria repanda 11 4 0.9962 Slope Angle 0.5 BIO 15 0.9436 

Platanthera zothecina 47 20 0.9555 Slope Aspect 0.6004 BIO 15 0.8757 

71

Species 

# Training 

Samples 


# Test 

Samples 


Test 

AUC 


Variable 

AUC 


Variable 

Potentilla angelliae 13 5 0.8715 Slope Angle 0.4887 BIO 9 0.9699 

Primula specuicola 27 11 0.9925 BIO 3 0.6973 BIO 16 0.8986 

Psoralidium junceum 7 3 0.9942 BIO 11 0.5 BIO 12 0.9351 

Psorothamnus arborescens 

var. pubescens 

Psorothamnus thompsoniae 

var. whitingii 

AUC 

10 4 0.9997 BIO 2 0.4595 BIO 16 0.9812 


Rosa stellata ssp. abyssa 15 6 0.999 Slope Aspect 0.5631 Geology 0.9772 

Salix arizonica 129 54 0.9961 Slope Aspect 0.7927 Geology 0.9772 

Schoenocrambe argillacea 57 24 0.9999 Slope Aspect 0.7184 BIO 9 0.9905 

Sclerocactus brevispinus 45 19 1 Geology*** 0.9011 BIO 7 0.9995 

Sclerocactus glaucus 10 4 0.9039 BIO 10 0.3434 BIO 15 0.9249 

Sclerocactus mesae-verdae 146 62 0.9997 Slope Aspect 0.648 BIO 6 0.9794 

Sclerocactus parviflorus 

var. intermedius 

11 4 0.9914 BIO 3 0.446 BIO 18 0.9473 

Sclerocactus sileri 29 12 0.9949 Slope Aspect 0.6299 BIO 15 0.9548 

Sclerocactus whipplei 35 14 0.9668 Slope Aspect 0.4761 BIO 15 0.852 

Sclerocactus wrightiae 81 34 0.9869 Slope Angle 0.5928 BIO 12 0.9858 

Shepherdia rotundifolia 91 38 0.9945 Slope Aspect 0.5764 BIO 15 0.924 

Silene petersonii 20 8 0.9766 Slope Aspect 0.5782 BIO 6 0.9093 

Silene rectiramea 11 4 1 Slope Aspect 0.5592 BIO 15 0.9866 

Sphaeralcea janeae 18 7 0.9984 Slope Angle 0.7861 BIO 13 0.9927 

Sphaeralcea psoraloides 40 17 0.9993 Slope Angle 0.5282 BIO 12 0.9812 

Talinum thompsonii 10 4 1 Slope Aspect 0.5 BIO 6 0.9978 

Thelypodiopsis juniperorum 


Townsendia aprica 52 21 0.9934 Slope Aspect 0.6216 BIO 11 0.9343 

Townsendia glabella 77 33 0.9767 Slope Angle 0.5305 BIO 15 0.9328 

Townsendia rothrockii 36 15 0.9902 BIO 2 0.6204 BIO 15 0.925 

Trifolium neurophyllum 20 8 0.9862 Slope Aspect 0.4605 BIO 3 0.9729 

Vanclevea stylosa 31 12 0.9924 Slope Aspect 0.6921 BIO 12 0.9618 

Xylorhiza glabriuscula var. 

linearifolia 

7 3 0.9981 BIO 11 0.4973 BIO 4 0.9204 

Zigadenus vaginatus 16 6 0.995 Slope Angle 0.5341 Geology 0.9368 

72


in the town of Shiprock near plant locations range from 

a high/low in January (coldest month) of 44.4 o F/17 o F to 

a high/low in July (hottest month) of 95 o F/58 o F (Ladyman 

2004). The federal listing of Mesa Verde cactus 

provides detailed information on habitat, soil and climatic 

requirements of the plant, providing great detail to 

compare to model data. 

Mesa Verde Cactus has 208 species location points. 

Models were developed using 146 for model training 

and 62 for model test validation points. The full, pruned 

and topo models were used to analyze suitable habitat 

for Mesa Verde cactus. The AUC score of the full 

model, incorporating all of the environmental variables, 

was 0.9977. The AUC score of the pruned model, developed 

from jackknife analysis, was 0.997. The pruned 

model showed geology as the highest predictive vari- 

able. In comparison, the topo model AUC score was 

0.831. The accuracy of the three niche models measured 

by AUC scores, demonstrated that all three models performed 

better than random (AUC 0.5). The full and 

pruned models were nearly identical in AUC scores 

(0.9977, 0.997). The topo model was out performed by 

both the full and pruned models. 

Variable isolation during jackknife analysis indicated 

geology, minimum temperature of coldest month, mean 

temperature of coldest quarter and temperature seasonality 

were the most predictive variables, with an AUC of 

0.9 or above (Table 3). Variable inclusion during jackknife 

analysis did not show any significant results; the 

AUC scores only changed 0.001 between models. The 

least predictive variable was slope aspect. 

Table 3: Mesa Verde Cactus model AUC scores. 

Sclerocactus mesae-verdae Jackknife Variable Exclusion Jackknife Variable Isolation 

#Training samples 146 AUC without Geology 0.997 AUC with only Slope Aspect 0.5824 

Iterations 500 AUC without BIO 4 0.9974 AUC with only BIO 2 0.6194 

Training AUC 0.9977 AUC without BIO 1 0.9975 AUC with only Slope Angle 0.7073 

#Test samples 62 AUC without BIO 10 0.9975 AUC with only BIO 12 0.7626 

Test AUC 0.9977 AUC without BIO 11 0.9975 AUC with only BIO 19 0.7773 

AUC Standard Deviation 0.0005 AUC without BIO 13 0.9975 AUC with only BIO 17 0.7802 

AUC without BIO 16 0.9975 AUC with only BIO 14 0.7907 














AUC without Slope Angle 0.9976 AUC with only BIO 6 0.9655 

AUC without Slope Aspect 0.9976 AUC with only Geology 0.9888 

73


The habitat suitability maps also varied between the 

models (Figure 2). The full model (Figure 2a) and the 

pruned model (Figure 2b) clearly show a smaller range 

than the topo model (Figure 2c). The area of suitable 

habitat for the full model was 919,973 acres, the pruned 

model 1,386,261 acres and the topo model 47,110,640 

acres. 

The spatial comparison of models found 571,307 

acres of suitable habitat agreement between the models 

(Figure 2d). Disagreement analysis of the three models 

shows 47,081,482 acres (Figure 2e) and suggests that 

the topo model generated the most disagreement between 

models. 

DISCUSSION 

Alternative Suitability Models 

The use of alternative models can offer a better understanding 

of how environmental variables play a role 

in a species distribution. The comparison of the three 

models may provide a deeper knowledge of speciesenvironment 

relationships and lead to a more rigorous 

assessment of potential distributions. The alternative 

models may also identify needs in data to further understand 

species-environment relationships. 

The three models used in this analysis all performed 

better than random. The full and pruned models performed 

the best (highest AUC scores) while the topo 

model over predicted suitable habitat. The jackknife 

analysis provided a way to understand variable importance 

for the use in the pruned model. This may help 

eliminate redundant variables and an overly large 

model. 

Most pruned models showed a larger area of suitable 

habitat when compared with the full models. This suggests 

two possible interpretations: 1) the full model may 

be over fit, and not predicting all potential suitable habitat, 

or 2) the pruned model may be over predicting. Another 

interesting aspect of using alternative models was 

the use of only topographic variables in the topo model. 

Although this model over predicted suitable habitat for 

Mesa Verde cactus, other species models show a closer 

fit. This model may provide a way to analyze a species 

distribution when only elevation data are available, as 

Parolo and others (2008) found with Arnica montana. 

Spatial Comparison 

The spatial comparison of habitat suitability map 

output from the three models provides quantification of 

uncertainty in the model predictions of suitable habitat. 

Levels of uncertainty have important management implications. 

Areas with high model agreement that a species 

is present but without known occurrences of that 

species are target areas for field surveys. Areas of disagreement 

may provide insight into variables that con- 

tribute to uncertainty that could be better resolved 

through additional field work. 

Final Habitat Maps and Use 

The goal of this project was to identify potential suitable 

habitat for species by using alternative models 

evaluated with AUC scores and a spatial comparison. 

When comparing the three models, the AUC scores 

range from good to excellent for Mesa Verde cactus. 

The spatial comparison identified the full and pruned 

models as having similar predicted areas, while the area 

predicted by the topo model was larger. This technique 

provides a more rigorous analysis of the potential distribution 

of suitable habitat. 

The methods described above provide a more 

straightforward ecological interpretation of how the environment 

affects a species’ distribution. This modeling 

technique can also provide a better understanding of 

endemic plant distributions. These models can be used 

for future field investigations to find new populations 

and to identify relationships between climate, geology 

and topography with endemics. 


This project was supported by the Gloria Barron Wilderness 

Society Scholarship and NSF grant 0753163. 

Comments from Dr. Deana Pennington and Daniela 

Roth greatly improved the manuscript and many thanks 

to those who have shared their data and time to improve 

this project. 


Araujo, M. and A. Guisan. 2006. Five (or so) challenges 

for species distribution modeling. Journal of Biogeography 

33: 1677-1688. 

Elith, J., C. Graham, R. Anderson, M. Dudik, S. Ferrier, 

A. Guisan, R. Hijmans, F. Huettmann, J. Leathwick, 

A. Lehmann, J. Lucia, L. Lohmann, B. Loiselle, 

G. Manion, C. Moritz, M. Nakamura, Y. Nakazawa, J. 

McC Overton, A.T. Peterson, S. Phillips, K. Richardson, 

R. Scachetti-Pereira, R. Schapire, J. Soberon, S. Williams, 

M. Wisz, and N. Zimmermann, 2006. Novel 

methods improve prediction of species’ distributions 

from occurrence data. Ecography 29: 129-151. 

Hanley, J.A. and B.J. McNeil. 1982. The meaning 

and use of the area under a Receiver Operating Characteristic 

(ROC) curve. Radiology 29: 773-785. 

Heil, K.D. 1984. Mesa Verde cactus (Sclerocactus 

mesae-verdae) Recovery plan. U.S. Fish and Wildlife 

Service, Region 2, Albuquerque, New Mexico. 

Kartesz, J. and A. Farstad. 1999. Multi-scale analysis 

of endemism of vascular plant species. In: Terrestrial 

ecoregions of North America: A conservation assessment. 

Eds. Ricketts, T.H., Dinerstein, E., Olson, D.M. 

and C. Loucks. Island Press, Washington, D.C. 

74


Figure 2. Mesa Verde Cactus distribution maps. Full model with all variables (a), pruned model (b), topo model (c), 

Suitable Habitat maps Maximum Agreement (d), Suitable Habitat maps Minimum Agreement (e). 

75


Ladyman, J.A.R. 2004. Status Assessment Report for 

Sclerocactus mesae-verdae. Prepared for The Navajo 

Nation Natural Heritage Program, Window Rock, AZ 

by JnJ Associates, LLC. Centennial, CO. 

Parolo, G., G. Rossi, and A. Ferrarini. 2008. Toward 

improved species niche modeling: Arnica montana in 

the Alps as a case study. Journal of Applied Ecology 45: 

1410-1418. 

Pearson, R.G., C.J. Raxworthy, M. Nakamura, and 

A.T. Peterson. 2007. Predicting species distributions 

from small numbers of occurrence records: a test case 

using cryptic geckos in Madagascar. Journal of Biogeography 

34: 102-117. 

Peterson, A. T., J. Soberon and V. Samcjez-Cordero. 

1999. Conservatism of Ecological Niches in Evolutionary 

Time. Science 285(5431): 1265. 

Peterson, A. T. and D. A. Vieglais 2001. Predicting 

Species Invasions Using Ecological Niche Modeling: 

New Approaches from Bioinformatics Attack a Pressing 

Problem. BioScience 51(5): 363-371. 

Peterson, A. T., M. Ortega-Huerta, V. Sanchez- 

Cordero, J. Soberon, R. Buddemeier and D. Stockwell. 

2002. Future Projections for Mexican Faunas Under 

Global Climate Change Scenarios. Nature (416). 

Phillips, S.J., R.P. Anderson, and R.E. Schapire. 

2006. Maximum entropy modeling of species geographic 

distributions. Ecological Modelling 190: 231- 

259. 

Ricketts, T.H., E. Dinerstein, D.M. Olson, and C. 

Loucks. 1999. Terrestrial ecoregions of North America: 

A conservation assessment. Island Press, Washington, 

D.C. 

Roth, D. 2001. Species account for Sclerocactus mesae-verdae. 

Navajo Nation Natural Heritage Program. 

Window Rock, AZ 

Stockwell, D. and D. Peters. 1999. The GARP Modelling 

System: Problems and Solutions to Automated 

Spatial Prediction. Intl. Journal of Geographical Information 

Science 13. 

Stockwell, D. and A.T. Peterson. 2002. Effects of 

sample size on accuracy of species distribution models. 

Ecological Modelling 148: 1-13. 

Welsh, S.L. 1978. Problems in plant endemism on 

the Colorado Plateau. Great Basin Naturalist Memoirs 

2:191-195. 

76


Biogeography of Rare Plants of the 

Ash Meadows National Wildlife Refuge, Nevada 

Leanna Spjut Ballard, 

Ballard Ecological Consulting, Millville, UT 

Abstract. The Ash Meadows National Wildlife Refuge encompasses more than 23,000 acres of unique desert that 

provides habitat for at least 25 plant and wildlife species found nowhere else in the world. Distinctive hydrology sustains 

high concentrations of endemic plants on the Refuge. Spring complexes, alkaline desert uplands, and velvet ash, 

emergent marsh and wet meadow communities known only to exist within the Refuge provide structure and habitat 

for several rare, endemic and endangered plants. Two studies designed to assist the Refuge with large-scale habitat 

restoration plans are underway. These studies include mapping all vegetation communities to a fine scale and locating 

and mapping the distribution of rare and listed plants on the Refuge. Mapping and classifying the vegetation communities 

to the alliance and association scale throughout the entire Refuge will provide a baseline of existing ecological 

conditions for monitoring change in the future. The rare plant surveys will also serve as a tool for monitoring the rare 

plants and the habitats in which they occur. Initially, vegetation classification standards were based upon community 

types derived from multiple earlier published sources. Due to the unique habitats and patterns of co-dominance of 

species occurring at the Refuge, several new alliance and association classification descriptions are being written to 

accurately describe plant communities. 

Description of the Ash Meadows National Wildlife 

Refuge 

The Ash Meadows National Wildlife Refuge 

(AMNWR) encompasses more than 23,000 acres of 

unique desert that provides habitat for at least 25 species 

found nowhere else in the world. The refuge may support 

the largest concentration of endemic species of any 

terrestrial landscape in the 48 contiguous United States. 

The Devil’s Hole pupfish (Cyprinodon diabolis) and 

one plant species, Ash Meadows niterwort (Nitrophila 

mohavenis), are listed as federally endangered, and most 

of the other endemic species are either listed as threatened 

or are managed as sensitive species. Unfortunately, 

past human activities in AMNWR have resulted in the 

invasion of numerous non-native plant species, several 

of which negatively affect native populations and habitats 

(Otis Bay and Steven Ecological Consulting 2006). 

Ash Meadows is essentially a watered island amidst the 

vast Mojave Desert. Groundwater discharging from a 

regional carbonate aquifer feeds the numerous springs 

that exist at Ash Meadows. Spring discharge maintains 

soil moisture in the lowlands while uplands receive water 

only from rainfall that averages less than 2.75 inches 

annually. Annual evaporation exceeds 98.50 inches. 

Ash Meadows is situated at approximately 2,200 feet 

elevation in the Mohave Desert, 40 miles east of Death 

Valley National Monument headquarters at Furnace 

Creek, California, and 90 miles northwest of Las Vegas, 

Nevada (Figure 1). It is located in the east-central portion 

of the Amargosa Desert. A series of Cambrian 

limestone and dolomite ridges form the eastern boundary 

of the Refuge near the center of the Amargosa Val- 

ley. The northern boundary crosses Quaternary playa 

deposits while the southern boundary primarily crosses 

Quaternary alluvial fan deposits. The pattern of sand 

dunes, badlands, alluvial fans, and broad meadows observed 

throughout Ash Meadows is the result of a history 

of playa deposition followed by erosion, formation 

of alluvial fans adjacent to surrounding ranges, and the 

transport of sand by both wind and infrequent flows in 

larger washes and drainages. 

The distinctive hydrology of Ash Meadows is the 

result of an extensive groundwater system and surface 

water drainage that culminates in the Carson Slough, a 

tributary to the Amargosa River (Otis Bay and Stevens 

Ecological Consulting 2006). Carson Slough is the primary 

drainage in Ash Meadows and is generally considered 

the core of the Ash Meadows ecosystem. The Crystal 

Spring drainage and the Jackrabbit/Big Spring drainages 

are significant tributaries of Carson Slough and 

drain large portions of the Refuge. Two primary aquifers, 

a regional carbonate aquifer and a local valley-fill 

aquifer, are present. Water is generally retained within 

the wetlands and alkali flats that sustain many of the 

Refuge’s endemic plants, as well as some endemic fish 

and wildlife species (BLM 2007). The persistence of 

this water since the late Pliocene/early Pleistocene has 

allowed for the continued existence of relict plants and 

animals which gained access to the region during pluvial 

climates. The isolation of these species in this harsh 

environment permitted their differentiation from related 

taxa and resulted in the distinctive character of many 

present-day occupants (Reveal 1980). The onset of 

more xeric conditions isolated Ash Meadows, thereby 

77


Figure 1. Location of the Ash Meadows National Wildlife Refuge. 

78


prohibiting genetic exchange with nearby populations, 

leading to progressive differentiation in plant and animal 

species now endemic to the area (Reveal 1980). 

Reveal (1980) concluded that four of the endemic Ash 

Meadows species (Astragalus phoenix, Mentzelia leucophylla, 

Grindelia fraxinopratensis and Centarium 

namophilum) are most closely related to congeners presently 

found in montane portions of the Intermountain 

Region. Their persistence to the present day is attributed 

to successful adaptation to a more xeric environment, 

the local persistence of water, and to relatively cool 

temperatures created by cool air drainage from the surrounding 

mountains (Beatley 1977, Reveal 1980). 

Current-day vegetation at AMNWR is composed of a 

typical Mohave creosote shrub vegetation community in 

addition to emergent marshes, wet meadows, distinctive 

spring complexes, alkaline desert uplands, and velvet 

ash community assemblages, several of which are 

known to exist only within the Refuge (Figure 2). These 

vegetation communities provide habitat for several rare 

and endangered plants, including endemic species, as 

well as federally-listed fish and wildlife species (Bio- 

West 2007). 

CREATION OF THE REFUGE 

Several legal and management documents led to the 

establishment of AMNWR in 1984. Devil’s Hole National 

Monument was declared by presidential proclamation 

in 1953, and federal water rights for it were adjudicated 

by the Supreme Court in 1976. The 1966 National 

Wildlife Refuge System Administration Act provided 

direction on Refuge management responsibilities 

and guidance. The Endangered Species Act of 1973, as 

amended, provided authority for appropriate protection 

and management of federally listed species. The U.S. 

Fish and Wildlife Service (USFWS) prepared a Warm 

Springs Pupfish Recovery Plan in 1976, and a Devil’s 

Hole Pupfish Recovery Plan in 1980. The Refuge was 

established on 18 June, 1984 with the purchase of 

12,654 acres of land from The Nature Conservancy. 

The Refuge now occupies a total of 23,488 acres in 

the Ash Meadows valley. Since designation, several 

documents have guided Refuge management. The 1987 

Ash Meadows Refuge Management Plan outlined general 

principles for management of Refuge ecosystems 

and listed species. The 1990 AMNWR Recovery Plan 

for the Listed Species of Ash Meadows outlined recovery 

needs for 12 listed species, and identified tasks to be 

completed to recover and downlist or delist endangered 

species (Sada 1990). In addition to the individual threatened 

and endangered plants and animals of Ash Meadows, 

the plan recognized the need for the recovery of 

Ash Meadows habitats, processes and ecosystems. The 

plan also included specific guidance on management 

objectives (Sada 1990). In 2000, an Environmental Assessment 

was completed (Otis Bay and Stevens Ecological 

Consulting 2006). 

ECOSYSTEM RESTORATION 

Managers at AMNWR are seeking to restore impacted 

wetland and desert upland habitats to conditions 

that existed 100 years ago in an effort to promote endemic 

species recovery. Large-scale ecosystem restoration 

plans include impacted habitats in the Carson 

Slough and Crystal Reservoir areas. Successful restoration 

projects have been completed at Kings Pool in the 

Point of Rocks area, and at Jackrabbit Spring, where 

visitors can view desert pupfish (Bio-West 2008). 

While the Refuge still supports a complex system of 

important native communities, portions have been significantly 

impacted by historic agricultural and mining 

activities. Impacts have included peat mining in marshlands 

surrounding Carson Slough in the 1960s and alfalfa 

farming and cattle grazing in the 1970s. These activities 

reduced discharge from all springs, and many 

spring outflows were channelized (Sada 1984). A complex 

irrigation system was constructed to support farming 

efforts, and resulting agricultural impacts included 

grading, cutting irrigation trenches, pumping spring 

pools, and creating water holding areas such as Crystal 

Reservoir (Otis Bay and Stevens Ecological Consulting 

2006). Land disturbances created by farming and grazing 

activities, as well as severe alteration of an ecosystem’s 

hydrology, can cause considerable change in 

vegetation community composition and allow for the 

encroachment of weedy and non-native species (Fraser 

and Martinez 2002). 

Two of the most obvious chronic threats to AMNWR 

species and ecosystems involve flow modification and 

land conversion associated with former agricultural development 

and invasive species. The most severe longterm 

threat to the Refuge is potential future groundwater 

extraction from the regional carbonate aquifer (Otis Bay 

and Stevens Ecological Consulting 2006). Currently, the 

Refuge is developing large-scale habitat restoration 

plans for Carson Slough, Crystal Reservoir, and other 

springs and areas around the Refuge. The proposed 

plans consider the removal of the remaining irrigation 

system and water control structures in an effort to restore 

the hydrology and geomorphology of the Refuge 

to a natural system. The hydrologic restoration will also 

support vegetation community and wildlife habitat restoration 

attempts. However, restoring an existing water 

system that has supported an area for decades could 

have unintended consequences. 

Within the Refuge, impacts resulting from the historic 

alterations to the landscape are evident both in the 

extensive monocultures of salt cedar (Tamarix ramosis- 

79


Figure 2. Vegetation community type map of the Ash Meadows National Wildlife Refuge. 

80


sima) and Russian knapweed (Acroptilon repens) found 

throughout Carson Slough, as well as in the parceled 

wet meadows created adjacent to Rogers Spring where 

spring water once flowed unobstructed into the slough. 

The Refuge is in the habitat restoration stage and will 

remain so for many years. Goals of the restoration plan 

include restoring natural hydrology and native vegetation 

communities, establishing a baseline of existing 

vegetation communities, and managing and recovering 

rare and endangered species occurring on the Refuge 

(McKelvey 2007). The U.S. Fish and Wildlife Service 

is conducting long-range, strategic management and 

restoration planning at AMNWR to accomplish the recovery 

goals of ecosystem and species restoration. The 

recovery objective for the Refuge is: “delisting for all 

species but the Devil’s Hole pupfish, which can only be 

downlisted to threatened status” (Sada 1990). 

PRELIMINARY VEGETATION MAPPING AND 

CLASSIFICATION 

The consulting firm BIO-WEST, Inc. undertook two 

studies designed to assist the AMNWR with its restoration 

efforts. The first study involved mapping all vegetation 

communities to a fine scale (0.25 acres), with the 

objective of providing a baseline data set for evaluation 

of management actions and future vegetation change, 

while the second included a comprehensive survey of 

distribution and abundance for twelve rare plant species. 

Completion of vegetation mapping on the Refuge has 

resulted in 6,237 delineated polygons (Figure 2). Of the 

delineated polygons, 5,913 (or 95 percent) have been 

assigned a preliminary alliance. Alliance assignments 

will be referred to as preliminary until classification is 

finalized. These classifications are being derived from 

multiple sources including Alliances of the Mojave Desert 

(USGS 2004), the National Vegetation Classification 

System (Grossman et al. 1998), and community 

data available on NatureServe (USGS 2004). 

As currently assigned, the alliances comprising the 

highest total acreage on the Refuge are the Atriplex confertifolia 

Shrubland Alliance, the Larrea tridentata- 

Ambrosia dumosa Shrubland Alliance, and the Isocoma 

acradenia Shrubland Alliance. Several delineated vegetation 

communities do not correspond to any previous 

classifications. Often this is the result of a Refuge community 

that has a typical dominant species occurring 

with an atypical co-dominant. An example of this is the 

Atriplex confertifolia Shrubland Alliance compared with 

the Atriplex confertifolia-Suaeda moquinii Shrubland 

Alliance. The first is a common community throughout 

the desert southwest. However, communities with both 

Atriplex confertifolia and Sueda moquinii occurring as 

co-dominants have not been classified. In these cases a 

new alliance classification may be written to best represent 

the vegetation community. Once final classifica- 

tions have been assigned, botanical descriptions will be 

developed for each of the newly created alliances. 

Association classifications are also currently in the 

developmental stage. Many of the common botanical 

associations are applicable to communities at the Refuge. 

However, a significant number of the delineated 

polygons contain associations of plants that are not 

commonly recognized in current classifications. These 

associations are being thoroughly researched in order to 

identify appropriate resources for classification assignments. 

As with the alliance classifications, we expect to 

develop several new association classifications to accurately 

describe the communities at the Refuge (BIO- 

WEST 2008). 

The vegetation mapping effort has resulted in a 

clearer picture of the diverse composition of vegetation 

communities that exist within AMNWR (Figure 2). 

Common vegetation communities that have been identified 

include Alkali Sink (an extensive shrubland community 

dominated by succulent shrubs such as Mojave 

seablite, [Suaeda moquinii] that occur adjacent to seasonally 

flooded wetlands and along desert washes), as 

well as a variety of wetland communities. Lowland Riparian 

Woodlands are found in the lowest elevations of 

the Refuge; they support a variety of canopy species 

such as velvet ash (Fraxinus velutina), mesquite 

(Prosopis), and narrow-leaf willow (Salix exigua). One 

of the more unique community types is the Alkali Playa 

community found west of Crystal Reservoir and Lower 

Marsh. This community may support the largest populations 

of rare and endemic plants on the Refuge. 

The western portion of the Refuge also supports 

well-established populations of salt cedar, Russian 

knapweed, and five-hook bassia (Bassia hyssopifolia). 

The salt cedar communities function as riparian woodlands 

and in some cases may provide important wildlife 

habitat. Several of the abandoned agricultural fields 

have been infested with five-hook bassia. However, 

these fields currently receive enough seasonal inundation 

to sustain recruiting populations of native wetland 

vegetation. 

Fairly intact transitional Upland and Desert Shrubland 

communities are present in the central and eastern 

portions of the Refuge. The Alkali Shrub community 

type transitions from a mesic phase as the topography 

rises in elevation from west to east. A noticeable 

change in vegetation composition occurs as whiteflower 

rabbitbrush (Chrysothamnus albidus)-dominated communities 

become replaced by alkali goldenbush 

(Isocoma acradenia) in the higher elevations where the 

water table is less accessible. Moving up into the alluvial 

fans, what was once classified as part of Creosote 

Shrubland is now classified as Salt Desert Shrubland 

composed of desert holly (Atriplex hymenelytra), shad- 

81


scale saltbush (Atriplex confertifolia), and spiny hopsage 

(Grayia spinosa). 

Creosote bush-dominated communities tend to persist 

in the alluvial fans east of Devil’s Hole and south 

along the eastern Refuge boundary in dry uplands and 

pavement soils. Approximately 5,000 acres of Creosote 

bush Shrubland is found throughout the NE corner of 

the refuge. It is dominated by creosote bush (Larrea 

tridentata) and white bursage (Ambrosia dumosa) and is 

one of the most common vegetation types in the Mojave 

Desert (MacMahon 2000). 

The central-eastern portions of the Refuge are home 

to some of the most extensive old field disturbances and 

remnants of agricultural activities conducted prior to the 

Refuge’s inception. Many of the non-native old fields 

contain Russian knapweed (Acroptilon repens), thistle 

spp. (Cirsium spp.) and tamarisk (Tamarix spp.). Old 

field non-native vegetation is largely the result of physical 

manipulation of land and land cover during agricultural 

uses. Old agricultural fields around the dam northeast 

of Point of Rocks are largely covered in annual 

grasses and forbs but some creosote bush and mesquite 

are also recolonizing. The Alkali Seep and transitional 

shrubland communities east of the Cold Spring private 

property have emerged as an important location for several 

endemic species such as Ash Meadows sunray, Ash 

Meadows blazingstar, and Ash Meadows ivesia. 

Alkali Meadow is a community exclusive to the 

Amargosa Valley and Owens Valley ecosystems. The 

community is a low-elevation grassland, typically with 

moist alkaline soils. Evaporation of surface water often 

leaves a crumbled salt crust over the soils. Alkali 

Meadow is dominated by inland salt grass (Distichlis 

spicata) and alkali sacaton (Sporabolis airoides). Arctic 

rush (Juncus arcticus) and whiteflower rabbitbrush are 

associated species. The federally listed spring loving 

centaury and Ash Meadows ivesia are found in this 

community. Alkali meadows are indicative of shallow 

water tables. Alkali flats peculiar to Ash Meadows sustain 

the highest concentrations of the federally-listed 

Amargosa niterwort. 

Extensive Alkali Shrubland communities dominated 

by Atriplex species (A. lentiformis, A. canescens, and A. 

confertifolia) are found in areas where groundwater is 

approximately 2-6 meters deep. At AMNWR, Alkali 

Shrubland covers 5,000 acres and comprises over 20% 

of the area. Alkali Meadow and Alkali Shrubland vegetation 

are distributed in close proximity to one another. 

In many places, there are raised mounds where the 

groundwater may be slightly deeper than surrounding 

alkali meadows. In these places, saltgrass, alkali sacaton, 

and Atriplex shrub cover increases. Other shrub 

species include matchbrush (Gutierrezia sarothrae), 

alkali goldenbush, and greasewood (Sarcobatus vermiculatus). 

Mesquite bosque vegetation is found pre- 

dominantly around spring vents and outflow channels. 

The dominant overstory species include mesquite 

(Prosopis spp.), Fremont cottonwood (Populus fremontii), 

and velvet ash (Fraxinus velutina). 

Emergent vegetation is found only at the springs and 

along permanent lakes and reservoirs. Emergent vegetation 

covers about 130 acres and comprises 0.5% of the 

refuge. Common species include Typha spp., spikerush 

(Eleocharis) spp., bulrushes, and rush species. Wetlands 

associated with the spring complexes include both native 

and non-native plant associations. 

RARE PLANT SURVEY METHODS 

Prior to the establishment of the Refuge, little quantitative 

information was available on the life history 

strategies, population genetics, demography, community 

associations, habitat requirements or abundance of plant 

species endemic to Ash Meadows. Implementation of 

the recovery plan for the seven listed endemic plant species 

will be successful only when these characteristics 

are known and disturbed environments are appropriately 

managed. Five additional at-risk plant species have been 

identified as species of concern at Ash Meadows, bringing 

the total to 12 targeted species for rare plant studies 

(Tables 1, 2). 

Beginning in 2007, BIO-WEST conducted the first 

comprehensive inventories of rare, endemic and listed 

plant species that occur within Refuge boundaries. The 

purpose of the rare plant studies was to obtain a baseline 

of existing ecological conditions and rare plant distributions 

as a foundation for future monitoring of changes in 

the status of rare plant species. The information provided 

by these studies will assist with planning future 

habitat-restoration activities and can be correlated with 

wildlife studies to understand benefits and potential detriments 

resulting from these management strategies. 

Floristic field surveys were initiated in March 2007 

and continued through late October 2008. Survey methods 

and intensity depended on the size of the area, investigator 

skill, size of the target species, and topography. 

Prior to conducting the field surveys, the area was 

analyzed to determine potential habitat for each species 

of interest. Potential habitat types for each species were 

identified in part by literature, background maps provided 

by Refuge staff, and observations from field visits 

in prior years. An intuitive controlled survey (the most 

commonly used and efficient method of surveying for 

rare plant abundance), was initially employed in 2007. 

The data from these initial surveys were used to design 

a more detailed methodology for surveying the distribution 

each target species during the 2008 field season. 

Surveyors at each population documented the estimated 

number of individuals, plant phenology, population distribution 

in terms of approximate area, and associated 

vegetation. As a general rule plant occurrences less than 

82


Table 1. Status of Endangered, Threatened, and Sensitive species occurring in Ash Meadows National 

Wildlife Refuge 

Taxon Name Agency Status NV Heritage 

Ranks 

Scientific Vernacular FWS BLM FS NV NNPS Global State 

Arctomecon merriamii White bearpoppy S S W 3 3 

Astragalus phoenix Ash Meadows milkvetch T S F T 2 2 

Calochortus striatus Alkali mariposa lily S S W 2 1 

Centaurium namophilum Spring-loving centaury T S F T 2Q 2 

Cordylanthus tecopensis Tecopa birds’-beak S T 2 2 

Enceliopsis nudicaulis var. 

corrugata 

Ash Meadows sunray T S F T 2 2 

Eriogonum concinnum Darin buckwheat W 2 2 

Grindelia fraxinopratensis Ash Meadows gumplant T S F T 2 2 

Ivesia kingii var. eremica Ash Meadows ivesia T S F T 1-2Q 1-2 

Mentzelia leucophylla 

Ash Meadows blazingstar 

T S F T 1Q 1 

Nitrophila mohavensis Amargosa niterwort E S F E 1 1 

Phacelia parishii Parish phacelia W 2-3 2-3 

Salvia funerea Death Valley sage W 3 1 

Sisyrinchium funereum 

Death Valley blue-eyed 

grass 

T 2-3 1 

Sisyrinchium halophilum Nevada blue-eyed grass 4-5 4 

Sisyrinchium radicatum 

Spiranthes infernalis 

St. George blue-eyed 

grass 

Ash Meadows ladiestresses 

W 2?Q 1-2 

T 1 1 

FWS (Endangered Species Act administered by USDI Fish and Wildlife Service): E - endangered; T - threatened 

BLM/FS (USDI Bureau of Land Management, USDA Forest Service): S - special status or sensitive; W - watch 

NV (state of Nevada): F - fully protected 

NNPS (Nevada Native Plant Society): W - watch list; T - threatened; E - endangered 

Nevada Heritage Program ranks (Global = worldwide; State = within Nevada): 1 - critically imperiled; 2 - imperiled; 

3 - vulnerable; Q - taxonomic status in question; ? - rank uncertain 

83


0.25 acres in size were documented as a point feature 

and occurrences larger than 0.25 acres were mapped as 

polygon features. Specific systematic methods of sampling 

for the 12 rare species of interest used during the 

2008 field season are shown in Table 2 (Ballard 2008). 

RARE PLANT SURVEY RESULTS 

Amargosa niterwort (Nitrophila mohavensis): 

Listed on May 20, 1985, Amargosa Niterwort is the 

only endangered plant at Ash Meadows and is restricted 

to the Amargosa River drainage (Knight and Clemmer 

1987). This member of the Chenopodiaceae is an extremely 

hardy dwarf rhizomatous perennial that is tolerant 

of high soil salinity and alkalinity. It occupies the 

most localized habitat of any plant species endemic to 

Ash Meadows and is often the only species present in its 

habitat. N. mohavensis is found in areas with heavy salt 

crusts created by evaporation of standing water. These 

sites are characterized by barren, moist alkali flats with 

sandy loam soils (~57% sand) encrusted with a layer of 

salt with a pH near 8.4. Distichlis spicata (inland saltgrass) 

is sometimes found either on the periphery, or 

occasionally intermixed within Amargosa niterwort 

populations (Mozingo and Williams 1980). Without ad- 

equate surface water, this habitat may be taken over by 

saltgrass. Reveal (1978a) noted that Amargosa niterwort 

is sensitive to disturbance and does not reinvade sites 

where salt crust overlying the soil has been disturbed. 

Additional associated species include Atriplex confertifolia 

(shadscale saltbush), Mojave seablite, and a more 

widely distributed congener Nitrophila occidentalis 

(boraxweed; Soil and Ecology Research Group 2004). 

Two other listed species, Ash Meadows ivesia (Ivesia 

kingii) and Tecopa birds beak (Cordylanthis tecopensis) 

are also found in this type of habitat. 

At the time of listing, the only known location for 

Amargosa niterwort was Tecopa, California (Otis Bay 

and Stevens Ecological Consulting 2006). Since that 

time, several populations have been documented at the 

Refuge and just outside its western boundary. The Nevada 

Natural Heritage Program (NNHP) estimated the 

population of this species across its entire range at 

13,000 individuals in 1997 (Morefield 2001). According 

to the recently published five-year review for the 

species, the Crystal Reservoir population was estimated 

at 10,000 ramets (above-ground stems), and the West 

Refuge Boundary population was estimated at approximately 

50 ramets (USFWS 2007b). The two popula- 

Table 2. Systematic surveying protocols according to species. 

Scientific Name Common Name Sampling Protocol 

Arctomecon merriamii White bearpoppy Individual count census and transect method 

Astragalus phoenix Ash Meadows milkvetch Transect method 

Calochortus striatus Alkali mariposa lily Individual count census 

Centaurium namophilum Spring-loving centaury Population census and negative sampling 

Cordylanthus tecopensis Tecopa bird’s-beak Population census 


corrugata 

Ash Meadows sunray 

84 

Transect method 

Eriogonum concinnum Darin buckwheat Not located 

Grindela fraxinopratensis Ash Meadows gumplant Population census 

Ivesia kingii var. eremica Ash Meadows ivesia Population census 

Mentzelia leucophylla Ash Meadows blazingstar Individual count census and transect method 

Nitrophila mohavensis Amargosa niterwort Individual count census 

Phacelia parishii Parish phacelia Not located 

Salvia funerea Death Valley sage Not located 

Sisyrinchium spp.* Blue-eyed grass Population census 

Spiranthes infernalis Ash Meadows ladies-tresses Individual count census and population census 

*Includes Sisyrinchium funereum, S. halophilum, and S. radicatum.


tions mentioned in the five-year review were the only 

populations known to occur on the Refuge at that time. 

Surveys conducted during 2008 extended the 

boundaries of the two known populations and added 

new occurrences. Some new populations were surveyed 

in the area just north of the outflow canal of Crystal 

Reservoir. In addition, occurrences noted during the 

2007 reconnaissance that were scattered across the west 

shore of Lower Marsh were resurveyed as well as populations 

found southwest of Crystal Reservoir. These areas 

included the drainage from Crystal Reservoir just 

north of the Refuge boundary and the Big Spring and 

Jackrabbit Spring drainage complex toward the western 

Refuge boundary. Also, a portion of critical habitat located 

in the west corner of the Refuge directly west of 

the Lower Marsh access road was mapped and inventoried. 

This population is referred to as the “Central 

Carson Slough” population by USFWS (2007b). The 

population was mapped in its entirety, including portions 

that fell just outside the Refuge boundary, and the 

entire population estimate for this polygon was included 

in the 2008 total. 

White Bearpoppy (Arctomecon merriamii): White 

bearpoppy is a Mohave Desert endemic known from 

Clark, Lincoln, and Nye Counties in Nevada, and from 

the Death Valley region of California. This species occurs 

in salt desert shrub communities on ridges, rocky 

slopes, gravelly canyon washes, and old lakebeds de- 

rived from carbonate rock sources, often in hard clay 

soils or with shadscale saltbush. It is a clump-forming 

perennial plant with large white flowers borne individually 

on the tips of leafless stems. A. merriamii can be 

distinguished from the golden-flowered Las Vegas bearpoppy 

(A. californica) by its scapose stems, larger capsules, 

and flower color. 

The NNHP rare plant fact sheet states that there have 

been approximately 129 occurrences documented 

throughout its Mojave Desert range. The estimated 

range-wide population is > 20,000 individuals (Morefield 

2001) (Table 3). While there is no documented 

evidence of the number of known individuals within the 

Refuge prior to this study, it is believed that the distribution 

has remained limited with low abundance of individuals 

(H. Hundt, AMNWR, 2007, pers. comm.). 

Field crews conducted reconnaissance in areas of 

potential habitat for this species between 2,000 and 

6,200 feet in elevation and within Salt Desert Scrub 

communities on alluvial gravel substrates. These areas 

included the northernmost portion of the Refuge just 

north and south of the Invite Road, the area surrounding 

Devils Hole, the alluvial fans surrounding Point of 

Rocks, and the extreme southeast corner of the Refuge. 

No plants were found during these searches. However, 

several previously undocumented populations were discovered 

and surveyed throughout the 2008 field season. 

These included the area just south of Peterson Road 

Table 3. Population estimates for surveyed plant species at Ash Meadows NWR. 

Scientific Name Common Name Most Recent Population 

Estimate 

85 

2008 Survey Population 

Estimate AMNWR 

Arctomecon merriamii White bearpoppy 20,000* 193 

Astragalus phoenix Ash Meadows milkvetch 1800 11,643 

Calochortus striatus Alkali mariposa lily unknown 6984 

Centaurium namophilum Spring-loving centaury 4290* 4,468,571 

Cordylanthus tecopensis Tecopa bird’s-beak 4379* 829,918 


corrugata 

Ash Meadows sunray 1849 50,954 

Grindela fraxinopratensis Ash Meadows gumplant 81,000 376,632 

Ivesia kingii var. eremica Ash Meadows ivesia 3862 486,798 

Mentzelia leucophylla Ash Meadows blazingstar 358 3763 

Nitrophila mohavensis Amargosa niterwort 10,050 78,406 

Sisyrinchium spp.* Blue-eyed grass unknown 99,822 

Spiranthes infernalis Ash Meadows ladies-tresses 1107 14,209 

* Range-wide estimate


near the Cold Spring private property, just northwest of 

Longstreet Road but south of Peterson Road, and near a 

large spring drainage between the eastern Refuge border 

and Longstreet Road. Many of the occurrences were 

found within habitat for species such as Ash Meadows 

sunray, Ash Meadows milkvetch, and Ash Meadows 

blazingstar. These habitats typically included mesic Alkali 

Shrublands with sandy soils and occasional deep 

washes. There is some evidence from the NNHP that 

occasionally moist sandy soils could serve as potential 

habitat indicators for this species (Moorefield 2001). 

The broad transect survey methods for the Ash Meadows 

sunray and Ash Meadows milkvetch likely contributed 

to the discovery of the new populations. The total 

population surveyed in 2008 was approximately 193 

individuals (Table 3). There is some potential for locating 

additional populations as the Ash Meadows sunray 

surveys are completed during the 2009 field season. 

Ash Meadows Milkvetch (Astragalus phoenix): 

Ash Meadows milkvetch is endemic to AMNWR. A 

federally protected species, it has been documented 

fairly extensively in the past, including studies directed 

at recovery of the species. The NNHP documents 13 

occurrences for a total estimated population of 1,943 

individuals within the Ash Meadows area (Morefield 

2001). Previous surveys conducted by the USFWS in 

2000 documented several populations within the Refuge 

totaling 1,800 individuals (Pavlik and Stanton 2006). 

Prior to 2006 populations were known from south of 

Rogers Spring and west through the northern portion of 

Purgatory, within the Cold Springs private property, 

south of Bradford Spring, east and west of Ash Meadows 

Road, and north and south of South Spring Meadows 

Road. Survey areas included potential habitat consisting 

of alkaline soils, desert washes, and barren flats 

(Reveal 1978b). Because this is commonly known to 

occur in habitats similar to those of Ash Meadows Sunray, 

both species could be surveyed together. 

Several new populations of Ash Meadows milkvetch 

were discovered during the Ash Meadows Sunray transect 

surveys. Large populations were discovered adjacent 

to the Cold Spring private property. Other notable 

populations were inventoried in the area between 

Rogers and Longstreet Springs, directly west of the 

junction of Ash Meadows Road and South Spring 

Meadows Road, and west of Jack Rabbit Spring. The 

estimated total population is approximately 11,643 individuals 

(Table 3). 

Alkali Mariposa Lily (Calochortus striatus): 

NNHP reports only four occurrences of alkali mariposa 

lily across its entire known range in Clark and Nye 

counties, Nevada and adjacent California. Morefield 

(2001) lists the estimated population of the species as 

“unknown.” 

Several populations documented during the 2007 

reconnaissance at AMNWR were surveyed in 2008. 

These populations included a number of locations immediately 

south of Collins Ranch, just west of Warm 

Springs and north of the access road to Bole Spring. 

Several new populations were located and surveyed including 

one at the bend in West Spring Meadows Road. 

A large population was surveyed within an Alkali 

Shrubland community east of Crystal Reservoir. An additional 

population was mapped and surveyed in the 

southeast corner of the Refuge. The recorded population 

for species surveyed during this study totals 6,984 individuals 

(Table 3). 

Spring-loving Centaury (Centaurium namophilum): 

Spring-loving centaury is an annual plant that 

is endemic to AMNWR and its immediate surroundings. 

It is currently listed as a threatened species by 

the USFWS. The last confirmed survey reported by 

NNHP was in 1986 and documented 19 occurrences for 

an estimated population in excess of 4,290 individuals 

(Morefield 2001). The draft five-year review mentions 

six mapped populations within the Refuge totaling over 

2,900 acres in comparison to the approximately 29 acres 

last reported to the NNHP (Morefield 2001, USFWS 

2008). According to the draft five-year review, population 

trends are insufficiently documented (USFWS 

2008). 

Survey area criteria for this species included seeps, 

wet meadows, and spring channel banks throughout 

AMNWR. In 2008, spring-loving centaury was found 

to be very widespread across the Refuge, populating 

habitats from seasonally flooded wetlands to seasonally 

moist Alkali Meadows and the edges of some Alkali 

Shrubland communities. It appeared that nearly any location 

on the Refuge containing surface or sub-surface 

water at any time during the year would produce a 

population. As surveys continued, a blooming trend for 

certain populations became apparent. Blooms were seen 

in “waves” for individual populations and subpopulations, 

or different parts of a single population would 

bloom at different times during the season. 

In the Peterson Reservoir area, extensive populations 

extended throughout surrounding drainages. As in the 

Rogers Spring and Carson Slough drainages, populations 

extend until they encounter what may be hydrologic 

barriers. Observed occurrences were so extensive 

that it became necessary to map areas of nonoccurrence. 

The total population from the 2008 surveys 

has been estimated at 4,468,571 individuals. 

Clearly the current population estimates are a significant 

increase from the last confirmed survey data provided 

to the NNHP (USGS 2004). It is clear that populations 

of this annual plant fluctuate widely from season 

to season; however, the likelihood that the number of 

86


individuals would dip as low as previously recorded 

estimates seems doubtful. 

Tecopa Bird’s-Beak (Cordylanthus tecopensis): 

Tecopa bird’s-beak is a hemiparasitic summer annual 

plant that is a Nevada Sensitive Species. It is known 

from approximately ten extant occurrences across a narrow 

range in California (Death Valley) and Nevada 

(AMNWR) (Morefield 2001). Its habitat includes Mohave 

Desert scrub and alkali flats and meadows below 

2,700 feet. It always grows with Distichlis spicata, 

which may be its principal host. Tecopa bird’s-beak is 

also a known associate of the spring-loving centaury 

and often occurs within the same habitat types (Otis Bay 

and Stevens Ecological Consulting 2006). Population 

estimates provided by the NNHP document >4,379 total 

individuals. 

Because of their similar habitat requirements, populations 

of Tecopa bird’s-beak were mapped and surveyed 

in conjunction with spring-loving centaury in 

2008. New occurrences were discovered along the 

shores of lower Crystal Marsh as well as on the west 

side of the marsh within old agricultural fields. The agricultural 

field population extended intermittently to the 

western Refuge boundary. In addition, a significant 

population was found associated with a new springloving 

centaury population in a narrow band of velvet 

ash located northwest and southeast of Collins Ranch. 

This area appears to be an important site for multiple 

rare and endemic species at the Refuge. The total population 

of Tecopa bird’s beak documented in the 2008 

field season is approximately 829,918 individuals. 

Ash Meadows Sunray (Enceliopsis nudicaulis var. 

corrugata): Ash Meadows sunray is an endemic variety 

of a widely distributed species that has been listed as 

threatened by the USFWS. The variety is almost strictly 

endemic to Ash Meadows with a few individuals reported 

from outside the Refuge in eastern California. It 

is largely restricted to strongly alkaline, poorly drained, 

saline soils associated with springs and dry washes but 

with the water table some distance below the surface. 

Lower elevation alkali clay soils in Ash Meadows have 

a shallow underlying water table that makes the habitat 

unsuitable. This species is associated with Ash Meadows 

milkvetch, shadscale saltbush, matchbrush, alkali 

goldenbush, basin yellow cryptantha (Cryptantha confertifolia) 

and white bearpoppy at elevations from 2,100 

to 2600 feet. It is generally found on dry to sometimes 

moist sites that are on open, hard, white clay hills with 

calcareous hardpans. Populations on the Refuge are 

found in occasionally moist alkaline soils, spring and 

seepage areas, and dry desert washes. The plants can 

also occasionally be found in salt desert shrubland and 

desert pavement habitats (Morefield 2001; Otis Bay and 

Stevens Ecological Consulting 2006). The last con- 

firmed population estimates for this plant were reported 

following its listing as a threatened species. 

The 2008 survey was directed at locating the plants 

throughout all potential habitat types occurring within 

the Refuge. Cruise transects 40 meters apart were used 

to survey large tracts of Ash Meadows. Of the more 

than 9,000-acres of potential habitat, nearly 6,000 acres 

have been surveyed to date. Ash Meadows sunray has 

been found throughout the areas mapped by the Refuge 

in 2006. In several cases known populations have been 

extended beyond previous distribution boundaries. New 

occurrences were documented west of known populations 

mapped along Ash Meadows Road, as well as on 

the alluvial fans east of Point of Rocks and south of 

Jackrabbit Spring. A single occurrence was also documented 

adjacent to Lower Crystal Marsh. The preliminary 

population estimate, calculated with approximately 

two-thirds of the survey complete, is 50,954 individuals. 

The remaining survey area includes habitat within the 

central portion of the Refuge that has long been known 

to support this plant. It is likely that upon completion of 

the surveys, the final population estimate will increase 

by several thousand. 

Ash Meadows Gumplant (Grindelia fraxinopratensis): 

Ash Meadows gumplant was listed as threatened 

by the USFWS in 1985. The plant is considered an 

endemic species primarily occurring within AMNWR 

with a limited distribution in neighboring Inyo County, 

California. NNHP documented 16 occurrences across 

the known range for the species and estimated a population 

of more than 13,000 individuals in 1986 (Morefield 

2001). The USFWS estimated the Refuge’s population 

at approximately 81,000 individuals following a 1998 

survey (USFWS 2007a). 

Distribution data provided by the USFWS in 2000 

and Refuge staff in 2006 indicate that populations of G. 

fraxinopratensis occur in spring drainages and marsh 

habitats throughout the Refuge. Notably, both data sets 

show a significant presence of this species along the 

Fairbanks and Rogers Spring drainages. However, BIO- 

WEST botanists involved with the 2007 reconnaissance 

and the 2008 surveys indicate that plants found at these 

locations are actually not members of this species. At 

this time BIO-WEST has been unable to confirm occurrences 

north or west of the Warm Springs complex. 

The Alkali Meadows south of Crystal Reservoir contain 

very large populations of Ash Meadows gumplant. 

Known populations were also documented in the Alkali 

Meadows of the Big Spring/Jackrabbit Spring drainage 

complex both east and west of South Spring Meadows 

Road. Another known population was inventoried south 

of Ash Meadows Road as it intersects South Spring 

Meadows Road. New populations were surveyed between 

the Warm Springs Complex and West Spring 

87


Meadows Road, as well as just north of Devils Hole 

Road in the Collins Ranch area, east of Crystal Reservoir 

and between the south end of Lower Crystal Marsh 

and the Refuge boundary. Due to a short bloom time, 

several occurrences west of the Refuge office were not 

surveyed in 2008. The total estimated population at this 

time is 376,632 individuals, a number that will likely 

increase upon completion of surveys. 

Ash Meadows Ivesia (Ivesia kingii): Endemic to the 

Refuge, Ash Meadows ivesia is a federally-listed threatened 

species. It is a perennial plant with a prostrate 

growth form. As is the case with many of the endemic 

Refuge plants, there is little current information regarding 

its abundance. According to the USFWS, the distribution 

of this species is limited to Nye County, Nevada, 

and likely limited to within AMNWR boundaries. Population 

estimates by the NNHP in the 1980s indicated 

there were as many as nine occurrences; these areas 

contained an estimated 3,862 individuals (Morefield 

2001). 

The majority of the 2008 surveys confirmed populations 

occurring within historic distribution areas on the 

Refuge. However, field crew members were unable to 

locate plants within several previously documented areas 

believed to contain populations and extant populations 

appeared significantly smaller. However, a visit in 

the Jackrabbit Spring area led to the discovery of several 

new populations. New populations, fairly large in size, 

were also surveyed in close proximity to West Spring 

Meadows Road where it makes a sharp turn to the west. 

Also, multiple, substantial populations were located and 

surveyed between Crystal Reservoir and West Spring 

Meadows Road. Another population was documented in 

a seepage area interrupting an upland habitat just east of 

the Cold Spring private property. Finally, populations 

occurring just north and south of Big Springs Road were 

surveyed. The current estimated population of Ash 

Meadows ivesia is 486,798 individuals (Table 3). 

Ash Meadows Blazingstar (Mentzelia leucophylla): 

This endemic biennial herb was listed as a threatened 

species in 1985. The plant’s distribution appears to be 

strictly limited to areas within the Refuge (Morefield 

2001; Otis Bay and Stevens Ecological Consulting 

2006). Recent information regarding species abundance 

is limited. Surveys in 1986 documented M. leucophylla 

from 8 locations with an estimated population of 358 

individuals (Morefield 2001). Distribution maps provided 

by the Refuge from 2006 internal surveys show 

confirmed populations at Purgatory, the Warm Springs 

Complex, and along West Spring Meadows Road. 

Additional populations of Ash Meadows blazingstar 

were discovered during surveys for Ash Meadows sunray 

and Ash Meadows milkvetch. Because of its biennial 

life form, M. leucophhylla was observed in rosette 

as well as in flower. Starting in mid-March, BIOWEST 

field crews began recording populations south of Peterson 

Road near the Cold Spring private property boundary. 

In addition, new blazingstar populations were documented 

just south of Rogers Spring and west of Longstreet 

Road, intermixed with a large white bearpoppy 

population southeast of the Warm Springs Complex access 

road and directly north of the “T” junction of South 

Spring Meadows Road and West Spring Meadows 

Road. These populations were surveyed and their 

boundaries extended. The estimated population is 3,763 

individuals (Table 3). 

Blue-Eyed Grass Species (Sisyrinchium species): 

Previously, three species of blue-eyed grass were 

thought to occur at the Refuge: Death Valley blue-eyed 

grass (Sisyrinchium funereum), Nevada blue-eyed grass 

(S. halophilum), and St. George blue-eyed grass (S. 

radicatum) (C. Baldino 2007 pers. comm.). However, 

Cholewa (in letter, 2003) suggested that S. halophilum 

probably does not extend as far south as the refuge. She 

based her statement on the discovery that many herbarium 

specimens that had been identified as that species 

were incorrect. Blue-eyed grass species are extremely 

difficult to differentiate, so there is ongoing debate and 

confusion as to exactly which species actually exist on 

the Refuge. 

Such problems are not new in Sisyrinchium. Specieslevel 

taxonomy of Sisyrinchium has long been disputed. 

Recent molecular work has clarified the limits of the 

genus and helped identify important morphological 

characters delineating it from closely related genera 

(Karst, unpublished). However, much more phylogenetic 

work, based on cladistic analysis of molecular 

data, is needed to understand species relationships 

within the genus (Cholewa and Henderson 2002). 

Species of Sisyrinchium are not easily distinguished. 

White flowers may occur in otherwise blue-flowered 

species, and vivipary occasionally occurs, where plants 

produce seeds that germinate before they detach from 

the parent. Furthermore, vegetative characteristics, 

while distinctive in some species, may overlap greatly in 

wide-ranging species. Writers of past floras sometimes 

were unaware of such phenotypic plasticity, or were 

inconsistent in their use of terminology. Some taxonomists 

have thought differences too subtle and chosen to 

lump species (Cholewa and Henderson 1984, 2002). 

Because of the taxonomic confusion surrounding the 

blue-eyed grass plants growing within refuge boundaries, 

the BIO-WEST 2008 field surveys combined all 

occurrences of blue-eyed grass into a Sisyrinchium spp. 

category. 

As documented by Cholewa and Henderson (2002), 

Sisyrinchium funereum populations occur mostly within 

Death Valley. These populations contain numerous individuals. 

Sisyrinchium radicatum is more widely distributed, 

growing in Clark, Nye, and Lincoln Counties, 

88


Nevada, and Washington County, Utah. In Clark 

County, it occurs in Pine Canyon and Ash Spring and is 

also known from Pine Creek and Red Rock Canyon- 

Calico Basin in the Spring Mountains. In Nye County, it 

is known from Big Springs in Ash Meadows, and Pahrump 

Valley. In Lincoln County, it is known from Pahranagat 

lakes and Pahranagat Valley – Ash Springs. The 

threats to these populations are currently unknown. Sisyrinchium 

demissum is a closely related species that 

overlaps the known range of the two species known to 

occur on AMNWR. 

Known populations of blue-eyed grass were visited 

and surveyed throughout the Refuge. Several previously 

undiscovered populations were documented including 

occurrences just south and east of Jackrabbit Spring. 

Large populations were surveyed in the area directly 

south of Crystal Reservoir, expanding habitat from the 

2007 findings. New populations were also located adjacent 

to the Cold Springs private property, with the large 

Ash Meadows ladies-tresses population northeast of 

Rogers Spring but south of Longstreet Road, and in a 

spring drainage leading from the eastern Refuge border 

in the north section of the Refuge. The estimated blueeyed 

grasses population is 99,822 individuals (Table 3). 

Ash Meadows ladies-tresses (Spiranthes infernalis): 

A Refuge endemic, Ash Meadows ladies- tresses 

is currently being considered for Federal listing (Otis 

Bay and Stevens Ecological Consulting 2006). NNHP 

survey records from 1998 show 15 locations with an 

estimated population of 1107 individuals (Morefield 

2001). 

New populations documented during the 2008 survey 

were found in several seep habitats The population was 

estimated at 14,209 individuals. Several occurrences 

were observed late in the field season when the plants 

had passed the flowering period and will be revisited 

during 2009. Because there can be significant variability 

in the number of individuals that bloom from season to 

season, it may be necessary to revisit surveyed locations 

to determine an accurate population estimate after new 

locations are added. 

The 2008 rare plant surveys of the twelve sensitive 

and endemic species that exist at AMNWR revealed 

larger populations and new locations of additional populations 

of these species than had been previously documented 

and aided in determining clearer population estimates, 

population location boundaries, and the associated 

vegetation communities these species exist in. Additional 

planned surveys in 2009 may aid in more accurately 

determining the biogeography of the rare, endemic 

and listed Plants of the Ash Meadows National 

Wildlife Refuge. 

REFERENCES 

Ballard, L.S. 2008. Sampling protocols for rare 

plants at Ash Meadows National Wildlife Refuge. BIO- 

WEST, Inc. 

Beatley, J.C. 1977. Threatened plant species of the 

Nevada Test Site, Ash Meadows, and central-southern 

Nevada. 66 pps. 

BIO-WEST, Inc. 2007. Ash Meadows National 

Wildlife Refuge. 2007 Draft Progress Report. Logan, 

Utah. 

BIO-WEST, Inc. 2008. Ash Meadows National 

Wildlife Refuge. 2008 Draft Progress Report. Logan, 

Utah. 

[BLM] U.S. Bureau of Land Management. 2007. 

Amargosa River Area of Critical Environmental Concern 

Implementation Plan. Barstow (CA): BLM. 19 p. 

plus appendices and maps. 

Cholewa, A. F. and D.M. Henderson. 1984. Biosystematics 

of Sisyrinchium Section Bermudiana 

(Iridadeae) of the Rocky Mountains. Brittonia 36: 342- 

364. 

Cholewa, A.F. and D.M. Henderson. 2002. Sisyrinchium. 

Pp 351-371. In: Flora of North America Editorial 

Committee. Flora of North America North of 

Mexico. Volume 26. Magnoliophyta: Liliidae: Liliales 

and Orchidales. 

Fraser J. and C. Martinez C. 2002. Restoring a desert 

oasis. Endangered Species Bulletin 27:18–19. 

Grossman, D.H., D. Faber-Landgendoen, A.S. 

Weakley, M. Anderson, P. Bourgeron, R. Crawford, K. 

Gooding, S. Landaal, K. Metzler, K. Patterson, M. Pyne, 

M. Reid, and L. Sneddon. 1998. International classification 

of ecological communities: Terrestrial vegetation of 

the United States. Volume I: The national Vegetation 

Classification Standard. The Nature Conservancy, Arlingon, 

Va. 

Knight, T.A., and G.H. Clemmer. 1987. Status of 

Populations of the Endemic Plants of Ash Meadows, 

Nye County, Nevada. Reno, NV: U.S. Fish and Wildlife 

Service, Great Basin Complex. Also available online at 

http://heritage.nv.gov/reports/ashmtext.pdf 

MacMahon, J.A. 2000. Warm Deserts. Pages 285- 

322 in North American Terrestrial Vegetation. M.G. 

Barbour and W.D. Billings eds. 2 nd edition. Cambridge 

University Press, Cambridge, UK. 

McKelvey S. 2007. The feasibility of restoring historic 

Carson Slough. Las Vegas: U.S. Fish and Wildlife 

Service. 16 p. 

Morefield, J.D., ed. 2001. Nevada Rare Plant Atlas. 

Compiled by the Nevada Natural Heritage Program. 

Portland, OR: Fish and Wildlife Service, U.S. Department 

of the Interior, Fish and Wildlife Service, Portland, 

Oregon and Reno, Nevada. http://heritage.nv.gov/atlas/ 

atlastxt.pdf. 

89


Mozingo, H.N. and M. Williams, 1980. Threatened 

and endangered plants of Nevada. An illustrated manual. 

U.S. Fish and Wildlife Service, Portland, OR, and 

U.S. Bureau of Land Management, Reno, NV. 

Otis Bay and Stevens Ecological Consulting. 2006. 

Ash Meadows geomorphic assessment and biological 

assessment: final report. Las Vegas: U.S. Fish and 

Wildlife Service. 149 p. plus appendices. 

Pavlik, B.M. and A.E. Stanton. 2006. Managing 

populations of rare plants at Ash Meadows National 

Wildlife Refuge. I. Demographic Survey and Habitat 

Quality Assessment to Recover Astragalus phoenix. 

Draft. San Francisco: BMP Ecosciences. 33 p. including 

appendices. 

Reveal, J.L. 1978a. Status report on Nitrophila mohavensis 

Munz and Roos. US Fish and Wildlife Service, 

Portland, OR. 

Reveal, J.L. 1978b. Status report on Astragalus 

phoenix. US Fish and Wildlife Service, Portland, OR. 

Reveal, J.L. 1980. Biogeography of the Intermountain 

Region. A speculative appraisal. Mentzelia 4: 1-92. 

Sada, D.W. 1990. Recovery Plan for the Endangered 

and Threatened Species of Ash Meadows, Nevada. U.S. 

Fish and Wildlife Service. Reno, Nevada. 

Soil Ecology and Research Group. 2004. Demographics 

and ecology of the Amargosa niterwort 

(Nitrophila mohavensis) and Ash Meadows gumplant 

(Grendelia fraxinopratensis) of the Carson Slough area. 

16 pps. Webpage: http://www.serg.sdsu.edu/SERG/ 

restorationproj/mojave%20desert/deathvalleyfinal.htm 

[USFWS]. U.S. Fish and Wildlife Service. 2007a. 

Ash Meadows gumplant (Grindelia fraxinopratensis) 

five-year review: summary and evaluation. Las Vegas: 

Nevada Fish and Wildlife Office, Southern Nevada 

Field Office. Pp. 5-7. 

[USFWS]. U.S. Fish and Wildlife Service. 2007b. 

Amargosa niterwort (Nitrophila mohavensis) five-year 

review: summary and evaluation. Las Vegas: Nevada 

Fish and Wildlife Office, Southern Nevada Field Office. 

Pp. 5-7. 

[USFWS]. 2008. Spring loving centaury (Centaurium 

nomophilum) draft five-year review: summary and 

evaluation. Las Vegas: Nevada Fish and Wildlife Office, 

Southern Nevada field office. Pp. 6-7. 

[USGS] U.S. Geological Survey. 2004. Mojave Desert 

ecosystem program: central Mojave vegetation database. 

Final report. Flagstaff (AZ): Colorado Plateau 

Field Station, USGS Southwest Biological Science Center. 

251 p. 

90


Assessing Vulnerability to Climate Change Among the 

Rarest Plants of Nevada’s Great Basin 

Steve Caicco, 

Conservation Planner, Planning Branch, National Wildlife Refuge System, 

Portland, OR 

, 

Abstract. The Great Basin of Nevada provides habitat for many narrowly distributed endemic plant species. To assess 

the vulnerability of 33 of the rarest of these plants to climate change, I used the elevation range of all reported 

locations as a surrogate measure of their bioclimatic envelopes. The results show that 14 of these taxa occur on or 

near the valley floors, nine taxa occur in montane habitats, and 10 taxa occur at high elevations. While the majority of 

the 33 taxa are restricted to highly specialized edaphic habitats, valley endemics are distributed through a smaller elevation 

span than montane or high elevation endemics. In addition, valley habitats have less variability in slope and 

aspect and their highly specialized habitats do not occur above the valley floor. These habitat restrictions are likely to 

constrain migration in response to climate change. Montane and high elevation habitats are more diverse topographically 

and, although often specialized, are more common both locally and regionally. This imposes fewer constraints 

on natural migration and offers more conservation options in the face of climate change. Our inability to accurately 

predict the actual parameters of climate change and its effects at a scale relevant to rare species will require a comprehensive 

inventory and monitoring effort to identify those species affected by climate change. An integrated long-term 

seed storage program will ensure adequate representation for genetic conservation. 

Pollen, woodrat midden, tree-ring, and lake level 

data spanning the past 50,000 years has demonstrated 

that the Great Basin is highly sensitive to climatic 

change (Wharton et al. 1990). During the 20 th Century, 

an average annual warming of 0.3 to 0.6 °C occurred 

and projections for the next century are for an additional 

rise of 2 to 5 °C (Cubashi et al. 2001; Wagner 2003). 

Alterations to the precipitation regimes are harder to 

predict, but seem likely to include a greater proportion 

falling as rain, decreasing winter snowpack, and earlier 

arrival of spring conditions, thereby affecting runoff and 

plant phenology (Mote et al. 2005; Snyder and Sloan 

2005). 

The basin-and-range topography that characterizes 

the Great Basin of Nevada has generated much interest 

among biogeographers and numerous seminal works 

have been published focused on the distribution and relationship 

of its vascular flora or specific aspects of 

plant endemism in this region (Billings 1978; Charlet 

1996; Harper and Reveal 1978; Holmgren 1972a; Pavlik 

1989; Reveal 1979; Shreve 1942; Wells 1983). A published 

proceedings of a symposium on intermountain 

geography includes numerous papers on many aspects 

of biogeography in the Great Basin (Harper and Reveal 

1978). Several sources of information are available on 

the rare plants of Nevada (Morefield 2001; Mozingo 

and Williams 1980) or parts of Nevada (Anderson et al. 

1991; Spahr et al. 1991; Weixelman and Atwood undated). 

A conservation blueprint for the Great Basin has 

been prepared that includes general discussions of the 

ecological systems represented and their associated species 

conservation targets and identifies a portfolio of 20 

priority landscape scale conservation sites; climate 

change and adaptation options are also discussed but no 

explicit assessment of species vulnerability was conducted 

(Nachlinger et al. 2001). 

Narrowly endemic plants are expected to be at far 

greater risk of extinction from climate change than are 

more widespread plants because of their limited range, 

small populations, and genetic isolation (Committee on 

Environment and Natural Resources 2008; Peters and 

Darling 1985). Alpine plants are often identified as at 

particular risk due to isolation and lack of an “escape 

route” (Grabherr et al. 1995; Halloy and Mark 2003). 

Among the observed and predicted effects of climate 

change on plant species are phenological changes, trophic 

level disruptions, range shifts and contractions, and 

extinctions (Parmesan 2006). Documented effects include 

an accelerated trend in upward shift of alpine 

plants in the Swiss Alps over the last few decades 

(Walther et al. 2005), a significant upward shift in optimum 

elevation of forest plants in Europe in the 20 th century 

(Lenoir et al. 2008), a decline of arctic-alpine plants 

from 1989-2002 in Glacier National Park (Lesica and 

McCune 2004), and an advance in the mean flowering 

dates of lilac and honeysuckle in the western United 

States of 2 and 3.8 days per decade, respectively (Cayan 

et al. 2001). 

Extinction is predicted for 3–21 percent of the flora 

in Europe, 38–45 percent in the Cerrado of Brazil, 32– 

91


63 percent in the alpine flora of New Zealand, and 41– 

51 percent of endemics in South Africa and Namibia 

(Fischlin et al. 2007). In the Great Basin, range contractions 

or extinctions have been predicted for higher elevation 

mammals, butterflies, birds and plants (Murphy 

and Weiss 1992; Van de Ven et al. 2007; Wagner 2003). 

Populations of pika (Ochotona princeps) have already 

been reduced by 28 percent compared to the number of 

populations known earlier in the 20 th century (Beever et 

al. 2003). Grayson (2000) has posited that during the 

Middle Holocene (5,000–8,000 years before present), a 

period characterized by a decrease in summer precipitation 

and an increase in temperatures, the small mammal 

fauna of the Great Basin decreased in species richness 

and evenness as a result of a series of local extinctions 

and near-extinctions coupled with an increase in taxa 

well-adapted to xeric conditions. 

The objective of this study was to conduct an initial 

assessment of the vulnerability of the rarest plants of the 

Great Basin of Nevada to climate change based on geographical 

and ecological data from the literature, stored 

in data bases and files maintained by the Nevada Natural 

Heritage Program, Carson City, Nevada, and stored 

in files maintained by the U.S. Fish and Wildlife Service 

at the Nevada Fish and Wildlife Office, Reno, Nevada. 

STUDY AREA 

The study area encompassed the Great Basin within 

the State of Nevada. The term “Great Basin” was first 

used in 1844 by the explorer John Fremont in reference 

to the large closed hydrologic basin lying between the 

Sierra Nevada of California and Nevada to the west and 

the Wasatch Front of Utah to the east with slight extensions 

into Oregon and Idaho (Tingley and Pizarro 2000). 

This analysis focuses on the floristic Great Basin within 

Nevada (Holmgren 1972a), an area of roughly 54,741 

km 2 that includes all of Nevada north of the Mojave Desert 

with the exception of the Carson Range along the 

eastern side of Lake Tahoe (Figure 1). 

In general, precipitation increases and temperature 

decreases with elevation in the Great Basin, although 

physiographic factors can exacerbate temporal and spatial 

variation in climatic patterns. The complex terrain, 

with its large differences in altitude and the consequent 

distortion of air currents creates high variability in local 

precipitation and short periods of intense rainfall followed 

by very long periods without precipitation (Hidy 

and Kleiforth 1990). Rapid heat loss at night results in 

cool air descending to valley floors where pooling in 

closed basins creates diurnal temperature inversions 

(Beatley 1975). Along the southern boundary of the 

study area, air and soil temperature regimes on two valley 

floor sites, separated by only 90 m horizontally and 

1.5 m vertically, were found to influence the distribution 

of dominant shrub species. The composition of herb- 

Figure 1. Location of the study area in the Great Basin 

of Nevada, an area of about 54,741 km 2 ; the larger 

dark outline is the boundary of the Great Basin Restoration 

Initiative as delineated by the U.S. Bureau of 

Land Management based primarily on floristic similarity. 

aceous perennials and winter annuals on the same two 

sites was similar, although temperature influenced the 

initiation of vegetative growth in herbaceous perennials 

and the germination success of winter annuals (Beatley 

1969, 1975). Predicting local climates and the climatic 

responses of highly localized endemic plant populations, 

therefore, is at best a challenging approximation. 

METHODS 

I used ecologic and geographic data stored by the 

Nevada Natural Heritage Program and the U.S. Fish and 

Wildlife Service, including various status assessment 

reports prepared for many of the rare plant taxa, to determine 

the reported minimum and maximum elevations 

of all known populations of the rarest plants within the 

study area. All taxa ranked as G1, G1G2, or T1 were 

included in this study. G1 ranked species are considered 

critically imperiled based on a very high risk of extinction 

due to extreme rarity (often five or fewer populations), 

very steep declines, or other factors; T1 ranks are 

applied to infraspecific taxa that meet the same criteria 

as for the G1 ranks (Natureserve 2009). I assessed a total 

of 167 reported locations of 33 taxa with ranks of 

G1, G1G2, or T1 ranks (Table 1). 

I used the reported minimum and maximum elevation 

of all known locations for each of the 33 taxa as a 

surrogate for the combined effects of temperature and 

92


precipitation, the principal climatic factors affecting 

plant distribution in the Great Basin (Billings 1949; 

Comstock and Ehleringer 1992; Fautin 1946). Van de 

Ven and others (2007) used actual temperature and precipitation 

data to construct bioclimatic models of potential 

climatic change in the White Mountains of California 

and Nevada at the western edge of the Great Basin, 

but such data are relatively sparse for the interior Great 

Basin. Climate stations are typically clustered near 

towns and skewed toward lower elevations; there are 

few climate stations above 2500 m. Moreover, due to 

the complex topography of the Western United States 

and the coarse resolution of most climate models, even 

the best climate models display biases at regional scales 

(Bonfils et al. 2008). All 33 taxa are highly localized 

endemics with geographic ranges that are considerably 

smaller than could be predicted by the best regional climate 

models. 

While approaches that incorporate additional factors 

such as slope, aspect, and geologic substrate have been 

used with some success to develop predictive models of 

potential habitat for a few rare plants in the Great Basin 

of Utah (Aitken and others 2007), field surveys conducted 

by experienced botanists for many of these 33 

taxa have consistently found that only a small portion of 

their predicted range contains suitable habitat. Therefore, 

the objective of this study was to compare the climatic 

niches of rare plants, rather than to develop predictive 

distribution models. 

RESULTS 

The rarest plants of the Great Basin of Nevada (i.e., 

those with a G1, G1G2, or T1 rank) can be placed into 

three broad elevation bands. Those bands reflect occurrence: 

1) below the lower limits of tree distribution on 

or near valley floors; 2) within a narrow montane zone 

dominated by pinyon-juniper woodlands and associated 

shrublands or subalpine; or 3) near or above timberline 

(Figure 2, Table 1). 

The lowest elevation group is comprised of 14 taxa 

that have a median population altitude below 2000 m. 

All but one of these taxa are endemic to Nevada (Figure 

2). Eleven taxa, Eriogonum tiehmii, E. argophyllum, 

E. ovalifolium var. williamsiae, E. diatomaceum, Castilleja 

salsuginosa, Johanneshowellia crateriorum, 

Boechera falcifructa, Mentzelia argillicola, M. tiehmii, 

Frasera gypsicola, and Potentilla basaltica have a reported 

elevational amplitude of less than 244 m, with 

nine of these distributed across less than 129 m of elevation 

(Figure 2). Two additional species, Sclerocactus 

blainei and Mimulus ovatus, are reported from an elevational 

amplitude of less than 50 m. The lone anomaly to 

the general pattern of restricted elevational range among 

the lowest elevation group was Penstemon floribundus, 

with populations spanning 1,009 m. 

The middle elevation group is comprised of nine 

taxa, all but one of which are thought to occur only in 

Nevada (Figure 2). The median altitudes of all known 

populations of these taxa fell between 2,000 m and 

2,746 m, although three taxa have some populations 

near or above 3,000 m. Six of these taxa, Tonestus graniticus, 

Lewisia maguirei, Collomia renacta, Eriogonum 

microthecum var. arceuthinum, E. douglasii var. 

elkoense, and Trifolium andinum var. podocephalum 

had elevational amplitudes of less than 280 m. This relatively 

narrow elevation span is comparable to those 

typical of the valley taxa, and an argument could be 

made to make only two groupings based on a division at 

2,500 m (Figure 2). Such a division would, however, 

mask an ecological distinction based on a notable difference 

between valley and montane endemics in their edaphic 

specialization, discussed in more detail in the following 

section. The remaining three taxa, Penstemon 

tiehmii, P. pudicus, and P. moriahensis had reported 

elevational amplitudes of 640 m, 782 m, and 1,128 m, 

respectively. 

The highest elevation group is comprised of ten taxa, 

seven of which are considered endemic to Nevada. Only 

three of the ten taxa, Draba serpentina, Boechera 

ophira, and Ipomopsis congesta var. nevadensis had 

reported elevational amplitudes of less than 200 m. Two 

taxa, Primula capillaris and Penstemon rhizomatosus, 

had reported elevational amplitudes of 381 m and 366 

m, respectively. The remaining five taxa, Viola lithion, 

Eriogonum holmgrenii, Potentilla cottami, Polemonium 

chartaceum, and Cymopterus goodrichii had reported 

elevational amplitudes between 747 m and 1,158 m. 

Only three, Polemonium chartaceum, Eriogonum holmgrenii, 

and Draba serpentina, were restricted to sites 

above treeline (Figure 2). 

DISCUSSION 

Overview 

Many of the rarest plant taxa in the Great Basin of 

Nevada are found below the lower limits of tree distribution. 

These low elevation taxa (median elevation below 

2,000 m) had the narrowest bioclimatic envelope, as 

estimated from their reported elevational amplitudes, 

with 11 of 14 (79 percent) spanning less than 244 m. If 

other elevation bands are considered, 17 of the 20 taxa 

(85 percent) with an elevational span of less than 280 m 

have a median elevation below 2,377 m (Figure 2). 

Among the ten highest elevation taxa, only three (30 

percent) have a reported elevational amplitude of less 

than 200 m (one of these is likely more widely distributed 

as discussed below), and four of them (40 percent) 

are known to occur over more than 700 m of elevation. 

These results oppose the prediction that alpine species 

are at particular risk due to isolation and lack of an 

93


“escape route” (Grabherr et al. 1995; Halloy and Mark 

2003). Perhaps this can be attributed to unusual features 

of plant endemism in the western United States in general 

and, more specifically, in the Intermountain West. 

Kruckeberg (1986) reformulated earlier works characterizing 

the processes of soil formation and vegetation 

development (Jenny 1941; Major 1951) to account for 

plant diversity in any region. He noted that topography, 

parent material, and the timing of geological processes 

or events created a patchiness or discontinuity of edaphic 

phenomena that creates additional opportunity for 

biological discontinuity. i.e., speciation. He termed endemic 

taxa that result from this process “geoedaphics” 

(Kruckeberg 1986). Rajakaruna (2004) provided a 

review that emphasized the role that unusual soil conditions 

play in the diversification of plant species. 

Although Kruckeberg (1986) emphasized the role of 

bedrock (and especially serpentine) outcrops in the evolution 

of geoedaphics, in an earlier paper Kruckeberg 

and Rabinowitz (1985), cast a broader net with respect 

to narrowly distributed endemics (sensu Mason 

1946a,b), noting that unique taxa associated with 

“gypsum, serpentine, limestone, alkaline and heavy 

metal soils are well known to field botanists in many 

parts of the world.” While few, if any, serpentine outcrops 

are known from the Great Basin in Nevada, the 

Calcareous Mountains Section of the Intermountain region, 

which lies primarily in the eastern half of Nevada 

and adjacent southwestern Utah, is recognized as the 

richest area of the Great Basin for plant endemism 

(Holmgren1972a). Elsewhere, examples of Great Basin 

plants restricted to exposures of carbonate bedrock are 

Figure 2. Elevation ranges of reported localities of the 33 rarest plants (NatureServe G1, G1G2, and T1 ranks) in the 

Great Basin of Nevada. The rectangles in each box-and-whisker plot show the 50 percent of the populations that occur 

between the 25 th and 75 th percentiles and the median elevation. Species are ranked by median elevation. Circles 

indicate plants known from only a single site. Shading shows plants endemic to Nevada. 

94


Table 1. Rare Plant Taxa in the Great Basin of Nevada Assessed 

for Vulnerability to Climate Change. 

Taxon 

No. of 

Pop’s a 

95 

Life 

Form b 

High Elevation Endemics (n=9) 

Habitat(s) 

Polemonium chartaceum H. Mason 1 (6) pf Rocky slopes, talus, fellfields 

Eriogonum holmgrenii Reveal 4 pf Quartzite/limestone outcrops, slopes, ridges 

Draba serpentina (Tiehm & P. Holmgren) Al- 

Shehbaz & Windham 

4 pf Quartzite slopes and cliffs 

Penstemon rhizomatosus N.H. Holmgren 6 pf Limestone talus and cliffs 

Cymopterus goodrichii Welsh & Neese 7 pf Quartzite and limestone talus 

Boechera ophira (Rollins) Al-Shehbaz 13 pf Steep slate or limestone scree/talus 

Ipomopsis congesta (Hook.) V.E. Grant var. 

nevadensis (Tidestr.) Tiehm 

1 pf Carbonate-derived soils or scree? 

Potentilla cottamii N.H. Holmgren 2 (2) pf Quartzite cracks and crevices 

Primula capillaris N.H. Holmgren & A. Holmgren 

4 pf Moist meadows in glacial till 

Montane Endemics (n=10) 

Viola lithion N.H. Holmgren & P.K. Holmgren 5 pf Limestone/dolomite cracks, crevices, 

ledges 

Penstemon moriahensis N.H. Holmgren 8 pf Gravelly and/or silty carbonate soils 

Penstemon tiehmii N.H. Holmgren 3 pf Soil pockets on steep volcanic talus/scree 

Penstemon pudicus Reveal & Beatley 6 pf Volcanic rocky soils, crevices, boulder 

piles 

Lewisia maguirei A.H. Holmgren 8 ge Carbonate scree/shallow soils steep slopes/ 

ridges 

Tonestus graniticus (Tiehm & L.M. Shultz) G.L. 

Nesom & D. Morgan 

1 pf Granite cliffs and outcrops 

Collomia renacta E. Joyal 2 an Volcanic lithosols 

Eriogonum microthecum Nutt. var. arceuthinum 

Reveal 

Trifolium andinum Nutt. var. podocephalum 

Barneby 

2 pf Tuffaceous knolls, bluffs, and rocky flats 

1 pf Tuffaceous bluffs and soils 

Eriogonum douglasii Benth. var. elkoense Reveal 1 pf Sandy to gravelly flats and slopes 

Valley Endemics (n=14) 

Eriogonum tiehmii Reveal 6 pf Rocky clays derived from mixed sedimentary 

rock 

Castilleja salsuginosa N.H. Holmgren 2 pf Seasonally moist alkaline clays/siliceous 

geothermal sinter 

Penstemon floribundus D. Danley 8 pf Volcanic talus, slopes, or colluvium 

Johanneshowellii crateriorum Reveal 7 an Sandy pumice flats and slopes

Taxon 



No. of 

Pop’s a 

Life 

Form b 

Habitat(s) 

Boechera falcifructa (Rollins) Al-Shehbaz 9 pf Zonal soils with big sagebrush zone 

Mentzelia argillicola N.H. Holmgren & P.K. 

Holmgren 

5 pf Alkaline clays/silts of Pliocene lake beds 

Frasera gypsicola (Barneby) D.M. Post 10 pf Alkaline clays/silts of Pliocene lake bed; 

spring mounds 

Mentzelia tiehmii N.H. Holmgren & P.K. Holmgren 

7 pf Alkaline clays/silts of Pliocene lake beds 

Eriogonum argophyllum Reveal 1 an Siliceous geothermal sinter 

Sclerocactus blainei S.L. Welsh & Thorne 3 ps Alkaline volcanic and calcareous clay soils 

Mimulus ”ovatus” c 9 pf Sandy to gravelly flats and slopes 

Eriogonum ovalifolium Nutt. var. williamsiae 

Reveal 

Eriogonum diatomaceum Reveal, J. Reynolds & 

Picciani 

1 pf Siliceous geothermal sinter 

11 pf Diatomaceous earth deposits 

Potentilla basaltica Tiehm & Ertter 9 pf Moist alkaline meadows 

a Number of reported populations in Nevada included in study; populations outside of Nevada included are shown in 

parentheses. 

b pf = perennial forb; ge = geophyte; an = annual; ps = perennial succulent. 

C Recent taxonomic revisions have left this western Nevada endemic, formerly included in Mimulus ovatus, without a 

name. 

well-documented in the White Mountains of California 

and Nevada (Marchand 1973; Mooney 1966; Mooney et 

al. 1962; Morefield 1992; Wright and Mooney 1965) 

and on altered andesites in western Nevada (Billings 

1950). In an analysis of plant distributions in the Mojave-Intermountain 

transition zone, Meyer (1978) found 

that endemic plants showed a high degree of habitat specialization 

and that edaphically restricted species were 

much better represented in xeric, than in mesic, community 

types. 

Kruckeberg and Rabinowitz (1986) also provided 

case histories of Astragalus phoenix and Mentzelia leucophylla, 

both narrow edaphic endemics known from 

flats, washes, and knolls of calcareous alkaline soils at 

Ash Meadows National Wildlife Refuge in Nye County, 

Nevada. Ash Meadows lies on the periphery of, and 

shares many ecological features with, the adjacent Great 

Basin; paleosoils on which these two species occur are 

the partially dissected remnants of a large Pleistocene 

playa. Numerous examples also exist of endemism or 

rarity in Great Basin plants associated with soils derived 

from volcanic ash (Grimes 1984), sand dunes (Holm- 

gren 1979; Pavlik 1989a), geothermal features (Holmgren 

1972b; Reveal 1972, 1981), Pliocene and Pleistocene 

lake and playa sediments, including gypsum 

mounds, (Forbis 2007; Holmgren and Holmgren 2002; 

Reveal 1972), diatomaceous earth deposits (Reveal et al. 

2002), and pumice deposits (Reveal 2004a). A detailed 

discussion of examples of edaphic endemism among the 

rarest plants of the valley, montane, and high elevation 

habitats of the study area is presented in subsequent sections. 

In most cases, the specialized habitats of these taxa 

are more properly characterized as a substrate rather 

than a well-developed soil. Sand deposits, shallow 

gravel sinters, pumice fields, and volcanic ash exposures 

in the valleys and the scree and talus slopes of the 

mountains are typically dynamic, unstable substrates 

shaped by erosional processes. Mineral material dominates 

the soil profile, little organic material is present, 

and soil horizons are poorly differentiated, if they are 

even present. In a few cases, such as paleosoils developed 

on ancient lake beds, moist alkaline clays, or playa 

edges, the soils are better developed, although in the 

96

case of the former lake beds a duripan is common within 

a short distance of the surface and the surface itself 

may be armored with desert pavement. The marked 

aridity of the Great Basin also slows the rate of soil development. 

While the degree of soil development has a 

substantial influence on the floristic composition and 

structure of more common, widespread plant communities, 

it appears to be less important in specialized edaphic 

endemics. In these habitats, physical soil factors 

(or perhaps, in some cases, soil chemistry) may have 

greater influence on the ability of species to establish 

and persist. 

Valley Endemics 

The 14 rarest plants below 2000 m all occur on valley 

floors within a matrix of zonal vegetation dominated 

by Artemisia tridentata ssp. wyomingensis or various 

salt desert shrubs (Figure 2, Table 1). Most occur on 

azonal soils developed from surficial deposits that overlie 

the underlying bedrock, in some cases by thousands 

of meters (Table 1). Eriogonum diatomaceum, for example, 

is restricted to diatomaceous earth deposits 

(Reveal et al. 2002). Mentzelia argillicola, M. tiehmii, 

and Frasera gypsicola are typically found on sediments 

comprised of calcareous silts, clays, and air-deposited 

ash beds that accumulated in Middle Pliocene-Early 

Pleistocene lakes (Tschanz and Pampeyan 1970), although 

the latter two species also occur on Pliocene 

spring mounds with high gypsum content (Forbis 2007). 

Frasera gypsicola is also rarely found in saline bottomlands 

(Smith 1994). Johanneshowellia crateriorum is 

known only from sandy pumice flats and slopes (Reveal 

2004a) associated with the Quaternary Lunar Crater volcanic 

field (Kleinhampl and Ziony 1985). The habitats 

of Sclerocactus blainei and Mimulus ovatus have been 

described as igneous or calcareous gravels with a clay 

matrix (Heil and Porter 1994; Welsh and Thorne 1985) 

and sandy to gravelly slopes derived from siliceous 

sinter or hydrothermally-altered andesite (Morefield 

2001), respectively. Eriogonum argophyllum, E. ovalifolium 

var. williamsiae, and Castilleja salsuginosa are 

all associated with geothermal features, either growing 

in siliceous sinter gravels (Erigonum spp.), or in moist 

alkaline clays or weathered travertine (Castilleja) 

(Holmgren 1972b; Reveal 1972, 1981). Potentilla basaltica 

is also restricted to alkaline wet meadows 

(Tiehm and Ertter 1984). 

The remaining low elevation species do not fit the 

same pattern of adaptation to azonal soils in the valleys. 

Boechera falcifructa appears to be the only lower elevation 

species that occurs on zonal soils; the association of 

known populations with cryptogammic soils crusts 

within the Artemisia tridentata ssp. wyomingensis zone 

suggests that it may once have been more widely distributed 

but subsequently reduced by livestock tramp- 


97 

ling (Morefield 1997). The only plant among the lowest 

elevation group to grow only on soils directly weathered 

from bedrock is Eriogonum tiehmii, which occurs on 

rocky clay soils derived from interbedded sedimentary 

rocks, including claystones, shales, tuffaceous sandstones 

and limestones (Morefield 1995; Reveal 1985). 

Penstemon floribundus is unique among the lower elevation 

taxa for the breadth of its altitudinal distribution 

(Figure 2); it is reported to occur on a wide variety of 

substrates derived from volcanic rocks (Danley 1985; 

Knight 1988). Extensive inventories for P. floribundus, 

endemic to the remote Jackson Range in northwestern 

Nevada, have not been conducted and the species may 

be more common*. 

Montane Endemics 

The nine plant species in this group occur within the 

narrow montane zone dominated by various species of 

Artemisia or the extensive woodlands of Pinus monophllya 

and Juniperus osteosperma typical of mountain 

ranges in the central Great Basin (Figure 2, Table 1). In 

higher mountains, these forests may be comprised of 

other conifers, including Abies concolor, P. flexilis, and 

less commonly, P. longaeva (Charlet 1996). In contrast 

to the valley endemics, eight of the nine montane taxa 

occur either on poorly developed soils directly weathered 

from underlying bedrock, or in scree, talus, or bedrock 

ledges, cliffs, and crevices (Table 1). The exception 

is Eriogonum douglasii var. elkoense reported from 

sandy to gravelly flats and slopes with Artemisia nova 

and mixed grasses (Reveal 2004b). The primary substrate 

affinities of the other eight taxa include tuffaceous 

volcanic sediments (Trifolium andinum var. podocephalum 

(Barneby 1989) and Eriogonum microthecum var. 

arceuthinum (Reveal 2004b)), volcanics (Collomia renacta 

(Joyal 1986), Penstemon pudicus (Reveal and 

Beatley 1971), and P. tiehmii (Holmgren 1998)), granite 

(Tonestus graniticus (Tiehm and Shultz 1985)), and carbonates 

(P. moriahensis (Holmgren 1979), Lewisia 

maguirei (Holmgren 1954; Williams 1981), and Viola 

lithion (Holmgren 1992)). In general, these substrates 

are common regionally and locally and the presumed 

rarity of these taxa is most likely determined by other 

ecological or historical factors. 

High Elevation Endemics 

Five of the nine high elevation endemics show a 

preference toward a particular geologic substrate (Table 

1). Those reported to occur only on quartzite or other 

siliceous substrates include Draba serpentina (Al- 

*Surveys conducted subsequent to the preparation of this 

manuscript have confirmed P. floribundus to be more common 

and to occur on the highest peaks of the Jackson Range 

(A. Tiehm, personal communication).


Shehbaz and Windham 2007; Tiehm and Holmgren 

1991), Boechera ophira (Morefield 2003) and Potentilla 

cottamii (Holmgren 1987). Penstemon rhizomatosus is 

reported only from carbonate substrates (Holmgren 

1998) while Cymopterus goodrichii is known from slate 

and limestone sedimentary rocks (Welsh and Neese 

1980). Ipomopsis congesta var. nevadensis may also 

occur on carbonate substrates but no systematic surveys 

have been completed (Morefield 2001). Eriogonum 

holmgrenii is reported from quartzitic, carbonate, and 

granitic substrates (Goodrich 1979; Reveal 1965), while 

Polemonium chartaceum occurs on rhyolite in the 

Sweetwater Range of California (Hunter and Johnson 

1983), on open slopes of metavolcanics and nonbasic 

rocks at the summit of White Mountain Peak in the 

White Mountains of California (Crowder and Sheridan 

1972; Morefield 1992; Rundel et al. 2008; Van de Ven 

et al. 2007), and on granitic rocks in the Boundary Peak 

area of the White Mountains in California and Nevada 

(Kartesz 1987; Pritchett and Paterson 1998). Similar to 

their montane counterparts, these taxa are typically 

found on poorly developed soils derived directly from 

weathered bedrock or in association with scree, talus, or 

bedrock ledges, cliffs, and crevices. These substrates 

and habitats are common regionally and locally and the 

presumed rarity of these taxa is most likely determined 

by other ecological or historical factors. The sole exception 

to this among the high elevation species is Primula 

capillaris, which occurs on moist meadow soils derived 

from glacial till (Holland 1995; Holmgren and Holmgren 

1974). 

Vulnerability to Climate Change 

The vulnerability of any plant species or population 

to climate change is influenced by many factors in addition 

to climate and substrate, including physiological 

tolerances, life-history strategies, phenological plasticity, 

relative probabilities of extinctions and colonizations, 

dispersal abilities, and disruptions in plantpollinator 

or herbivorous insect-host plant interactions 

(Parmesan 2006). Unfortunately, few data exist on most 

of these factors for any plants, including the 33 plants 

examined herein. This analysis, therefore, provides preliminary 

predictions that will be tested as climate 

change progresses over the coming decades. 

Previous studies have concluded that with progressive 

climate change we can expect ongoing upward 

shifts in both forest and alpine plant species and subsequent 

declines in arctic-alpine plants at the southern 

margin of their range (Lenoir et al. 2008; Lesica and 

McCune 2004; Walther et al. 2005). The results reported 

here suggest that losses of endemic plants in the 

Great Basin may be highest within the lower elevation 

sagebrush and salt desert zones. This is consistent with 

similar results reported from Utah where the highest 

levels of plant endemism were found between 1000 m 

and 2000 m (Ramsey and Shultz 2004). Meyer (1978) 

also found that the percentage of edaphically restricted 

species in the Mojave-Intermountain transition zone 

dropped sharply with an increase in altitude. This may 

be because xeric environments tend to be more heterogeneous 

than mesic habitats in response to variables 

other than moisture and heterogeneous environments 

tend to restrict the migration of specialized plant species, 

thus reinforcing endemism. 

Previously published studies of climate change in the 

Great Basin have largely focused on the montane biogeography 

of birds and mammals (e.g., Beever 2003; 

Brown 1971, 1978; Lawlor 1998), phytogeography 

(Billings 1978; Harper et al. 1978) or dominant tree and 

shrub distributions (Wells 1983; West et al. 1978). Most 

often these studies have applied colonization-extinction 

models (sensu MacArthur and Wilson 1967) to mountain 

ranges as “sky islands” within an homogenous 

“sagebrush sea” rather than actual landscape patterns of 

local plant endemism. While such imagery has popular 

appeal, it can mask the reality that low elevation ”sea” 

contains many distinctive ecological islands supporting 

edaphic plant endemics, contributing significantly to the 

overall biodiversity of the Great Basin. 

This is not simply a matter of scale because ecological 

islands in the valleys are more-or-less equivalent 

counterparts in both area and isolation to subalpine and 

alpine habitats on ridges and peaks. The principal difference 

with respect to vulnerability to climate change is 

that montane and higher elevation habitats possess a 

nearly continuous array of microclimates produced by 

variations in slope and aspect across a wide range of 

elevation. In contrast, valley habitats have much less 

variation in either slope or aspect within a much narrower 

elevation range. Thus, lower elevation plant endemics 

are likely to have fewer ecological options as 

their bioclimatic envelope shifts. This greater vulnerability 

to climate change is compounded by the fact that 

valley habitats are also more susceptible to invasion by 

non-native plants and habitat modification or destruction 

by human activities (e.g. transportation or energy 

corridors and off-road recreation). 

Conceptual Model 

Based on the foregoing analysis, I am proposing a 

conceptual model for assessing the potential effects of 

climate change on rare plants in the Great Basin of Nevada; 

this model also may have general application 

throughout areas of the west with similar “basin-andrange” 

topography (Figure 3a). The model begins with a 

generic Great Basin landscape of playas, sand dunes, 

and terraces comprised of older lake sediments on valley 

fill. At the base of the bounding range, the valley fill 

is overlain by alluvial fan deposits. The bounding range 

98


in this model is comprised of bedded carbonates overlain 

by more erosion-resistant quartzite, a typical landscape 

throughout much of eastern Nevada. In other parts 

of the Great Basin, the ranges may be comprised of volcanic 

rocks or, locally, intrusive granitic rocks. 

The valley currently provides suitable habitat for 

endemic plants that are typically restricted to specialized 

edaphic conditions, such as playa edges, the periphery 

of sand dunes, or specific microhabitats within older 

lake sediments. Montane endemics are restricted to carbonate 

or siliceous rock types or, in some cases, are 

more restricted to habitats characterized by coarse materials 

weathered from various substrates, such as gravels, 

angular slates, talus, or scree. At the highest elevations 

are the mountaintop endemics in subalpine-alpine habitats 

or, in the lower mountain ranges, on ridge tops 

within lower vegetation zones (Figure 3b). 

As climate changes, populations of those endemic 

plants with the most restrictive and least common habitat 

constraints are likely to shrink in areal extent and 

become more isolated from one another. This is most 

likely to happen in the valleys, where many of the rarest 

endemics occur. But all plants restricted to highly specialized 

habitats, such as mountain tops or carbonate 

substrates, are vulnerable when physiological limits are 

exceeded as their bioclimatic envelope shifts to unsuitable 

habitats. Less specialized endemics may be less 

susceptible to habitat shifts but are still likely to decrease 

in extent since less area is available at higher elevation 

(Figure 3c). The eventual outcome of this scenario 

is likely to be extirpation of populations and eventual 

species extinctions throughout the landscape, with 

the highest rates likely in the valleys where the greatest 

edaphic specializations occur (Figure 3d). As noted 

3a. 

3b. 

3c. 3d. 

Figure 3. Conceptual model of the effects of climate change on endemic plants in the Great Basin. a) Generalized 

landscape typical of the Great Basin in eastern Nevada showing a playa lake, sand dunes, and Pliocene-Pleistocene 

lake sediments on the valley floor, overlain by alluvial, and a fault-block mountain range comprised of a band of 

carbonate sediments overlain by erosion resistant quartzite that forms the ridge; b) the current landscape occupied 

by endemic plant populations restricted to playa edges, sand dunes, carbonate rocks, montane plants that occur on 

both carbonates and quartzite, and higher elevation plants; c) as climate change progresses, populations of plants 

restricted to specialized habitats contract in place, while those less specialized migrate upward or onto more suitable 

aspects as their bioclimatic envelope shifts; d) eventually, populations of plants on highly specialized habitats 

are extirpated and the taxa go extinct while other species continue to migrate upward where less habitat area is 

available. 

99


earlier, however, the physiological limits of most species 

are unknown and may not be exceeded. Or species 

may be able to adapt at a pace commensurate with the 

rate of climate change. Alternatively, climate change 

may facilitate the migration of competitors into the specialized 

habitats of these species. Many uncertainties 

remain not only about the parameters of climate change 

itself, but also about its potential effects on rare plant 

species. 

Conservation Strategies 

We cannot predict with any certainty which of these 

33 plant taxa, or for that matter the roughly 150 additional 

G2 taxa (those with fewer than 20 known occurrences), 

will be most affected by climate change. Nevertheless, 

several conservation actions can be identified to 

better ensure the long-term conservation of plant diversity 

in the Great Basin. Julius and others (2008) identified 

adaptation options with an overall goal of maximizing 

resilience to climate change and noted that establishing 

current baselines, identifying thresholds, and monitoring 

for changes will be essential elements of any adaptation 

approach. 

Clearly, inventory and monitoring are critical elements 

of a conservation strategy for these rare plants 

Yet only 23 percent of the 167 reported populations of 

the 33 rarest taxa have even been visited in the past decade, 

and population estimates are available for fewer 

than half of these populations (USFWS, unpublished 

data). Quantitative monitoring programs are in place, or 

in progress, for only five of the taxa. Without a concerted 

effort to establish baselines and a commitment to 

long-term monitoring, many populations could vanish 

without our knowledge, potentially compromising the 

long-term viability of species and certainly resulting in a 

loss of genetic resources. 

To inventory and monitoring, we can add ex situ approaches 

as an, admittedly less than ideal, but necessary 

element at least for genetic conservation. Ex situ conservation 

has long been regarded as an option of last resort 

among plant conservationists, in part, because of concern 

that it might be viewed as an acceptable alternative 

to the conservation of wild habitats. In the face of climate 

change, however, many botanists now recognize 

that ex situ conservation has a place among a portfolio 

of scientifically based techniques that support the primary 

objective of retaining plant diversity in the wild. 

Such techniques are requisite for restoration and relocation 

actions, especially when integrated with regional 

conservation for both ecosystems and suites of species 

(Guerrant and Pavlik 1997; Maunder et al. 2004; Pavlik 

1996). 

Fortunately, both the infrastructure and successful 

models for comprehensive and integrated approaches 

for ex situ conservation of plant genetic resources exist. 

The Center for Plant Conservation (CPC; www.center 

for plantconservation.org), is dedicated solely to preventing 

the extinction of America’s imperiled native 

flora. Hosted at the Missouri Botanical Garden in St. 

Louis, Missouri, CPC coordinates a network of 33 participating 

institutions throughout the country which 

maintain plant material (seeds, cuttings, etc.) of the most 

imperiled plant species in their region as part of the National 

Collection of Endangered Plants, totaling some 

700 species. 

Representation of the rarest plants of the Great Basin 

in the National Collection, however, is poor. Seeds of 

only two of the 33 taxa in this study, Eriogonum argophyllum 

and Eriogonum ovalifolium var. williamsiae, 

are in long-term conservation storage at participating 

institutions at the Red Butte Garden in Salt Lake City, 

Utah, and the Berry Botanical Garden in Portland, Oregon, 

respectively. Among the 11 participating institutions 

in the western United States, the Berry Botanical 

Garden has the most comprehensive collection with 53 

taxa conserved, although only a few of these are from 

the Great Basin, as delineated here, and it is uncertain 

how representative these samples are of the full range of 

genetic diversity among these few taxa. The Red Butte 

Garden, the designated primary seed repository for the 

Great Basin has 22 taxa conserved, most of which are 

plants endemic to Utah. 

Seed collection and long-term storage is also coordinated 

by the Bureau of Land Management’s Seeds of 

Success project established in 2001 in partnership with 

the Royal Botanic Gardens, Kew (http://www.nps.gov/ 

plants/sos/) to collect, conserve, and develop native 

plant materials for stabilizing, rehabilitating and restoring 

lands in the United States. This partnership has now 

grown to include many partners who have collected 

over 6,689 seed accessions. While the focus of Seeds of 

Success is on common species, it nevertheless provides 

a useful model for a comprehensive, landscape-based 

program of targeted seed collection. 

CONCLUSIONS 

While climate change poses ecosystemic challenges 

to many species, narrowly distributed and highly specialized 

taxa are particularly at risk. Among the rarest 

plants in Great Basin of Nevada, the majority of those 

likely to be at greatest risk are restricted to azonal edaphic 

habitats in the valleys. A comprehensive and integrated 

program of adaptation options is essential to 

maximize the resilience of their ecosystems to change. 

In particular, inventory, monitoring, and ex situ conservation 

are needed to ensure that baseline data are available 

against which demographic changes in these taxa 

can be evaluated and to ensure that genetic resources 

representative of the diversity within these taxa are conserved. 

Conservation of the full range of genetic diver- 

100


sity will enhance the capability of future botanists to 

conserve these species in the wild, whether through 

population enhancement or translocation to new suitable 

sites. 


This analysis would not have been possible without 

the extended efforts over the past half-century to document 

the flora of the Great Basin. The contributors to 

this effort are too numerous to list but their names reappear 

through the citations herein and in many of the 

specific epithets. The Nevada Natural Heritage Program 

plays an indispensable role in maintaining data on rare 

plant species. Thanks especially to my reviewers, Dr. 

Alan de Queiroz, Dr. Bruce Pavlik, and Dr. James 

Morefield; their comments helped clarify my thinking 

and made a substantial improvement in the manuscript. 

Finally, thanks to the Nevada Fish and Wildlife Office 

of the U.S. Fish and Wildlife Office for allowing me 

time to complete this work. The findings and conclusions 

in this article are those of the author and do not 

necessarily represent the views of the U.S. Fish and 

Wildlife Service. 

REFERENCES 

Aitken, M., D.W. Roberts, and L.M. Shultz. 2007. 

Modeling distributions of rare plants in the Great Basin, 

western North America. Western North American Naturalist. 

67(1):26–38. 

Al-Shehbaz, I.A. and M.D. Windham. 2007. New or 

noteworthy North American Draba (Brassicaceae). Harvard 

Papers in Botany. 12(2):409–419. 

Anderson, S., M. White, and D. Atwood. 1991. 

Humboldt National Forest sensitive plant field guide. 

Ogden, UT: Department of Agriculture, U.S. Forest Service: 

unpaginated. 

Barneby, R.C. 1989. Trifolium andinum var. podocephalum. 

In: Cronquist, A.; Holmgren, A.H.; Holmgren, 

N.H.; Reveal, J.L.; Holmgren, P.K. Intermountain 

Flora: vascular plants of the Intermountain West, U.S.A. 

Volume Three, Part II (Fabales). Bronx, NY: The New 

York Botanical Garden: 217–218. 

Beatley, J. 1969. Biomass of desert winter annual 

plant populations in southern Nevada. Oikos. 20(2): 

261–273. 

Beatley, J. 1975. Climates and vegetation pattern 

across the Mojave/Great Basin Desert transition of 

southern Nevada. American Midland Naturalist. 93: 

53–70. 

Beever, E., P.F. Brussard, and J. Berger. 2003. Patterns 

of apparent extirpation among isolated populations 

of pikas (Ochotona princeps) in the Great Basin. Journal 

of Mammalogy. 84:37–54. 

Billings, W.D. 1949. The shadscale vegetation zone 

of Nevada and eastern California in relation to climate 

and soils. American Midland Naturalist. 42(1):87–109. 

Billings, W.D. 1950. Vegetation and plant growth as 

affected by chemically altered rocks in the western 

Great Basin. Ecology. 31:62–74. 

Billings, W.D. 1978. Alpine phytogeography across 

the Great Basin. In: Harper, K.T.; Reveal, J.L. Intermountain 

biogeography: a symposium. Great Basin 

Naturalist Memoirs 2, Provo, UT: Brigham Young University 

Printing Service: 105–118. 

Bonfils, C., B.D. Santer, D,W, Pierce, H.G. Hidaloo, 

B. Govindasamy, T. Das, T.P. Barnett, D.R. Cayan, C. 

Doutriaux, A.W. Wood, A. Mirin, and T. Nozawa. 

2008. Detection and attribution of temperature changes 

in the mountainous Western United States. Journal of 

Climate. 21:6404–6424. 

Brown, J.H. 1971. Mammals on mountaintops: nonequilibrium 

insular biogeography. American Naturalist. 

105:467–478. 

Brown, J.H. 1978. The theory of insular biogeography 

and the distribution of boreal birds and mammals. 

In: Harper, K.T.; Reveal, J.L. Intermountain biogeography: 

a symposium. Great Basin Naturalist Memoirs 2, 

Provo, UT: Brigham Young University Printing Service: 

209–227. 

Cayan, D.R., S.A. Kammerdiener, M.D. Dettinger, J. 

M. Caprio, and D.H. Peterson. 2001. Changes in the 

onset of spring in the western United States. Bulletin of 

the American Meteorological Society. 82:399–415. 

Charlet, D. 1996. Atlas of Nevada conifers: a phytogeographic 

reference. Reno, NV. University of Nevada 

Press: 320 p. 

Committee on Environment and Natural Resources. 

2008. Scientific assessment of global change on the 

United States. Washington, D.C., U.S. Climate Change 

Science Program, National Science and Technology 

Council: 261 p. 

Comstock, J.P. and J.R. Ehleringer. 1992. Plant adaptation 

in the Great Basin and Colorado Plateau. Great 

Basin Naturalist. 52(3):195–215. 

Crowder, D.F. and M.F. Sheridan. 1972. Geologic 

map of the White Mountain Peak Quadrangle, Mono 

County, California. U.S. Geologic Survey Report GQ– 

1012. 

Cubashi, U, G.A. Meehl, and G.J. Boer. 2001. Projections 

of future climate change. In: Houghten, J.T., 

Ding, Y.; Griggs, D.J.; Noguer, M.; van der Linden, 

P.J.; Da, X.; Maskell, K.; Johnson, C.A., eds. Climate 

change 2001: the scientific basis. Contributions of 

Working Group I to the Third Assessment Report of the 

Intergovernmental Panel on Climate Change. Cambridge, 

Cambridge University Press: [Online] Available: 

http://www.grida.no/publications/other/ipcc%5Ftar/? 

src=/CLIMATE/IPCC_TAR/wg1/109.htm. [March 13, 

2009]. 

101


Danley, D.M. 1985. A new Penstemon (Scrophulariaceae) 

from northwestern Nevada. Brittonia. 37: 

321–324. 

Fautin, R.W. 1946. Biotic communities of the northern 

desert shrub biome in western Utah. Ecological 

Monographs. 16(4):251–310. 

Fischlin, A., G.F. Midgley, J.T. Price, R. Leemans, 

B. Bopal, C. Turley, M.D.A. Rounsevell, O.P. Dube, J. 

Tarazona, and A.A. Velichko. 2007. Ecosystems, their 

properties, goods, and services. In: Parry, M.L.; Canziani, 

O.F., Palutikof, J.P.; van der Linden, P.J.; and 

Hanson, C.E., eds. Climate change 2007: Impacts, adaptation 

and vulnerability. Contributions of Working 

Group II to the Fourth Assessment Report of the Intergovernmental 

Panel of Climate Change. Cambridge, 

Cambridge University Press: [Online] Available: 

http://www.ipcc.ch/ipccreports/ar4-wg2.htm. [May 26, 

2009] 

Forbis, T. 2007. White River rare plant conservation 

strategy. Unpublished report prepared by The Nature 

Conservancy, Reno, Nevada. On file at Nevada Fish and 

Wildlife Office, U.S. Fish and Wildlife Service, Reno, 

NV. 20 p. 

Goodrich, S. 1979. Status report on Eriogonum 

holmgrenii. Unpublished report prepared for the U.S. 

Fish and Wildlife Service, Portland, OR. On file at Nevada 

Fish and Wildlife Office, U.S. Fish and Wildlife 

Service, Reno, NV. 5 p. 

Grabherr, G., M. Gottfried, A. Gruber, and H. Pauli. 

1995. Patterns and current changes in alpine plant diversity. 

In: Chapin, F.S.; Körner, C., eds. Arctic and Alpine 

Biodiversity. Berlin, Heidelberg: Springer Verlag. p. 

167–181. 

Grayson, D.K. 2000. Mammalian responses to Middle 

Holocene climatic change in the Great Basin of the 

western United States. Journal of Biogeography. 

27:181–192. 

Grimes, J.W. 1984. Notes on the flora of Leslie 

Gulch, Malheur County, Oregon. Madroño 31:69–85. 

Guerrant, E.O., Jr. and B.M. Pavlik. 1997. Reintroduction 

of rare plants: genetics, demography, and the 

role of ex situ methods. In: Fiedler, P.L.; Jain, S.K., 

eds. Conservation Biology: the theory and practice of 

nature conservation, preservation, and management. 

Second Edition. London: Chapman and Hall:80–108. 

Halloy, S.R.P. and A.F. Mark. 2003. Climate-change 

effects on alpine plant biodiversity: a New Zealand perspective 

on quantifying threat. Arctic, Antarctic, and 

Alpine Research. 35(2):248–254. 

Harper, K.T. and J.L. Reveal. 1978. Intermountain 

biogeography: a symposium. Great Basin Naturalist 

Memoirs 2, Provo, UT: Brigham Young University 

Printing Service: 268 pp. 

Heil, K.D. and J.M. Porter. 1994. Sclerocactus 

(Cactaceae): a revision. Haseltonia. 2:20–46. 

Hidy, G.M. and H.E. Klieforth. 1990. Atmospheric 

processes affecting the climate of the Great Basin. In: 

Osmond, C.B.; Pitelka, L.F.; Hidy, G.M., eds. Plant 

biology of the Basin and Range. Ecological Studies 80, 

Springer-Verlag, Berlin. 19–45. 

Holland, R.F. 1995. Current knowledge and conservation 

status of Primula capillaris Holmgren & Holmgren 

(Primulaceae), Ruby Mountains primrose, in Nevada. 

Unpublished report prepared for Nevada Natural 

Heritage Program, Carson City, Nevada, and Nevada 

Fish and Wildlife Office, U.S. Fish and Wildlife Service, 

Reno, Nevada. On file at Nevada Fish and Wildlife 

Office, U.S. Fish and Wildlife Service, Reno, NV. 19 p., 

plus appendices. 

Holmgren, A.H. 1954. A new Lewisia from Nevada. 

Leaflets of Western Botany. 7:135–137. 

Holmgren, N.H. 1972a. Plant geography of the Intermountain 

Region. In: Cronquist, A.; Holmgren, A.H.; 

Holmgren, N.H.; Reveal, J.L. Intermountain Flora: vascular 

plants of the Intermountain West, U.S.A. Volume 

One. Bronx, NY: The New York Botanical Garden: 77– 

161. 

Holmgren, N.H. 1972b. Five new species of Castilleja 

(Scrophulariaceae) from the Intermountain Region. 

Bulletin of the Torrey Botanical Club. 100(2):83– 

93. 

Holmgren, N.H. 1979. New penstemons (Scrophulariaceae) 

from the intermountain region. Brittonia. 31: 

217–242. 

Holmgren, N.H. 1987. Two new species of Potentilla 

(Rosaceae) from the Intermountain Region of western 

U.S.A. Brittonia. 39(3):340–344. 

Holmgren, N.H. 1992. Two new species of Viola 

(Violaceae) from the Intermountain West, U.S.A. Brittonia. 

44(3):300–305. 

Holmgren, N.H. 1998. Two new species of Penstemon 

(Scrophulariaceae: Sect. Saccanthera) from Nevada, 

U.S.A. Brittonia. 50(2):159–164. 

Holmgren, N.H. and A.H. Holmgren. 1974. Three 

new species from the Great Basin. Brittonia. 26(3):309– 

315. 

Holmgren, N.H. and P.K. Holmgren. 2002. New 

Mentzelias (Loasaceae) from the Intermountain Region 

of the western United States. Systematic Botany. 27 

(4):747-762. 

Hunter, K.B. and R.E. Johnson. 1983. Alpine flora of 

the Sweetwater Mountains, Mono County, California. 

Madroño 30(4):89–105. 

Jenny. 1941. Factors of soil formation. New York: 

McGraw-Hill. 109 pp 

Joyal, E. 1986. A new species of Collomia 

(Polemoniaceae) from the Great Basin. Brittonia. 38 

(3):243–248. 

Julius; S.H., J.M. West, G.M. Blate, J.S. Baron, B. 

Griffith, L.A. Joyce, P. Kareiva, B.D. Keller, M.A. 

102


Palmer, C.H. Peterson, and J.M. Scott. 2008. Executive 

Summary. In: Julius, S.H.; West, J.M. eds. Preliminary 

review of adaptation options for climate-sensitive ecosystems 

and resources. A report by the U.S. Climate 

Change Science Program and the Subcommittee on 

Global Change Research. Washington, D.C., U.S. Environmental 

Protection Agency: p. 1–1 to 1–6. 

Kartesz, J.T. 1987. A flora of Nevada. Reno, NV: 

University of Nevada, Reno. 1,729 p. Dissertation. 

Kleinhampl, F.J. and J.L. Ziony. 1985. Geology of 

northern Nye County, Nevada. Bulletin 93a. Reno, NV: 

University of Nevada, Reno, Nevada Bureau of Mines, 

Mackay School of Mines. 172 p. 

Knight, T. 1988. Status report for Penstemon floribundus 

Danley. Unpublished report prepared by the Nevada 

Natural Heritage Program for the Winnemucca 

District, Bureau of Land Management, Winnemucca, 

NV. On file at Nevada Fish and Wildlife Office, U.S. 

Fish and Wildlife Service, Reno, NV. 23 p., plus appendices. 

Kruckeberg, A.R. 1986. An essay: The stimulus of 

unusual geologies for plant speciation. Systematic Botany. 

11:455–463. 


aspects of endemism in higher plants. Annual Review 

of Ecology and Systematics. 16:447–479. 

Lawlor, T.E. 1998. Biogeography of Great Basin 

mammals: paradigm lost? Journal of Mammalogy 

79:1111–1130. 

Lenoir, J., J.C. Gégout, P.A. Marquet, P. de Ruffray, 

and H. Brisse. 2008. A significant upward shift in plants 

species optimum elevation during the 20 th century. Science. 

320:1768–1771. 

Lesica, P. and B. McCune. 2004. Decline of arcticalpine 

plants at the southern margin of their range following 

a decade of climatic warming. Journal of Vegetation 

Science. 15:679–690. 

MacArthur, R.H. and E.O. Wilson. 1967. The theory 

of island biogeography. Princeton, NJ. Princeton University 

Press: 203 p. 

Major, J. 1951. A functional, factorial approach to 

plant ecology. Ecology. 32(3):392–412. 

Marchand, D.E. 1973. Edaphic control of plant distribution 

in the White Mountains, eastern California. 

Ecology, 54:233–250. 

Mason, H.L. 1946a. The edaphic factor in narrow 

endemism. I. The nature of environmental differences. 

Madroño 8:209–226. 

Mason, H.L. 1946b. The edaphic factor in narrow 

endemism. II. The geographic occurrence of plants of 

highly restricted patterns of distribution. Madroño 8: 

241–272. 

Maunder, M., E.O. Guerrant, Jr., K. Havens, and 

K.W. Dixon. 2004. Realizing the full potential of Ex 

Situ contributions to global plant conservation. In: 

Guerrant, E.O., Jr.; Havens, K., Maunder, M., eds. Exsitu 

Plant Conservation. Washington: Island Press: 

389–418. 

Meyer, S. 1978. Some factors governing plant distributions 

in the Mojave-Intermountain transition zone. In: 

Harper, K.T.; Reveal, J.L. 1978. Intermountain biogeography: 

a symposium. Great Basin Naturalist Memoirs 2, 

Provo, UT: Brigham Young University Printing Service: 

197–207. 

Mooney, H.A. 1966. Influence of soil type on the 

distribution of two closely related species of Erigeron. 

Ecology, 47:950–958. 

Mooney, H.A., G. St. Andre, and R.D. Wright. 1962. 

Alpine and subalpine vegetation patterns in the White 

Mountains of California. American Midland Naturalist. 

68:257–273. 

Morefield, J.D. 1992. Spatial and ecologic segregation 

of phytogeographic elements in the White Mountains 

of California and Nevada. Journal of Biogeography. 

19(1):33–50. 

Morefield, J.D. 1995. Current knowledge and conservation 

status of Eriogonum tiehmii Reveal (Polygonaceae), 

Tiehm buckwheat. Unpublished report prepared 

by the Nevada Natural Heritage Program, Carson City, 

Nevada. On file at Nevada Fish and Wildlife Office, 

U.S. Fish and Wildlife Service, Reno, NV. 21 p., plus 

appendices. 


status of Arabis falcifructa Rollins (Brassicaceae), 

the Elko rockcress. Unpublished report prepared 


Nevada, for the Bureau of Land Management, Reno, 

Nevada. On file at Nevada Fish and Wildlife Office, 

U.S. Fish and Wildlife Service, Reno, NV. 28 p., plus 

appendices. 

Morefield, J.D., ed. 2001. Nevada rare plant atlas. 

Unpublished report compiled by the Nevada Natural 

Heritage Program, Carson City, Nevada, for the U.S. 

Fish and Wildlife Service, Portland, OR, and Reno, NV. 

[Online]. Available: http://heritage.nv.gov/atlas/ 

atlas.html). [March 12, 2009]. 


status of Arabis ophira Rollins (Brassicaceae), 

the Ophir rockcress. Unpublished status report prepared 


NV. On file at Nevada Fish and Wildlife Office, U.S. 

Fish and Wildlife Service, Reno, NV. 25 p., plus appendices. 

Mote, P.W., A.F. Hamlet, M.P. Clark., and D.P. Lettermaier. 

2005. Declining mountain snowpack in western 

North America. Bulletin of the American Meteorological 

Society. 86:39–49. 

Mozingo, H.M and M. Williams. 1980. Threatened 

and endangered plants of Nevada: an illustrated manual. 

Portland, OR, U.S. Fish and Wildlife Service, and U.S. 

103


Bureau of Land Management, Reno, NV: 268 p. 

Murphy, D.D. and S.B. Weiss. 1992. Effects of climate 

change on biological diversity in western North 

America: species losses and mechanisms. In: Peters, 

R.L.; Lovejoy, T.E., eds. Global warming and biological 

diversity. Castleton, NY: Hamilton Printing: Chapter 26. 

Nachlinger, J., K. Sochi, P. Comer, G. Kittel, and D. 

Dorfman. 2001. Great Basin: an ecoregion-based conservation 

blueprint. Reno, Nevada: The Nature Conservancy. 

159 p., plus appendices. 

NatureServe. 2009. NatureServe Explorer. [Online] 

Available: http://www.natureserve.org/explorer/ 

ranking.htm. [May 26, 2009] 

Parmesan, C. 2006. Ecological and evolutionary responses 

to recent climate change. Annual Review of 

Ecology and Systematics. 37:637–669. 

Pavlik, B.M. 1989. Phytogeography of sand dunes 

in the Great Basin and Mojave Deserts. Journal of Biogeography. 

16(3):227–238. 

Pavlik, B.M. 1996. A framework for defining and 

measuring success during reintroduction of endangered 

plants. In: Falk, D; Millar, C.; Olwell, P., eds. Restoring 

diversity: strategies for reintroduction of endangered 

plants. Washington: Island Press:127–156. 

Peters, R.L. and J.D.S. Darling. 1985. The greenhouse 

effect and nature reserves. BioScience. 35 

(11):707–717. 

Pritchett, D.W. and R. Patterson. 1998. Morphological 

variation in California alpine Polemonium species. 

Madroño 45(3):200–209. 

Rajakaruna, N. 2004. The edaphic factor in the origin 

of plant species. International Geology Review. 46:471– 

478. 

Ramsey, D. and L. Shultz. 2004. Evaluating the geographic 

distribution of plants in Utah from the Atlas of 

Vascular Plants in Utah. Western North American 

Naturalist. 64(4):421–432. 

Reveal, J.L. 1965. A new alpine Eriogonum from 

Nevada. Leaflets of Western Botany. 10:183–186. 

Reveal, J.L. 1972. New species and combinations in 

Eriogonum. Phytologia. 23(1):168–172. 

Reveal, J.L. 1979. Distribution and phylogeny of 

Eriogonoideae (Polygonaceae). In: Harper, K.T.; Reveal, 

J.L. 1978. Intermountain biogeography: a symposium. 

Great Basin Naturalist Memoirs 2, Provo, UT: 

Brigham Young University Printing Service: 169–190. 

Reveal, J.L. 1981. Notes on endangered buckwheats 

(Eriogonum: Polygonaceae) with three newly described 

from the United States. Brittonia. 33(3):441–448. 

Reveal, J.L. 1985. New Nevada entities and combinations 

in Eriogonum (Polygonaceae). Great Basin 

Naturalist. 45(2):276–280. 

Reveal, J.L. 2004a. Johanneshowellia (Polygonaceae: 

Eriogonoideae), a new genus from the Intermountain 

West. Brittonia. 56(4):299–306. 

Reveal, J.L. 2004b. New entities in Eriogonum 

(Polygonaceae: Eriogonoideae). Phytologia. 86(3):121– 

159. 

Reveal, J.L. and J. Beatley. 1971. A new Penstemon 

(Scrophulariaceae) and Grindelia (Asteraceae) from 

southern Nye County, Nevada. Bulletin of the Torrey 

Botanical Club. 98:332–335. 

Reveal, J.L., J. Reynolds, and J. Picciani. 2002. 

Eriogonum diatomaceum (Polygonaceae: Eriogonoideae), 

a New Species from Western Nevada, U.S.A. 

Novon. 12:87–89. 

Rundel, P.W., A.C. Gibson, and M.R. Sharifi. 2008. 

The alpine flora of the White Mountains, California. 

Madroño 55(3):202–215. 

Shreve, F. 1942. The desert vegetation of North 

America. The Botanical Review. 8(4):195–247. 

Smith, F. 1994. Status report for Frasera gypsicola 

(Barneby 1942) D. Post. Unpublished report prepared 

for Nevada Natural Heritage Program, Carson City, Nevada, 

and Nevada Fish and Wildlife Office, U.S. Fish 

and Wildlife Service, Reno, Nevada. On file at Nevada 

Fish and Wildlife Office, U.S. Fish and Wildlife Service, 

Reno, NV. 12 p., plus appendices. 

Snyder, M.A. and L.C. Sloan. 2005. Transient future 

climate over the western United States using a regional 

climate model. Earth Interactions 9, Paper No. 11:1-21. 

[Online]. Available: http://EarthInteractions.org. [March 

12, 2009]. 

Spahr, R., L. Armstrong, D. Atwood, and M. Rath. 

1991. Atwood. Threatened, Endangered, and Sensitive 

Species of the Intermountain Region. Ogden, UT: Department 

of Agriculture, U.S. Forest Service: unpaginated. 

Tiehm, A. and B. Ertter. 1984. Potentilla basaltica 

(Rosaceae), a new species from Nevada. Brittonia. 36 

(3):228–231. 

Tiehm, A. and P.K. Holmgren. 1991. A new variety 

of Draba oreibata (Brassicaceae) from Nevada, U.S.A. 

Brittonia. 43(1):20–23. 

Tiehm, A. and L. Shultz. 1985. A new Haplopappus 

(Asteraceae: Astereae) from Nevada. Brittonia. 37 

(2):165–168. 

Tingley, J.V. and K.A. Pizarro. 2000. Traveling 

America’s loneliest road: a geologic and natural history 

tour through Nevada along U.S. Highway 50. Nevada 

Bureau of Mines and Geology, Special Publication 

26, Mackay School of Mines, University of Nevada, 

Reno. 

Tschanz, C.M. and E.H. Pampeyan. 1970. Geology 

and mineral resources of Lincoln County, Nevada. Bulletin 

73. Reno, NV: University of Nevada, Reno, Nevada 

Bureau of Mines, Mackay School of Mines. 187 p. 

Van de Ven, C.M., S.B. Weiss, and W.G. Ernst. 

2007. Plant species distributions under present conditions 

and forecasted for warmer climates in an arid 

104


mountain range. Earth Interactions. 11:1-33. [Online]. 

Available: http://EarthInteractions.org. [March 12, 

2009]. 

Wagner, F.H., ed. 2003. Preparing for a changing 

climate – the potential consequences of climate variability 

and change. Rocky Mountain/Great Basin regional 

climate-change assessment. A report of the Rocky 

Mountain/Great Basin Regional Assessment Team for 

the U.S. Global Change Research Program. Logan, UT. 

Utah State University. 240 p. 

Walther, G., S. Beißner, and C.A. Burga. 2005. 

Trends in the upward shift of alpine plants. Journal of 

Vegetation Science. 16:541–548. 

Weixelman, D. and N.D. Atwood. Undated. Toiyabe 

National Forest sensitive plant field guide. Ogden, UT. 

U.S. Department of Agriculture, U.S. Forest Service: 

123 p. 

Wells, P.V. 1983. Paleobiogeography of montane 

islands in the Great Basin since the last glaciopluvial. 

Ecological Monographs. 53(4):341–382. 

Welsh, S.L. and E. Neese. 1980. A new species of 

Cymopterus (Umbelliferae) from the Toiyabe Range, 

Lander County, Nevada. Madroño. 27(2):97–100. 

Welsh, S.L. and K.H. Thorne. 1985. New Sclerocactus 

(Cactaceae) from Nevada. Great Basin Naturalist. 45 

(3):553–555. 

West, N.E., R.J. Tausch, K.H. Rea, and P.T. Tueller. 

1978. In: Harper, K.T.; Reveal, J.L. Intermountain biogeography: 

a symposium. Great Basin Naturalist Memoirs 

2, Provo, UT: Brigham Young University Printing 

Service: 119–136 

Wharton, R.A., P.E. Wigand, M.R. Rose, R.L. 

Reinhardt, D.A. Mouat, H.E. Kleiforth, N.L. Ingraham, 

J.O. Davis, C.A. Fox, and J.T. Ball. 1990. The North 

American Great Basin: a sensitive indicator of climatic 

change. In: Osmond C.B.; Pitelka, L.F.; Hidy, G.M., 

eds. Plant biology of the Basin and Range. Ecological 

Studies 80. New York, NY: Springer-Verlag: 323–359. 

Williams, M.J. 1981. Status report on Lewisia 

maguirei. Unpublished report prepared for the U.S. Fish 

and Wildlife Service, Portland, OR. On file at the Nevada 

Fish and Wildlife Office, U.S. Fish and Wildlife 

Service, Reno, NV. 19 p. 

Wright, R.D. and H.A. Mooney. 1965. Substrateoriented 

distribution of Bristlecone pine in the White 

Mountains of California. American Midland Naturalist. 

73(2):257–284. 

105


Sentry Milkvetch (Astragalus cremnophylax 

var. cremnophylax) Update 

Janice Busco 

Grand Canyon National Park, Grand Canyon, AZ 

Abstract. The Grand Canyon endemic Astragalus cremnophylax var. cremnophylax (Sentry milkvetch) was listed as 

an endangered species in 1990. There are 725 plants within three populations on the South Rim, all found in shallow 

soils upon large flat Kaibab limestone platforms. Habitat specificity and reduced number of plants make Sentry milkvetch 

vulnerable to extinction. Recovery plan actions completed in 2008 include seed collection and parking lot removal 

to allow habitat restoration and population expansion. Planned recovery plan actions include establishment of 

an ex situ population, seed production, and development of techniques for population augmentation and creation of 

artificial populations. 

The Grand Canyon National Park endemic Astragalus 

cremnophylax Barneby var. cremnophylax (Sentry 

milkvetch) was listed as an endangered species by the 

US Fish and Wildlife Service and protected from trampling 

by South Rim sightseers in 1990. There are approximately 

725 individuals known in three locations, 

all on the South Rim of Grand Canyon. Sentry milkvetch 

occurs in shallow, well-drained soils or porous 

limestone pavement in crevices and depressions in large 

flat Kaibab limestone platforms in unshaded openings in 

the pinyon-juniper woodland along the canyon edge. 

The underlying bedrock limestone stores water and is 

critical to its growth and development (USFWS 2006). 

Sentry milkvetch is a small, mat-forming perennial 

plant (Figure 1) and has a thick taproot and woody caudex. 

Pale purple flowers appear from late April to early 

May, with seed set in late May to June. Its tiny seeds 

tend to fall in the mat of the plant; therefore the plant 

does not spread and remains isolated. 

Threats to Sentry milkvetch include small population 

size, vulnerability to drought and stochastic events, digging 

by ground squirrels and bighorn sheep (Figure 2), 

low reproductive capacity, limited seed dispersal, limited 

habitat, and reduced genetic diversity and vigor 

(Allphin et al. 2005). 

Figure 1. A mature Sentry milkvetch with a quarter for 

scale. 

RECOVERY CRITERIA AND OBJECTIVES 

In order to downlist the species, the Sentry Milkvetch 

Recovery Plan (USFWS 2006) requires achievement, 

maintenance and long-term protection of at least 

four viable Sentry milkvetch populations of at least 

1,000 individuals each, for a total of at least 4,000 individuals 

in the wild. Recovery will be attained when 

there are eight viable Sentry milkvetch populations of 

1,000 individuals each, with long-term protection. 

Figure 2. Bighorn sheep damage to Sentry milkvetch. 

106


RECOVERY PLAN ACTIONS 

Completed Actions 2006-2008 

In 2006, recovery plan actions for the species began. 

Grand Canyon NP has completed annual monitoring of 

the Maricopa Point population each year (2006-2008) 

and performed a complete census of the Maricopa Point 

population in 2008. We installed permanent photopoints 

in all three populations in summer 2008. Grand 

Canyon NP and the Arboretum at Flagstaff collected 

2600 seeds from 74 individuals in the summer of 2008 

(Figure 3). The Arboretum at Flagstaff completed seed 

germination studies, initiated greenhouse seed production. 

established an ex situ population, and conducted 

mycorrhizal studies of the species. Grand Canyon NP 

completed parking lot removal, trail rerouting and shuttle 

bus stop relocation from Sentry milk-vetch habitat 

adjacent to the Maricopa Point population to allow habitat 

restoration and population augmentation and expansion 

(Figure 4). In addition, two suitable areas for artificial 

population establishment were selected in Spring 

2008. 

Planned Actions 2009-2013 

Recovery plan actions planned for 2009-2013 include 

restoration of disturbed habitat and completion of 

seeding and planting trials at Maricopa Point. Acquisition 

of a passive solar greenhouse in 2009 will provide a 

dedicated facility for seeding trials, seed and plant production 

for introduction trials at Maricopa Point, and 

will create an ex situ Sentry milkvetch population. 

Through these trials we plan to develop techniques for 

creation of artificial populations, increase the number of 

individuals at Maricopa Point, and establish new pilot 

populations in suitable areas near Maricopa Point 

(Figure 5). A discovery survey of the westernmost portions 

of the south rim will be completed in 2009. Soil 

seed bank studies and ecological observations to determine 

pollinators will be conducted in 2009-2010. Additionally, 

interpretative materials will be developed and 

displayed at the Grand Canyon Visitor Center. Cooperating 

agencies in planned recovery actions include 

Grand Canyon National Park, US Fish and Wildlife Service, 

The Arboretum at Flagstaff, Grand Canyon Association, 

National Park Service, Center for Plant Conservation, 

Northern Arizona University Environmental 

Monitoring and Assessment (EMA) and Coconino National 

Forest. 


Allphin, L., N. Brian, and T. Matheson. 2005. Reproductive 

success and genetic divergence among varieties 

of the rare and endangered Astragalus cremnophylax 

(Fabaceae) from Arizona, USA. Conservation Genetics 

6: 803-821. 

Figure 3. Seed collection, summer 2008. 

Figure 4. Parking lot removal adjacent to population, 

September 2008, readied the site for restoration to be 

completed in 2010-2012. 

Figure 5. Suitable area selected for Sentry milkvetch 

artificial population. 

107


Figure 7. A Mason bee (Osmia ribifloris ribifloris) pollinating 

Sentry milkvetch. 

Figure 6. A Painted lady butterfly visits Sentry milkvetch. 

USFWS 2006. Sentry Milk-vetch (Astragalus 

cremnophylax Barneby var. cremnophylax Barneby) 

Recovery Plan. U.S. Fish and Wildlife Service, Albuquerque, 

New Mexico, i-vii + 44pp. 

ADDENDUM 

This paper reported on the earliest efforts of Grand 

Canyon National Park to begin implementation of the 

2006 USFWS recovery plan for Sentry milkvetch. 

Since my presentation at the University of Utah in 

2009, four years of intensive work to recover this species 

has been completed. Some of this work is documented 

in Falk and others (2011) and Busco and others 

(2011). 

Preliminary pollination studies in spring 2010 established 

the identity of three pollinators - two mason 

bee (Osmia ribifloris and O. ribifloris ribifloris, Figure 

7) and a hoverfly (Syrphidae) (Busco et al. 2011). 

We have confirmed the presence of these generalist 

pollinators throughout Sentry milkvetch populations 

in repeat studies in 2011 and 2012. 

At the time of my presentation in 2009, only 725 

individuals of Sentry milkvetch were known. Today 

there are an estimated 3552 known naturally-occurring 

individuals in wild populations, and 425 plants in 

reintroduction areas. The increase in numbers within 

wild populations is largely the result of the discovery 

of new groups of Sentry milkvetch plants on limestone 

fingers above the rim and on lower limestone 

levels below the rim during revisits to these populations 

in 2010-2012 (Figure 8), as well as the result of 

continued protection of the Maricopa Point population. 

While two of the three populations are apparently 

stable or increasing in number, the third small 

population is increasingly threatened. The area below 

the rim of this population has crumbled away and 

fallen into the canyon; the few remaining individuals on 

a solitary boulder above may likely follow. 

Plant reintroductions at Maricopa Point began in July 

2010 and continue to this date (Figure 9). The first 

small planting trial was completed in July 2010 – 5 Sentry 

milkvetch plants that were planted from greenhousegrown 

plants at that time are all alive today. Seeds were 

less successful in that reintroduction - 10 groups of three 

seeds each were sown in April 2011 and today one seedling 

from this cohort is alive and has reached reproductive 

maturity. Of eighty greenhouse-grown plants and 

Figure 8. Newly discovered Sentry milkvetch population 

in Grand Canyon National Park. 

108


Figure 9. First large-scale reintroduction site for Sentry 

milkvetch. 

240 seeds sown in the restored parking lot area adjacent 

to the Maricopa Point population in July 2011, 51% survive. 

Fourteen have set seed on site and produced 58 

seedlings. In addition, another 44 milkvetch seedlings 

have become established from seed that either washed 

in from the adjacent Sentry milkvetch population or 

were present in the soil seed bank and germinated. Of 

seeds planted in 2011, 15.8% produced seedling plants 

that are now alive. In total, there are now 181 sentry 

milk-vetch plants growing in this reintroduction site 

where habitat beneath a parking lot removed in 2008 

was uncovered and restored. 

We seeded a second reintroduction area at Maricopa 

Point in 2012. Two additional seeding trials were carried 

out at the Maricopa Point reintroduction areas in 

July 2012 with 518 seeds. We also tested seeding techniques 

and watering regimens for this species that will 

provide information for continued large-scare reintroduction 

efforts. 

Finally, we have received a three-year National Park 

Service grant to complete installation and establishment 

of two new reintroduction plantings. This funding will 

continue the momentum of successful reintroduction 

efforts. If we can successfully establish these two new 

populations and maintain the two large naturally-occurring 

sentry milk-vetch populations, Grand Canyon National 

Park will be well on its way to downlisting the 

species within the next ten years. 

Additional References 

Busco, J., E. Douglas, and J. Kapp. 2011. Preliminary 

pollination study on Sentry milk-vetch (Astragalus 

cremnophylax Barneby var. cremnophylax), Grand Canyon 

National Park’s only Endangered plant species. 

The Plant Press 35(1):11-12. 

Falk, M., J. Busco, L. Makarick, and A. Mathis. 

2011. The return of the “Watchman of the Gorge.” Endangered 

Species Bulletin, Summer 2011. Pp. 40-41. 

109


A Tale of Two Single Mountain Alpine Endemics: 

Packera franciscana and Erigeron mancus 

James F. Fowler, Carolyn Hull Sieg, Brian M. Casavant, and Addie E. Hite 

USFS Rocky Mountain Research Station, Flagstaff, AZ 

Abstract. Both the San Francisco Peaks ragwort, Packera franciscana and the La Sal daisy, Erigeron mancus are 

endemic to treeline/alpine habitats of the single mountain they inhabit. There is little habitat available for these plant 

species to migrate upward in a warming climate scenario. For P. franciscana, 2008 estimates indicate over 18,000 

ramets in a 4 m band along a recreational trail in the Arizona San Francisco Peaks, a trail-side population centroid of 

3667 m, and that the population is producing and dispersing seed. We also mapped the 2008 distribution of E. 

mancus patches along the La Sal Mountain crestline in Utah. 

Both the San Francisco Peaks ragwort, Packera franciscana 

(Greene) W.A. Weber and A. Löve, and the La 

Sal daisy, Erigeron mancus Rydberg, are endemic to 

treeline and alpine habitats of the single mountain they 

inhabit. Packera franciscana is known only from the 

San Francisco Peaks in Arizona (Greenman 1917, Trock 

2006) (Figure 1) where it has been reported to mostly 

occur between 3525 m and 3605 m elevation (Dexter 

2007) or more generally 3200-3800 m (Trock 2006) 

with a range size of 85 ha (Dexter 2007). Since the elevation 

of the highest peak on the mountain is 3851 m, 

there is little habitat available for the plant to migrate 

upward in a warming climate scenario, and it has been 

widely speculated that the species is vulnerable to extinction 

due to climate change. In 1985 the distribution 

of P. franciscana on the San Francisco Peaks was 

mapped (Dexter 2007), but prior to our study, no published 

data were available on species abundance. Erigeron 

mancus only inhabits the La Sal Mountains in Utah 

(Cronquist 1947) (Figure 1) where it occurs in alpine 

meadows between 3000-3800 m elevation (Nesom 

2006). In sharp contrast to P. franciscana which predominately 

inhabits loose talus slopes (USFWS 1983), 

E. mancus occupies stable substrates, which greatly facilitates 

field measurements. No published information 

about the population biology of these species is available. 

Consequently, P. franciscana was listed as a 

Threatened species under the Endangered Species Act 

by the U.S. Fish and Wildlife Service (1983) and E. 

mancus is on the Forest Service Region Four Sensitive 

Plant List. 

Kruckeberg and Rabinowitz (1985) note that narrow 

endemics can be locally abundant in specific habitats 

but geographically restricted, a description that may fit 

both species. Biologists have noted that P. franciscana 

is fairly abundant locally (Trock 2006, USFWS 1983) 

and our observations concur. We know of no density or 

Figure 1. Map of the two study areas as isolated single 

mountains on the Colorado Plateau. 

population size data to support this observation, yet such 

data are critical for recovery of the species under the 

Endangered Species Act. In a changing climate scenario 

with increased temperatures and changes in 

amount, type, and patterns of precipitation, it becomes 

difficult to predict population trends. Our study will 

110


define baseline population densities along permanent 

transects under the current climate and allow the detection 

of future population trends. Specifically, our objectives 

are to: 1) establish a statistically robust sampling 

protocol for long-term population density trends; 2) determine 

the elevation of patch centroids along these 

transects to allow early detection of altitudinal migration 

driven by climate change; and 3) provide data for ongoing 

formal species assessments, management responses, 

and, in the case of P. franciscana, revision of the 20- 

year old Species Recovery Plan (Phillips and Phillips 

1987). 

METHODS 

In September 2008 (after the monsoon rains), we 

established an elevational transect along a designated 

recreational trail through Packera franciscana habitat to 

estimate the density of P. franciscana ramets, mid-September 

flowering/fruiting phenology, and the population 

centroid elevation as it intersects the trail (Figure 2). 

Sample points were established at 25 m intervals along a 

transect starting at 3550 m elevation and extending 1425 

m along the trail to an elevation of 3798 m. At each 

sample point we counted P. franciscana ramets (upright 

stems) within 12 individual 1 m 2 frames arranged to 

Figure 2. Location of the Packera franciscana trailside transect on the outslope of the volcanic caldera at and above 

treeline. 

111


allow flexibility for trail curvature (Figure 3). Sampling 

frames were omitted when they overlapped previously 

counted frames, covered recent trail maintenance areas, 

or covered vertical drop-off > 5 m. Counts of ramets 

with flower, fruit, or both were also made within each 

frame. Coordinates for latitude, longitude, and elevation 

were made for each sample point with a Trimble® 

Geo XT 2005 Series GPS. Descriptive statistics were 

calculated with SAS/STAT 9.1 (2002-2003). Population 

centroid was calculated as the mean elevation of 

occurrence weighted by the number of ramets / sample 

point. 

In July 2008, we mapped polygons of E. mancus 

patches with the Trimble® Geo XT 2005 Series GPS in 

three areas near and within Mt. Peale Research Natural 

Area, which is located in the Middle Group of peaks on 

La Sal Mountain. These polygons were plotted on a 

topographic map with ArcMap 9.2. 

RESULTS 

The September 2008 density estimate for P. franciscana 

along the recreational trail was 3.19 ramets / m 2 

ramets in the 4 m band along the transect. A population 

centroid was located at 3667 m elevation. We counted a 

total of 1881 ramets of which 91 percent were vegetative, 

eight percent were in fruit, and one percent were 

flowering. Only seven ramets were in both flower and 

fruit. 

Figure 3. Arrangement and sampling sequence of 1 m 2 

frames to measure ramet density. 

Erigeron mancus mapping work in July 2008 revealed 

a relatively continuous series of E. mancus 

patches along the west ridge up to Mt. Laurel in the La 

Sal Middle Group of peaks, from the talus field at 3725 

m down to 3475 m just above treeline, as well as along 

the La Sal Middle Group crestline at 3650 m (Figure 4). 

Our observations indicate that it can be abundant within 

its microhabitat niche on dry, windy ridgelines but less 

abundant to absent on nearby more mesic midslopes 

DISCUSSION 

Phillips and Peterson (1980) reported a P. franciscana 

population density range of 50-370 plants per 100 

m 2 on the San Francisco Peaks but did not clearly define 

plants as ramets or genets (clumps or clones) (Figure 5). 

However, later references to clump size would indicate 

that they were using the latter concept. On a per 100 m 2 

basis, our density measurements are similar at the upper 

end of their density range (319 vs. 370), which is probably 

a reflection of the different “plant” definitions. 

Given the difficulty of defining and counting clumps 

and clones in the field, ramets provide a more accurate 

way to assess population density. Even though ramet 

density may inflate the number of functional plants, it is 

an accurate reflection of photosynthetic and reproductive 

potential. Phillips and Peterson (1980) also reported 

that 13% of the P. franciscana plants were adult 

(sexually reproducing) which again is comparable to the 

9% of ramets we sampled which were flowering and/or 

fruiting. These results and our estimate of >18,000 P. 

franciscana ramets in a very small portion of its range 

would indicate that the species is persisting and reproducing. 

We interpret the successful production of fruit, which 

we observed actively dispersing by upslope winds in 

mid-September, as an indication that P. franciscana can 

sexually reproduce on the San Francisco Peaks. Seed 

viability studies may provide additional support for this 

interpretation. Examination of plant root systems would 

be necessary to determine if ramets originate from seed 

or from existing perennial rhizotamous clones. Rhizomes 

can produce large patches of ramets which may 

be the primary method of reproduction (USFWS 1983) 

but we also found single isolated ramets during our sampling 

which could be the result of seed dispersal or rhizome 

fragments moving downslope in the talus substrate 

P. franciscana inhabits. Plants inhabiting the upper 

portions of talus slopes would seem to be the result 

of seed dispersal since avalanches and downslope creep 

of talus fields would carry existing P. franciscana plants 

downslope. We noted dead P. franciscana plants at the 

base of some avalanche chutes. The population centroid 

of 3667 m we measured is above the 3525-3605 m elevation 

range for most P. franciscana noted by Dexter 

(2007) and near the upper end of the 3350-3750 m main 

112


Figure 4. Erigeron mancus 

patches along the ridge up to 

Laurel Mt. and along the crestline 

of the Middle Group of La 

Sal mountain. 

occurrence range in earlier reports (Phillips and Peterson 

1980, U.S. Fish and Wildlife Service 1983). However, 

our transect is located on a drier west-southwest 

slope which may account for the higher occurrence elevation. 

More mesic slopes may have lower patch centroids; 

a hypothesis we intend to test by establishing a 

northeast aspect, trail-side transect in 2009. 

We plan the second trail-side transect and annual 

measurements of both transects to detect P. franciscana 

population trends. Sampling in subsequent years may 

indicate trends in population density, changes in September 

phenology, or elevational migration within its 

habitat. We also plan to measure the change in E. 

mancus density along an elevational transect through the 

E. mancus patches shown in Figure 4. By measuring 

patch widths along this elevational transect, we can cal- 

culate patch size and, using our density measurements, 

can then estimate population size for this area. Changes 

in population density and the elevation of population 

centroids over time for both species may allow detection 

of climate change effects as well as provide managers 

with accurate data on which to base land and recreation 

use decisions. 


Thanks to Amanda Kuenzi and Suzy Neal for help 

with fieldwork and to the Coconino and Manti-La Sal 

National Forests for funding support. Thanks also to 

Shaula Hedwall and Barb Phillips who reviewed the 

previous version of this manuscript and provided access 

to internal reports. 

113


Figure 5. Clonal habit of Packera franciscana. 


Cronquist, A. 1947. Revision of the North American 

species of Erigeron, north of Mexico. Brittonia 6(2): 

121-302. 

Dexter, L.R. 2007. Mapping impacts related to the 

Senecio franciscanus Greene. Final Report FS Agreement 

Number: 06-CR-11030416-777. 

Greenman, J.M.. 1917. Monograph of the North and 

Central American species of the Genus Senecio, Part II. 

Annals of the Missouri Botanical Garden. 4: 15-36. 


aspects of endemism in higher plants. Annual 

Reviews in Ecology and Systematics. 16: 447-479. 

Nesom, G.L. 2006. Erigeron. In: Flora of North 

America north of Mexico, Volume 20, Magnoliophyta: 

Asteridae, part 7: Asteraceae, part 2. Oxford University 

Press: 256-348. 

Phillips, A.M. III and E.M. Peterson. 1980. Status 

report: Senecio franciscanus. Office of Endangered 

Species, U.S. Fish and Wildlife Service, Albuquerque, 

NM. 13 p. 

Phillips, B.G. and A.M. Phillips III. 1987. San Francisco 

groundsel (Senecio franciscanus) recovery plan. 

U.S. Fish and Wildlife Service, Albuquerque, NM. 

SAS/STAT. 2002-2003. SAS/STAT® software, version 

9.1 of the SAS System for Windows 2000. SAS 

Institute Inc., Cary, NC. 

Trock, D.K. 2006. Packera. In: Flora of North 

America north of Mexico, Volume 20, Magnoliophyta: 

Asteridae, part 7: Asteraceae, part 2. Oxford University 

Press: 570-602. 

United States Fish and Wildlife Service. 1983. Endangered 

and threatened wildlife and plants; final rule to 

determine Senecio franciscanus (San Francisco Peaks 

groundsel) to be a threatened species and determination 

of its critical habitat. Federal Register 48: 52743- 

52747. 

Addendum 

The planned population size and density estimates 

for E. mancus were completed in summer 2009 and 

published in 2010 (Fowler and Smith 2010). We also 

added the second trailside transect for P. franciscana in 

2009 and published the 2008-2009 results in Fowler and 

Sieg (2010). A second P. franciscana manuscript covering 

the 2010-2012 time frame is in preparation. 

Fowler, J.F. and C.H. Sieg. 2010. Density and elevational 

distribution of the San Francisco Peaks ragwort, 

Packera franciscana (Asteraceae), a threatened 

single-mountain endemic. Madroño 57(4):213-219. 

Fowler, J.F. and B. Smith. 2010. Erigeron mancus 

(Asteraceae) density as a baseline to detect future climate 

change in La Sal Mountain habitats. Journal Botanical 

Research Institute Texas 4(2):747-753. 

114


Long-term Population Demographics and Plant Community Interactions 

of Penstemon harringtonii, an Endemic Species 

of Colorado’s Western Slope 

Thomas A. Grant III 

Program in Ecology, Colorado State University, Fort Collins, CO 

Michelle E. DePrenger-Levin, 

Denver Botanic Gardens Research and Conservation Dept, Denver, CO 

and Carol Dawson 

Bureau of Land Management Colorado State Office, Lakewood, CO 

Abstract. Penstemon harringtonii is an endemic species of Colorado’s western slope. Known from only six counties, 

Harrington’s penstemon is threatened primarily by habitat degradation and destruction in rural areas that are experiencing 

relatively rapid development and recreational pressure. Annual demographic monitoring since 1996 has 

not identified statistically significant changes in the overall number of rosettes, although significant inter-annual 

variation occurs at the two study sites. Additional research has focused upon the interactions of P. harringtonii with 

Artemisia tridentata (big sagebrush) and local plant species richness, and competition for soil moisture. Weak negative 

correlations between P. harringtonii and A. tridentata have been documented at both study sites, although the 

two sites have opposite trends in the correlation of P. harringtonii and species richness. Ordination techniques (nonmetric 

multi-dimensional scaling, NMS) are being explored as a means to find patterns that could increase our understanding 

of the rare species’ interactions with the dominant shrub (A. tridentata), local plant species diversity, and 

soil moisture. NMS found a positive correlation between species richness and the higher density P. harringtonii 

quadrats, although no strong relationships were identified between the rare species and soil moisture. Additional sites 

will be sampled in 2009 and 2010 to test hypotheses concerning the potential drivers of P. harringtonii density and 

provide guidance in the development of appropriate management and restoration methods. 

Long-term monitoring of Penstemon harringtonii 

Penland (Scrophulariaceae) was initiated in 1996 to determine 

population trends of this rare species. The penstemon 

is threatened by land development for homes 

and ski areas, oil and natural gas development, overgrazing, 

and off-road vehicle use (CoNPS 1997, Panjabi 

and Anderson 2006). Formerly a Category 2 candidate 

for listing under the Endangered Species Act (ESA), P. 

harringtonii is endemic to western Colorado (USDA 

NRCS 2009) and populations have been documented 

from six counties within the state (Eagle, Garfield, 

Grand, Pitkin, Routt, and Summit)(Panjabi and Anderson 

2006, Spackman et al. 1997). Currently, P. harringtonii 

is listed as a sensitive species by the U.S. Bureau 

of Land Management Colorado State Office and the 

USDA Forest Service Region 2. The species is ranked 

G3/S3 by The Nature Conservancy Natural Heritage 

ranking system (Spackman et al. 1997). Based upon the 

S3 ranking, the species is considered vulnerable to extirpation 

in the state and only 74 occurrences have been 

documented (Panjabi and Anderson 2006). 

The majority of populations occur in habitats dominated 

by Artemisia tridentata (big sagebrush), a hyd- 

raulic lifting species (Caldwell et al. 1998). The interaction 

between A. tridentata and growth of herbaceous 

plants such as P. harringtonii is unclear, although soil 

moisture and precipitation are acknowledged as limiting 

factors for primary production in semiarid sagebrush 

steppe ecosystems (Horton and Hart 1998). 

Initial goals of the project were to document population 

trends at two study sites, although this has been 

expanded into understanding the species’ relationship to 

A. tridentata density, soil moisture, and plant community 

composition with the ultimate goal of improving 

our management of the species and its habitat. We hypothesized 

that areas with lower densities of A. tridentata 

will have higher densities of P. harringtonii and 

greater species richness due to reduced competition for 

water. Although destruction or degradation of P. harringtonii’s 

habitat is the greatest threat, an understanding 

of the species’ interactions with big sagebrush and 

soil moisture may augment management of extant populations 

and restoration of degraded areas. Additionally, 

an understanding of the relationships between sagebrush, 

soil moisture and species richness may assist our 

ability to manage for diverse ecosystems. 

115


METHODS 

Plant Species 

Penstemon harringtonii is a potentially long-lived 

perennial forb in the Scrophulariaceae. A distinctive 

characteristic of the species is the exsertion of the two 

lower stamens from the blue to pink/lavender corollas 

(Figure 1). Flower production and seedling recruitment 

are thought to be episodic and probably related to seasonal 

precipitation and available soil moisture (Panjabi 

and Anderson 2006). The species is found in open sagebrush 

(A. tridentata) and less commonly in pinyonjuniper 

plant communities between 1951 and 2865m 

elevation. All major threats are based on increased human 

use and development in the region for housing, ski 

areas, resource extraction, grazing, and recreation 

(CoNPS 1997, Spackman et al 1997). 

Study Sites 

Two geographically and ecologically diverse populations 

of P. harringtonii have been monitored since 

1996. The Eagle study site is a sagebrush-steppe community 

near the town of Eagle, CO and is at an elevation 

of 2100m (Buckner and Bunin 1992). The site was 

roller-chopped in the 1980s (BLM personal communication) 

to decrease shrub cover and promote graminoid 

forage for cattle grazing and has relatively low sagebrush 

cover (6.99%). The Gypsum study area is located 

near the town of Gypsum, CO at an elevation of 2200m 

(Buckner and Bunin 1992). Relative to the Eagle site, 

this area has much higher cover of sagebrush (25.57%) 

and lower densities of P. harringtonii. Sagebrush cover 

was determined using the line-intercept method on the 

aerial cover of the shrubs and consisted of ten 60m transects 

per site. Based on a two-sample t-test, the amount 

of sagebrush cover was significantly different between 

the two study sites (P < 0.0001, alpha = 0.05, n = 10). 

Long-term Demographic Study 

At each site a 40 x 60m macroplot was installed at a 

location containing P. harringtonii and 1 x 60m quadrats 

were sampled within the macroplot based on a 

stratified random sampling method. The goal of the 

monitoring was to be statistically capable of detecting a 

20% change in the populations of P. harringtonii and 

was designed with the analysis having a power of 95% 

with a 1% chance of making a false-change error (Type 

I). Sample size and power analyses were conducted on 

the two initial years of data (1996 and 1997) to determine 

the appropriate number of quadrats for each site 

(Elzinga et al. 1998). The eight quadrats sampled at the 

Eagle site and 12 at Gypsum were sufficient to meet the 

desired power of the study. Within each quadrat the following 

data were collected: x and y coordinates, number 

of rosettes, and presence or absence of flowers, fruits 

Figure 1. Penstemon harringtonii (Photo by Carol Dawson). 

and herbivory. The quadrats were censused annually in 

early to mid-June. Repeated measures Analysis of Variance 

(ANOVA) and paired t-tests were utilized to determine 

statistically significant differences in rosette numbers 

over time or between years, respectively. Data were 

log transformed for statistical analysis. 

Plant Community Analysis 

In the summer of 2005 additional sampling of the 

macroplots was conducted as part of a Denver Botanic 

Garden internship investigating the relationship between 

A. tridentata density, soil moisture and the rare penstemon. 

Species richness and the density of sagebrush and 

Harrington’s penstemon were determined for 4 x 2m 

plots randomly located within the existing macroplots at 

the Eagle and Gypsum study sites. Sample size and 

power analysis determined that 17 and 20 plots were 

necessary for the Eagle and Gypsum study areas, respectively 

(Elzinga et al. 1998). Sample size was estimated 

using the 2005 data and a confidence level of 

90%. Analysis of the plant community data was conducted 

using Non-metric Multi-dimensional Scaling 

(NMS), a non-parametric multivariate ordination technique 

capable of detecting and describing vegetation 

patterns between the sites and correlating this inform- 

116


ation to species richness, soil moisture, and sagebrush 

density. The NMS used a Sorenson distance matrix and 

the presence or absence of plant species for the primary 

data matrix. The secondary matrix consisted of species 

richness per plot, categorical values for the ratio of Penstemon 

to Artemisia densities, and a dummy variable 

designating within which site the data are associated. 

Multi-Response Permutation Procedures (MRPP) provided 

a non-parametric statistical method to test for differences 

between groups (study sites) by comparing the 

heterogeneity within groups against the probability of 

occurrence by random chance. This type of randomization 

or permutation test provides an ‘A’ statistic and a 

‘P’ value that is used to determine if the two study sites 

are statistically significant from each other based on 

species compositions. Both NMS and MRPP were conducted 

in PC-Ord version 5.0 software (MjM Software 

1999). Linear regression and correlation coefficients 

(R 2 ) were determined between penstemon and sagebrush 

densities, soil moisture, and species richness using Microsoft 

Excel. 

RESULTS 

Long-term Demographic Study 

The 13 years of monitoring documented statistically 

significant variation in the number of rosettes per quadrat 

using a repeated measures ANOVA (Between subjects: 

P


Figure 4. Non-metric Multi-dimensional Scaling (NMS) of the plant species composition at the two study sites (Eagle 

and Gypsum). Triangles represent quadrats and plus signs are plant species with species codes as labels. Red vectors 

represent correlation coefficients with R 2 > 0.20 and are directionally aligned to plant species to which the vector 

variables are positively correlated. The length of the vector represents the magnitude of the correlation. 

data from the primary matrix. The vector for ‘species 

richness’ is oriented towards the sample space primarily 

occupied by the Eagle plots and represents a positive 

correlation between the Eagle samples and plots with 

higher species richness. The ‘dominance’ vector represents 

a ratio of P. harringtonii and A. tridentata densities 

and documents the positive correlation of the higher 

penstemon ratios with the Eagle samples. ‘Site’ is a 

categorical or ‘dummy’ variable necessary to code the 

sites and distinctly segregates the two study sites. Ordination 

analysis of soil moisture data and densities of 

Harrington’s penstemon or sagebrush did not determine 

any significant relationships or strong correlations. Although 

simple linear regressions determined that the 

Gypsum site had a weak negative relationship (R 2 = 

0.15) between the penstemon and sagebrush densities. 

Using linear regression to compare species richness with 

penstemon or sagebrush density, the two study sites had 

weak R 2 values, but the general trends were always 

opposite between the sites. At the Eagle site, species 

richness was positively related to penstemon density (R 2 

= 0.0856) or sagebrush density (R 2 = 0.0857). Conversely, 

at the Gypsum location the densities of penstemon 

and sagebrush were negatively correlated to species 

richness (R 2 = 0.108 and R 2 = 0.125, respectively). 

DISCUSSION 

The monitoring data documents that the P. harringtonii 

populations at Eagle and Gypsum are stable over 

the 13 years of monitoring (Figures 2 and 3), although 

they fluctuated greatly and reached the lowest population 

sizes during the droughts of the early 2000s. Increasing 

drought severity or frequency could have negative 

consequences for this species, especially if its habitat 

becomes more fragmented by development or degraded 

due to overuse. These results cannot be extrapolated 

to all populations of Harrington’s penstemon, but 

provide quantitative data concerning the population 

118


trends and supports the idea that the species is reproductive 

(data not presented) and recruitment is probably 

occurring at these locations. This monitoring project 

provided minimal data to inquire into the interactions of 

the plant community and resource use within the ecosystem. 

The multivariate ordination (NMS) analysis 

found interesting relationships between species richness 

and penstemon density (Figure 4), although linear regressions 

had weak results. We hypothesized that water 

is a driving resource in the system and that higher sagebrush 

density indirectly reduces species richness and 

penstemon density due to its control over the soil moisture 

and possibly space and nutrients. Our current data 

cannot support this hypothesis, but the initial analysis of 

the plant community data found interesting and relevant 

interactions that could lead to a better understanding of 

the rare species’ population and community dynamics. 

Sampling of additional sites and increased measuring of 

soil moisture will assist in the development of theories 

that relate sagebrush and penstemon densities to the 

availability of water and the effects of inter-specific 

competition. 

The Eagle study site was roller-chopped in the 1980s 

to reduce sagebrush cover and improve the system for 

grazing. This disturbance may have reduced sagebrush’s 

control over soil moisture and provided an appropriate 

disturbance for increased recruitment and establishment 

of P. harringtonii. An understanding of this ecosystem’s 

plant community dynamics and competition for water 

may provide clues to improving habitat of the rare penstemon, 

especially if the rate of habitat loss continues to 

accelerate in this rapidly developing region. If we improve 

our knowledge of the resources (water) and processes 

(disturbance) that regulate P. harringtonii it may 

be possible to develop management and restoration 

practices that promote higher density populations of the 

species by the manipulation of sagebrush cover and surface 

disturbance. 

REFERENCES 

Buckner, D.L. and J.E. Bunin. 1992. Final report: 

1990/91 status report for Penstemon harringtonii 

Penland. Unpublished report prepared for the Colorado 

Natural Areas Program, Denver, CO by Esco Associates, 

Inc., Boulder, CO. 

Caldwell, M.M., T.E. Dawson, and J.H. Richards. 

1998. Hydraulic lift: Consequences of water efflux from 

the roots of plants. Oecologia 113:151-161. 

Colorado Native Plant Society. 1997. Rare Plants of 

Colorado. 2nd Edition. Falcon Press. Helena, MT. 

Elzinga, C.L., D.W. Salzer and J.W. Willoughby. 

1998. Measuring and Monitoring Plant Populations. 

BLM Technical Reference 1730-1 (BLM/RS/ST- 

98/005+1730). 

Horton, J.L. and S.C. Hart. 1998. Hydraulic lift: a 

potentially important ecosystem process. TREE 13(6): 

232-235. 

Panjabi, S.S. and D.G. Anderson. (2006, June 30). 

Penstemon harringtonii Penland (Harrington’s beardtongue): 

a technical conservation assessment. [Online]. 

USDA Forest Service, Rocky Mountain Region. Available: 

http://www.fs.fed.us/r2/projects/scp/assessments/ 

penstemonharringtonii.pdf [1/7/09]. 

PC-ORD Version 5.0. 2005. MjM Software, Gleneden 

Beach, OR, USA. 

Spackman, S., B. Jennings, J. Coles, C. Dawson, M. 

Minton, A. Kratz, and C. Spurrier. 1997. Colorado Rare 

Plant Field Guide. Prepared for the Bureau of Land 

Management, the U.S. Forest Service and the U.S. Fish 

and Wildlife Service by the Colorado Natural Heritage 

Program. 

USDA, NRCS. 2009. The PLANTS Database (http:// 

plants.usda.gov, 27 May 2009). National Plant Data 

Center, Baton Rouge, LA 70874-4490 USA. 

119


Conservation and Restoration Research 

at The Arboretum at Flagstaff 

Kristin E. Haskins and Sheila Murray 

The Arboretum at Flagstaff, Flagstaff, AZ 

Abstract. The Colorado Plateau is experiencing increased climate change effects and population expansion. Many 

native plant species are at risk for becoming rare or threatened, and it is challenging to secure local, native plant seed 

for use in restoration. Here we highlight two examples of our conservation efforts for Astragalus cremnophylax var. 

cremnophylax in the Grand Canyon and Purshia subintegra in the Verde Valley of central Arizona, and discuss our 

involvement in a local native plant propagation movement conducted by the US Forest Service and the Museum of 

Northern Arizona. The Arboretum has been working toward rare plant conservation and restoration efforts for over 

25 years. 

SENTRY MILK-VETCH CONSERVATION 

Astragalus cremnophylax var. cremnophylax (Figure 

1) was listed as a species of concern in 1980 and was 

bumped up to endangered species status in 1990. This 

tiny legume is found only at Grand Canyon National 

Park (GCNP) in limestone outcrops. Threats to this species 

include development of the park and climate 

change. 

In 2005, The Arboretum began working with GCNP 

and The U.S. Fish and Wildlife Service to conserve this 

rare species. The initial task was to increase seed availability 

through ex-situ propagation. Typically, seeds are 

propagated in sterile potting mix that can provide mixed 

results in terms of seed germination and growth. In an 

effort to improve seedling performance, we examined 

the effects of different soil treatments: 1) potting soil 

with Rhizobium added and 2) potting soil with a native 

Figure 2. Astragalus cremnophylax var. cremnophylax 

grown with a native soil inoculum (left) and in standard, 

sterile potting soil (right). Photo by Sheila Murray. 

inoculum added versus traditional potting soil. After 

five months of growth, seedlings propagated with a native 

soil inoculum had significantly greater aboveground 

volume than seedlings grown in either of the other two 

treatments (Figure 2). We are currently tracking these 

plants to determine if these differences will translate 

into increased seed production. Ultimately, we aim to 

conserve Sentry Milk-vetch by expanding established 

and adding new populations at GCNP. 

Figure 1. Astragalus cremnophylax var. cremnophylax 

in bloom. Photo by Julie Crawford. 

ARIZONA CLIFFROSE RESTORATION 

Purshia subintegra, or Arizona cliffrose, is known 

from four populations in central Arizona, with the largest 

population occurring in the Verde Valley. This xeric, 

evergreen member of the Rosaceae was listed as endangered 

in 1984. Major threats include development, overgrazing 

and climate change. The Arizona Department of 

Transportation funded work by The Arboretum from 

120


1996-2000 in response to road construction that would 

result in a loss of Arizona cliffrose habitat. The Arboretum 

developed a protocol for propagating Arizona cliffrose 

via cuttings (Figure 3), as recent droughts had prevented 

the species from producing seed. Additionally, 

The Arboretum examined ways in which the propagated 

cuttings (Figure 4) could be put back in the field onto 

protected sites. Since out-planting in 2001, the Research 

Department has been involved in monitoring the new 

populations. We are happy to report that the new populations 

are doing well. 

ARIZONA NATIVE PLANT PROPAGATION 

In 2007, The Arboretum began a collaborative project 

with the U.S. Forest Service and The Museum of 

Northern Arizona to collect and propagate native seeds 

for use in local restoration efforts (Figure 5). This project 

arose in response to a high demand and lack of supply 

of local seed genotypes that were crucially needed 

after large scale forest fires hit the area in 2002. 

The first phase of the project is to collect and propagate 

seeds of native species that appeal to land managers 

for use in re-vegetation projects. Plant species are being 

chosen for their wildlife forage quality and likelihood of 

propagation success. We are also focusing on species 

that are not already in commercial production. Our goal 

is to start small, but eventually produce a reliable source 

for local seed genotypes that can be used by local land 

managers. 


The Research Department at The Arboretum at Flagstaff 

(a.k.a. Kris and Sheila) would like to thank all of 

our wonderful volunteers and the following groups for 

financial support: The U.S. Fish and Wildlife Service, 

Arizona Department of Transportation, and The U.S. 

Forest Service. 

Figure 4. An Arboretum volunteer helps re-pot Purshia 

subintegra in the research greenhouse. Photo by K. 

Haskins. 

Figure 3. Sheila Murray collects cuttings of Purshia 

subintegra in the Verde Valley, AZ. Photo by Joyce 

Maschinski. 

Figure 5. The research greenhouse at The Arboretum; 

where plant propagation will take place. Photo by K. 

Haskins. 

121


The Digital Atlas of Utah Plants: 

Determining Patterns of Biodiversity and Rarity 

Leila M. Shultz, R. Douglas Ramsey, Wanda Lindquist, and C. Garrard 

Floristics Lab and Remote Sensing/GIS Lab, Utah State University, Logan, UT 

Abstract. The digital Atlas of Utah Plants is a web-based revision of the Atlas of the Vascular Plants of Utah by B. 

Albee, L. Shultz, and S. Goodrich, published in 1988 by the University of Utah Museum of Natural History. The hard 

copy version provided distribution maps for 2,438 native or naturalized plant species and an appendix with generalized 

locations for 399 additional species that are either extremely rare, recently introduced, or at the edges of their 

ranges – and known from one population. The new digitized version allows analysis of original data from pre-1988 

herbarium collections (mapped at coarse level, resolution at approximately 10 km 2 ) and brings in herbarium records 

for post-1988 collections (most of which are mapped at high resolution with global positioning devices) as well as 

more than 6,000 observations and records of rare species from the database for the Utah Natural Heritage program. 

While specific location sites for rare species are not displayed on the web-based version, the presence of rare species 

is highlighted on the grid map displayed for each species. The digital version provides a tool for tracking reports of 

new records as well as a tool for analyzing patterns of diversity. Open access to these records is currently available 

and species lists have been compiled for each major ecoregion in the state. 

The on-line version of the Atlas of Utah Plants by 

Leila Shultz, R. Douglas Ramsey, and Wanda Lindquist 

is a revision of the Atlas of the Vascular Plants of Utah 

(Albee et al.1988). The new digital version displays a 

source code for each mapped point, new records, general 

locations for rare species, and nomenclatural updates. 

The original Atlas was based on the authors’ examination 

of approximately 400,000 herbarium specimens 

of Utah plants housed within the natural history collections 

of Brigham Young University (BRY), University 

of Utah (UT), Utah State University (UTC), the Forest 

Service Herbarium in Ogden (OGDF), and several Bureau 

of Land Management and Park Service herbaria. 

Although the original maps were hand-plotted on a gridded 

base map from specimens that were not entered in a 

database, the herbarium location for each voucher was 

color-coded on the map. These original maps are archived 

at the University of Utah Museum of Natural 

History’s Garrett Herbarium (Shultz et al. 1998; Ramsey 

and Shultz 2004). 

METHODS 

Technicians at the Remote Sensing/Geographic Information 

System laboratory of Utah State University 

hand-digitized the original maps. The first on-line version 

was sorted by family name with access through a 

web site. Development of GIS layers follows the methodology 

reported in Ramsey and Shultz (2004). The 

1992 version remained unchanged until the development 

of the digital version (Shultz et al. 2007). 

Voucher specimens on which the atlas is based represent 

more than 150 years of work by scores of dedicated 

professionals and amateurs, many of whom devoted 

their lives to tracking down unusual plants in some of 

the most isolated and physically challenging areas in 

North America. Authors of the original atlas (Beverly 

Albee, Leila Shultz, and Sherel Goodrich) spent approximately 

seven years critically examining and mapping 

locations for these approximately 400,000 collections 

that are housed primarily in Utah herbaria. Each 

point on the original maps is color-coded to show which 

herbarium houses the voucher for a particular record. 

Original maps are archived at the University of Utah’s 

Garrett Herbarium (UT). While the lack of specific 

voucher data is somewhat of a handicap, this problem 

will be corrected as on-line retrieval systems are developed 

through the developing consortium of herbaria in 

Utah. Once these data are available, users should be able 

to retrieve specific location records through the internet 

links to individual herbaria, or a consortium of herbaria 

(see the Intermountain Herbarium website for new reports, 

at http://herbarium.usu.edu/holdings_ specimen _ 

database.htm). 

RESULTS 

Digitized version. The digital version allows for 

analysis of distribution patterns and patterns of diversity 

within the state. The composite geographic information 

system for 2,840 plant species provides approximately 

77,000 locations (some with multiple records) in 10 X 

10 km grids (roughly equal to a township). The digital 

122


version displays new geographic records, new species, 

and the more than 500 nomenclatural changes that have 

taken place since publication of the hard copy in 1988 

(Figure 1). Additions include approximately 8,000 new 

location records (including 6,000 locality records from 

the Utah Natural Heritage program data files that are 

mapped at the resolution of township and range), and 

new entries for more than 400 rare species that were not 

included in the 1988 publication (including new reports 

in Welsh et al. 2003). It is based on a real-time mapping 

system that draws from numerous data layers and displays 

records on a map of Utah with a choice of backgrounds 

(satellite image, state map with county boundaries, 

or one with Nature Conservancy ecoregion boundaries). 

When viewing the distribution map, clicking on a 

point allows one to see the source of the information 

(see explanations below) as well as the elevation of the 

specific location. 

Nomenclatural updates and addition of rare species. 

Name changes resulting from nomenclatural revisions 

(Flora of North America 1993 – 2006) involve approximately 

16% of the names in the Utah flora. We worked 

to make the transition to new names as painless as possible: 

"old names" are retained, but shown in italics. 

When you click on an italicized name, you will be taken 

to the accepted "new name". Common names and nomenclatural 

changes are based on the USDA Plants Database 

(2009). Rare plants are highlighted in red letters, 

with the species taken from the state sensitive plant list 

(based on the Utah Native Plant Society [2009] and Bureau 

of Land Management Sensitive Species List). 

Species checklists. Species checklists for each of the 

major ecoregions in the state are provided on the web 

page. The checklists are found by clicking on the name 

of an eco-region on the right side of the web page. Information 

for each species listed includes the common 

Figure 1. Sample Page from the Digital Atlas of Utah Plants (http://earth.gis.usu.edu/plants) showing distribution of 

Abies concolor (White fir). A color-corrected satellite image is chosen as the background (note other choices available, 

including ecoregions), courtesy of the Remote Sensing/GIS Lab at Utah State University. 

123


Figure 2. Species richness patterns based on voucher specimens (Albee et al. 1988), relative to a hexagonal grid sampling 

frame (EPA 649 km 2 sampling frame). Areas highest in species richness are shown in red, with lowest species 

richness shown in blue. Patterns of species richness generally correlate with elevation: the higher the richness, the 

higher the elevation (Ramsey and Shultz 2004). 

name, family name, growth form (tree, shrub, grass, or 

forb), known elevational range within Utah, acronym 

code (USDA PLANTS database), and notation as to 

whether the species is native or introduced. Rare plants 

are also highlighted in red on these checklists. 

Analysis of biodiversity patterns. The digital version 

allows reports of total species richness by ecoregion as 

well as diversity within a uniform grid system (Figure 

2). The layers also can be manipulated to analyze patterns 

of species diversity between ecoregions. For comparisons 

of ecoregions within the state, see Shultz and 

others (2000). 

DISCUSSION 

Anyone attempting to represent the distribution of a 

biological organism knows that distributions are not 

124 

static and that maps can do no better than represent the 

distribution of a species at a specified point in time. 

Changing landscapes have a profound effect on the distribution 

of a species. In addition, the development of 

an atlas of plants contains a number of inherent problems 

-- primarily regarding accurate identification and 

scale. Due to the diversity of the vascular plants (with 

more than 20,000 species in North America according to 

the Flora of North America Editorial Committee 1992-- 

2006), reports of plant species generally cannot be 

trusted unless accompanied by a voucher specimen. 

That constraint severely limits the sample size and 

skews the kind of species represented by vouchers. In 

general, common species are under-represented in herbarium 

collections. However, rare occurrences are usually 

well-represented in herbarium collections and con-


sequently, distribution maps of rare species are highly 

reliable as to the total range of a species. 

When mapping from herbarium vouchers, botanists 

have the advantage of having a verifiable report that can 

be re-examined (and re-mapped) if the distribution is 

questionable. There is much biological and climatically 

important information to be gained from the mapping of 

a species distribution, but the user of a map should understand 

how the data are collected in order to understand 

what kinds of analyses can be performed. The 

first thing a user should understand is that a dot on a 

map will not guide the user to a specific spot on the 

ground. Imprecise location data for older herbarium 

specimens makes it impossible to map most reported 

plant distributions at a fine scale. New herbarium records, 

however, generally provide location data collected 

by global positioning systems. While these readings 

might be off by a hundred meters or so, the level of 

accuracy represents a considerable improvement over 

the older records. In order to understand how to interpret 

the reported distributions, one must first understand 

that the maps represent ranges of species rather than 

precise locations on the landscape. 

In developing the first atlas for Utah plants (Albee et 

al. 1988), the authors were fully aware that development 

of a collection database would be preferable to simply 

creating dot maps. However, constraints of time, cost of 

equipment, and limits of available technology in the 

“early days” of computers made it impossible to consider 

the development of a collection database. By our 

rough estimate, we calculated that such an undertaking 

would take at least twenty years. In addition, we knew 

that we would be overwhelmed by the problems inherent 

in mapping from a literal translation of herbarium 

data. We chose instead to spend our time checking the 

identification of each specimen, using the most current 

monographic or floristic treatments available. If a collection 

could not be mapped to the accuracy of township 

and range, we did not represent it with a dot. We could 

not map between twenty and thirty percent of all collections, 

and specimens from locations that were already 

mapped were not mapped again. We did, however, 

color-code each dot as to the herbarium from which the 

record was obtained. A questionable distribution, or 

one of particular interest, can thus be traced by consulting 

the archived maps at the University of Utah. A correction 

of distribution maps thus requires re-examination 

of specimens – a procedure that is highly recommended 

in the event of new studies or generic revisions. 

The lack of database-generated maps is not a great 

handicap at this time, but there should be considerable 

improvement in the future. If we (the authors) had access 

to a graphics tablet or a system for creating bar 

codes when we initiated the project, we would have 

simply placed a bar code on the specimen and linked it 

to a dot on a base map underlain by a graphics tablet. 

That kind of system would have been time-efficient, 

allowing later linkage to a database. Undoubtedly, new 

maps will be generated as collection databases are developed, 

and we can only hope that they will represent 

greater scale as well as a better way to track species distributions 

through time. 

For the present, the digitized atlas provides good estimates 

of species ranges within Utah, a mechanism for 

generating species lists for any specified area, relatively 

current nomenclature, and highly accurate estimates as 

to the number and distribution of rare species in the 

state. The authors of this paper encourage the use of the 

digitized atlas and invite readers to visit the “Virtual 

Utah” website at http://earth.gis.usu.edu/utah/. 


A grant from the Bureau of Land Management made 

this modern (post-2005) digital revision possible. Curators 

of the S.L. Welsh Herbarium of Brigham Young 

University (BRY), the Garrett Herbarium of University 

of Utah (UT), and the Intermountain Herbarium of Utah 

State University (UTC), provided enormous support 

throughout the years of examination of specimens. 

Since 2004, digital records from UTC and Utah Valley 

University (UVSC) have been added, for which we 

thank Michael Piep and Renee Van Buren. Staff associated 

with collections housed with various National 

Parks, Forests, and Bureau of Land Management offices 

in the state helped by providing access to collections. R. 

Douglas Ramsey had the vision to see the potential in 

developing the geospatial format and provided the technical 

support that made the project happen. Ben Franklin 

of the Division of Wildlife Resources sent records of 

rare species collections and observations in buffered, 

digitized format--a contribution of inestimable value. 

Walt Fertig provided his voucher data for Grand Staircase 

Escalante Monument, specimens deposited at BRY 

and UTC. Bonnie Banner, Thad Tilton, Tom Van Neil, 

Kent Braddy, and Chris Garrard of the Remote Sensing 

Lab of USU helped develop the geo-referenced database 

and create the original digital interface. Wanda 

Lindquist provided the programming expertise that allowed 

us to incorporate nomenclatural revisions, link to 

new data layers, and create species checklists. She designed 

the new web interface and integrated the complex 

system of new data layers. 


Albee, B.A., L.M. Shultz, and S. Goodrich. 1988. 

Atlas of the Vascular Plants of Utah. Univ. of Utah Museum 

of Nat. History, Occasional Publ. no.7. 670 p. 

Flora of North America Editorial Committee, 

eds. 1993+. Flora of North America North of Mexico. 

12+ vols. New York and Oxford. (Vol. 1, 1993; 

125


vol. 2, 1993; vol. 3, 1997; vol. 4, 2003; vol. 5, 2005; 

vol. 19, 2006; vol. 20, 2006; vol. 21, 2006; vol. 22, 

2000; vol. 23, 2002; vol. 25, 2003; vol. 26, 2002.) 

Ramsey, R.D. and L. Shultz. 2004. Evaluating the 

Geographic Distribution of Plants in Utah from the Vascular 

Atlas of Utah Plants. W. N. Amer. Nat. 64: 421- 

432. 

Shultz, L.M., N. Morin, and R.D. Ramsey. 1998. 

Floristics in North America: Tracking Rare Species 

Electronically. In C.I. Peng & P. P. Lowrey II, eds., Institute 

of Botany, Academia Sinica Monograph 16: 

259—273. 

Shultz, L.M., R.D. Ramsey, W. Lindquist. 2007. 

Digital Atlas of Utah Plants: http://earth.gis.usu.edu/ 

plants [revision ongoing, when citing, please give date 

of access] 

Shultz, L.M., R.D. Ramsey, and P. Terletzky. 2000. 

Using an atlas of vascular plants to determine patterns 

of biodiversity in Utah. Ecological Society of America 

85 th Annual Meetings August 6—10. ESA Abstracts: 

337. 

USDA, NRCS. 2009. The PLANTS Database (http:// 

plants.usda.gov, Accessed May 2007). National Plant 

Data Center, Baton Rouge, LA 70874-4490 USA. 

Utah Native Plant Society. 2009. Utah Rare Plant 

Guide (http://www.utahrareplants.org/rpg_species. 

html#All) 


Higgins. 2003. A Utah Flora, third edition. Brigham 

Young University, Provo, UT. 

Addendum 

A tool that allows users to extract species lists in database form, with accompanying information about species, 

has been added to the Digital Database website. Users can draw a polygon of any size within the Utah borders (using 

the ‘limit area’ command) and download a species list in .dbf format (using the ‘get list’ command). The list will be 

accompanied by information that includes common names, currently accepted acronyms, growth form, elevational 

range, nativity, etc. in an electronic format that can be incorporated into spreadsheets or database programs. The 

reference for the site is: 

Shultz, L.M., R.D. Ramsey, W. Lindquist, and C. Garrard. Utah State University, Logan, UT. (http://earth.gis.usu. 

edu/plants/). 

126


Molecular Genetic Diversity and Differentiation in Clay Phacelia 

(Phacelia argillacea Atwood: Hydrophyllaceae) 

Steven Harrison 

Brigham Young University, Provo, UT 

Susan E. Meyer, 

US Forest Service Rocky Mountain Research Station, Shrub Sciences Laboratory, Provo UT 

and Mikel Stevens, 

Brigham Young University, Provo, UT 

Abstract. Clay phacelia (Phacelia argillacea Atwood) was listed as federally endangered in 1978. It is known from 

only two populations in Spanish Fork Canyon, Utah. Samples were taken in each of three years from each of these 

two populations. We used AFLP markers to assess the genetic relatedness between the two populations and degree of 

differentiation between P. argillacea and three of its congeners. Six AFLP primer combinations resulted in 535 reliable 

marker loci of which 124 were polymorphic. Phacelia argillacea is genetically distinct from both its close and 

distant congeners. The two P. argillacea populations were not strongly differentiated, suggesting that gene flow between 

these populations probably occurred historically. In contrast, cohorts establishing in different years within a 

population were often genetically differentiated. Sampling in a single year would seriously underestimate genetic 

diversity in this species. 

Phacelia argillacea Atwood (clay phacelia) is a narrow 

endemic presently known from two locations approximately 

8 kilometers apart in the Spanish Fork Canyon, 

Utah County, Utah (Figure 1). The genus Phacelia 

is the largest in the Hydrophyllaceae. Phacelia argillacea 

is a member of the Crenulatae group of section Phacelia, 

subgenus Phacelia and is thought to be most 

closely allied to P. glandulosa Nutt., a species of wide 

distribution in extreme eastern Utah, western Colorado, 

Wyoming, eastern Idaho, and Montana (Atwood 1975). 

Another apparently closely allied species is the recently 

described P. argylensis Atwood (Welsh et al. 2003), 

known only from the type location in Argyle Canyon, 

Carbon County, Utah (Figure 1). The Crenulatae group 

(which consists of approximately 35 species) differs 

from other species of Phacelia in producing four-seeded 

capsules, faveolate seeds with a central ridge on the 

ventral side, and have a chromosome number of n=11 

(Atwood 1975). 

Recent taxonomic work in the genus Phacelia has 

confirmed the placement of P. glandulosa within the 

Crenulatae but has not included either P. argillacea or 

P. argylensis (Garrison 2007, Gilbert et al. 2005). Our 

goal was to examine molecular genetic diversity within 

and among the two known populations of P. argillacea 

as a necessary step in designing strategies for introducing 

new populations of this species on public land. We 

also wanted to make a preliminary assessment of the 

degree of differentiation between P. argillacea and its 

close congeners P. glandulosa and P. argylensis. We 

included the more distant congener P. crenulata Torr. 

ex S. Wats. as an out group in the analysis. 

Phacelia argillacea is an annual or biennial and has 

years when few or no actively growing plants are present. 

Its seeds germinate from spring to late summer or 

early fall and produce a rosette of leaves which grows 

during the winter months and bolts in the spring to produce 

a flowering shoot (Armstrong 1992, Meyer personal 

observation). The species is an edaphic endemic 

that is confined to steep hillsides of the Green River 

shale formation. Because of its restricted habitat and 

small and widely fluctuating population size, P. argillacea 

was declared an endangered species in 1978 (US 

Fish and Wildlife Service 1978, 1989). 

In 1990, The Nature Conservancy purchased the 

Tucker site, which at the time was the only known extant 

population for P. argillacea, and fenced it to prevent 

damage from grazing and trampling by deer and 

sheep and from disturbance caused by highway and railroad 

construction (Armstrong 1992). The plant was 

later rediscovered at the Railroad site, further down the 

canyon (Figure 1, inset). This population is on private 

land and is not fenced or managed for conservation. 

Known impacts to the Railroad site are highway widening 

coupled with the construction of a retaining wall, 

subsequent erosion, and trailing and grazing of domestic 

and native ungulates. P. argillacea may also be threatened 

by invasive weeds and drought. 

127


Figure 1. Collection sites for four species of Phacelia included in the AFLP (Amplified Fragment Length Polymorphism) 

study. Tissue samples were field-collected or greenhouse-grown tissue for P. argillacea and P. glandulosa 

for Site 2, while samples from the other P. glandulosa sites and for P. argylensis were obtained from specimens in 

the Brigham Young University herbarium. (Inset shows Spanish Fork Canyon, Utah, with extant P. argillacea subpopulations 

in black, reintroduction sites in green). 

Through U.S. Fish and Wildlife funding, a working 

group was organized in 2004 to focus on the introduction 

of P. argillacea into suitable and presumably previously 

occupied habitat on public land in Spanish Fork 

Canyon, an action recommended in the recovery plan 

for this species (US Fish and Wildlife Service 1989). 

Because many endangered plants with small population 

sizes and fragmented populations, such as P. argillacea, 

suffer higher risk of extinction due to genetic drift and 

inbreeding as well as stochastic environmental effects, 

an introduction program is almost essential for rare 

plants like P. argillacea (Kang et al. 2005). In 2007, 

seeds produced from greenhouse-grown individuals 

from the Tucker site were introduced at two new sites 

on US Forest Service land. The inset in Figure 1 depicts 

the location of the reintroduction sites (Mill Fork and 

Tie Fork) with reference to the extant P. argillacea 

populations. 

Analysis of the genetic diversity within and between 

the P. argillacea populations was also necessary because 

the species had never been studied at the molecular 

level. Analysis of the genetic diversity of an endan- 

gered plant species is a key element in the estimation of 

the viability of a population and can assist in conservation 

programs (Ronikier 2002, Kang et al. 2005). Data 

were also needed to address the question of whether P. 

argillacea was truly a distinct species or simply a disjunct 

population of P. glandulosa. Amplified Fragment 

Length Polymorphism (AFLP) was the molecular 

marker chosen to quantify genetic diversity. AFLPs 

were chosen because they have the potential to resolve 

genetic differences for individual identification and require 

no prior sequence knowledge of the organism. In 

this study AFLP analysis was used to address the following 

questions with regard to reintroduction of P. argillacea: 

(a) What is the level of genetic diversity in the 

two populations? (b) Is there genetic differentiation between 

the two populations? (c) Do samples collected 

within a population within a single year represent a random 

sample of genetic variation, or is there genetic differentiation 

between years? (d) Is P. argillacea genetically 

distinct from its close congeners P. argylensis and 

P. glandulosa? The answers to these questions should 

help to inform reintroduction efforts for this organism. 

128


MATERIALS AND METHODS 

Sample Collection 

Phacelia argillacea leaf tissue samples were collected 

from Tucker in the 2006 and 2008 field seasons 

and from Railroad in the 2006, 2007, and 2008 field 

seasons. A single basal leaf was removed nondestructively 

from each individual. The 2004 Tucker samples 

represent half-sibling progeny from wild-collected seeds 

of 15 maternal individuals (a total of 53 plants) in the 

2004 field season. These individuals were grown for 

seed production. Phacelia crenulata samples represented 

individuals greenhouse-grown from seeds of a 

bulk wild collection. Phacelia argylensis and some P. 

glandulosa samples were collected from Brigham 

Young University Herbarium; specimens were annotated 

as sampled for this study. The remaining P. glandulosa 

samples represent bulk population samples from 

two closely adjacent populations collected by Frank 

Smith in 2007 (Figure 1). 

DNA Extraction and AFLP Analysis 

Fresh leaf tissue samples for DNA extraction were 

dried over silica gel, lypophilized, or frozen at -80C immediately 

after collection. DNA was extracted from 

tissue samples using a Qiagen Plant Mini Kit (QIAGEN, 

Inc., Valencia, CA) with minor modifications in the protocol 

to achieve a higher concentration of DNA. 

AFLP analysis was carried out following Vos et al 

(1995) with minor modifications. The enzymes EcoRI 

and MseI were used for DNA digestion. Each plant 

sample was fingerprinted with six primer combinations. 

The primer extensions used were EcoAA/MseA, 

EcoAA/MseG, EcoAA/MseT, EcoAC/MseA, EcoAC/ 

MseG, and EcoAC/MseT. Fragment separation and detection 

was carried out on a LI-COR 4300 DNA Analysis 

System (LI-COR Biosciences, Lincoln, NE) on a 

6.5% polyacrylamide gel. Only unambiguous bands 

(50 – 350 bp) were scored for presence or absence. 

Bands that were monomorphic among all samples were 

discarded from analysis of polymorphic bands. Principal 

components analysis was performed on the complete 

data set, on data from the three close congeners alone, 

on data from P. argillacea alone, and on data from the 

Tucker half-sib families alone. We used SAS software 

(SAS Institute, Cary, NC) for the analysis. 

RESULTS 

AFLP analysis produced a total of 535 reliably reproducible 

bands, 124 of which were polymorphic. Seventy-five 

of these bands were polymorphic only between 

P. crenulata and the other Phacelia species (P. glandulosa, 

P. argylensis, and P. argillacea; Table 1). This 

clearly demonstrated that the P. glandulosa group is 

strongly genetically differentiated from P. crenulata, the 

putative distant congener in the study. The three close 

congeners were much more genetically similar. Phacelia 

argillacea exhibited nine bands that were polymorphic 

with P. argylensis, seven polymorphic bands 

within P. glandulosa from herbarium material, and 

seven polymorphic bands within P. glandulosa collected 

by Frank Smith in western Colorado. An unexpected 

result was that the Smith collections were even more 

differentiated from other P. glandulosa than was P. argillacea, 

with 15 bands polymorphic between the two 

groups. In contrast, P. argylensis was closely similar to 

the herbarium-collected P. glandulosa group, with only 

3 polymorphic bands. Within P. argillacea, we observed 

a total of 30 polymorphic bands, however no polymorphic 

bands were found between the Tucker and 

Railroad populations, suggesting low genetic differentiation. 

When we analyzed data from all four Phacelia species 

included in the study, the first principal component 

represented 79% of the total variation and clearly separated 

P. crenulata form the other three species, reflect- 

Table 1. Number of polymorphisms identified using the AFLP (amplified fragment length polymorphism) 

technique between or within pairs of species or, in the case of P. glandulosa, within-species groups. 

P. argylensis P. glandulosa (H) P. glandulosa (F) P. crenulata 

P. argillacea 9 7 7 78 

P. argylensis 3 18 93 

P. glandulosa (H) 15 87 

P. glandulosa (F) 83 

(H) Herbarium-collected samples from individual herbarium specimens. 

(F) Field-collected bulk samples from two closely adjacent populations. 

129


ing the distant relationship between it and the other 

Phacelia samples. This was a consequence of the large 

number of AFLP bands that were polymorphic between 

P. crenulata and the other three species (Table 1). 

PCA was then used to analyze the relationships between 

P. argylensis, P. glandulosa, and P. argillacea. 

The first two principal components represented 9.2 and 

7.2% of the total variation. The P. argillacea samples 

clearly grouped separately from samples of the other 

two species (Figure 2). In contrast, P. argylensis 

grouped closely with the herbarium-collected P. glandulosa 

group, calling its status as a separate species into 

question. The field-collected western Colorado P. glandulosa 

was most distant from the herbarium-collected P. 

glandulosa group, as was also indicated by the large 

number of polymorphic bands between these two sets of 

collections, suggesting that it perhaps represents an undescribed 

taxon within the group (Table 1). 

When PCA was applied to data from the P. argillacea 

samples from both populations and among different 

years, the first two principal components, which explained 

7.5 and 2.9% of the total variation, provided 

enough separation to compare populations and years 

(Figure 3). The resulting data grouped each sample with 

cohorts from the same year more closely than by population. 

For example, there was no overlap between 

Tucker 2006 samples and the 2004 and 2008 samples 

from the same population. Similarly, there was no overlap 

between Railroad 2008 samples and the 2006 and 

2007 samples of the same population. In addition, individuals 

from the Railroad population were surrounded 

Figure 2. Scores on the first two axes from Principal Components Analysis of AFLP (amplified fragment length 

polymorphism) data for three species of Phacelia. The P. glandulosa point near the lower left hand corner of the 

graph represents two bulked field-collected samples from closely adjacent populations; all other points represent individual 

plants, or multiple individuals with identical genotypes. 

130



polymorphism) data for P. argillacea individuals collected from each of two populations in each of three years. 

by Tucker individuals on the plot, showing no clear genetic 

differentiation between the two populations. 

PCA was also used to differentiate the greenhousegrown 

half-sib individuals from Tucker 2004. The first 

two principal components, even though they explained 

only 3.4 and 1.7% of the total variation, generally separated 

the samples into their respective families (Figure 

4). The graphs indicate that members of half-sib families 

tend to resemble each other more closely than samples 

belonging to different half-sib families. 

DISCUSSION 

Although the AFLP marker system is not well suited 

for phylogenetic analysis per se, it is useful for examining 

the degree of genetic distinctness among closely 

related populations and species. In this study, PCA 

analysis of the AFLP bands suggests that P. argillacea 

is distinct from both its closest congeners (P. argylensis 

and P. glandulosa). Additionally, these results indic- 

ated that P. argylensis is more closely related to P. glandulosa 

than is P. argillacea, and may not be distinct 

from P. glandulosa. Our analysis also suggests that the 

differences within P. glandulosa as presently described 

may be greater than the differences between P. glandulosa 

and P. argillacea. A close examination of the population 

that was the source of the field-collected P. glandulosa 

samples from western Colorado may reveal that 

these samples represent a distinct and previously undescribed 

taxon. 

The AFLP analysis revealed that P. argillacea appears 

to have a surprising amount of genetic diversity 

for a species of such limited distribution. Of the total 

polymorphic bands encountered, 24%, or 30 bands, 

were polymorphic just within P. argillacea. These 

polymorphisms were distributed within populations, as 

there were no polymorphic bands between the Tucker 

and Railroad populations, and the PCA showed no distinct 

pattern by population. Instead, the PCA of P. 

131


4 

Family 7 

4 

Family 11 

4 

Family 14 

4 

Family 16 

2 

2 

2 

2 

0 

0 

0 

0 

-2 

-2 

-2 

-2 

-4 

-6 -4 -2 0 2 4 

4 

2 

Family 17 

-4 

-6 -4 -2 0 2 4 

4 

2 

Family 21 

-4 

-6 -4 -2 0 2 4 

4 

2 

Family 24 

-4 

-6 -4 -2 0 2 4 

4 

2 

Family 26 

PRINCIPAL COMPONENT 2 

0 

-2 

-4 

-6 -4 -2 0 2 4 

4 

2 

0 

-2 

4 

2 

0 

-2 

Family 27 

-4 

-4 

-6 -4 -2 0 2 4 -6 -4 -2 0 2 4 

Family 39 

-4 

-6 -4 -2 0 2 4 

0 

-2 

-4 

-6 -4 -2 0 2 4 

4 

2 

0 

-2 

4 

2 

0 

-2 

Family 33 

Family 41 

-4 

-6 -4 -2 0 2 4 

0 

-2 

-4 

-6 -4 -2 0 2 4 

4 

2 

0 

-2 

Family 35 

-4 

-6 -4 -2 0 2 4 

4 

2 

0 

-2 

Family 42 

-4 

-6 -4 -2 0 2 4 

PCA 2 

0 

-2 

-4 

-6 -4 -2 0 2 4 

4 

2 

0 

-2 

Family 37 

-4 

-6 -4 -2 0 2 4 

One Individual 

Two Individuals 

Three Individuals 

Four Individuals 

Five Individuals 

PRINCIPAL COMPONENT 1 


polymorphism) data for greenhouse-grown individuals belonging to 15 half-sib familes collected from the P. argillacea 

Tucker population in 2004. Point size reflects number of individuals with identical AFLP genotypes. 

argillacea indicated that individuals tended to group 

together by year much more than by population. These 

data suggest a persistent seed bank, which means a fraction 

of the seeds not only remain in the soil, but are viable 

for at least one year after production (Thompson 

and Grime 1979). A site characterization study of P. 

argillacea supported this hypothesis by suggesting that 

the seed bank reservoir contains an accumulation of 

seeds from many different years (Armstrong 1992). 

Preliminary results from a long term seed retrieval study 

with P. argillacea also support the existence of a longlived 

seed bank in this species. Few or no seeds have 

germinated in the field during the first two years, and 

most are still in a state of primary dormancy (Meyer 

unpublished data). This type of seed bank structure has 

also been reported in Phacelia secunda. Seeds from P. 

secunda were collected and allowed to germinate, and 

after three years a considerable fraction of the seeds remained 

viable but ungerminated (Cavieres 2001). 

PCA also suggests that collections from a single year 

and population of P. argillacea would depict a very narrow 

genetic diversity within the organism. A persistent 

seed bank can function as a genetic memory by accumulating 

seed genotypes from different years (Cabin et al. 

1998). In the case of the rare annual Clarkia springvillensis, 

analysis of seed bank samples illustrated significantly 

higher within-seed bank genetic diversity when 

compared to the adult population (McCue and Holtsford 

1998). The same could be true for P. argillacea, as evidenced 

by the pattern seen in the PCA (Figure 3). The 

seed bank must have a higher genetic diversity than the 

established plants in any one year, because of the wide 

range of diversity seen when comparing years. A similar 

situation was found in Phacelia dubia, which has small 

132


population size and was thought to suffer from genetic 

drift and bottlenecking. However, analysis of the seed 

bank and adult plants from different years showed that 

no alleles were lost. The seed bank was found to store 

the full range of different genotypes (del Castillo 1994). 

Our study suggests that P. argillacea exhibits this type 

of age-structured seed bank and genetic pattern. In addition, 

the age and genotype of a seed may play a part in 

permitting it to germinate and establish in a particular 

kind of year. Allowing only certain genotypes to germinate 

each year would produce a pattern similar to the 

one found in Figure 3, with little or no overlap in genotypes 

between years. 

CONCLUSIONS 

In the current reintroduction efforts with P. argillacea, 

by selecting seeds collected in just one year we 

may be severely limiting the genetic base of this species. 

As the data from this study and similar studies suggest, 

in cases of a persistent seed bank, the parents of 

each year’s crop can differ from the seedling cohorts 

found in the years before and after (Figure 4). By using 

only greenhouse-grown seeds produced from the Tucker 

2004 seed collection, we were inadvertently selecting 

for only a few specific genotypes. With individuals from 

just one year, the reintroduction program will almost 

surely suffer from inbreeding and genetic bottlenecks. 

To broaden the genetic base of this organism and allow 

for establishment of successful new populations of P. 

argillacea, the reintroduction program needs to include 

collections from several years and from both populations. 


This work was carried out with the aid of funding 

from the US Fish and Wildlife Service Preventing Extinction 

Program. We thank Heather Barnes (USFWS), 

Denise Van Keuren (formerly of the Uinta National Forest), 

Renee Van Buren and Kimball Harper (Utah Valley 

University), Katie Temus Merill and Ben Brulotte 

(Brigham Young University), Wendy Yates, Jennifer 

Lewinsohn, and Rita Dodge (Red Butte Gardens), Frank 

Smith (Western Ecological Services), and Bitsy Shultz 

and Susan Garvin Fitts (Utah Native Plant Society) for 

logistical assistance. Permission to collect tissue samples 

from specimens at the Brigham Young University 

Herbarium is also gratefully acknowledged. 


Armstrong, V.R. 1992. Site characteristics and habitat 

requirements of the endangered Clay Phacelia 

(Phacelia argillacea Atwood, Hydrophyllaceae). M.S. 

thesis. Brigham Young University. 

Atwood, N.D. 1975. A revision of the Phacelia 

Crenulatae group (Hydrophyllaceae) for North America. 

Great Basin Naturalist 35: 1-190. 

Cabin, R.J., R.J. Mitchell, and D.L. Marshall. 1998. 

Do surface plant and soil seed bank populations differ 

genetically? A multipopulation study of the desert mustard 

Lesquerella fendleri (Brassicaceae). American 

Journal of Botany 85(9): 1098-1109. 

Cavieres, L.A. and M.T.K. Arroyo. 2001. Persistent 

soil seed banks in Phacelia secunda (Hydrophyllaceae): 

experimental detection of variation along an altitude 

gradient in the Andes of central Chile (33°S). Journal of 

Ecology 89: 31-39. 

del Castillo, R.F. 1994. Factors influencing the genetic 

structure of Phacelia dubia, a species with a seedbank 

and large fluctuations in population size. Heredity 

72: 446-458. 

Garrison, L. 2007. Phylogenetic relationships in Phacelia 

(Boraginaceae) inferred from nrITS sequence data. 

M.S. thesis. San Francisco State University. 

Gilbert, C., J. Dempcy, C. Ganong, R. Patterson, and 

G.S. Spicer. 2005. Phylogenetic relationships within 

Phacelia subgenus Phacelia (Hydrophyllaceae) inferred 

from nuclear rDNA ITS sequence data. Systematic Botany 

30: 627-634. 

Kang, M., Q. Ye, and H. Huang. 2005. Genetic consequence 

of restricted habitat and population decline in 

endangered Isoetes sinensis (Isoetaceae). Annals of Botany 

96: 1265-1274. 

McCue, K.A. and T.P. Holtsford. 1998. Seed bank 

influences on genetic diversity in the rare annual Clarkia 

springvillensis (Onagraceae). American Journal of 

Botany 85: 30-36. 

Ronikier, M. 2002. The use of AFLP markers in conservation 

genetics – a case study on Pulsatilla vernalis 

in the Polish Lowlands. Cellular and Molecular Biology 

Letters 7: 677-684. 

Thompson, K and J.P. Grime. 1979. Seasonal variation 

in the seed banks of herbaceous species in ten contrasting 

habitats. Journal of Ecology 67: 893-921. 

U.S. Fish and Wildlife Service. 1978. Determination 

of five plants as endangered species. Federal Register 

43: 44809-44812. 

U.S. Fish and Wildlife Service. 1989. Clay phacelia, 

Phacelia argillacea Atwood, recovery plan. USFWS 

Region 6, Denver Colorado. iii + 19 pp. 

Vos, P. et al. 1995. AFLP: a new technique for DNA 

fingerprinting. Nucleic Acids Research. 23: 4407-4414. 


Higgins. 2003. A Utah Flora, third edition. Brigham 

Young University, Provo, UT. 

133


A Taxonomic Revision of Astragalus lentiginosus var. maricopae 

and Astragalus lentiginosus var. ursinus 

Two Taxa Endemic to the Southwestern United States 

Jason A. Alexander, curator 

Utah Valley University Herbarium, Utah Valley University, Orem, UT 

and research associate, Wesley E. Niles Herbarium, University of Nevada, Las Vegas 

Abstract. Two taxa in the Astragalus lentiginosus complex of Section Diphysi, Astragalus lentiginosus var. maricopae 

and A. lentiginosus var. ursinus, have been historically overlooked by taxonomists and have had an uncertain 

taxonomic status. Astragalus lentiginosus var. maricopae is a highly endangered endemic (likely totaling less than 

5,000 individuals primarily due to habitat loss from development) and confined to a small region of igneous and granitic 

alluvial fans in the vicinity of Scottsdale and the Verde River drainage in northern Maricopa Co., Arizona. The 

second variety, A. lentiginosus var. ursinus, is a highly restricted limestone talus endemic (totaling less than 5,000 

individuals) and is confined to a small region of the Beaver Dam Mountains in Mohave Co., Arizona and Washington 

Co., Utah. Two morphological principal coordinates analyses (PCoA) were used on vouchers of these two varieties 

and nearly 150 specimens from related taxa in Section Diphysi. The results of the first PCoA showed that the floral 

and pod morphology of A. lentiginosus var. maricopae contributed highly to its distinctiveness when compared to 

other varieties, especially A. lentiginosus var. wilsonii (its geographically closest relative ). These results combined 

with field observations indicate that A. lentiginosus var. maricopae is a morphologically unique and highly endangered 

taxon that is threatened by disturbance and development throughout its known range. Based on the second 

PCoA, A. lentiginosus var. ursinus trends toward smaller pods and flowers than its geographically nearest relative (A. 

lentiginosus var. mokiacensis) and is herein recognized at the varietal level. Astragalus lentiginosus var. ursinus is 

more ecologically specialized than A. lentiginosus var. maricopae. However, most of the population is in a wilderness 

area and is threatened by recreational activities, not extirpation by suburban development. 

Marcus E. Jones (1923) was one of the first taxonomists 

to attempt to write a comprehensive treatment of 

species previously considered related to Astragalus lentiginosus 

Douglas ex Hook., a species described almost 

a century earlier by Hooker (1831). After Hooker's publication, 

new species were added to this complex by Asa 

Gray (1849, 1865) and Sereno Watson (1871), but the 

complex remained poorly known until the late 19th century. 

In 1898, Jones proposed a set of new combinations, 

placing some species from Section Diphysi A. 

Gray (sensu Gray 1863) as varieties within a greatly 

expanded concept of Astragalus lentiginosus (Jones 

1898). Jones' concept of A. lentiginosus remained relatively 

unchanged and culminated in his Revision of 

North-American Species of Astragalus (Jones 1923) 

which was ignored by taxonomists for two decades 

(Barneby 1964). Jones' core varietal concepts in Astragalus 

lentiginosus are largely accepted today, mainly due 

to the eloquence and precision of Rupert C. Barneby in 

his 1945 and 1964 monographs. 

Barneby (1945) was the first to make explicit and 

unambiguous the relationships between the 40 varieties 

of Astragalus lentiginosus and their placement into Astragalus 

Section Diphysi. Although many new varieties 

would be described and old ones further refined over the 

next 60 years (see Barneby 1956, 1989, Isely 1998, 

134 

Kearney & Peebles 1960, Munz & Keck 1959, Welsh 

1978, 1993, 2003), the recognition of 42 varieties remains 

even in the latest monograph (Welsh 2007). Despite 

the many revisions of this complex, the best description 

of the extremes of the diversity within the A. 

lentiginosus complex was published by Barneby in his 

first monograph: 

"The varieties of A. lentiginosus, as known at 

present, are not of equal stature: some, indeed, 

are doubtfully distinct, while others appear to 

be isolated and might, in another group of the 

genus, pass as species of the first rank. It is noticeable, 

however, that every example of the latter 

type is comparatively little known, whereas 

all those represented by extensive collections 

are found to intergrade at some point with a 

related variety" (1945: 70). 

The Astragalus lentiginosus complex can be divided 

into two major groups based on pod morphology. The 

first group, comprising the majority of the varieties of 

this complex, is distinguished by the presence of a deciduous, 

bladdery inflated, biloculate, ovoid to orbicular 

pod. These characters were viewed as the most representative 

by Barneby (1964) and used to distinguish


Section Diphysi from other putatively related sections. 

The thickness or texture of the valve walls, the type and 

distribution of pubescence on the valves, and the degree 

of closure of the locules by the septum (if it is complete 

and fused to the funicular flange throughout or just 

within a portion of the body of the pod) is highly variable 

throughout the range of this species. 

The second group within this complex is characterized 

by having scarcely inflated (cylindrical to ventricose 

in shape and slightly inflated dorsally, if at all), 

thick papery to leathery, elliptic, narrowly oblong, to 

linear pods. The septum is generally incomplete in this 

group, either semi-bilocular (the septum partially divides 

the two locules) or sub-unilocular (the septum is 

less than half the width of the locule). Unlike the first 

group, the pods are either deciduous or long-persistent. 

These scarcely inflated taxa were first comprehensively 

described by Rydberg (1929) as Section Palantia 

Rydb. within the genus Tium Medik., based on the similarity 

of the pod morphology. The remaining members 

of Section Diphysi were split and included in the old 

world genus Cystium Steven. The degree to which these 

characteristics define a section or species is a major 

source of disagreement among all monographs of this 

complex (Barneby 1945, 1964, Isely 1998, Jones 1923, 

Rydberg 1929, Welsh 2007). The long-persistent pods 

in one taxon, A. lentiginosus var. mokiacensis (A. Gray) 

M.E. Jones, was the primary character that lead Barneby 

(1989) to retain it within the Section Preussiani M.E. 

Jones, a section distantly related to Section Diphysi. 

Vastly different interpretations of the significance of 

this character have led to often disparate views of the 

species boundaries and delimitations surrounding these 

taxa (Alexander 2005, Barneby 1964, 1989, Welsh 

2007). When the taxa with persistent pods are delimited 

as varieties, Astragalus lentiginosus becomes the only 

documented North American species of Astragalus to 

have forms with both persistent and deciduous pods. 

All other examples proposed by taxonomists have been 

split at the species or the sub-sectional level in recent 

monographs. 

In the most recent revision, Alexander (2008) considered 

all of the scarcely inflated varieties of A. lentiginosus 

to be a single evolutionary lineage and referred to 

them collectively as the Palantia, based on Rydberg's 

sectional name. The Palantia consists of A. lentiginosus 

var. bryantii (Barneby) J.A. Alexander, A. lentiginosus 

var. iodanthus (S. Watson) J.A. Alexander, A. lentiginosus 

var. maricopae Barneby, A. lentiginosus var. 

mokiacensis (including A. lentiginosus var. trumbullensis 

S.L. Welsh & Atwood), A. lentiginosus var. palans 

(M.E. Jones) M.E. Jones, A. lentiginosus var. pseudiodanthus 

(Barneby) J.A. Alexander, A. lentiginosus var. 

ursinus (A. Gray) Barneby, and A. lentiginosus var. wilsonii 

(Greene) Barneby. The name, Palantia, is not used 

herein in a nomenclatural sense as a sub-generic or sectional 

name. It is unusual in botany for a collective 

name to be required for clarity when referring to groups 

of morphologically similar varieties. However, with 

over 40 varieties and 4 or 5 lineages that are similar 

morphologically, a collective naming convention for A. 

lentiginosus is necessary to refer to these groups. As 

such, the Palantia is used as a convenient, informal 

name for the varietal group with scarcely inflated pods 

and is italicized following the common literary convention 

for unfamiliar Latin words. 

Within the Palantia, two taxa, Astragalus lentiginosus 

var. maricopae and A. lentiginosus var. ursinus have 

been historically overlooked by taxonomists and have 

an uncertain taxonomic status. Astragalus ursinus A. 

Gray (first reduced to a variety of A. lentiginosus by 

Barneby 1964) was the first to be described in 1878, 

along with A. mokiacensis A. Gray (Gray 1878). Gray 

was the first to propose a close relationship between 

these taxa and A. lentiginosus var. iodanthus (at that 

time, and until only recently, this taxon was delimited as 

a species). The affinity of the types of A. ursinus to extant 

populations has been controversial since the taxon 

was first described (Alexander 2005, 2008, Barneby 

1964, Jones 1923, Welsh 1978, 2007, Welsh & Atwood 

2001). In some recent taxonomic treatments (Barneby 

1989, Welsh 1993), the types have been regarded as an 

insignificant variant of A. lentiginosus var. palans. 

However, in others A. lentiginosus var. ursinus is recognized 

as an insignificant variant of A. lentiginosus var. 

mokiacensis (Alexander 2005, Welsh 2007, Welsh & 

Atwood 2001). 

Based on a combination of field surveys, morphometric 

analyses and chloroplast haplotype analyses, Astragalus 

lentiginosus var. ursinus was found to be genetically 

distinct from its geographically closest relative, 

A. lentiginosus var. mokiacensis (Alexander 2008, Alexander 

& Liston, in prep). In addition, Astragalus lentiginosus 

var. ursinus is a highly restricted limestone talus 

endemic (totaling less than 5,000 individuals), and is 

confined to a small region of the Beaver Dam Mountains 

in Mohave County, Arizona and Washington 

County, Utah (Alexander 2008). 

The second species, Astragalus lentiginosus var. 

maricopae, was first described in Barneby's (1945) 

monograph. It is often confused with A. lentiginosus 

var. yuccanus due to similar floral morphology (size and 

color) and has remained poorly known since it was first 

described. The floral and pod morphology are highly 

distinct when compared to the other members of the 

Palantia. If it were placed within any other section of 

the genus, it would be recognized at the species-level. It 

has universally been recognized as a variety of A. lentiginosus 

in all major monographs, but has only recently 

been found to be a restricted endemic (Alexander 2008). 

135


Though A. lentiginosus var. maricopae is morphologically 

distinct from its geographically closest relative, A. 

lentiginosus var. wilsonii, it is not as genetically differentiated 

as A. lentiginosus var. ursinus is from A. lentiginosus 

var. mokiacensis (Alexander 2008, Alexander 

& Liston, in prep). Alexander (2008) also found that A. 

lentiginosus var. maricopae is a highly endangered endemic 

(likely totaling less than 5,000 individuals primarily 

due to habitat loss from suburban development) 

and confined to a small region of igneous and granitic 

alluvial fans in the vicinity of Scottsdale and the Verde 

River drainage in northern Maricopa County, Arizona. 

In this study, morphological principal coordinates 

analyses (PCoA), cluster analyses, and cladistic analyses 

are used to detect the degree of morphological differentiation 

between Astragalus lentiginosus var. maricopae, 

A. lentiginosus var. ursinus and the remaining 

taxa of the Palantia and whether this differentiation corresponds 

to species or varietal delimitations in preparation 

for monographic revision of the A. lentiginosus 

complex. 

MATERIALS AND METHODS 

Field observations and voucher specimens were 

made from spring 2001 to summer 2004 throughout the 

range of the Palantia. Most populations were visited 

several times and were observed during early flower, 

maturity, and senescence. Vouchers for this study are 

deposited at NY, OSC, RSA, UNLV, and UVSC. 

Herbarium specimens were examined at UC in December 

of 1999, BRY in August of 2000, GH in August 

of 2002, NY in October of 2003, and UNLV in July of 

2002 and 2003. Additional herbarium specimens were 

obtained on loan from BRY, CAS, DS, K, POM, RM, 

and RSA. 

Specimens from taxa in Astragalus Section Diphysi 

were examined for two morphological PCoA studies. 

The first, the 153 specimen Palantia PCoA, focused on 

heavily sampling all members of the group and was designed 

to evaluate the morphological distinctiveness of 

A. lentiginosus var. maricopae. 103 specimens of A. 

lentiginosus var. palans were used representing all major 

regions of its range, including the type population. 

Also included were multiple specimens of A. lentiginosus 

var. bryantii (10), A. lentiginosus var. ursinus (10), 

and A. lentiginosus var. wilsonii (15). Due to the poor 

nature of most herbarium specimens of A. lentiginosus 

var. maricopae, 5 specimens were collected in the field 

in 2005 for the morphological analysis. To ensure that 

the range of regional variation was present in the PCoA, 

representative specimens (one except where noted) were 

included from these morphologically similar and geographically 

proximal varieties: A. lentiginosus var. ambiguus 

Barneby (the type specimen); A. lentiginosus var. 

araneosus (Sheld.) Barneby; a population of a A. lent- 

iginosus (from Chloride, Mohave County, Arizona, interpreted 

herein as an intermediate to A. lentiginosus 

var. yuccanus M.E. Jones) considered part of A. lentiginosus 

var. ambiguus in Barneby (1964); A. lentiginosus 

var. iodanthus; A. lentiginosus var. mokiacensis (2 type 

specimens); A. lentiginosus var. pseudiodanthus; A. lentiginosus 

var. stramineus (Rydb.) Barneby (the type 

specimen); A. lentiginosus var. vitreus Barneby (the 

type specimen); and A. lentiginosus var. yuccanus (the 

type specimen). 

The second, the 43 specimen Astragalus lentiginosus 

var. mokiacensis PCoA, focused on further evaluating 

the morphological distinctiveness of A. lentiginosus var. 

ursinus. Fifteen specimens from throughout the range 

of A. lentiginosus var. ursinus were examined, including 

the type specimens. For comparison, 28 specimens from 

throughout the range of A. lentiginosus var. mokiacensis 

(including the type specimens) were examined. Multiple 

specimens and types of the synonym, A. lentiginosus 

var. trumbullensis S.L. Welsh & N.D. Atwood were 

included in this study. These data are a modified version 

of the data used in Alexander (2005). 

In both PCoA's, duplicate specimens or specimens 

from the same locality were used to determine the character 

states of missing data. Qualitative characters that 

were found to be polymorphic within a single individual 

were excluded. 

The morphological matrices for both studies were 

transformed into a Gower (1971) similarity matrix, a 

process that is not sensitive to data sets with mixed ordinal, 

nominal, continuous, and ratio data types. The matrix 

was then used in the PCoA. A Kendall's tau correlation 

between the PCoA axes and all morphological characters 

was used to determine the magnitude of the contribution 

of characters to the overall analysis (Easdale et 

al. 2007, Hammer et al. 2001). All correlations and 

PCoA analyses were performed using Paleontological 

Statistics (PAST) version 1.76 (Hammer et al. 2001). A 

Euclidean distance dendrogram was also created using 

PAST for the cluster analysis. 

PAUP* for Windows version 4.0 beta 10 (Swofford, 

2002) was used to assess the relationships among 30 

specimens of the Palantia and Section Diphysi using 

cladistic methodologies. 

The PCoA, cluster, and parsimony analyses were 

used to address the following questions: 

1) Do populations of A. lentiginosus var. maricopae and 

A. lentiginosus var. ursinus form groups discrete from 

the other members of the Palantia ? 

2) Which morphological characters contribute to the 

observed groups? 

3) Are these groups morphologically differentiated from 

the closely related and geographically proximal varieties 

of A. lentiginosus? 

136


RESULTS 

Distribution maps (Figures 1, 2) show the localities 

of specimens of Astragalus lentiginosus var. maricopae 

and A. lentiginosus var. ursinus examined for this study. 

Specific vouchers labeled in figures are shown in Table 

1. A detailed list of the vouchers examined can be found 

in the Taxonomic Treatment and in Appendix 1. 

Of the 24 morphological characters examined for the 

153 specimen Palantia PCoA, two were constant and 

not used (Table 2), one was discarded due to character 

state scoring issues, and 22 were variable. The first 

component of the PCoA explained 22.1% of the total 

variance (Table 3). The largest correlations to the first 

axis were from pod raceme orientation (podro), degree 

of pod incurve (podpi), pod pedicel orientation (podpo), 

banner color (bannc), and pod persistence (poddp). The 

second component of the PCoA explained 11.8% of the 

total variance. The largest correlations to the second 

axis were from pod pubescence (podpu), leaf abaxial 

pubescence (leafab), pod shape in cross section (podsc), 

and pod inflation (podin). All other components of the 

PCoA explained less than 10% of the total variance. The 

scatterplot of the first two components of the PCoA is 

shown in Figure 3. 

Of the 24 morphological characters examined for the 

43 specimen Astragalus lentiginosus var. mokiacensis 

PCoA, 12 were found to be variable in this subset of the 

data (Table 4). The first component of the PCoA explained 

48.4% of the total variation (Table 5). The largest 

correlations to the first axis were from pod shape in 

Figure 1. Distribution map of Astragalus lentiginosus 

var. maricopae in Arizona. The distributions of Astragalus 

lentiginosus var. bryantii and A. lentiginosus var. 

wilsonii are shown for reference. Table one contains the 

legend of letter codes used for specific vouchers shown. 

Vouchers and major populations are labeled as follows: 

(caret), A. lentiginosus var. bryantii; (upward triangle), 

A. lentiginosus var. maricopae; (circle), A. lentiginosus 

var. wilsonii; (X), Cameron population of A. lentiginosus 

var. wilsonii putatively intermediate to A. lentiginosus 

var. palans. 

Figure 2. Distribution map of Astragalus 

lentiginosus var. ursinus. The distribution 

of A. lentiginosus var. mokiacensis 

is shown for reference. Table 1 contains 

the legend of letter codes used for 

specific vouchers shown. Taxa are labeled 

as follows: (X) A. lentiginosus var. 

ursinus; (upward triangle) A. lentiginosus 

var. mokiacensis, mokiacensis minor 

variant; (diamond) A. lentiginosus var. 

mokiacensis, Gold Butte minor variant; 

(caret) A. lentiginosus var. mokiacensis, 

trumbullensis minor variant. For a taxonomic 

treatment for the morphological 

variants shown herein for A. lentiginosus 

var. mokiacensis, see Alexander 

(2005, 2008). 

137


Table 1. Specific vouchers identified in the PCoA, cluster analysis, and parsimony figures. 

The Map Code column identifies vouchers on the distribution maps and PCoA figures. Label data for vouchers of A. 

lentiginosus var. maricopae and A. lentiginosus var. ursinus can be found in the Taxonomic Treatment and the data 

for all other varieties can be found in Appendix 1. Vouchers collected by Alexander with letters in brackets (i.e. Alexander 

2367 [A]) are specific individuals on vouchers measured for the morphological analyses. Type: H=Holotype, 

L=Lectotype, I=Isotype, IL=Isolectotype, P=Paratype, v=type vicinity 

Taxon Type Map 

Code 

Herbarium 

Voucher 

A. lentiginosus var. ambiguus H, I B RSA, CAS Ripley & Barneby 3403 

A. lentiginosus var. ambiguus 

(intermediate to var. yuccanus) 

D OSC, UNLV Alexander 2325 

A. lentiginosus var. araneosus I A NY, ORE, GH Jones s.n. from June 1880; 

Jones 1807 from June 1880 

A. lentiginosus var. bryantii H N CAS Bryant s.n. 15 Dec. 1939 

A. lentiginosus var. bryantii N 2 GH Holmgren et al. 15609 

A. lentiginosus var. iodanthus v J NY, ORE Jones s.n. from May 1882; 

Jones 3837 from May 1882 

A. lentiginosus var. maricopae v S OSC, UNLV Alexander 1621 [A] 

A. lentiginosus var. maricopae v S 1 OSC, UNLV Alexander 1621 [C] 

A. lentiginosus var. maricopae H S 2 US Harrison 1790 

A. lentiginosus var. mokiacensis 

(putative type locality) 

A. lentiginosus var. mokiacensis 

(published type locality) 

L, IL L GH, NY Palmer 105 

L,IL L 2 GH, NY Palmer 105 

A. lentiginosus var. mokiacensis Q NY, POM Jones 5058 

A. lentiginosus var. palans H,I E POM, GH Eastwood s.n. June 1892 

A. lentiginosus var. palans 

(type of A. amplexus) 

H,I F RM, GH Payson 335 

A. lentiginosus var. palans P G GH Eastwood s.n. May 1892 

A. lentiginosus var. palans H NY Barneby 13104 

A. lentiginosus var. palans H 1 NY Demaree 43807b 

A. lentiginosus var. palans H 2 NY Holmgren & Holmgren 12796 

A. lentiginosus var. palans H 3 NY, POM, 

BRY 

Jones 5218 

A. lentiginosus var. palans H 4 NY Jones 5218a 

A. lentiginosus var. palans H 5 NY Raven 13079 

A. lentiginosus var. palans H 6 NY Ripley & Barneby 8662 

A. lentiginosus var. palans H 7 NY Weber 4735 

A. lentiginosus var. pseudiodanthus v K OSC, UNLV Alexander 1631 

A. lentiginosus var. stramineus H,I T NY, GH Palmer s.n. in 1870 

138

A. lentiginosus var. ursinus 

(published type locality) 


Table 1, continued 

Taxon Type Map 

Code 

Herbarium 

Voucher 

L,IL M GH, NY Palmer s.n. 1877 

A. lentiginosus var. ursinus M 1 OSC, UNLV Alexander 2120 [A] 

A. lentiginosus var. ursinus 

(Mokiak Pass elements mounted with the 

type by Gray) 

M 2 GH Palmer s.n. 1877 

A. lentiginosus var. vitreus H,I U POM, NY Maguire & Blood 4413 

A. lentiginosus var. wilsonii v R OSC, UNLV Alexander 2367 [A] 

A. lentiginosus var. wilsonii v R 2 OSC, UNLV Alexander 2334 [D] 

A. lentiginosus var. wilsonii (type locality) H,I R 3 ND Wilson s.n. from May 1893 

A. lentiginosus var. wilsonii 

(population intermediate with var. bryantii, 

var. mokiacensis, or var. ursinus) 

A. lentiginosus var. wilsonii 

(population putatively intermediate to var. 

palans) 

P CAS Eastwood 5748 

P 2 NY Demaree 43807 

A. lentiginosus var. yuccanus H C POM Jones 3886 

cross section (podsc), pod shape in longitudinal section 

(podsl), pod pedicel orientation (podpo), leaf abaxial 

pubescence (leafab), keel length (keell), calyx teeth 

shape (calyxs), and pod orientation on raceme. The second 

axis of the PCoA explained 11.7% of the total variation. 

The largest correlation to the second axis was 

from leaf adaxial pubescence (leafad). Moderate correlations 

were obtained for pod stipe length (pods), calyx 

teeth shape (calyxs), and wing color (wingc). A scatterplot 

of the first two coordinates of the PCoA is shown in 

Figure 4. 

In PAUP*, an heuristic search of 100 random addition 

sequences with TBR branch swapping was started 

with the data set of 21 morphological characters from 30 

specimens of the Palantia and related members of A. 

lentiginosus. All 21 characters were parsimony informative. 

Sixteen most parsimonious trees of length 110 

were recovered (HI = 0.5636; RI = 0.6416; CI = 0.4364; 

RC= 0.2800). Figures 5-8 are the strict consensus of 

trees of length 110 with major characters state changes 

mapped on the clades. The clades in this analysis have 

low support. A bootstrap analysis of 10,000 replicates 

resulted in only five clades having 70% or higher support 

(pseudiodanthus & iodanthus, clade A, 73%; 

vitreus 4413 to yuccanus 3886, clade B, 73%; yuccanus 

3886 & ambiguus 2325, clade C, 87%; wilsonii 2367A 

& 2334D clade, 75%; maricopae 1621A & 1621C, 

clade E, 95%). Only banner color (Figure 5), pod persistence 

(excluding the reversal to a deciduous pod in A. 

lentiginosus var. wilsonii, Eastwood 5748; Figure 6), 

pod raceme orientation (Figure 7) and degree of pod 

incurve (not shown) had a high consistency with little or 

no character state reversals. 

A dendrogram (Figure 9) of a Euclidean similarity 

matrix obtained from a cluster analysis showed nearly 

the same topology as the tree obtained from the parsimony 

analysis. 

DISCUSSION 

Outgroup selection in this study proved to be problematic. 

Barneby (1964) proposed that a plant similar to 

the small flowered Astragalus lentiginosus var. salinus 

(a taxon with bladdery inflated pods) was the ancestor to 

the members of the A. lentiginosus complex and that 

this complex was closely related to Section Inflati A. 

Gray, a large species complex with unilocular, bladdery 

inflated pods. Nuclear inter-transcribed spacer (ITS) 

DNA sequence data have shown that A. lentiginosus has 

an identical sequence to that of A. purshii Douglas ex 

Hook. (and an almost identical sequence to that of A. 

utahensis (Torr.) Torr. & A. Gray) of Section Argophylli 

A. Gray (a section composed primarily of taxa with 

139


Table 2. List of 24 morphological characters and 

their abbreviations used in the 153 specimen 

Palantia PCoA. Two were constant (C: calyxd, calyxo) 

and not used in the PCoA, parsimony or cluster 

analysis. One (O: calyxs) was not used due to the finding 

that it was variable on the same plant (and within 

most calyces). For a list of character states for the characters 

below, see Appendix 2. Characters were coded as 

multistate continuous variation (R), binary state (B), or 

multistate (M). 

1. Adaxial leaflet pubescence (Leafad) M 

2. Abaxial leaflet pubescence (Leafab) M 

3. Leaf and stem hair length (leafh) R 

4. Leaflet number (leafn) M 

5. Inflorescence length in flower (inflw) R 

6. Calyx tube length (calyxl) R 

7. Calyx pubescence density (calyxd) M, C 

8. Calyx teeth shape (calyxs) M, O 

9. Calyx teeth orientation (calyxo) M, C 

10. Keel length (keell) R 

11. Keel color (keelc) M 

12. Banner color (bannc) M 

13. Inflorescence length in fruit (infr) R 

14. Pod pedicel orientation (podpo) M 

15. Pod length X width ratio (podr) R 

16. Pod deciduous or persistent (poddp) B 

17. Pod shape, longitudinal section (podsl) M 

18. Pod shape, cross section (podsc) M 

19. Pod orientation on raceme (podro) M 

20. Pod orientation, degree of pod incurve (podpi) R 

21. Pod inflation (podin) M 

22. Pod valve texture (podt) M 

23. Pod pubescence (podpu) M 

24. Pod valve color (podvc) M 

scarcely inflated, unilocular, leathery, deciduous pods), 

and not to members of Section Inflati (4-6 base pair divergence; 

Alexander, unpublished data, Wojciechowski 

et al. 1993, 1999). The ITS sequence data suggest that a 

deciduous, unilocular, leathery, scarcely inflated pod is 

the putative ancestral state in this complex. Based, in 

part, on these data, members of Section Argophylli were 

used as outgroups for chloroplast haplotype analyses in 

Knaus (2008). Taxa in other putatively closely related 

sections (Section Monoenses Barneby, Section Cystiella 

Barneby, Section Circumdati (M.E. Jones) Barneby, or 

Section Platytropides Barneby; all of which have taxa 

with inflated pods) have not been fully investigated in 

molecular analyses. The selection of any member of 

Section Argophylli as an outgroup automatically polarizes 

the ancestral state of the group as a unilocular or 

partially bilocular, scarcely inflated, deciduous pod. In 

addition, haplotypes within the Argophylli sampled by 

Knaus (2008) were found to be nearly twenty steps 

more distant from the haplotypes examined in the 

Palantia. Finding that A. lentiginosus var. iodanthus and 

A. lentiginosus var. pseudiodanthus (both of which have 

been universally delimited as species until Alexander 

2009), are not highly genetically or morphologically 

distinct from A. lentiginosus is a problem for outgroup 

selection in this study, primarily with the parsimony 

analysis and the genetic analyses (Alexander & Liston, 

in prep). The terminal taxa in this study are also not recognized 

at the species level, which violates assumptions 

in parsimony analyses. As a result, robust phylogenetic 

conclusions cannot be made with these data. 

Instead, the cladistic analyses were used to investigate 

patterns of character state changes within the 

Palantia. Astragalus lentiginosus var. iodanthus, and A. 

lentiginosus var. pseudiodanthus were selected as outgroups 

based on their relatively greater genetic and morphologic 

distance from the Palantia based on the results 

from Alexander (2008), Knaus (2008), and Alexander & 

Liston (in prep). Even if the inflated varieties related to 

A. lentiginosus var. yuccanus were used as outgroups 

for the Palantia, the morphological trends discussed 

herein would not change (see Figures 5-8) since the major 

clades of the Palantia would still be split in two 

clades. More thorough molecular analyses of species 

potentially closely related to Section Diphysi (with both 

inflated and scarcely inflated pods) are needed before a 

robust molecular and morphological phylogenetic study, 

with a satisfactory outgroup, can be attempted. 

Despite this, conclusions of the taxonomic status of 

Astragalus lentiginosus var. maricopae and A. lentiginosus 

var. ursinus can be made. The results of the PCoA 

analyses and the genetic analyses (Alexander & Liston, 

in prep) indicate that species level delimitations for 

many of the Palantia have much weaker support than 

previously thought by Alexander (2005). Astragalus 

140


Table 3. Results of the 153 specimen Palantia PCoA. This analysis used 21 variable characters. Each is 

listed in the Kendall rank correlations below. 

Eigenvalues 

1 2 

1.36 0.72 

Percent of Total Variance Explained 

22.06 11.75 

Kendall rank correlations and probabilities 

(between PCoA coordinates and morphological characters, significance of p


Figure 3. Scatterplot of the 153 specimens Palantia PCoA using 21 variable, morphological characters. The first and 

second axes represent 33.5% of the variation. Taxa are labeled as: (circle), Astragalus lentiginosus var. palans including 

the Cameron, Arizona specimens; (caret), A. lentiginosus var. bryantii; (upward triangle), A. lentiginosus var. 

maricopae; (X), A. lentiginosus var. ursinus; (+), A. lentiginosus var. wilsonii including South Rim, Arizona specimens. 

Relevant specimens of these and other taxa (diamonds) are labeled with letters. Table one contains the legend 

of letter codes used for specific vouchers shown. 

Table 4. List of 12 morphological characters from 

the 43 specimen Astragalus lentiginosus var. 

mokiacensis PCoA. This analysis used a modified version 

of the data used in Alexander (2005). For a list of 

character states for character below, see Appendix 2. 

Characters were coded as multistate continuous variation 

(R), binary state (B), or multistate (M). 

1. Adaxial leaflet pubescence (leafad) M 

2. Abaxial leaflet pubescence (leafab) M 

3. Calyx tube length (calyxl) R 

4. Calyx teeth shape (calyxs) M 

5. Keel length (keell) R 

6. Wing color (wingc) M 

7. Pod length X width ratio (podr) R 

8. Pod pedicel orientation (podpo) M 

9. Pod shape, longitudinal section (podsl) M 

10. Pod shape, cross section (podsc) M 

11. Pod orientation on raceme (podro) M 

12. Pod stipe length (pods) R 

lentiginosus var. maricopae, A. lentiginosus var. 

mokiacensis, and A. lentiginosus var. ursinus, all with 

persistent pods (an otherwise dependable species-level 

character in Astragalus according to Barneby 1964), 

were confirmed to be closely related to varieties of A. 

lentiginosus with deciduous pods (A. lentiginosus var. 

palans and A. lentiginosus var. wilsonii). A haplotype 

network derived from an analysis of chloroplast microsatellites 

(Alexander & Liston, in prep) shows A. lentiginosus 

var. maricopae is neither highly genetically 

differentiated from A. lentiginosus var. palans, A. lentiginosus 

var. wilsonii, nor A. lentiginosus var. ursinus. 

Astragalus lentiginosus var. ursinus was found to be 

more genetically similar to the long distance disjunct, A. 

lentiginosus var. wilsonii, than to its geographically 

proximal relative, A. lentiginosus var. mokiacensis 

(Alexander 2008, Alexander & Liston, in prep). 

Though Astragalus lentiginosus var. maricopae is 

not highly genetically differentiated from its geographically 

nearest relative, A. lentiginosus var. wilsonii, it is 

distinct morphologically. The PCoA analysis shows that 

the specimens of A. lentiginosus var. maricopae form a 

morphologically distinct group away from A. lentiginosus 

var. mokiacensis, A. lentiginosus var. wilsonii, and 

A. lentiginosus var. ursinus. The distance is not farther 

than A. lentiginosus var. pseudiodanthus is from the lar- 

142


ger cluster of A. lentiginosus var. palans. Also, the inflated 

members of A. lentiginosus sampled (see Figure 

3: A. lentiginosus var. araneosus, A; versus A. lentiginosus 

var. stramineus, T) in this study are also spread an 

equivalent distance apart. The presence of a yellow 

flower and cylindrical pods contributed highly to the A. 

lentiginosus var. maricopae group. Though the flower 

color of A. lentiginosus var. maricopae was reported by 

Barneby (1964) to be ochroleucous in the type description, 

field observations in 2005 and 2006 revealed that 

the flower is yellow to light yellow in color, but not as 

deep a yellow as that found in European Astragalus, 

Thermopsis, or Trifolium. Ochroleucous flowers in Astragalus 

tend to have a cream tint and dry a whitish-tan, 

or tend to be distinctly white, basally, and grade to a 

yellowish tint, apically, especially in age. The flower 

color, the distinctiveness of the pod morphology, and 

the range disjunction could be utilized as support for a 

species-level delimitation for this taxon. However, A. 

lentiginosus var. maricopae is not the only variety in 

this complex with yellowish flowers. Though some individuals 

of the southern California endemic, A. lentiginosus 

var. nigricalycis M.E. Jones, seem to have creamish 

to greenish-white flowers, most have yellow flowers 

that dry to a darker yellow in age. Also, A. lentiginosus 

var. bryantii has pods that are narrower, longer, and 

more tubular than those in A. lentiginosus var. maricopae 

(see the taxonomic treatment below for more specific 

morphological differences). When considering 

both the genetic and morphological data, A. lentiginosus 

var. maricopae is just one of several taxa at the extreme 

edge of the range of variation in A. lentiginosus and one 

of the most morphologically distinct varieties in the 

Palantia. 

In contrast, Astragalus lentiginosus var. ursinus is 

genetically distinct from its nearest relative, A. lentiginosus 

var. mokiacensis (Alexander 2008, Alexander & 

Liston, in prep). The two are, however, much more sim- 

Table 5. Results of 43 specimen Astragalus lentiginosus var. mokiacensis PCoA. This analysis used 12 

variable characters. Each is listed in the Kendall rank correlations below. 

Eigenvalues 

1 2 

1.78 0.43 

Percent of Total Variance Explained 

48.4 11.69 

Kendall rank correlations and probabilities 

(between PCoA coordinates and morphological characters, significance of p


Figure 4. Scatterplot of the 43 specimen, A. lentiginosus var. mokiacensis PCoA using 12 variable, morphological 

characters. The first and second axes represent 60.1 % of the variation. Taxa and variants are labeled as: (circle) 

mokiacensis minor variant; (X) trumbullensis minor variant; (diamond) Gold Butte minor variant; (+) A. lentiginosus 

var. ursinus (See Table 1 for the legend of letter codes used for specific vouchers shown). For a taxonomic treatment 

of the morphological variants shown herein for A. lentiginosus var. mokiacensis, see Alexander (2005, 2008). 

Figure 5. Strict consensus of 16 most 

parsimonious trees of length 110 

showing banner color (bannc) character 

state changes. The letters above 

branches are the only highly supported 

clades in the tree (above 70% bootstrap 

support; see the results section 

for values for each of these lettered 

clades). Table 1 contains the legend of 

letter codes used for specific vouchers 

shown to the right of the taxon labels. 

144




showing pod deciduous or persistent 

(podpd) character state changes. The 

letters above branches are the only 

highly supported clades in the tree 

(above 70% bootstrap support; see the 

results section for values for each of 

these lettered clades). Table one contains 

the legend of letter codes used for 

specific vouchers shown to the right of 

the taxon labels. 



showing pod raceme orientation 

(podro) character state changes. The 

letters above branches are the only 


(above 70% bootstrap support; see 

the results section for values for each 

of these lettered clades). Table one 

contains the legend of letter codes 

used for specific vouchers shown to 

the right of the taxon labels. 

145


Figure 8. Strict consensus of 16 

most parsimonious trees of length 

110 showing pod inflation (podin) 

character state changes. The letters 

above branches are the only 


(above 70% bootstrap support; see 

the results section for values for 

each of these lettered clades). Table 

one contains the legend of letter 

codes used for specific vouchers 

shown to the right of the taxon 

labels. 

Figure 9. Cluster analysis 

dendrogram of the morphological 

data from 30 specimens 

of the Palantia based on 

a Euclidean distance matrix 

calculated by PAST ver. 1.76. 

Table one contains the legend 

of letter codes used for specific 

vouchers shown to the 

right of the taxon labels. 

146


ilar morphologically than A. lentiginosus var. maricopae 

is to A. lentiginosus var. wilsonii. The morphological 

similarity and geographic proximity of A. lentiginosus 

var. mokiacensis and A. lentiginosus var. ursinus (the 

nearest populations are at least 20 km apart in adjacent 

mountain ranges) has been the primary evidence for 

placing them into the same taxon in the latest monograph 

(Welsh 2007). The morphological distinctiveness 

of A. lentiginosus var. ursinus is subtle, but it is, nevertheless, 

present. In the A. lentiginosus var. mokiacensis 

PCoA analysis, specimens of A. lentiginosus var. ursinus 

appear to grade into the specimens of A. lentiginosus 

var. mokiacensis (including the type specimens of A. 

lentiginosus var. trumbullensis). The parsimony and 

cluster analyses also show A. lentiginosus var. ursinus 

in the same region of the tree as A. lentiginosus var. 

mokiacensis. However, A. lentiginosus var. ursinus does 

not share haplotypes with A. lentiginosus var. mokiacensis. 

It shares the most haplotypes with A. lentiginosus 

var. wilsonii (Alexander & Liston, in prep). Based on 

the lack of shared haplotypes with A. lentiginosus var. 

mokiacensis and a trend toward smaller pods and flowers 

(see the taxonomic treatment below for more specific 

morphological differences), A. lentiginosus var. 

ursinus is recognized herein at the varietal level following 

the delimitation proposed by Barneby (1945, 1964). 

Taxonomic Treatment 

The taxonomic revision herein is a first step in a full 

monograph of the Astragalus lentiginosus complex. As 

such, the specimens used in the morphological analysis 

are labeled in a separate voucher list. Where applicable, 

a list of specimens examined by the author but not yet 

included in morphological analyses is also included. 

Species delimitations in the taxonomic revision follow a 

phenetic species concept (Sokal, 1973; Luckow 1995). 

The original goal of this study was to apply a phylogenetic 

species concept, however, the genetic and morphologic 

data obtained could not be analyzed robustly 

using cladistic methodologies. Primarily, population 

level data are largely ignored since the smallest taxonomic 

units of phylogenetic analyses are species (Nixon 

& Wheeler 1990, Cracraft 1983, Luckow 1995). Table 1 

is a list of notable specimens identified in maps in this 

revision. Following the key are complete taxonomic 

treatments for A. lentiginosus var. maricopae and A. 

lentiginosus var. ursinus. Treatments for the other taxa 

in the key can be found in Alexander (2008). 

Key to the Palantia and related varieties of Astragalus lentiginosus 

1. Pod, in longitudinal section, linear, lanceolate, oblong, or elliptic, the shape cylindrical and not inflated or ventricose 

and scarcely inflated dorsally or laterally, the valves stiffly papery to coriaceous, bilocular, semibilocular, 

or sub-unilocular, the septum to 2.5 mm wide and not fused to the funicular flange. 

2. Pods long-persistent, sessile on a minute boss on the receptacle or contracted at the base into an incipient stipe 

0.4 to 0.7 (1.0) mm long. 

3. Banner light yellow, without a central white or striped spot (immaculate); keel slightly darker than the banner, 

drying yellowish brown and immaculate; wings slightly darker than the banner, drying yellowish brown. 

…………………………………………………………………………………..A. lentiginosus var. maricopae 

3. Banner light to dark purple with a white & purple striate central spot; keel light purple & dark purple maculate; 

wings light to dark purple, sometimes with white tips. 

4. Pods straight or slightly incurved, 20-28 (-32) mm long, 4-7.1x longer than wide, the pedicel ascending or 

spreading, straight or curved; leaflets glabrous to moderately pubescent adaxially, at least sparsely 

pubescent abaxially; [=A. mokiacensis, A. lentiginosus var. trumbullensis]……………………………… 

………………………………………………………………………………….A. lentiginosus var. mokiacensis 

4. Pods incurved, 13-20 (-23) mm long, 2.4-4.7x longer than wide; the pedicel ascending or erect, straight 

or curved; leaflets glabrous adaxially, glabrous or sparsely pubescent abaxially; [=A. ursinus] ………. 

………………………………………………………………………………………A. lentiginosus var. ursinus 

2. Pods deciduous (sometimes tardily so) by a cellular abscission layer between the receptacle and gynoecium, 

sessile on a minute boss on the receptacle or contracted at the base into an incipient stipe or gynophore 0.3 to 

0.5 (0.7) mm long. 

5. Pods semi-bilocular to nearly bilocular, the septum 1-2.5 mm wide, the body incurved 120-180°, incurved 

less than 90°, or straight, the valves in cross section cordate, obcordate, or terete, in longitudinal section 

elliptic, linear, or oblong. 

6. Pods falcate to hamate (nearly circular), occasionally lunate, incurved to 120-180°, the pedicel deflexed 

or declined…………………………………………………………………. A. lentiginosus var. palans 

6. Pods lunate to falcate, incurved less than 90° to nearly straight, the pedicel erect, ascending, or spreading. 

147


7. Pods spreading, 6-8.8 times longer than wide, incurved less than 90°, the valves stiff papery, in longitudinal 

section linear or narrowly oblong…………………………………A. lentiginosus var. bryantii 

7. Pods ascending to erect, 4-6 times longer than wide, incurved less than 90° or nearly straight, the 

valves leathery to thick leathery (subligneous), often with prominent reticulate veins, in longitudinal 

section narrowly elliptic or oblong ……………………………………….A. lentiginosus var. wilsonii 

5. Pods semi-bilocular to sub-unilocular, the septum to 1.5 mm wide, the body incurved to 180° (in most individuals 

nearly circular), the valves in cross section oblong, obcordate, or triangular, in longitudinal section 

narrowly elliptic, lanceolate, or linear. 

8. Stems arising from a superficial root-crown; herbage glabrous to strigulose, rarely villosulous with hairs 

to 1.0 mm long; habitat various; [=A. iodanthus] ………………………A. lentiginosus var. iodanthus 

8. Stems arising from a subterranean root-crown; herbage densely villous to villosulous with hairs 0.7-1.2 

mm long; habitat sandy pockets of alluvial fans and stabilized dunes; [=A. pseudiodanthus] ……….. 

………………………………..………………………………….. A. lentiginosus var. pseudiodanthus 

1. Pod, in longitudinal section, ovoid to globose, the shape terete to didymous, bladdery inflated, the inflation dorsiventrally 

and laterally, or bladdery-ventricose and the inflation greater dorsally and laterally than ventrally, 

the valves papery membranous to stiffly papery (occasionally coriaceous), bilocular, the septum over 2 mm 

wide and weakly fused to the 

1. Astragalus lentiginosus var. maricopae 

Astragalus lentiginosus var. maricopae Barneby, Leafl. 

W. Bot. 4:140. 1945. 

TYPE: U.S.A. ARIZONA: MARICOPA CO.: roadside 

near Tempe, 4 May 1926, G.J. Harrison 1790 

(HOLOTYPE: US!). Map: Figure 1. 

Short lived perennial herbs, 3-8 dm tall; stems ascending, 

single or several in clumps from a superficial root 

crown; herbage glabrous to sparsely strigulose with basifixed 

hairs; stipules 3-8 mm long, ovate-, lance- or deltoid-acuminate, 

mostly recurved, partially or fully amplexicaul-decurrent, 

none connate; leaves 6-16 cm long; 

leaflets 15-23 (25), ovate, suborbicular, or obovate, the 

apex obtuse or emarginate, 5-22 mm long; peduncles 

erect, 5-14 cm long; racemes 13-30 (35) flowered, early 

elongating, flowers ascending to spreading, the axis becoming 

(3) 5-12 (20) cm long in fruit; calyx 7-9 mm 

long, white-, black-strigulose or mixed, the campanulate 

or cylindric tube 4.5-5.5 (6.5) mm long, the teeth, subulate 

to lance-acuminate, 1-2.5 mm long; petals light 

lemon yellow, drying ochroleucous to brownish; banner 

14-16.5 mm long; keel (10) 11-13 mm long, immaculate; 

wings 12-15 mm long, whitish with light lemon 

yellow tip; ovary glabrous; ovules 22-26; fruiting pedicels 

persistent, ascending or spreading, straight or 

curved; pod persistent, ascending or spreading, straight 

or incurved less than 90, in longitudinal section linear or 

narrowly oblong, in cross section cordate or terete, (17) 

19-23 x 3-4 mm, 5-6 (6.2)x longer than wide, sessile on 

a minute boss on the receptacle or contracted at the base 

into an incipient stipe to 0.5 mm long, the valves thinly 

fleshy, becoming coriaceous, stramineous, semibiloculate 

to nearly biloculate (but not fused to the funicular 

flange), the septum 1.5-2 mm wide, the beak 

unilocular; dehiscence apical, through the beak while 

still attached to the raceme. 

Habitat. In mixed shrub communities, in sandy, 

gravely washes (sometimes among boulders) derived 

from Precambrian granites and Tertiary volcanic rocks. 

Distribution. In northern Maricopa County, found in 

the foothills and alluvial fans in vicinity of Cave Creek, 

Fish Creek, Scottsdale, and Tempe; to be expected in 

the foothills and alluvial fans from Scottsdale and 

Tempe east to the mountains along both sides of the 

Verde River drainage, southward to its confluence with 

the Salt River, and west to the alluvial fans in the vicinity 

of Saguaro Lake (see Lehto 510 from 1962 below). 

Phenology. Flowering from February - April; fruiting 

from April - June. 

Astragalus lentiginosus var. maricopae has been 

largely overlooked by most botanists since it was first 

described in 1945. Based on similar floral morphology, 

is has been confused with A. lentiginosus var. yuccanus. 

On the valley floor and alluvial fans northeast of Phoenix, 

Scottsdale, and Tempe, this taxon has become very 

rare (and nearly extirpated throughout its historically 

known range) due to extensive suburban housing and 

golf course development. The population sampled for 

molecular analysis in the vicinity of Scottsdale has already 

been developed, since home construction was 

well underway when the samples were collected. This 

variety is the most unique morphologically, and the 

most endangered of all the Palantia. 

Voucher specimens examined for the morphologic 

analysis. USA. ARIZONA: MARICOPA CO.: 

west of intersection of Westland Drive and Pima Rd, 

Scottsdale, February 2005 (fl, fr), Alexander 1621 

[individuals A, C, D, E] (OSC, UNLV); along Horseshoe 

Dam Rd, 0.5 mi below dam, 02 March 1989 (fl), 

C.L. Jones 5, (GH, NY, RSA) 

Voucher specimens examined (to be included in 

future morphological analyses). U.S.A. ARIZONA: 

148


MARICOPA CO.: 10 mi E of Scottsdale rd, 27 mi NE 

of Scottsdale, 24 March 1960 (fl, imm fr), Crosswhite et 

al 496 (NY); Scottsdale rd, between Bell Rd & Carefree, 

27 April 1974 (fl, fr), Engard et al. 203 (NY); 26 

mi NE of Scottsdale along Hwy 87, 28 March 1973 (fl), 

Higgins 6445 (NY); Hwy 87, 2.7 mi SW of Saguaro 

Lake, 14 April 1962 (fl, fr), Lehto 510 (NY); Cave 

Creek, 23 April 1977 (fl, fr), Lehto 21306 (NY); Carefree, 

23 April 1977 (fl, fr), Lehto 21308 (NY). 

2. Astragalus lentiginosus var. ursinus 

Astragalus lentiginosus var. ursinus (A. Gray) Barneby, 

Leafl. West. Bot. 4: 133. 1945. Astragalus ursinus A. 

Gray, Proc. Am. Acad. Arts Sci. 13: 367. 1878. Tium 

ursinum (A. Gray) Rydb., N. Amer. Fl. 24: 398. 1929. 

TYPE: U.S.A. ARIZONA OR UTAH: MOHAVE CO. 

OR WASHINGTON CO.: Beaver Dams [west slope of 

Beaver Dam Mountains, the type locality was erroneously 

published by Gray as Bear Valley, Iron Co., 

Utah], 20-28 Apr 1877, E. Palmer s.n. (LECTOTYPE: 

GH, designated by Alexander, in prep; ISOLECTO- 

TYPE: K). Map: Figure 2. 

Perennial herbs, 2-4 dm tall; stems erect and ascending 

in clumps from a superficial root crown; herbage 

glabrous to sparsely strigulose with basifixed hairs; stipules 

3-8 mm long, triangular- or deltate-acuminate, 

mostly reflexed, partially or fully amplexicauldecurrent, 

none connate; leaves 2-9 cm long; leaflets 11- 

17 (19), suborbicular, obovate, or oblong, the apex obtuse 

or emarginate, 5-11 mm long; peduncles erect, 4-10 

cm long; racemes 7-15 (20) flowered, early elongating, 

flowers ascending, the axis becoming 3-9 (-11) cm long 

in fruit; calyx 4-8 mm long, white-, black-strigulose or 

mixed, the campanulate or cylindric tube 3-6 mm long, 

the teeth deltate, subulate to lance-acuminate, 0.8-3 mm 

long; petals pink purple, drying violet; banner 12-16 

mm long, purple with a white, purple striate spot; keel 

8.5-13 mm long, light to dark purple maculate; wings 

10.5-16 mm long, purple with dark purple tip or purple 

with a white tip; ovary glabrous or sparsely strigulose; 

ovules 22-24; fruiting pedicels persistent, erect or ascending, 

straight or curved; pod long-persistent, erect or 

ascending, in longitudinal section oblong or narrowly 

elliptic, in cross section cordate or terete, straight or incurved 

less than 90°, 10-23 x 4-5 mm, 2.4-4.7x longer 

than wide, sessile on a minute boss on the receptacle or 

contracted at the base into an incipient stipe to 0.7 (1.0) 

mm long, the valves thinly fleshy, becoming coriaceous, 

stramineous to reddish, semi-biloculate to nearly biloculate 

(but not fused to the funicular flange), the septum 2- 

2.5 mm wide, not extended into the beak, the ventral 

suture sometimes prominent, the beak unilocular; dehiscence 

apical, through the beak while still attached to the 

raceme. 

Habitat. In mixed shrub communities with Larrea and 

Yucca brevifolia, in gravely washes and talus slopes 

derived from the Permian Kaibab Formation (limestone), 

Toroweap Formation (limestones and sandstones), 

Hermit Formation (sandstones and siltstones), 

Queantoweap Sandstone, Permian-Pennsylvanian Callville 

Limestone, and Mississippian Redwall Limestone; 

with Penstemon petiolatus and other limestone crevice 

species on limestone cliffs of various Paleozoic limestones, 

especially the Kaibab Formation and Callville 

Limestone. 

Distribution. Washington Co., Utah, in the southern 

end of the Beaver Dam Mountains in the vicinity of 

Bulldog Knolls and Bulldog Canyon, north to Cedar 

Pockets Wash on the slopes of the peak south of Jarvis 

Peak; in adjacent Mohave Co., Arizona, south to the 

mouth of the Virgin River Gorge, and Hedricks Canyon 

in the Virgin Mountains; to be looked for in the vicinity 

of Mokaac Mountain, Wolf Hole Mountain, Quail Canyon 

or Quail Hill on the northern edge of the Shivwits 

Plateau, Mohave Co., Arizona. 

Phenology. Flowering from March - April; fruiting 

from April - May. 

The type collections of Astragalus ursinus are a 

drought depauperate, limestone crevice form of A. lentiginosus 

var. ursinus. The depauperate morphology of 

the types, especially with respect to the small flower 

size, has contributed to a perennial fog of confusion surrounding 

this variety's taxonomic relationships. It is 

only slightly differentiated morphologically from A. 

lentiginosus var. mokiacensis and imperfectly distinguished 

from some A. lentiginosus var. mokiacensis 

populations in habitat preference (with respect to the 

small number of plants per population growing within 

limestone crevices only). Both varieties have populations 

that inhabit limestone talus slopes below cliff 

faces. The genetic analysis (Alexander 2008, Alexander 

& Liston, in prep) shows that A. lentiginosus var. ursinus 

is distinct from A. lentiginosus var. mokiacensis and 

more closely related to A. lentiginosus var. palans and 

A. lentiginosus var. wilsonii. Additionally, Astragalus 

lentiginosus var. ursinus and A. lentiginosus var. wilsonii 

are the only two members of the Palantia with 

erect to ascending fruiting pedicels and erect to ascending, 

incurved to nearly straight pods. 

Of further note, Alexander (2005) cites the lectotypifications 

for Astragalus mokiacensis and A. ursinus as 

being published in Taxon. However, these two lectotypifications 

have not yet been published due to technical 

circumstances beyond the author's control. The citation 

of the lectotype above should not be considered the 

formal lectotypification of Astragalus ursinus. 

Voucher specimens examined for the morphological 

analysis. U.S.A. ARIZONA OR UTAH: MOHAVE 

CO. OR WASHINGTON CO.: Beaver Dams [Beaver 

149


Dam Mountains, mislabeled as Bear Valley and Beaver 

Valley by Gray], 20-28 Apr 1877 (fl, fr), Palmer s.n. 

(GH, K). ARIZONA: MOHAVE CO.: Virgin Narrows 

about 5 mi NE of Littlefield, 2 Jun 1977 (fl, fr), 

Gierisch 3954 (BRY, NY); Hedricks Canyon [Virgin 

Mountains], T40N R15W S23, 2 Apr 1981 (fl, imm fr), 

Gierisch & Morgart 4824 (BRY, NY). UTAH: WASH- 

INGTON CO.: Beaver Dam Mountains, Bull Dog Canyon, 

25 Apr 2003 (fr), Alexander 1388 (OSC, UNLV), 6 

May 2005 (fl,fr), Alexander 2120 (OSC,UNLV), 6 May 

2005 (fl, fr), Alexander 2121 (OSC,UNLV), 6 May 

2005 (fl, fr), Alexander 2127 (OSC,UNLV); Bulldog 

Knolls, 6 May 2006 (fl, fr), Alexander 2132 

(OSC,UNLV), 6 May 2006, Alexander 2134 

[individuals A,B,C] (OSC, UNLV), 6 May 2006 (fl, fr), 

Alexander 2135 (OSC, UNLV); Bull Dog Knolls, S 

slope of S knoll, T43S R18W S28, 1097 m, 21 Apr 

1986 (fl, fr), Baird 2324 (BRY); Bulldog Knolls, north 

knoll, T43S R18W S21, 341 m, 28 Mar 1986 (fl, imm 

fr), Higgins & Barnum 16267 (BRY, NY); Bulldog 

Canyon, Beaver Dam Mountains, W slope, T43S R18W 

S26, 381 m, 15 Apr 1983 (fl, fr), Neese & Welsh 13062 

(BRY, NY); Cedar Pockets wash, along a sequential 

drainage, T43S R17W S20, 1219 m, 5 May 1986 (fr), 

Welsh & Atwood 23742 (BRY). 


Numerous individuals provided advice, funding and 

encouragement during various stages of this project including: 

Wesley E. Niles; Aaron Liston, my thesis advisor; 

Richard Halse of the OSC Herbarium; Kathryn 

Birgy of the UNLV Herbarium; Arnold Tiehm and the 

Nevada Native Plant Society (and the NNPS Small 

Grants Program); Lisa Decesare of the Library of the 

Gray Herbarium, Harvard University; the Moldenke 

Fund for Plant Systematics and the Hardman Fund at 

Oregon State University; and my committee members, 

Mary Santelmann, David Lytle, Rich Cronn, and Paul 

Farber. Additional thanks go to the late R.C. Barneby 

and the Barneby Fund for Research in Legume Systematics, 

which funded a month of research at NY. The 

following herbaria's online databases were consulted for 

this project: CAS, DS, GH, JEPS, K, MO, NY, OSC, 

ORE, RSA, UC, UNLV, US, WILLU. I thank the curators 

and staff of the following herbaria for access to 

their collections or loans of their specimens: BRY, 

CAS, DS, GH, JEPS, K, NY, OSC, ORE, RM, RSA, 

UC, UNLV, US, WILLU. 


Alexander, J.A. 2005. The taxonomic status of Astragalus 

mokiacensis (Fabaceae), a species endemic to the 

Southwestern United States. Brittonia 57:320-333. 

Alexander, J.A. 2008. A taxonomic revision of Astragalus 

mokiacensis and allied taxa within the Astragalus 

lentiginosus complex of Section Diphysi. Ph.D. 

dissertation. Oregon State University, Corvallis, OR. 

Alexander, J.A. 2009. The types of Astragalus Section 

Diphysi (Fabaceae), a complex endemic to Western 

North America, Part I: lectotypifications, epitypifications, 

and new combinations of several taxa. Journal of 

the Botanical Research Institute of Texas 3:211-218. 

Barneby, R.C. 1945. Pugillus Astragalorum - IV: the 

section Diplocystium. Leaflets of Western Botany 4: 65- 

147. 

Barneby, R.C. 1956. Pugillus Astragalorum - XVII: 

four new species and one variety. Leaflets of Western 

Botany 8: 14-23. 

Barneby, R.C. 1964. Atlas of North American Astragalus. 

(2 volumes). Memoirs of the New York Botanical 

Garden. 13:1-1188. 

Barneby, R.C. 1989. Intermountain Flora. Volume 

III Part B. Eds. A. Cronquist, A.H. Holmgren, N.H. 

Holmgren, J.L. Reveal, P.K. Holmgren. The New York 

Botanical Garden. Bronx, New York. 279 pp. 

Cracraft, J. 1983. Species concepts and speciation 

analysis. Current Ornithology 1: 159-187. 

Easdale, T.A., Gurvich, D.E., Sersic, A.N. & Healey, 

J.R. 2007. Tree morphology in seasonally dry montane 

forest in Argentina: Relationships with shade tolerance 

and nutrient shortage. Journal of Vegetation Science 18: 

313-326. 

Gower, J.C. 1971. A General Coefficient of Similarity 

and Some of Its Properties. Biometrics 27:857-871. 

Gray, A. 1849. Plantae Fendlerianae Novi-Mexicanae. 

Memoirs of the American Academy of Arts and 

Sciences. Series 2. 1:1-116. 

Gray, A. 1863. A Revision and Arrangement (mainly 

by the fruit) of the North American species of Astragalus 

and Oxytropis. Proceedings of the American Academy 

of Arts and Sciences. 6:188-237. 

Gray, A. 1865. Characters of some new plants of 

California and Nevada, chiefly from the collections of 

Professor William H. Brewer, botanist of the State Geological 

Survey of California, and of Dr. Charles L. 

Anderson, with revisions of certain genera or groups. 

Proceedings of the American Academy of Arts and Sciences. 

6:519-556. 

Gray, A. 1878. Contributions to the botany of North 

America. Proceedings of the American Academy of 

Arts and Sciences. 13:361-374. 

Hammer, O., Harper, D.A.T., and P.D. Ryan. 2001. 

PAST: Palaeontological Statistics software package for 

education and data analysis. Palaeontological Electronica 

4:9 pp. 

Hooker, W.J. 1831. Flora Borelli-American or the 

botany of the northern parts of British America. 2 

(12):97-160. 

150


Isely, D. 1998. Native and naturalized Leguminosae 

(Fabaceae) of the United States. Brigham Young University. 

Provo, Utah. 1007 pp. 

Jones, M.E. 1898. Contributions to Western Botany 

8. Contributions to Western Botany. 8:1-43. 

Jones, M.E. 1923. Revision of North American species 

of Astragalus. Text distributed Feb. 15, 1923; 

Plates, June 20, 1923. Salt Lake City, Utah. 288 pp. 

Kearney, T.H. and R.H. Peebles. 1960. Arizona 

Flora. University of California Press. Berkeley, California. 

1079 pp. 

Knaus, B.J. 2008. A Fistful of Astragalus: phenotypic 

and genotypic basis of the most taxon rich species 

in the North American Flora. Ph.D. dissertation. Oregon 

State University, Corvallis, OR. 

Luckow, M. 1995. Species Concepts: assumptions, 

methods, and applications. Systematic Botany. 20:589- 

605. 

Munz, P.A. and D.D. Keck. 1959. A California flora. 

University of California Press. Berkeley, California. 

Nixon, K.C., and Q.D. Wheeler. 1990. An amplification 

of the phylogenetic species concept. Cladistics 6: 

211-223. 

Rydberg, P.A. 1929. Astragalinae. North American 

Flora 24(5): 251-462. 

Sokal, R. R. 1973. The species problem reconsidered. 

Systematic Zoology. 22: 360-374. 

Swofford, D. L. 2002. PAUP*: phylogenetic analysis 

using parsimony. ver. 4 beta 10. Sinauer Associates, 

Inc., Sunderland, Massachusetts. 

Watson, S. 1871. Botany. In C. King, Report of the 

geological exploration of the fortieth parallel, Vol. 5. 

Washington: Government Printing Office. 525 pp. 

Welsh, S.L. 1978. Utah flora: Fabaceae (Leguminosae). 

Great Basin Naturalist. 38:225-367 

Welsh, S.L. 1993. Leguminosae (Fabaceae). Pp 379- 

463 in S. L . Welsh, N. D. Atwood, S. Goodrich, and L. 

C. Higgins, eds., A Utah Flora. 2nd ed. Monte L. Bean 

Life Science Museum, Provo, Utah. 

Welsh, S.L. 2003. Leguminosae (Fabaceae). Pp 350- 

427 in S. L . Welsh, N. D. Atwood, S. Goodrich, and L. 

C. Higgins, eds., A Utah Flora. 3rd ed. Monte L. Bean 

Life Science Museum, Provo, Utah. 

Welsh, S.L. 2007. North American species of Astragalus 

(Leguminosae): A taxonomic review. Monte L. 

Bean Life Science Museum. Provo, Utah. 932 pp. 

Welsh, S.L., and N.D. Atwood. 2001. New taxa and 

nomenclatural proposals in miscellaneous families- 

Utah and Arizona. Rhodora. 103: 71-95. 

Wojciechowski, M. F., M. J. Sanderson, B. G. Baldwin, 

& M. J. Donoghue. 1993. Monophyly of aneuploid 

Astragalus (Fabaceae): evidence from nuclear ribosomal 

DNA internal transcribed spacer sequences. American 

Journal of Botany. 80: 711-722. 

Wojciechowski, M.F., M.J. Sanderson, J.-M. Hu. 

1999. Evidence on the monophyly of Astragalus 

(Fabaceae), and its major subgroups based on nuclear 

ribosomal DNA ITS and chloroplast DNA trnL intron 

data. Systematic Botany. 24: 409-437. 

APPENDIX 1 

List of vouchers used in morphometric analyses not 

cited in the Taxonomic Revision 

Note: All vouchers listed herein have been seen by 

the author. The "!" designation is not used herein as a 

result. 

Astragalus lentiginosus var. ambiguus Barneby 

U.S.A. ARIZONA: MOHAVE CO.: Peach Springs, 11 

May 1941, Ripley & Barneby 3403 (RSA). 

Astragalus lentiginosus var. araneosus (Sheld.) 

Barneby U.S.A. UTAH: MILLARD CO.: Frisco, June 

1880, Jones 1807 [June 1880, Jones s.n., interpreted 

herein as an isotype] (NY, ORE, GH). 

Astragalus lentiginosus var. bryantii (Barneby) 

J.A. Alexander U.S.A. ARIZONA: COCONINO CO.: 

head of Phantom Canyon in Grand Canyon, 15 December 

1939, Bryant s.n. (CAS); 10 yds N of Colorado 

River, 11 mi S of Phantom Ranch, directly N of Grand 

Canyon Village, 11 April 1960, Crosswhite 642 (NY); 

at mouth of Hermit Creek, in sand, Grand Canyon of the 

Colorado River, 10 April 1917, Eastwood 5991 (GH); 

"Utah Flat", Grand Canyon N.P., ca. 0.83 mi NW of 

Phantom Ranch and Bright Angel Creek, 09 April 1993, 

Hodgson & Anderson 2085 (NY); Colorado River, Bass 

Rapids, 108 miles below Lees Ferry, 01 May 1971, 

Holmgren, et al. 15502 (NY); Colorado River, Grand 

Canyon near confluence of Clear Creek, 3.5 miles upriver 

from Kaibab Suspension Bridge (near Phantom 

Ranch), 1 mile up Clear Creek Canyon, 08 May 1971, 

Holmgren, et al. 15609 (GH, NY); Colorado River, 

Grand Canyon, Shinumo Creek, 108.5 river mi below 

Lees Ferry, 17 air mi NW of Grand Canyon Village, 10 

May 1971, Holmgren, et al. 15615 (NY); Grand Canyon 

N.P., at confluence of Bright Angel Creek and Colorado 

River, 20 March 1968, Spellenberg 1826. 

Astragalus lentiginosus var. iodanthus (S. Watson) 

J.A. Alexander U.S.A. NEVADA: CARSON CITY 

(ORMSBY) CO.: Empire City [Carson City vicinity], 

19 May 1882, Jones 3837 [May 1882, Jones s.n., interpreted 

herein as a duplicate] (NY, ORE). 

151


Astragalus lentiginosus var. mokiacensis (A. Gray) 

M. E. Jones U.S.A. ARIZONA: MOHAVE CO.: Hidden 

Canyon, ca. 0.5 mi W of corral, 36.5366°N 

113.7388°W, 1159 m, 25 May 2002 (fr), Alexander 

1304 (OSC, UVSC), 02 May 2003 (fl, fr), Alexander 

1398 (NY, OSC, UNLV, UVSC); Hidden Canyon wash, 

36.5291°N 113.7283°W, 1189 m, 02 May 2003 (fl, fr), 

Alexander 1500 (OSC, UNLV); Whitmore Canyon, below 

Kinney Point, 02 May 2003 (fl, fr), Alexander 1502 

(OSC, UNLV, UVSC); on the N slope of Garnet Mountain 

[Iron Mountain], 20 Apr 1997 (fl, fr), Alexander et 

al. 846 (UNLV); 11 mi S of Mt. Trumbull village, 

Parashant (Trail) Canyon, 26 Apr 1974 (fl, fr), Atwood 

6029 (BRY, NY); 30 mi S of Mt. Trumbull village, 

Andrus Canyon, 28 Apr 1974 (fl, imm fr), Atwood 6056 

(NY); 9 mi S of Mt. Trumbull village, head of Parashant 

(Trail) Canyon, 28 Apr 1974 (fl, imm fr), Atwood 6087 

(BRY, NY); Andrus Canyon, 3 mi W of Andrus Point, 

T32N R10W S6, 26 Apr 1999 (fr), Atwood & Furniss 

24293 (BRY, NY, RM, UNLV, US); Andrus Canyon, 3 

mi W of Andrus Point, T32N R10W S6, 26 Apr 1999 

(fl, imm fr), Atwood & Furniss 24300 (BRY, NY); 

Andrus Canyon, 1 mi W of Andrus Point, T32N R10W 

S10, 26 Apr 1999 (fl, fr), Atwood & Furniss 24302 

(BRY, NY); drainage below Andrus Spring, T33N 

R12W S20, 19 Apr 2000 (fl, imm fr), Atwood et al. 

25058 (BRY, NY); 1 mi S of Trail Canyon summit, 

Parashant-Andrus rd, T33N R10W S2, 19 Apr 2000 (fl, 

fr), Atwood et al. 25095 (BRY, NY); 2 mi S of Mt. 

Trumbull school house, near Griffiths Knoll, T34N 

R10W S1-S2, 1600 m, 21 Apr 2000 (fl, fr), Higgins et 

al. 20277 (BRY, NY); Bar Ten Ranch [Hells Hollow], 

T33N R9W S14, 21 Apr 2000 (fl, fr), Higgins et al. 

21171 (BRY, NY, OSC); 12.5 km (7.8 miles) south of 

Mt. Trumbull, Whitmore Canyon, T34N R9W S29, 

1600 m, 25 May 1979 (fl, fr), Holmgren et al. 9172 

(BRY, NY); Shivwits Plateau, wash 0.5 mi W of Cupe 

Spring [Cupe Seep], Grassy Point, Lake Mead National 

Recreation Area, 21 May 1977 (fl, fr), Leary 1646 

(UNLV); Grand Canyon of the Colorado River near 

Peach Springs [erroneously labeled as “Fort Mohave”], 

May 1884 [Apr 27] (fr), Lemmon 3116 (GH); Peach 

Spring, on hills, Grand Canyon of the Colorado River, 

Jun 1884 (fr), Lemmon 3326 (GH, UC); Mokiak Pass, 

[Palmer's “Juniper Mountains” in the vicinity of Grand 

Wash], 28 Apr-2 May or 2-4 Jun 1877 (fl, fr), Palmer 

105 (GH, K, NY, MO, POM, US); top of Grand Wash 

Cliffs above Vulture Canyon [Andrus Canyon vicinity], 

at lowermost end of Grand Canyon, T32N R10W S6, 

1280 m, 19 Mar 1977 (fl, imm fr), Phillips III 77-1 

(NY); [Whitmore Canyon vicinity], T34N R9W S18, 

1585 m, 2 Jun 1978 (fl, fr), Smith & Gierisch 1091 

(BRY, NY, 2 sheets); Cottonwood Wash, sandy wash 

bottom, T37N R15W S28, 1539 m, 20 May 1987 (fl, fr), 

Thorne & Atwood 5256 BRY); NEVADA: CLARK 

CO.: Gold Butte, NW edge, Granite Spring vicinity, 

36.2847°N 114.1920°W, 10 Apr 2001 (fl, fr), Alexander 

1147 (OSC, UNLV); Gold Butte, southwest foothills, 

36.2712°N 114.2137° (W, 10 Apr 2001 (fl, fr), Alexander 

1148 (OSC, UNLV); Quail Springs Wash, Gold 

Butte area, 36.2585°N 114.2028°W, 1220 m, 3 May 

2003 (fl, fr), Alexander 1503 (OSC); Grapevine Spring, 

Gold Butte area, 36.23990°N 114.1742°W, 4200 ft 

(1280 m), 3 May 2003 (fl, fr), Alexander 1505 (OSC, 

UNLV); Twin Springs Wash, 36.1819°N 114.2222°W, 

976 m, 17 May 2003 (fl, fr), Alexander 1510 (OSC, 

UNLV); Black Mountains, northwest slope, on a ridge E 

of Pinto Valley, Lake Mead National Recreation Area, 

Z11 723894 m E 4012957 m N, 900 m, 13 Apr 1997 (fl, 

fr), Alexander & Birgy 795 (UNLV); E of Gold Butte, 

ca. 1.6 mi S of Summit Pass, in gravely wash, 31 May 

2001 (fr), Alexander & Carlson 1160 (OSC); Mica 

Spring [Gold Butte area], 1219 m, 13 Apr 1894 (fl, fr), 

Jones 5058 (BRY, NY, UC, US); 3.1 mi S of Gold 

Butte in Cataract Wash, 20 May 1977 (fl, imm fr), 

Leary & Niles 1910 (NY, UNLV); Grapevine Spring 

Creek, near Jumbo Peak, near reservoir, Gold Butte 

area, T19S R70E S3, 1280 m, 26 May 1977 (fl, fr), 

Leary & Niles 1947 (UNLV); Quail Spring Wash, about 

1.3 mi SW of Gold Butte, 25 Apr 1997 (fl, fr), Niles et 

al. 4806 (UNLV); Garden Spring area, 3 mi NE of Gold 

Butte, T19S R70E S11, 1150 m, 19 Apr 1997 (fl, fr), 

Niles et al. 4932 (NY, UNLV); Garnet Valley, north 

base of Bonelli Peak, T20S R69E S25, 9 May 1997 (fl, 

fr), Niles et al. 4988 (UNLV); New Spring Wash, ca 1 

rd mi S of Summit Pass, 1128 m, 15 Apr 1986 (fr), Pinzl 

7032 (NY, UNLV). 

Astragalus lentiginosus var. palans (M.E. Jones) 

M.E. Jones U.S.A. ARIZONA: COCONINO CO.: ca 

0.5 mi S of Page, 01 May 2000, Atwood & Welsh 

25360 (NY, RM); E abutment of Glen Canyon Dam, S 

of hwy, 5 May 1998, Atwood & Welsh 26924 (NY); 

Glen Canyon damsite, 7 June 1961, Barneby 13114 

(CAS, NY, RSA); ca 2 rd mi S of Page, along AZ 89, 11 

May 1991, Christy 493 (NY); Navajo Power Plant pipeline, 

4 mi SE of Page, 15 April 1972, Davey s.n. (NY); 

base of Leche-E Rock, 15 April 1973, Hevly & Jenness 

s.n. (NY); Lake Powell, NW of Page, ca 0.5 mi NE of 

Glen Canyon Dam, 3 May 1996, Hufford 1130 (NY); 

Bitter Springs on rd to Page, 19 May 1973, LeDoux et 

al. 753 (NY), LeDoux et al. 781 (NY); US Hwy 89, ca 8 

mi SW of Page, 22 May 1973, Spellenberg et al. 3228 

(NY); NE part of Antelope Island, ca 5 mi NNE of 

Page, 14 April 1987, Tuhy & Holland 2927 (NY); NA- 

VAJO CO.: 12 mi N of Kayenta, 6 June 1961, Barneby 

13104 (CAS, NY, RSA); Mystery Valley, in region of 

Monument Valley, 16 April 1963, McClintock s.n. 

(CAS), 11 April 1963, McClintock s.n. (CAS); 4 mi W 

of Kayenta, 30 April 1981, Welsh 20381 (NY, RSA); 

152


COLORADO: DELTA CO.: Dominguez Creek, W of 

the Gunnison River below Bridgeport, 0-1.5 mi up 

Dominguez Canyon, 19 May 1982, Atwood & Thompson 

8773 (NY); MESA CO.: NW end of Sinbad Valley 

at head of Salt Creek Canyon, 26 May 1983, Atwood 

9262 (CAS, NY); 2 mi E of Bedrock, Paradox Valley, 

23 May 1984, Atwood et al. 9728 (NY); 6 mi SSW of 

Grand Junction, Rough Canyon, base of sandstone wall, 

18 May 1988, Dorn 4886 (NY, RM); open slope 10 mi 

S of Gateway, 11 June 1949, Harrington 4423 (RM); 

Grand Junction, Colorado Monument Park, 03 June 

1921, Osterhout 6142 (RM, RSA); Colorado N.M., west 

entrance, white hills, 17 May 1982, Siplivinsky 3273 

(RM); Colorado N.M., head of Ute Canyon, within park 

boundary, 25 May 1982, Siplivinsky 3379 (RM); near 

hq. Colorado N.M., 6 mi S of Fruita, mesa summit, 21 

May 1948, Weber 3840 (CAS, DUD, RM, US); MON- 

TROSE CO.: La Sal Creek, 4 mi S of Paradox, near cliff 

dwellers mine, 20 May 1982, Atwood & Thompson 

8798 (NY, RM); Monogram Mesa, 6 mi W of Vancorum, 

3 June 1961, Barneby 13046 (CAS, NY, RSA); 

Paradox Valley, N slope, 25 April 1986, Franklin 2838 

(NY, RM); ca 18 air mi NW of Naturita, ca 2.5 air mi S 

of Bedrock, E side Dolores River Canyon, E facing 

slope on W side canyon, 17 June 1995, Moore 5502 

(RM); Slick Rock Canyon, Dolores River, ca 18 air mi 

W of Naturita, at mouth and surrounding area of Leach 

Creek, SW facing wash, 05 July 1995, Moore 6334 

(RM); Slick Rock Canyon, Dolores River, ca 18 air mi 

W of Naturita, at mouth and surrounding area of Leach 

Creek, SW facing wash, 05 July 1995, Moore 6335 

(RM); near Naturita, SW Colorado, dry hills, 22 May 

1914, Payson 335 (RM); SAN MIGUEL CO.: Colorado 

side of Island Mesa, 27 May 1998, Atwood & Trotter 

23617 (NY); roadside, rocky cedar breaks,1 mi S of 

Gladel, 09 June 1951, Penland & Hartwell 4178 (RM); 

W end Gypsum Valley, rocky mouth of Hamm Canyon, 

09 June 1949, Weber 4735 (CAS, DUD, RM, RSA, UC, 

US); UTAH: CARBON CO.: ca 1.5 mi N of Emery Co. 

line along US Hwy 50-6, 29 April 1965, Welsh 3875 

(NY); EMERY CO.: San Rafael Swell, Chimney Rock 

flats, 17 May 1979, Harris 116 (RM); near junction of 

Lower Black Box Rd to Swazys Leap and Sulphur 

Spring, 19 May 1992, Heil & Hyder 7132 (RSA); San 

Rafael Swell, near Temple Mtn, 30 April 1968, Higgins 

& Reveal 1277 (NY, RM); San Rafael Swell, 18 May 

1915, Jones s.n. (UC), 12 May 1914, Jones s.n. (POM); 

Red Plateau, SW of Woodside, along rd to Castledale, 

6.9 mi W of US 50, 5 June 1958, Raven 13079 (CAS, 

GH, NY); Red Plateau, SE of Woodside, 13 June 1947, 

Ripley & Barneby 8675 (RSA); rocky draw, San Rafael 

Swell, 22 mi W of Green River on I-70, 1mi E of rest 

area just below Rattlesnake Bench, 28 April 1979, 

Shultz et al. 3113 (NY, RM, RSA); Summerville 

[Wash], at Woodside, 27 April 1977, Welsh & Taylor 

14624 (NY); GARFIELD CO.: upper E end of Choprock 

Bench, ca 30 mi ESE of Escalante, 6 May 1987, 

Tuhy & Holland 3127 (NY, RSA); head of North Wash, 

28 April 1981, Welsh 20368 (NY, RSA); GRAND CO.: 

Kane Springs Rd, 13 mi S of Moab, [no date], Atwood 

7478 (NY); Sandflat Rd, E of Moab 12 mi., ca 0.25 mi 

W of Forest Service boundary, 20 May 1982, Atwood & 

Thompson 8787 (NY,RM); Arches N.M., 5 mi NW of 

Moab, 21 May 1984, Atwood et al. 9695 (NY); 11 mi N 

of Moab, 18 May 1955, Barneby 12753 (CAS, NY, 

RSA); 1 mi N of Moab, 14 April 1940, Beath 48 (RM); 

Arches road, 04 May 1947, Beath s.n. (RM); near 

Moab, 17 May 1940, Beath & Goodding 6-373 (RM); 8 

mi NW of Moab, along UT Hwy 160, 28 April 1961, 

Bright 135 (NY); between Moab and bridge, 13 May 

1933, Cottam 5623 (RM); hillside in the Colorado River 

Canyon, a little below Salt Wash, NE of Moab, 9 May 

1961, Cronquist 8974 (GH, NY, RSA); slope E above 

Hwy 128, 0.15 mi N of Pole Canyon, 05 May 1985, 

Franklin 1391 (NY, RM); Red Hills SE above White 

Ranch along Colorado River, 07 May 1985, Franklin 

1421 (NY, RM); 3.7 mi due E of Moab, sand flats 

among sandstone fins, 17 April 1986, Franklin 2679 

(RM); Mat Martin Point, sand pockets on slickrock, 12 

May 1986, Franklin 2933 (RM); Sevenmile Mesa, point 

W of confluence of Dolores River and Fisher Cr., ca 25 

mi due NE of Moab, 18 May 1986, Franklin 3104 

(RM); Onion Creek, 22 May 1984, Goodrich et al. 

20398 (NY); N end of La Sal Mountains, Fisher Valley, 

22 May 1984, Goodrich et al. 20414 (NY); 1 mi NE of 

Moab, 0.5 mi along UT Hwy 128, along Colorado 

River, 30 April 1961, Hanson 148 (NY); on edge of 

Colorado River, Moab, clay buttes, 09 May 1933, Harrison 

5945 (RM); cliffs above headquarters, S edge of 

Arches N.M., sandy terraces among rocks, 25 April 

1947, Harrison 11123 (RSA, US), Harrison s.n. (UC); 

south of Courthouse Towers, Arches N.M., sandy flat, 

17 May 1950, Harrison 11383 (US, UC); The 'Neck" 13 

miles due WSW of Moab, sand rock ridge, 11 June 

1941, Harrison et al. 10286 (US); Arches N.M., 29 June 

1948, Howell 24752 (CAS,RSA); near Skyline Arch, 

Arches N.M., 7 September 1968, Howell & True 44906 

(CAS,NY); 10 mi N of Moab, sandy plain, 17 June 

1955, Isely 6460 (RSA, US); Castle Valley, May 1931, 

Jones s.n. (POM); red sandstone banks just north of 

Kane Springs, ca. 15 miles south of Moab, 21 April 

1966, Ledingham 4694 (UC); 5 mi N of Moab, 08 June 

1939, Porter 1796 (RM), Porter 1797 (RM); San Sige 

[illegible] Hollow, Grand River Canon [Grand River 

Canyon near Moab, June 1899], 1899, Purpus s.n. (UC); 

sandy slopes near headquarters, Arches N.M., near 

Moab, 6 June 1962, Rever & Belcher 73 (NY); W of 

Thompson, 13 June 1947, Ripley & Barneby 8662 

(RSA); Spanish Valley, 4.5 mi E of Moab, along roadside 

in town, 03 May 2000, Spencer 1487 (NY, RM); 

153


Ida Gulch, 2.5 mi SE of Hwy 128 along Colorado River, 

E of Castle Valley Rd, slopes below Priests and Nuns, 

11 May 1986, Thorne et al. 4614 (RM); Sego, 2 mi N of 

Thompson, 30 April 1965, Welsh 3881 (NY); Castle 

Valley, ca 4.5 mi E of junction with UT Hwy 128, 3 

May 1968, Welsh 7018 (NY); Castle Valley, ca 4.5 mi E 

of junction with UT Hwy 128, 29 May 1968, Welsh & 

Moore 7170 (NY); KANE CO.: 5 mi S of Glen Canyon 

N.R.A. boundary, along Hole-in-the-Rock Rd., 23 April 

1996, Atwood & Furniss 20795 (NY); Willow Tank, 41 

mi SE of Escalante, 8 June 1961, Barneby 13127 (CAS, 

NY, RSA); abandoned "Pareah" townsite near the Paria 

River, 3 May 1986, Shultz & Shultz 9931 (NY); Willow 

Tank, ca 17 mi S of Garfield Co. line, along road to 

Hole-in-the-Rock, 04 May 1962, Welsh 1687 (NY, US); 

Lake Powell, Driftwood Canyon, Driftwood Garden, 27 

May 1986, Welsh 22102 (NY); Paria townsite, 5 May 

1970, Welsh & Atwood 9727 (NY); SAN JUAN CO.: 

south of Mexican Hat along Hwy 63, 14 May 1970, Atwood 

2486 (NY); Navajo Twins, N side of Bluff, 20 

May 1985, Atwood & Furniss 11009 (NY); E of Monticello, 

28 April 1998, Atwood & Trotter 23411 (NY); ca 

15 mi W of Mexican Hat, along bench above San Juan 

River, 28 April 1998, Atwood & Trotter 23423 (NY); E 

slope of Navajo Mountain, 23 June 1973, Atwood & 

Trotter 5334 (NY); Copper Canyon, S of Monitor Mesa, 

22 May 1985, Atwood et al. 11057 (NY); 14 mi W of 

Blanding, 19 May 1955, Barneby 12785 (CAS, NY, 

RSA); 10 mi W of Blanding on US Hwy 95, 30 April 

1961, Bright 83 (NY); Goose Necks, 01 May 1935, Cottam 

5837 (RM); between Blanding and Bluff, hillsides, 

17 April 1928, Cottam 6693 (RM, 2 sheets), Cottam 

6703 (RM); Goosenecks of San Juan River, 12 April 

1938, Cronquist 1104 (RM); Montezuma Canyon, 29 

May 1892, Eastwood s.n. (GH); Montezuma Canon, 01 

June 1892, Eastwood s.n. (RM, POM, UC); 25 mi W of 

Hanksville, 14 May 1940, Goodding & Beath 48 (RM); 

8 mi E of Halls Crossing, 23 May 1983, Higgins & 

Welsh 13216 (NY); ca 5 mi E of Mexican Hat, 24 May 

1983, Higgins & Welsh 13291 (NY); slopes above 

Gretchen Bar, 54 mi below Hite on the Colorado River, 

04 May 1954, Holmgren & Goddard 9980 (CAS); black 

streak cliffs, vicinity of Bluff, 24 June 1944, Holmgren 

& Hansen 3439 (NY); Mendenhall Loop, riverside below 

the Mendenhall Stone Cabin, ca mi 31, 2.5 km W of 

Mexican Hat, 11 June 1997, Holmgren & Holmgren 

12796 (NY); Bluff, 24 May 1919, Jones s.n. (POM); 14 

mi due SW of La Sal summit of Rone Bailey Mesa, 4 

June 1985, Neese & Welsh 16993 (NY); Bluff, south 

exposure, 03 April 1966, Pederson 12 (GH); W of 

Bluff, 22 May 1943, Ripley & Barneby 5603 (RSA); 

along San Juan River, near Bluffs, 25-29 August 1911, 

Rydberg & Garrett 9914 (NY, RM, US); 12.2 mi S of 

Moab, 22 May 1976, Shultz 2151 (RSA), Shultz & 

Bolander 2151 (NY); Bluff area, 1 mi E of Needles 

Overlook, along roadsides in sand, 22 May 1976, Shultz 

& Shultz 1854 (NY,US); Rone Bailey Mesa ca 22 mi 

NNW of Monticello, 01 July 1984, Tuhy 1581 (RM); 

head of Red Canyon, ca 12 mi SW of Natural Bridges 

N.M., 29 April 1981, Welsh 20374 (RSA); Bluff, 29 

April 1961, Welsh 1501 (NY); Comb Reef, 17 mi W of 

Blanding, 1 mi from Perkins Ranch, 30 April 1961, 

Welsh 1510 (NY); head of Red Canyon, ca 12 mi SW of 

Natural Bridges N.M., 29 April 1981, Welsh 20374 

(NY); Whirlwind Draw, Clay Hills Divide, 30 April 

1966, Welsh 5206 (NY); W of Grand View Point, top of 

Murphy's Ridge, 18 May 1968, Welsh 7076 (NY); S Six 

Shooter Peak platform, Davis Canyon, 11 May 1982, 

Welsh et al. 21100 (NY); between Natural Bridges N.M. 

and Blanding along UT Hwy 95, 22 April 1967, Wetherell 

& Finzel 633 (NY); WASHINGTON CO.: 2 mi W 

of Springdale, 15 April 1938, Cronquist & Gaupin 1106 

(BRY); Hurricane, 4 mi due SW of Rockville, 22 May 

1977, Foster 3766 (NY); Coalpits Wash, 5 May 1988, 

Franklin & Thorne 5930 (BRY); 9 mi E of Hurricane 

along Hwy 59, 20 May 1971, Higgins & Welsh 4227 

(NY); sandy desert, 5 miles east of Virgin, 11 May 

1937, Hitchcock 3025 (GH); Virgin, 14 May 1894, 

Jones 5215e (US); Rockville, 14 May 1894, Jones 

5215e (DUD, 3 sheets, NY, UC); Rockville, 15 May 

1894, Jones 5218 (DUD, GH, NY, RM), 14 May 1894, 

Jones 5218 (RM, US, UC); Virgen City, 14 May 1894, 

Jones 5218a (US); Rockville, 15 May 1894, Jones 

5218a (DUD, POM); Zion N.P, Petrified Forest, 13 

April 1972, R.A. Nelson 9930 (RM); Zion N.P, Petrified 

Forest, 13 April 1972, R.A. Nelson 9937 (RM); 2 mi N 

of Hurricane, 10 April 1966, Stevens 140 (BRY); Petrified 

Forest sector, 2 mi N of Rockville, 06 May 1988, 

Welsh & Clark 23986 (BRY, RM); WAYNE CO.: Capitol 

Reef entrance to Grand Wash, 24 May 1975, Harrison 

1688 (RM); Horseshoe Canyon, ca 1 mi up Horseshoe 

Canyon, Canyonlands N.P., 20 May 1990, Heil et 

al. 5895 (RSA); Capitol Reef entrance to Grand Wash, 

24 May 1975, K. Harrison 1688 (NY); canyon bed of 

Fremont Canyon, N of Fruita, 05 May 1940, Maguire 

18115 (RM); 23.8 mi W of Loa, 3 May 1977, Neese & 

White 2752 (NY); Barrier (Horseshoe) Canyon, 19 April 

1970, Welsh 9593 (NY); road to North Point ca 1 mi NE 

of French Spring above Orange Cliffs, 30 May 1970, 

Welsh & Atwood 9870 (NY). 

The following voucher specimens examined for the 

morphological analysis from the vicinity of Cameron 

are putative populations of A. lentiginosus var. palans 

intermediate to A. lentiginosus var. wilsonii. 

U.S.A. ARIZONA: COCONINO CO.: roadside, 6 mi W 

of Cameron, 08 April 1938, Cronquist & Gaupin 1105 

(RM); Painted Desert, Cameron, low areas and roadsides, 

22 April 1961, Demaree 43807 (DUD, NY, US); 

ca 5 mi W of Cameron, 8 April 1978, Gierisch 4183 

154


(NY); 4 mi W of Cameron on Hwy 64, 18 March 1968, 

Hitchcock 25614 (NY, OSC); Rainbow Lodge, south 

end of Navajo Mountain, 11 June 1938, Peebles & 

Smith 13939 (GH, NY, US); 10 mi SW of Cameron, 

Hwy 64, 7 April 1957, Strickland 21 (NY). 

Astragalus lentiginosus var. pseudiodanthus 

(Barneby) J.A. Alexander U.S.A. NEVADA: NYE 

CO.: sand dunes N of Tonopah, May 2003 (fl,fr), Alexander 

1631 (OSC, UNLV). 

Astragalus lentiginosus var. stramineus (Rydb.) 

Barneby U.S.A. ARIZONA: MOHAVE CO.: 

"southeastern Utah" [southwestern slope Beaver Dam 

Mountains, near Littlefield], June 1870, Palmer s.n. 

(NY, US). 

Astragalus lentiginosus var. vitreus Barneby 

U.S.A. UTAH: WASHINGTON CO.: 5 mi W of Leeds, 

19 May 1933, Maguire & Blood 4413. (POM, CAS). 

Astragalus lentiginosus var. wilsonii (Greene) 

Barneby U.S.A. ARIZONA: YAVAPAI CO.: sandy 

washes north of Cottonwood, 1 June 2005 (fl, fr), Alexander 

2334 [individuals B, E, J, K] (NY, OSC, UNLV, 

UVSC); south of the boundary of Montezuma Castle 

National Monument, 2 June 2005 (fr), Alexander 2339 

[individual F] (OSC, UNLV), 2 June 2005 (fl, fr), Alexander 

2340 [individuals A, D]; in scrubland 1.3 mi west 

of I-17, near junction of FR 9204F and AZ Hwy 179 2 

June 2005 (fr), Alexander 2367 [individuals A, B] (NY, 

OSC, UNLV, UVSC); on roadcut 0.1 mi west of I-17, 

along AZ Hwy 179, 02 June 2005 (fr), Alexander 2368 

(OSC, UNLV). 

The following vouchers specimens examined for the 

morphological analysis are putative populations of A. 

lentiginosus var. wilsonii along the South Rim of the 

Grand Canyon that are intermediate to A. lentiginosus 

var. bryantii, A. lentiginosus var. mokiacensis, or A. lentiginosus 

var. ursinus. U.S.A. COCONINO CO.: Kaibab 

Trail, 10 May 1940 (fl, fr), Collom KT24 (US); 

Grand View Trail, Grand Canyon of the Colorado 

River, 16 June 1916 (fl, fr), Eastwood 5748 (CAS, GH, 

UC); Grand Canyon [Bright Angel Trail vicinity], June 

1915 (fr), Macbride & Payson 945 (GH, RM); Kaibab 

Trail, Grand Canyon N.P., 23 May 1938 (fl, fr), Nelson 

& Nelson 2799 (RM); near El Tovar, South Rim Grand 

Canyon, 3 June 1947 (fr), Ripley & Barneby 8472 

(RSA). 

Astragalus lentiginosus var. yuccanus M.E. Jones 

U.S.A. ARIZONA: MOHAVE CO. : Yucca, 13 May 

1884, Jones 3886 (POM, GH, NY, US). 

Appendix II 

List of morphological characters and character 

states used in the morphometric analyses 

Adaxial leaflet pubescence (leafad): 1. densely 

pubescent (surface obscured, overlap); 2. moderately 

pubescent (some hair overlap, gap less than 0.2 mm); 3. 

sparsely pubescent to subglabrate (uneven to evenly 

across surface with no overlap, or just confined to midrib 

& base); 4. entirely glabrous (or just a few hairs at 

base) 

Abaxial (lower) leaflet pubescence (leafab): 1. 

densely pubescent (surface obscured, overlap); 2. moderately 

pubescent (some hair overlap, gap less than 0.2 

mm); 3. sparsely pubescent to subglabrate (uneven to 

evenly across surface with no overlap, or just confined 

to midrib & base); 4. entirely glabrous (or just a few 

hairs at base) 

Leaf and stem hair length (leafh): average of 3, in 

mm. 

Leaflet number (leafn): average of 3, in mm. 

Inflorescence length in flower (inflw): average of 

3, in mm. 

Calyx tube length (calyxl): average of 3, in mm. 

Calyx pubescence density (calyxd): 1. densely 

strigulose (surface obscured, overlap); 2. moderately 

strigulose (some hair overlap); 3. evenly & sparsely 

strigulose; 4. entirely glabrous (or just a few scattered 

hairs). 

Calyx teeth shape (calyxs): 1. deltoid; 2. lancesubulate, 

subulate; 3. subulate-setaceous; 4. lanceattentuate; 

5. lance-acuminate. 

Calyx teeth orientation (calyxo): 1. erect to spreading 

; 2. loosely long recurving. 

Keel length (keell): average of 3, in mm. 

Keel color (keelc): 1. light to dark purple maculate; 

2. tan to pink maculate tipped; 3. ochroleucous, not 

maculate; 4. yellow. 

Wing color (wingc): 1. purple, dark purple tipped; 2. 

purple, white tipped; 3. white, pink-purple tipped; 4. 

ochroleucous; 5. yellow. 

Banner color (bannc): 1. purple & white with purple 

striate central spot; 2. whitish to ochroleucous & 

pink tinged; 3. tan ochroleucous to whitish; 4. yellow. 

Inflorescence length in fruit (infr): average of 3, in 

mm. 

Pod pedicel orientation (podpo): 1. ascending to 

spreading, straight; 2. ascending to spreading, arched; 3. 

appressed to erect, straight; 4. recurved, arched 

Pod length & width ratio (podr): average of 3, in 

mm. 

Pod deciduous or persistent (poddp): 1. deciduous; 

2. persistent. 

155


Pod shape, longitudinal section (podsl): 1. seculate 

& abruptly acute apically; 2. seculate & attenuate apically; 

3. elliptic, narrowly elliptic, lunate & acute apically; 

4. oblong, narrowly oblong & attenuate apically; 

5. linear; 6. lanceolate, widest at base & attenuate apically; 

7. oval, ovoid, broadly ovoid; 8. obovoid, broadly 

obovoid; 9. subglobose; 10. clavate-oblanceolate. 

Pod shape, cross section (podsc): 1. subterete, dorsiventrally 

compressed, when obcordate, the suture shallowly 

sulcate; 2. obcordate & laterally compressed, 

broadly to narrowly (obcompressed), suture deeply sulcate; 

3. didymous & dorsi-ventrally compressed, 

broadly to narrowly; 4. cordate & dorsi-ventrally compressed, 

broadly to narrowly; 5. cordate & laterally 

compressed, broadly to narrowly. 

Pod orientation on raceme (podro): 1. ascending, 

erect & beak & 1/2 pod curved inward; 2. ascending to 

spreading, straight; 3. spreading, declined & beak to 1/2 

pod recurved; 4. spreading & slightly curved inward 

from middle (lunate, falcate); 5. spreading & beak to 1/2 

pod incurved to 180° (hamate). 

Pod orientation, degree of pod incurve, coded 

range (podpi): 1.


Ecology of Rusby’s Milkvetch (Astragalus rusbyi), 

a Rare Endemic of Northern Arizona Ponderosa Pine Forests 

Judith D. Springer, Michael T. Stoddard, and Daniel C. Laughlin, 

Ecological Restoration Institute, Northern Arizona University, Flagstaff, AZ 

Debra L. Crisp and Barbara G. Phillips, 

Coconino National Forest, Flagstaff, AZ 

Abstract. Rusby’s milkvetch (Astragalus rusbyi Greene) is endemic to basaltic soils northwest and west of Flagstaff, 

Arizona. Recent interest in this species is due in part to its addition to the U.S. Forest Service Region 3 sensitive species 

list in 1999 and its occurrence in ecological restoration projects and proposed fuels reduction projects that involve 

tree thinning and prescribed burning. Some of its habitat has been subjected to large wildfires over the last few 

decades, and other areas have undergone ecological restoration treatments, while much of its range in ponderosa pine 

forest is slated to undergo such treatments in the near future. In a ponderosa pine restoration study area northwest of 

Flagstaff, A. rusbyi was an indicator species of remnant grass patches and increased following tree thinning and prescribed 

burning. However, in an area less than 3 km away, there appeared to be no relationship to restoration treatments, 

trees per ha, pine basal area, or canopy cover, but A. rusbyi did appear to be sensitive to an extreme drought 

event in 2002 and may have remained dormant in that year, a pattern that has been observed in other rare Astragalus 

species. A. rusbyi has a foliar nitrogen content of 4.4% and a foliar C:N mass ratio of 11. It is classified as a competitive 

ruderal species, meaning it is able to compete well with other understory species, but is not very tolerant of 

stresses, such as deep shade. We currently do not have a thorough understanding of the ecology of this species, or the 

effects of ecological restoration or fuels reduction treatments. In this paper we will discuss ecology of other members 

of the genus Astragalus and explore the relationships of A. rusbyi to moisture, vegetation treatments and overstory 

mortality. 

Astragalus is believed to be the largest genus of 

flowering plants in the world, with over 2500 species 

worldwide and over 400 species in North America 

alone, primarily in arid regions of the western U.S. The 

highest diversity in North America is centered in the 

Great Basin and on the Colorado Plateau (Barneby 

1989, Sanderson 1991). Astragalus species are often 

found in marginal habitats or on specialized soil types, 

and their geographic ranges are strongly skewed towards 

narrow endemism (Barneby 1964, 1989; Sanderson 

1991). Along with a limited dispersal ability possessed 

by many members of the genus, there is widespread 

local differentiation and geographic speciation, 

particularly in the areas of the western U.S. where it 

achieves the highest levels of diversity (Sanderson 

1991, Lesica et al. 2006). However, due to restricted 

ranges and habitats, some Astragalus species may exhibit 

low genetic variability and reduced fitness from 

inbreeding depression (Karron et al. 1988, Allphin et al. 

2005, Breinholt et al. 2009). Neoendemism is common 

in the intermountain regions of North America where 

there are large numbers of both widespread, recently 

evolved species, as well as narrowly endemic species, 

which are often associated with extreme edaphic conditions 

and reduced competition from dominant species 

(Lesica et al. 2006). Lesica and his co-authors suggest 

that restricted ranges and high local abundances of 

neoendemic species may be due more to patterns and 

processes of speciation than to ecological tolerance. The 

small ranges exhibited by many Astragalus species in 

the western U.S. may be due to recent speciation and an 

insufficient amount of time for these species to have 

increased their ranges significantly (Lesica et al. 2006). 

Reticulate evolution may not be widespread in the Astragalus 

genus, for many members of the genus appear 

to exhibit allopatry (geographic isolation) along with 

high levels of local endemism and little hybridization 

(Sanderson 1991). 

The type specimen of A. rusbyi was collected by 

Henry Hurd Rusby on July 2, 1883 on Mt. Humphreys, 

near Flagstaff Arizona (Welsh 2007) and was first described 

by Edward Lee Greene in 1884 (Greene 1884). 

A. rusbyi is a slender perennial averaging 15-40 cm in 

height. It also has a fairly deep taproot (D.C. Laughlin, 

personal communication, 2008). It grows primarily in 

meadows in ponderosa pine forests and in aspen groves 

(Barneby 1964, Welsh 2007), but it also may be found 

in moderately dense ponderosa pine forests. Populations 

are mainly concentrated on basaltic soils in two areas in 

northern Arizona: around the San Francisco Peaks 

157


(primarily on the southern and western sides of the 

Peaks) and also the vicinity of Kendrick Mountain 

(Figure 1). Collections from areas to the south, including 

Oak Creek Canyon and Yavapai County are of uncertain 

validity. It is ranked G3 (vulnerable) by Nature- 

Serve (2009) and is on the U.S. Forest Service sensitive 

species list for Region 3 (Southwestern Region). 

A. rusbyi has been placed in the section Strigulosi, 

which contains approximately 35 species found mainly 

in the Mexican highlands north to Arizona and New 

Mexico and generally associated with oak and pine forests 

(Barneby 1964, Spellenberg 1974). Section Strigulosi 

is thought by some authors to be the most primitive 

group of Astragalus species in North America. The majority 

of the evidence, including research on chromosome 

numbers and more recent molecular phylogenetic 

data, currently points to an Old World origin for Astragalus, 

presumably in the steppes and mountains of 

southwestern and south-central Asia and the Himalayan 

Plateau (Spellenberg 1976, Wojciechowski 2005). 

Astragalus rusbyi has a chromosome number of 11 

(2n=22) (Spellenberg 1974). Morphologically, it appears 

to be most closely related to two other members of 

section Strigulosi, A. egglestonii and A. longissimus, 

with which it shares technical features (Barneby 1964). 

Like most members of Strigulosi, A. rusbyi flowers 

from mid-summer onward into the fall, varying in abundance 

in response to the amount and timing of summer 

rains. Astragalus rusbyi is differentiated from these 

species of Astragalus by minor differences in characters 

of its pendulous, stipitate, bilocular, trigonously compressed 

pods (Barneby 1964). However, the ranges of 

these three species do not overlap and they maintain 

geographic isolation from one another with no observed 

intermediate populations. 

ECOLOGY OF ENDEMIC WESTERN SPECIES 

Astragalus is a very large genus with little ecological 

information available for the vast majority of species. 

However, many uncommon and rare species share characteristics 

with each other that may directly affect monitoring 

and conservation planning for A. rusbyi. For example, 

some species exhibit prolonged vegetative dormancy, 

such as A. scaphoides (Bitterroot milkvetch) and 

A. sinuatus (Whited’s milkvetch) of the sagebrush 

steppe of the Pacific Northwest, or dormancy during dry 

years, as demonstrated by A. schmolliae (Schmoll’s 

milkvetch) of Mesa Verde National Park, in southwestern 

Colorado (Anderson 2004, Gamon 1995, Lesica and 

Steele 1994). A. scaphoides plants may utilize prolonged 

vegetative dormancy as a bet-hedging strategy in 

an effort to conserve resources while avoiding the risk 

inherent in funneling resources into aboveground 

growth (Gremer et al. 2012). Lesica (1995) found that 

A. scaphoides plants may remain dormant belowground 

for as long as five years before reappearing. Dormancy 

may be inferred in other species, such as A. ripleyi, due 

to an increased number of visible plants in years of 

above average precipitation (Ladyman 2003). Barneby 

(1964) notes that A. rusbyi, like most members of section 

Strigulosi varies in “vigor and abundance in proportion 

to amount and timing of summer rains,” but prolonged 

vegetative dormancy has not been established. 

Many Astragalus species exhibit large underground 

storage organs (A. scaphoides), a vigorous creeping root 

system (A. cicer – chickpea milkvetch), or long taproots 

(A. ampullarioides – Shivwits milkvetch) (Jennifer Gremer, 

unpublished data; Mark Miller, personal observation; 

Horvath 2002). These extensive rooting systems 

exhibited in the genus may be linked to observed patterns 

of vegetative dormancy. A. ripleyi (Ripley’s milk- 

Figure 1. Known range of Astragalus rusbyi in northern Arizona based on herbarium collections and research studies 

(figure compiled by J.E. Crouse and D. L. Crisp). 

158


vetch), a species of ponderosa pine forests in northcentral 

New Mexico and south-central Colorado, reproduces 

by seed, but plants tend to allocate resources toward 

survival of individual plants, and it is believed to 

build up root stock reserves when aboveground parts are 

consumed (Ladyman 2003). Direct evidence supports an 

ability to lie dormant for two years, but monitoring has 

not yet been used to establish if it can remain dormant 

for a longer period. 

Although a number of Astragalus species contain 

toxins (such as miserotoxin or swainsonine) or accumulate 

selenium, thus making them poisonous to livestock, 

there are also many species that are highly desirable to 

herbivores. Lesica (1995) found fecundity losses in A. 

scaphoides due to livestock and insect herbivory ranging 

from 14-90% at two sites. Observations have confirmed 

that inflorescences are consumed by ants and 

moth larvae. Loss of seeds to weevil predation ranged 

from 0-33%. Sugars such as those found in flower nectar 

may increase palatability. Lesica also found very low 

recruitment, accounting for less than 17% of population 

growth. However, despite heavy losses in reproductive 

output and low recruitment, populations can continue to 

persist and increase in size. He suspects that persistence 

of many populations of long-lived plants may be more 

reliant on growth and survival of established plants than 

on recruitment from seed. Herbivory by cattle and game 

has also been observed in A. terminalis (railhead milkvetch), 

and seed predation in A. ripleyi may be the cause 

of significant seed loss (Heidel and Vanderhorst 1996, 

Ladyman 2003). Apparently, like A. scaphoides, this 

species has low recruitment rates and allocates a significant 

amount of resources toward maintenance of the 

root system. A. ripleyi is also consumed by a number of 

arthropods (aphids, treehoppers, carpenter ants), rodents 

and large mammals, including cattle, elk, deer, sheep 

and goats. A ninety percent reduction in fruit production 

due to herbivory was observed in A. ampullarioides 

(Shivwits milkvetch), which the authors suggest could 

have a significant impact on reproductive output (Miller 

et al. 2007). The toxicity of A. rusbyi is unknown. 

The effects of disturbances, such as tree thinning or 

burning, on Astragalus species vary widely. A. ripleyi is 

thought to be a “fire evader” rather than a stress tolerator 

(Ladyman 2003). Following fire, plants have been 

observed in areas where they have not been detected 

before, presumably emerging from dormant root systems 

underground. However, the stress-tolerator category 

may be appropriate, for a pattern of rapid colonization 

following fire and drought has also been observed 

in this species (Ladyman 2003). 

Thinning activities in pinyon-juniper woodlands at 

Mesa Verde National Park appeared to cause an increase 

in Poa fendleriana (muttongrass) that could result in 

undesirable competition impacts on A. schmolliae 

(Anderson 2004). Grass seeding post-fire also has the 

potential to cause negative impacts on this species. 

Drought is deleterious, but it is likely tolerant of fire 

because of a deep taproot. However, monitoring indicates 

that while fire may confer short-term benefits, it 

may also have long-term detrimental impacts (Anderson 

2004). 

OBSERVATIONS FROM FIELD STUDIES 

A. rusbyi has a very small range in northern Arizona, 

with the bulk of its population limited to a band approximately 

18 x 7 km (11 x 4.5 mi) in size to the west 

and north of the San Francisco Peaks and a few scattered 

populations to the west (Figure 1). Some of its 

habitat has been subjected to large wildfires over the last 

few decades; other areas have undergone ecological restoration 

treatments (tree thinning and prescribed burning); 

and much of its range in ponderosa pine forest is 

slated to undergo such treatments in the near future. Increasing 

tree densities of ponderosa pine, and a cessation 

of frequent fires in ponderosa pine forests since 

Euro-American settlement of this area of the Southwest 

have been well documented (Covington and Moore 

1994, Fulé et al. 1997). 

We currently do not have a thorough understanding 

of the basic ecology of this species. Additionally, we 

have insufficient knowledge of the effects of increased 

tree densities, tree thinning, or fire on the population 

dynamics. However, some limited information is available 

from large landscape scale studies within its range. 

Fisher and Fulé (2004) installed 121 20x50 m permanent 

monitoring plots on the south side of the San Francisco 

Peaks (specifically Agassiz Peak). Plots were established 

in five forest types: ponderosa pine, mixed 

conifer, aspen, spruce/fir and bristlecone pine. Overstory 

measurements and plant community data were 

collected between 2000 and 2003. A. rusbyi was found 

to be an indicator species for ponderosa pine forest, with 

an indicator value of 36.5 (p


Figure 2. Frequency of Astragalus rusbyi by treatment over time at G.A. Pearson Natural Area near Flagstaff, AZ. 

treatments (G.A. Pearson Natural Area), Laughlin and 

others (2008) found A. rusbyi to be an indicator species 

of both thinned treatments and treatments that involved 

thinning plus burning (mean indicator value of 25.0; 

p


D.C. Laughlin (personal communication, 2009) collected 

trait data on 137 ponderosa pine understory species, 

including A. rusbyi, and found that, on average, it 

has a higher specific leaf area (24 mm 2 /mg) and higher 

nitrogen and phosphorous concentrations in its foliar 

tissue (4.4% and 0.18%, respectively) (Figure 4). Because 

of the high nitrogen content, it has a relatively 

high net photosynthetic rate. On average, it also has a 

lower leaf dry matter content (0.19 mg/mg) and foliar 

C:N mass ratio (10.7). Combined with its high photosynthetic 

rate and comparatively tall stature (with an 

average height of 31 cm), it is able to compete well with 

other understory species, but is not very tolerant of 

stresses such as deep shade. The combined trait data 

place it in the category of a competitive ruderal species 

(Hodgson et al. 1999). 

CONCLUSIONS 

As previously mentioned, little is known of the ecology 

of this locally abundant but narrowly endemic species, 

and much of its known range is slated to undergo 

various thinning and prescribed burning activities 

in the very near future. Many Astragalus species are 

long-lived, recruit slowly by seed, and maintain longlived 

seeds in the soil seed bank. Whether A. rusbyi utilizes 

a similar strategy is unknown but could be determined 

from additional research. We currently do not 

have a thorough understanding of the population dynamics 

of this species over time. Rigorous long-term 

demographic monitoring would be valuable in determining 

population baselines and is essential for understanding 

the ecology and conservation and habitat needs of 

this species. Such monitoring can also reveal patterns 

that might be caused by precipitation fluctuations. From 

the information available, it appears to have a large taproot, 

which should give some resistance to the impacts 

of drought and fire, but high-intensity fire or burning at 

peak growth times could be detrimental. It has shown 

positive to no effects from tree thinning and prescribed 

burning operations in ecological restoration research 

studies, but additional research that specifically targets 

this species would be useful before we can draw firm 

conclusions. 

Figure 3. Proportion of permanent monitoring plots through time containing Astragalus rusbyi at an ecological restoration 

study area near Flagstaff, AZ. Treatments were randomly assigned within each block and included (a) no thinning, 

no burning (control), (b) 1.5-3 tree replacement (high-intensity thinning), (c) 2-4 tree replacement (mediumintensity 

thinning), and (d) 3-6 tree replacement (low-intensity thinning). All treatment units were thinned in 1999 

and subsequently treated with prescribed fire in spring 2000 (Block 3) and spring 2001 (Blocks 1 and 2). 

161


Figure 4. Astragalus rusbyi is located at the positive end of the ‘leaf economic spectrum’ because of its high specific 

leaf area and tissue nitrogen concentration. This scatterplot illustrates the results of a principal components analysis 

of 133 plant species and 10 functional traits (adapted from Laughlin 2009) that occur in southwestern USA ponderosa 

pine forests. Species symbols are coded by plant functional types. A few species names are highlighted: ARPU = 

Aristida purpurea, BLTR = Blepharoneuron tricholepis, CAGE = Carex geophila, HECO = Hesperostipa comata, 

IRMI = Iris missouriensis, LUAR = Lupinus argenteus, MUMO = Muhlenbergia montana, OXLA = Oxytropis lambertii, 

PIPO = Pinus ponderosa, QUGA = Quercus gambelii. 


We would like to thank Wally Covington, Margaret 

Moore, Pete Fulé, Marta Fisher, Julie Korb, Mark 

Daniels, Jon Bakker, and Cheryl Casey and the staff and 

students of the Northern Arizona University Ecological 

Restoration Institute and School of Forestry who assisted 

with data collection. We would also like to thank 

Martin Wojciechowski and Jill Craig for their reviews 

of the paper and the U.S. Forest Service for providing 

funding to carry out the majority of the research described 

in this paper. 

REFERENCES 

Allphin, L., N. Brian and T. Matheson. 2005. Reproductive 

success and genetic divergence among varieties 

of the rare and endangered Astragalus cremnophylax 

(Fabaceae) from Arizona, USA. Conservation Genetics. 

6: 803-821. 

Anderson, D.G. 2004. Populations Status Survey of 

Schmoll’s Milkvetch (Astragalus schmolliae C.L. Porter). 

Prepared for National Park Service Mesa Verde 

National Park. Colorado Natural Heritage Program. Fort 

Collins, CO. 81 pp. 

Barneby, R.C. 1964. Atlas of North American Astragalus. 

Memoirs of The New York Botanical Garden. 

13:1–1188. 

162


Barneby, R.C. 1989. Fabales. Intermountain Flora, 

vol. 3, part B. A. Cronquist et al. (editors). The New 

York Botanical Garden, New York. 

Breinholt, J.W., R.Van Buren, O.R. Kopp, and C.L. 

Stephen. 2009. Population genetic structure of an endangered 

Utah endemic, Astragalus ampullarioides (Fabaceae). 

American Journal of Botany. 96(3):661-667. 

Covington, W.W. and M.M. Moore. 1994. Southwestern 

ponderosa pine forest structure: changes since 

Euro-American settlement. Journal of Forestry. 92: 39- 

47. 

Fisher, M.A. and P.Z. Fulé. 2004. Changes in forest 

vegetation and arbuscular mycorrhizae along a steep 

elevation gradient in Arizona. Forest Ecology and Management. 

200: 293–311. 

Fulé, P.Z, W. W. Covington and M.M. Moore. 1997. 

Determining reference conditions for ecosystem management 

of southwestern ponderosa pine forests. Ecological 

Applications.7: 895-908. 

Gamon, J.G. 1995. Report on the status of Astragalus 

sinuatus Piper. Olympia, WA: Washington Natural 

Heritage Program, Washington Department of Natural 

Resources. 

Greene, H.L. 1884. New plants of the Pacific Coast. 

Bulletin of the California Academy of Sciences. 1(1 – 

Feb. 1884): 8. 

Gremer, J.R., E.E. Crone, and P. Lesica. 2012. Are 

dormant plants hedging their bets? Demographic consequences 

of prolonged dormancy in variable environments. 

The American Naturalist 179(3):315-327. 

Heidel, B.L. and J. Vanderhorst. 1996. Sensitive 

plant surveys in Beaverhead and Madison counties, MT. 

Unpublished report to the Bureau of Land Management. 

Montana Natural Heritage Program, Helena. 85 pp. plus 

appendices. 

Hodgson, J.G., P.J. Wilson, R. Hunt, J.P. Grime, and 

K. Thompson. 1999. Allocating C-S-R plant functional 

types: a soft approach to a hard problem. Oikos. 85:282- 

294. 

Horvath, J. 2002. Factsheet Astragalus cicer (Cicer 

milkvetch). University of Saskatchewan, Rangeland 

Ecosystems and Plants, PL EC 434.3. 

Karron, J.D., Y.B. Linhart, C.A. Chaulk, and C.A. 

Robertson. 1988. Genetic structure of populations of 

geographically restricted and widespread species of Astragalus 

(Fabaceae). American Journal of Botany. 75 

(8): 1114-1119. 

Ladyman, J.A.R. 2003. Astragalus ripleyi Barneby 

(Ripley’s milkvetch): A technical conservation assessment. 

Prepared for the USDA Forest Service, Rocky 

Mountain Region, Species Conservation Project. 48 pp. 

Laughlin, D.C., J.D. Bakker, M.L. Daniels, M.M. 

Moore, C.A. Casey, J.D. Springer. 2008. Restoring plant 

species diversity and community composition in a ponderosa 

pine-bunchgrass ecosystem. Plant Ecology. 

197:139-151. 

Lesica, P. 1995. Demography of Astragalus scaphoides 

and effects of herbivory on population growth. 

Great Basin Naturalist. 55(2): 142-150. 

Lesica, P. and B.M. Steele. 1994. Prolonged dormancy 

in vascular plants and implications for monitoring 

studies. Natural Areas Journal. 14(3): 209-212. 

Lesica, P., R. Yurkewycz, and E.E. Crone. 2006. 

Rare plants are common where you find them. American 

Journal of Botany. 93(3): 454-459. 

Miller, M.E., R.K. Mann and J.D. Yount. 2007. Ecological 

Investigations of the Federally Endangered 

Shivwits Milk-Vetch (Astragalus ampullarioides) – 

2006 Annual Report. U.S. Department of the Interior, 

U.S. Geological Survey, in cooperation with the National 

Park Service, Zion National Park. Open-File Report 

2007-1050. 34 pp. 

NatureServe. 2009. NatureServe Explorer: An online 

encyclopedia of life [web application]. Version 7.1. 

NatureServe, Arlington, Virginia. Available http:// 

www.natureserve.org/explorer. (Accessed: May 27, 

2009 ). 

Sanderson, M.J. 1991. Phylogenetic Relationships 

within North American Astragalus L. (Fabaceae). Systematic 

Botany. 16(3): 414-430. 

Spellenberg, R. 1974. Chromosome number as an 

indication of relationships of Astragalus, section Strigulosi 

(Leguminosae), with descriptive notes on A. altus. 

The Southwestern Naturalist. 18(4): 393-396. 

Spellenberg, R. 1976. Chromosome numbers and 

their cytotaxonomic significance for North American 

Astragalus (Fabaceae). Taxon. 25(4): 463-476. 

Welsh, S.L. 2007. North American Species of Astragalus 

Linnaeus (Leguminosae): A Taxonomic Revision. 

Brigham Young University, Provo, UT. 932 pp. 

Wojciechowski, M.F. 2005. Astragalus (Fabaceae): 

A molecular phylogenetic perspective. Brittonia. 57(4): 

382-396. 

163


Long-term Responses of Penstemon clutei (Sunset Crater Beardtongue) 

to Root Trenching and Prescribed Fire: Clues for Population Persistence 

Judith D. Springer 1 , Peter Z. Fulé 2 , and David W. Huffman 1 

1 Ecological Restoration Institute, Northern Arizona University, Flagstaff, AZ, and 

2 School of Forestry, Northern Arizona University, Flagstaff, AZ 

Abstract. Penstemon clutei A. Nelson (Sunset Crater beardtongue) is narrowly endemic to the cinder hills and volcanic 

fields northeast of Flagstaff, Arizona. Disturbances such as wildfire, tornadoes, logging activity, and tree mortality 

from bark beetle outbreaks appear to stimulate regeneration of this species, but the manner in which populations 

persist between events is still largely unknown. From 1994-2000, we examined P. clutei responses to prescribed 

burning and root trenching treatments that were experimentally implemented as proxies for surface fire and reduced 

tree densities that might be observed following natural disturbance. We revisited this experiment in 2008 to assess 

long-term effects of the treatments. We also collected soil samples at this time to evaluate the importance of a persistent 

seed bank in population dynamics. In 2008, the mean number of P. clutei plants on trenched plots had declined 

with time, but was still significantly higher than on the control plots (mean density of 7.4 plants in trenched plots vs. 

0.6 plants in control plots). There was no significant difference in density between burned and unburned plots. Only 

21 P. clutei seedlings emerged from 176 soil seed bank samples, and we found no correlation between the number of 

P. clutei plants aboveground and the number of emergents from the samples. A targeted study to obtain samples near 

the base of reproductively mature plants produced 9 emergents from 30 samples. Results from this work suggest that 

disturbances that reduce competition for soil resources may be associated with long-term population persistence. Latent 

seed banks appear to be of only minor importance in recovery after disturbance; however, additional research 

with larger sample sizes would allow for greater confidence in this conclusion. We also recommend that additional 

long-term research be conducted on the response of this species to specific disturbances and stressors such as wildfire, 

tree mortality from bark beetle outbreaks, and water limitations. 

Penstemon clutei (Sunset Crater beardtongue) is a 

narrow endemic that occurs on volcanic soils to the 

northeast of Flagstaff in northern Arizona. The species 

is primarily restricted to tephra deposits from the Sunset 

Crater eruption (estimated dates of eruption vary from 

approximately 1040-1100 AD) at an elevation of approximately 

2135 m (7000 ft), but a disjunct population 

is also present on older cinder cones about 20 km to the 

northwest of Sunset Crater (Figure 1). P. clutei is typically 

found in open ponderosa pine forests and pinyonjuniper 

woodlands in areas containing a sparse understory, 

commonly on fairly coarse and dry, cindery soils 

that lie over a series of finer textured sandy or silty 

bands, which may alternate with coarse layers of cinders 

(Abella and Covington 2006, Phillips et al. 1992). The 

type specimen was collected by Willard Clute in July 

1923 north of the San Francisco Peaks in “lava sand,” 

and was described and named by Aven Nelson from the 

University of Wyoming (Nelson 1927). Growing to 

about 50-75 cm (20-30 in) in height, P. clutei has bluish-green 

glaucous leaves with serrated margins and 

gradually inflated, deep pink corollas. It is a very showy 

and attractive specimen plant and is readily available to 

gardeners through the horticultural trade. Flowering 

times vary by year, but it has been observed to flower 

from April through early September. It is ranked G2 

(imperiled) by NatureServe (2009) and is on the U.S. 

Forest Service sensitive species list for Region 3 

(Southwestern Region) (Arizona Game and Fish Department 

2003; D. C. Crisp, personal communication, 

2009). 

There is speculation that P. clutei is descended from 

P. pseudospectabilis (desert penstemon) and that geographic 

isolation occurred following the Sunset Crater 

eruption (Bateman 1980), or it may be intermediate between 

P. pseudospectabilis and P. palmeri (Palmer’s 

penstemon) (Clokey and Keck 1939). Phylogeny reconstruction 

of the genus Penstemon using nuclear and 

chloroplast sequence data and parsimony analysis produced 

incongruent results (Wolfe et al. 2006). Strict 

consensus trees generated from ITS (Internal Transcribed 

Spacers) placed P. clutei in a polytomy with P. 

bicolor (pinto beardtongue), P. floridus (Panamint 

beardtongue), P. palmeri, and P. rubicundus (Wassuk 

Range beardtongue). In contrast, strict consensus trees 

generated from chloroplast sequence data placed P. clutei 

as sister to P. centranthifolius (scarlet bugler) (Wolfe 

et al. 2006). With both methods (ITS and chloroplast 

sequence), the genera within the tribe Cheloneae had 

high bootstrap values. However, few terminal line- 

164


Figure 1. Range of Penstemon clutei (Sunset Crater beardtongue) in northern Arizona. 

ages of sister taxa within the Penstemon clade had bootstrap 

values above 70%, which is the generally accepted 

value for moderate to strong support. The contradictory 

results are likely due to hybridization and/or rapid speciation 

among penstemons. Wolfe and her co-authors 

(1998) have documented hybridization among some 

Penstemon species and have also demonstrated that pollen-mediated 

gene flow occurs via hummingbird vectors. 

Many narrow endemics are found in extreme edaphic 

conditions, including recent volcanic soils (Lesica et al. 

2006). Characteristic of many of these species are high 

population growth rates but poor dispersal rates. Their 

restricted ranges may in some cases be due more to recent 

evolution than to ecological tolerance: it may simply 

be the case that some species have not yet had time 

to spread across the landscape and may therefore be 

relatively young (neoendemics) (Lesica et al. 2006). 

Neoendemism is common in intermountain regions of 

western North America, and P. clutei is likely a fairly 

recently evolved species. 

Little is currently known about the ecology of P. clutei, 

and information in the scientific literature is sparse. 

It is believed to be a short-lived perennial like many 

other taxa in the genus Penstemon, perhaps living five 

to seven years on average. No long-term population 

studies following individual plants have been conducted 

to date in its natural habitat, so estimates of its longevity 

are purely speculative at this time. Observations in the 

field have suggested a link to disturbances. Prolific 

growth was observed following the Burnt Fire in 1973 

(Goodwin 1979) and the Hochderffer Fire in 1996 (Fulé 

et al. 2001). It has also been observed growing in large 

numbers in the path left by a tornado (Crisp 1996) as 

well as surrounding Pinus edulis (pinyon pine) trees that 

were killed by drought and bark beetles in 2002-2003 

(J.D. Springer, personal observations, 2008 and 2009). 

Phillips and others (1992) noted vigorous plants and 

high seedling numbers in areas of past disturbance, especially 

from logging operations. Plants were particularly 

prevalent near decaying logs and stumps. Large 

numbers of reproductively mature plants are also sometimes 

found in a ring surrounding recent Pinus ponderosa 

(ponderosa pine) snags (Fulé et al. 2001). 

Because plants had been noted to emerge in abundance 

following wildfire, two prescribed burning studies 

165


were established by the U.S. Forest Service, but results 

were inconclusive (Nagiller 1992). We initiated a study 

in 1992 to test the hypothesis that restoration of historic 

ecosystem conditions may enhance the sustainability of 

this species (Fulé et al. 2001). This study encompassed 

several components, including seed germination studies, 

a field seeding trial, a prescribed burning study and a 

trenching experiment. We initiated the prescribed burning 

component in 1994 to test the hypothesis that prescribed 

fire would increase P. clutei density by removing 

litter and competing vegetation. The results suggest 

that prescribed burning caused a significant decline in 

density by as much as 75%. However, density also declined 

in two of the three control areas (in one area also 

by as much as 75%). So, while prescribed burning appears 

to be responsible for the death of mature plants, 

natural population declines may also occur in the absence 

of disturbance. 

After evaluating results from the prescribed burning 

experiment, we investigated the possibility that vigorous 

responses following fires were a result of mortality of 

overstory trees and removal of root competition (Fulé et 

al. 2001). We initiated a study in 1998 to test the hypothesis 

that cutting root competition through trenching 

would increase P. clutei density. In 1999, one year following 

the trenching, there was a significant difference 

in density between trenched (mean of 104.9 plants/plot) 

and control plots (14.0 plants/plot), mostly in the form 

of seedlings. By 2000, densities had declined to an average 

of 30.6 vs. 1.5 plants in the trenched and control 

plots, respectively, mostly due to the death of seedlings. 

Two preliminary conclusions were drawn from the 

trenching study: 1) trenching had a positive effect on P. 

clutei reproduction, and this trend was still evident a 

year later, and 2) increases in P. clutei were likely due 

to reduced root competition with overstory trees. Although 

our earlier germination experiments indicated 

that P. clutei did not exhibit innate seed dormancy under 

laboratory conditions (see Fulé et al. 2001), we were 

puzzled over the dramatic field response to root trenching. 

A field seeding trial of P. clutei showed very poor 

rates of establishment (0.1-0.6%), with no seedlings establishing 

after an April seeding, and only a minimal 

number establishing following an October seeding (Fulé 

et al. 2001). Determining whether P. clutei maintains a 

persistent soil seed bank is crucial for conservation and 

management efforts. The combined results from our 

previous studies suggest that it does not form a persistent 

seed bank, that there may be dissimilarities in germination 

rates between plants from different habitats, 

and that field emergence is extremely low and/or seedling 

mortality is high. Collecting P. clutei seeds from 

additional habitats could yield new information on 

whether this species exhibits cyclic dormancy patterns 

or dormancy that differs in contrasting habitats. 

It remains largely unknown, then, how long populations 

of P. clutei plants persist on the landscape following 

disturbance, what mechanism this species employs 

to colonize an area following disturbance, or how it is 

able to disperse across the landscape. In an effort to gain 

answers to some of these questions, we revisited the 

study area ten years after root trenching and 13-14 years 

following prescribed burning. Our objectives were to 

assess the long-term effects of the prescribed burning 

and trenching treatments and to evaluate the importance 

of a persistent seed bank in population dynamics. 

METHODS 

Fulé and others (2001) described methods of our previous 

prescribed burning and root trenching studies in 

detail, but we will also summarize them here. The experimental 

studies described in this and the 2001 paper 

were established in 1992-1994 and conducted on Coconino 

National Forest lands in the vicinity of O'Leary 

Peak, adjacent to Sunset Crater National Monument 

(Figure 1). The elevation of the study area is approximately 

2100-2300 m (6890-7550 ft). Soils are cindery 

and deep, well-drained Vitrandic Ustochrepts and Typic 

Ustorthents (Miller et al. 1995). Weather records from 

Sunset Crater National Monument, 1 km south of the 

study area, include an annual precipitation average of 

42.7 cm (16.8 in) (1969-2008), with most precipitation 

occurring in winter and during the summer monsoon 

(July-September). However, annual precipitation has 

varied widely in recent decades from a low of 23.6 cm 

(9.3 in) in 1989 to 66 cm (26.0 in) in 1992. The average 

minimum temperature in January is -11 o C and the average 

maximum temperature in July is 29 o C. 

We established an experiment to study the effects of 

prescribed burning on the P. clutei community in 1994. 

Forty P. clutei plant-centered plots were established, 

each with a 2.5 m radius circle (area = 19.6 m 2 ) centered 

0.3 m northwest of an existing plant. P. clutei was tallied 

in four categories: seedling, second-year plant, mature 

plant, and dead. Field experience indicated that the 

distinctions between the living plant categories were 

approximate. Plots were randomly selected for burn or 

control treatments, and burning was conducted in September 

1994. Burn season effects were tested on a second 

experimental site immediately north of the fall 

burning site, and twenty randomly selected plots were 

burned in April 1995. The fall burn plots were re-measured 

in July 1995. All 80 plots, including spring and fall 

burns, were re-measured in August/September 1996; 

August 1997; August 1998; and August/September 

2008. Changes in P. clutei density were analyzed with 

repeated measures analysis of variance (Systat 8.0, 

SPSS Science, Chicago). Data were square-root transformed 

to meet ANOVA assumptions. In 2008, data 

166


were cube-root transformed, and the software used was 

JMP 8.0 (SAS, Cary, NC). 

In October 1997 we established ten new experimental 

plots for root trenching within the low-elevation 

burning study area. Each plot was paired with a nearby 

control plot from the original burn experiment. The 

number of trenched plots was eventually dropped to 

eight due to off-road vehicle damage and other factors. 

The below ground effects of tree removal were simulated 

by digging a narrow trench approximately 1 m 

deep around each plot. Trenches were located 50 cm 

outside the plot boundary to avoid physical disturbance 

within the measured area and were lined with plastic 

sheeting to minimize tree root regrowth (Milne 1979). 

Trenches were backfilled immediately after lining, and 

plots were re-measured in August 1998, September 

1999, August 2000 and August/September 2008. P. clutei 

density was analyzed as described for the prescribed 

burning study above. 

In 2008, we collected two soil seed bank samples 

from each of the 80 prescribed burning study plots and 

also from the trenched plots (176 total). Samples were 

collected to a depth of 5 cm, approximately 15 cm away 

from the original plot center to the east and west, and 

each core was approximately 70 cm 3 in volume. We also 

collected 30 targeted seed bank samples 15 cm to the 

east and west of reproductively mature individuals that 

were located outside of plots. Seeds are dispersed in the 

fall and winter, and germination is thought to occur during 

late spring and early summer rains, so we collected 

soil seed bank samples in late August and early September, 

presumably after germination occurred, but before 

new seeds were dispersed, in an effort to capture seeds 

in the persistent soil seed bank. We sieved samples to 

remove large cinders and placed the soil samples on potting 

soil in gallon-sized pots. Samples were placed in 

the greenhouse in September 2008 and received artificial 

light, one application of Miracle Gro® and daily 

watering for five months, using the seed emergence 

method (Ter Heedt et al. 1996). 

RESULTS 

In 2008, thirteen and fourteen years after spring and 

fall burning, respectively, there was no significant difference 

in P. clutei density between burned and unburned 

plots (Figure 2). Mean density in burned plots 

(combined spring and fall burns) was 0.9 live plants, 

and mean density in control plots was 0.6 live plants. 

Over the course of this study, there has been a general 

decline in P. clutei in both burned and control plots. 

However, there was still a significant difference in mean 

density between trenched and control plots ten years 

Figure 2. Mean density of Penstemon clutei plants following a prescribed burn study near Sunset Crater National 

Monument in northern Arizona. Pre-treatment data were collected in 1994 and 1995. Bars indicate standard error. 

167


Figure 3. Mean density of Penstemon clutei plants following a root trenching experiment near Sunset Crater National 

Monument in northern Arizona . Pre-treatment data were collected in 1997. Bars indicate standard error (each plot 

was 19.6 m 2 ; n=80; p value in 2008 was


point to a number of studies showing that root trenching 

can lead to an increase in the availability of mineral nitrogen. 

Research by Selmants and Hart (2008) indicates 

that there are large carbon and nitrogen pools and fluxes 

under the canopies of one-seed juniper (Juniperus 

monosperma) in soils around Sunset Crater. Soils under 

juniper and pinyon pine canopies are generally higher in 

both carbon and nitrogen than are intercanopy sites, but 

these levels vary according to the age of the soils, which 

are of volcanic origin in this region. 

Abella and Covington (2006) obtained samples from 

a number of soil types across the Coconino National 

Forest, including black and red cinder soils in the vicinity 

of Sunset Crater, and determined that black cinder 

soils contain the driest surface soils among those tested. 

These soils are very sandy (>90% concentration at 0-15 

cm depth), and contained the fewest plant species per 

plot. Red cinder soils are also quite sandy (averaging 

63% concentration at 0-15 cm depth). Soil samples 

taken from red cinders contained no calcium carbonate, 

but these soils had higher organic carbon and total nitrogen 

than the black cinder soils, and they also had higher 

soil moisture. P. clutei populations in pinyon-juniper 

woodlands are typically found on these older, red cinder 

soils. 

The soils on which P. clutei grows, then, are arguably 

some of the harshest in northern Arizona and are 

susceptible to extreme environmental fluctuations. 

Seedling mortality is high, but once plants reach maturity, 

they have adapted to the harsh, arid environment by 

means of a large taproot or thick lateral roots (D.W. 

Huffman, personal observations, 2008) and thick leaves. 

The species also must have developed adaptations to be 

able to rapidly colonize following disturbance, perhaps 

through longevity, rapid dispersal, high germinability, 

or a persistent soil seed bank. Soil seed banks buffer 

populations against environmental variation, and seed 

dormancy is a mechanism of escape from unfavorable 

conditions in time (compared to dispersal, which is an 

escape in space) (Doak et al. 2002). Short-lived perennials 

in areas of high environmental variation, which includes 

most rare plants in the Southwest, often rely on 

the soil seed bank for recruitment (Doak et al. 2002). P. 

lemhiensis (Lemhi penstemon) and P. palmeri have 

been documented to buffer populations against environmental 

fluctuations by maintaining a persistent soil seed 

bank (Heidel and Shelly 2001, Meyer and Kitchen 

1992). Conversely, long-lived perennials are often more 

reliant on growth and survival of established plants than 

on recruitment from seed or soil seed banks (Lesica 

1995). If a species does not exhibit innate dormancy, it 

is unlikely that it forms a soil seed bank. Because P. 

clutei plants have been observed to appear in large numbers 

following a disturbance such as the Hochderffer 

Fire (P.Z. Fulé, personal observation, 1997 and 1998), 

conventional thinking is that this species forms a persistent 

soil seed bank (Phillips et al. 1992), but it may also 

maintain genetic diversity through existing reproductively 

mature plants scattered across the landscape, or 

exhibit rapid dispersal rates following disturbance. 

While it does seem from our study that P. clutei may 

form a minor persistent seed bank, the degree of its importance 

in recovery following disturbance is unknown, 

and larger sample sizes from additional habitats are necessary 

in order to make inferences about its significance 

for recruitment following disturbance. 

Meyer et al. (1995) found a diversity of germination 

timing mechanisms in 38 Intermountain West Penstemon 

species. Seeds of many of these species diverge 

into two fractions. One fraction does not exhibit dormancy 

and will germinate readily under optimal conditions 

in the first year. The other fraction may respond to 

chilling cues and become nongerminable, allowing for 

between-year carryover in the soil seed bank. Meyer and 

her co-authors (1995) found this strategy to be especially 

common in populations of penstemons from middle 

elevation areas that have unpredictable winters. 

Meyer and Kitchen (1992) discovered that P. palmeri 

seeds undergo cyclic dormancy changes in the field. 

Moist chilling induces secondary dormancy in about 

half of the seeds, while moisture combined with summer 

temperatures removes secondary dormancy. These 

mechanisms allow for a persistent soil seed bank and for 

seeds that can persist from year to year without burial. 

The result is that some seeds germinate in the spring, 

while those seeds that are rendered dormant by chilling 

are carried over in the soil seed bank. Another fruitful 

area of research for this species could include seed augmentation 

studies to determine if a paucity of viable 

seeds may be limiting establishment. Abella (2008) conducted 

such a study with P. virgatus (upright blue 

beardtongue) in a ponderosa pine forest not far from our 

study site and found that under particular experimental 

conditions, the site environment (e.g., tree overstory) 

apparently was more limiting to recruitment than either 

leaf litter thickness or seed availability. 

Our results also indicate that prescribed burning 

alone does not seem to be a useful management tool for 

this species, as it appears to kill reproductively mature 

individuals leading to potential decreases in available 

seeds for future recruitment. An experiment involving 

use of prescribed fire as a management tool for P. lemhiensis 

returned variable results (Heidel and Shelly 

2001). Fire appeared to cause mortality of adult plants 

ranging from 25-75%. However, the burning caused 

an increased recruitment rate of 4600-6400%. As we 

pointed out in our previous paper (Fulé et al. 2001), 

patchy tree mortality does appear to benefit P. clutei. 

Tree mortality from the 2002-2003 bark beetle outbreak 

among pinyon pines appears to be correlated with dra- 

169


matic increases in the number of reproductively mature 

individuals on the south-facing slopes of cinder cones in 

the vicinity of Sunset Crater and Indian Flat (J.D. 

Springer, personal observations, 2008 and 2009). 

Although some P. clutei populations contain hundreds 

or thousands of individuals, populations are often 

widely dispersed, and there are a few major threats that 

could jeopardize this species in the future. The entire 

range of P. clutei has not yet been mapped; however, a 

large portion of its known range falls within the Cinder 

Hills OHV (off-highway vehicle) area (Figure 1). Most 

of this area is not fenced and OHV use spills outside the 

boundaries shown in the map. Although P. clutei appears 

to benefit from disturbance, whether disturbance 

is beneficial or detrimental depends on the type of disturbance 

and the amount of impact. No quantifiable data 

has yet been collected on the impacts of OHV activity 

on this species, but anecdotal evidence points to OHVs 

as a direct factor in adult P. clutei mortality (J.D. 

Springer, personal observations, 2008). OHV activity 

causes above- and belowground soil impacts, resulting 

in decreased soil moisture, increased soil bulk density, 

and increased water infiltration time, which have been 

shown to negatively impact plant species in the area, 

such as ponderosa pine (Kennedy 2005). 

While OHV use and impacts can be controlled, potential 

negative changes to P. clutei habitat from climate 

change cannot. Climate models predict a more arid climate 

in the southwestern U.S. in the coming decades 

(Seager et al. 2007). This species already lives in a 

harsh environment, and any major decreases in available 

soil moisture could significantly impact its long-term 

viability. Additional threats include potential hybridization 

with other Penstemon species brought to the area 

for horticulture or highway revegetation purposes, herbivory, 

insect damage and urban expansion. 

Determining the long-term population dynamics of 

this species is integral to future conservation management 

planning and points out the direct need for longterm 

monitoring, particularly in the face of potential 

climate change and unmanaged OHV use in the center 

of its habitat. Teasing out whether P. clutei population 

declines occur from disturbance, absence of disturbance, 

senescence, competition, drought, climate change, interactive 

effects, or other as yet undetermined factors will 

be critical for understanding future conservation and 

management needs for this species. 

ACKNOWLDEGMENTS 

We thank Deb Crisp, Barb Phillips, and Frank Thomas 

(Coconino National Forest), Steve Rosenstock 

(Arizona Game and Fish Department), Paul Whitefield 

(National Park Service), and Susie Smith (Northern Arizona 

University) for their assistance in gathering locations 

and for their input on ecology and experimental 

design; Scott Abella (UNLV and Public Lands Institute) 

and Nancy Brian (National Park Service) for their review 

of the manuscript; and Joe Crouse, Mark Daniels, 

Chris McGlone, Mike Stoddard, Chelsea Green, Cat 

McGowan, Don Normandin and students and staff of 

the Ecological Restoration Institute at Northern Arizona 

University for assistance with data collection, mapping, 

greenhouse study maintenance and statistical analysis. 


Abella, S.R. 2008. Plant recruitment in a northern 

Arizona ponderosa pine forest: testing seed- and leaf 

litter-limitation hypotheses. Pp. 119-127 in Olberding, 

Susan D., and Moore, Margaret M., tech coords. 2008. 

Fort Valley Experimental Forest—A Century of Research 

1908-2008. Proceedings RMRS-P-53CD. Fort 

Collins, CO: U.S. Department of Agriculture, Forest 

Service, Rocky Mountain Research Station. 408 p. 

Abella, S.R. and W.W. Covington. 2006. Forest ecosystems 

of an Arizona Pinus ponderosa landscape: multifactor 

classification and implications for ecological 

restoration. Journal of Biogeography. 33: 1368-1383. 

Arizona Game and Fish Department. 2003. Penstemon 

clutei. Unpublished abstract compiled and edited 

by the Heritage Data Management System, Arizona 

Game and Fish Department, Phoenix, AZ. 5 pp. 

Bateman, Gary C. 1980. Natural Resource Survey and 

Analysis of Sunset Crater and Wupatki National Monuments. 

Final Report (Phase III). Prepared for the Office 

of Natural Resources Management, Southwest Region, 

National Park Service. 

Clokey, I.W. and D.D. Keck. 1939. Reconsideration 

of certain members of Penstemon subsection spectabilis. 

Bulletin of the Southern California Academy of Science. 

38: 8-13. 

Coomes, D.A. and P.J. Grubb. 2000. Impacts of Root 

Competition in Forests and Woodlands: A Theoretical 

Framework and Review of Experiments. Ecological 

Monographs. 70(2): 171-207. 

Crisp, D.L. 1996. Monitoring of Penstemon clutei A. 

Nels. on tornado salvage. In USDA Forest Service General 

Technical Report RM-GTR-283, pp. 243-246. 

Rocky Mountain Forest and Range Experiment Station, 

Fort Collins, Colorado. 

Doak, D.F., D. Thomson, and E.S. Jules. 2002. Population 

Viability Analysis for Plants: Understanding the 

Demographic Consequence of Seed Banks for Population 

Health. In: Steven R. Beissinger and Dale R. 

McCullough, Eds. Population Viability Analysis. University 

of Chicago Press. 

Fulé, P.Z., J.D. Springer, D.W. Huffman, and W.W. 

Covington. 2001. Response of a rare endemic, Penstemon 

clutei, to burning and reduced belowground competition. 

Pp 139-152 in Maschinski. J., and L. Holter 

170


(tech. coords.), Proceedings: Third Rare and Endangered 

Plant Conference. Proceedings RMRS-P-23. 

Goodwin, G. 1979. Observations on Penstemon clutei 

on the Coconino National Forest. Unpublished report 

on file at Supervisor's Office, Coconino National Forest, 

Flagstaff, AZ. 7 pp. 

Heidel, B. and J.S. Shelly. 2001. The Effects of Fire 

on Lemhi Penstemon (Penstemon lemhiensis) – Final 

Monitoring Report 1995-2000. Montana Natural Heritage 

Program. Prepared for Beaverhead-Deerlodge National 

Forest and Dillon Field Office – Bureau of Land 

Management. 

Kennedy, K.J. 2005. Above- and belowground impacts 

of off-road vehicles negatively affect establishment 

of a dominant forest tree. Master’s thesis, Northern 

Arizona University, Flagstaff. 39 pp. 

Lesica, P. 1995. Demography of Astragalus scaphoides 

and effects of herbivory on population growth. 

Great Basin Naturalist. 55(2): 142-150. 

Lesica, P., R. Yurkewycz, and E.E. Crone. 2006. 

Rare plants are common where you find them. American 

Journal of Botany. 93(3): 454-459. 

Meyer, S.E., and S.G. Kitchen. 1992. Cyclic seed 

dormancy in the short-lived perennial Penstemon 

palmeri. Journal of Ecology 80: 115-122. 

Meyer, S.E., S.G. Kitchen and S.L. Carlson. 1995. 

Seed germination timing patterns in intermountain Penstemon 

(Scrophulariaceae). American Journal of Botany. 

82(3): 377-389. 

Miller, G., N. Ambos, P. Boness, D. Reyher, G. 

Robertson, K. Scalzone, R. Steinke and T. Subirge. 

1995. Terrestrial ecosystems survey of the Coconino 

National Forest. USDA Forest Service, Southwestern 

Region. 

Milne, M.M. 1979. The effects of burning, root 

trenching, and shading on mineral soil nutrients in 

southwestern ponderosa pine. Master’s thesis, Northern 

Arizona University, Flagstaff. 

Nagiller, S. 1992. Untitled report of prescribed burning 

results on file at Elden Ranger District, Coconino 

National Forest, Flagstaff, AZ. 2 pp. 


encyclopedia of life [web application]. Version 7.1. 

NatureServe, Arlington, Virginia. Available http:// 

www.natureserve.org/explorer. (Accessed: May 27, 

2009 ). 

Nelson, A. 1927. A new Penstemon from Arizona. 

American Botanist. 33: 109-110. 

Phillips, A. M. III, M. B. Murov, and R. J. van Ommeren. 

1992. Distribution and ecology of Sunset Crater 

beard tongue (Penstemon clutei) in the Cinder Hills 

area, Coconino National Forest, Arizona. Unpublished 

report on file at the Supervisor's Office, Coconino National 

Forest, Flagstaff, Arizona. 11 pp. 

Seager, R., M. Ting, I. Held, Y. Kushnir, J. Lu, G. 

Vecchi, H. Huang, N. Harnik, A. Leetmaa, N. Lau, C. 

Li, J. Velez, and N. Naik. 2007. Model Projections of an 

Imminent Transition to a More Arid Climate in Southwestern 

North America. Science. 316(5828): 1181- 

1184. 

Selmants, P.C. and S.C. Hart. 2008. Substrate age 

and tree islands influence carbon and nitrogen dynamics 

across a retrogressive semiarid chronosequence. Global 

Biogeochemical Cycles. 22, GB1021, doi: 

10.1029/2007GB003062, 2008. 

Ter Heedt, G.N.J., G.L. Verwiej, R.M. Bekker and 

J.P. Bakker. 1996. An improved method for seed-bank 

analysis: seedling emergence after removing the soil by 

sieving. Functional Ecology. 10(1): 144-151. 

Wolfe, A.D., Q. Ziang and S.R. Kephart. 1998. Assessing 

hybridization in natural populations of Penstemon 

(Scrophulariaceae) using hypervariable intersimple 

sequence repeat (ISSR) bands. Molecular Ecology. 

7:1107-1125. 

Wolfe, A.D., C.P. Randle, S.L. Datwyler, J.J. 

Morawetz, N. Arguedas and J. Diaz. 2006. Phylogeny, 

taxonomic affinities and biogeography of Penstemon 

(Plantaginaceae) based on ITS and cpDNA sequence 

data. American Journal of Botany. 93(11): 1699-1713. 

171


¡Viva thamnophila! Ecology of Zapata Bladderpod 

(Physaria thamnophila), an Endangered Plant of the 

Texas-Mexico Borderlands 

Dana M. Price, 

US Army Corps of Engineers, Albuquerque District 

Christopher F. Best, 

US Fish and Wildlife Service, Austin Ecological Services Field Office 

Norma L. Fowler, 

Section of Integrative Biology, University of Texas at Austin, 

and Alice L. Hempel, 

Department of Biology and Health Sciences, Texas A&M University- Kingsville 

Abstract. Conserving rare plants is dependent upon our ability to identify, manage and restore their habitat. We examined 

the plant community associates and habitat requirements of Physaria thamnophila, an endangered herbaceous 

perennial of the Tamaulipan shrubland of south Texas, at four sites from 2003 to 2007. At each site, vegetation 

height structure and species composition were sampled concurrently with intermittent censuses of P. thamnophila. 

We found significant and interesting differences among sites and years, as well as between our quantitative results 

and previous descriptions of P. thamnophila’s community and habitat. Existing plant community descriptions and 

mapped soil types do not provide a close match with our field observations. Finally, we discuss the application of 

these results to conservation of P. thamnophila and restoration of its community. 

An endangered plant that is a member of a fragmented 

and altered remnant plant community poses 

many challenges to conservation workers. In addition to 

threats to its (usually) few and small known populations, 

two additional challenges are faced. One, the detection 

of additional populations is often hampered by uncertainty 

in identifying its habitat; and two, uncertainty 

about its habitat requirements hampers management and 

restoration efforts. Physaria thamnophila (Zapata bladderpod; 

Brassicaceae) is one such plant. This endangered 

herbaceous perennial of south Texas grows in 

remnants of Tamaulipan thornscrub, likely on a specific 

but as yet poorly defined soil type and geologic substrate. 

Prior to the present study, the structure and composition 

of the specific plant community in which Zapata 

bladderpod occurs had been examined quantitatively 

for only one site (Sternberg 2005). 

The U.S. Fish and Wildlife Service (USFWS) Recovery 

Plan (USFWS 2004) highlights the need for better 

knowledge of the habitat and community in which P. 

thamnophila lives for several purposes, including discovering 

additional populations, locating sites for establishing 

new populations, and restoring and managing its 

habitat. This study addresses these goals by providing a 

detailed, quantitative description of the habitat and 

community of four sites with persistent (as defined by 

NatureServe 2002) P. thamnophila populations. The 

results will not only be useful for P. thamnophila conservation, 

but can improve our ability to manage and 

restore one of the communities that comprise the 

Tamaulipan thornscrub ecosystem. 

SPECIES 

Physaria thamnophila (Rollins and E.A. Shaw) 

O’Kane and Al-Shehbaz (formerly Lesquerella thamnophila) 

is a federally listed endangered species (USFWS 

2004) with a global conservation ranking of G1 

(critically imperiled; NatureServe 2009). It is a shortlived 

perennial with a rosette of silvery- or gray-green 

leaves covered with stellate trichomes (Figure 1) and 

one to several flexuous, sprawling to ascending flowering 

stems. The yellow flowers give rise to subglobose 

silicles on recurved pedicels (Rollins and Shaw 1973) 

(Figure 2). It usually flowers in spring (February to 

April), but can flower during midsummer or as late as 

September in response to rain (Poole et al. 2007; Sternberg 

2005). Germination has not been observed, but 

probably occurs in response to rainfall in cool weather. 

Under prolonged dry conditions, all of the plant’s leaves 

may die, making some surviving plants very difficult to 

locate. 

172


Figure 1. Physaria thamnophila rosette leaves. 

SITES 

All known populations of P. thamnophila (eight verifiably 

extant, one not accessible, and one historic population; 

data on file, Natural Diversity Database, Texas 

Parks and Wildlife Department, Austin, TX) are in Starr 

and Zapata Counties, Texas, between Zapata and Roma, 

a distance of 69 km (Figure 3). All populations occur in 

remnant Tamaulipan thornscrub vegetation (Poole et al. 

2007). For this study, we examined all sites with at least 

100 plants to which we had access. Other sites were excluded 

due to small populations, lack of access to private 

land, or discovery too late to include in the study. 

Three of the four sites in this study (Arroyo Ramirez, 

Arroyo Morteros, and Cuellar) were in the Lower Rio 

Grande Valley National Wildlife Refuge. The fourth 

study site, Santa Margarita, was on private land. All 

four sites were within 18 km of each other in Starr 

County, Texas, between Falcon Dam (26° 33’ N, 99° 

09’W) and Roma (26° 24’N, 99° 01’W). 

The climate in this region is hot and often dry. At 

Falcon Dam, the nearest station in the National Oceanic 

and Atmospheric Administration (NOAA) data base, the 

months of July and August have the highest means of 

daily maximum temperatures (37.6 o C and 37.5 o C, respectively) 

and January has the lowest mean daily minimum 

temperature (7.7 o C) (NOAA 2009). Between 

1963 and 2007, average annual precipitation at Falcon 

Dam was 515 mm. However, annual precipitation varied 

widely, as during those 45 years, there were 7 years 

with more than 600 mm and 9 years with less than 400 

mm annual precipitation. During our study, 2003 and 

2004 were wet (804 mm and 671 mm, respectively) and 

2005 was dry (353 mm). However, precipitation in the 

area occurs largely as isolated thunderstorms, and likely 

varied among sites. 

Most known populations occur on the edges of terraces 

above the Rio Grande flood plain. Physaria thamnophila 

populations have been found on the Jackson, 

173 

Figure 2. Physaria thamnophila fruiting stem. 

Yegua and Laredo formations (Bureau of Economic Geology 

1976; Poole et al. 2007; USFWS 2004), all of 

which are Eocene calcareous sandstones and clays. All 

four study sites were on Eocene sandstones: Cuellar and 

Arroyo Morteros on the Yegua Formation, and Arroyo 

Ramirez and Santa Margarita on the Jackson Group 

(Bureau of Economic Geology 1976). Arroyo Morteros, 

Arroyo Ramirez, and Santa Margarita were located 

along the edge of the cliff that marks the edge of the 

flood plain of the Rio Grande; Cuellar was ~ 3000 m 

from the flood plain in an area without a cliff. 

Known P. thamnophila populations occur on shallow, 

well-drained sandy loam soil. Soils at known sites 

are mapped as members of the Zapata, Maverick, Catarina, 

or Copita series as described by the Natural Resources 

Conservation Service (NRCS) (Poole et al. 

2007; USFWS 2004). These highly calcareous soils are 

derived from Eocene sandstone, clay and shale. Catarina 

soils are derived from Frio and Yegua formation parent 

material; these and Maverick soils contain up to 15 percent 

gypsum. Copita soil is derived from weakly consolidated 

calcareous sandstone of the Jackson Formation 

and is only slightly (2 percent) gypsiferous (NRCS 

2009; Thompson et al. 1972). The soils at our study 

sites have been described as follows: Arroyo Morteros: 

Copita, Zapata and Catarina; Arroyo Ramirez: Jimenez-


Figure 3: Physaria thamnophila distribution, Starr and Zapata Counties, Texas. 

174


Quemado; Cuellar: Catarina; Santa Margarita: Maverick, 

eroded and Jimenez-Quemado (NRCS 2009). However, 

soils maps for these areas lack precision and the 

inclusion of P. thamnophila sites within Jimenez- 

Quemado soil polygons is incorrect (see Discussion). In 

some sites, fossil oyster shells or gypsum crystals were 

conspicuous. 

METHODS 

Vegetation Data Collection 

Vegetation data were collected once at Arroyo Ramirez, 

Arroyo Morteros, and Santa Margarita and twice at 

Cuellar in conjunction with intermittent censuses of the 

study sites (Fowler et al. 2011). Cuellar vegetation was 

sampled in 2002 and 2007, Arroyo Ramirez in 2003, 

Santa Margarita in 2004, and Arroyo Morteros in 2007. 

Due to limitations of time and personnel, we set up 

study plots at one site each year in 2002-2005. In each 

site, permanent circular plots marked with rebar were 

located randomly along transects. Some of the Cuellar 

census plots did not have vegetation data collected in 

one or both years and were dropped from analyses. 

At Cuellar, 30 plots were located in an uncleared 

area and 30 plots in an area that had been brush-cleared 

using a ‘Woodgator©’ (roller-chopper) in December 

2000. This decreased shrub canopy (see Results) and 

significantly increased herbaceous species richness and 

grass abundance in the cleared area (Fowler et al. 2011). 

A transect ran along the margin between the two areas, 

with the plots on either side. In each of the other three 

sites, transects were located to sample the entire Physaria 

thamnophila population. The full extent of the 

Santa Margarita population was unknown at the beginning 

of this study, and the Arroyo Ramirez and Arroyo 

Morteros populations were discovered in fall 2002 and 

summer 2004, respectively. Therefore, we first conducted 

reconnaissance surveys to map the populations’ 

extents using GPS. The two roughly linear populations 

that followed rocky outcrops, Arroyo Ramirez and 

Santa Margarita, were each sampled by running a discontinuous 

transect (excluding unoccupied habitat) 

along the long axis, through the approximate midline of 

the population. The transect length was then divided 

into 30 strata, with plots located within each stratum at 

random distances either side of the transect line. There 

were 30 plots at Santa Margarita (Figure 4) and 34 plots 

at Arroyo Ramirez. At Arroyo Morteros, where the 

population occupied a large, irregular polygon, we 

placed 58 plots along ten parallel transects of different 

lengths, spaced 38m apart. The initial plot in each transect 

was located a random distance between 0.1 m to 10 

m along the transect, with successive plots spaced 13.8 

m apart. 

By 2007, the cleared portion of Cuellar was quite 

similar to the other three sites in herbaceous species 

richness and grass abundance (Fowler et al. 2011). Arroyo 

Morteros had on average somewhat greater herbaceous 

species richness than the other sites (12 species 

per plot, on average, versus 6 to 10 in the other sites; 

Fowler et al. 2011). 

Vegetation data from each plot were collected in five 

circular subplots. Each subplot was 0.25m in radius. 

One of these subplots was centered on the plot’s central 

point, and the other four were centered around points 

located on the circumference of the circle at the distal 

ends of four radial lines emanating from the central 

point (two parallel and two perpendicular to the transect). 

In each subplot, the presence/absence of each species 

was recorded in each of 4 height categories. These 

categories were 0.0 to 0.5 m, 0.5 to 1.0 m, 1.0 to 2.0 m, 

and 2.0 to 3.0 m above ground. No plants were taller 

than 3 m. Species names follow USDA PLANTS database 

(2012) with the exception of Physaria and Paysonia, 

for which we follow the treatment of Al-Shehbaz 

and O’Kane (2002). Most common species, particularly 

woody dominants, were identified in the field. We collected 

voucher specimens when species were encountered 

in flower; specimens are being prepared for deposit 

at the University of Texas herbarium (TEX-LL). 

It was not always possible to definitively identify 

species at the time when they were first observed. As a 

result, in a few cases the interim identifications used for 

unknown species were not consistent between years. 

Therefore, for analysis we pooled (a) the two Chamaesyce 

(Euphorbiaceae) species; (b) Chaetopappa bellioides 

and Aphanostephus skirrhobasis (Asteraceae); 

and (c) Chamaesaracha sordida and Physalis cinerascens 

(Solanaceae). This reduced the number of 'species' 

in the analyses from 150 to 147. For simplicity, each of 

these three pairs of species is referred to simply as a 

species. ‘Bare ground’ was recorded as a 148 th ‘species’. 

A few plants could never be identified; almost all of 

these were without fruits or flowers and many were 

grasses. These have been left as unknown species 1, 

unknown grass, etc. in Table 1. 

The number of subplots in each plot in which the 

species occurred was used to quantify the abundance of 

each species. The abundance of a given species in a plot 

therefore had a value between 0 and 5. For P. thamnophila 

only, counts of numbers of individuals in each plot 

were also available. However, censuses of P. thamnophila 

were conducted in all sites in 2006 and 2007 only. 

We calculated the density of P. thamnophila in each 

plot by dividing the number of individuals in the plot by 

the plot’s area, for each plot separately. P. thamnophila 

densities in 2006 and in 2007 were then averaged, for 

each census plot separately. 

175


Figure 4. Sampling design for discontinuous sites. 

176


Table 1. Species composition in Physaria thamnophila study plots by site 

Scientific Name Family >1 

m 

tall? 

# 

Sites 

w/sp a 

CL02 

cl 

% of Subplots Occupied in Each Site b 

CL02 

un 

CL07 

cl 

CL07 

un 

AM07 AR03 SM04 

Abutilon fruticosum Malvaceae 1 - - 0.74 - - - - 

Acacia berlandieri Fabaceae 1 - - - - - - 0.67 

Acacia rigidula Fabaceae X 5 47.20 33.06 60.00 39.26 46.55 24.12 18.00 

Acalypha monostachya Euphorbiaceae 2 - - - - - 1.18 9.33 

Acleisanthes obtusa Nyctaginaceae 2 - - 2.22 - - 2.35 - 

Acourtia runcinata Asteraceae 1 - - - - - - 0.67 

Allionia incarnata Nyctaginaceae 3 - - - - 0.34 8.82 3.33 

Aloysia macrostachya Verbenaceae X 3 - 2.42 - 13.33 - 0.59 3.33 

Argythamnia humilis var. 

humilis 

Euphorbiaceae 5 0.80 - 0.74 2.22 0.69 1.18 0.67 

Aristida purpurea Poaceae 5 17.60 1.61 28.89 4.44 11.72 42.94 66.67 

Aristolochia sp. Aristolochiaceae 1 - - - - - - 0.67 

Astragalus nuttallianus Fabaceae 2 - - 0.74 2.22 - - - 

Astragalus sp. 2 Fabaceae 1 - - - - 0.34 - - 

Ayenia pilosa Sterculiaceae 5 - - 2.96 2.22 0.34 4.12 2.67 

Bahia absinthifolia Asteraceae 1 - - - - 0.69 - - 

Bothriochloa laguroides 

ssp. torreyana 

Poaceae 1 - - - - - 1.18 - 

Bouteloua repens Poaceae 1 - - - - - - 0.67 

Bouteloua trifida Poaceae 5 16.80 - 8.15 1.48 9.66 8.24 2.67 

Cardiospermum dissectum 

Sapindaceae 1 - - - - - 16.47 - 

Celtis ehrenbergiana Ulmaceae X 2 - - 0.74 - 1.03 - - 

Celtis laevigata Ulmaceae 1 - - - - - 0.59 - 

Cenchrus spinifex Poaceae 2 - - 2.22 - 0.69 - - 

Cevallia sinuata Loasaceae 1 - - - - 3.45 - - 

Chaetopappa bellioides 

and Aphanostephus skirrhobasis 

var. kidderi 

Chamaesaracha sordida 

and Physalis cinerascens 

Asteraceae 5 1.60 - 19.26 3.70 50.00 4.12 14.67 

Solanaceae 3 - - - 4.44 26.55 - 8.67 

177




m 

tall? 

# 

Sites 

w/sp a 

CL02 

cl 


CL02 

un 

CL07 

cl 

CL07 

un 


Chamaesyce laredana and 

C. cinerascens 

Euphorbiaceae 4 24.00 14.52 5.93 11.85 6.55 - 6.00 

Cissus trifoliata Vitaceae 2 - - - 0.74 0.69 - - 

Citharexylum brachyanthum 

Verbenaceae X 4 0.80 0.81 1.48 2.96 2.07 0.59 - 

Commelina erecta Commelinaceae 4 - - - 1.48 0.69 0.59 2.67 

Condalia spathulata Rhamnaceae 1 - - - - 0.34 - - 

Convolvulus equitans Convolvulaceae 3 - - 0.74 0.74 0.69 - - 

Cooperia sp. Liliaceae 2 - - - - - 1.76 1.33 

Croton incanus Euphorbiaceae X 1 - - - - - 12.94 - 

Cynanchum barbigerum Asclepiadaceae X 4 - - 1.48 2.96 1.72 2.94 - 

Cyperus sp. Cyperaceae 5 0.80 - 2.96 0.74 6.21 1.18 1.33 

Dalea nana Fabaceae 2 4.00 - 8.89 - - 0.59 - 

Dalea pogonathera Fabaceae 1 - - - - 0.34 - - 

Digitaria cognata Poaceae 4 - - 2.96 - 0.34 7.65 16.00 

Diospyros texana Ebenaceae X 4 0.80 - 0.74 - 3.45 2.35 4.00 

Echinocactus texensis Cactaceae 1 - - - - - 0.59 - 

Echinocereus enneacanthus 

Cactaceae 3 - - - - 0.34 1.76 0.67 

Echinocereus poselgeri Cactaceae 4 - 0.81 - 1.48 0.34 1.76 0.67 

Echinocereus reichenbachii 

ssp. fitchii 

Cactaceae 1 - - - - - - 0.67 

Ephedra antisyphilitica Ephedraceae X 3 0.80 - 0.74 - - 18.24 4.00 

Eragrostis curtipedicellata 

Poaeae 3 - - 0.74 0.74 3.79 - - 

Eriogonum greggii Polygonaceae 1 - - - - - - 26.67 

Erioneuron pilosum Poaceae 2 - - - - - 4.12 0.67 

Escobaria emskoetteriana Cactaceae 2 - - - - - 0.59 0.67 

Evolvulus alsinoides Convolvulaceae 3 - - 8.15 9.63 8.97 - - 

Eysenhardtia texana Fabaceae X 5 - - 2.22 1.48 6.21 7.65 4.67 

178




m 

tall? 

# 

Sites 

w/sp a 

CL02 

cl 


CL02 

un 

CL07 

cl 

CL07 

un 


Ferocactus hamatacanthus 

var. sinuatus 

Cactaceae 2 - - - - - 0.59 2.00 

Fleischmannia incarnata Asteraceae 1 - - - - 0.34 - - 

Florestina tripteris Asteraceae 1 - - - - 0.34 - - 

Forestiera angustifolia Oleaceae X 5 0.80 8.06 2.22 8.15 23.10 4.71 6.00 

Galium sp. Rubiaceae 4 - - 5.93 4.44 1.38 - 9.33 

Galphimia angustifolia Malphigiaceae 4 - - 2.22 1.48 3.79 - 2.00 

Gamochaeta pensylvanica Asteraceae 1 - - - - 1.03 - - 

Gilia incisa Polemoniaceae 1 - - 2.22 - - - - 

Grusonia schottii Cactaceae 1 0.80 - - - - - - 

Guaiacum angustifolium Zygophyllaceae 4 3.20 2.42 2.96 1.48 - 3.53 1.33 

Heliotropium confertifolium 

Heliotropium curassavicum 

Boraginaceae 2 - - - - 1.03 - 3.33 

Boraginaceae 1 - - - - 0.34 - - 

Herissantia crispa Malvaceae 2 - - - - 0.69 1.18 - 

Heterotheca sp. Asteraceae 1 - - - - 0.34 - - 

Hibiscus martianus Malvaceae 3 - - 0.74 0.74 - 0.59 - 

Ibervillea lindheimeri Cucurbitaceae 2 - - - - 0.34 - 0.67 

Jatropha dioica Eupohorbiaceae X 5 4.80 3.23 5.93 7.41 11.38 5.88 3.33 

Jefea brevifolia Asteraceae 2 0.80 - - - - - 1.33 

Justicia pilosella Acanthaceae 2 - - - - 0.34 1.18 - 

Karwinskia humboldtiana Rhamnaceae X 5 1.60 5.65 4.44 9.63 7.93 12.35 16.00 

Koeberlinia spinosa Capparaceae X 1 - 0.81 - - - - - 

Krameria ramosissima Krameriaceae 5 8.00 4.84 7.41 7.41 3.45 16.47 10.00 

Lantana achyranthifolia Verbenaceae 1 - - - - 6.90 - - 

Lantana urticoides Verbenaceae 2 - - - 1.48 1.72 - - 

Lepidium lasiocarpum 

var. wrightii 

Brassicaceae 1 - - - - 2.76 - - 

Leucophyllum frutescens Scrophulariaceae X 5 30.40 52.42 62.96 72.59 13.79 11.76 11.33 

179




m 

tall? 

# 

Sites 

w/sp a 

CL02 

cl 


CL02 

un 

CL07 

cl 

CL07 

un 


Linum lundellii Linaceae 5 - - 14.07 10.37 2.07 1.18 2.67 

Lippia graveolens Verbenaceae X 5 6.40 12.10 13.33 19.26 3.10 22.94 8.00 

Lupinus texensis Fabaceae 1 - - - - - - 1.33 

Lycium berlandieri Solanaceae 2 - - - - 0.34 0.59 - 

Macrosiphonia lanuginosa 

Apocynaceae 1 - - - - - - 0.67 

Mammillaria heyderi Cactaceae 1 - - - - - 2.35 - 

Mammillaria sphaerica Cactaceae 1 - - - - - 0.59 - 

Manfreda longiflora Agavaceae 1 - - - - - 0.59 - 

Maurandya antirrhiniflora 

Scrophulariaceae X 1 - - - - 6.55 - - 

Melampodium cinereum Asteraceae 5 25.60 9.68 29.63 27.41 23.79 3.53 4.67 

Melinis repens Poaceae 1 - - - - 0.34 - - 

Mimosa texana 

(M. wherryana) 

Fabaceae X 5 8.80 13.71 13.33 13.33 11.72 5.29 12.67 

Nama hispidum Hydrophyllaceae 4 - - 20.74 28.89 41.38 - 1.33 

Oenothera laciniata Onagraceae 3 - - 8.89 7.41 10.34 - - 

Opuntia engelmannii Cactaceae X 4 0.80 - 1.48 0.74 2.41 2.94 - 

Opuntia leptocaulis Cactaceae X 4 - 1.61 - 1.48 0.34 1.18 0.67 

Opuntia sp. Cactaceae 1 - - - - - 0.59 - 

Oxalis dichondrifolia Oxalidaceae 3 - - 5.19 2.22 1.38 - - 

Palafoxia texana Asteraceae 2 - - 1.48 - 7.24 - - 

Panicum hallii Poaceae 1 - - - - 0.69 - - 

Parietaria pensylvanica Urticaceae 4 - - 3.70 1.48 9.66 - 1.33 

Parkinsonia texana var. 

texana (Cercidium texanum) 

Fabaceae X 4 1.60 3.23 4.44 2.96 3.10 0.59 - 

Parthenium confertum Asteraceae 3 4.00 2.42 2.96 3.70 22.76 - - 

Paspalum setaceum Poaceae 1 - - - - 0.34 - - 

Passiflora tenuiloba Passifloraceae 2 - - - - 0.34 2.35 - 

180




m 

tall? 

# 

Sites 

w/sp a 

CL02 

cl 


CL02 

un 

CL07 

cl 

CL07 

un 


Paysonia lasiocarpa Brassicaceae 1 - - - - 1.72 - 

Pennisetum ciliare Poaceae 3 - - 0.74 - 11.38 9.41 - 

Phacelia congesta Hydrophyllaceae 1 - - - - 2.07 - - 

Phemeranthus aurantiacus 

Portulacaceae 2 - - - 0.74 0.34 - - 

Phyllanthus polygonoides Euphorbiaceae 5 - - 21.48 9.63 8.97 7.06 4.67 

Physaria thamnophila Brassicaceae 5 20.80 1.61 22.22 5.93 24.14 25.29 17.33 

Plantago hookeriana Plantaginaceae 3 - 0.81 2.96 2.22 9.66 - - 

Polygala lindheimeri Polygalaceae 5 30.40 50.00 24.44 48.15 5.52 20.59 14.67 

Portulaca sp. Portulaceae 1 - - - - 1.72 - - 

Rivina humilis Phytolaceae 1 - - - - - 1.76 - 

Salvia ballotiflora Lamiaceae 1 - - - - 0.69 - - 

Schaefferia cuneifolia Celastraceae X 3 2.40 0.81 2.22 - 0.34 - - 

Senna bauhinioides Fabaceae 3 - - 1.48 2.96 0.69 - - 

Setaria leucopila Poaceae 4 0.80 - - - 1.03 1.76 0.67 

Setaria ramiseta Poaceae 2 - - 6.67 4.44 - - - 

Setaria texana Poaceae 1 - - - - 7.93 - - 

Sida abutifolia Malvaceae 3 - - 7.41 6.67 4.48 - - 

Sideroxylon celastrinum Sapotaceae X 3 - - - - 0.34 11.76 4.67 

Sonchus oleraceus Asteraceae 2 - - 1.48 0.74 - - - 

Spermolepis echinata Apiaceae 1 - - - - 2.07 - - 

Sporobolus cryptandrus Poaceae 5 6.40 - 5.19 1.48 1.72 5.88 3.33 

Synthlipsis greggii Brassicaceae 3 - - - - 4.14 2.94 10.00 

Tetraclea coulteri Verbenaceae 3 - - 2.96 0.74 2.41 - - 

Thamnosma texana Rutaceae 5 28.80 11.29 39.26 28.15 6.90 10.00 7.33 

Thymophylla pentachaeta Asteraceae 5 38.40 3.23 71.11 54.07 50.00 11.18 5.33 

Tiquilia canescens Boraginaceae 4 21.60 6.45 13.33 11.85 6.55 0.59 - 

Tridens muticus Poaceae 5 18.40 1.61 7.41 0.74 2.07 12.94 14.67 

Turnera diffusa Turneraceae 1 - - - - - 58.82 - 

181




m 

tall? 

# 

Sites 

w/sp a 

CL02 

cl 


CL02 

un 

CL07 

cl 

CL07 

un 


Unidentified cactus seedling 

Cactaceae 1 - - - - - - 0.67 

Unknown grass Poaceae 5 8.00 0.81 25.93 24.44 21.03 11.18 1.33 

Unknown legume Fabaceae 1 - - - 0.74 - - - 

Unknown 1 1 1.60 - - - - - - 

Unknown 2 2 3.20 2.42 0.74 1.48 - - - 

Unknown 3 1 1.60 - - - - - - 

Unknown 4 1 0.80 - - - - - - 

Unknown 5 5 2.40 1.61 4.44 7.41 0.69 5.29 0.67 

Unknown 6 1 - - - - - - 1.33 

Unknown 7 1 - - - - - 1.76 - 

Urochloa ciliatissima Poaceae 1 - - - - 1.03 - - 

Urochloa texana Poaceae 1 - - 0.74 - - - - 

Verbena sp 1. Verbenaceae 2 - - 2.22 - 2.76 - - 

Verbena sp 2. Verbenaceae 3 - - 2.22 0.74 1.72 - - 

Wedelia texana (A. Gray) 

B.L. Turner 

Asteraceae 5 2.40 - 19.26 0.74 4.48 0.59 0.67 

Yucca treculeana Agavaceae X 2 - - - 0.74 0.34 - - 

Zanthoxylum fagara Rutaceae X 1 - - - - 1.03 - - 

Ziziphus obtusifolia Rhamnaceae 4 - - 0.74 - 0.69 0.59 0.67 

Bare ground (no plants in 

subplot) 

5 0.80 7.26 - - 0.34 4.12 3.33 

a 

Number of sites with this species: Cuellar cleared and uncleared counted as different ‘sites’ due to 

treatment difference 

b Site codes: CL cl = Cuellar cleared; CL un = Cuellar uncleared; AM = Arroyo Morteros; AR = Arroyo 

Ramirez; SM = Santa Margarita. Numbers 02-07 refer to year of vegetation data collection. 

182


RESULTS 

Differences in Vegetation Among Sites 

A total of 150 vascular plant species occurred in at 

least one vegetation plot in a least one site (Table 1). 

Sixty-six of these species occurred in only one of the 

four sites (pooling Cuellar treatments). While the vegetation 

of the four sites was similar in many ways, the 

relative abundances of species differed among sites 

(Table 1). In Arroyo Morteros, Thymophylla pentachaeta, 

Acacia rigidula, and Chaetopappa bellioides/ 

Aphanostephus skirrhobasis var. kidderi were most 

abundant; in Arroyo Ramirez, Aristida purpurea, 

Turnera diffusa, Acacia rigidula, and Polygala lindheimeri; 

in Santa Margarita, Aristida purpurea and 

Eriogonum greggii; and in Cuellar, Acacia rigidula, 

Leucophyllum frutescens, Polygala lindheimeri, and 

Melampodium cinereum. Sites also differed in the abundances 

of less common species (Table 1). A MANOVA 

(multivariate analysis of variance) comparing the abundances 

of all species (except P. thamnophila and ‘bare 

ground’, for a total of 146 ‘species’ after the pooling 

described in the Methods) among the five site x treatment 

combinations (Cuellar treatments not pooled) was 

highly significant (Cuellar 2007 vegetation data used; 

Hotelling-Lawley Trace, F = 9.44, df = 548, 114.71, 

P < 0.0001). 

The four sites also differed in vegetation structure 

(Figure 5). The lack of vegetation above 1m in the 

cleared portion of Cuellar in 2002 was due to its treatment, 

while the uncleared portion of Cuellar had the 

densest canopies. The regrowth in the Cuellar cleared 

portion is apparent in a comparison of the 2002 and 

2007 graphs (Figure 5). Among uncleared sites, Arroyo 

Ramirez had the most open canopies, and Arroyo 

Morteros the tallest. 

Relationships Between P. thamnophila Presence and 

Neighboring Species 

Twenty-nine species, including P. thamnophila, were 

present in 25 percent or more of the subplots. These species, 

except for P. thamnophila, were used as the dependent 

variables in a MANOVA. In this MANOVA, 

site and P. thamnophila presence/absence in the 2007 

census were the independent variables. Cuellar 2007 

vegetation data were used in this analysis and the two 

Cuellar treatments were considered to be 2 different 

‘sites’, for a total of 5 site-treatment combinations. Both 

of the independent variables were highly significant 

(site-treatment combination: Hotelling-Lawley Trace F 

= 20.67; df = 112,476; P < 0.0001; presence/absence: 

Hotelling-Lawley Trace F = 1.93; df = 28,143; P = 

0.0066). 

Because the MANOVA found significant effects of 

P. thamnophila presence we followed it with ANOVAs 

(univariate analyses of variance). Six of 28 ANOVAs on 

the same 28 species had P-values less than 0.05 associated 

with the effect of P. thamnophila presence/absence. 

In each of these ANOVAs, presence/absence, df = 1, 

was the second factor in a hierarchical sums of squares 

table that had site, df = 3, as the first factor. Thus these 

P-values reflect the amount of variation that presence/ 

absence accounted for above-and-beyond the variation 

accounted for by site-year combination. Plots that had 

P. thamnophila had on average more Diospyros texana 

(P = 0.0424), Acacia rigidula (P = 0.0043), unidentified 

seedling grasses (P = 0.0068), Mimosa texana (P = 

0.0012), and Chamaesaracha sordida + Physalis cinerascens 

(P = 0.0293), and less Tiquilia canescens (P = 

0.0293), than plots without P. thamnophila. These positive 

and negative associations with P. thamnophila 

should be regarded as tentative, due to multiple testing 

issues, deviations from the multivariate normal assumption, 

and correlations among abundances of the different 

species. Only Mimosa texana meets the Bonferroni criterion 

for significance (for 28 tests, P < 0.00183 to give 

an overall P < 0.05). 

Relationships Between P. thamnophila Density and 

Neighboring Species 

A stepwise regression was used (SAS PROC REG) 

with the average 2006/2007 density of P. thamnophila 

as the dependent variable. Site was included as a class 

variable (df = 3). All 148 ‘species’ except bare ground 

and P. thamnophila were available to the regression procedure. 

A criterion of P < 0.05 was used to retain variables 

in the model. Eighteen taxa met this criterion. 

Thirteen of them were negatively associated with P. 

thamnophila: Acleisanthes obtusa, Acourtia runcinata, 

Citharexylum brachyanthum, Thymophylla pentachaeta, 

Echinocactus texensis, Echinocereus reichenbachii ssp. 

fitchii, Galphimia angustifolia, Nama hispidum, Oenothera 

laciniata, Portulaca sp., Sideroxylon celastrinum, 

Synthlipsis greggii, and Tridens muticus. Five were 

positively associated with P. thamnophila: Acacia rigidula, 

Cyperus sp., Evolvulus alsinoides, Melampodium 

cinereum, and Passiflora tenuiloba. 

DISCUSSION 

Community Composition: General 

Each of the study sites had a rich plant community 

that contained shrubs, forbs, graminoids, and cacti 

(Table 1). The shrubs Acacia rigidula and Leucophyllum 

frutescens dominated all four sites. While all four 

study sites were similar enough to be described as having 

the same plant community, there were some interesting 

differences among them. These include Turnera diffusa, 

found in 59 percent of samples at Arroyo Ramirez 

but absent from other sites; Eriogonum greggii, itself a 

183


Figure 5: Canopy height structure at Physaria thamnophila 

sites. 

184


somewhat rare plant, found in 27 percent of subplots at 

Santa Margarita only; and the shrub Forestiera angustifolia, 

which occurred in 23 percent of subplots at Arroyo 

Morteros, but was infrequent at other sites. Since 

there was no evidence that P. thamnophila was declining 

at any of the sites, each of these slightly different 

plant communities is evidently suitable for P. thamnophila. 

Community Composition: Comparisons with Other 

Studies 

The plant community of our study sites, quantified in 

Table 1, does not match previously identified plant communities 

of the region. It also differs in some important 

details from previously published lists of the plant species 

associated with Physaria thamnophila. These differences 

may be attributed to site selection, scale of observation, 

method of community description (sampling 

vs. observation of visual dominants), and, for herbaceous 

species, year effects. 

The Recovery Plan (USFWS 2004) stated that P. 

thamnophila occurs in “an open Leucophyllum frutescens 

(cenizo) - Acacia berlandieri (guajillo) shrubland 

alliance” (Acacia rigidula - Leucophyllum frutescens - 

Acacia berlandieri Shrubland Alliance; NatureServe 

2009). The plant community we observed may be a 

component association of this alliance. However, A. 

berlanderi was rarely encountered in our study. Among 

the plant associations described by NatureServe (2009), 

the best match to the community we observed is the 

Acacia rigidula - Leucophyllum frutescens - Acacia berlandieri 

Shrubland Association (CEGL007759). However, 

this “broadly defined type” is not described in sufficient 

detail to make it useful in defining P. thamnophila 

habitat. Our sites somewhat resembled the Acacia 

rigidula - Leucophyllum frutescens - Hechtia glomerata 

Shrubland Association (CEGL007760; NatureServe 

2009). However, this association occurs on saline clay 

soils. Furthermore, Hechtia glomerata, while observed 

in the vicinity of P. thamnophila sites, was not encountered 

in study plots. The more general Acacia rigidula 

Shrubland Association (CEGL003874) is too broadly 

defined to be useful in identifying P. thamnophila habitat. 

Jahrsdoerfer and Leslie (1988) described a “Chihuahuan 

(or Falcon) thorn forest” community in this 

area, but listed only two dominant species: A. rigidula 

and Mimosa biuncifera. M. biuncifera does not occur in 

this area, but the name has been misapplied to M. texana; 

and thus, it is likely they were referring to M. texana. 

Other plant community classifications for South 

Texas, including McLendon (1991) and Bezanson 

(2000), describe associations similar to NatureServe’s. 

These communities include species that were not important 

in our study, and occur on different substrates. 

NRCS Ecological Site Descriptions also list different 

species mixes than those we encountered. Copita soils 

belong to the Gray Sandy Loam Ecological Site; Catarina, 

to the Saline Clay Ecological Site, and Mavericks 

to the Rolling Hardland Ecological Site. The plant communities 

of these ecological sites are described as having 

more mid-grasses and different forbs than our study 

sites, as well as a different mix of dominant shrubs. 

Jimenez and Quemado soils belong to the undescribed 

Gravelly Ridge Ecological Site, while their associated 

“unnamed, minor components” have not been assigned 

to an ecological site. Zapata soils are in the undescribed 

Shallow Ridge Ecological Site (NRCS 2009) 

Wu and Smeins (1999) listed sixteen woody species 

associated with P. thamnophila, including A. rigidula, 

L. frutescens, Karwinskia humboldtiana, Krameria 

ramosissima, and Jatropha dioica, all of which were 

frequently encountered in our plots. However, they also 

listed species that we did not find to be closely associated 

with Zapata bladderpod, including Prosopis glandulosa, 

Hechtia glomerata, Acacia berlandieri, and A. 

greggii. They noted a diverse herbaceous understory but 

did not list species. Although their report does not describe 

the method for characterizing vegetation, we suspect 

that this study reported visual dominants or species 

common in the general area, but not necessarily associated 

with P. thamnophila at a finer scale. 

Sternberg (2005) listed 22 woody and herbaceous 

species associated with P. thamnophila at Cuellar. Even 

though his sampling method covered the entire population 

area and ours focused on the cleared area and adjacent 

uncleared habitat, most of the frequently encountered 

species were important in both studies. However, 

there were some notable differences. For example, 

Sternberg did not report the woody species Mimosa texana, 

nor did he encounter several herbaceous species 

that we observed in 10 percent or more of subplots. 

Among the important herbaceous species that Sternberg 

did not observe, a few were most abundant in the 

cleared portion of Cuellar (Tridens muticus, Aristida 

purpurea), while others were only identified in 2007, 

following a relatively wet winter (for example, Phyllanthus 

polygonoides and the annuals Nama hispidum and 

Linum lundellii). 

Relationships Between P. thamnophila and Individual 

Plant Species 

A number of species that were common on upland 

soils in the region were absent or uncommon in our 

study sites. Acacia berlandieri, Prosopis glandulosa, 

Ziziphus obtusifolia, Zanthoxylum fagara, Cordia boissieri, 

and Lycium berlandieri are common members of 

upland plant communities in the region (Bezanson 2000; 

NatureServe 2009; NRCS 2009; USFWS 2004), but 

were absent or uncommon in our study sites. A. ber- 

185

landeri, cited as a community dominant in the Recovery 

Plan (USFWS 2004), appeared in the vegetation data in 

only one of our four study sites. This species is a dominant 

at Santa Margarita immediately upslope of the P. 

thamnophila population that is on a different soil type. 

Prosopis glandulosa never occurred within our study 

plots. The inclusion of Prosopis in the Recovery Plan 

was based on a site that we did not study, an abandoned 

trailer park which has subsequently been overtaken by 

invasive buffelgrass (Pennisetum ciliare), leaving only a 

small remnant on highway right-of-way (J.M. Poole, 

Texas Parks and Wildlife Department, personal communication 

20 July 2009). Other species mentioned in the 

Recovery Plan that we encountered infrequently include 

Celtis ehrenbergiana, Yucca treculeana, Ziziphus obtusifolia, 

and Guaiacum angustifolium, which are common 

visual dominants in the region that were not closely 

associated with P. thamnophila at our sites. These species 

may be better suited to sites with deeper soils or 

more favorable soil chemistry. 

Two common invasive grasses of the area, buffelgrass 

(Pennisetum ciliare (L.) Link) and Kleberg bluestem 

(Dichanthium annulatum (Forssk.) Stapf), are also 

conspicuous by their absences or low abundances in the 

four study sites. It seems likely that P. thamnophila, like 

other native grasses and forbs of south Texas (Sands et 

al. 2009), is out-competed by these invasive grasses. 

The plots in which P. ciliare was encountered should be 

monitored to determine its effect on P. thamnophila. 

Associated species’ negative or positive correlation 

with P. thamnophila may have a number of explanations. 

Some of the negative correlations may reflect very 

localized competition. For example, a plot in which 

Thymophylla pentachaeta or Nama hispidum were very 

abundant (especially in 2007) may have been colonized 

by these species in response to winter rain, and they in 

turn excluded P. thamnophila. Some negative correlations 

may reflect microsites not suitable for P. thamnophila, 

such as hardpan, where Tiquilia canescens was 

relatively common. Synthlipsis greggii seems to use different 

microsites and topographic positions than P. 

thamnophila; however, we did not quantify this observation. 

Acacia rigidula’s positive correlation with P. thamnophila 

may indicate facilitation, probably because as a 

legume it may create a soil patch with relatively high 

nitrogen content. Alternatively, its presence may indicate 

that the plot is not bare hardpan, but rather is favorable 

to vegetation in general. The other shrub that was 

positively correlated with P. thamnophila was the legume 

Mimosa texana. This shrub, although not rare, has 

a restricted range and may be more indicative of P. 

thamnophila habitat. A “characteristic species of the 

arid, sandy-soil Falcon Woodlands, which cover a small 

upland part of Starr and Zapata Counties” (Ideker 1999), 


186 

M. texana was present at all study sites in 5 to14 percent 

of subplots. 

The positive correlation of P. thamnophila with perennial 

herbaceous species such as Melampodium 

cinereum and Evolvulus alsinoides is probably related to 

its similar microhabitat requirements and response to 

precipitation and shrubs. Other species with significant 

correlations were present in low frequency and are inconclusive. 

Edaphic Requirements of P. thamnophila 

Although P. thamnophila populations are mapped on 

several soil series, our observations in the field indicate 

very similar soils and geologic substrates at all sites. All 

four populations of P. thamnophila occur on a sandstone 

substrate (Figure 6), on yellowish, highly erodible, 

highly calcareous soils. All other Texas populations to 

which we and other observers have had access have 

similar yellowish sandy soils and occur on sandstone. 

Wu and Smeins (1999) report Copita and Zapata sandy 

loam soils as the substrate for P. thamnophila, and clarified 

that sites mapped as Catarina soils (saline, gypsiferous 

clay) are actually on sandy inclusions. Their analyses 

of soil from four P. thamnophila sites found very 

high calcium, high sulfur, and very low nitrogen levels. 

We believe that use of NRCS digital soil maps at a 

level of detail beyond which they were intended has led 

to confusion about the soils on which P. thamnophila 

occurs. P. thamnophila has never been found on 

Jimenez-Quemado soils (contra Poole 1989 and 

USFWS 2004). The parent material of these soils is 

gravelly alluvium, deposited by ancient, high-velocity 

streams on the high terraces over the Rio Grande 

(Thompson et al. 1972). The Jimenez-Quemado soil 

polygons contain inclusions of “unnamed, minor components” 

and rock outcrops. These outcrops, rather than 

Jimenez or Quemado soils, are likely habitat for P. 

thamnophila. 

Figure 6: Sandstone substrate with Physaria thamnophila 

plant.


The underlying geology at the sites is complex. The 

Yegua and Jackson Formations were deposited in part 

of the Gulf Coast geosyncline known as the Rio Grande 

Embayment during Eocene cycles of sedimentation. 

Land subsidence and marine transgressions and regressions 

produced fluctuating sea levels and deposition of 

“complexly interbedded sands, silts, and clays” as well 

as marine shales (Preston 2009). Dumble (1902) provided 

a detailed site-specific description of this complex 

stratigraphy. Reporting on outcrops of “buff sandstone” 

near Roma (in the vicinity of our study sites), he describes 

sections of “greenish-yellow clays” with gypsum, 

oyster beds, buff clays, sandy clays, and indurated 

sandstone. 

At two of our study sites that were on bluffs along 

the Rio Grande, two layers of buff to yellowish sandstone 

were interspersed with other substrates or oyster 

shell deposits. Fossil oyster shell beds occurred upslope 

from all four study sites. Gypsum crystals occurred on 

the soil surface in many places. The alternating layers of 

sandstone with fossil oyster shell, shale and clay may 

explain the presence of gypsum at the sites, even though 

the sandy soils (Copita and Zapata) are only weakly 

gypsiferous. The layers of different substrates may create 

microsites where water is more available due to 

seepage from relatively permeable layers located over 

impermeable or less permeable layers. Several species 

we encountered in this study are members of genera that 

are documented as tolerating gypsum, particularly 

Tiquilia, but also Nama, Eriogonum, and Acleisanthes 

(Moore and Jansen 2007). We have no evidence that P. 

thamnophila is a gypsum endemic; however, it tolerates 

gypsum. 

All four sites were undergoing active gully erosion 

to the degree that some slopes could be called badlands. 

All four sites also appeared to have high rates of sheet 

erosion as well, especially in the bare areas between 

shrubs. While erosion probably does not directly benefit 

P. thamnophila, high erosion rates likely reduce the 

number of competing species that can live in the site, 

and perhaps their densities. High erosion rates may be 

one factor contributing to the positive association between 

shrubs and P. thamnophila, especially P. thamnophila 

seedlings within our sites (Fowler et al. 2011). By 

slowing the rate of erosion in their immediate vicinity, 

shrubs may increase seedling survival there. High erosion 

rates may also explain why roller-chopping part of 

Cuellar increased the density of P. thamnophila there 

(Fowler et al. 2011), as the woody debris left by this 

treatment may also have reduced the erosion rate. However, 

we cannot exclude other positive effects that 

shrubs may have upon P. thamnophila. 

These edaphic features (high erosion rates, highly 

calcareous soils, perhaps the presence of gypsum) could 

be used to search for sites where additional populations 

might occur. They also provide some guidance for identifying 

sites suitable for introduction or reintroduction. 

Although we do not believe that P. thamnophila requires 

high soil erosion rates, the presence of gypsum, 

or even calcareous soils, all of these probably reduce the 

number and density of competing species. Ecologically, 

P. thamnophila is apparently a stress-tolerator (sensu 

Grime 1977, 2001), rather than a strong competitor or a 

ruderal species. This may also be true of many of its 

associates. 

Conservation Applications 

The Recovery Plan (USFWS 2004) called for (a) 

identification of sites where additional populations 

might occur; (b) identification of sites most suitable for 

attempts to establish new populations; (c) identification 

of tracts most appropriate for mitigation purposes, 

should that be necessary; (d) development of management 

plans; and (e) development of habitat restoration 

objectives. The vegetation structure and composition 

data presented here, especially that of Table 1 and Figure 

5, provide quantitative objectives for each of these 

tasks. Whereas visually dominant species mentioned in 

previous work can still be used to search for general 

areas of suitable native thornscrub vegetation, the species 

composition presented here provides a finer scale 

focus for identification of suitable habitat. The qualitative 

description of likely soil types provided above 

could be used to guide searches for new populations and 

identify sites for introduction or reintroduction. It would 

also be helpful to know the gypsum content of the soils. 

Management and restoration projects can use Table 1 

and Figure 5 to help set quantitative objectives. 

It should be noted that this study was not designed to 

compare sites with and without P. thamnophila, so results 

do not definitively identify what it is about these 

four sites that made them different from similar sites in 

the region. Additionally, we did not attempt to characterize 

small, remnant, disturbed sites, which are also 

important to the species’ conservation. In any case, with 

so few P. thamnophila populations, the absence of P. 

thamnophila from any particular site is hard to interpret. 

P. thamnophila may be absent from suitable sites due to 

chance, poor dispersal, disease, herbivory, or other factors, 

which is often a problem in studying endangered 

species (Hanski and Ovaskainen 2002). Experienced 

biologists will no doubt make their own judgments from 

the data we provide. 


We thank the Lower Rio Grande Wildlife Refuge 

staff for their assistance; landowner Jorge Gonzales and 

family for access to Santa Margarita Ranch; and Tom 

and Elena Patterson, Thomas Adams, Robyn Cobb, 

Loretta Pressley, Jim Manhart, and Alan Pepper for 

187


assistance with field work. Cullen Hanks reviewed the 

status of Physaria populations and the map, and Alan 

Pepper, Karen Clary, Jackie Poole, and Martin Terry 

reviewed the manuscript. Their comments have greatly 

improved this work. 


Al-Shehbaz, I.A. and S.L. O’Kane Jr. 2002. 

Lesquerella is united with Physaria (Brassicaceae). Novon 

12:319-329. 

Bezanson, D. 2000. Natural vegetation types of 

Texas and their representation in conservation areas. 

M.A. thesis, University of Texas, Austin. [Online]. 

Available: http://abisw.org/bezanson/ [June 1, 2009]. 

Bureau of Economic Geology. 1976. Geologic atlas 

of Texas, McAllen-Brownsville sheet. University of 

Texas, Austin, TX. 

Dumble, E.T. 1902. Geology of southwestern Texas. 

Transactions of the American Institute of Mining Engineers 

33:913-987. 

Fowler, N.L., C.F. Best, D.M. Price, and A.L. Hempel. 

2011. Ecological requirements of the Zapata bladderpod 

Physaria thamnophila, an endangered 

Tamaulipan thornscrub plant. Southwestern Naturalist 

56(3):341-352. 

Grime, J. P. 1977. Evidence for the existence of three 

primary strategies in plants and its relevance to ecological 

and evolutionary theory. American Naturalist 

111:1169-1194. 

Grime, J. P. 2001. Plant strategies, vegetation processes, 

and ecosystem properties, 2nd ed. John Wiley and 

Sons, New York, NY. 456 p. 

Hanski, I., and O. Ovaskainen. 2002. Extinction debt 

at extinction thresholds. Conservation Biology 16:666- 

673. 

Ideker, J. 1999. Wherry mimosa and black mangrove: 

Natives. Native Plant Society of Texas News 17 

(5):8 [Online]. Available: http://www.npsot.org/ 

NPSOTNews/ [June 8, 2009]. 

Jahrsdoerfer, S. E. & D. M. Leslie. 1988. Tamaulipan 

brushland of the Rio Grande Valley of south Texas: 

Description, human impacts, and management options. 

U.S. Fish and Wildlife Service, Biol. Rep. 88(36), xiii + 

63 p. 

McLendon, T. 1991. Preliminary description of the 

vegetation of south Texas exclusive of coastal saline 

zones. Texas Journal of Science 43:13-32. 

Moore, M. J. and R. K. Jansen. 2007. Origins and 

biogeography of gypsophily in the Chihuahuan Desert 

plant group Tiquilia subg. Eddya (Boraginaceae). Systematic 

Botany 32(2): 392–414. 

NatureServe. 2002. NatureServe Element Occurrence 

data standard. Version published February 6, 2002 

[Online]. Available: http://www.natureserve.org/ 

prodServices/eodata.jsp [June 5, 2009]. 


encyclopedia of life. Version 7.0. NatureServe, Arlington, 

Virginia. [Online]. Available: http://www.nature 

serve.org/explorer [June 5, 2009]. 

NOAA (National Oceanic and Atmospheric Administration). 

2009. National Climatic Data Center, Asheville, 

NC [Online]. Available: http://www.ncdc.noaa. 

gov/oa/ncdc.html [June 5, 2009]. 

NRCS (Natural Resources Conservation Service) 

Soil Survey Staff. 2009. United States Department of 

Agriculture. Web Soil Survey. [Online]. Available: 

http://websoilsurvey.nrcs.usda.gov/ [April 30, 2009]. 

Poole, J. M. 1989. Status report on Lesquerella thamnophila. 

Report prepared for U.S. Fish & Wildlife Service, 

Albuquerque. 21p. + figs. 

Poole, J. M., W. R. Carr, D. M. Price, and J. R. 

Singhurst. 2007. Rare Plants of Texas. Texas A&M 

University Press, College Station, Texas. 640p. 

Preston, R.D. 2009. The Yegua-Jackson aquifer. 

Chapter 3, p. 51-59 in: Mace, R.E. et al., eds. Aquifers 

of the Gulf Coast of Texas. Texas Water Development 

Board Report 365, Austin, Texas. [Online]. Available: 

http://www.twdb.state.tx.us/publications/reports/ 

GroundWaterReports/GWReports/GWreports.asp [June 

9, 2009]. 

Rollins, R.C. and E.A. Shaw. 1973. The genus 

Lesquerella (Cruciferae) in North America. Harvard 

University Press, Cambridge, Mass. 

Sands, J. P., L. A Brennan, F. Hernández, W. P. 

Kuvlesky, J. F. Gallagher, D. C. Ruthven, and J.E. 

Pittman. 2009. Impacts of buffelgrass (Pennisetum 

ciliare) on a forb community in South Texas. Invasive 

Plant Science and Management, 2(2):130-140. 

Sternberg, M. A. 2005. Zapata bladderpod 

(Lesquerella thamnophila Roll. & Shaw): its status and 

association with other plants. Texas Journal of Science 

57(1):75-86. 

Thompson, C.M., R.R. Sanders, and D. Williams. 

1972. Soil survey of Starr County, Texas. Soil Conservation 

Service, United States Department of Agriculture, 

in cooperation with the Texas Agricultural Experiment 

Station. 88pp. + figs. 

USDA NRCS. 2012. The PLANTS Database (http:// 

plants.usda.gov, 21 February 2012). National Plant 

Data Team, Greensboro, NC. 

USFWS (U.S. Fish and Wildlife Service). 2004. Zapata 

Bladderpod (Lesquerella thamnophila) Recovery 

Plan. U.S. Fish and Wildlife Service, Albuquerque, New 

Mexico. i-vii + 30 pp., Appendices A-B. [Online]. 

Available: http://ecos.fws.gov/speciesProfile/profile/ 

speciesProfile.action?spcode=Q13L [May 27, 2009]. 

Wu, X.B. and F.E. Smeins. 1999. Multiple-Scale 

Habitat Models of Rare Plants: Model Development and 

Evaluation. Report Submitted to Texas Department of 

Transportation, Pharr and Austin, TX. 

188


Intraspecific Cytotype Variation and Conservation: 

An Example from Phlox (Polemoniaceae) 

Shannon D. Fehlberg 

Desert Botanical Garden, Phoenix, AZ 

and Carolyn J. Ferguson, 

Herbarium and Division of Biology, Kansas State University, Manhattan, KS 

Abstract. Information on genetic structure in rare plants, such as patterns of genetic diversity, differentiation and 

gene flow, is useful when planning management strategies for conservation. However, few studies of genetic structure 

in rare plants include an investigation of intraspecific cytotype variation. A number of reviews have suggested 

that cytotype variation, or variation in chromosome structure or number, may be more widespread in natural populations 

than previously thought. A recent review of federally listed plant species found that 75% of species belonged to 

genera exhibiting both interspecific and intraspecific cytotype variation. Cytotype variation among populations of 

rare plants raises intriguing questions about the origin(s), extent, evolutionary relationships, and ecological differentiation 

of various cytotypes. While traditional cytological methods may not be feasible for population level sampling, 

advances in the accuracy and cost of flow cytometry now make examination of intraspecific cytotype variation possible. 

We initiated an investigation of cytotype variation in the genus Phlox (Polemoniaceae) using flow cytometry. 

Phlox comprises ca. 65 species in North America and includes many poorly studied endemic taxa in the western 

United States. Our results indicate noteworthy variation in chromosome numbers (ploidy level) across a subset of 

western taxa. A detailed study of two endemic species of conservation concern, P. amabilis and P. woodhousei, revealed 

that these species are made up of diploid, tetraploid and hexaploid populations. Ongoing analyses suggest that 

these populations are spatially, ecologically, and genetically differentiated. These results caution against the common 

assumption of homoploidy of species based on limited data and indicate the value of incorporating an understanding 

of cytotype variation into conservation biology studies. 

The interdisciplinary field of conservation biology 

aims to provide the knowledge and tools necessary for 

the long-term preservation of biodiversity. Studies of 

population genetics are one important source of information 

for conservation biology. These studies can provide 

basic information about populations of rare plants 

such as levels of genetic variation, the distribution of 

genetic variation within and among populations, the degree 

of population fragmentation and isolation, the patterns 

of historical and contemporary gene flow, the levels 

of inbreeding, the effective population size, the presence 

of taxonomic distinctiveness, and the occurrence of 

hybridization (Booy et al. 2000; Ellis and Burke 2007; 

Murray and Young 2001). Ultimately, this basic information 

can be combined with other sources of data to 

form more effective conservation strategies. 

One component of genetic variation that is often 

overlooked in studies of rare plants is intraspecific cytotype 

variation (De Lange et al. 2008; Severns and Liston 

2008). Intraspecific cytotype variation ranges from 

chromosomal inversions and translocations to variation 

in the number of whole genomes present (polyploidy). 

Recent reviews have suggested that intraspecific cytotype 

variation may be more widespread in natural plant 

populations than previously thought (Soltis et al. 2007; 

Suda et al. 2007). It may also be widespread in populations 

of rare, threatened and endangered plants. A survey 

of cytotype variation in 416 US federally listed 

plant taxa found that cytotype data were available for 

only 182 of these taxa (44%). Of these 182 taxa, 158 

belonged to genera with interspecific cytotype variation 

(87%; New World congeners with aneuploidy or polyploidy), 

and 121 belonged to genera with at least one 

taxon showing intraspecific cytotype variation (66%; 

Severns and Liston 2008). Intraspecific cytotype variation 

can be important when planning conservation 

strategies because it has the potential to affect patterns 

of genetic diversity and gene flow. Knowledge of such 

variation may also be important for clarifying taxonomy, 

identifying unique evolutionary lineages (perhaps 

accompanied by ecological differentiation), and determining 

appropriate populations for reintroduction or 

augmentation (De Lange et al. 2008; Murray and Young 

2001; Severns and Liston 2008). 

Recent advances in the application of flow cytometry 

to evolutionary and population biology studies now 

make the assessment of cytotype variation across large 

taxonomic and spatial scales possible (Kron et al. 2007). 

189


Flow cytometry is a method that quantifies nuclear 

DNA content by measuring the relative fluorescence of 

isolated nuclei that have been stained with a fluorescent 

dye. This method offers several advantages over traditional 

cytological methods including: 1) sample preparation 

and processing is relatively easy and rapid (allowing 

for large sample sizes), 2) sample material does not 

need to be actively dividing (a variety of tissues can be 

used, including dried and frozen material), and 3) sampling 

is relatively non-destructive (enabling studies of 

sensitive plants; Kron et al. 2007; Suda et al. 2007). 

Thus, flow cytometry provides a practical means by 

which intraspecific population level cytotype data can 

be gathered. 

The genus Phlox has been an important system for 

plant evolutionary studies of polyploidy, hybridization, 

and ecology (Ferguson and Jansen 2002; Ferguson et al. 

1999; Grant 1959; Levin and Schaal 1970; Levin and 

Smith 1966; Wherry 1955). Phlox comprises ca. 65 species 

of annual and perennial herbs distributed predominantly 

in North America with a center of diversity in the 

western United States. We are using flow cytometry to 

study cytotype variation across this genus. The present 

study focuses on a broad sample of western, upright perennial 

Phlox species with special emphasis on two endemic 

species of conservation concern found in coniferous 

forests in Arizona and New Mexico, P. amabilis and 

P. woodhousei. These two species are closely related 

(Ferguson et al. 2012) and very similar based on gross 

morphology, sharing a characteristic woody-based upright 

perennial growth form, thick linear leaves and 

notched petals. However, their geographic distributions 

do not overlap, and they are readily distinguished by 

differences in style length and stigma placement relative 

to anther position. These taxa have been variously classified 

as distinct species or conspecifics (Cronquist et al. 

1984; Wherry 1955). Our objectives for the present 

study were to examine 1) taxonomic and large-scale 

spatial patterns of cytotype variation across a subset of 

western Phlox taxa and 2) fine-scale spatial patterns of 

cytotype variation within and among populations of P. 

amabilis and P. woodhousei. 

METHODS 

The present study includes samples from seven species 

of upright, perennial Phlox from Arizona, California, 

Colorado, Idaho, Montana, Nevada, Utah and Wyoming 

(Table 1). For determination of cytotype, several 

leaves were collected from one to five individual plants 

at each sampling location for each taxon and stored on 

ice until nuclear extraction. Voucher specimens for each 

population were deposited at the Kansas State University 

Herbarium (KSC). 

We determined the cytotype of each sample of Phlox 

using flow cytometry. Flow cytometry measures nuclear 

DNA content, which can then be interpreted in terms of 

ploidy level, especially when closely related taxa are 

studied and when knowledge of ploidy level is independently 

assessed through conventional chromosome 

counts (Suda et al. 2007; Halverson et al. 2008). For 

each sample of Phlox, we placed 100-300 mg of chilled 

leaf tissue into a petri dish with 1.5 ml of chopping 

buffer, modified from Bino and others (1993) as described 

by Davison and others (2007). We chopped the 

leaves finely with a new razor blade and filtered the resulting 

liquid through a 30 µm filter into a microcentrifuge 

tube. Tubes were centrifuged at 500 x g for 7 min, 

the supernatant removed, the pellet re-suspended in 700 

µl propidium iodide staining solution (50 mg/ml; 

BioSure), and 2µl of chicken erythrocyte nuclei singlets 

added (CEN internal standard; BioSure). Samples were 

protected from light and stored on ice for at least 30 min 

before analysis on a Becton Dickinson FACS Calibur 

flow cytometer at the Kansas State University Flow Cytometry 

Facility. The amount of fluorescence was measured 

for ~10,000 nuclei per sample. Resulting histograms 

were visually inspected for the presence of clear 

nuclear populations from the Phlox sample and the CEN 

internal standard, and mean peak values were calculated 

using the program Cell Quest (Becton Dickinson). Nuclear 

DNA content was calculated as the Phlox sample 

mean peak value divided by the CEN internal standard 

mean peak value multiplied by the 2C-value of the CEN 

internal standard (2.5 pg; Dolezel and Bartos 2005). 

Ploidy level was inferred for each sample based on the 

calculated DNA content. Inferred ploidy level was 

linked with chromosome count data for several samples. 

RESULTS AND DISCUSSION 

A total of 140 samples from seven species of Phlox 

collected from 63 locations were assessed for cytotype 

variation (Table 1). When cytotype was measured in 

multiple individuals from a single location, average nuclear 

DNA content was calculated. We did not detect 

any cytotype variation within populations based on our 

limited sampling. The results from flow cytometry were 

interpreted as measures of ploidy level rather than absolute 

measures of DNA content (see Suda et al. 2007). 

Results from chromosome counts confirmed ploidy levels 

of 2x, 4x, and 6x (for five samples; Table 1). 

Cytotype varied throughout the species studied and 

appeared to reflect both taxonomy and geography 

(Table 1). Populations of P. caryophylla and P. cluteana 

were diploid, while populations of P. aculeata were 

tetraploid; all of these taxa are fairly narrow endemics. 

In general, populations of the wide-ranging P. longifolia 

and P. stansburyi were geographically structured, with 

190


Table 1. Samples included in this study. Voucher information, collection locality, number of individuals 

cytotyped (N), DNA content and inferred cytotype are noted. Taxon recognition for P. stansburyi follows 

Ferguson and others (2012); intraspecific classification for P. longifolia (including P. longifolia and P. viridis 

sensu Wherry [1955]) is under ongoing investigation and is not proposed here. 

Taxon Voucher County State Location Detail 

N 

DNA content 

(pg) 

P. aculeata Sidells 116 Twin Falls Co. ID 1 14.65 4X 

P. aculeata CF 789 Butte Co. ID 3 15.43 4X 

P. aculeata CF 787 Bonneville Co. ID 2 18.67 4X 

P. caryophylla CF 653 Archuleta Co. CO 1 7.61 2X 

P. cluteana CF 651 Apache Co. AZ 1 8.14 2X 

P. cluteana SDF 51008-2 Apache Co. AZ 3 8.84 2X 

P. longifolia SDF 42908-3 Navajo Co. AZ 3 8.37 2X 

P. longifolia SDF 43008-1 Coconino Co. AZ 3 8.60 2X 

P. longifolia SDF 42908-1 Navajo Co. AZ 3 8.74 2X 

P. longifolia SDF 51008-1 Apache Co. AZ 3 8.88 2X 



P. longifolia SDF 50708-1 Mojave Co. AZ 3 9.27 2X 



P. longifolia CF 728 Eagle Co. CO 1 15.25 4X 

P. longifolia CF 727 Mesa Co. CO 1 16.71 4X 

P. longifolia CF 788 Butte Co. ID 3 7.60 2X 

P. longifolia CF 806 Lemhi Co. ID 1 14.27 4X 

P. longifolia CF 603 Clark Co. ID 1 14.67 4X 

P. longifolia CF 792 Butte Co. ID 3 15.25 4X 

P. longifolia CF 589 Custer Co. ID 1 15.41 4X 

P. longifolia CF 794 Custer Co. ID 3 15.67 4X 

P. longifolia CF 786 Jefferson Co. ID 2 18.67 4X 

P. longifolia CF 601 Beaverhead Co. MT 1 13.63 4X 

P. longifolia CF 710 White Pine Co. NV 1 17.11 4X 

P. longifolia CF 724 Beaver Co. UT 1 8.38 2X 

P. longifolia CF 702 Sevier Co. UT 1 15.00 4X 

P. longifolia CF 611 Summit Co. UT 1 21.90 4X 

P. longifolia CF 734 Carbon Co. WY 1 13.98 4X a 

P. longifolia CF 215 Sublette Co. WY 1 14.65 4X 

Cytotype 

191




192 

N 

DNA content 

(pg) 

P. longifolia CF 617 Beaver Co. UT 1 15.25 4X 

P. longifolia CF 606 Lincoln Co. WY 1 14.51 4X 

P. stansburyi var. 

brevifolia 


brevifolia 


brevifolia 


brevifolia 


brevifolia 

P. stansburyi ssp. 

stansburyi 

P. stansburyi ssp. 

stansburyi 

SDF 50208-2 Yavapi Co. AZ 4 8.60 2X 

SDF 50308-2 Yavapi Co. AZ 3 9.18 2X 

SDF 50508-3 Coconino Co. AZ 4 9.63 2X 

CF 630 Mono Co. CA 1 8.31 2X 

CF 718 Kane Co. UT 1 10.49 2X 

SDF 42808-1 Cochise Co. AZ 3 8.18 2X 

SCS 18 Lander Co. NV 1 15.85 4X 

P. stansburyi ssp. superba SDF 50908-2 Coconino Co. AZ 3 18.86 4X 

P. stansburyi ssp. superba CF 634 Inyo Co. CA 1 10.54 2X 

P. stansburyi ssp. superba CF 704 Nye Co. NV 1 7.31 2X 



P. stansburyi ssp. superba CF 707 White Pine Co. NV 1 18.87 4X 

P. amabilis CF 780 Yavapai Co. AZ Camp Woods 1 8.28 2X 

P. amabilis CF 775 / SDF 

51507-2 

Yavapai Co. AZ Thumb Butte 3 8.41 2X b 

P. amabilis SDF 51707-1 Mojave Co. AZ Black Rock 1 8.81 2X 

P. amabilis SDF 51807-2 / 

50708-4 

Mojave Co. AZ Death Valley 

Spring 

5 16.92 4X 

P. amabilis SDF 50208-4 Yavapai Co. AZ Watson Lake 2 17.58 4X 

P. amabilis SDF 50308-1 Yavapai Co. AZ Mingus Mtn 4 24.52 6X b 

P. amabilis SDF 51407-2 Coconino Co. AZ Hobble Mtn 3 24.54 6X 

P. amabilis SDF 50508-1 Coconino Co. AZ Kaibab Lake 3 26.36 6X 

P. woodhousei CF 770 Coconino Co. AZ Oakcreek 1 8.81 2X 

P. woodhousei SDF 50108-3 Coconino Co. AZ Stoneman 3 8.89 2X b 

P. woodhousei SDF 51407-1 Coconino Co. AZ Bill Williams 1 9.02 2X 

P. woodhousei SDF 50807-1 Catron Co. NM Reserve 1 16.50 4X 

P. woodhousei SDF 50108-1 Gila Co. AZ McFadden 

Peak 

5 16.89 4X 

Cytotype




N 

DNA content 

(pg) 

P. woodhousei SDF 42908-2 Navajo Co. AZ Showlow 3 17.76 4X 

P. woodhousei SDF 50108-2 Coconino Co. AZ Strawberry 6 18.13 4X b 

P. woodhousei SDF 43008-2 Gila Co. AZ Sharp Creek 8 20.06 4X 

P. woodhousei SDF 51107-2 Gila Co. AZ Sierra Ancha 1 27.16 6X 

a inferred ploidy level supported by published mitotic chromosome count from the same population (Löve, 1971; counted by 

Daniel J. Crawford; a voucher specimen from New York Botanical Garden [Crawford 76, June 1970] was also consulted). 

b inferred ploidy level supported by meitoic chromosome counts conducted in laboratories of the authors. 

Cytotype 

tetraploid populations located in the northern portion of 

the range and diploid populations located in the southern 

portion of the range. Cytotype in some populations 

of P. stansburyi subsp. superba could not be reliably 

determined due to wide variation in DNA content, and 

further study of this taxon is needed. This preliminary 

survey indicates that cytotype variation may be a useful 

character for clarifying historically complicated tax- 

onomic divisions (as suggested by Kron et al. 2007) 

among these Phlox species. 

An in-depth survey of cytotype variation in P. 

amabilis and P. woodhousei revealed that these species 

were made up of diploid, tetraploid and hexaploid populations 

and that some cytotypes were restricted to specific 

portions of the range (Table 1; Figure 1). For example, 

most populations of P. woodhousei were associ- 

Figure 1. Map showing the locations and ploidy levels of 18 populations of Phlox amabilis and P. woodhousei sampled 

for this study. 

193


ated with the Mogollon Rim formation, but diploid 

populations appeared to be restricted to the western end 

of the rim. Hexaploid populations of P. amabilis were 

restricted to the easternmost portion of the range and 

appeared to be associated with magnesium rich igneous 

rock formations (data not shown). It is possible that 

such differences in distribution among cytotypes within 

species reflect ecological differentiation, which could 

result in local adaptation (Schonswetter et al. 2007; 

Buggs and Pannell 2007; Paun et al. 2007). Preliminary 

analysis of genetic variation in P. amabilis and P. woodhousei 

also supports differentiation among cytotypes 

(data not shown). We are continuing to investigate this 

unexpected variation in cytotype in both species using 

insights from genetics and ecology. 

CONCLUSIONS 

The results of this survey of cytotype variation 

among these western Phlox taxa demonstrate the value 

of such knowledge for the study of plant diversity, evolution, 

and conservation. The use of flow cytometry allowed 

us to gather data easily and rapidly on cytotype 

variation across expanded sample sizes that included 

multiple species from multiple locations throughout 

their ranges as well as population level sampling in species 

of conservation interest. We found that cytotype 

was variable not only among species but also among 

populations within species. This variation appears to be 

related to taxonomy, geography, and possibly ecology. 

Continued studies of cytotype variation in Phlox will 

provide valuable data for resolving taxonomic divisions 

as well as insight into the evolutionary diversification of 

this group. The detailed analysis of cytotype variation in 

P. amabilis and P. woodhousei, combined with results 

from ongoing population genetic work, suggest the presence 

of unique evolutionary lineages and ecological differentiation, 

both of which are important when planning 

conservation strategies. Taken together with recent reviews, 

these results emphasize the value of incorporating 

an understanding of cytotype variation into conservation 

biology studies. 


We gratefully acknowledge support from the KSU 

Ecological Genomics Institute, the KSU Center for the 

Understanding of Origins and NSF DEB-0089656. We 

thank Thelma Green and Maureen Ty for technical assistance, 

Jerry Davison and Suzi Strakosh for assistance 

with the initial stages of flow cytometry, and Mark 

Mayfield and Ed Turcotte for unpublished chromosome 

count data. We thank the New York Botanical Garden 

(NY) for a loan of herbarium specimens. We thank 

Montezuma Castle National Monument (MOCA-2008- 

SCI-0004), Grand Canyon National Park (GRCA-2008- 

SCI-0015), the National Forest Service (various permits, 

Regions 1-6), the California BLM, and the Navajo Nation 

(Permits 414 and 051402-062) for permission to 

collect plants. This is publication 09-372-A of the Kansas 

Agricultural Experiment Station. 


Bino, R.J., S. Lanteri, H.A. Verhoeven, and H.L. 

Kraak. 1993. Flow cytometric determination of nuclear 

replication stages in seed tissues. Annals of Botany 72: 

181-187. 

Booy, G., R.J.J. Hendriks, M.J.M. Smulders, J.M. 

Van Groenendael, and B. Vosman. 2000. Genetic diversity 

and the survival of populations. Plant Biology 2: 

379-395. 

Buggs, R.J.A. and J.R. Pannell. 2007. Ecological 

differentiation and diploid superiority across a moving 

ploidy contact zone. Evolution 61: 125-140. 

Cronquist A., A.H. Holmgren, N.H. Holmgren, J.L. 

Reveal, and P.K. Holmgren. 1984. Intermountain Flora, 

Vol. 4. New York Botanical Garden, Bronx, New York. 

Davison, J., A. Tyagi, and L. Comai. 2007. Largescale 

polymorphism of heterochromatic repeats in the 

DNA of Arabidopsis thaliana. BMC Plant Biology 7: 

44. 

De Lange, P.J., P.B. Heenan, D.J. Keeling, B.G. 

Murray, R. Smissen, and W.R. Sykes. 2008. Biosystematics 

and conservation: a case study with two enigmatic 

and uncommon species of Crassula from New 

Zealand. Annals of Botany 101: 881-99. 

Dolezel, J. and J. Bartos. 2005. Plant DNA flow cytometry 

and estimation of nuclear genome size. Annals 

of Botany 95: 99-110. 

Ellis, J.R. and J.M. Burke. 2007. EST-SSRs as a resource 

for population genetic analyses. Heredity 99: 

125-132. 

Ferguson, C.J. and R.K. Jansen. 2002. A chloroplast 

DNA phylogeny of eastern Phlox (Polemoniaceae): implications 

of congruence and incongruence with the ITS 

phylogeny. American Journal of Botany 89: 1324-1335. 

Ferguson, C.J., F. Krämer, and R.K. Jansen. 1999. 

Relationships of eastern North American Phlox 

(Polemoniaceae) based on ITS sequence data. Systematic 

Botany 24: 616-631. 

Ferguson, C.J., S.C. Strakosh, and R. Patterson. 

2012. Phlox. Pp. 1068-1071 In: B. G. Baldwin, D.H. 

Goldman, D.J. Keil, R. Patterson, T.J. Rosatti, and D.H. 

Wilken, eds. The Jepson Manual, Vascular Plants of 

California, second edition. University of California 

Press, Berkeley. 

Grant, V. 1959. Natural history of the Phlox family: 

systematic botany. Martinus Nijhoff, The Hague, Netherlands. 

Halverson, K., S.B. Heard, J.D. Nason, and J.O. 

Stireman III. 2008. Origins, distribution, and local cooccurrence 

of polyploidy cytotypes in Solidago altis- 

194


sima (Asteraceae). American Journal of Botany 95: 50- 

58. 

Kron, P., J. Suda, and B.C. Husband. 2007. Applications 

of flow cytometry to evolutionary and population 

biology. Annual Review of Ecology, Evolution, and 

Systematics 38: 847-876. 

Levin, D.A. and B.A. Schaal. 1970. Reticulate evolution 

in Phlox as seen through protein electrophoresis. 

American Journal of Botany 57: 977-987. 

Levin, D.A. and D.M. Smith. 1966. Hybridization 

and evolution in the Phlox pilosa complex. The American 

Naturalist 100: 289-301. 

Löve, A. 1971. IOPB chromosome number reports 

XXXI. Taxon 20:157-160. 

Murray, B.G. and A.G. Young. 2001. Widespread 

chromosome variation in the endangered grassland forb 

Rutidosis leptorrhynchoides F. Muell. (Asteraceae: Gnaphalieae). 

Annals of Botany 87: 83-90. 

Paun, O., F.M. Fay, D.E. Soltis, and M.W. Chase. 

2007. Genetic and epigenetic alterations after hybridization 

and genome doubling. Taxon 56: 649-656. 

Schonswetter, P., M. Lachmayer, C. Lettner, D. Prehsler, 

S. Rechnitzer, D.S. Reich, M. Sonnleitner, I. 

Wagner, K. Huelber, G.M. Schneeweiss, P. Travnicek, 

and J. Suda. 2007. Sympatric diploid and hexaploid cytotypes 

of Senecio carniolicus (Asteraceae) in the Eastern 

Alps are separated along an altitudinal gradient. 

Journal of Plant Research 120: 721-725. 

Severns, P.M. and A. Liston. 2008. Intraspecific 

chromosome number variation: a neglected threat to the 

conservation of rare plants. Conservation Biology 22: 

1641-1647. 

Soltis, D.E., P.S. Soltis, D.W. Schemske, J.F. Hancock, 

J.N. Thompson, B.C. Husband, and W.S. Judd. 

2007. Autopolyploidy in angiosperms: have we grossly 

underestimated the number of species? Taxon 56: 13- 

30. 

Suda, J., P. Kron, B.C. Husband, P. Travnicek. 2007. 

Flow cytometry and ploidy: applications in plant systematics, 

ecology and evolutionary biology. In: J. 

Dolezel, J. Greilhuber, and J. Suda, eds. Flow Cytometry 

with Plant Cells: Analysis of Genes, Chromosomes 

and Genomes. Wiley-VCH, Weinheim. Pp. 103-130. 

Wherry, E.T. 1955. The genus Phlox. Morris Arboretum 

Monographs 3. Philadelphia, Pennsylvania. 

195


Prioritizing Plant Species for Conservation in Utah: 

Developing the UNPS Rare Plant List 

Walter Fertig, Chair, Utah Native Plant Society Rare Plant Committee 

Abstract. Rare plant lists are an important tool for identifying and prioritizing species for conservation attention. 

Over a dozen systems have been derived for ranking the rarity and conservation priority of plant and animal species, 

each differing in emphasis, methods, and biological and anthropogenic criteria. In 2007 I developed a new ranking 

protocol for the flora of Wyoming that combines aspects of the NatureServe, International Union for Conservation of 

Nature (IUCN), and US Fish and Wildlife Service systems and the classic paper “Seven forms of Rarity” by Deborah 

Rabinowitz. The so-called “Wyoming protocol” was adopted by the Utah Native Plant Society’s Rare Plant Committee 

to develop an updated rare plant list for Utah. In this protocol, species or varieties are assessed using seven qualitative 

criteria: Utah’s contribution to global distribution, number of populations in the state, number of individuals, 

habitat specificity, intrinsic rarity, magnitude of threats, and population trend. Individual criteria are rated on a binary 

scale (0 for unthreatened, 1 for at risk) based on expert opinion. Species for which no data are available are scored 

“unknown”. The values for each criterion are summed to derive a minimum and maximum potential score for each 

taxon. The minimum score is calculated by summing each individual score and treating any unknown criteria as 0. 

The maximum score is derived in the same way, except that unknown criteria are given a value of 1. The two summary 

scores are averaged to determine a conservation priority rank. Those taxa that are at risk for a large number of 

criteria have higher conservation priority ranks than those species that are at risk for only a few criteria. This simple 

method allows practitioners to rapidly identify the relatively small subset of species of high or extremely high conservation 

priority (those with limited ranges, few populations, low numbers, high habitat specificity, high intrinsic rarity, 

high threats, and downward trends) and those species with significant data gaps in need of additional study. Being 

able to differentiate among species based on their priority score enables conservationists and managers to better allocate 

limited resources to those taxa most in need. The UNPS Utah rare plant list developed by the Rare Plant Committee 

and attendees of a breakout ranking session at the Fifth Southwestern Rare Plant Conference is presented in 

Appendices 1-4, with modifications adopted at subsequent Utah Rare Plant meetings from 2010-2012. 

Experts predict that one-fifth to one-third of all vascular 

plant species in the United States are threatened 

with local or range-wide extinction (Center for Plant 

Conservation 2000). This number is only likely to increase 

as plant habitat becomes increasingly fragmented 

and disturbed by development, climate change, or invasion 

by non-native weeds. Not all plant species, however, 

are equally imperiled. Some species are naturally 

rare due to their limited range, high habitat specificity, 

or low population size (Rabinowitz 1981), but may not 

be in imminent danger because their population trends 

are stable or threats are presently low. Because so many 

species are potentially vulnerable and conservation resources 

(time, funding, and personnel) are nearly always 

inadequate, conservation biologists have a dilemma determining 

which species should be the highest priority 

for attention (Noss and Cooperrider 1994; Regan 2005). 

Rare species lists can be an important tool for identifying 

and prioritizing those taxa (species, subspecies, 

and varieties) most vulnerable to extinction. Over the 

past 40 years conservation biologists have proposed 

more than a dozen ranking systems for creating state or 

national rare species lists (Andelman et al. 2004). 

Ranking schemes often differ widely in their emphasis 

on inherent rarity, degree of threat, vulnerability of extinction 

as well as their scoring methods and overall 

complexity and transparency (Akcakaya et al. 2000; 

Faber-Langendoen et al. 2009; IUCN 2001; O’Grady et 

al. 2004; Rabinowitz 1981; Regan et al. 2004; Spence 

2012, US Fish and Wildlife Service 1983). These systems 

also utilize different criteria for ranking, including 

abundance, number of populations, geographic range, 

area of occupancy, population trend, intrinsic rarity, 

taxonomic distinctiveness, ecological significance, 

population viability, habitat condition or degree of fragmentation, 

magnitude and imminence of threats, and 

number of protected populations (Andelmann et al. 

2004; Beissinger et al. 2000; Breininger et al. 1998; 

Holsinger 1992; IUCN 2001; Keith 1998; Mace et al. 

2008; Millsap et al. 1990; Panjabi et al. 2005; Rabinowitz 

1981; Regan et al. 2004; Spence 2012; US Fish 

and Wildlife Service 1983). 

Ideally, a ranking system should have a strong biological 

basis, recognize the significance of threats and 

trends, be easy to apply and update with available information 

(while recognizing the importance of data gaps), 

and be transparent (Fertig 2011). Each ranking system 

has its merits, but none meet all of these criteria. For 

196


example, the US Fish and Wildlife Service (1983) system 

for listing species under the US Endangered Species 

Act is heavily weighted towards threats and taxonomic 

distinctiveness at the expense of other aspects of rarity 

(Master et al. 2000). The IUCN (2001) protocol focuses 

chiefly on population size, trends, and likelihood of extinction 

but is dependent on quantitative viability data 

that are not always available for vascular plants. One 

advantage of the IUCN protocol, however, is its recognition 

of “data deficient” species (Akçakaya et al. 2000). 

Rabinowitz (1981) introduced a simple, but elegant, 

binary ranking system using just three components of 

rarity: geographic range, abundance, and habitat specificity. 

But several additional biological and anthropogenic 

criteria (such as threat) were not incorporated, 

which limits the suitability of the Rabinowitz system for 

prioritizing among different kinds of rare species. 

The most widely used ranking protocol today is the 

natural heritage system, first developed by The Nature 

Conservancy in the 1970s (Master 1991) and now administered 

by NatureServe. In this system, full species 

or varieties are assigned a conservation rank on a scale 

of 1 (critically imperiled) to 5 (demonstrably secure) 

across their entire global range (G rank) and at a subregional 

scale (state/province, or S rank). Traditionally, G 

and S ranks were based on the number of occurrences 

(discrete biological populations), abundance, or risk of 

extinction as determined by expert opinion (Master et al. 

2000). In the past decade, NatureServe protocols have 

become more quantitative and consider additional ranking 

criteria, including long and short-term trends, area 

of occupancy, condition of occurrences, intrinsic rarity, 

and threat (Faber-Langendoen et al. 2009; Regan et al. 

2004). Unfortunately, the revised NatureServe ranking 

protocol has become more complex and less transparent. 

Individual ratings are weighted differently, some criteria 

are used only conditionally, and scores are tallied by a 

“black box” computer algorithim (Faber-Langendoen et 

al. 2009). 

As part of my doctoral dissertation on plant conservation 

in Wyoming (Fertig 2011), I developed a hybrid 

ranking protocol by borrowing components of each of 

the preceding systems. As a starting point, I adopted 

most of the rarity factors from NatureServe (Regan et al. 

2004), added the uncertainty components of IUCN 

(2001), and included an emphasis on threats from the 

US Fish and Wildlife Service (1983). The ranking system 

itself is a modification of the qualitative, binary 

scoring employed by Rabinowitz (1981), expanded to 

include additional criteria. I added a simple scoring 

component to classify plant species into six different 

rarity classes reflecting each taxon’s overall conservation 

priority (Fertig 2009, 2011). 

In 2007, I beta-tested the “Wyoming protocol” at the 

annual Utah rare plant meeting sponsored by the Utah 

Native Plant Society (UNPS) and Red Butte Garden. 

Following the meeting, the UNPS state board voted to 

reestablish a rare plant committee and charged it with 

applying this ranking system to the entire Utah vascular 

plant flora in order to create a new, prioritized list of 

rare plant species for the state. A draft version of the 

list was presented at a special session of the Fifth Southwestern 

Rare and Endangered Plant Conference, held at 

the University of Utah in March 2009. Based on feedback 

from meeting participants and other experts, the 

list was revised and published in November 2009 

(Fertig 2009). The list has since been updated twice 

(Fertig 2010a, 2012) based on additional input from the 

UNPS Rare Plant Committee, attendees of the Society’s 

annual rare plant meeting, and review of new literature. 

The purpose of this paper is to briefly describe the 

Wyoming ranking protocol and its application to the 

flora of Utah. Appendices 1-4 include the current lists 

of Utah plants on the Extremely High Priority, High 

Priority, Watch, and Need Data lists. The paper concludes 

with a comparison of the current list to previous 

rare plant lists for Utah and a discussion of additional 

applications and future directions. 

METHODS 

Ranking Criteria 

The Wyoming protocol is based on seven biological 

and anthropogenic factors that influence the conservation 

priority of a vascular plant species. These criteria 

are: 

1. Geographic range. Geographic range takes into 

account the state’s contribution to the total global distribution 

of a species. Six geographic range categories are 

recognized (Table 1). Local and regional endemics 

have highly restricted global distributions, ranging from 

single populations covering a few acres to less than 

250,000 km 2 (an area about the size of the state of Wyoming). 

Widespread species, defined as occupying a 

global range in excess of 250,000 km 2 , can still be considered 

rare if state populations are widely isolated from 

the core of the species’ range (disjunct) or are at its very 

edge (peripheral). A small number of plant species may 

occur widely but are limited to small, often scattered or 

discontinuous habitats and occupy less than 5% of the 

state (sparse). Species that are introduced to the state are 

not included in the rankings. 

2. Number of Populations. This criterion is based on 

the number of extant populations of a species within 

Utah (occurrences outside the state are not considered). 

Populations are defined as aggregations of individual 

plants within a specific geographic area that are separated 

from other populations by a physical barrier, extensive 

area of unsuitable habitat, or sufficient distance 

to prevent gene flow (usually about 1-2 km). The num- 

197


Table 1. Scores for Ranking Factors. 

Ranking Factor Category or Condition Points 

1. Geographic Range 

(only taxa native to the 

state are scored) 

2. Number of 

Populations 

3. Abundance 

Local endemic (global range less than 16,500 km 2 or about 1 degree of latitude x 2 

degrees of longitude) 

Regional endemic (global range covering 16,501-250,000 km 2 or an area about the 

size of Wyoming) 

Disjunct (globally widespread but state population is isolated from the main contiguous 

range of the species by a gap of more than 800 km) 

Peripheral (globally widespread but state population is at the margin of its continuous 

range and occupies less than 5% of the state’s area near state boundary) 

Sparse (globally widespread, but distribution patchy and discontinuous in the state 

and covering less than 5% of the state’s area) 

Widespread (occurs widely across North America [covering more than 250,000 

km 2 ] and across the state [occupying well over 5% of the area]) 

Unknown 0-1 

Low (fewer than 25 extant populations in state) 1 

Medium to High (25 or more extant populations in state) 0 

Unknown 0-1 

Low (depends on life history of species, but typically less than 30,000 individuals 

for perennials [higher numbers allowable for annuals] or occupying an area of less 

than 3000 acres in state) 

Medium to High (known from well over 30,000 individuals for perennials or occupying 

an area greater than 3000 acres in state) 

Unknown 0-1 

4. Habitat Specificity High (“Specialist” restricted to one or a few specialized geologic substrates, soil 

types, or vegetation types) 

Medium to Low (“Generalist” found in numerous geologic substrates, soil types, or 

vegetation types) 

Unknown 0-1 

5. Intrinsic Rarity High (unusual life history, dependence on rare or specialized pollinators, poor dispersal, 

low fecundity, poor seedling survival, etc.) 

6. Magnitude and 

Imminence of Threats 

Medium to Low (no unusual life history or biological attributes limiting establishment 

or persistence 

Unknown 0-1 

High (current or foreseeable threats significant or broad in scale or scope 1 

Medium to Low (threats minimal or limited to small percentage of populations 

now or in the foreseeable future) 

Unknown 0-1 

7. Population Trend Decreasing (short to long-term decline in number, size, or vigor of populations) 1 

Increasing, stable, or oscillating around a mean 0 

Unknown 0-1 

198 

2 

1 

1 

1 

1 

0 

1 

0 

1 

0 

1 

0 

0


ber of populations is not necessarily equivalent to the 

number of collections of a species, especially if these 

collections are from the same general area. 

3. Abundance. Abundance refers to the number of 

individual plants known statewide. If census data are 

lacking, abundance can be approximated from the relative 

dominance of a species within its area of occupied 

habitat. 

4. Habitat Specificity. This factor assesses the degree 

to which a species is a habitat specialist restricted 

to a particular soil or geologic substrate (edaphic endemics) 

or vegetation type, or is a generalist found in a 

wide variety of substrates or plant communities. 

5. Intrinsic Rarity. Analogous to habitat specificity, 

intrinsic rarity addresses those attributes of a species’ 

life history that may limit its establishment or persistence. 

Examples include low fecundity, poor dispersal, 

low seedling survival, low genetic diversity, or dependence 

on specialized pollinators. 

6. Magnitude and Imminence of Threats. This criterion 

assesses the scope, severity, and immediacy of current 

or future negative impacts on a species. Potential 

threats include habitat destruction, over-collection, herbivory, 

trampling or soil compaction from recreation, or 

competition from invasive plants. 

7. Population Trend. Trend is the change in population 

size, extent, and vigor over time. 

Assigning scores to each criterion 

Following the model of Rabinowitz (1981), six of the 

seven preceding criteria are scored using a binary rating 

(high/low or increasing/decreasing). A score of 1 is assigned 

to those conditions that make a species highly 

vulnerable to extinction or extirpation, while a score of 

0 is given for conditions that only moderately impact or 

do not adversely affect a species’ persistence in the 

state. The only exception is geographic range in which 

three scores are possible (0, 1, or 2) to allow greater 

weighting of locally endemic species. If there is insufficient 

data to rate a specific criterion, or available information 

is inconclusive, a value of “U” (unknown) may 

be assigned. Scores for individual criteria are shown in 

Table 1. 

Scoring is based on a review of pertinent literature, 

specimen databases, and expert knowledge and should 

be supported by corroborating data. Scores can be tabulated 

in a spreadsheet or in a simple data form (see Table 

2 for an example). 

Determining conservation priority 

Once the ranking form or table is completed, the individual 

scores for each of the seven ranking factors are 

summed to derive both a minimum and potential (maximum) 

score (Table 2). These scores can range from 0 

to 8. The minimum score includes only those factors for 

which information is available, with any unknowns 

scored as 0. The potential score includes the same values 

but with unknowns given a “worst case” score of 1. 

The minimum and potential scores are then averaged 

(with the sum rounded down) to derive an overall score 

(Table 2). 

The final score can be used to assign each species to 

one of the following six conservation priority categories: 

Extremely High (7 or 8 points): species at extreme 

risk of extirpation across its range due to all seven of the 

following conditions: limited geographic range, small 

number of populations, low number of individuals, high 

habitat specificity, high intrinsic rarity, high threats, and 

downward population trend. 

High (6 points): species at high risk of extirpation 

rangewide or in the state. High priority species are 

scored as vulnerable for at least six of the seven ranking 

criteria. 

Watch (5 points): species currently secure but vulnerable 

to downward changes in status. These taxa are 

scored as vulnerable for at least five of the seven ranking 

criteria. 

Medium (4 points): species secure rangewide but 

vulnerable to extirpation in the state. Medium priority 

species are scored as vulnerable for at least four of the 

seven ranking criteria. 

Low (0-3 points): species secure rangewide and in 

the state. These species are scored as vulnerable for 

three or less of the seven ranking criteria. 

Need Data: insufficient data available to score species 

for at least three of the seven ranking criteria. If 

information were available. these species would likely 

be ranked as Extremely High, High, Watch, or Medium 

priority rather than Low priority. 

An example of ranking a Utah species 

The following example demonstrates the application 

of the Wyoming protocol. Penstemon gibbensii is a narrow 

endemic of extreme NE Utah (Daggett County), 

adjacent NW Colorado, and SC Wyoming, earning it 2 

points for geographic range. In Utah, it is known from a 

single occurrence in the Browns Park area (1 point for 

low number of populations) containing approximately 

700 plants (1 point for low number of individuals) (Utah 

Division of Wildlife Resources 1998). It is restricted to 

barren white shales of the Browns Park Formation (1 

point for high habitat specificity). Little is known about 

the pollination biology or life history of P. gibbensii 

(Heidel 2009), suggesting an “unknown” score is appropriate 

for intrinsic rarity. Threats from trampling, soil 

erosion, and over-collection by gardeners are high 

throughout its range (1 point for threats). Trends in 

Utah are unknown, although some populations in Wyoming 

appear to be declining (Heidel 2009). The mini- 

199


mum score for P. gibbensii is 6 points, while the potential 

score is 8. The average of the two scores is 7, earning 

P. gibbensii a place on the UNPS Extremely High 

priority list. 

RESULTS 

The UNPS Rare Plant Committee met in January 

2009 to apply the Wyoming protocol to the entire flora 

of Utah. Based on the fourth edition of A Utah Flora 

(Welsh et al. 2008) and other recent literature (such as 

The Flora of North America and Intermountain Flora) 

we started with a pool of 4273 taxa* of vascular plants 

known to be native or introduced in Utah. We immediately 

removed 1017 cultivated and naturalized (nonnative) 

taxa, as we deemed these not to be of conservation 

importance in the state. Of the 3160 species native 

to Utah, we eliminated another 1421 common and widespread 

taxa (mostly ranked S4 or S5 by NatureServe) of 

*Taxa include full species and unique subspecies and varieties, 

treated here in the phylogenetic sense of Cracraft (1987) 

as the smallest evolutionary units that are diagnosably distinct. 

Table 2. Sample Ranking Form 

Species 

Date Scored 

Ranking Factors 

Evaluators 

Scores 

Select one score per ranking factor in either column A or B 

Column A 

1. Geographic Range Local Endemic (2) ____ 

Regional Endemic, 

Disjunct, Peripheral, 

or Sparse (1) ____ 

Widespread (0) ____ 

2. Number of Populations Low (1) ____ 

Medium to High (0) ____ 

3. Abundance Low (1) ____ 

Medium to High (0) ____ 

4. Habitat Specificity High (1) ____ 

Medium to Low (0) ____ 

5. Intrinsic Rarity High (1) ____ 

Medium to Low (0) ____ 

6. Magnitude and Imminence 

of Threats 

High (1) 

Medium to Low (0) 

____ 

____ 

7. Population Trend Downward (1) ____ 

Stable, Oscillating 

or Upward (0) ____ 

TOTALS 

Conservation Priority* 

_____________________ 

Sum of scores in 

Column A 

____ 

Minimum 

(based on total score in 

Column A) ____ 

Unknown (1) 

Unknown (1) 

Unknown (1) 

Unknown (1) 

Unknown (1) 

Unknown (1) 

Unknown (1) 

Column B 


Column B 

____ 

____ 

____ 

____ 

____ 

____ 

____ 

____ 

Potential 

(based on sum of scores 

in Columns A + B) ____ 

Comments 


Column A + B 

____ 

Averaged 

(based on average of 

scores in Columns 

A + B rounded down) ____ 

*Conservation Priority is based on the averaged point total: Extremely high priority = total score of 7 or 8 points, 

High priority = total score of 6 points, Watch list = total score of 5, Medium priority = total score of 4 points, Low 

priority = total score of 0-3 points. 

200


low conservation priority. The remaining 1739 taxa 

were then ranked by the committee using the Wyoming 

protocol. At this point another 927 taxa were assigned 

to the Low Priority list, leaving 812 species among the 

other categories (Extremely High, High, Watch, Medium, 

and Need Data). These remaining taxa comprised 

the first draft of the UNPS rare plant list, presented at 

the 2009 Southwest Rare Plant Conference. 

Revisions suggested by the conference attendees and 

other specialists resulted in changes to the final score for 

over 100 taxa. These changes were reflected in the final 

UNPS rare plant list, published in the Society’s membership 

publication, the Sego Lily, in November 2009 

(Fertig 2009). The list has been revised twice since 

then. Twenty taxa were changed in rank due to new 

data and 11 newly described or discovered species were 

added (Fertig 2010, 2012). 

Extremely High Priority At present, 31 Utah plant 

taxa are recognized as species of Extremely High conservation 

priority (Table 3, Appendix 1). This group 

represents just 1% of the entire flora of the state. To 

qualify as Extremely High priority, a species must have 

a limited geographic range, few populations, low number 

of individuals, high habitat specificity and intrinsic 

rarity, high threats, and a downward population trend. 

Nearly half (15 taxa) of these species are presently listed 

as Threatened or Endangered under the ESA, and another 

four species are candidates or proposed for listing. 

Another 14 of these species are designated as Sensitive 

by the BLM or Forest Service. Just two of the 31 Extremely 

High priority plant taxa lack any formal designation: 

Iris pariensis (a species with taxonomic questions 

that may be extinct) and Viola clauseniana 

(endemic to Zion National Park and increasingly threatened 

by competition from exotic plants and possible 

over-collection; Fertig 2010b). Since the initial UNPS 

list was published one species has been dropped from 

the Extremely High priority list (Sclerocactus wetlandicus, 

changed to High priority) and one species has 

been added (Carex specuicola, upgraded from the High 

priority list). 

High Priority The High priority list presently contains 

119 taxa, up from 114 recognized in 2009 (Table 

3, Appendix 2). This group consists of only 3.8% of the 

native flora of Utah. High priority plant species generally 

have limited geographic ranges, low population 

size, few known occurrences, and high habitat specificity, 

but have lower intrinsic rarity, fewer threats, or stable 

trends (or these factors are unknown) compared to 

Extremely High priority taxa. Eight of the state’s 24 

federally listed Threatened or Endangered species are 

ranked High priority by UNPS, and two other species 

from this group are currently candidates for potential 

listing. Half of the High priority taxa (60 species) are 

listed as Sensitive by the BLM or Forest Service. The 

remaining 51 species have no status and consist mostly 

of species that were once considered for listing under 

the ESA (41 former Category 2 or 3C taxa), or are recently 

described (Atwood et al, 1991; Welsh and Atwood 

2009). 

Watch List The UNPS Watch list currently contains 

264 taxa (Table 3, Appendix 3), or 8.4% of the native 

Utah flora. This category contains local or regional 

endemics with high habitat specificity or intrinsic rarity, 

but which are either locally abundant or apparently secure 

at present. If current conditions were to change 

significantly, however, these species could easily trend 

downward and become species of greater concern (and 

be rescored as Extremely High or High priority). At 

least 115 species in the Watch category were initially 

considered for potential listing under the ESA (Atwood 

et al. 1991, Ayensu and DeFilipps 1978; Greenwalt 

1975, Welsh 1978, Welsh et al. 1975, Welsh and Chatterley 

1985) in the 1970s and 1980s, but were subsequently 

dropped from consideration after better survey 

data found them to be less imminently threatened or 

rare. Astragalus montii is the only species currently 

listed under the ESA that is categorized on the Watch 

list (Erigeron maguirei, recently delisted, is also on the 

Watch list). Another 70 species in the Watch category 

are presently listed as Sensitive by the BLM or Forest 

Service. Over one-third of the changes to the UNPS list 

since 2009 have involved additions or status changes 

affecting the Watch category. 

Medium Priority Another 329 plant species in Utah 

are currently ranked as Medium priority for conservation 

attention (Table 3)*. Species in this category are 

usually widespread globally but rare within the state, 

with few known occurrences, low numbers, or potentially 

high threats. A small subset of Medium priority 

species (including 13 listed as Sensitive by the BLM or 

Forest Service) are locally common regional endemics 

with stable trends and low threats that might otherwise 

be treated on the Watch list. Medium priority taxa account 

for 10.4% of the native flora of Utah. 

Need Data A total of 115 taxa are presently on the 

UNPS Need Data list (Table 3, Appendix 4), or 3.6% of 

the state’s native flora. This is the fastest growing category, 

as it tends to be the repository for newly described 

species or those discovered for the first time within 

Utah. Recent additions include Astragalus kelseyae, 

Eremogone loisiae, Eriogonum domitum, and Navarretia 

furnissii, all named as new species since 2009 

(Corbin 2011, Grady and Reveal 2011; Holmgren and 

Holmgren 2011, Johnson et al. 2012). The Need Data 

list also includes species with unresolved taxonomic 

questions and those that have been reported for the state 

*In the interests of space, the Medium priority list is not included 

in this paper, but is available by request from UNPS. 

201


Figure 1. Utah Counties with three-letter codes used in 

Table 3 and Appendix 1-4. 

Figure 2. TNC Ecoregions of Utah, used in Table 3, as 

defined by Stein et al. (2000). 

Table 3. Summary of UNPS Rare Plant List, 2009-2012 

See text for explanation and scoring of each of the ranking categories. Counties are depicted in Figure 1. Ecoregions 

are defined as geographic regions with similar climate, topography, and vegetation, as defined by The Nature Conservancy 

(Stein et al. 2000) and are shown in Figure 2. 

State/County/TNC Ecoregion 

Extremely 

High 

202 

High Watch Need 

Data 

Medium 

State Utah Statewide 31 119 264 115 329 858 

County 

Beaver County (Bvr) 0 8 13 8 24 53 

Box Elder County (Box) 1 2 10 6 28 47 

Cache County (Cch) 0 3 7 4 28 42 

Carbon County (Crb) 1 1 9 5 3 19 

Daggett County (Dag) 1 2 13 3 16 35 

Davis County (Dav) 0 0 2 2 6 10 

Duchesne County (Dch) 4 14 28 9 24 79 

Emery County (Emr) 5 8 24 17 23 77 

Garfield County (Grf) 1 17 49 17 49 133 

Grand County (Grn) 1 12 26 17 29 85 

Iron County (Irn) 0 5 18 7 23 53 

Juab County (Jub) 1 6 13 12 17 49 

Kane County (Kan) 3 25 52 12 67 159 

Millard County (Mil) 0 6 21 17 21 65 

Total



County 

TNC 

Ecoregion 

State/County/TNC Ecoregion 

Extremely 

High 

High Watch Need 

Data 

Medium 

Morgan County (Mor) 0 0 2 0 1 3 

Piute County (Piu) 0 9 13 3 14 39 

Rich County (Rch) 0 1 4 5 9 19 

Salt Lake County (Slt) 0 8 12 2 20 42 

San Juan County (Snj) 2 13 39 15 63 132 

Sanpete County (Snp) 2 7 12 4 8 33 

Sevier County (Sev) 3 13 18 6 21 61 

Summit County (Sum) 0 1 5 3 15 24 

Tooele County (Toe) 1 3 13 5 13 35 

Uintah County (Uin) 6 15 36 10 23 90 

Utah County (Uta) 1 9 14 12 14 50 

Wasatch County (Was) 0 3 8 4 8 23 

Washington County (Wsh) 6 18 76 17 115 232 

Wayne County (Way) 6 12 13 11 22 64 

Weber County (Web) 0 3 6 4 9 22 

Colorado Plateau (CP) 13 47 98 41 124 323 

Columbia River Basin (CRB) 1 1 2 2 19 25 

Great Basin (GB) 2 16 43 28 63 152 

Mohave Desert (MD) 4 11 44 12 67 138 

Utah High Plateaus (UHP) 9 36 69 28 63 205 

Utah-Wyoming Rocky Mtns 

(UWRM) 

Total 

0 17 50 16 65 148 

Wyoming Basins (WyB) 7 12 21 11 19 70 

in the literature with ambiguous supporting data. The 

majority of taxa in this category need additional information 

on trend, threats, and abundance. At least 15 of 

these species are designated as BLM or Forest Service 

Sensitive. Eriogonum corymbosum var. nilesii is a candidate 

for potential listing under the ESA, although a 

recent monograph questions the veracity of Utah reports 

(Reveal in Holmgren et al. 2012). 

Low Priority The remaining 2302 native plant taxa of 

Utah are presently scored as Low conservation priority 

(72.8% of the total native flora). These species are usually 

widespread rangewide and within Utah, have numerous 

populations in the state, low habitat specificity 

and intrinsic rarity, stable to increasing trends, and few 

threats. These species are still important for the sake of 

conserving overall biodiversity, but rarely warrant individualized 

attention. 

All told, 858 of Utah’s 3160 native plant taxa (27.2%) 

have been identified as Extremely High, High, Watch, 

or Medium priority, or are on the UNPS need data list 

using the Wyoming protocol system. The distribution 

of these species across the state is not random. With 

232 taxa of conservation concern (27% of the state total), 

Washington County has the highest number of species 

of conservation concern of any county in Utah 

(Figure 1, Table 3). This high richness can be explained 

in part by Washington County’s location at the confluence 

of four major floristic regions: the Mojave Desert, 

203

Great Basin, Colorado Plateau, and Rocky Mountains. 

The next three counties with the greatest number of 

species of conservation concern (Kane, Garfield, and 

San Juan) are all, like Washington County, located on or 

near the southern boundary of the state. Additional 

counties with a high number of species of concern include 

Uintah, Duchesne, Grand, and Emery counties 

near the eastern border of Utah. By contrast, the number 

of rare species is relatively low in the northern and 

western tier of counties. Surprisingly few plant species 

of conservation concern occur in the greater Salt Lake 

City area, though this may be an artifact of undersampling 

or reflect significant habitat losses over the 

last 150 years of settlement (Fertig 2009). 

Ecoregions are defined as geographic areas with a 

similar climate, topography, and vegetation. The Nature 

Conservancy has developed a national ecoregional classification 

(Stein et al. 2000) that recognizes seven ecoregions 

in Utah (Figure 2*). Of these, the Colorado Plateau 

ecoregion has the highest number of plant species 

of conservation concern with 323 taxa, or 37.6% of the 

state total (Table 3). This region, which includes the 

canyon country and La Sal and Abajo mountains of 

southeast Utah, also has the highest number of endemic 

species in the state (Welsh and Atwood 2009). Although 

comparable in area to the Colorado Plateau, the 

Great Basin ecoregion has less than half as many species 

of concern (152 taxa). The Mohave Desert ecoregion 

of extreme southwestern Utah is the second smallest 

in area in the state (after the Columbia River Basin 

in the Grouse Creek and Raft River mountains of northwest 

Utah) but has the highest concentration of species 

of concern per unit area (138 taxa in all). The Utah 

High Plateaus, which extends from the Tavaputs Plateau 

and Book Cliffs of eastern Utah to the Wasatch Plateau, 

and Markagunt and Paunsaugunt plateaus of southcentral 

Utah, has the second highest concentration of 

endemics and taxa of conservation concern (205 species, 

or 23.9% of the state total) (Table 3). 

DISCUSSION 

The UNPS rare plant list is just the latest in a long 

series of comparable publications dating back to the 

passage of the Endangered Species Act (ESA) of 1973. 

No plants were included in the very first official list of 

species protected under the ESA, but Congress directed 

the Smithsonian Institution to develop the first national 

list of vascular plants that might qualify for listing as 

*Welsh and Atwood (2009) have developed a similar system 

of “geoendemic areas” to identify floristic regions of Utah. 

Their map depicts 12 subregions and differs from the TNC 

system in lumping the Columbia River Basin with the Great 

Basin and in more finely subdividing the Utah-Wyoming 

Rocky Mountains, Utah High Plateaus, Colorado Plateau, and 

Mohave Desert ecoregions. 


204 

Threatened or Endangered. That list appeared in 1975 

and was based on the best available information at the 

time (Ayensu and DeFilipps 1978, Greenwalt 1975). 

The Smithsonian Institution cited 761 plant taxa as potentially 

Endangered, 1238 as Threatened, and 100 as 

extinct in the continental United States (another 1088 

endangered, threatened, and extinct species were reported 

for Hawaii). Of these, 156 species were from 

Utah, including 56 listed as endangered, 91 threatened, 

and 9 extinct (Greenwalt 1975). 

Welsh and others (1975) reviewed the Smithsonian 

publication and developed the first Utah-specific compilation 

of endangered, threatened, extinct, endemic, and 

rare plant species in 1975. Welsh and his co-authors 

recognized 66 Utah plant taxa as possibly endangered, 

198 as threatened, 7 as extinct, and 20 as extirpated 

(extinct in Utah, but extant elsewhere). Most of the recommendations 

by Welsh and others (1975) were incorporated 

into a revised Smithsonian list (Ayensu and De- 

Filipps 1978) that became part of a proposal to list 

nearly 1700 plant species as Threatened or Endangered 

in 1976 (the proposal was ultimately dismissed). 

These initial rare species lists were plagued by incomplete 

data and taxonomic problems. Of the 156 

Utah species considered endangered, threatened, or extinct 

by the Smithsonian Institution in 1975, only 39 

(25%) are still considered taxa of Extremely High or 

High conservation priority today. At least 13 of these 

species (8.3%) are no longer recognized as legitimate 

taxa. Another 28 species (18%) are now known to be 

much more common or less threatened and are classified 

as Low priority by UNPS. Eight of the nine species 

considered extinct in 1975 have been rediscovered (only 

Cuscuta warneri is still thought to be extirpated in 

Utah). Among the additional 225 state endemics and 

other potentially rare species evaluated by Welsh and 

others (1975), one half (112 taxa) are now scored as 

Low priority and 26 (11.5%) are no longer recognized 

taxonomically. 

Over the next two decades new Utah rare plant lists 

were developed by the US Fish and Wildlife Service, 

Bureau of Land Management, US Forest Service, and 

non-governmental organizations (such as The Nature 

Conservancy and Utah Native Plant Society). The composition 

of these lists continued to evolve to reflect 

ever-improving knowledge of the distribution, abundance, 

and status of the state’s flora (Atwood et al. 

1991; Utah Division of Wildlife Resources 1998; Utah 

Native Plant Society 1980, 1982; Welsh 1978; Welsh 

and Chatterley 1985; Welsh and Thorne 1979). Threat 

of potential listings under the ESA prompted a large 

scale effort to survey rare species and remote corners of 

the nation for new taxa. During the period from 1975 to 

1994 nearly 1200 new vascular plant taxa were described 

across North America, or approximately 60 new


species per year (Hartman and Nelson 1998). In Utah 

alone, over 250 new plant taxa have been named since 

the early 1970s (Welsh et al. 2008). Not surprisingly, 

most of these newly discovered species are narrow endemics 

with few or small populations and specialized 

habitat requirements, making them potential candidates 

for rare plant lists. Fifty-five percent of the current 

UNPS Extremely High priority list (17 species) and 

52% of the High priority list (62 species) have only 

been named since 1978. 

In the years since 1975, field surveys and taxonomic 

research have resulted in many plant taxa being removed 

from consideration as species of concern due to 

lack of threat, stable trends, or documented abundance. 

At least 772 of the current 2302 Utah plant taxa scored 

as Low priority by UNPS (33.5%) have been listed as 

potentially Endangered, Threatened, extinct, or otherwise 

rare at one time. Of the 28 Utah species that have 

been listed as Threatened or Endangered by the US Fish 

and Wildlife Service since 1978, four have subsequently 

been delisted (Astragalus perianus, Erigeron maguirei, 

Echinocereus engelmannii var. purpureus, and Echinocereus 

triglochidiatus var. inermis) because surveys 

have found them to be much more common, or the taxa 

are no longer recognized. 

Future Applications 

Rare plant lists have a short shelf life. The UNPS list 

has already been revised twice since it first appeared in 

2009 and will need to be updated again in the coming 

year. With the publication of the final volume of the 

Intermountain Flora (Holmgren et al. 2012) at least 36 

new native plant species have been documented in Utah 

which have not been evaluated by the UNPS Rare Plant 

Committee. Several of these species are narrow endemics 

that are likely to be ranked as Extremely High, High, 

or Watch list species when sufficient data are available 

for review. Other species currently on the Need Data 

list will also likely be placed in higher priority categories 

in the near future. Undoubtedly, there are more rare 

species still awaiting discovery in the years ahead. Results 

of on-going monitoring studies and field inventories 

will also improve our understanding of many species 

and result in shifts in their conservation priority. 

In addition to Utah, the Wyoming protocol has been 

recently applied to the entire flora of Wyoming (Fertig 

2011) and to Zion National Park (Fertig 2010b). In 

Zion, the park’s initial list of over 200 species of concern 

was streamlined to 51 taxa, of which only 13 were 

deemed Extremely High or High priority. Some species 

were given a slightly different rank in the park compared 

to the state as a whole, reflecting differences in 

scale and data sufficiency (Fertig 2010b). The Idaho 

and Arizona native plant societies have also expressed 

interest in using this methodology to rank rare plants in 

their respective states. As it is used more frequently, the 

protocol will hopefully be strengthened and improved. 

It is important to remember that the UNPS rare plant 

list has no binding legal authority and is only as 

accurate as the information used for ranking. The list 

and the listing process remain useful, however, because 

they provide a simple, repeatable, and transparent 

method to prioritize conservation action among hundreds 

of rare species. With conservation resources 

stretched thin and time running out, this form of triage 

may be critical to preserving Utah’s most vulnerable 

botanical treasures. 


These lists were developed with the input of my fellow 

members of the Utah Native Plant Society Rare 

Plant Committee: Duane Atwood (BYU herbarium, retured), 

Rita [Dodge] Reisor (Red Butte Garden), Robert 

Fitts (UT Conservation Data Center), Ben Franklin (UT 

Conservation Data Center, retired), and Jason Alexander 

(Utah Valley University). The original list benefited 

from input from more than 40 participants in the Utah 

rare plant breakout session at the 2009 Southwest Rare 

Plant Conference. Additional helpful comments and 

suggestions have been provided by attendees of the annual 

Utah Rare Plant meetings from 2010 to 2012. Special 

thanks to the following for their input on various 

iterations of the UNPS list: Ron Bolander (BLM state 

botanist), Jessie Brunson (USFWS), Debi Clark 

(Canyon De Chelley NM), Cheryl Decker (NPS SE 

Utah Group), Larry England (USFWS, retired), Tony 

Frates (UNPS Conservation Committee and webmaster), 

Kipp Lee (UNPS), Kezia Nielson (Zion NP), 

Teresa Prendusi (USFS regional botanist), Gary Reese 

(consultant), Daniela Roth (formerly USFWS), Jim 

Spencer (NRCS, Roosevelt UT), Blake Wellard 

(University of Utah grad student) and Dorde Woodruff 

(Utah cactus expert). My apologies (and thanks) to 

other contributors whom I have omitted inadvertently. 


Akçakaya, H.R., S. Ferson, M.A. Burgman, D.A. 

Keith, G.M. Mace, and C.R. Todd. 2000. Making consistent 

IUCN classifications under uncertainty. Conservation 

Biology 14 (4):1001-1013. 

Andelman, S.J., C. Groves, and H.M. Regan. 2004. 

A review of protocols for selecting species at risk in the 

context of US Forest Service viability assessments. 

Acta Oecologica 26:75-83. 

Atwood, D., J. Holland. R. Bolander, B. Franklin, 

D.E. House, L. Armstrong, K. Thorne, and L. England. 

1991. Utah Threatened, Endangered, and Sensitive 

Plant Field Guide. US Forest Service Intermountain 

Region, National Park Service, Bureau of Land Management, 

Utah Natural Heritage Program, US Fish and 

205


Wildlife Service, Environmental Protection Agency, 

Navajo Nation, and Skull Valley Goshute Tribe. 

Ayensu, E.S. and R.A. DeFilipps. 1978. Endangered 

and Threatened Plants of the United States. Smithsonian 

Institution and World Wildlife Fund, Washington, 

DC. 403 pp. 

Beissinger, S.R., J.M. Reed, J.M. Wunderle Jr., S.K. 

Robinson, and D.M. Finch. 2000. Report of the AOU 

conservation committee on the Partners in Flight species 

prioritization plan. The Auk 117:549-561. 

Breininger, D.R., M. Barkaszi, R.B. Smith, D.M. 

Oddy, and J.A. Provancha. 1998. Prioritizing wildlife 

taxa for biological diversity conservation at the local 

scale. Environmental Management 22:315-321. 

Center for Plant Conservation. 2000. America’s 

Vanishing Flora, Stories of Endangered Plants from the 

Fifty States and Efforts to Save Them. Center for Plant 

Conservation, St. Louis, MO. 59 pp. 

Corbin, B.L. 2011. A new species of Astragalus 

(Fabaceae) from the Wasatch Mountains of Utah. 

Madroño 58(3):185-189. 

Cracraft, J. 1987. Species concepts and the ontology 

of evolution. Biology and Philosophy 2:329-346. 

Faber-Langendoen, D., L. Master, J. Nichols, K. 

Snow, A. Tomaino, R. Bittman, G. Hammerson, B. Heidel, 

L. Ramsay, and B. Young. 2009. NatureServe conservation 

status assessments: Methodology for assigning 

ranks. NatureServe, Arlington, VA. 42 pp. 

Fertig, W. 2009. Developing a Utah rare plant list. 

Sego Lily 32(6):1-17. 

Fertig, W. 2010a. Updates to UNPS rare plant list. 

Sego Lily 33(6):15. 

Fertig, W. 2010b. Rare plants of Zion National 

Park. Moenave Botanical Consulting, Kanab, UT. 95 

pp. 

Fertig, W. 2011. Strategies for plant conservation in 

Wyoming: Distributional modeling, gap analysis, and 

identifying species at risk. Doctoral dissertation, Department 

of Botany, University of Wyoming, Laramie, 

WY. 451 pp. 

Fertig, W. 2012. Rare plant committee updates 

UNPS rare plant list. Sego Lily 35(3):7. 

Grady, B.R. and J.L. Reveal. 2011. New combinations 

and a new species of Eriogonum (Polygonaceae: 

Eriogonoideae) from the Great Basin Desert, United 

States. Phytotaxa 24:33-38. 

Greenwalt, L.A. 1975. Endangered and threatened 

wildlife and plants. Federal Register 40:44412-44429. 

Hartman, R.L. and B.E. Nelson. 1998. Taxonomic 

novelties from North America north of Mexico: A 20- 

year vascular plant diversity baseline. Monographs in 

Systematic Botany from the Missouri Botanical Garden 

67:1-59. 

Heidel, B. 2009. Survey and monitoring of Gibbens’ 

penstemon (Penstemon gibbensii) in south-central 

Wyoming. Report prepared for the Bureau of Land 

Management. Wyoming Natural Diversity Database, 

Laramie, WY. 36 pp. 

Holmgren, N.H. and P.K. Holmgren. 2011. A new 

species of Eremogone (Caryophyllaceae) from northern 

Utah and southeastern Idaho, U.S.A. Brittonia 63(1):1- 

6. 

Holmgren, N.H., P.K. Holmgren, J.L. Reveal, and 

collaborators. 2012. Intermountain Flora, Vascular 

Plants of the Intermountain U.S.A. Volume two, part A, 

subclasses Magnoliidae-Caryophyllidae. New York Botanical 

Garden, Bronx, NY. 731 pp. 

Holsinger, K.E. 1992. Setting priorities for regional 

plant conservation programs. Rhodora 94 (879):243- 

257. 

IUCN. 2001. IUCN Red List Categories and Criteria 

Version 3.1. IUCN Species Survival Commission. 

IUCN, Gland, Switzerland and Cambridge, UK. 30 pp. 

Johnson, L.A., L.M. Chan, K. Burr, and D. Hendrickson. 

2012. Navarretia furnissii (Polemoniaceae), 

a new diploid species from the intermountain western 

United States distinguished from tetraploid Navarretia 

saximontana. Phytotaxa 42:51-61. 

Keith, D.A. 1998. An evaluation and modification 

of World Conservation Union red list criteria for classification 

of extinction risk in vascular plants. Conservation 

Biology 12 (5):1076-1090. 

Mace, G.M., N.J. Collar, K.J. Gaston, C. Hilton- 

Taylor, H.R. Akçakaya, N. Leader-Williams, E.J. 

Milner-Gulland, and S.N. Stuart. 2008. Quantification 

of extinction risk: IUCN’s system for classifying threatened 

species. Conservation Biology 22(6):1424-1442. 

Master, L.L. 1991. Assessing threats and setting 

priorities for conservation. Conservation Biology 5: 

148-157. 

Master, L.L., B.A. Stein, L.S. Kutner, and G.A. 

Hammerson. 2000. Vanishing assets, conservation 

status of U.S. species. Pp. 93-118. In: Stein, B.A., L.S. 

Kutner, and J.S. Adams, eds. Precious Heritage, the 

Status of Biodiversity in the United States. The Nature 

Conservancy and Association for Biodiversity Information, 

Oxford University Press, New York. 

Millsap, B.A., J.A. Gore, D.E. Runde, and S.L. Cerulean. 

1990. Setting priorities for the conservation of 

fish and wildlife species in Florida. Wildlife Monographs 

111:1-57. 

Noss, R.F. and A.Y. Cooperrider. 1994. Saving Nature’s 

Legacy, Protecting and Restoring Biodiversity. 

Island Press, Washington, DC. 416 pp. 

O’Grady, J.J., M.A. Burgman, D.A. Keith, L.L. Master, 

S.J. Andelman, B.W. Brook, G.A. Hammerson, T. 

Regan, and R. Franklin. 2004. Correlations among extinction 

risks assessed by different systems of threatened 

species categorization. Conservation Biology 18 

(6):1624-1635. 

206


Panjabi, A.O., E.H. Dunn, P.J. Blancher, W.C. 

Hunter, B. Altman, J. Bart, C.J. Beardmore, H. Berlanga, 

G.S. Butcher, S.K. Davis, D.W. Demarest, R. 

Dettmers, W. Easton, H. Gomez de Silva Garza, E.E. 

Iñigo-Elias, D.N. Pashley, C.J. Ralph, T.D. Rich, K.V. 

Rosenberg, C.M. Rustay, J.M. Ruth, J.S. Wendt, and 

T.C. Will. 2005. The Partners in Flight handbook on 

species assessment. Version 2005. Partners in Flight 

Technical Series No. 3. Rocky Mountain Bird Observatory 

website: http://www.rmbo.org/downloads/ 

Handbook2005.pdf. 

Rabinowitz, D. 1981. Seven forms of rarity. Pp. 

205-217. In: Synge, H. (ed.). The Biological Aspects of 

Rare Plant Conservation. John Wiley and Sons, Ltd. 

Regan, T.J., M.A. Burgman, M.A. McCarthy, L.L. 

Master, D.A. Keith, G.M. Mace, and S.J. Andelman. 

2005. The consistency of extinction risk classification 

protocols. Conservation Biology 19 (6):1969-1977. 

Regan, T.J., L.L. Master, and G.A. Hammerson. 

2004. Capturing expert knowledge for threatened species 

assessments: a case study using NatureServe conservation 

status rank. Acta Oecologica 26:95-107. 

Spence, J. 2012. A new look at ranking plant rarity 

for conservation purposes, with an emphasis on the flora 

of the American Southwest. Calochortiana 1:25-34. 

Stein, B.A., L.S. Kutner, and J.S. Adams. 2000. 

Precious Heritage, the Status of Biodiversity in the 

United States. The Nature Conservancy and Association 

for Biodiversity Information, Oxford University 

Press, New York. 399 pp. 

US Fish and Wildlife Service. 1983. Endangered 

and Threatened species listing and recovery priority 

guidelines. Federal Register 48 (184):43098-43105. 

Utah Division of Wildlife Resources. 1998. Inventory 

of sensitive species and ecosystems in Utah. Endemic 

and rare plants of Utah: an overview of their distribution 

and status. Report prepared for Utah Reclamation 

Mitigation and Conservation Commission and US 

Department of Interior. 566 pp. + app. 

Utah Native Plant Society. 1980. Utah threatened 

and endangered plants. Utah Native Plant Society 

Newsletter 3(1):1-4. 

Utah Native Plant Society. 1982. Report on Utah 

rare/threatened/endangered plant conference. Sego Lily 

5(1):5-9. 

Welsh, S.L. 1978. Endangered and threatened plants 

of Utah: a reevaluation. Great Basin Naturalist 38:1-18. 

Welsh, S.L. and N.D. Atwood. 2009. Plant Endemism 

and Geoendemic Areas of Utah. Privately published. 

97 pp. 


Higgins. 2008. A Utah Flora, 2004-2008 summary 

monograph, fourth edition, revised. Brigham Young 

University, Provo, UT. 1019 pp. 

Welsh, S.L., N.D. Atwood, and J.L. Reveal. 1975. 

Endangered, threatened, extinct, endemic, and rare or 

restricted Utah vascular plants. Great Basin Naturalist 

35:327-376. 

Welsh, S.L. and L.M. Chatterley. 1985. Utah’s rare 

plants revisited. Great Basin Naturalist 45:173-236. 

Welsh, S.L. and K.H. Thorne. 1979. Illustrated 

Manual of Proposed Endangered and Threatened Plants 

of Utah. US Fish and Wildlife Service, Bureau of Land 

Management, and US Forest Service. 316 pp. 

207

Pot Score 

Min Score 

Trend 

Threat 

Intrin Rar 

Hab Spec 

# Indiv 

# Pops 

Range 


Appendix 1. UNPS Rare Plant List: Extremely High Priority Species 

The following table lists 31 species scored as Extremely High priority for conservation attention in Utah based on the 

Wyoming protocol ranking system. Species are listed alphabetically by family and scientific name. Synonyms for 

family and species names are included in parentheses. See text for an explanation of the seven ranking criteria and 

scoring methods used to derive the minimum and potential scores. County codes are explained in Table 3. Legal 

Status: Bureau of Land Management (BLM) and US Forest Service (USFS) Sensitive = S; US Fish and Wildlife Service 

(USFWS) Candidate = C, Endangered = E, Proposed = P; Threatened = T. 

Family Species Common Name 

County Dist. 

& Legal Status 

Agavaceae 

Yucca sterilis 

(Y. harrimaniae var. s.) 

Creeping yucca 2 1 1 U 1 1 U 6 8 Dch?, Uin; 

BLM:S 

Asteraceae 

(Compositae) 

Townsendia aprica 

Last Chance townsendia 

2 1 1 1 0 1 1 7 7 Emr, Sev, Way; 

USFWS: T 

Brassicaceae 

(Cruciferae) 

Lepidium barnebyanum 

Schoenocrambe 

argillacea 

(Hesperidanthus a.) 

Schoenocrambe 

barnebyi 

(Hesperidanthus b.) 

Schoenocrambe 

suffrutescens 

(Hesperidanthus s.) 

Barneby’s pepperwort 

2 1 1 1 U 1 U 6 8 Dch; USFWS: E 

Clay reed-mustard 2 1 1 1 U 1 U 6 8 Uin; USFWS: T 

Barneby’s reedmustard 

Shrubby reedmustard 

Cactaceae Pediocactus despainii Despain’s pincushion 

cactus 

2 1 1 1 U 1 U 6 8 Emr, Way; 

USFWS: E 

2 1 1 1 U 1 1 7 8 Dch, Uin; 

USFWS: E 

2 1 1 1 1 1 1 8 8 Emr, Way; 

USFWS: E 

Pediocactus winkleri 

Sclerocactus brevispinus 

(S. whipplei var. ilseae) 

Sclerocactus wrightiae 

Winkler’s pincushion 

cactus 

Pariette hookless 

cactus 

Wright’s fishhook 

cactus 

2 1 1 1 1 1 1 8 8 Way; USFWS: T 

2 1 1 1 1 1 U 7 8 Dch, Uin; 

USFWS: T 

2 0 1 1 1 1 1 7 7 Emr, Way; 

USFWS: E 

Chenopodiaceae Atriplex canescens var. 

gigantea 

(Not recognized by 

Holmgren et al. 2012) 

Giant fourwing 

saltbush 

2 1 1 1 U 1 U 6 8 Jub; BLM: S 

Cyperaceae Carex specuicola Navajo sedge 1 1 1 1 1 1 1 7 7 Snj; USFWS:T 

Originally on 

High priority list 

Fabaceae 

(Leguminosae) 

Astragalus 

ampullarioides 

Shivwits milkvetch 

2 1 1 1 1 1 1 8 8 Wsh; USFWS: E 

Astragalus anserinus 

Goose Creek milkvetch 

2 1 1 1 0 1 1 7 7 Box; BLM: S; 

USFS: S; 

USFWS: C 

208

Pot Score 

Min Score 

Trend 

Threat 

Intrin Rar 

Hab Spec 

# Indiv 

# Pops 

Range 


Appendix 1. UNPS Rare Plant List: Extremely High Priority Species, continued 


County Dist. 


Fabaceae 

(Leguminosae) 

Astragalus holmgreniorum 

Holmgren’s milkvetch 

2 1 1 1 1 1 1 8 8 Wsh; USFWS: E 

Astragalus iselyi Isely’s milkvetch 2 1 1 1 0 1 1 7 7 Grn, Snj; 

BLM:S; USFS:S 

Astragalus lentiginosus 

var. pohlii 

Pohl’s milkvetch 2 1 1 1 1 1 U 7 8 Toe; BLM: S 

Trifolium variegatum 

var. parunuweapensis 

Parunuweap 

clover 

2 1 1 1 U 1 U 6 8 Kan; BLM: S 

Hydrophyllaceae Phacelia argillacea Clay phacelia 2 1 1 1 0 1 1 7 7 Utah; USFWS: E 

Iridaceae 

Lamiaceae 

(Labiatae) 

Phacelia utahensis Utah phacelia 2 1 1 1 U 1 1 7 8 Snp, Sev; 

BLM:S 

Iris pariensis 

(Included in Iris missouriensis 

in FNA) 

Salvia columbariae var. 

argillacea 

Paria iris 2 1 1 U U 1 1 6 8 Kan 

Chinle chia 2 1 1 1 1 1 1 8 8 Kan, Wsh; BLM: 

S 

Loasaceae Mentzelia argillosa Arapien stickleaf 2 1 1 1 0 1 1 7 7 Snp, Sev; BLM: 

S 

Malvaceae Sphaeralcea gierischii Gierisch’s globemallow 

2 1 1 1 U 1 1 7 8 Wsh; BLM: S; 

USFWS:C 

Papaveraceae Arctomecon humilis Dwarf bearclaw 

poppy 

2 1 1 1 1 1 1 8 8 Wsh; USFWS:E 

Polemoniaceae 

Gilia caespitosa 

(Aliciella c.) 

Rabbit Valley 

gilia 

2 1 1 1 0 1 1 7 7 Way; BLM: S; 

USFS: S 

Ranunculaceae 

Ranunculus aestivalis 

(R. acris var. aestivalis) 

Autumn buttercup 2 1 1 1 1 1 1 8 8 Grf; USFWS:E 

Gibbens’ beardtongue 


Penstemon gibbensii 

2 1 1 1 U 1 U 6 8 Dag; BLM: S 

Penstemon grahamii 

Graham’s penstemon 

2 1 1 1 1 1 1 8 8 Crb, Uin; BLM: 

S; USFWS:P 

Penstemon scariosus 

var. albifluvis 

White River penstemon 

2 1 1 1 0 1 1 7 7 Uin; BLM: S; 

USFWS: C 

Violaceae Viola clauseniana Clausen’s violet 2 1 1 1 1 U 1 7 8 Wsh 

209

Pot Score 

Min Score 

Trend 

Threat 

Intrin Rar 

Hab Spec 

# Indiv 

# Pops 

Range 


Appendix 2. UNPS Rare Plant List: High Priority Species 

The following table lists 119 species scored as High Priority for conservation attention in Utah based on the Wyoming 

protocol ranking system. Species are listed alphabetically by family and scientific name, with synonyms in parentheses. 

See text for an explanation of the seven ranking criteria and scoring methods used to derive the minimum 

and potential scores. County codes are explained in Table 3. Legal Status: Bureau of Land Management (BLM) and 

US Forest Service (USFS) Sensitive = S; US Fish and Wildlife Service (USFWS) Candidate = C, Endangered = E, 

Proposed = P; Threatened = T. 


County Dist. 


Apiaceae 


Cymopterus coulteri 

Two-leaf springparsley 

2 1 U 1 0 1 U 5 7 Jub, Snp, Sev, 

Toe 

Cymopterus higginsii 

Higgins’ springparsley 

2 1 1 1 0 1 0 6 6 Kan 

Lomatium latilobum 

Canyonlands 

lomatium 

2 1 1 1 0 1 U 6 7 Grn, Snj; BLM: 

S; USFS: S 

Lomatium scabrum var. 

tripinnatum 

Virgin lomatium 2 1 U 1 0 1 U 5 7 Wsh 

Apocynaceae 

Cycladenia humilis var. 

jonesii 

(C. jonesii) 

Jones’ cycladenia 1 1 1 1 1 1 0 6 6 Emr, Grf, Grn, 

Kan; USFWS:T 

Asclepiadaceae Asclepias welshii Welsh’s milkweed 2 1 1 1 0 1 0 6 6 Kan; USFWS:T 

Asteraceae 

(Compositae) 

Ambrosia x sandersonii 

(Hymenoclea s.) 

Chrysothamnus nauseosus 

var. glareosus 

(Ericameria nauseosa 

var. glareosa) 

Cirsium virginense 

(included in C. mohavense 

in FNA) 

Enceliopsis nudicaulis 

var. bairdii 

Sanderson’s bursage 

Marysvale rabbitbrush 

2 1 1 0 1 1 U 6 7 Wsh 

2 1 1 1 0 U 1 6 7 Piu 

Virgin thistle 1 1 1 1 0 1 1 6 6 Wsh; BLM: S 

Baird’s nakedstem 2 1 1 1 0 1 1 7 7 Wsh 

Erigeron higginsii 

(included in E. canaani 

in FNA) 

Higgins’ daisy 2 1 1 1 0 1 U 6 7 Wsh 

Erigeron kachinensis Kachina daisy 2 1 1 1 0 1 U 6 7 Snj; BLM: S; 

USFS: S; originally 

on Watch 

list 

Erigeon vagus var. 

madsenii 

Madsen’s daisy 2 1 1 1 1 0 U 6 7 Grf, Irn, Kan 

Haplopappus armerioides 

var. gramineus 

(Stenotus a. var. g.) 

Grass goldenweed 2 1 U 1 0 1 U 5 7 Dch, Uit 

210

Pot Score 

Min Score 

Trend 

Threat 

Intrin Rar 

Hab Spec 

# Indiv 

# Pops 

Range 


Appendix 2. UNPS Rare Plant List: High Priority Species, continued 


County Dist. 


Asteraceae 

(Compositae) 

Haplopappus lignumviridis 

(Ericameria l.) 

Haplopappus scopulorum 

var. canonis 

Senecio castoreus 

(Packera c.) 

Senecio malmstenii 

(Packera m.) 

Senecio musiniensis 

(Packera m.) 

Thelesperma subnudum 

var. maliterrimum 

(T. pubescens) 

Townsendia goodrichii 

Townsendia jonesii var. 

lutea (included in T. 

aprica by some authors) 

Townsendia strigosa var. 

prolixa 

Viguiera soliceps 

(Heliomeris soliceps) 

Xylorhiza cronquistii 

Greenwood 

goldenbush 

Canyon spindly 

goldenbush 

Beaver Mountain 

groundsel 

2 1 1 1 1 0 U 6 7 Sev; BLM: S 

2 1 1 1 0 U U 5 7 Snj 

2 1 1 1 0 1 U 6 7 Bvr, Piu; USFS: 

S 

Podunk groundsel 2 1 1 1 1 0 U 6 7 Grf, Irn, Kan; 

USFS: S 

Musinea groundsel 2 1 1 1 U 0 U 5 7 Snp; USFS: S 

Uinta greenthread 2 1 1 1 0 U U 5 7 Dch, Uin; BLM: 

S; USFS: S 

Goodrich’s townsendia 

2 1 1 1 0 U U 5 7 Dch, Uin 

Sigurd townsendia 2 1 1 1 0 1 U 6 7 Jub, Piu, Sev; 

BLM: S; USFS: 

S 

Strigose townsendia 

2 1 U 1 0 1 U 5 7 Dch, Grn; BLM: 

S 

Tropic goldeneye 2 1 0 1 1 1 0 6 6 Kan 

Cronquist’s woodyaster 

2 1 1 1 1 0 U 6 7 Grf, Kan 

Xylorhiza glabriuscula 

var. linearifolia 

Boraginaceae Cryptantha grahamii Graham’s 

cryptanth 

Cryptantha semiglabra 

Moab woodyaster 2 1 U 1 0 1 U 5 7 Grf, Grn, Snj, 

Way 

Pipe Spring 

cryptanth 

2 1 U 1 0 1 U 5 7 Dch, Uin; BLM: 

S 

2 1 1 1 0 1 U 6 7 Wsh? 

Brassicaceae 

(Cruciferae) 

Arabis falcatoria 

(Boechera falcatoria) 

Arabis harrisonii 

(Boechera harrisonii) 

Draba ramulosa 

Falcate rockcress 2 1 1 1 0 U U 5 7 Box, Jub; USFS: 

S 

Harrison’s rockcress 

Belknap Peak 

draba 

2 1 1 1 0 U U 5 7 Uta 

2 1 1 1 0 1 U 6 7 Bvr, Piu; USFS: 

S 

Draba sobolifera Creeping draba 2 1 1 1 0 1 U 6 7 Bvr, Piu; USFS: 

S 

Lepidium integrifolium 

Entire-leaf pepperwort 

1 1 1 1 0 1 1 6 6 Bvr, Rch, Snp, 

Sev, Uin 

211

Pot Score 

Min Score 

Trend 

Threat 

Intrin Rar 

Hab Spec 

# Indiv 

# Pops 

Range 




County Dist. 


Brassicaceae 

(Cruciferae) 

Cactaceae 

Capparaceae 

(Cleomaceae) 


alpinum 


stellae 

Lepidium ostleri 

Physaria chambersii var. 

canaanii 

Physaria grahamii 

(some authors include P. 

acutifolia vars. repanda 

& purpurea) 

Physaria rubicundula 

var. tumulosa 

(P. t., Lesquerella t.) 

Ferocactus acanthodes 

(F. cylindraceus) 

Pediocactus sileri 

Sclerocactus 

wetlandicus 

(S. whipplei var. 

glaucus) 

Cleomella hillmanii var. 

goodrichii 

(C. palmeriana var. g.) 

Chenopodiaceae Krascheninnikovia 

lanata var. ruinina 

Crassulaceae 

Dudleya pulverulenta 

var. arizonica 

(D. arizonica) 

Wasatch pepperwort 

Stella’s pepperwort 

Ostler’s pepperwort 

Canaan Peak twinpod 

2 1 1 1 0 U U 5 7 Slt; USFS: S 

1 1 1 1 1 1 U 6 7 Grf, Kan 

2 1 1 1 0 1 U 6 7 Bvr; BLM: S; 

USFWS: C 

2 1 1 1 0 U U 5 7 Grf 

Graham’s twinpod 2 1 1 1 0 U U 5 7 Dch, Grn, Uin, 

Uta, Was 

Kodachrome bladderpod 

Desert barrel cactus 

Siler’s pincushion 

cactus 

Uinta Basin hookless 

cactus 

Goodrich’s stinkweed 

Ruin Park winterfat 

Arizona liveforever 

2 1 1 1 0 1 U 6 7 Kan; USFWS:E 

1 1 1 1 0 1 1 6 6 Wsh; originally 

on Watch list 

1 1 1 1 0 1 1 6 6 Kan, Wsh; 

USFWS:T 

2 0 1 1 0 1 1 6 6 Dch, Uin; 

USFWS:T 

(formerly on 

ExH list) 

2 1 1 1 0 1 U 6 7 Uin; BLM: S 

2 1 1 1 0 1 U 6 7 Grn, Snj 

1 1 1 1 0 1 1 6 6 Wsh 

Cuscutaceae Cuscuta warneri Warner’s dodder 1 1 1 U 1 1 1 6 7 Mil; may be extirpated 

in UT 

Cyperaceae 

Fabaceae 

(Leguminosae) 

Carex haysii 

(included in C. curatorum 

by some authors) 

Hays’ sedge 2 1 1 1 0 1 U 6 7 Wsh; originally 

on Watch list 

Astragalus ampullarius Gumbo milkvetch 1 1 1 1 1 1 U 6 7 Kan, Wsh; BLM: 

S 

Astragalus cronquistii 

Cronquist’s milkvetch 

2 1 1 1 0 1 0 6 6 Snj; BLM: S 

Astragalus cutleri Cutler’s milkvetch 2 1 1 1 0 U U 5 7 Snj 

Astragalus desereticus Deseret milkvetch 2 1 1 1 U 1 0 6 7 Uta; USFWS:T 

212

Pot Score 

Min Score 

Trend 

Threat 

Intrin Rar 

Hab Spec 

# Indiv 

# Pops 

Range 




County Dist. 


Fabaceae 

(Leguminosae) 

Astragalus diversifolius 

Meadow milkvetch 

1 1 1 1 0 1 1 6 6 Jub, Toe; USFS: 

S 

Astragalus equisolensis 

(A. desperatus var. 

neeseae) 

Horseshoe milkvetch 

2 1 1 1 0 1 U 6 7 Uin; BLM: S 

Astragalus hamiltonii 

Hamilton’s milkvetch 

2 1 1 1 0 1 U 6 7 Uin; BLM: S 

Astragalus harrisonii 

Harrison’s milkvetch 

2 1 1 1 0 1 U 6 7 Grf, Way 

Astragalus loanus 

Glenwood milkvetch 

2 1 1 1 0 U U 5 7 Sev; BLM: S 

Astragalus sabulosus 

var. sabulosus 

Astragalus sabulosus 

var. vehiculus 

Cisco milkvetch 2 1 1 1 0 1 U 6 7 Grn; BLM: S 

Stage milkvetch 2 1 1 1 0 1 U 6 7 Grn; BLM: S 

Astragalus serpens Plateau milkvetch 2 1 1 1 0 U U 5 7 Piu, Sev, Way 

Astragalus striatiflorus 

Escarpment milkvetch 

2 1 1 1 0 1 U 6 7 Kan, Wsh; BLM: 

S 

Astragalus welshii Welsh’s milkvetch 2 1 1 1 0 U U 5 7 Grf, Irn, Kan, 

Mill Piu, Way; 

BLM: S 

Trifolium friscanum 

(T. andersonii var. f.) 

Frisco clover 2 1 1 1 0 1 U 6 7 Bvr; BLM: S; 

USFWS: C 

Fagaceae 

Quercus gambelii var. 

bonina 

Goodhope oak 2 1 1 0 1 1 U 6 7 Snj 

Fumariaceae 

Corydalis caseana var. 

brachycarpa 

Gentianaceae Frasera ackermaniae Ackerman’s frasera 

Hydrangeaceae 


Jamesia americana var. 

macrocalyx 

Hydrophyllaceae Phacelia argylensis 

Case’s corydalis 2 1 1 0 0 1 1 6 6 Slt, Uta, Was, 

Web; USFS: S 

2 1 1 1 U 0 U 5 7 Uin; BLM: S 

Wasatch jamesia 2 1 1 1 0 U U 5 7 Jub, Slt, Uta, 

Was; USFS: S 

Argyle Canyon 

phacelia 

2 1 1 1 0 1 U 6 7 Dch; BLM: S 

Phacelia cephalotes Chinle phacelia 1 1 1 1 1 1 U 6 7 Kan, Snj, Wsh 

Phacelia cronquistiana 

Cronquist’s phacelia 

1 1 1 1 1 1 U 6 7 Kan; BLM: S 

Phacelia demissa var. 

heterotricha 

Brittle phacelia 2 1 0 1 1 1 U 6 7 Piu, Sev, Way 

213

Pot Score 

Min Score 

Trend 

Threat 

Intrin Rar 

Hab Spec 

# Indiv 

# Pops 

Range 




County Dist. 


Hydrophyllaceae 

Phacelia demissa var. 

minor 

Brittle phacelia 2 1 0 1 1 1 U 6 7 Dch, Uin 

Phacelia indecora Bluff phacelia 2 1 1 1 0 1 U 6 7 Snj; BLM: S; 

originally on 

Watch list 

Phacelia pulchella var. 

atwoodii 

Phacelia pulchella var. 

gooddingii 

Phacelia sabulonum 

(P. pulchella var. 

sabulonum) 

Atwood’s pretty 

phacelia 

Goodding’s pretty 

phacelia 

2 1 0 1 1 1 0 6 6 Kan; BLM: S 

1 1 1 1 1 1 U 6 7 Wsh 

Tompkins phacelia 2 1 0 1 1 1 U 6 7 Grf, Kan 

Loasaceae Mentzelia shultziorum Shultz’s stickleaf 2 1 1 1 0 1 U 6 7 Grn; BLM: S 

Malvaceae 

Petalonyx parryi 

Sphaeralcea fumariensis 

(S. grossulariifolia var. 

fumariensis) 

Sphaeralcea janeae 

Parry’s sandpaperplant 

Smoky Mountain 

globemallow 

Jane’s globemallow 

Sphaeralcea psoraloides Scurfpea globemallow 

Onagraceae Camissonia exilis Meager camissonia 

Oenothera caespitosa 

var. stellae 

Stella’s eveningprimrose 

1 1 1 1 0 1 1 6 6 Wsh; BLM: S 

2 1 1 1 0 1 U 6 7 Kan: BLM: S 

2 1 1 1 0 U U 5 7 Grn, Snj, Way; 

BLM: S 

2 1 1 1 0 U U 5 7 Emr, Grn, Way; 

BLM: S 

1 1 1 1 1 1 U 6 7 Kan 

2 1 1 1 0 1 U 6 7 Emr, Grf, Kan, 

Snp 

Oenothera murdockii 

Murdock’s evening-primrose 

2 1 1 1 0 1 U 6 7 Kan, Wsh; BLM: 

S 

Ophioglossaceae Botrychium lineare Slender moonwort 1 1 1 U 1 1 U 5 7 Slt; USFS: S 

Orchidaceae 

Cypripedium calceolus 

var. parviflorum 

Spiranthes diluvialis 

(S. romanzoffiana var. 

d.) 

Large yellow ladies-slipper 

1 1 1 0 1 1 1 6 6 Cch, Grn, Slt, 

Sum, Uta, Web; 

USFS: S 

Ute ladies-tresses 1 1 0 1 1 1 1 6 6 Cch, Dag, Dch, 

Grf, Slt, Toe, 

Uin, Uta, Way, 

Web; USFWS:T 

Poaceae Elymus simplex Alkali wildrye 1 1 1 1 U 1 1 6 7 Dag 

Polemoniaceae 

Gilia imperialis 

(G. latifolia var. i.) 

Cataract gilia 2 1 1 1 0 U U 5 7 Emr, Grf, Kan, 

Snj, Way 

214

Pot Score 

Min Score 

Trend 

Threat 

Intrin Rar 

Hab Spec 

# Indiv 

# Pops 

Range 




County Dist. 


Polemoniaceae 

Polygonaceae 

Gilia tenuis 

(Aliciella t.) 

Ipomopsis congesta var. 

ochroleuca 

Phlox hoodii var. 

madsenii 

Eriogonum brevicaule 

var. huberi 

(included in var. laxifolium 

by Holmgren et 

al. 2012) 


var. mitophyllum 

(E. mitophyllum) 


var. promiscuum 

(included in var. laxifolium 



Eriogonum corymbosum 

var. cronquistii 

(E. cronquistii) 


var. heilii 


var. matthewsiae 

(included in var. albiflorum 

by Holmgren et al. 

2012) 


var. smithii 

(E. smithii) 

Mussentuchit gilia 2 1 1 1 0 U U 5 7 Emr, Sev; BLM: 

S 

Arapien gilia 2 1 U 1 0 1 U 5 7 Snp, Sev 

Madsen’s carpet 

phlox 

Huber’s wild 

buckwheat 

Lost Creek wild 

buckwheat 

Mount Bartles 

wild buckwheat 

Cronqist’s wild 

buckwheat 

Heil’s wild buckwheat 

Springdale wild 

buckwheat 

Flat top wild buckwheat 

2 1 1 1 0 U U 5 7 Way 

2 1 1 1 0 U U 5 7 Dch 

2 1 1 1 0 1 U 6 7 Sev; BLM: S 

2 1 1 1 0 U U 5 7 Crb 

2 1 1 1 0 U U 5 7 Grf 

2 1 1 1 0 1 U 6 7 Way 

2 1 1 1 0 1 U 6 7 Wsh 

2 1 1 1 0 U U 5 7 Emr, Way; 

BLM: S 

Eriogonum esmeraldense 

var. tayei 

(Included in var. esmeraldense 



Eriogonum nummulare 

var. ammophilum 

(E. ammophilum) 

Eriogonum racemosum 

var. nobilis 

(included in E. zionis by 


Taye’s wild buckwheat 

Ibex wild buckwheat 

Bluff wild buckwheat 

2 1 1 1 0 U U 5 7 Sev 

2 1 1 1 0 U U 5 7 Mil; BLM: S 

2 1 1 1 0 U U 5 7 Kan, Snj; BLM: 

S 

215

Pot Score 

Min Score 

Trend 

Threat 

Intrin Rar 

Hab Spec 

# Indiv 

# Pops 

Range 




County Dist. 


Polygonaceae Eriogonum soredium Frisco wild buckwheat 

Portulacaceae 

(Montiaceae) 

Primulaceae 

Talinum thompsonii 

(Phemeranthus validulus) 

Dodecatheon dentatum 

var. utahense 

(D. utahense) 

Primula domensis 

Thompson’s 

talinum 

Hooker’s shooting-star 

House Range 

primrose 

2 1 1 1 0 1 U 6 7 Bvr; BLM: S 

2 1 1 1 0 U U 5 7 Emr; BLM: S 

2 1 1 1 1 0 0 6 6 Slt; USFS: S 

2 1 1 1 0 U U 5 7 Mil; BLM: S 

Primula maguirei 

Maguire’s primrose 

2 1 1 1 1 0 U 6 7 Cch; USFWS:T 

Ranunculaceae 

Aquilegia holmgrenii 

(formerly included in A. 

elegantula) 

Holmgren’s columbine 

2 1 1 1 0 U U 5 7 Grf 

Aquilegia rubicunda 

Link Trail columbine 

2 1 1 1 0 U U 5 7 Emr, Sev; USFS: 

S 

Aquilegia scopulorum 

var. goodrichii 

Goodrich’s columbine 

2 1 1 1 0 U U 5 7 Dch; BLM: S 

Rosaceae 

Ivesia shockleyi var. 

ostleri 

Shockley’s ivesia 2 1 1 1 0 U U 5 7 Bvr; BLM: S 

Aquarius paintbrush 


Castilleja aquariensis 

Castilleja parvula var. 

revealii 

Penstemon flowersii 

Penstemon goodrichii 

Penstemon x jonesii 

Reveal’s paintbrush 

Flowers’ penstemon 

Goodrich’s penstemon 

Fuchsia penstemon 

2 1 0 1 1 1 0 6 6 Grf; USFS: S 

2 1 1 1 1 0 U 6 7 Grf, Irn, Kan; 

USFS: S 

2 1 1 1 0 1 U 6 7 Dch, Uin 

2 1 1 1 0 1 U 6 7 Dch, Uin; BLM: 

S 

2 1 1 0 1 U U 5 7 Kan, Wsh 

Penstemon pinorum Pinyon penstemon 2 1 1 1 0 1 U 6 7 Irn; BLM: S; 

USFS: S 

Penstemon tidestromii 

(includes P. leptanthus) 

Tidestrom’s penstemon 

2 1 1 1 0 1 U 6 7 Jub, Snp, Uta 

Penstemon wardii Ward’s penstemon 2 1 1 1 0 1 U 6 7 Mil, Piu, Snp, 

Sev; BLM: S; 

USFS: S 

Violaceae Viola beckwithii Beckwith’s violet 1 1 1 U U 1 1 5 7 Box, Cch, Slt, 

Uta 

216

Pot Score 

Min Score 

Trend 

Threat 

Intrin Rar 

Hab Spec 

# Indiv 

# Pops 

Range 


Appendix 3. UNPS Rare Plant List: Watch List 

The following table lists 264 species on the Watch List for potential conservation attention in Utah based on the 

Wyoming protocol ranking system. Species are listed alphabetically by family and scientific name, with synonyms in 

parentheses. See text for an explanation of the seven ranking criteria and scoring methods used to derive the minimum 

and potential scores. County codes are explained in Table 3. Legal Status: Bureau of Land Management 

(BLM) and US Forest Service (USFS) Sensitive = S; US Fish and Wildlife Service (USFWS) Candidate = C, Endangered 

= E, Proposed = P; Threatened = T. 


County Dist. 


Adoxaceae Adoxa moschatellina Moschatel 1 1 1 1 0 U 1 5 6 Snj 

Agavaceae 

Apiaceae 


Agave utahensis var. 

utahensis 

Utah century plant 1 1 1 1 0 1 U 5 6 Wsh 

Nolina microcarpa Beargrass 1 1 1 0 0 1 1 5 5 Wsh 

Yucca kanabensis 

(Y. angustissima var. k.) 

Kanab yucca 1 1 1 1 1 0 U 5 6 Kan, Wsh 

Yucca schidigera Splinter yucca 1 1 1 0 0 1 1 5 5 Wsh; originally 

High priority 

Yucca toftiae 

(Y. angustissima var. 

toftiae) 

Toft’s yucca 2 1 1 1 0 0 U 5 6 Grf, Kan, Snj; 

originally High 

priority 

Angelica wheeleri Utah angelica 1 1 1 0 0 1 1 5 5 Cch, Jub, Piu, 

Slt, Sev, Uta; 

USFS:S 

Cymopterus acaulis var. 

parvus 

Cymopterus beckii 

Cymopterus evertii 

Cymopterus minimus 

Cymopterus trotteri 

(Oreoxis trotteri) 

Lomatium graveolens 

var. clarkii 

Small springparsley 

Beck’s springparsley 

Evert’s springparsley 

Least springparsley 

Trotter’s springparsley 

2 1 U 1 0 0 U 4 6 Mil, Toe 

1 1 0 1 0 1 1 5 5 Kan, Snj, Way; 

BLM: S; USFS: 

S 

1 1 1 1 0 U U 4 6 Uin 

2 1 1 1 0 0 U 5 6 Grf, Irn, Kan; 

USFS: S 

2 1 1 1 0 0 U 5 6 Grn; BLM: S 

Clark’s lomatium 2 1 1 1 0 0 0 5 5 Wsh 

Lomatium junceum Rush lomatium 2 1 1 0 0 U U 4 6 Emr, Grf, Sev, 

Way 

Musineon lineare Utah musineon 2 1 1 1 0 0 U 5 6 Box, Cch 

Asclepiadaceae 

Asclepias cutleri Cutler’s milkweed 1 1 1 1 0 1 U 5 6 Grn, Snj 

Cynanchum utahense Swallow-wort 1 1 1 0 0 1 1 5 5 Wsh 

Asteraceae 

(Compositae) 

Artemisia campestris 

var. petiolata 

Petiolate wormwood 

2 1 1 0 0 U U 4 6 Dch; USFS: S 

217

Pot Score 

Min Score 

Trend 

Threat 

Intrin Rar 

Hab Spec 

# Indiv 

# Pops 

Range 


Appendix 3. UNPS Rare Plant List: Watch List, continued 


County Dist. 


Asteraceae 

(Compositae) 

Artemisia nova var. 

duchesnicola 

Aster kingii var. 

barnebyana 

(Tonestus k. var. b., Herrickia 

k. var. b.) 

Aster kingii var. kingii 

(Tonestus k. var. k., Herrickia 

k. var. k.) 

Aster welshii 

(Symphyotrichum w.) 

Baccharis viminea var. 

atwoodii 

Chrysopsis jonesii 

(Heterotheca jonesii) 


var. iridis 


var. i.) 


var. psilocarpus 


var. psilocarpa) 

Cirsium eatonii var. 

harrisonii 

Duchesne sagebrush 

2 1 1 1 0 0 U 5 6 Uin 

Barneby’s aster 2 1 1 1 0 0 U 5 6 Jub, Mil; USFS: 

S 

King’s aster 2 1 1 1 0 0 U 5 6 Slt, Uta; USFS: 

S 

Welsh’s aster 1 1 1 1 0 1 U 5 6 Bvr, Dch, Grf, 

Irn, Kan, Piu, 

Sum, Uta, Wsh, 

Way 

Atwood’s seepwillow 

Jones’ goldenaster 

Rainbow rabbitbrush 

Huntington rabbitbrush 

2 1 U 0 0 1 U 4 6 Emr, Grn, Snj 

2 1 1 1 0 0 0 5 5 Grf, Kan, Wsh; 

USFS: S 

2 1 0 1 0 1 U 5 6 Snp, Sev 

2 1 1 0 0 U U 4 6 Crb, Dch, Emr, 

Sev, Was 

Harrison’s thistle 2 1 U 1 0 0 U 4 6 Bvr, Piu 

Cirsium joannae Joanna’s thistle 2 1 1 1 0 0 U 5 6 Kan, Wsh 

Cirsium murdockii 

(C. eatonii var. m.) 

Murdock’s thistle 2 1 0 1 0 U U 4 6 Dag, Dch, Uin 

Cirsium ownbeyi Ownbey’s thistle 1 1 1 1 0 1 U 5 6 Dag, Uin 

Enceliopsis argophylla 

Silverleaf enceliopsis 

1 U U 1 0 1 1 4 6 Wsh 

Erigeron arenarioides Wasatch daisy 2 1 1 1 0 0 0 5 5 Box, Slt, Toe, 

Uta, Web 

Erigeron canaani Canaan daisy 2 1 1 1 0 0 0 5 5 Kan, Wsh 

Erigeron carringtoniae 

(included in E. untermannii 

in FNA) 

Carrington’s daisy 2 1 1 1 0 0 U 5 6 Emr, Snp; USFS: 

S 

Erigeron cronquistii Cronquist’s daisy 2 1 1 1 0 0 U 5 6 Cch; BLM: S; 

USFS: S 

218

Pot Score 

Min Score 

Trend 

Threat 

Intrin Rar 

Hab Spec 

# Indiv 

# Pops 

Range 




County Dist. 


Asteraceae 

(Compositae) 

Erigeron garrettii Garrett’s daisy 2 1 1 1 0 0 U 5 6 Slt, Uta, Was; 

USFS:S 

Erigeron goodrichii Goodrich’s daisy 2 1 U 1 0 0 U 4 6 Dag, Dch, Sum?, 

Uin, Uta 

Erigeron huberi 

(included in E. radicatus 

in FNA) 

Erigeron maguirei 

(includes var. harrisonii) 

Huber’s daisy 2 1 U 1 0 0 U 4 6 Dch 

Maguire’s daisy 2 1 1 1 0 0 0 5 5 Emr, Way; 

BLM: S; 

USFWS former 

T 

Erigeron religiosus Religious daisy 2 0 1 1 0 1 0 5 5 Grf, Kan, Snj, 

Wsh 

Erigeron sionis 

(includes vars. sionis & 

trilobatus) 

Zion daisy 2 1 1 1 0 0 0 5 5 Grf, Irn, Kan, 

Wsh 

Erigeron untermannii Untermann’s daisy 2 1 0 1 0 U U 4 6 Dch; BLM: S; 

USFS: S 

Erigeron ursinus var. 

meyerae 

Meyer’s daisy 2 1 1 0 U 0 U 4 6 Wsh 

Erigeron zothecinus Alcove daisy 2 1 1 1 0 0 U 5 6 Grf, Grn, Kan, 

Snj 

Geraea canescens Desert sunflower 1 1 1 0 0 1 1 5 5 Wsh 

Gutierrezia pomariensis Orchard snakeweed 

Haplopappus racemosus 

var. sessiliflorus 

(Pyrrocoma racemosa 

var. sessiliflora) 

Haplopappus zionis 

(Ericameria z.) 

Hymenoxys helenioides 

(Picradenia helenioides) 

Racemose goldenweed 

Cedar Breaks 

goldenweed 

Sneezeweed hymenoxys 

2 1 U 1 0 0 U 4 6 Dch, Uin 

1 1 1 1 0 U U 4 6 Mil 

2 1 1 1 0 0 U 5 6 Grf, Irn, Kan; 

BLM: S 

1 1 1 0 1 U U 4 6 Crb, Emr, Grf, 

Snp, Sev, Way 

Hymenoxys lapidicola Rock hymenoxys 2 1 1 1 0 0 0 5 5 Uin; BLM: S 

Hymenoxys lemmonii Alkali hymenoxys 1 1 1 1 0 U U 4 6 Uin 

Layia platyglossa var. 

breviseta 

Lepidospartum latisquamum 

Perityle emoryi 

Coastal tidytips 1 1 1 1 0 U 1 5 6 Snj 

Nevada broom 1 1 U 1 0 1 U 4 6 Mil 

Emory’s rockdaisy 

1 1 1 0 0 1 1 5 5 Wsh 

219

Pot Score 

Min Score 

Trend 

Threat 

Intrin Rar 

Hab Spec 

# Indiv 

# Pops 

Range 




County Dist. 


Asteraceae 

(Compositae) 

Perityle specuicola Alcove rock-daisy 2 1 1 1 0 0 U 5 6 Grn, Snj; BLM: 

S 

Peucephyllum schottii Pygmy-cedar 1 1 1 1 0 1 U 5 6 Wsh 

Platyschkuhria integrifolia 

var. oblongifolia 

Senecio dimorphophyllus 

var. intermedius 

(Packera dimorphophylla 

var. intermedia) 

Senecio fremontii var. 

inexpectans 

San Juan bahia 1 1 1 1 0 U U 4 6 Snj 

La Sal groundsel 2 1 1 0 0 U U 4 6 Dch, Grn, Snj, 

Snp, Sum 

Unexpected 

groundsel 

2 1 1 1 0 0 U 5 6 Grn, Snj 

Senecio werneriifolius 

var. barkleyi 

Barkley’s groundsel 

220 

2 1 1 1 0 0 U 5 6 Grf, Kan 

Solidago spectabilis Nevada goldenrod 1 1 U 1 0 1 U 4 6 Mil, Wsh 

Sphaeromeria ruthiae 

(Artmeisia ruthiae) 

Stephanomeria tenuifolia 

var. myrioclada 

(S. minor var. myrioclada) 

Stephanomeria tenuifolia 

var. uintahensis 

(S. minor var. u.) 

Townsendia beamanii 

Ruth’s chickensage 

Slender wirelettuce 

2 1 1 1 0 0 0 5 5 Kan, Wsh 

1 1 1 1 0 U U 4 6 Box 

Uinta wire-lettuce 2 1 1 0 0 U U 4 6 Uin 

Beaman’s townsendia 

Townsendia condensata Cushion townsendia 

Townsendia mensana 

Townsendia montana 

var. caelilinesis 

Townsendia montana 

var. minima 

Xylorhiza confertifolia 

Plateau townsendia 

Skyline townsendia 

Bryce Canyon 

townsendia 

Henrieville 

woody-aster 

Boraginaceae Cryptantha barnebyi Barneby’s cryptanth 

2 1 1 0 U 0 U 4 6 Snj; BLM: S 

1 1 1 1 0 U U 4 6 Bvr, Piu 

2 0 U 1 0 1 U 4 6 Dch, Uin 

2 1 0 1 0 U U 4 6 Dch, Snp, Was 

2 1 1 1 0 0 U 5 6 Grf, Irn, Kan, 

Wsh 

2 1 0 1 0 1 U 5 6 Grf, Kan, Way 

2 1 0 1 0 1 U 5 6 Uin; BLM: S 

Cryptantha compacta Mound cryptanth 2 1 0 1 0 U U 4 6 Bvr, Mil, Toe; 

BLM: S 

Cryptantha creutzfeldtii 

Creutzfeldt’s 

cryptanth 

2 1 0 1 0 1 U 5 6 Crb, Emr; BLM: 

S; USFS: S

Pot Score 

Min Score 

Trend 

Threat 

Intrin Rar 

Hab Spec 

# Indiv 

# Pops 

Range 




County Dist. 


Boraginaceae Cryptantha elata Tall cryptanth 1 1 1 1 0 U U 4 6 Grn 

Brassicaceae 

(Cruciferae) 

Cryptantha johnstonii 

Cryptantha jonesiana 

Cryptantha ochroleuca 

(included in C. compacta 


Hackelia ibapensis 

Arabis shockleyi 

(Boechera s.) 

Arabis vivariensis 

(Boechera. fernaldiana 

ssp. v.) 

Descurainia pinnata var. 

paysonii 

(D. incisa var. p.) 

Johnston’s cryptanth 

San Rafael cryptanth 

Yellowish cryptanth 

Deep Creek stickseed 

Shockley’s rockcress 

2 1 1 0 0 1 U 5 6 Emr 

2 1 0 1 0 U U 4 6 Emr 

2 1 1 1 0 0 0 5 5 Grf; USFS: S 

2 1 1 1 0 0 U 5 6 Jub 

1 1 1 1 0 U U 4 6 Bvr, Jub, Mil, 

Toe 

Park rockcress 2 1 1 1 0 0 U 5 6 Uin; BLM: S 

Payson’s tansymustard 

1 1 1 1 0 U U 4 6 Grn, Snj, Uin 

Draba kassii Kass’ draba 2 1 1 1 0 0 U 5 6 Toe 

Draba maguirei var. 

burkei 

(D. burkei) 

Draba maguirei var. 

maguirei 

Lepidium huberi 


claronense 


heterophyllum 


neeseae 

Burke’s draba 2 1 1 1 0 0 U 5 6 Box, Mor, Web; 

USFS: S 

Maguire’s draba 2 1 U 1 0 0 U 4 6 Box, Cch, Web; 

USFS: S 

Huber’s pepperwort 

1 1 1 1 0 U U 4 6 Uin; BLM: S 

Claron pepperwort 2 1 1 1 0 0 U 5 6 Grf, Kan, Piu 

Cedar Canyon 

pepperwort 

Neese’s pepperwort 

2 1 1 0 0 1 U 5 6 Irn, Mil, Piu, Sev 

2 1 1 1 0 0 U 5 6 Grf; USFS: S 

Lepidium nanum Low pepperwort 1 1 1 1 0 U U 4 6 Toe 

Physaria acutifiolia var. 

purpurea (included in 

P. grahamii in FNA) 

Physaria arizonica 

(Lesquerella arizonica) 

Physaria chambersii var. 

sobolifera 

Purple twinpod 2 1 1 1 0 0 U 5 6 Emr, Grn, Sev, 

Way 

Arizona bladderpod 

1 1 1 1 0 U U 4 6 Grf, Kan, Wsh 

Claron twinpod 2 1 1 1 0 0 U 5 6 Grf 

Physaria floribunda Mesa twinpod 1 1 1 1 0 U U 4 6 Grn 

221

Pot Score 

Min Score 

Trend 

Threat 

Intrin Rar 

Hab Spec 

# Indiv 

# Pops 

Range 




County Dist. 


Brassicaceae 

(Cruciferae) 

Physaria garrettii 

(Lesquerella garrettii) 

Thelypodiopsis ambigua 

var. erecta 

Thelypodiopsis sagittata 

var. ovalifolia 

(Thelypodium sagittatum 

var. ovalifolium) 

Garrett’s twinpod 2 1 1 1 0 0 U 5 6 Dav, Slt, Uta, 

Was; USFS: S 

Kanab thelypody 1 1 1 1 0 1 U 5 6 Kan, Wsh?; 

BLM: S 

Palmer’s thelypody 

1 1 1 1 0 U U 4 6 Grf, Irn, Jub, 

Kan, Mil 

Thelypodium flexuosum Zigzag thelypody 1 1 1 1 0 U U 4 6 Bvr, Toe 

Buddlejaceae Buddleja utahensis Utah butterflybush 1 1 1 1 0 1 0 5 5 Wsh 

Cactaceae 

Echinocereus triglochidiatus 

var. mojavensis 

(E. mojavensis) 

Mohave claretcup 1 1 1 0 0 1 1 5 5 Bvr, Mil, Wsh 

Mamillaria tetrancistra Pincushion cactus 1 1 1 0 0 1 1 5 5 Wsh 

Neolloydia johnsonii 

(Echinomastus j.) 

Opuntia echinocarpa 

(Cylindropuntia echinocarpa) 

Opuntia pulchella 

(Grusonia p.) 

Sclerocactus blainei 

Johnson’s neolloydia 

222 

1 1 1 0 0 1 1 5 5 Wsh 

Pale cholla 1 1 1 0 0 1 1 5 5 Bvr?, Wsh 

Sand cholla 1 1 U 1 0 1 U 4 6 Box, Jub, Mil, 

Toe, Wsh? 

Blaine’s fishhook 

cactus 

1 1 U 1 0 1 1 5 6 Irn 

Caryophyllaceae Silene nachlingerae Jan’s catchfly 1 1 1 1 0 U U 4 6 Bvr; USFS: S 

Chenopodiaceae Atriplex gardneri var. 

bonnevillensis 

(A. bonnevillensis) 

Cuscutaceae 

Atriplex obovata 

Atriplex pleiantha 

(Proatriplex p.) 

Atriplex wolfii var. 

tenuissima 

Bonneville saltbush 

New Mexico saltbush 

Four Corners 

orach 

1 1 U 1 0 1 U 4 6 Jub, Mil 

1 1 U 1 0 1 U 4 6 Snj 

1 1 1 1 0 U U 4 6 Snj 

Slender orach 1 1 1 1 0 U U 4 6 Crb, Dch, Emr, 

Grf, Piu, Snp, 

Sev, Uin 

Corispermum welshii Welsh’s bugseed 1 1 U 1 0 1 U 4 6 Grf, Kan, Mil, 

Snj? 

Cuscuta applanata Winged dodder 1 1 1 0 1 U U 4 6 Wsh 

Cuscuta cuspidata Toothed dodder 1 1 1 0 1 U U 4 6 Slt, Uta, Web 

Cyperaceae Carex crawei Crawe’s sedge 1 1 1 1 0 U U 4 6 Kan 

Carex curatorum 

(C. scirpoidea var. c.) 

Canyonlands 

sedge 

1 1 1 1 0 U U 4 6 Kan, Snj, Uin

Pot Score 

Min Score 

Trend 

Threat 

Intrin Rar 

Hab Spec 

# Indiv 

# Pops 

Range 




County Dist. 


Cyperaceae Carex diandra Lesser panicled 

sedge 

1 1 1 1 0 U U 4 6 Dch, Grf? 

Carex lasiocarpa Slender sedge 1 1 1 1 0 U U 4 6 Dag, Sev?, Uin 

Carex leptalea Bristly-stalk sedge 1 1 1 1 0 U U 4 6 Dag, Dch, Uin 

Carex livida Pale sedge 1 1 1 1 0 U U 4 6 Dch, Uin 

Carex microglochin Subulate sedge 1 1 1 1 0 U U 4 6 Dag, Dch, Emr 

Cladium californicum Saw-grass 1 1 U 1 0 U 1 4 6 Kan, Snj 

Lipocarpha aristulata 

(L. drummondii, Hemicarpha 

micrantha) 

Scirpus nevadensis 

(Amphiscirpus n.) 

Slender-rush 1 1 1 1 0 1 U 5 6 Kan 

Nevada bulrush 1 1 1 1 0 U U 4 6 Jub, Rch 

Euphorbiaceae Euphorbia nephradenia Utah spurge 1 1 1 1 0 1 U 5 6 Emr, Grf, Kan, 

Way; BLM: S 

Fabaceae 

(Leguminosae) 

Astragalus calycosus 

var. monophyllidus 

One-leaf milkvetch 

1 1 1 1 0 U U 4 6 Sev 

Astragalus chloodes Grass milkvetch 2 1 0 1 0 U U 4 6 Uin 

Astragalus concordius 

(formerly included in A. 

piutensis) 

Hairy-pod milkvetch 

2 1 U 0 0 1 U 4 6 Irn, Wsh 

Astragalus detritalis Debris milkvetch 1 1 1 1 0 1 U 5 6 Dch, Uin 

Astragalus henrimontanensis 

(A. argophyllus 

var. stocksii) 

Dana’s milkvetch 2 1 1 0 0 U U 4 6 Grf; USFS: S 

Astragalus jejunus 


var. mokiacensis 

(A. m.; some authors 

include var. ursinus) 

Astragalus limnocharis 

var. limnocharis 

Astragalus limnocharis 

var. tabulaeus 

(included in A. montii by 

some authors) 

Starveling milkvetch 

1 1 1 1 0 U U 4 6 Rch; USFS: S 

Mokiak milkvetch 1 1 1 1 0 1 0 5 5 Wsh 

Navajo Lake milkvetch 

Table Cliff milkvetch 

2 1 1 1 0 0 U 5 6 Irn, Kan; USFS: 

S 

2 1 1 1 0 0 U 5 6 Grf; USFS: S 

Astragalus lutosus Dragon milkvetch 1 1 1 1 0 1 U 5 6 Dch, Uin, Uta, 

Was 

Astragalus malacoides 

Kaiparowits milkvetch 

2 1 1 1 0 0 U 5 6 Grf, Kan 

223

Pot Score 

Min Score 

Trend 

Threat 

Intrin Rar 

Hab Spec 

# Indiv 

# Pops 

Range 




County Dist. 


Fabaceae 

(Leguminosae) 

Astragalus montii 

Astragalus monumentalis 

Heliotrope milkvetch 

Monument milkvetch 

2 1 0 1 0 U U 4 6 Snp, Sev; 

USFWS: T 

2 1 0 1 0 U U 4 6 Grf, Snj 

Astragalus naturitensis Naturita milkvetch 1 1 1 1 0 U U 4 6 Snj 

Astragalus piscator Fisher milkvetch 2 1 1 1 0 0 U 5 6 Grn, Snj, Way 

Astragalus saurinus 

Dinosaur milkvetch 

2 1 0 1 0 U U 4 6 Uin 

Astragalus uncialis Currant milkvetch 2 1 1 0 0 1 U 5 6 Mil; USFS: S 

Astragalus wetherillii 

Astragalus zionis var. 

vigulus (“A. tephrodes”) 

Hedysarum boreale var. 

gremiale 

Hedysarum occidentale 

var. canone 

Oxytropis besseyi var. 

obnapiformis 

Oxytropis oreophila var. 

jonesii 

Pediomelum aromaticum 

var. aromaticum 


var. barnebyi 


var. tuhyi 

Wetherill’s milkvetch 

224 

1 1 1 0 U U 1 4 6 Grn 

Guard milkvetch 2 1 1 0 0 1 U 5 6 Wsh; USFS: S 

Rollins’ sweetvetch 

Coal Cliffs sweetvetch 

2 1 1 0 0 U U 4 6 Uin 

2 1 1 0 0 U U 4 6 Crb, Dch, Emr; 

USFS: S 

Maybell locoweed 1 1 U 1 0 1 U 4 6 Dag 

Jones’ locoweed 2 1 U 1 0 0 U 4 6 Emr, Grf, Grn, 

Irn, Snp, Uin 

Aromatic breadroot 

Barneby’s breadroot 

1 1 U 1 0 1 1 5 6 Emr?, Grn 

1 1 1 1 0 1 U 5 6 Kan, Wsh; BLM: 

S 

Tuhy’s breadroot 2 1 0 1 0 U U 4 6 SnJ; BLM: S 

Pediomelum epipsilum Kane breadroot 2 1 0 1 0 1 U 5 6 Kan; BLM:S 

Pediomelum mephiticum Skunk breadroot 1 1 1 0 0 1 1 5 5 Wsh 

Pediomelum pariense Paria breadroot 2 1 1 1 0 0 U 5 6 Grf, Kan; USFS: 

S 

Pediomelum retrorsum 

(P. megalanthum var. r.) 

Psoralidium lanceolatum 

var. stenostachys 

Psorothamnus arborescens 

var. pubescens 

Psorothamnus nummularius 

(P. polydenius 

var. jonesii) 

Peach Springs 

breadroot 

Rydberg’s scurfpea 

Beauty indigobush 

1 1 U 1 0 1 U 4 6 Wsh 

2 1 U 1 0 0 U 4 6 Dav, Jub, Mil, 

Slt, Toe, Web 

1 1 1 1 0 U U 4 6 Kan 

Jones’ indigo-bush 2 1 0 1 0 U U 4 6 Emr; BLM: S

Pot Score 

Min Score 

Trend 

Threat 

Intrin Rar 

Hab Spec 

# Indiv 

# Pops 

Range 




County Dist. 


Fabaceae 

(Leguminosae) 

Psorothamnus polydenius 

var. polydenius 

Glandular indigobush 

1 1 U 1 0 1 U 4 6 Wsh 

Trifolium beckwithii Beckwith’s clover 1 1 1 0 0 1 1 5 5 Piu?, Sev 

Gentianaceae 

Hydrangeaceae 


Swertia gypsicola 

(Frasera gypsicola) 


zionis 

White River swertia 

1 1 1 1 0 1 U 5 6 Mil; BLM: S 

Zion jamesia 2 1 1 1 0 0 0 5 5 Kan, Wsh; 

USFS: S 

Jamesia tetrapetala Basin jamesia 1 1 1 1 0 U U 4 6 Mil; BLM: S; 

USFS: S 

Hydrophyllaceae Phacelia austromontana Southern phacelia 1 1 1 0 0 1 1 5 5 Wsh 

Phacelia cottamii Cottam’s phacelia 2 1 0 1 0 1 U 5 6 Crb, Emr, Sev 

Phacelia glandulosa 

Phacelia mammillarensis 

Glandular scorpion-weed 

Nipple Bench phacelia 

1 1 U 1 0 1 U 4 6 Grn, Uin 

2 1 0 1 0 1 U 5 6 Grf, Kan 

Phacelia palmeri Palmer’s phacelia 1 1 1 1 0 1 0 5 5 Wsh 

Phacelia perityloides 

var. laxiflora 

Crevice phacelia 1 1 1 1 0 1 0 5 5 Wsh 

Phacelia salina 

Phacelia splendens 

Bitter Creek scorpion-weed 

Eastwood’s phacelia 

1 1 1 1 0 U U 4 6 Snp, Toe 

1 1 U 1 0 1 U 4 6 Gra 

Iridaceae 

Juncaceae 

Lamiaceae 

(Labiatae) 

Liliaceae 

Phacelia tetramera 

Four-parted phacelia 

1 1 1 0 1 U U 4 6 Web 

Tricardia watsonii Three hearts 1 1 1 0 U 1 1 5 6 Wsh 

Sisyrinchium douglasii 

(Olsynium d.) 

Juncus tweedyi 

(J. brevicaudatus) 

Stachys rothrockii 

Allium geyeri var. chatterleyi 

Purple-eyed grass 1 1 1 U 0 1 U 4 6 Toe 

Tweedy’s rush 1 1 1 0 0 1 1 5 5 Box 

Rothrock’s hedgenettle 

1 1 U 1 0 1 U 4 6 Kan 

Chatterley’s onion 2 1 1 0 0 U U 4 6 Snj; USFS: S 

Allium passeyi Passey’s onion 2 1 0 1 0 U U 4 6 Box 

Loasaceae Eucnide urens Desert rock-nettle 1 1 1 1 0 1 U 5 6 Wsh 

Mentzelia goodrichii 

Goodrich’s 

stickleaf 

2 1 1 1 0 0 U 5 6 Dch; BLM: S; 

USFS: S 

225

Pot Score 

Min Score 

Trend 

Threat 

Intrin Rar 

Hab Spec 

# Indiv 

# Pops 

Range 




County Dist. 


Loasaceae 

Malvaceae 

Najadaceae 

Mentzelia multicaulis 

var. flumensevera 



Sphaeralcea caespitosa 

var. caespitosa 

Najas caespitosa 

(included in N. flexilis 

by most authors) 

Sevier Canyon 

stickleaf 

Uinta Basin 

stickleaf 

Jones’ globemallow 

2 1 1 1 0 0 U 5 6 Piu, Sev 

1 1 U 1 0 1 U 4 6 Dch, Uin 

2 1 0 1 0 1 0 5 5 Bvr, Mil; BLM: 

S 

Fish Lake naiad 1 1 1 0 U 1 1 5 6 Sev; USFS: S 

Oleaceae Menodora spinescens Spiny menodora 1 1 1 1 0 1 0 5 5 Wsh 

Onagraceae Camissonia atwoodii Atwood’s camissonia 

2 1 0 1 0 1 U 5 6 Kan 

Camissonia bairdii Baird’s camissonia 1 1 1 1 0 1 U 5 6 Wsh; BLM: S 

Camissonia claviformis 

var. aurantiaca 


var. claviformis 


var. cruciformis 

Camissonia gouldlii 

Clubpod camissonia 



Gould’s camissonia 

1 1 1 0 0 1 1 5 5 Wsh 

1 1 1 0 0 1 1 5 5 Wsh 

1 1 1 0 0 1 1 5 5 Wsh 

1 1 1 1 0 1 U 5 6 Mil, Wsh; BLM: 

S 

Epilobium nevadense 

Oenothera deltoides var. 

decumbens 

Ophioglossaceae Botrychium multifidum 

Orchidaceae 

Papaveraceae 

Habenaria zothecina 

(Platanthera z.) 

Eschscholzia mexicana 

(E. californica var. m.) 

Nevada willowherb 

St. George 

evening-primrose 

Leathery grape 

fern 

1 1 1 1 0 U U 4 6 Irn, Mil, Wsh; 

BLM: S; USFS: 

S 

1 1 U 1 0 1 U 4 6 Wsh 

1 1 1 0 1 U U 4 6 Dch 

Alcove bog-orchid 1 1 1 1 0 1 U 5 6 Emr, Grf, Grn, 

Snj, Uin 

Mexican goldenpoppy 

1 1 1 0 0 1 1 5 5 Wsh 

Papaver coloradense 

(P. uintanense, P. 

kluanense, P. radicatum) 

Alpine Rocky 

Mountain poppy 

1 1 1 1 0 U U 4 6 Dag, Dch, Sum; 

USFS: S 

Platystemon californicus Creamcups 1 1 1 0 0 1 1 5 5 Wsh 

Poaceae 

(Gramineae) 

Andropogon glomeratus Bushy bluestem 1 1 1 1 0 U U 4 6 Grf, Kan, Snj, 

Way 

Festuca dasyclada Utah fescue 1 1 1 1 0 1 U 5 6 Emr, Grf, Snp 

226

Pot Score 

Min Score 

Trend 

Threat 

Intrin Rar 

Hab Spec 

# Indiv 

# Pops 

Range 




County Dist. 


Poaceae 

(Gramineae) 

Polemoniaceae 

Imperata brevifolia Satintail 1 1 1 1 0 U U 4 6 Kan, Snj 

Panicum hallii Hall’s panicgrass 1 1 1 0 U U 1 4 6 Bvr 

Stipa arnowiae Arnow’s ricegrass 1 1 1 0 1 U U 4 6 Grf, Grn, Irn, 

Jub, Kan, Uin, 

Wsh 

Ipomopsis spicata var. 

spicata 

(Gilia s. var. s.) 

Ipomopsis tridactyla 

(I. spicata ssp. t., Gilia 

tridactyla) 

Phlox lutescens 

(P. austromontana var. 

lutescens) 

Spike gilia 1 1 1 1 0 U U 4 6 Dag 

Cedar Breaks gilia 2 1 1 1 0 0 U 5 6 Irn, Piu 

Yellowish phlox 2 1 1 0 0 U U 4 6 Grf, Grn, Snj 

Phlox opalensis Opal phlox 1 1 1 1 0 1 U 5 6 Dag 

Polygonaceae Eriogonum acaule Stemless wild 

buckwheat 

Eriogonum aretioides 


var. loganum 

(E. loganum) 

Eriogonum cernuum var. 

psammophilum 

(var. not recognized by 



var. albiflorum 

(E. thompsoniae var. a.) 

Widtsoe wild 

buckwheat 

Logan wild buckwheat 

Sand Dune nodding 

wild buckwheat 

Virgin wild buckwheat 

1 1 1 1 0 U U 4 6 Rch 

2 1 1 1 0 0 U 5 6 Emr, Grf; USFS: 

S 

2 1 1 0 0 U U 4 6 Cch, Mor, Rch; 

USFS: S 

2 1 U 1 0 0 U 4 6 Grf, Kan, Snj; 

var. not recognized 

by Holmgren 

et al. (2012) 

1 1 U 1 0 1 U 4 6 Wsh 


var. aureum (sensu 

stricto) 

Eriogonum ephedroides 

(E. brevicaule var. 

ephedroides) 

Eriogonum exaltatum 

(E. insigne) 

Eriogonum heermannii 

var. subspinosum 

Eriogonum mortonianum 

Eriogonum scabrellum 

Golden buckwheat 2 1 U 1 0 0 U 4 6 Wsh; does not 

include ‘var. 

glutinosum’ 

Ephedra wild 

buckwheat 

Ladder wild buckwheat 

Tabeau Peak wild 

buckwheat 

Morton’s wild 

buckwheat 

Westwater wild 

buckwheat 

2 1 0 1 0 1 U 5 6 Uin 

1 1 1 1 0 1 0 5 5 Irn, Kan, Wsh 

1 1 1 1 0 1 0 5 5 Wsh 

2 1 1 0 0 U U 4 6 Kan 

1 1 1 1 0 U U 4 6 Emr, Grf, Grn, 

Kan, Snj 

227

Pot Score 

Min Score 

Trend 

Threat 

Intrin Rar 

Hab Spec 

# Indiv 

# Pops 

Range 




County Dist. 


Polygonaceae 

Polypodiaceae 

Primulaceae 

Ranunculaceae 

Eriogonum wrightii var. 

wrightii 

Wright’s wild 

buckwheat 

1 1 U 1 0 1 U 4 6 Wsh 

Koenigia islandica Koenigia 1 1 1 1 0 U U 4 6 Dch 

Pterostegia drymarioides 

Adiantum pedatum var. 

aleuticum 

Pterostegia 1 1 U 1 0 1 U 4 6 Wsh 

Northern maidenhair 

fern 

1 1 1 1 1 0 U 5 6 Grf, Slt, Wsh 

Cheilanthes wootonii Wooton’s lip-fern 1 1 1 1 0 1 0 5 5 Wsh 

Cystopteris bulbifera 

Gymnocarpium dryopteris 

Dodecatheon pulchellum 

var. zionense 

(includes subvar. huberi) 

Bulblet bladder 

fern 

1 1 1 1 0 U U 4 6 Slt, Snj, Wsh 

Oak fern 1 1 1 1 0 U U 4 6 Piu 

Zion shooting-star 1 1 1 1 0 0 1 5 5 Crb, Grn, Kan, 

Snj?, Wsh 

Primula specuicola Cave primrose 1 1 1 1 0 1 U 5 6 Grf, Grn, Kan, 

Snj, Way 

Aquilegia atwoodii 

(included in A. fosteri by 


Atwood’s columbine 

2 1 1 1 0 0 U 5 6 Uin; BLM: S 

Aquilegia barnebyi Shale columbine 1 1 1 1 0 1 U 5 6 Dch, Uin 

Aquilegia desolaticola 

Aquilegia fosteri 

(A. formosa var. fosteri, 

A. desertorum; may include 

A. atwoodii) 

Aquilegia grahamii 

(A. micrantha var. g.) 

Desolation Canyon 

columbine 

Foster’s columbine 

Graham’s columbine 

2 1 1 1 0 0 U 5 6 Grn; BLM: S 

2 1 1 1 0 0 0 5 5 Wsh 

2 1 1 1 0 0 U 5 6 Uin; USFS: S 

Rhamnaceae 

Rosaceae 

Aquilegia loriae 

(A. micrantha var. l.) 

Trautvetteria caroliniensis 

Ceanothus greggii var. 

franklinii 

Crataegus douglasii var. 

duchesnensis 

Lori’s columbine 2 1 1 1 0 0 0 5 5 Kan 

Carolina tassel-rue 1 1 1 1 0 U U 4 6 Snj 

Franklin’s desertlilac 

Duchesne hawthorn 

2 1 1 0 0 U U 4 6 Grf?, Grn, Snj 

2 1 1 0 0 U U 4 6 Dch, Uin, Was 

Ivesia utahensis Utah ivesia 2 1 1 0 0 1 U 5 6 Slt, Sum, Uta, 

Was; USFS: S 

Potentilla angelliae 

Angell’s cinquefoil 

228 

2 1 1 0 0 1 U 5 6 Way; USFS: S

Pot Score 

Min Score 

Trend 

Threat 

Intrin Rar 

Hab Spec 

# Indiv 

# Pops 

Range 




County Dist. 


Rosaceae Rubus neomexicanus New Mexico 

thimbleberry 

Rutaceae 

Ptelea trifoliata var. 

lutescens 

1 1 1 1 0 U U 4 6 Grf, Snj 

Hoptree 1 1 1 1 0 U 1 5 6 Grf?, Kan, Wsh 

Salicaceae Salix arizonica Arizona willow 1 1 1 1 0 1 U 5 6 Grf, Irn, Snp, 

Sev; USFS: S 

Saururaceae Anemopsis californica Yerba mansa 1 1 1 1 0 U U 4 6 Uta, Wsh 


Castilleja parvula var. 

parvula 

Maurandya antirrhiniflora 

Mimulus bigelovii var. 

cuspidatus 

Tushar paintbrush 2 1 1 1 0 0 U 5 6 Bvr, Grf, Piu; 

USFS: S 

Maurandya 1 1 1 0 0 1 1 5 5 Wsh 

Bigelow’s monkeyflower 

1 1 1 0 0 1 1 5 5 Wsh 

Mohavea breviflora Desert snapdragon 1 1 1 0 0 1 1 5 5 Wsh 

Penstemon abietinus Firleaf penstemon 2 1 1 0 0 U U 4 6 Sev, Uta 

Penstemon acaulis var. 

acaulis 

Penstemon ammophilus 

Penstemon angustifolius 

var. vernalensis 

Penstemon atwoodii 

Penstemon barbatus var. 

trichander 

Penstemon bracteatus 

Penstemon compactus 

Penstemon duchesnensis 

(P. dolius var. duchesnensis) 

Penstemon franklinii 

Stemless penstemon 

Sandloving penstemon 

2 1 0 1 0 1 U 5 6 Dag; BLM: S; 

USFS: S 

2 1 1 1 0 0 U 5 6 Grf, Kan, Wsh 

Vernal penstemon 2 1 1 0 0 1 U 5 6 Dag, Uin 

Atwood’s penstemon 

2 1 1 1 0 0 U 5 6 Grf, Kan 

Scarlet penstemon 1 1 1 1 0 U U 4 6 Snj 

Red Canyon penstemon 

Bear River penstemon 

Duchesne penstemon 

Franklin’s penstemon 

2 1 1 1 0 0 U 5 6 Grf; USFS: S 

2 1 1 1 0 0 U 5 6 Cch; USFS: S 

2 1 0 1 0 1 U 5 6 Dch 

2 1 1 0 0 1 U 5 6 Irn; BLM: S 

Penstemon idahoensis Idaho penstemon 1 1 1 1 0 U U 4 6 Box; BLM: S; 

USFS: S 

Penstemon marcusii 

Penstemon navajoa 

Marcus Jones’ 

penstemon 

Navajo Mountain 

penstemon 

2 1 U 1 0 1 U 5 7 Crb, Emr 

2 1 1 0 0 1 U 5 6 Snj 

229

Pot Score 

Min Score 

Trend 

Threat 

Intrin Rar 

Hab Spec 

# Indiv 

# Pops 

Range 




County Dist. 



Penstemon petiolatus 

Penstemon scariosus 

var. cyanomontanus 

Penstemon sepalulus 

Crevice penstemon 

Blue Mountain 

penstemon 

Littlecup penstemon 

1 1 1 1 0 1 0 5 5 Wsh 

2 1 1 1 0 0 U 5 6 Uin 

2 1 U 1 0 0 U 4 6 Jub, Uta, Wsh? 

Selaginellaceae Selaginella utahensis Utah spike-moss 2 1 1 1 0 0 0 5 5 Kan, Wsh 

Violaceae 

Viola frank-smithii 

Viola purpurea var. 

charlestonensis (V. 

charlestonensis) 

Bear River Range 

violet 

Charleston Mountain 

violet 

2 1 1 1 0 0 U 5 6 Cch; USFS: S 

1 1 1 1 0 1 0 5 5 Kan, Wsh; 

USFS: S 

Zygophyllaceae Fagonia laevis Fagonia 1 1 1 1 0 1 U 5 6 Wsh 

230


Appendix 4. UNPS Rare Plant List: Need Data List 

The following table includes 115 species with three or more ranking criteria scored as “unknown”. A large number 

of these species have only recently been named or discovered within Utah, and additional field surveys are needed to 

confirm their abundance, distribution, habitat needs, life history patterns, potential threats, and trends. Some species 

on the list have taxonomic questions that still need to be resolved. All of the plants included here have the potential 

to be ranked as extremely high or high priority, or as watch species, once needed studies are completed. Species are 

arranged alphabetically by family and species. Additional information is provided on county-level distribution (see 

Table 3 for codes) and data needs. Legal Status: Bureau of Land Management (BLM) and US Forest Service (USFS) 

Sensitive = S; US Fish and Wildlife Service (USFWS) Candidate = C. 

Family Species Common Name County Dist. & 

Legal Status 

Information Needed 

Apiaceae 


Asteraceae 

(Compositae) 

Cymopterus basalticus 

Cymopterus crawfordensis 

Artemisia biennis var. 

diffusa 

Artemisia parryi 

Artemisia tridentata var. 

parishii 



(Ericameria x u.) 

Crepis runcinata var. 

aculeolata 

Shadscale springparsley 

Crawford Mountain 

spring-parsley 

Mystery wormwood 

Parry’s wormwood 

Parish’s big sagebrush 

Bvr, Mil 

Rch 

Grf 

Grn, Snj 

SW UT 

Info needed on # of indiviuals, threats, & 

trends 

Recently named, info needed on # of individuals, 

habitat specificity, threats, trends 

Taxonomic questions, info needed on # of 

populations, intrinsic rarity, threats, trends 

Info needed on # of individuals, threats, & 

trends 

Info needed on distribution in UT, # of 

individuals, # of populations, threats, trends 

Uinta rabbitbrush Dag, Uin Info needed on # of individuals, threats, & 

trends 

Utah hawksbeard Kan Taxonomic questions, info needed on habitat 

specificity, threats, & trends 

Erigeron katiae Katie’s daisy Rch Newly described, info needed on habitat 


Erigeron mancus La Sal daisy Grn, Snj Recent research needs to be reviewed relating 

to # of populations, threats, & trends; 

USFS: S 

Erigeron watsonii Watson’s daisy Reported Info needed on # of individuals, habitat 


Haplopappus acaulis 

var. atwoodii 

(not recognized in FNA) 

Haplopappus crispus 

(Ericameria crispa) 

Haplopappus leverichii 

(Isocoma leverichii, I. 

humilis) 

Haplopappus racemosus 

var. paniculatus 


var. paniculata) 

Haplopappus racemous 

var. prionophyllus 


var. prionophylla) 

Atwood’s goldenweed 

Pine Valley 

goldenbush 

Canyon goldenweed 



Jub 

Mil?, Wsh; 

Wsh 

Mil 

Cch, Dch, Uta 

Treated as var. glabratus by Welsh et al. 

(2008); info needed on # of individuals, 

habitat specificity, threats, & trends 


trends; BLM: S; USFS: S 

Taxonomic questions, info needed on intrinsic 

rarity, threats, & trends; not seen 

since 1971 


trends 


trends 

231


Appendix 4. UNPS Rare Plant List: Need Data List, continued 


Legal Status 


Asteraceae 

(Compositae) 

Brassicaceae 

(Cruciferae) 

Hofmeisteria pluriseta 

(Pleurocoronis p.) 

Lygodesmia grandiflora 

var. doloresensis 

(L. doloresensis) 

Arrowleaf Wsh? Info needed on # of individuals, threats, & 

trends 

Dolores River 

skeletonplant 

Grn? 

Confirmation needed whether this species 

is in UT; info needed on intrinsic rarity, 

threats, & trends; BLM: S 

Senecio bairdii Baird’s groundsel Box Newly described, info needed on habitat 

specificity, intrinsic rarity, threats, trends 

Senecio streptanthifolius 

var. platylobus 

Senecio werneriifolius 

var. malmstenoides 

Arabis goodrichii 

(Boechera g.) 

Arabis holboellii var. 

derensis 

(Boechera inyoensis, 

included in A. beckwithii 


Arabis lasiocarpa 

(Boechera l.) 

Arabis perennans var. 

thorneae 

(Boechera selbyi var. t.) 

Arabis pulchra var. 

duchesnensis 

(Boechera duchesnensis) 

Arabis thompsonii 

(Boechera t., B. pallidifolia) 

Boechera glareosa 

(“Arabis glareosa”) 

Draba abajoensis 

(D. spectabilis var. glabrescens) 

Draba pedicellata var. 

pedicellata 

Wasatch groundsel Uta, Web 

Mt. Nebo groundsel 

Goodrich’s rockcress 

Desert Experimental 

Range rockcress 

Jub, Uta 

Mil 

Mil 

Wasatch rockcress Box, Cch, Rch, 

Slt, Uta 

Thorne’s rockcress Uin 

Duchesne rockcress 

Thompson’s rockcress 

Dch 

Snj 

Newly described, info needed on # of individuals, 

habitat specificity, threats, trends 


threats, & trends 

Newly described, info needed on habitat 

specificity, threats, & trends; BLM: S 


individuals, threats, & trends 

Info needed on habitat specificity, threats, 

& trends 

Recently described, info needed on habitat 


Taxonomic questions; info needed on # of 

individuals, habitat specificity, & trends 

Newly described, info needed on habitat 


Dorn’s rockcress Uin Recently described narrow endemic of CO 

& UT (holotype from S side of Blue Mountain), 

info needed on # of individuals, habitat 

specificity, number of populations, 


Abajo Peak draba Grn, Snj Recently described, info needed on # of 

individuals, intrinsic rarity, threats, & 

trends; USFS:S 

Cusick’s draba Toe Recently documented in UT, info needed 

on # of individuals, threats, and trends 

Draba pennellii Schell Creek draba Jub Recently documented in UT, info needed 

on # of individuals, threats, & trends; 

USFS: S 

Draba santaquinensis Santaquin draba Uta Recently described narrow endemic from 

southern Wasatch Range, info needed on # 

of individuals, habitat specificity, # of 

populations, threats, & trends; USFS: S 

232




Legal Status 


Brassicaceae 

(Cruciferae) 

Lepidium moabense 

(included in L. eastwoodiae 


Physaria acutifolia var. 

repanda (included in P. 

grahamii in FNA) 

Physaria hemiphysaria 

var. hemiphysaria 

Moab pepperplant Grf, Grn, Kan, Snj Taxonomic questions, info needed on # 

of populations, intrinsic rarity, & trends 

Indian Canyon 

twinpod 

Skyline bladderpod 

Crb, Dch, Emr, Sev, 

Uin, Uta, Was 

Dch, Emr, Snp, Uta, 

Was 

Info needed on # of individuals, threats, 

& trends 


& trends 

Physaria hemiphysaria 

var. lucens 

Physaria navajoensis 

(Lesquerella navajoensis) 

Tavaputs bladderpod 

Crb 


& trends 

Navajo bladderpod Kan? Taxonomic questions; info needed on # 

of individuals, intrinsic rarity, threats, & 

trends 

Physaria neeseae Neese’s twinpod Grf, Wsh? Newly described, info needed on # of 


Thelypodiopsis aurea Golden thelypody Snj Info needed on # of individuals, threats, 

& trends 

Thelypodiopsis vermicularis 

Wormwood thelypody 

Box, Irn, Jub, Mil, 

Snp, Sev, Toe, Uta 


& trends 

Thelypodium rollinsii 

Rollins’ thelypody Bvr, Crb, Jub, 

Mil, Piu, Snp, Sev 


& trends 

Cactaceae 

Coryphantha vivipara 

var. deserti 

Mohave pincushion 

cactus 

Irn, Wsh 

Info needed on # of individuals, number 

of populations, & trends 

Echinocactus polycephalus 

var. polycephalus 

Echinocactus polycephalus 

var. xeranthemoides 

(E. xeranthemoides) 

Opuntia pinkavae 

(O. basilaris var. woodburyi) 

Sclerocactus pubispinus 

var. pubispinus 

Manyhead barrel 

cactus 

Manyhead barrel 

cactus 

Pinkava’s pricklypear 

Great Basin fishhook 

Wsh? 

Kan? 

Kan, Wsh 

Bvr, Box, Irn, Jub, 

Mil, Toe 

Reports from UT need confirmation; 

info needed on # of individuals, # of 

populations, & trends 

Reports from UT need confirmation; 



May be represented by two forms in 

Utah; frequently hybridizes with other 

taxa; taxonomic problems; info needed 

on # of individuals, # of populations, & 

trends 

Info needed on # of individuals, # of 


Sclerocactus spinosior 

(S. pubispinus var. s.) 

Desert valley fishhook 

Jub, Mil, Sev 

Info needed on # of individuals, # of 


Caryophyllaceae Eremogone loisiae Lois’ sandwort Box, Cch, Dav, Jub, 

Rch, Slt, Snp, Toe, 

Uta, Web 

Recently named, info needed on # of 

individuals, threats, and trends 

233




Legal Status 


Chenopodiaceae Atriplex gardneri var. 

welshii (A. welshii) 

Fabaceae 

(Leguminosae) 

Atriplex powellii var. 

minuticarpa 

(A. minuticarpa) 

Astragalus brandegei 

Welsh’s saltbush Grn Taxonomic questions; info needed on # 

of individuals, threats, & trends 

Green River orach Emr, Gra, Way 

Brandegee’s milkvetch 

Emr, Grf, Irn, Piu, 

Sev, Way 


& trends 


& trends 

Astragalus callithrix 

Callaway milkvetch 

Mil 


& trends 

Astragalus desperatus 

var. petrophilus 

Rock-loving milkvetch 

Emr 


& trends 

Astragalus eastwoodiae 

Eastwood’s milkvetch 

Emr, Grf, Grn, Snj, 

Way 


& trends 

Astragalus hornii Horn’s milkvetch Wsh? Reports for UT need confirmation; info 

needed on # of individuals, # of populations, 

& trends 

Astragalus kelseyae 

Astragalus laccoliticus 

(A. chamaeleuce var. 

laccoliticus) 


var. negundo 


var. stramineus 

Kelsey’s milkvetch 

Laccolite milkvetch 

Box Elder freckled 

milkvetch 

Web 

Grf, Way 

Box 

Newly described narrow endemic, info 

needed on # of individuals, habitat 

specificity, intrinsic rarity, & trends 


& trends 

Newly described, info needed on # of 


Straw milkvetch Wsh? Reports from UT need confirmation; 



Astragalus pardalinus Panther milkvetch Emr, Grf, Grn, Way Info needed on # of individuals, threats, 

& trends 

Astragalus pattersonii 

Patterson’s milkvetch 

Crb, Emr, Grf, Snj, 

Sev, Uin, Way 


& trends 

Astragalus pinonis Pinyon milkvetch Bvr, Jub Info needed on habitat specificity, 


Astragalus preussii var. 

laxiflorus 

Littlefield milkvetch 

Wsh? 

Reports for UT need confirmation; info 


& trends 

Astragalus pubentissimus 

var. peabodianus 

Astragalus rafaelensis 

Peabody’s milkvetch 

San Rafael milkvetch 

Emr, Grn Taxonomic questions, info needed on # 

of individuals, threats, & trends; BLM: 

S 

Emr, Grn 


& trends 

Astragalus woodruffii 

Woodruff’s milkvetch 

Emr, Grf, Way 


& trends 

234




Legal Status 


Fabaceae 

(Leguminosae) 

Dalea flavescens var. 

epica (D. epica) 

Hole-in-the-Rock 

prairie-clover 

Grf, Snj 


individuals, threats, & trends; BLM: S 

Lupinus flavoculatus Yellow-eye lupine Wsh Info needed on # of individuals, intrinsic 

rarity, & trends 

Pediomelum castoreum 

Trifolium andinum var. 

canone 

Beaver Dam 

breadroot 

Canyon Mountains 

clover 

Wsh? 

Mil 

Reports for UT need confirmation; info 


& trends 




navajoense 


wahwahense 

Vicia americana var. 

lathyroides 

Navajo clover Snj Newly described, info needed on habitat 


Wah Wah clover Bvr Newly described, info needed on # of individuals, 


Pavant vetch Mil Newly described, info needed on habitat 


Gentianaceae Lomatogonium rotatum Marsh felwort Dag Info needed on # of individuals, threats, & 

trends 

Hydrangeaceae 



rosea 

Hydrophyllaceae Phacelia crenulata var. 

orbicularis 

Rosy cliff jamesia Irn 

Henry Mountains 

phacelia 

Grf, Way 


populations, intrinsic rarity, threats, trends 


trends 

Phacelia petrosa Forgotten phacelia Grf, Snj Info needed on # of individuals, threats, & 

trends 

Liliaceae Calochortus ciscoensis Cisco mariposa Dch, Grn, Uin Newly described, info needed on # of individuals, 


Loasaceae 


var. librina 

Horse Canyon 

stickleaf 

Crb, Emr 


trends; BLM: S 

Mentzelia thompsonii 

Thompson’s 

stickleaf 

Grn, Uin 


trends 

Petalonyx nitidus 

Shiny-leaf sandpaper-plant 

Wsh? 

Reports from UT need confirmation; info 


& trends 

Nyctaginaceae 

Abronia fragrans var. 

harrisii 

Harris’ fragrant 

sand-verbena 

Emr, Grf, Uin 


populations, threats, & trends 

Onagraceae Camissonia bolanderi Bolander’s camissonia 

Emr, Way? 



Ophioglossaceae Botrychium boreale 

(B. pinnatum) 

Northern grapefern 

Sum 

Taxonomic questions; info needed on habitat 

specificity, intrinsic rarity, threats, & 

trends 

Botrychium crenulatum Dainty moonwort Was Info needed on # of individuals, # of populations, 

threats, & trends; USFS: S 

235




Legal Status 


Ophioglossaceae Botrychium hesperium 

Papaveraceae 

Poaceae 

(Gramineae) 

Western moonwort 

Botrychium lanceolatum Lance-leaf grapefern 

Jub, Sum 

Jub 

Confirmation needed, info needed on # of 

individuals, # of populations, threats, & 

trends 

Info needed on habitat specificity, intrinsic 

rarity, threats, & trends 

Botrychium paradoxum Paradox moonwort Grf Confirmation needed; info needed on # of 

individuals, # of populations, threats, & 

trends; USFS: S 

Argemone corymbosa 

var. parva (A. parva) 

Bouteloua uniflora 

San Rafael 

prickly-poppy 

Grf, Grn, Snj, 

One-flower grama Reported, Zion 

NP 

Recently described; info needed on # of 


Confirmation needed; info needed on # of 

individuals, habitat specificity, # of populations, 


Leersia oryzoides Rice cutgrass Dav, Uta, Web Info needed on # of individuals, threats, & 

trends 

Stipa scribneri 

(Achnatherum s.) 

Scribner needlegrass 

Way 

Info needed on # of populations, threats, & 

trends 

Polemoniaceae 

Ipomopsis congesta var. 

goodrichii 

Goodrich gilia Dch Info needed on # of individuals, threats, & 

trends 

Langloisia schottii 

(Loeseliastrum s.) 

Navarretia furnissii 

Schott’s langloisia Wsh 

Furniss’s navarettia 

Cch, Sum, Was 


trends 

Recently named, info needed on # of individuals, 

intrinsic rarity, threats, & trends 

Phlox albomarginata 

White-margined 

phlox 

Rch 

Info needed on # of individuals, habitat 


Phlox austromontana 

var. jonesii 

(P. jonesii) 

Phlox austromontana 

var. prostrata 

Jones’ phlox Kan, Wsh Taxonomic questions; info needed on # of 


Silver Reef phlox Kan, Wsh Taxonomic questions; info needed on # of 


Polygonaceae 


var. viridulum 

(E. viridulum) 

Duchesne wild 

buckwheat 

Dch, Uin 


trends 

Eriogonum contortum 

Grand Valley wild 

buckwheat 

Emr, Grn 


trends 


var. hylophilum 

(E. hylophilum) 

Gate Canyon wild 

buckwheat 

Dch 


trends 


var. nilesii 

Las Vegas wild 

buckwheat 

Kan? 

Authenticity of UT reports has been questioned 

by James Reveal (Holmgren et 

al.2012); info needed on # of individuals, 

habitat specificity, # of populations, & 

trends; BLM: S; USFWS: C 

236




Legal Status 


Polygonaceae 

Rosaceae 



var. revealianum 

Eriogonum domitum 

Reveal’s wild 

buckwheat 

House Range wild 

buckwheat 

Eriogonum howellianum Howell’s wild 

buckwheat 

Eriogonum jamesii var. 

higginsii 

(E. arcuatum) 

Eriogonum lonchophyllum 

var. lonchophyllum 

Eriogonum microthecum 

var. tegetiforme 

(Lumped with var. lapidicola 



Eriogonum panguicense 

var. alpestre 

Eriogonum spathulatum 

var. kayeae 

(included in var. spathulatum 

by Holmgren et al. 

2012), Welsh et al. 

(2008) include E. artificis. 

Eriogonum spathulatum 

var. natum 

(E. natum) 

Potentilla diversifolia 

var. madsenii 

Penstemon acaulis var. 

yampaensis 

(P. yampaensis) 

Penstemon cyananthus 

var. judyae 

Higgins’ wild 

buckwheat 

Longleaf wild 

buckwheat 

Grf, Kan, Piu, 

Way 

Mil 

Jub, Mil, Toe 

Snj 

Emr, Grn, Snj, 

Uin 

Slender buckwheat Jub?, Mil, Wsh 

Cedar Breaks wild 

buckwheat 

Kaye’s wild buckwheat 

Son’s wild buckwheat 

Madsen’s cinquefoil 

Irn 

Bvr 

Mil 

Kan 

Yampa penstemon Dag 

Judy’s penstemon Uta 

var. heilii recently pulled out, updated 

status info needed on remaining pops, including 

# of individuals, threats, trends 

Described in 2011, endemic to House 

Range; info needed on # of individuals, 

habitat specificity, threats, & trends 


trends 

Variety not recognized by Reveal in Holmgren 

et al. (2012); info needed on # of individuals, 


Vars not recognized by Holmgren et al. 

(2012), including var. saurinum; info 

needed on # of individuals, threats, trends 



Taxonomic questions; info needed on 

threats & trends (move to Watch list in 

future) 

Taxonomic questions. E. artificis considered 

a separate species by Reveal in Holmgren 

et al. (2012); info needed on # of individuals, 



trends 




trends 

Recently described, info needed on # of 


Penstemon moffatii 

Mofatt penstemon Dch, Emr, Grf, 

Grn, Snj, Uta, 

Way 


trends 

Penstemon nanus Dwarf penstemon Bvr, Irn?, Mil Info needed on # of individuals, threats, & 

trends 

237



PO Box 520041 

Salt Lake City, UT 84152-0041 

The Utah Native Plant Society was founded in 1978 

with a mission to promote the conservation, appreciation, 

and stewardship of native plants in the wild and in 

home cultivation. Through its publications, annual 

member meeting/dinner, field trips, and chapter meetings, 

UNPS is active in connecting citizens of Utah and 

the west with the native flora that makes the Beehive 

State so special. UNPS also funds an annual scholarship 

and small grants program using proceeds from its 

on-line store and generous contributions from members. 

Members of UNPS receive the Society’s bimonthly 

newsletter, the Sego Lily, discounts on posters, cds, and 

other merchandise at the UNPS store (www.unps.org), 

and are enrolled in their local chapter. 

Join hundreds of other Utah plant lovers (or plant 

lovers from other states or countries) in becoming a 

member of UNPS by submitting the form below or going 

online (www.unps.org). 

Utah Native Plant Society Membership 

__ New Member 

__ Renewal 

__ Gift Membership 

Membership Category 

__ Student $9.00 

__ Senior $12.00 

__ Individual $15.00 

__ Household $25.00 

__ Sustaining $40.00 

__ Supporting Organization $55.00 

__ Corporate $500.00 

__ Lifetime $250.00 

Mailing 

___ US Mail 

___ Electronic 

Contribution to UNPS scholarship fund ____ $ 

Name _________________________________ 

Street _________________________________ 

City ______________________ State ________ 

Zip ___________ 

Email ___________________ 

Chapter _______________________________ 

__ Please send a complimentary copy of the Sego Lily to 

the above individual. 

Please enclose a check, payable to Utah Native Plant Society 

and send to: 


PO Box 520041 

Salt Lake City, UT 84152-0041 

Join or renew on-line at unps.org 

238

December 2012 Number 1 - Utah Native Plant Society

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?