Content uploaded by Nelda Matheny
Author content
All content in this area was uploaded by Nelda Matheny on Nov 04, 2014
Content may be subject to copyright.
Content uploaded by Nelda Matheny
Author content
All content in this area was uploaded by Nelda Matheny on Oct 17, 2014
Content may be subject to copyright.
112
Clark and Matheny: Urban Forest Sustainability Model
A MODEL OF URBAN FOREST SUSTAINABILITY:
APPLICATION TO CITIES IN THE UNITED STATES
by James R. Clark and Nelda P. Matheny
Abstract. The applicability of a model for urban forest
sustainability was evaluated through a written survey. The
model considers the character of the vegetation resource,
community awareness of, and attitudes about, urban forests,
and the management programs for the resource. Results
from 25 U.S. cities were evaluated using the 20 criteria and
four levels of performance found in the model. Although
surveys were sent directly to mayors, urban forestry
professionals completed the questionnaires. The average
score for responding cities was
48.8,
out of a maximum score
of 80. The range in scores was 27 to 61. Criteria of
sustainability with highest overall scores dealt with
awareness of trees as a community resource and
neighborhood action; the lowest scores involved the
participation of private landholders in urban forest
management and regional cooperation.
Creation and management of urban forests to
achieve sustainability is the long-term goal of ur-
ban foresters as well as elected officials,
busi-
ness leaders, and citizens. Clark et al. (1997)
previously described a model for assessing a
community's progress towards this
goal.
In their
model,
a sustainable urban forest was defined
as "the naturally occurring and planted trees in
cities which are managed to provide the inhabit-
ants with a continuing level of economic, social,
environmental and ecological benefits today and
into the future." The model considers three broad
categories:
• Vegetation Resource: knowledge of the
existing tree resource
• Community Framework: interaction and
cooperation of constituent groups
• Resource Management: current manage-
ment programs
To assess progress towards a sustainable ur-
ban forest, communities must have criteria to
measure the current state of these three compo-
nents.
Historically, any such criteria have been
qualitative in nature and lacked universal appli-
cability. In the 1996 State of the Urban Forest
Report (Sacramento Urban Forest Task Force
1996) for the Sacramento region, sustainability
was assessed qualitatively, through 16 region-
ally specific criteria. In an approach focused on
municipal programs, Thompson
etal.
(1994)
iden-
tified four criteria of sustainable urban forests with
a variety of qualitative assessment measures. In
contrast to these examples, the model developed
by Clark et
al.
provided a method of assessment
that was region-independent and quantitatively
based (Table 1). This paper presents an evalua-
tion of the applicability of this model based upon
the results of a survey of cities across the United
States.
Methods and Materials
A written survey was developed that adapted the
criteria of sustainability and performance indica-
tors identified by Clark et al. (1997) into a series
of open- and close-ended questions. In the open-
ended questions, respondents were asked to pro-
vide opinions, examples, and experiences. In the
close-ended questions, the possible responses
corresponded directly to the four levels of perfor-
mance suggested for each criteria of
sustainability. For example, one of the criteria in
the Community Framework component is the in-
teraction among citizen, government, and
busi-
ness groups. The performance indicators of this
criterion are as follows:
• Optimal: formal interaction (such as tree
board with staff coordination)
• High: informal and/or general cooperation
• Moderate: no interaction among constitu-
encies
• Low: conflicting goals among constituen-
cies
In the survey, the question corresponding to
this criterion was, "What is the level of interac-
tion among government, citizens, institutions, cor-
porations, and other constituent groups?"
Journal of Arboriculture 24(2): March 1998
113
Respondents could choose from among the fol-
lowing answers:
• Excellent, we have regular interaction
(formal working group, city tree board,
mayor's advisory committee, etc.).
• We have a tree board with communitywide
representation, but it is not supported by
city staff.
• Good, we have many informal contacts
throughout the city.
• Fair, we have good interaction with some,
conflicts with others.
• There is little interaction.
• Various groups are often in conflict.
Respondents were also asked, "Are there
particular issues or topics that foster interaction
Table
1.
Criteria for urban forest sustainability. Each criterion includes four levels of performance towards
attaining the key objective (Clark et al. 1997).
Component
Criteria Key objective
Vegetation
Resource
Community
Framework
Resource
Management
Canopy cover
Age distribution of trees in community
Species mix
Native vegetation
Public agency cooperation
Involvement of large private and
institutional landholders
Green industry cooperation
Neighborhood action
Citizen-government-business
interaction
General awareness of trees as
community resource
Regional cooperation
Citywide management
Citywide funding
City staffing
Assessment tools
Protection of existing trees
Species and site selection
Standards for tree care
Citizen safety
Recycling
Achieve climate-appropriate degree of tree
canopy communitywide.
Provide for uneven age distribution.
Provide for species diversity.
Preserve and manage regional biodiversity.
Maintain the biological integrity of native
remnant forests. Maintain wildlife corridors to
and from the city.
Ensure all city departments operate with
common goals and objectives.
Large private landholders embrace citywide
goals and objectives through specific resource
management plans.
The green industry operates with high
professional standards and commits to
citywide goals and objectives.
At the neighborhood level, citizens understand
and participate in urban forest management.
All constituencies in the community interact for
the benefit of the urban forest.
The general public understands the value of
trees to the community.
Provide for cooperation and interaction among
neighboring communities and regional groups.
Develop and implement a management plan
for trees and forests on public and private
property.
Develop and maintain adequate funding to
implement a citywide management
plan.
Employ and train adequate staff to implement
citywide management
plan.
Develop methods to collect information about
the urban forest on a routine basis.
Conserve existing resources, planted and
natural,
to ensure maximum function.
Provide guidelines and specifications for
species use, on a context-defined basis.
Adopt and adhere to professional standards
for tree care.
Maximize public safety with respect to trees.
Create a closed system for tree waste.
114
Clark and Matheny: Urban Forest Sustainability Model
among groups? that inhibit interaction among
groups?"
Forty-five cities
in
the United States were
iden-
tified as potential participants in the survey. Par-
ticipant cities were selected based upon the
following criteria:
• population between 30,000 and 750,000
(Chicago, Illinois, was an exception)
• geographic diversity, including natural
forest diversity
• membership in U.S. Conference of Mayors
• active involvement with the programs of
the Trust for Public Land
• existing municipal urban forestry program
or nonprofit group
Surveys were sent directly to the mayor of each
city, with a cover letter from the Conference of
Mayors. Communities that did not return surveys
within the designated completion period were con-
tacted and encouraged to complete the survey.
Survey responses were converted to numeri-
cal scores based upon specific indicators of per-
formance
(i.e.,
no or negative response = 0, low
= 1, fair = 2, high = 3, optimal = 4). Half-units
(e.g.,
2.5) were used in some cases. For ex-
ample,
in the question dealing with interaction
among constituent groups scores were assigned
as follows:
Response Score
Excellent, we have regular interaction... 4
We have a tree board with community- 3.5
wide representation...
Good ... 3
Fair...
2
There is little interaction 1
Groups are in conflict 0
Numerical scores were equally weighted and
summarized for each of the 20 criteria in the
sustainability
model,
yielding a maximum possible
score of
80.
Results were summarized in spread-
sheet format then imported into Statview, a sta-
tistical analysis program.
Results
Twenty-five cities returned completed surveys, a
response rate of 56% (Table 2). The geographic
distribution of responding cities ranged across the
Table 2. Population and size of responding cities.
City
Livermore, CA
Modesto, CA
San Francisco, CA
Colorado Springs, CO
Denver, CO
Atlanta, GA
Chicago, IL
Fort Wayne, IN
Topeka, KS
Louisville, KY
Baton Rouge, LA
Lowell,
MA
Baltimore, MD
Cincinnati,
OH
Cleveland,
OH
Providence, Rl
Charleston, SC
Rock
Hill,
SC
Sioux Falls, SD
Ft. Worth, TX
Mesquite, TX
Salt Lake City, UT
Seattle, WA
Milwaukee, Wl
Cheyenne, WY
Population
(1990)
56,741
180,000
700,000
281,140
467,610
415,200
3,719,000
180,000
119,883
269,063
380,105
104,000
750,000
365,000
505,616
160,000
80,414
41,643
100,814
447,619
101,484
160,000
516,259
628,088
50,008
Area
(ha)
5,076
8,761
12,691
35,224
40,049
34,188
59,173
—
14,426
15,540
122,248
—
22,533
19,425
19,943
—
11,370
6,195
11,655
76,455
10,878
27,972
23,828
24,812
4,403
continental United States. City population ranged
from 41,643 (Rock
Hill,
South Carolina) to
3,719,000
(Chicago, Illinois). The average popu-
lation was 431,187 (median population was
269,063). The size of respondent cities ranged
from
4,403
to 122,248 ha (17.0 to 472 mi
2
) with
an average area of 27,454 hectares (106 mi
2
).
Although the surveys were addressed to the
individual mayors in each city, for the most part
they were completed by professional staff mem-
bers within city government. Job titles included
urban forester, city forester, city arborist, park
superintendent, landscape supervisor, natural
resource manager, and director of landscape and
forestry. In two cases (Chicago, Illinois, and
Lowell,
Massachusetts), surveys were completed
by program directors (Chicago: Greenstreets;
Lowell:
Parks and Conservation Trust). For some
cities,
more than one person took part in com-
pleting the survey.
The average score for the respondent cities
was 48.8, with a range from 27 to
61
(maximum
possible score of 80) (Figure
1;
Table 3 and Table
Journal of Arboriculture 24(2): March 1998
115
2
n
c
1
CO
•6
E3
Resouce management
H Community framework
E3
Vegetation resource
Responding City
Figure 1. Compiled urban forest sustainability model scores of responding cities.
4).
A wide range in scores existed among the
three components of the sustainability model. For
the vegetation resource component, the average
value was 8, with a range of 4 to 13.5 (maximum
possible score: 16). For the community frame-
work component, the average score was 17.9,
with a range of 12 to 25 (maximum score: 28).
The range in scores for the resource manage-
ment component was 6 to 29, with an average of
22.4 (maximum score: 36). No significant rela-
tionship was found between overall score and city
population or
size.
Both high and low scores were
obtained by large and small cities.
The average numerical score for all 20 criteria
in the sustainability model for all respondent cities
was 2.43 (SE
mean
0.04). The range in scores was
1.78 to 3.06 (maximum score: 4) (Figure 2; Table
5).
The two lowest-scoring criteria were both in
the Community Framework component. One dealt
with the involvement of holders of large private
and institutional lands in management of the ur-
ban forest, with a score of 1.78. Only 13 cities had
information about this constituent group. The sec-
ond low-scoring question involved regional coop-
eration among communities, with a score of 1.98.
The highest-scoring criteria were also in the
Community Framework component. The criterion
dealing with community awareness of trees and
urban forests had an average score of
3.06.
Other
high-scoring criteria were neighborhood action in
urban forest planning and management (2.96),
cooperation among green industry groups (2.88),
capacity of city staff to implement a citywide man-
agement plan (2.88), and the use of tree care
standards as a management tool (2.86).
Responses to the open-ended questions re-
flected the views of municipal urban forestry/
arboriculture staff on a variety of topics. For ex-
ample,
several questions considered the role of
constituent groups in urban forest management
(Table 6 and Table 7). Staff completing the sur-
veys viewed city government and community
groups as the most important constituent groups
(Table 6). In contrast, business districts, corpo-
rations, and private landholders were seen as less
important and not critical. In two cases, respon-
116
Clark and Matheny: Urban Forest Sustainability Model
dents viewed city government and the general
public as a liability to urban forest management.
Respondents to the survey also commented
on the importance that various city departments
place on the urban forest (Table 7). The most
positive comments were made about forestry and
parks units. In contrast, 20% to 25% of the re-
spondents viewed the public utility and public
works units in a negative way
(e.g.,
"These de-
partments view trees and urban forests as a li-
ability").
Community groups were considered either
important or vital to the management of public
trees (Table 8). None of the responding cities re-
garded community groups in a negative manner.
Contributions made by such groups included gen-
eral awareness of the need for management, a
source of direct labor, and financial support. Is-
sues to which community groups responded in a
positive manner included tree planting and the
environmental benefits provided by
trees.
Issues
evoking negative attitudes included uplifted side-
walks,
litter, and interference with signs and lights.
Respondents provided detailed information
about their city's urban forest resource. Sixty per-
cent of respondents knew the number of trees on
streets; 20% knew the number of park
trees.
How-
ever, only one respondent knew the number of
trees in the entire city (public and private land).
Sixty-four percent of respondents knew the extent
of canopy cover across their community and 40%
had specific target goals for the amount of cover.
For cities with urban forest management plans
(64%
of respondents), such plans were largely
restricted to public trees. Only one city (Sioux
Falls,
South Dakota) had an urban forest man-
agement plan that considered all trees in the com-
munity, both public and private.
Existing urban forest management included
use of a number of tools for assessing tree spe-
cies,
age, condition, and location (Table 9). Re-
source information was incorporated into
geographic information systems (GIS) in 40% of
surveyed communities. Other assessment tools
cited by respondents included inventories, aerial
photographs, and windshield surveys. Use of
comprehensive assessment tools was clearly
reflected in the scores for the Vegetation Re-
Table 3. Summary of survey results for the model
of urban forest sustainability.
Component
Vegetation
Resource
Community
Framework
Resource
Maximum Range
score
Management
Overall
Table 4. Urban
for responding
City
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Veg.
Res.
6
6.5
11
5.5
7
7.5
8.5
5.5
7
7.5
10
8
13.5
8
4
9.5
9
10.5
11
10
6.5
10
10
10
11
16
28
36
80
low
4
12
6
27
high
13.5
25
29
61
forest sustainability model
cities.
Score
Comm.
Frame.
15
19
14
16
14
18
15.5
16.5
15.5
19.5
16
16.5
12
18.5
16
20
15.5
17
17.5
19
23
23.5
20
25
24.5
Res.
Mgt.
6
16
17
21
22
18
19.5
23
23.5
19.5
21
23
22.5
22.5
29.5
22
28
25.5
25.5
25
26
23.5
28.5
25.5
25.5
Average
8.5
17.9
22.4
48.8
scores
Total
27
41.5
42
42.5
43
43.5
43.5
45
46
46.5
47
47.5
48
49
49.5
51.5
52.5
53
54
54
55.5
57
58.5
60.5
61
source category. Communities that were aware
of the species mix had knowledge of the age dis-
tribution as
well.
Discussion
The survey was designed to evaluate the utility
of the model developed by Clark et al. (1997) in
assessing the sustainability of urban forests. To
that
end,
the survey asked both closed- and open
ended questions about each criterion and per-
Journal of Arboriculture 24(2): March 1998
117
Figure 2. Compiled scores to urban forest sustainability model survey questions.
formance indicator contained in the model. By as-
signing a numerical rating to each performance
indicator, researchers could compute for each city
a summary score associated with progress to-
wards a sustainable urban forest. In this way, a
quantitative assessment of sustainability could be
developed.
Under the model developed by Clark et al., a
score of 80 represents a sustainable urban for-
est. Among the responding cities, the average
score was 48.8, with a range from 27 to
61
(Table
4 and Figure 1). The model divides 20 criteria
into three categories. The average score (as a
percentage of the maximum) for each of these
components was Vegetation Resource, 53%;
Community Framework, 64%; and Resource
Management, 62%
Results for each city among the three catego-
ries were not uniform. There was wide variation
among components—communities with high
scores in one of the components did not neces-
sarily have high scores
in
the others. For example,
the community with the lowest score (4 of a pos-
sible 16) in the Vegetation Resource component
had the highest score (29 of 36) in the Resource
Management component.
Scores obtained from this survey reflect two
of the important concepts of the sustainability
model:
1) sustainable urban forests require hu-
man intervention and 2) trees growing on pri-
vate lands comprise the majority of urban forests
(Clark et al. 1997). For the first concept, the
model includes the active interaction and par-
ticipation of government, businesses, citizens,
and other constituent groups. Because urban
forests are a mix of naturally occurring and
planted trees in specific locations, their creation
and preservation require active planning and
management by a diverse group of owners,
managers, and stakeholders.
118
Clark and Matheny: Urban Forest Sustainability Model
Integrating management of private forests and
lands into a citywide plan is a significant
chal-
lenge to communities. For many of the model's
performance indicators, the difference between
a score of 2 and 3 is based upon consideration
of trees on private lands. Unfortunately, few cit-
ies are aware of management efforts on private
lands,
let alone actively interact with those man-
agers.
A broadening of focus to consider all trees
in the city must occur in order to achieve sub-
stantial progress towards sustainability. More
actively involving private landholders in coopera-
tive management programs and goals requires
new thinking and approaches other than those
historically employed with individual citizens. This
does not mean that public agencies should over-
see,
direct, or otherwise regulate management
of trees on private lands. Rather, there is a need
for shared vision and agreement on goals, ob-
jectives, and management approaches.
Another aspect of broadening the focus is re-
gional cooperation in urban forest management.
Most of the respondents noted their involvement
Table 5. Scores for urban forest sustainability
model criteria.
Model components
and criteria
Mean
Standard
error
Vegetation Resource
Canopy cover
Age/size distribution
Species distribution
Native vegetation
Community Framework
Public agency cooperation
Private/institutional landholders
Green industry cooperation
Neighborhood action
Citizen-government-business
interaction
Awareness of trees as a
community resource
Regional cooperation
Resource Management
Citywide tree management plan
Funding
Staffing
Assessment tools
Protection existing trees
Species and site selection
Standards for tree care
Citizen safety
Recycling
2.04
2.18
2.16
2.14
2.42
1.78
2.88
2.96
2.78
3.06
1.98
2.18
2.28
2.88
2.46
2.24
2.20
2.86
2.68
2.44
0.18
0.16
0.14
0.19
0.12
0.20
0.13
0.14
0.16
0.18
0.14
0.15
0.19
0.16
0.15
0.19
0.16
0.16
0.21
0.19
in such professional organizations as the Inter-
national Society of Arboriculture as examples of
such cooperation. The model, however, consid-
ers regional cooperation to be the positive inter-
action of neighboring communities in urban forest
management. This could take the form of consis-
tency in tree preservation and resource conser-
vation policies, pest management programs, or
storm water runoff. The recently completed Chi-
cago Urban Forest Ecosystem Project
(McPherson et al. 1994) and Sacramento Urban
Forest Ecosystem Study are examples of the
potential to develop and implement regionally
based resource assessment and management
programs. Another example of the potential for
regional cooperation involves the integration of
riparian forests into urban storm drainage sys-
tems (Ellis 1995).
When we evaluated the survey results, sev-
eral limitations of the sustainability model and
survey approach became clear. First and fore-
most, the sustainability score represents the
views of the respondent
(i.e.,
the person com-
pleting the survey) resulting in a strong bias on
at least two levels. The first bias of the respon-
dent involves his or her perceptions and attitudes
about the sustainability model. While we might
hope that the municipal arborist or urban forester
who completed these surveys represented the
broader community views about a topic, there is
no way to ensure this fact. The second respon-
dent bias involves the cities and their elected of-
ficials.
Surveys were sent to the mayor of each
community, who then transferred it to the respon-
dent. That this transfer occurred represents a
positive interest in the urban forest.
Table 6. Constituent groups important to urban for-
est sustainability.
Importance of constituent groups to urban forest
Absolutely Nice but No. of re-
Group
vital Important not critical sponses
City government 18 6 1 25
Community groups 12 9 3 24
Public 9 12 1 22
Green industry 5 14 3 22
Business districts 3 7 12 22
Corporations 1 10 11 22
Private landholders 2 8 10 20
Journal of Arboriculture 24(2): March 1998
119
Another limitation of the survey involves pri-
vately owned
trees.
The sustainability model con-
siders all trees within a community. The more
information a city possesses about all trees in its
communities, the higher its score. In some cases,
the survey was completed by a team of staff
members from such departments as parks,
streets, and forestry. Scores from these commu-
nities are likely to be higher than those from cit-
ies for which the survey was completed by a
single staff member whose responsibility involved
only parks or only streets.
Finally, several of the model's criteria involve
preservation of existing trees, particularly in na-
tive forests. Although it was not intentional, cities
without a native forest resource were undoubt-
edly penalized in their overall score. We
mini-
mized this penalty by averaging scores for other
criteria within the component group and applying
the average to the questions dealing with a na-
tive forest resource.
We caution that the results obtained from this
survey are not representative of the municipal
urban forestry programs nationwide, for several
reasons. First, we did not sample communities
at random. Rather, we included cities about which
we had prior knowledge of their urban forestry
program or government structure. On this basis,
the participant cities would be expected to score
higher than those from a randomly selected
Table 7. Attitudes of city departments about the
urban forest (as perceived by the respondents).
Attitude about the urban forest
Department Excellent Fair
b
Poor
0
Responses
Public utility
Parks
School district
Forestry
Planning
Boards and
commissions"
Public works
Public housing
Assessment
districts
6
17
6
22
14
14
5
4
6
14
7
17
8
7
11
9
4
25
24
24
22
22
21
21
13
10
group.
Second, respondents to the survey repre-
sent a positive self selection
(i.e.,
only communi-
ties with positive feelings about urban forests took
the time to complete the survey). This includes
both urban forestry professionals and elected
officials. For cities that responded to the survey,
the mayor felt strongly enough about the issue to
pass the request on to the appropriate
staff.
While we believe the sustainability model can
be a powerful tool to communities in assessing
sustainability, results of the survey identified some
limitations. For example, the model's criteria are
equally weighted. Given the central importance
of the tree resource in making an urban forest,
this component area may warrant stronger
weighting.
We also recognize that a
long,
com-
plicated survey is not the most efficient method
of assessment and are investigating more con-
cise,
well-structured evaluation forms.
Table
8.
Involvement of community groups in man-
agement of urban forests.
Topic
No.
of responses
•Trees are great
"Neutral
Trees are liability
"E.g.,
design review, tree board
Importance of community groups to management of public
trees
Critical 8
Important 10
Neutral 5
Negative 0
Contributions of community groups to urban forest
Direct financial support 6
Indirect financial support 7
Direct labor 11
General awareness 15
Issues associated with positive citizen attitudes
Support/demand tree planting 5
Environmental benefits 5
(including energy conservation)
General support 3
Neighborhood improvement 3
Articles/editorials in newspaper 2
Increased property values 2
Dislike large-scale removal 2
Other 7
Issues associated with negative citizen attitudes
Uplifted sidewalks 7
Litter (leaves, fruit); demand 6
removal,
prevent planting
Interfere with signs and lights 4
Homeowner responsibility for care 2
View pruning 2
Vandalism 2
Other 7
120
Clark and Matheny: Urban Forest Sustainability Model
Table 9. Urban forest assessment tools used by
responding cities.
Assessment tool
Geographic information
system
Complete inventory
Partial inventory
Mapping from aerial
photographs
Other
(e.g.,
windshield
survey)
Resource type
Public tree;
street
10
9
11
11
10
park
10
6
11
12
7
3
other
1
—
—
2
—
Private
trees
3
—
1
6
2
Literature Cited
Clark, J.R., N.P. Matheny, G. Cross, and V. Wake.
1997.
A model of urban forest sustainability. J.
Arboric. 23:17-30.
Ellis,
J. 1995. Integrated approaches for achieving
sustainable development of urban storm drainage.
Water Sci. Technol. 32:1-6.
McPherson, E.G., D.J Nowak, and R.A Rowntree.
1994.
Chicago's Urban Forest Ecosystem: Results
of the Chicago Urban Forest Climate Project. USDA
For. Serv. Northeast For. Exp. Sta. Gen. Tech. Rpt.
NE-186. Radnor, PA. 201 pp.
Sacramento Urban Forest Task Force. 1996. State of
the Urban Forest Report 1996. City of Sacramento/
Sacramento Tree Foundation. Sacramento, CA.
44 pp.
Thompson, R., N. Pillsbury, and R. Hanna. 1994. The
elements of sustainability in urban forestry. Urban
Forest Ecosystems Institute.
Calif.
Poly. St. Univ.
San Luis Obispo, CA. 56 pp.
Acknowledgements. We are particularly indebted
to the elected officials and staff of the cities who re-
sponded to the survey. The research presented in this
paper would not have been possible without their con-
tributions. Thanks also to Gordon Bradley and Genni
Cross for their review and advice on creating the sur-
vey, to
Tom
McClimon and the U.S. Conference of May-
ors for their help in distributing the surveys to the
participant cities, to Judy Thrapp for her assistance in
producing and compiling the surveys, to Joshua Sum-
mit for his invaluable guidance in analyzing survey
data,
and to Greg McPherson, Bob Miller, and an
anonymous reviewer for their very helpful comments
and suggestions. Funding for this project was provided
by a grant from the National Urban and Community
Forestry Advisory Council through the USDA Forest
Service Urban and Community Forestry Challenge
Cost-share Program (No. G-5-94-20-095).
HortScience, Inc.
P.O.
Box 754
Pleasanton, CA 94566
