Ranking Tool Created for Medicinal plants at Risk of Being Overharvested in the Wild


 

 

Research Communica on 

Ethnobiology Le ers. 2014. 5: 77‐88. DOI: 10.14237/ebl.5.2014.169. 77 

mation about plant populations was unavailable, the 
organization deliberately chose to be conservative in 
their assessments and include species on the list 
(Gladstar 2000). Plants for which multiple species 
may be harvested and sold under the same common 
name (e.g. “Echinacea” or “eyebright”) were scored 
collectively as a genus. As of December 2012, the 
United Plant Savers At-Risk List included 14 species 
and 7 genera, while the To-Watch List included 17 
species and 5 genera. 

The At-Risk and To-Watch Lists immediately 
filled a unique role in plant conservation. Many other 
agencies have created lists: the federal Endangered 
Species Act works to protect the very rarest of 
species; NatureServe provides a standard ranking 
system used by all US States to score plant species 
based on rarity and abundance; and the Convention 
on International Trade in Endangered Species 
(CITES) regulates international trade in threatened 
species. None of these other lists, however, considers 
issues specific to medicinal plants, including market 
forces and method of harvest. The At-Risk and To-
Watch Lists were therefore widely disseminated and 
used by different audiences. Governmental agencies, 

Introduction 
Competing interests for land use, a growing human 
population, and a growing herbal products industry 
place pressure on populations of wild-harvested 
medicinal plants. As a reaction to these growing 
threats, the United Plant Savers (UpS) formed in 1995 
with a mission to “protect native medicinal plants of 
the United States and Canada and their native habitats 
while ensuring an abundant supply of medicinal 
plants for generations for come” (UpS 2013). One 
step towards accomplishing this mission was to create 
lists of medicinal plant species deemed most vulnera-
ble to over-harvest (“At-Risk”) and those less 
vulnerable but still of great concern (“To-Watch”).  

First officially published in 2000 as part of the 
United Plant Savers’ Planting the Future book (Gladstar 
and Hirsch), the At-Risk and To-Watch Lists were 
developed following a long series of discussions that 
included input from herbalists, ecologists, land 
managers and herb growers. The listed plants were 
considered to be sensitive to human activity based on 
market analysis, habitat specificity, impacts of harvest, 
and lack of techniques or material for large scale 
cultivation (Gladstar 2000). Where scientific infor-

Ranking Tool Created for Medicinal plants at Risk of Being 
Overharvested in the Wild 

Lisa Marie Castle
1*
, Susan Leopold

2
, Rachel Cra

3
, Kelly Kindscher

3 

Author address: 
1
Southwestern Oklahoma State University, Department of Biological Sciences, 100 Campus Drive, Weather‐

ford,  OK  73096 
2
United  Plant  Savers,  PO  Box  776,  Athens,  OH  45701, 

3
Kansas  Biological  Survey,  2101  Constant  Ave., 

Lawrence, KS 66047
 

*
Corresponding Author: lisa.castle@swosu.edu  

Received: September 26, 2013  Volume 5:77‐88 
Published: May 30, 2014  © 2014 Society of Ethnobiology 

Abstract:  We developed an adaptable, transparent tool that can be used to quan fy and compare vulnerability to overhar-
vest for wild collected medicinal plants. Subsequently, we are crea ng a list of the most threatened medicinal plants in 
temperate North America. The new tool scores species according to their life history, the effects of harvest, their abundance 
and range, habitat, and demand. The resul ng rankings, based on explicit criteria rather than expert opinion, will make it 
easier to discuss areas of vulnerability and set conserva on priori es. Here we present scores for 40 species assessed using 
the At-Risk Tool and discuss the traits that led to different scores for six example species: echinacea (Echinacea angus folia 
DC. Asteraceae), peyote (Lophophora williamsii (Lem. ex Salm-Dyck) J.M. Coult. Cactaceae), sandalwood (Santalum spp. L. 
Santalaceae), s nging ne le (Ur ca  dioica L. Ur caceae), American ginseng (Panax  quinquefolius  L. Araliaceae) and 
mayapple (Podophyllum peltatum L. Berberidaceae). 

Keywords: medicinal plants, sustainable harvest, plant conserva on 



 

 

Research Communica on 

Ethnobiology Le ers. 2014. 5: 77‐88. DOI: 10.14237/ebl.5.2014.169. 78 

conservation organizations, popular media, herb 
sellers and growers, consumers, and herbalists have all 
referred to the UpS lists (Cech 1998, Westfall and 
Glickman 2004, McCoy et al. 2007, Roberson 2008). 
As use of the lists has expanded, so has recognition of 
their limitations. Despite a stated desire for the lists to 
be continuously reviewed and for species to be 
removed from the lists as the threat of over-harvest is 
reduced (Gladstar 2000), the difficulties of including 
input from so many stakeholders made the lists 
effectively static. Neither additional plants nor 
additional information could be easily added. Without 
publicly available criteria, inclusion on the At-Risk 
List did not provide any information about the factors 
causing a species to be vulnerable, information about 
areas of greatest conservation concern, or means of 
comparing one species to another.  

Working with the United Plant Savers, we set out 
to create a tool to answer these criticisms and better 
aid in setting conservation priorities for wild-
harvested medicinal plants. Our primary goal was to 
create a tool that: 1) meets the needs of the diverse 
users of the United Plant Savers lists, 2) leads to 
numerical scores of vulnerability to over-harvest that 
are comparable across species, 3) is based on explicit, 
science-based criteria, and 4) is transparent and 
adaptable such that new information or new plants 
can be added at any time. Our second goal was to use 
the assessment tool to score plants and use the 
numerical scores to inform updates to the At-Risk 
and To-Watch Lists.  

Materials and Methods 

Creation of the Assessment Tool 
The format of the assessment tool was patterned after 
the Blue Oceans Group’s Seafood Mini Guides 
(Brownstein et al. 2003). As with susceptibility of 
seafood to over-fishing, vulnerability to over-harvest 
depends on many different factors, from intrinsic life 
history traits to market forces. Based on literature, 
logic, and discussions with conservation practitioners, 
five main factors that influence a species’ vulnerability 
to overharvest were determined: life history, effect of 
harvest on individual plants, population size, habitat, 
and demand (Peters 1994, Cunningham 2001, 
Schippmann et al. 2002, Schippmann et al. 2006). 
While it was recognized that each of these factors is 
important, it was impossible to quantify their relative 
importance, so like the seafood assessment or 

Rabinowitz’s “seven forms of rarity” (1981), we 
considered each factor equally. The At-Risk Assess-
ment Tool is divided into five sections based on these 
factors. Each section begins with a broad multiple 
choice question that leads to a score of 4 to 12 points, 
which is then modified by three to five questions that 
can add or subtract up to two points each. The range 
for point values was selected for ease of use with 
integer values and whole number totals. Absolute 
magnitude of the scores is not meaningful outside of 
the context of the tool. The complete list of main and 
modifying questions and their associate point values 
can be found on The At-Risk Assessment Tool in 
Appendix A.  A species’ score is the sum of all the 
section scores and the higher the point total, the more 
vulnerable the plant species is to overexploitation.  

The questions in each section are discussed 
below. 

Life History section  
This group of questions assesses how quickly a plant 
can grow and spread. For example, a long-lived 
perennial that is destructively killed when harvested 
will score 12 points for life history, while a harvested 
annual will score only 4. Vegetative reproduction in 
the wild will decrease this score, and the necessity of a 
specialist pollinator will increase it.  

Effect of Harvest on Individuals section  
This group of questions will produce higher scores for 
plant species that are root harvested rather than those 
for which the leaves are used. Re-sprouting or quick 
recovery will decrease this score and a long harvest 
season (thus being open to year round exploitation) 
will increase it. 

Population Size section 
These questions’ scores will primarily be determined 
by the extent and density of naturally occurring 
populations. The size of the range and habitat 
specificity modify this section score. 

Habitat section 
The main question in this section is about the 
vulnerability and extent of a plant species’ habitat. 
Habitats that are both limited and specifically threat-
ened score 12 points, and habitats that are widespread 
and no more threatened than all natural areas score 4. 
Modifying questions concern fragmentation, soil type, 
and particular habitat threats (e.g. urbanization, 
logging, invasive species). 



 

 

Research Communica on 

Ethnobiology Le ers. 2014. 5: 77‐88. DOI: 10.14237/ebl.5.2014.169. 79 

Demand section 
These questions’ scores will primarily be determined 
by the annual demand for wild harvested plant 
product. Availability of widely accepted substitutes 
and ease of cultivation will decrease this score, while 
unique uses and failed past attempts at cultivation will 
increase it. 

The process of refining the list of questions was 
collaborative and iterative. At each stage, plant 
ecologists and medicinal herbs specialists were asked 
to independently score plant species. Areas where 
scores were most divergent were noted and discus-
sions ensued about whether the differences were due 
to incomplete information or different interpretations 
of the questions. After several rounds of scoring and 
discussion, a committee of 14 met in person in July 
2005 for an At-Risk Tool Review meeting at Herb-
Pharm in Williams, Oregon. At this time, we placed 
quantitative bounds on question responses (e.g. that 
high demand is more than 10 tons dry weight collect-
ed annually in the United States, or that a large range 
extends more than 300 miles) and questions for which 
the answers are unknown for most species (e.g. seed 
bank details, or how disturbance affects reproductive 
output) were removed from the tool. We recognized 
that one of the weaknesses of the tool is that good 
data on the actual number of plants, seed produced, 
and other reproductive characteristics do not exist for 
most species of medicinal plants. 

Selection of Species and Scoring 
All of the taxa on the United Plant Savers At-Risk or 
To-Watch Lists as of October 2012 were scored using 
the tool. Stinging nettle (Urtica dioica L. Urticaceae) 
and elderberry (Sambucus canadensis L. Caprifoliaceae) 
were also scored because we wanted to include species 
that likely (and did) have low scores because they were 
not at risk. Each species was independently assessed 
by at least three scorers who work in the field of plant 
ecology. Many species were also assessed by under-
graduate students in plant science classes at Glenville 
State College and Southwestern Oklahoma State 
University as part of an assignment investigating 
medicinal plants and internet information retrieval. To 
assess a species, the scorer enters the point value for 
each main and modifying question on a score sheet. 
The score sheet also contains a space for the scorer to 
enter a note about the relevant knowledge for that 
question and a space to identify the source of that 
knowledge. Where possible, scorers relied on primary 
and well-documented secondary source information 

including the North Carolina Consortium on Natural 
Medicines Grower’s Guides, the USDA Plants 
Database and the American Herbal Products Associa-
tion Tonnage Reports (Greenfield and Davis 2004, 
USDA, NRCS 2013, Dentali and Zimmerman 2012). 
Where better-documented data were unavailable, plant 
information was also gleaned from web pages of 
wildcrafters, retail herbal companies, and home 
gardeners. 

“Master Scores” for each species were assigned by 
Kindscher or Castle. To do this, all of the independ-
ent score sheets were compiled and a score for each 
question was assigned based on the consensus 
responses from the individual score sheets or from 
reconciling differences based on source data. 

Results 
Scores on the 40 species assessed ranged from a low 
(least vulnerable to overharvest) of 8 for nettles (U. 
dioca) to a high (most vulnerable to overharvest) of 75 
for sandalwood (Santalum paniculatum Hook & Arn.  
Santalaceae and related species) (Figure 1). The most 
vulnerable species scored, including sandalwood, had 
high scores in all five areas assessed. Many species on 
the original At-Risk List had high scores for several 
categories, but not for all, which highlights different 
areas of greatest concern for different species. 

Collectively, those species on the 2012 At-Risk 
List had higher average scores than those on the 2012 
To-Watch List and those on the To-Watch List had 
higher average scores than the species assessed that 
were not listed by United Plant Savers. Many individu-
al species from the 2012 To-Watch List, however, 
scored higher than some individual species originally 
deemed At-Risk, which indicates a need to examine 
the dividing line between At-Risk and To-Watch. 

Discussion 

Case studies of species on the At-Risk List 
The following case studies present a range of species 
that were scored with the ranking tool. 

1) Eastern Deciduous Perennials: American Ginseng 
(Panax quinquefolius L. Araliaceae), score of 63, and 
Mayapple (Podophyllum peltatum L. Berberidaceae), 
score of 34 
Many of the species on the original At-Risk List are 
herbaceous perennials from the understory of the 
deciduous forest in the Eastern United States and 
Canada. Most of these species are long-lived and 
harvested for root or rhizome use (Klein 2000), giving 



 

 

Research Communica on 

Ethnobiology Le ers. 2014. 5: 77‐88. DOI: 10.14237/ebl.5.2014.169. 80 

them high scores for the main questions in life history 
and effects of harvest on individuals, yet the overall 
scores can be quite different. 

Comparing mayapple (overall score of 34) with 
American ginseng (overall score of 63) highlights the 
areas in which two perennial species from the same 
woodlands can differ. Mayapples can thrive along 
roadsides, spread rhizomatously, and are easy to grow 
in a garden setting, decreasing their life history score 
compared with disturbance intolerant ginseng. While 
both species have a naturally wide range, covering 
hundreds of kilometers, ginseng patches are much less 
dense and more difficult to find than mayapple 
patches. Adverse effects of the plant-killing harvest on 
population size and structure in ginseng populations 
have been documented (McGraw 2001, Mooney and 

McGraw 2009, Rock et al. 2012), thereby increasing 
the relative vulnerability of ginseng to overharvest. 

Both mayapple and ginseng live in a habitat of 
fairly stable size that is threatened by invasive species, 
overgrazing by deer, and expanded human develop-
ment. Impacts of these habitat changes on ginseng 
populations have been studied and documented 
(McGraw and Furedi 2005, Wixted and McGraw 
2010). Mayapple populations, frequently visible from 
roadsides, appear less disturbed by these habitat 
changes; but, as with most medicinal plant species, 
effects of habitat changes on the populations have not 
been directly studied. 

Demand for ginseng remains high and, although 
cultivation is possible, wild-harvested roots continue 

 

0

10

20

30

40

50

60

70

80

U
rt
ic
a
 d
io
ca
 (
O
T
H
) 
8

S
a
m
b
u
cu
s 
ca
n
a
d
e
n
si
s 
(O
T
H
) 
1
9

M
it
ch
e
ll
a
 r
e
p
e
n
s 
(T
W
) 
2
7

P
o
d
o
p
h
y
ll
u
m
 p
e
lt
a
tu
m
 (
T
W
) 
3
4

U
lm

u
s 
ru
b
ra
 (
A
T
R
) 
3
4

S
p
ig
e
li
a
 m

a
ri
la
n
d
ic
a
 (
T
W
) 
3
5

C
a
st
e
la
 e
m
o
ry
i 
(T
W
) 
3
9

C
h
im

a
p
h
il
a
 u
m
b
e
ll
a
ta
 (
T
W
) 
4
0

A
ct
a
e
a
 r
a
ce
m
o
sa
 (
A
T
R
) 
4
0

E
u
p
h
ra
si
a
 s
p
p
. 
(A
T
R
) 
4
0

A
n
e
m
o
p
si
s 
ca
li
fo
rn
ic
a
 (
T
W
) 
4
1

A
sc
le
p
ia
s 
tu
b
e
ro
sa
 (
T
W
) 
4
1

C
o
ll
in
so
n
ia
 c
a
n
a
d
e
n
si
s 
(T
W
) 
4
1

D
io
sc
o
re
a
 v
il
lo
sa
 (
A
T
R
) 
4
1

B
a
p
ti
si
a
 t
in
ct
o
ri
a
 (
T
W
) 
4
2

B
u
rs
e
ra
 m

ic
ro
p
h
yl
la
 (
T
W
) 
4
4

E
ch
in
a
ce
a
 a
n
g
u
st
if
o
li
a
 (
A
T
R
) 
4
4

C
a
u
lo
p
h
y
ll
u
m
 t
h
a
li
ct
ro
id
e
s 
(A
T
R
) 
4
5

A
le
tr
is
 f
a
ri
n
o
sa
 (
A
T
R
) 
4
6

T
ri
ll
iu
m
 s
p
p
. 
(A
T
R
) 
4
6

A
ri
st
o
lo
ch
ia
 s
e
rp
e
n
ta
ri
a
 (
A
T
R
) 
4
7

S
a
n
g
u
in
a
ri
a
 c
a
n
a
d
e
n
si
s 
(A
T
R
) 
4
7

Li
g
u
st
ic
u
m
 p
o
rt
e
ri
 (
A
T
R
) 
4
8

E
p
ip
a
ct
is
 g
ig
a
n
te
a
 (
T
W
) 
4
9

S
a
lv
ia
 a
p
ia
n
a
 (
T
W
) 
4
9

C
h
a
m
a
e
li
ri
u
m
 l
u
te
u
m
 (
A
T
R
) 
4
9

Lo
p
h
o
p
h
o
ra
 w
il
li
a
m
si
i 
(A
T
R
) 
4
9

H
y
d
ra
st
is
 c
a
n
a
d
e
n
si
s 
(A
T
R
) 
5
0

Lo
m
a
ti
u
m
 d
is
se
ct
u
m
 (
A
T
R
) 
5
0

A
ll
iu
m
 t
ri
co
cc
u
m
 (
O
T
H
) 
5
0

D
ic
e
n
tr
a
 c
a
n
a
d
e
n
si
s 
(T
W
) 
5
1

F
ra
n
g
u
la
 p
u
rs
h
ia
n
a
 (
T
W
) 
5
1

A
d
ia
n
tu
m
 p
e
d
a
tu
m
 (
T
W
) 
5
2

A
ra
li
a
 r
a
ce
m
o
sa
 (
T
W
) 
5
3

D
ro
se
ra
 s
p
p
. 
(A
T
R
) 
5
8

D
io
n
a
e
a
 m

u
sc
ip
u
la
 (
A
T
R
) 
6
1

P
a
n
a
x 
q
u
in
q
u
e
fo
li
u
s 
(A
T
R
) 
6
3

C
y
p
ri
p
e
d
iu
m
 s
p
p
. 
(A
T
R
) 
6
6

P
ip
e
r 
m
e
th
y
st
ic
u
m
 (
T
W
) 
6
8

S
a
n
ta
lu
m
 s
p
p
. 
(A
T
R
) 
7
5

Life History Effects of Harvest Population Size Habitat Demand

Figure 1. Scores of species and genera scored using the At‐Risk Tool. Parentheses indicate previous lis ng by the United 

Plant Savers (as of December 2012) as At‐Risk (ATR), To‐Watch (TW) or unlisted (“Other”, OTR). Colored regions indicate 

magnitude of the subscores for life history, effects of harvest on individuals and popula ons, popula on size, habitat, and 

demand. 



 

 

Research Communica on 

Ethnobiology Le ers. 2014. 5: 77‐88. DOI: 10.14237/ebl.5.2014.169. 81 

to sell at a considerable premium (Hankins 1998, 
Chamberlin et al. 2013). The demand for mayapple, 
meanwhile, is low: a recent internet search found 
many more retailers selling plants for ornamental 
garden use than selling medicinal mayapple products 
in any form. While no long-lived perennial harvested 
for roots or rhizomes is invulnerable to overharvest, 
mayapple exhibits traits that make it considerably less 
vulnerable than American ginseng. 

2) Nettle (Urtica dioica), score of 8 
Stinging nettles is a well-known herb that is common 
in moist soil and waste places. Its low score suggests 
it is not at risk of over-harvest and it exemplifies 
many of the traits of species for which wild harvest is 
not currently of conservation concern. It is a perenni-
al that reproduces the first year from seed, spreads by 
rhizomes, and is almost invasive. As only the leaves 
and stems are generally collected, harvest does not 
impact the population much. A nettle patch can be 
harvested more than once per year, and certainly 
every year as it re-grows easily. The plant is naturally 
abundant with a large range and many dense popula-
tions that appear to be stable, and painful to harvest, 
across its range. Its specific habitat is not threatened 
and there are large areas of low, moist waste ground 
along creeks and streams. Finally, although the annual 
demand for nettles is moderate, the plant is high 
yielding and can easily be cultivated. Overall this plant 
is not at risk for being over-harvested. 

3) Sandalwood (Santalum spp.), score of 75 
Native Hawaiian sandalwood is vulnerable to overhar-
vest and possibly at risk of  extinction due to the fact 
that it takes more than 40 years to mature, and 
harvesting involves removing the entire tree. Further-
more the sandalwood tree is a hemi-parasite species 
that needs certain host plants in order to grow, 
making it a tricky species to reforest successfully. 
Sandalwood’s extraordinary fragrance, versatility, and 
medicinal properties have put it in high demand for 
centuries, all over the world. This is why Hawaii’s 
native sandalwood population was almost completely 
decimated during the infamous sandalwood trade that 
took place during 1815-1825. Despite this history, 
Hawaii remains the only region in the world where 
sandalwood is commercially harvested without 
regulation. Native Hawaiian sandalwood represents a 
quarter of  the diversity of  the genus Santalum. Six 
separate species are found throughout the Hawaiian 
Islands, and within these species are several unique 

varieties, all endemic to the Islands. One variety, 
Santalum freycinetianum var. lanaiense Rock, has already 
been officially recognized as endangered (USDA, 
NRCS 2013). Currently, S. paniculatum is the only 
species that is currently commercially harvested on 
the Big Island (Tummons 2010). 

Sandalwood was placed on the UpS At-Risk List 
by a board vote in 2011. After hearing concerns about 
exploitation of sandalwood in Hawaii, UpS members 
scored the plant using a draft version of the tool, and 
then came to the consensus decision to add  Sandal-
wood  to the At-Risk List. It is the only taxon that has 
been added to the list after being scored with the tool. 

 Sandalwood has a high life history score as it is a 
long-lived tree that does not reproduce easily. It 
scored high on effects of harvest on populations 
because the entire plant is harvested, many plants do 
not re-grow, and for those that do, they are slow 
growing. It scored very high in abundance and range, 
which are both very limited due to limited habitats on 
an island. The habitat of native sandalwood is both 
scarce and threatened. Finally, sandalwood scores 
high in demand because there is a large market for the 
volatile oil and no commercial cultivation exists in the 
US. Overall, sandalwood, with a score of 75, exempli-
fies all of the traits of a plant species at great risk of 
being over-harvested. 

4) Echinacea (Echinacea angustifolia DC. Asteraceae),  
score of 44 
The Echinacea genus in North America includes nine 
species with very different ranges and medicinal use 
patterns. The most popular medicinal Echinacea is E. 
angustifolia, which is primarily harvested in the wild 
(Price and Kindscher 2007; Kindscher et al. 2008). 
Also in the genus are Echinacea purpurea (L.) Moench, 
which is widely cultivated, and two species that have 
been listed as threatened or endangered (E. sanguinea 
Nutt. and E. tennesseensis (Beadle) Small). To clarify 
assessment, we are considering Echinacea species 
separately, rather than as a collective genus as 
originally listed on the At-Risk List. We have scored 
only E. angustifolia here, but caution that E. sanguinea 
and E. tennesseensis, because of their small population 
sizes and limited ranges, are species more vulnerable 
to overharvest. 

The life history score for E. angustifolia is low to 
moderate because, although it is a long-lived perenni-
al, it tolerates disturbance, produces lots of seeds, and 
most interesting, half of the harvested plants are able 



 

 

Research Communica on 

Ethnobiology Le ers. 2014. 5: 77‐88. DOI: 10.14237/ebl.5.2014.169. 82 

to re-sprout after the top 6-10 inches (15-25 cm) of 
root is harvested (Kindscher et al. 2008). The score 
for “Effects of Harvest” is high because roots are 
harvested (which can kill the plant), it takes years for 
plants to be large enough to harvest, harvest is 
conducted nearly year-round, and, when recovery 
from harvest occurs, it takes several years for the 
roots to be large enough for harvest again. The 
abundance and range score is very low because many 
scattered populations exist over a large range of Great 
Plains states and E. angustifolia can grow in many 
broad habitats, although some populations have been 
decreasing due to grazing, herbicide use, and other 
land management practices. The habitat vulnerability 
score is moderately low as the rocky prairie habitat is 
widespread and not particularly threatened. Since the 
great plowing of prairies occurred, habitats have 
remained generally available with no special threats, 
and the plants do not require any unique soil type. 
The demand score is moderately high as market 
demand is high, but yield per acre is moderate, and 
cultivated sources are known and available. Overall, 
E. angustifolia, with a score of 44, has only a moderate 
risk of being over-harvested. 

5) Peyote (Lophophora williamsii, (Lem. ex Salm-
Dyck) J.M. Coult. Cactaceae) , score of 49 
L. williamsii, or peyote, is an unusual medicinal plant 
as it is not legal for most people to harvest or possess 
it, yet it is a central part of a Native American 
sacrament. There is considerable concern about the 
over-harvest of this plant due to its limited range, the 
tightening of trade across the Mexican border, and the 
growth of the Native American Church (Terry et al. 
2011). Peyote scores high on life history as it is a long-
lived cactus that produces slowly, but it does vegeta-
tively propagate and tolerate some disturbance. It also 
scores high on the effects of harvest because either 
crowns or whole plants are collected and the harvest 
season is very long. When the crowns are harvested, 
regrowth takes at least four years (Terry, personal 
communication January 2013; Terry et al., 2011). As 
for abundance and range, it has a moderately high 
score as populations are not dense, most of the large 
range is in Mexico and inaccessible to US harvesters, 
and population declines have been documented. 
Peyote’s main habitat score is fairly low, as its habitat 
is widespread and has not changed greatly over the 
last decades. Modifying questions increase this section 
score because the plant is only found on calcareous 
soil and the habitat is threatened by invasive species 

and development.  The highly regulated market keeps 
legal demand low, but the demand section score is 
increased because the yield per acre is less than 10 
pounds, no substitutes are acceptable, and it is not 
currently commercially cultivated.  

Conclusions: Applications and Limitations 
We believe that the At-Risk Tool can be a useful 
method of summarizing a plant species’ vulnerability 
to over-exploitation and will be helpful for setting 
conservation priorities.  The tool does not provide a 
clear numerical cut-off between species that warrant 
protection and species that can be harvested without 
concern.  It does, however, provide a snapshot of 
relative vulnerability based on magnitude of total 
score, and a quick synopsis of areas of greatest 
concern based on the subscores.  

Additional Influences on Vulnerability and Missing Infor-
mation  
We recognize that other factors not included in the 
tool may influence a species’ vulnerability and these 
circumstances will require broader discussion. We also 
recognize that the information on plant populations, 
habitats, and demand, that are required to use the 
tool, are not always well documented and are rarely 
available in the same form across species. While this 
lack of consistent baseline data highlights a limitation 
of the tool, it also demonstrates an important 
secondary function of the tool: use of the tool 
requires a compilation of available information into 
one place and draws attention to areas where data is 
entirely lacking. Such gaps in knowledge can be the 
starting point for future studies, and use of the tool 
brings them to light. An important attribute of the 
tool is that it is adaptable to new information and that 
scores can be easily updated or generated as new 
information is learned or conditions change. 

Using Scores and the At-Risk Tool 
We are hopeful that, like the original At-Risk and To-
Watch Lists, the new scores will be used by a wide 
range of audiences, from land management organiza-
tions setting conservation priorities to herbalists 
recommending appropriate alternatives to the most 
vulnerable herbs. The availability of subscores for 
each area should aid in planning and decision-making. 
For example, even though they both score 49, 
different strategies should be employed to protect 
white sage (Salvia apiana Jeps. Lamiaceae), which has a 
highly threatened habitat in the southern California 
desert and moderate demand for leaves, than to 



 

 

Research Communica on 

Ethnobiology Le ers. 2014. 5: 77‐88. DOI: 10.14237/ebl.5.2014.169. 83 

protect the stream orchid (Epipactis gigantea Douglas 
ex. Hook. Orchidaceae), which has low demand but 
an inherently more vulnerable life history, being a 
water-loving perennial orchid. 

The tool can be used to quickly model how 
potential changes might affect a species’ vulnerabil-
ity. If a plant were determined next year to be the 
next Viagra, only better, then the At-Risk Tool 
could be used to determine to what degree we 
should be concerned about its potential overharvest. 
It can also be used to project effects of changes in 
range and habitat due to climate change, invasive 
species or urban expansion. 

Opportunities for Expansion 
One current limitation of use of the tool is small 
sample size of plants scored. Scores would be of 
greater value if they could be compared to scores 
from a broader sample of medicinal plants. We 
heartily encourage others interested and knowledge-
able about medicinal plants in North America to 
assess plants using the At-Risk Tool and submit the 
results and comments to the United Plant Savers. 

While designed for wild-harvested medicinal 
plants in temperate North America, the questions 
on the At-Risk tool can apply to many other groups 
of plants and regions with only slight modifications 
to questions.  The tool can also serve as a dynamic 
teaching instrument for students and the general 
public who are concerned about wild plant popula-
tions. As use of the At-Risk Tool expands to other 
species and other regions, we anticipate productive 
discussions for adjustments to provide scores that 
will best inform practitioners working with the goal 
of sustainable plant populations. 

Acknowledgements 
There are many people who have helped with 
developing this tool. The At-Risk List was devel-
oped by the United Plant Savers, and their staff and 
board over the years have been instrumental in 
maintaining and promoting the list. The current 
board also requested that we work on this tool and 
provided some funds for its development. Among 
those on the board and staff, advisory board, and 
others who have helped, are: Beth Baugh, Tim 
Blakely, David Bunting, Richo Cech, Jim Chamber-
lain, Trish Flaster, Edward Fletcher, Steven Foster, 
Cascade Anderson Geller, Rosemary Gladstar, 
Mindy Green, Christopher Hobbs, Gary Kauffman, 
Sara Katz, Lynda LeMole, Michael McGuffin, and 

Mark Wheeler. Hillary Loring, Quinn Long, Jennifer 
Moody, and students from Glenville State College 
2007 and 2009 Botany classes and Southwestern 
Oklahoma State University 2012 Plant Taxonomy all 
scored individual plant species. Zella Classen assisted 
with data management and fact checking.  

Declarations 
Permissions: None declared.  

Sources of funding: The United Plant Savers provided 
funds to Kindscher, Castle, and their students to 
meet, print posters, and complete the project.  

Conflicts of interest: Susan Leopold is employed by the 
United Plant Savers. 

References Cited 
Brownstein, C., M. Lee, and C. Safina. 2003. Harness-

ing Consumer Power for Ocean Conservation. 
Conservation Magazine 4:39-42. 

Cech, R. A. 1998. Balancing Conservation with 
Utilization: Restoring Populations of Commercially 
Valuable Medicinal Herbs in Forests and Agrofor-
ests. Paper presented at the North American 
Conference on Enterprise Development through 
Agroforestry. Minneapolis, MN, October 4-7. 

Chamberlain, J. L., S. Prisley, and M. McGuffin. 2013. 
Understanding the relationships between American 
ginseng harvest and hardwood forests inventory and 
timber harvest to improve co-management of the 
forests of eastern United States. Journal of Sustainable 
Forestry (in press). 

Cunningham A. B. 2001. Applied Ethnobotany: People, 
Wild Plant Use, and Conservation. Earthscan Publica-
tions, London, UK. 

Dentali, S. and M. Zimmermann. 2012. Tonnage 
Surveys of Select North American Wild-Harvested 
Plants, 2006-2010. American Herbal Products 
Association. Silver Springs, MD. 

Gladstar, R. 2000. Introduction. Pp. 1-12 in Planting 
the Future, edited by R. Gladstar and P. Hirsch. 
Healing Arts Press, Rochester, VT. 

Gladstar, R. and P. Hirsch. 2000. Planting the Future. 
Healing Arts Press, Rochester, VT. 

Greenfield, J. and J. Davis. 2004. Medicinal Herb 
Production Guides. North Carolina Consortium on 
Natural Medicines and Public Health. North 
Carolina State University, Raleigh, NC. 



 

 

Research Communica on 

Ethnobiology Le ers. 2014. 5: 77‐88. DOI: 10.14237/ebl.5.2014.169. 84 

Hankins, A. 1998. Producing and marketing wild 
simulated ginseng in forest and agroforestry 
systems. Paper presented at the North American 
Conference on Enterprise Development through 
Agroforestry. Minneapolis, MN, October 4-7. 

Kindscher, K., D. M. Price, and L. Castle. 2008. Re-
sprouting of Echinacea angustifolia Augments 
Sustainability of Wild Medicinal Plant Populations. 
Economic Botany 62(2):139-147.  

Klein, R. 2000. Wise Old Plants. Pp. 24-38 in: Planting 
the Future, edited by R. Gladstar and P. Hirsch. 
Healing Arts Press, Rochester, VT. 

McCoy, J. A., J. M. Davis, N. D. Camper, I. Khan, 
and A. Bharathi. 2007. Influence of Rhizome 
Propagule Size on Yields and Triterpene Glycoside 
Concentrations of Black Cohosh [Actaea racemosa 
(L.) Syn Cimicifuga racemosa (L.) nuttal]. HortScience 42
(1): 61-64. 

McGraw, J. B. 2001. Evidence for Decline in Stature 
of American Ginseng Plants from Herbarium 
Specimens. Biological Conservation 98:25-32. 

McGraw, J. B. and M. A. Furedi. 2005. Deer Brows-
ing and Population Viability of a Forest Understory 
Plant. Science 307(5711):920-922. 

Mooney, E. H. and J. B. McGraw. 2009. Relationship 
Between Age, Size and Reproduction in Popula-
tions of American Ginseng, Panax quinquefolius 
(Araliaceae), Across a Range of Harvest Pressures. 
Ecoscience 16 (1): 84-94. 

Peters, C. M. 1994. Sustainable Harvest of Non-timber 
Plant Resources in Tropical Moist Forest: An Ecological 
Primer. The New York Botanical Garden, Bronx, 
NY. 

Price, D. H. and K. Kindscher. 2007. One Hundred 
Years of Echinacea angustifolia Harvest in the 
Smoky Hills of Kansas, USA. Economic Botany 61:86-
95.  

Rabinowitz, D. 1981. Seven forms of rarity.  Pp 205-
217 in: The biological aspects of rare plant conservation. 
Edited by J. Synge. Wiley, New York. 

Roberson, E. 2008. Medicinal Plants at Risk. A Native 
Plant Conservation Campaign Report. Center for 
Biological Diversity. Tucson, AZ. 

Rock, J., G. Kauffman, and N. Murdock. 2012. 
Harvesting of Medicinal Plants in the Southern 
Appalachian Mountains. Journal of Medicinal Plant 

Conservation. Winter 2012: 12-13.Schippmann U., D. 
Leaman, and A. B. Cunningham. 2006. A Compari-
son of Cultivation and Wild Collection of Medicinal 
and Aromatic Plants Under Sustainability Aspects. 
Pp. 75-95 in Medicinal and Aromatic Plants, edited by 
R.J, Bogers, L.E. Craker, and D. Lange. Wageningen 
UR Frontis Series, Vol. 17. Springer, Dordrecht, the 
Netherlands. 

Schippmann U., D. J. Leaman, and A. B. Cunning-
ham. 2002. Impact of Cultivation and Gathering of 
Medicinal Plants on Biodiversity: Global Trends and 
Issues. In Biodiversity and the Ecosystem Approach in 
Agriculture, Forestry and Fisheries. Inter-Departmental 
Working Group on Biological Diversity for Food 
and Agriculture, FAO, Rome.  

Terry, M., K. Trout, B. Williams, T. Herrera, and N. 
Fowler. 2011. Limitations to natural production of 
Lophophora williamsii (Cactaceae) I. Regrowth and 
survivorship two years post harvest in a South 
Texas population. Journal of the Botantical Research 
Institute of Texas 5:661−675. 

Tummons, P. 2010. Dispute Over Hokukano Sandal-
wood Logging Ends Up Before Federal Bankruptcy 
Judge. Environment Hawaii 21:1. 

Westfall, R. E. and B. W. Glickman. 2004. Conserva-
tion of Indigenous Medicinal Plants in Canada. In 
Proceedings of the Species at Risk 2004 Pathways to 
Recovery Conference, edited by T. D. Hooper. Species 
at Risk 2004 Pathways to Recovery Organizing 
Committee, Victoria, B.C. 

UpS. 2013. The United Plant Savers Homepage 
(http://www.unitedplantsavers.org/, accessed 9 
Sept. 2013). East Barre, VT, USA. 

USDA, NRCS. 2013. The PLANTS Database 
(http://plants.usda.gov, accessed 1 July 
2013).National Plant Data Team, Greensboro, NC 
27401-4901 USA. 

Wixted, K. L. and J. B. McGraw. 2010. Competitive 
and Allelopathic Effects of Garlic Mustard (Alliaria 
petiolata) on American Ginseng (Panax quinquefolius). 
Plant Ecology 208:347-357. 

Biosketch 

Lisa Marie Castle  inves gates plant‐human interac ons 
and  teaches  as  an  assistant  professor  of  biology  at 
Southwestern Oklahoma State University.  



 

 

Research Communica on 

Ethnobiology Le ers. 2014. 5: 77‐88. DOI: 10.14237/ebl.5.2014.169. 85 

Susan  Leopold  is  Execu ve  Director  of  United  Plant 
Savers  and  serves  as  board  member  of  Botanical 
Dimensions and the Center for Sustainable Economy.  

Rachel  Cra   is  a  PhD  Candidate  at  the  University  of 
Kansas,  Department  of  Sociology,  and  a  Graduate 
Research Assistant at the Kansas Biological Survey. 

Kelly  Kindscher  is  a  plant  ecologist  at  the  Kansas 
Biological Survey and a professor  in the Environmental 
Studies Program at the University of Kansas.  



 

 

Research Communica on 

Ethnobiology Le ers. 2014. 5: 77‐88. DOI: 10.14237/ebl.5.2014.169. 86 

Appendix A: United Plant Savers At‐Risk Assessment Tool 
 
I. Life History: How vulnerable are plants based on their life history? 
  
1. Life span 
  +4  Annual or Biennial  (1‐2 years)       
  +4  Perennial Plant that is not destruc vely harvested 
  +8  Short Lived Perennial  (2‐5 years)       
+12  Long Lived Perennial  (> 5 years)       
 
1.1 Age at first reproduc on 
 ‐2   1 or less 
  0  2 to 4               
+2   5 or more             
 
1.2 Ability to withstand disturbance (e.g. ability to grow a er vegeta on and soil have been mowed, plowed, grazed or oth‐

erwise disturbed)  
 ‐2  Thrives on disturbance (weedy or early succession species)   
  0  Tolerates some disturbance or some types of disturbances     
+2   Intolerant (very conserva ve species) 
 
1.3 Ability to reproduce vegeta vely under normal condi ons 
 ‐2  Reproduces vegeta vely regularly in the wild and from small parts of plant (includes suckers, runners, bulblets and 

tubers) 
  0  Occasionally reproduces vegeta vely in the wild 
+2  Rarely to never seen to reproduce vegeta vely in the wild     
 
1.4 Ability to reproduce from seed under normal condi ons 
 ‐2  Seedlings regularly seen or easy to cul vate from seed 
  0  Seedlings occasionally seen 
+2  Seedlings rarely to never seen 
 
1.5 Interac ons with other organisms required for growth and reproduc on (e.g. known obligatory mychorrhizal associa‐

ons, pollinator specificity, parasi sm) 
 ‐2  No special associa ons needed to grow it in places outside of natural range 
  0  Not known                 
+2  Known limi ng associa ons  
 
II. Effects of Harvest on Individuals and Popula ons: How does harvest affect plants? 
 
2. Part of plant most commonly harvested 
  +4   Harvest is of leaves and twigs only. 
  +8  Harvest is of seeds, fruits, flowers, stem bark or off‐shoots. 
+12   Harvest is of roots, root bark or en re plant. 
 
2.1 Post‐Harvest Recovery of Individual Plants 
‐2  Plants are able to reproduce normally the season following harvest. 
‐1  Harvest limits the next season’s growth 
 0  At least some plants in a harvested popula on can re‐grow a er harvest, but re‐growth takes several growing sea‐

sons   
+2  Individual plants cannot re‐grow a er harvest 



 

 

Research Communica on 

Ethnobiology Le ers. 2014. 5: 77‐88. DOI: 10.14237/ebl.5.2014.169. 87 

2.2 Harvest Interval 
 ‐2  A plant can be harvested mul ple  mes in one year 
  0  1 to 2 years 
+2   3 years or more  
 
2.3 Length of Harvest Season 
 ‐2  Harvestable for one month or less 
  0  Harvestable for 1 to 3 months 
+2  Harvestable for more than 3 months per year 
 
III. Popula on Size: How many plants are there? 
 
3. Is the plant naturally abundant? 
  +4  Many dense popula ons exist.  (There are many popula ons in which someone could harvest all day in a very local 

area.) 
  +6  A few dense popula ons exist and many more sca ered popula ons exist.  (There are a few popula ons in which 

someone could harvest all day without moving and many in which one could harvest all day by moving across some 
local acreage.) 

  +8  Many sca ered popula ons exist.  (There are many places in which someone could harvest all day by driving to 
several local patches.) 

+10  Few sca ered popula ons exist and many more sparse popula ons.  (There are a few places in which a harvester 
could harvest all day moving around a bit, but most places the harvester would need to drive distances to harvest 
all day.) 

+12  Popula ons are few and sparse. 
 
3.1 Range 
 ‐2  Large (plant found across an area greater than 300 miles) 
  0  Medium (plant found across an area 100 to 300 miles)   
+2  Small (plant found across an area less than 100 miles) 
 
3.2 Change in overall popula on size in primary harvest range 
 ‐2  Popula on known to be increasing 
  0  Popula on stable or status unknown 
+2  Declines in popula on size documented 
 
3.3 Degree of habitat specializa on 
 ‐2  Can grow in roadsides, vacant lots or other disturbed areas 
  0  Can grow in broad habitat categories (e.g. “eastern deciduous forest” or “great plains grassland”) 
+2  Can only grow in a very limited habitat (e.g. “moist acidic glades in eastern deciduous forest” or “limestone out‐

crops in tall‐grass prairie”)  
 
IV. Habitat: How vulnerable is the habitat? 
 
4. How vulnerable is the plant’s physical habitat? 
  +4  Habitat is widespread and no more threatened than all land areas. 
  +8  Habitat is limited OR specifically threatened 
+12  Habitat is limited AND specifically threatened 
 
4.1 Habitat Acreage Change 
 ‐2  Habitat acreage is expanding (e.g. forest edge, roadsides, “suburban savannas”) 
  0  Habitat acreage unchanged or not drama cally reduced 
+2  Habitat acreage has been reduced by 50% or more over last 100 years. 



 

 

Research Communica on 

Ethnobiology Le ers. 2014. 5: 77‐88. DOI: 10.14237/ebl.5.2014.169. 88 

4.2 Habitat Fragmenta on 
 ‐2  Large tracts of con nuous acres of habitat exist 
  0  Habitat areas intermediate or unknown 
+2  Only very small habitat patches exist 
 
4.3 Confined to a limited or very vulnerable soil type 
  0  No             
+2  Yes (includes hydric or salty soils)   
 
4.4 Habitat Threats 
Add 1 point for each of these habitat threats (to a maximum of 2 points) 

logging expansion  rapid development  mining    over‐grazing  take‐over from invasive species 
use land for recrea on growing rapidly     widespread regular herbicide use   vulnerability to disease 
acid deposi on     fire suppression 
  
V. How much is needed?: What is the demand?  Are there alterna ves? Can the plant be cul vated? 
 
5  Annual Demand for Wild Harvested Plant 
  +4  Less than 1 ton dry weight 
  +8  1 to 10 tons dry weight 
+12  More than 10 tons dry weight 
 
5.1 Yield per Acre 
 ‐2  Ten pounds or more 
  0  One to ten pounds 
+2  Less than one pound 
 
5.2 Availability of good subs tute to wild‐harvested plant 
 ‐2  Subs tute known and widely accepted 
  0  Subs tute known but not widely accepted as such     
+2  No known subs tutes 
 
5.3 Cul va on and poten al for cul va on 
 ‐2  Currently cul vated and commercially viable     
 ‐1  Not commercially cul vated but cul va on on a commercial scale hor culturally achievable (plant material availa‐

ble, no special associa ons required)     
  0  Growth on a commercial scale not easily achievable or economically viable (plant material not available or special 

associa ons required) 
+2  Growth on a commercial scale probably not achievable (plant material not available AND special associa ons re‐

quired)